Journal of Baltic Science Education, Vol. 20, No. 1, 2021

Vol.20, No.
1, 2021
ISSN 1648-3898 /Print/
ISSN 2538–7138 /Online/
The International Journal of the Scientia Socialis Ltd., & SMC “Scientia Educologica”
Journal of Baltic Science Education, Vol. 20, No. 1, 2021
Editorial Board
ISSN 1648–3898 /Print/
Editor-in-Chief ISSN 2538–7138 /Online/
Prof., Dr. Vincentas Lamanauskas Siauliai University, Lithuania
Deputy Editor-in-Chief
Prof., Dr. Andris Broks University of Latvia, Latvia
Deputy Editor-in-Chief
Prof., Dr. Miia Rannikmäe University of Tartu, Estonia
Executive Secretary
Dr. Laima Railienė SMC “Scientia Educologica”, Lithuania
Editors:
Prof., Dr. Boris Aberšek University of Maribor, Slovenia
Prof., Dr. Agnaldo Arroio University of Sao Paulo, Brazil
Prof., Dr. Martin Bilek Charles University, Czech Republic
Dr. Paolo Bussotti University of Udine, Italy
Prof., Dr. Bulent Cavas Dokuz Eylul University, Turkey
Prof., Dr. Salih Cepni Uludag University, Turkey
Dr. Peter Demkanin Comenius University, Slovakia
Dr. André du Plessis Nelson Mandela University, South Africa
Dr. Markus Sebastian Feser University of Hamburg, Germany
Prof., Dr. Jack Holbrook University of Tartu, Estonia
Prof., Dr. Ryszard M. Janiuk Maria Curie Sklodowska University, Poland
Dr. Milan Kubiatko Jan Evangelista Purkyně University in Ústí nad Labem, Czech Republic
Dr. Todar Lakhvich Belarusian State Medical University, Republic of Belarus
Prof., Dr. Jari Lavonen University of Helsinki, Finland
Dr. Rita Makarskaitė-Petkevičienė Vilnius University, Lithuania
Prof., Dr. Paul Pace Malta University, Malta
Prof., Dr. Valfrids Paškevičs Daugavpils University, Latvia
Prof., Dr. Jongwon Park Chonnam National University, Korea
Prof., Dr. Raffaele Pisano University of Lille, France
Prof., Dr. Pavol Prokop Comenius University in Bratislava, Slovakia
Prof., Dr. Konstantinos Ravanis University of Patras, Greece
Dr. Alona Rauckienė-Michaelsson Klaipėda University, Lithuania
Dr. Dušica Rodić University of Novi Sad, Serbia
Prof., Dr. Heimo Saarikko Helsinki University, Finland
Prof., Dr. Hae-Ae Seo Pusan National University, Korea
Dr. Uladzimir K. Slabin University of Oregon, USA
Prof., Dr. Andrej Šorgo University of Maribor, Slovenia
Prof., Dr. Borislav V. Toshev Sofia University, Bulgaria
Prof., Dr. Georgios Tsaparlis University of Ioannina, Greece
Dr. Muhammet Usak Kazan Federal University, Republic of Tatarstan, Russian Federation
A scientific journal JBSE issued by the Scientia Socialis Ltd., in cooperation with SMC Scientia Educologica, Lithuania,
emphasizes theoretical, experimental and methodical studies in the field of science education. JBSE is an international
academic journal. In order to maintain the high standards appropriate to such a journal, all contributions received are
submitted for anonymous review by two experts, additionally to review by the Editor. The decision of the Editor on
the acceptance of articles is final and no correspondence can be entered into on reasons for rejection of a submitted
contribution.
Published since 2002 Address:

The journal is published bimonthly. Scientia Socialis, Ltd.
IF: 1.024 (2018) Donelaičio Street 29, LT-78115 Siauliai, Lithuania
H Index: 14 (2019) E-mail: [email protected]
SJR: 0.435 (2019) Phone: +370 687 95668
ICDS: 10.7 (2017) Home page: http://www.scientiasocialis.lt/jbse/
ICV: 159.24 (2019) Skype: scientia12
ISSN 1648–3898 (Print) © Scientia Socialis Ltd., a member of the Lithuanian

Publishers Association, Lithuania, 2020
ISSN 2538–7138 (Online)
The articles appearing in this journal are indexed/abstracted in Copernicus Index, EBSCO:
2 Academic Search Premier, Web of Science Core Collection (Social Sciences Citation Index ),
SCOPUS, ProQuest, Road, Crossref, ERIC and DOI.
ISSN 1648–3898 /Print/ CONTENTS
CONTENTS
Editorial
THE CONCEPT OF INERTIA: AN INTERDISCIPLINARY APPROACH

Paolo Bussotti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Articles
THE ROLE OF MATH AND SCIENCE EXPOSURE ON THE EFFECT OF 5E INSTRUCTIONAL MODEL
IN PHYSICS CONCEPTIONS
Michael Allan A. Bahtaji . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
THE EFFECTS OF STEM ACTIVITIES ON 6TH GRADE STUDENTS’ CONCEPTUAL DEVELOPMENT

OF SOUND
Ayşegül Dedetürk, Aslı Saylan Kırmızıgül, Hasan Kaya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
DEVELOPING A SELF-EFFICACY SCALE TOWARD PHYSICS SUBJECTS FOR LOWER-

SECONDARY SCHOOL STUDENTS
Mustafa Fidan, Meric Tuncel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
IMPACT OF INSTRUCTION ON SCIENCE PERFORMANCE: LEARNING INITIATIVE AS A

MEDIATOR AND GENDER AS A LIMITED MODERATOR
Tao Jiang, Ji-gen Chen, Yin-yin Wu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
STUDENT, PARENT AND TEACHER PERCEPTIONS ON THE BEHAVIORAL CHARACTERISTICS

OF SCIENTIFIC CREATIVITY AND THE IMPLICATIONS TO ENHANCE STUDENTS’ SCIENTIFIC
CREATIVITY
Insun Lee, Jongwon Park . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
THE CONCEPTUAL STRUCTURE OF CHEMICAL EQUILIBRIUM IN UPPER-SECONDARY SCHOOL

STUDENTS: EVIDENCE FROM FACTOR ANALYSIS
Yuhua Mai, Yangyi Qian, Linshen Li, Haihang Lan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
PEDAGOGICAL MODEL FOR DECOLONISING, INDIGENISING AND TRANSFORMING SCIENCE

EDUCATION CURRICULA: A CASE OF SOUTH AFRICA
Maxwell Jnr Opoku, Angela James . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
THE PREDICTORS OF NINTH GRADE STUDENTS’ ATTITUDES TOWARDS PLANTS

Meryem Selvi, Emel Çelepçıkay İslam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
GRADE 6 & 9 STUDENT AND TEACHER PERCEPTIONS OF TEACHING AND LEARNING

APPROACHES IN RELATION TO STUDENT PERCEIVED INTEREST/ENJOYMENT TOWARDS
SCIENCE LEARNING
Moonika Teppo, Regina Soobard, Miia Rannikmäe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
EFFECT OF A STEM-ORIENTED COURSE ON STUDENTS’ MARINE SCIENCE MOTIVATION,

INTEREST, AND ACHIEVEMENTS
Liang-Ting Tsai, Cheng-Chieh Chang, Hao-Ti Cheng. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
ANALYSIS OF TWO-TIER QUESTION SCORING METHODS: A CASE STUDY ON THE LAWSON’S

CLASSROOM TEST OF SCIENTIFIC REASONING
Shao-Na Zhou, Qiao-Yi Liu, Kathleen Koenig, Qiu-ye Li, Yang Xiao, Lei Bao. . . . . . . . . . . . . . . . . . . 146
Information
INFORMATION FOR CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
PROBLEMS OF PSYCHOLOGY IN THE 21ST CENTURY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
NATURAL SCIENCE EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
EDUCATION POLICY, MANAGEMENT AND QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
3
Paolo Bussotti
University of Udine, Italy
“Every body perseveres in its state of rest, or of uniform motion in a right line, unless it is compelled to change
that state by forces impressed thereon” (p. Newton 1846, p. 83).
This is the famous first axiom or law of motion stated by Newton in his masterpiece The Mathematical principles
of natural philosophy (ivi). Everywhere, in the courses of physics at the high school level the inertia principle is the
first to be taught. However, there are many doubts that most of learners fully grasp its numerous and fundamen-
tal nuances, which are necessary for a satisfying introduction to physics. Therefore, I propose an interdisciplinary
approach for the explanation of this principle in which history of science and analysis of the daily experiences are
joined to offer a complete comprehension of the concept of inertia.
Most of the young people aged 15-16 years are tied to their immediate sensorial experiences, which do not
induce to think that the inertia law might be valid: if you throw an object, you observe that it does not prosecute
along a straight line, but it tends to fall on the ground following a strange trajectory. At all appearances, the trajec-
tories of the objects are different according to the material of which the object is composed. Also, in conditions in
which apparently no impressed force exists, the movements of the objects are anything but uniform and rectilinear.
If a learner is interested in the movements of the skies –also granted that he/she knows the Copernican system -,
he/she will be impressed by the regularities of such movements, which, certainly, are not rectilinear uniform. But
which force can exist in the skies that makes the bodies to deviate from a rectilinear path travelled with uniform
speed? Let us suppose that a young be extremely gifted, that he/she is able to overcome all these difficulties: there
will be, however, a crucial difficulty which very likely will be not solved in the learner’s mind. It is so expressible:
“dear teacher, you speak of rest and uniform motion in a right line, but in respect to what do you determine these
kinetic states? For me, in my car, I am at rest and the external world is moving; for the external world, I am moving.
What is the situation?”
No fear, no hurry: the same questions and doubts which are in the minds of the pupils were also in the minds of
the scientists of the 16th and 17th centuries, and they solved them in the admirable period between 1543 (Copernicus
published The revolutions of the celestial spheres, Copernicus 1978) and 1687 (Newton published The Mathematical
principles of natural philosophy, Newton 1846). This is the golden age of Scientific Revolution.
To recall in synthesis the steps through which the scientists overcame such difficulties is the best way to ap-
proach the basic concepts of physics. Furthermore, this will allow the teachers to connect the lessons of physics
with those of history and philosophy (in the countries where this discipline is taught at high school level). This is
a profound interdisciplinary approach.
Let us divide this didactical proposal into steps:
1. Relativity of the motions. It is necessary to clearly explain that no body is at rest or in motion in an absolute
sense. It depends on the choice you do. In most cases, for us it is convenient to consider the Earth at rest, though
we know that it is not a rest. Our sensorial experience is misleading because it induces us to think the Earth to
be at rest. However, if the teacher is careful, some sensorial and visual experiences can indicate the relativity of
motion. Each of us has experienced that when a train starts slowly, the passengers have the impression that the
train remains at rest and that the station is moving. History of science can be used to avoid that the concept of
4
https://doi.org/10.33225/jbse/21.20.04
ISSN 1648–3898 /Print/ THE CONCEPT OF INERTIA: AN INTERDISCIPLINARY APPROACH
(pp. 4-9)
relativity of motion be trivialised. It is appropriate to remember that Aristotle in his Physics thought the Earth to be
absolutely at rest, and that the kinetic states of the other bodies should (not might) be determined in respect to
the Earth. Though not all of the scientists shared Aristotle’s view, his Physics was one of the most successful books
in the scientific-philosophical panorama of the Antiquity and Middle Ages. Copernicus felt the need to highlight
the relativity of the motions (he was not the first one to guess such relativity, but in his epoch, it was not granted
at all). Copernicus wrote:
Every observed change of place is caused by a motion of either the observed object or the observer or, of course, by
an unequal displacement of each. For when things move with equal speed in the same direction, the motion is not
perceived, as between the observed object and the observer, I mean. It is the earth, however, from which the celes-
tial ballet is beheld in its repeated performances before our eyes. Therefore, if any motion is ascribed to the earth, in
all things outside it the same motion will appear, but in the opposite direction, as though they were moving past it
(Copernicus 1978, book I, chapter 5, pp. 11-12).
Thus, a recall to history is a good means for the pupils to understand that the concepts of physics are at the basis
of our view of the world and that without the notion of relativity of the movements, the Copernican system would
have been impossible. This is of help to guess the interconnections among knowledge which, at a superficial view,
might appear unrelated but that, in fact, are. Of course, collaboration with the teacher of History might be conceived
to insert Copernicus’ thought within the picture of the Renaissance.
2. Approach to inertia. After having spoken of the relativity of any motion, the teacher of physics might enter the
classroom developing this experiment without any initial explanation. Let us consider two joined equal inclined
planes as in the Figure 1:
Figure 1
First Phase of the Experiment: Two Inclined Planes with the Same Inclination
A little ball is in A at the altitude h. If you precise that the two planes have to be considered completely smooth,
probably all of the learners will intuitively claim that when the ball has reached the point B, it will ascend on the
plane BC in C at the same altitude it had in A, not higher, not lower.
Let us not suppose that the plane BC’ is less inclined than AB on the horizon, as in the following Figure 2.
Figure 2
Third Phase of the Experiment: Two Inclined Planes with Different Inclinations
5
(pp. 4-9)
The learners will recognise that the little ball will arrive at C’ at the same altitude it had in A, not higher, not lower.
Finally, suppose that the plane BC’ becomes always less inclined on the horizon until becoming perfectly horizontal
like this (Figure 3).
Figure 3
Third Phase of the Experiment: Inclined Planes of Various Inclinations until a Horizontal Plane
Now ask the learners: consider the situation in which the second plane is completely horizontal; the ball arrives at
B, what happens then? Probably most of them will answer correctly: the ball arrives at B with a certain speed, after
that it continues its motion along the horizontal plane maintaining the same speed as in B. There is no sufficient
reason because something different happens. Well, you will tell your pupils: “very good, you have arrived on your
own at the inertia principle. Bravo! Not only: you have followed the same train of thoughts through which Galileo
reached the concept of rectilinear inertia”. For, Galileo in the Second Day of his Dialogue Concerning the Two Chief
World Systems (1632, Galilei 1967) developed a reasoning whose synthesis has been just described. The teacher
might read this important passage of the Dialogue (ivi, pp. 145-146) to the class, so that the learners, through
Galileo’s detailed descriptions and explanations, could realise all the nuances of this concept and the difficulties
encountered by the most brilliant minds of mankind to reach it.
Thus, the human aspect of science will be also pointed out: the scientists have reached the notions which nowa-
days guide our knowledge of the world through a difficult road, full of obstacles. Only their hard work permitted
to overcome the obstacles. A clever mind is nothing without the work.
The way in which Galileo (and the students!) have reached the idea of inertia allows us to develop some interesting
conceptual and didactical considerations:
a) The whole reasoning relies on the assumption that the surfaces of the two inclined planes are perfectly
smooth, without any friction. In fact, no such surface exists in the reality. No space devoid of any fric-
tion exists. What does this mean? Is the inertia principle not valid? Here, the teacher has to be careful
to distinguish the world and our model of the world. The inertia principle belongs to our model of
the world. It shows a tendency of the real bodies, which is indispensable to understand their kinetic
behaviour. It is clear that in reality no motion is perfectly inertial. The model has to offer a scheme in
which, abstracting from some properties P of the bodies, we are able to focus our attention on proper-
ties Q we consider more profound. In the reality P and Q cohabit. In the case of inertia, the property
P is represented by the existence of friction, constrains and obstacles, the property Q is the inertial
tendency of the rectilinear uniform motions. If we wanted to study the motions under the condition
that a friction exists, we will include the friction in the properties Q and so on. Each situation analysed
by physics deals with our theoretical model which is a simplified view of reality. Of course, our model
is the more satisfactory the more aspects of reality it will explain and foresee, but every model will
abstract from some aspects of reality.
b) Our experiment did not demonstrate the mathematical, logical validity of the inertia principle. No
such demonstrations can exist in classical physics, because, as the name itself clarifies, that of inertia
is a principle, an axiom, a postulate; it is not proved through a series of reasonings. Rather, it is one of
the axiomatic bases to develop theorems and propositions. Galileo’s and our experiments make the
principle plausible, reasonable; but do not demonstrate it. Newton was perfectly right to assume inertia
principle as an axiom.
c) Since a uniform rectilinear motion continues without the need of any external action, this means that
such an action – which the physicists from Newton onwards call force - has not to produce velocities,
6
(pp. 4-9)
but change of velocities, that is accelerations (I give for granted that the learners know the vectors’ el-
ementary properties). Let us now see, the basis of the concept of force.
3. Forces. In the world you observe that most motions are not uniform and rectilinear. Therefore, inertia principle
is not sufficient in our physics. It is necessary to identify an action which modifies the kinetic state of the bodies. It
is clear that such action has to be proportional to acceleration. Newton identified the proportionality factor which
ties force and acceleration with the mass (considered as quantity of matter) of a body (Definition VIII of Newton’s
Principles). From here the famous formula F=ma. This definition of force seems reasonable: after all, from an intui-
tive standpoint, if we have two objects – let us suppose two carts – of the same form and of the same material, we
struggle the more to move the cart the bigger it is, because it has more mass, more matter than the smaller one.
This definition of force seems acceptable also considering the forces which produce circular motions. Consider a
sling and a stone at its extremity. We make to rotate the sling. As in the previous case, we struggle to move the sling
the bigger the stone is. At all appearances, things work perfectly. Following Newton, you will identify the inertial
mass (he called it innate force, vis insita, Definition III) with the resistance of a body to modify its state of rest or of
uniform rectilinear motion. This is fully coherent with the general definition of mass as quantity of matter: the more
massive a body is, the more it is difficult to modify its inertial state. Now, following this partially historical itinerary,
you are ready to introduce the third Newton’s principle of dynamics: to any action an equal and opposite reaction
corresponds. This is intuitive as well: if you hit a wall, you feel a pain, it is as if the wall hit you.
The learners should have no difficulty to understand the order of thoughts traced until now. Some elements of
the history of physics have been useful to highlight the kind of difficulties the learners can encounter as well as
their solutions.
However, something is missing in our construction and something very important: the forces we have introduced
are supposed to act through a direct contact, but on the Earth, there is another action, which acts on all the bod-
ies independently of their composition, of their form, of their nature. This force is gravity. But gravity does not act
through a direct contact. Nonetheless, there is no doubt that it changes the state of motion of a body. Furthermore:
what is the usual way in which we measure the mass of a body? Certainly, we do not check how difficult to move
a cart is, or how strong the resistance of a stone is when we put a sling in rotation. There is by far a more usual
and easier solution: we pose a body on a balance, we read on the balance its weight, and we divide the weight by
the acceleration of gravity g. Thus, we obtain the mass. This is our normal procedure. It is fully consistent with our
definition F=ma. In this case a=g. But here there is a conceptual problem on which the learners have to reflect from
the beginning of their education in physics: if we try to determine the mass of an object through the resistance it
opposes to any change of its kinetic state, it seems that we determine something completely different in respect
to the mass of the same object considered as the way in which it is attracted by the Earth. This is the gravitational
mass of a body, not its inertial mass. Why do we speak of “mass” in both cases? The answer is that classical physics
has no answer to this question. If, empirically, you measure the inertial mass and the gravitational mass of the same
body, you find the same value for the two masses. Numerous experiments have always given the same results. But
in Newton’s physics there is no theoretical explanation on the coincidence of the two masses.
The set of these considerations is necessary for the learners to achieve a correct idea of physics, of its conceptual
problems and of its theoretical value. They will guess that we cannot claim that science explains everything without
posing some unexplained presuppositions which have the role of the axioms in mathematics. Not only: some-
times we have to accept situations as that connected to the problem as inertial mass / gravitational mass which
belong neither to the presuppositions nor to the consequences of a theory. It happens accidentally in classical
physics, in spite of the fact that the coincidence between the two masses implies fundamental consequences. The
learners – or at least some of them – will be curious to know whether a theory which explains such coincidence
exists, and whether – more in general – a theory can offer complete explanations of all the phenomena it deals
with. You will immediately explain that, though science cannot explain everything, it is the most reliable form of
knowledge available for mankind. On these bases, you might also develop a brief reasoning on the general nature
of knowledge, which might be very useful and stimulating for the pupils. Thus, the proposed conceptual-historical
approach might induce many interesting questions in the minds of the learners.
The inertial reference frames. We have seen that in our common experience every motion is relative, we have seen
what inertia is, and we have seen what a force is. Now there is the fundamental and most difficult question: if every
7
(pp. 4-9)
motion is relative, how is it possible to recognize an inertial kinetic state from one which is not? In other terms: to
what should inertia be referred since inertia is such a fundamental property which, being connected to the mass,
seems an absolute and not a relative property? This is one of the most difficult conceptual theoretical questions in
classical physics. Reflecting on this problem, the learners – before dealing with more specific concepts, problems
and developments - will fully grasp the conceptual frame in which classical physics is inserted. Here the historical
development plays a decisive role. Newton, at the dawn of modern physics, conceived one of the most genial rea-
sonings ever developed in science (final section of the Definitions in his Principles). He thought to have identified
an absolute motion and an absolute space to be considered as the real inertial reference frame. The concept of
inertia should be referred to such absolute space. Newton’s argument is based on the celebrated bucket experi-
ment. I recommend the teachers to explain this experiment in any detail to their pupils. I will restrict to refer to the
experiment and to Newton’s interpretations. Broader considerations on this experiment might be the occasion for
a further Editorial in our JBSE!
Figure 4
The Various Phases of the Bucket Experiment
Let us consider a bucket hanging at a rope as in the Figure 4. Act like this:
(1) Twist the rope until it is very hard and keep the rope twisted;
(2) Fill the bucket of water;
(3) Free the rope. You observe that the bucket begins to rotate, but you will observe no movement in
the water (picture 1, Figure 4);
(4) After a little bit time the bucket, because of the friction between the particles of the water and those
of the bucket’s walls communicates the movement to the water, which begins to rotate and to ascend
along the bucket’s walls (picture 2, Figure 4);
(5) After a while, the bucket stops its movement, but the water continues to rotate (picture 3, Figure 4);
(6) Finally, both the bucket and the water are at rest (picture 4, Figure 4).
Newton interpreted the rotational movement of the water as due to its inertia: during their movement, each
particle of water tends to escape along its instantaneous tangent because of the inertia principle. This inertia ten-
dency cannot be actualised because of the bucket’s walls. Thence, the particles of water heap up on the bucket’s walls.
Once clarified that the ascent of the water’s particles is due to their inertia, Newton wondered to what the
inertia might be referred. His reasoning can be summarised in the following Table 1:
Table 1
A Summary of Newton’s Reasoning
Relative motion water/bucket Water’s surface Inertial forces
1. Max Flat No
2. No Curved Yes
3. Max Curved Yes
4. No Flat No
8
(pp. 4-9)
Is it possible that the inertia of water depends on the relative motion between water and bucket? No, it is not,
because you note the effects of the inertial tendency (to summarise, but with a little linguistic impropriety, called
“inertial forces”) both in the case in which the relative motion water/bucket does not exist (phase 2) and in that
in which it is maximal (phase 3). You note the lack of inertial forces both in the case in which the relative motion
water/bucket is maximal (phase 1) and in that in which no relative motion exists (phase 4).
Thence, Newton concluded that inertia cannot be referred either to the bucket or, even the less, to other objects
close to the water. He referred inertia to an Absolute Space. In this way, the concept of inertia is well founded; it
is possible, at least in principle, if not in practice, to distinguish the inertial reference frames from the others: they
are the frames which are at rest or move rectilinearly and uniformly in respect to the Absolute Space.
Is this picture completely satisfactory? The answer to a next Editorial!
References
Copernicus, N. (1978). On the Revolutions. Edited by F. Dobrzycki. Translation and Commentary by E. Rosen. Polish Scientific
Publishers and Macmillan Press. Latin edition 1543.
Galilei, G. (1967). Dialogue concerning the two chief world systems. Translation by S. Drake, Introduction by A. Einstein. University
of California Press. Second Edition. Italian edition 1632.
Newton, I. (1846). The mathematical principles of natural philosophy. Reprint of A. Motte’s Translation (1727). Daniel Adee. Three
Latin editions: 1687, 1713, 1726.
Received: January, 10, 2021 Accepted: February 02, 2021
Cite as: Bussotti, P. (2021). The concept of inertia: An interdisciplinary approach. Journal of Baltic Science Education, 20(1), 4-9.
Paolo Bussotti PhD, Researcher, University of Udine, via Palladio 8, 33100 Udine,
Italy.
E-mail: [email protected]
ORCID: https://orcid.org/0000-0002-7883-8618
9
THE ROLE OF MATH AND
SCIENCE EXPOSURE ON THE
ISSN 1648-3898 /Print/ EFFECT OF 5E INSTRUCTIONAL
ISSN 2538-7138 /Online/ MODEL IN PHYSICS
CONCEPTIONS
Abstract. 5E instructional model is Michael Allan A. Bahtaji

commonly utilized in science teaching to
promote conceptual learning. However, the
benefit of 5E instructional model cannot be
fully attained if the knowledge and skills
essential to learning were not properly
established from past. Similarly, the devel- Introduction
opment of new concepts cannot be fully
attained if the important processes incul- Instructional model, like the 5E instructional model, is a learning process
cated in the 5E model were not considered model designed to promote students’ decision making, problem solving abil-
in teaching. In this research, the effects of ity and deeper understanding of concepts. The model encouraged students
5E model instruction and previous learning to use one’s own knowledge and personal experiences to understand new
experiences on the conceptual learning of concepts, design and perform inquiry processes, work collaboratively with
students in Newtonian mechanics were peers, and apply the concepts learned to various perspectives (Rodriguez
evaluated. Using 2 x 2 factorial research et al., 2019; Yaman & Karaşah, 2018). These processes, which can be traced
design, the conceptual understanding of from the early social constructivist perspective and which can be observed
172 undergraduate students was evaluated within the social context, are important in the development of new scientific
before and after the experiment. These concepts and are crucial to the progress of 21st century skills or information
participants were distributed into four age skills (Sotáková et al., 2020; Vygotsky, 1978; Yaman & Karaşah, 2018).
groups with different programs. The result However, the pedagogical goal of 5E model will likely not succeed if the
revealed that both 5E instructional model essential skills and knowledge needed for the processes to take place were
and previous learning experiences contrib- not properly established from previous learning experiences (Ruppert et al.,
uted to the conceptual learning of students 2019). For example, in the learning process “engaging”, students who lack
in mechanics. Those students who received fundamental skill and knowledge in doing science inquiry and hypothesis
5E instructional model and whose exposure testing are more likely to acquire numerous trial and error during an ex-
to science and math courses is high, gained periment (van Riesen et al., 2018). Similarly, students who lack fundamental
the highest increase in the mean score knowledge about force and motion are more likely to experience difficulties
after the experiment. This suggests that in relating variables about force and motion and unlikely to gain conceptual
the effectiveness of 5E instructional model understanding in physics. These skills and knowledge, which developed prior
can be fully achieved if the knowledge and to students’ exposure to science and math courses from their previous years,
skills essential to learning and conceptual are considered important for the success of 5E instructional model. More-
change were properly established from pre- over, the development of schema is associated to early learning experiences
vious learning experiences and activities. (Graham et al., 2012). The association suggests that the ability of a person
Keywords: 5E model, conceptual change, to interpret a certain phenomenon is related to the appropriateness of the
conceptual learning, learning cycle, prior schemas that have been developed from past learning experiences (Rowland
knowledge et al., 2007). The implication suggests that exposures to science and math
courses from the past learning experiences could serve as a good parameter
to estimate the appropriateness of schemas needed to interpret problems
in Newtonian mechanics. Thus, higher math and science exposure could
Michael Allan A. Bahtaji
Technological University of the Philippines, lessen students’ misconceptions in Newtonian mechanics which can result
Philippines to correct interpretations of force and motion. Moreover, the development
of schemas that students used in making sense of things around them hap-
pens during social interaction. Learning through social interactions creates
10
ISSN 1648–3898 /Print/ THE ROLE OF MATH AND SCIENCE EXPOSURE ON THE EFFECT OF 5E INSTRUCTIONAL
MODEL IN PHYSICS CONCEPTIONS
ISSN 2538–7138 /Online/ (pp. 10-20)
a plane to which the students internalize the information they received from their surroundings. The impact of
students’ learning experiences on the process of conceptual change reflects the importance of prior knowledge
in the development of physics conceptions. However, despite of its significant role in learning, prior knowledge
will less likely support conceptual change if instructional learning processes, like the processes stipulated in the
5E instructional model, will not be considered both by the learners and the teachers during the learning process
(Arslan et al., 2015). For example, to activate curiosity and to make learning meaningful, students must realize the
relevance of new concept they are learning, which happens during internalization processes or during sense mak-
ing process. Despite of the importance of both 5E instructional model and learning experiences on the conceptual
learning of the students in Newtonian mechanics, few studies have explicitly explored the interaction effect of
5E model and learning experiences in physics learning, and almost none have been conducted in the domain of
Newtonian mechanics in the tertiary level. In response to the said challenge, this research has explicitly examined
the effect of both instructional learning cycle model and previous exposures to science and math courses in the
conceptual learning of students in Newtonian force using 2 x 2 factorial research design.
Conceptual Change

The 5E instructional model is established on a framework called constructivist learning theory. This framework
explains that building of knowledge and meaning is a result of one’s interaction with the environment. The funda-
mental framework that explains how learning happens in social interaction is the social constructivist perspective
pioneered by Vygotsky (1978). According to this framework, social situations play an important role in learning. These
situations include the exchange of information as a result of communication and collaboration between students.
Moreover, the important role of prior knowledge in teaching and learning has been heavily studied in the field of
science education research since the 1970s (Scott et al., 2007). One of the common research findings shows that
learners across all levels hold concepts that are different to the concepts accepted in the scientific field. Wherein,
designed teaching intervention is often used to improve students’ conceptions. The common term usually coined
with these odd conceptions is misconceptions (Rowlands et al., 2007). Misconception is said to be originated from
daily life interactions with the environment, which in common situation are considered very useful in dealing with
everyday life activities (Brown & Hammer, 2008). This is the reason why learners stand firm with the misconceptions
they have and restrain instructional interventions designed to correct the misconceptions (Brown & Hammer, 2008;
Rowlands et al., 2007; Scott et al., 2007). In the cognitive framework of learning, restructuring and reorganizing of
knowledge are the processes that happen during conceptual change and are considered very important in learn-
ing and teaching. In some situation, learners need to undergo the process of conceptual change before they can
gain a valid and justified understanding about their world. Restructuring of knowledge is very difficult for learners
because it is during this process that learners need to connect existing concepts with the present concept they
need to learn. Reviewing all research in the field of conceptual change, the majority of them have considered the
cognitive aspect of conceptual change conceptualized by the group of Posner (1982). The group suggested that
there are four conditions that need to attain before conceptual change happens. First, learners need to be dissatis-
fied with the current conceptions, second, learners should comprehend the new conceptions introduced to them,
third, learners find the new conceptions reasonable and probable, and fourth, learners realize the meaning and
use of the new conceptions. Based on the model proposed by Posner et al., (1982), cognitive conflict is considered
the main factor that triggers cognitive disequilibration, which then results in the process of conceptual change.
Learners realize the implausibility of their misconceptions and consider the plausibility of new conceptions. Aside
from being subjected to plausibility, conceptual change is also characterized as progressive and continual rather
than appearing instantly (Rowland et al., 2007; Scott et al., 2007). Constructivist theory highlighted the importance
of prior knowledge in the process of conceptual change, arguing that the process of conceptual change is gradual
which involves refinement of prior knowledge into a more structured and complex form, instead of replacing prior
knowledge with a new accepted one (Schneider et al., 2012). Another theory that describes the formation of new
conceptions is called the schema theory. Schema theory explains how learners organize their concepts to form a
new concept. Schema refers to a group of concepts that learners used to make sense of their world. The schema
used by students to make sense of their world can be associated to students’ learning experiences, thus those
students who have gained high exposures to a field they had studied before may develop a schema conducive to
the same field they are studying at present (Rowlands et al., 2007). In this research, the effect of students’ level of
exposure to science and math on students’ conceptual learning was evaluated.
11
THE ROLE OF MATH AND SCIENCE EXPOSURE ON THE EFFECT OF 5E INSTRUCTIONAL
(pp. 10-20) ISSN 2538–7138 /Online/
5E Instructional Model
Reviewing some research papers in the field of science education that evaluated the effect of instructional
models in learning, common result reveals that 5E instructional model is effective in improving students’ concep-
tual learning (Arslan et al., 2015; Ceylan et al., 2009; Ören & Tezcan, 2009). The idea about learning cycle was first
introduced by Atkin and Karpus (1962) using three stages namely exploring, discovering and creating. Relatively,
in the model developed by Bybee and colleagues (2006), they suggested five stages of learning cycle activities,
namely engaging, exploring, explaining, elaborating and evaluating. These processes were commonly known
as 5E instructional model. In the engaging process, learners develop curiosity toward the concepts they need to
learn and explore their prior knowledge to establish meaning about those concepts. During this process, learners
are prompted to engage cognitively with the concepts they need to learn. When questions formed during the
engagement process were answered, learners are said to be in the process of exploring. During this process, learn-
ers engage in inquiry learning activities to gather more evidence about the concepts they are studying. To deepen
understanding of the concepts learned, students engage in discussion activities that will clarify questions they
formed. During this stage, learners are said to be in the process of explaining. To make those concepts meaning-
ful and useful, learners need to realize the uses and applications of those concepts to their life. During this time,
learners are said to be in the process of elaboration. In the fifth cycle, evaluation stage, learners are encouraged
to evaluate the processes they designed or evaluate the concepts they learned to further enhance the concepts
they learned. To study the contribution of 5E instructional model in the conceptual learning of students in physics,
its main effect was evaluated.
Exposure to Science and Math Courses
Prior knowledge is considered one important factor that influences students’ learning development (Hai-
likari et al., 2008). The quality of understanding students can attain during the learning process depends on the
degree of prior knowledge acquired by the students from their previous experiences. The quality and amount
of knowledge acquired by students from previous learning experiences can be associated with the students’
exposure to science and math courses taken before. Those students who credited more courses in the field of
science and math could possibly acquire more prior knowledge in a field they have taken before, thus easily
utilize the processes needed to undergo conceptual change. Previous exposures to science and math courses can
contribute to the development of schema helpful in understanding the concepts of force and motion. Previous
learning experiences in science and math courses contribute to the acquisition of concepts that learners can
utilize in solving problems related to force and motion. It is during the process that learners retain information
that is useful in their future learning activities. Previous learning experiences improve learners’ problem-solving
skills, critical thinking and creativity which are important in the process of conceptual change. Previous learning
experiences promote the development of skill in knowledge restructuring. It is also the stage when skill in as-
similation and accommodation is developed. Thus, one factor that needs to be considered in engaging learners
to learning cycle is the level of learning experiences they gained from the past. Since the ability of the student to
undergo conceptual change depends on the prior knowledge established from previous learning experiences,
the effectiveness of 5E instructional model relies also on the knowledge and skills students established from
previous learning experiences. Students’ high school background affects students’ performance in physics. In
one of the research conducted before, the authors found that those students who credited physics courses in
their senior-level high school scored higher compared to those students who were not able to credit any physics
course in their senior-level high school (Harlow et al., 2014). The number of math courses completed in the past
also contribute to the performance of students in FCI. In the research conducted by McCullough (2002) among
300 undergraduate students, the findings revealed that those students who completed the highest number of
courses in math in their early studies performed better in the FCI compared those students who completed the
lowest number of math courses in their past studies. Thus, to prove the role of previous learning experiences
on the conceptual learning of the students, the effect of students’ exposure to science and math courses to
conceptual learning of the students was evaluated.
To explicitly examine the interaction effect of both teaching method and exposure to science and math courses
on the conceptual progress of the students in Newtonian force, the following questions have been formulated.
12
ISSN 2538–7138 /Online/ (pp. 10-20)
1. How did the 5E instructional model contribute to the conceptual progress of the students in physics,
particularly in the field of Newtonian mechanics?
2. How did the level of exposure in science and math courses contribute to the conceptual progress of
the students in Newtonian mechanics?
3. What is the interaction effect of both the teaching method and science and math exposure in the
conceptual progress of the students in Newtonian mechanics?
Research Methodology
General Background
2 x 2 factorial research design was utilized to evaluate the interaction effect of both teaching method and
exposure to science and math (EMS) on students’ conceptual understanding in Newtonian mechanics. Two factors
(teaching method and ESM), consisting of two levels, were explored in this research. The factor ‘teaching method’
consists of two levels, the ‘5E instructional model method’ and the ‘conventional method’. The 5E instructional model
method uses learning processes, such as 5Es, to guide students to develop deeper understanding of the concepts
they need to learn. Conventional method, on the other hand, can be described as an expository teaching approach
that usually starts through the review of previous lessons and ends through problem solving activities. The review
is followed by an introduction of new concepts, presentation of solutions to problems, then problem solving activi-
ties for students. Aside from the teaching method, the ESM factor also consists of two levels, the ‘high ESM’ and the
‘low ESM’. High ESM means higher exposure to science and math courses, while low ESM means lower exposure
to science and math courses. The combination of two different levels has resulted to 2 x 2 factors or groups: 5E
learning cycle method with ‘high ESM’ group; 5E instructional model method with ‘low ESM’ group; conventional
method with ‘high ESM’ group; conventional method with ‘low ESM’ group. Figure 1 shows the group distribution.
Figure 1
Group Distribution Using 2 x 2 Factorial Design
Teaching method
5E instructional model method Conventional method
Exposure to science and math High ESM Group 1 Group 3

(ESM)
Low ESM Group 2 Group 4
Participants
This research was carried through at a tertiary university in the Philippines. The participants are 178 2nd year
undergraduate university students (95 males, 83 females; age distribution from 18 to 22) who are all enrolled in
physics course that includes Newtonian mechanics in the course syllabus. The participants are distributed into
four sections comprising of 43 students in engineering technology program, 42 students in teaching program,
47 students in computer science program, and 46 students in apparel and fashion design program. Considering
the purpose of a research, it has been suggested that the minimum number of participants for each group must
not be less than 30, thus a sample size higher than the suggested minimum sample size will least likely affect the
reliability of the research results (Cohen et al., 2018, p. 203). The participants in the first and third groups credited
the highest number of science and math courses prior to physics course, thus attained higher ESM. On the other
hand, the participants in the second and fourth groups credited the least number of science and math courses
prior to the current course in physics, thus attained least ESM. The participants were considered because their early
exposure to science and math courses differs across the groups, which makes the groups comparable in terms of
their early ESM. Other than that, most of the students who enter the school comes from the different regions of
the country. Thus, they may be considered a good representation of the population. Furthermore, the treatment
given to the participants was assured as the teacher handling the course is the same. This maintains the validity
of the results of the research.
13
(pp. 10-20) ISSN 2538–7138 /Online/
Research Instrument

To find out how the teaching methods and learning experiences contributed to the conceptual understanding
of the students in Newtonian force, their physics conceptions on the said domain were pre-tested and post-tested
using the Force Concept Inventory (FCI) instrument. The FCI is a 30-item concept inventory modelled to assess
students’ conceptions in Newtonian mechanics (Hestenes et al., 1992). Each item contains five options, one option
is correct while four options are incorrect. The instrument contains items that probe students’ understanding about
force and motion in one or two dimensions. The distractors in the instrument represent students’ misconceptions.
Topics in Newtonian mechanics like conservation of energy and momentum were excluded in the instrument. To
test its reliability, the instrument was piloted to 73 university students in engineering technology program. The
result (α = .82) revealed a good statistical reliability. The instrument has been already utilized in other studies locally,
and the results of reliability test indicate acceptable reliability. For example, in the research conducted by Morales
(2017) on the conceptual progress of undergraduate students in Newtonian concepts, the computed coefficient
alpha (α = .71) was statistically interpreted as good reliability.
Procedure

To figure out the interaction effect of the 5E instructional model and ESM on the conceptual gain of the students
in Newtonian mechanics, three phases have been followed in this research. The first phase includes sampling, filling-
out inform consent and pre-testing. The non-probabilistic convenience sampling has been considered to directly
determine the role of learning experiences on the effect of teaching method on the conceptual understanding of
students in mechanics. This sampling method has been considered so that the researcher could gather information
that could provide answers to the research questions. Through proper consent, students were informed about the
purpose of the research, the benefit that the research might contribute to them and to science education, and
the confidentiality of the research. Students after this step were pre-tested using the FCI. Two teaching methods
have been applied in the second phase of this research, the 5E instructional model method and the usual method
used in teaching physics, which in this research was labelled as ‘conventional method’. The first and the second
class received 5E instructional model learning condition while the third and the fourth class received the usual
method of teaching. Students’ conceptual learning in Newtonian mechanics was re-evaluated using the concept
inventory at the last phase of the research. All the scores gained by the students before and after the instructions
were analyzed using proper statistical tools.
Data Analysis

The interaction effect of both the learning condition and ESM on the conceptual learning of the students
in Newtonian mechanics was analyzed using Analysis of Covariance (ANCOVA). The validity of ANCOVA test was
determined using the homogeneity regression slope test. Scattered plots (post-test scores vs pre-test scores)
with respect to the independent variables were created to establish the validity of AVCOVA test. The intercorrela-
tion between the dependent variables and the independent variables has been conducted to confirm the main
effect of 5E instructional model and ESM on the post-test scores of the students in FCI. To determine which learn-
ing condition and level of ESM has significantly contributed to students’ pre-test and post-test scores in the FCI,
pairwise comparisons were computed. To evaluate the students’ gain in score after the interventions, the pre-test
mean-score and post-test mean score for each group were also compared using t-test.
Research Results
To determine the main effect of both teaching method and exposure to science and math (ESM) on the
post-test scores of the students in Force Concept Inventory (FCI), Analysis of Covariance (ANCOVA) was computed.
Homogeneity regression slope test was conducted to confirm the validity of the ANCOVA test. Results revealed a
non-statistical difference between the teaching method and ESM (F = 2.64, p > .01), between the teaching method
and pre-test scores (F = 2.16, p > .01), and between the pre-test scores and ESM (F = 0.58, p > .01). Results also
showed that the regression slopes between the teaching method, ESM, and the pre-test scores are homogenous
14
ISSN 2538–7138 /Online/ (pp. 10-20)
to each other (F = .19, p > .01). The homogeneity or the non-significant difference between the slopes affirmed
the validity of ANCOVA test. The scattered plots between pre-test and post-test scores in term of independent
variables were shown in Figure 2.
Figure 2
Regression Slopes in Terms of Teaching Method and Exposures to Science and Math (ESM)
Note. High ESM = high exposures to science & math; Low ESM = low exposures to science & math.
The results of ANCOVA showed that pre-test scores (covariance) have significantly contributed to the FCI
scores of the students in post-test (F = 6.48, p < .05, ɳ2 = .04). Moreover, results also showed that both teaching
methods (F = 21.42, p < .01, ɳ2 = .11) and ESM (F = 226.19, p < .01, ɳ2= .57) have significantly contributed to the
post-test scores of the students in the FCI. The ANCOVA results also indicated that the ‘combination effect’ of the
independent variables (teaching method and ESM) also contributed to the post-test scores of the students in FCI (F
= 7.52, p < .01, ɳ2= .04). The results confirmed the hypothesis that 5E instructional model and students’ exposures
to science and math courses have a main effect on the post-test scores of the students in FCI.
The intercorrelations among the independent and dependent variables were presented in Table 1, similarly,
the significant difference between the variables was also presented in the said table. Using the conventional way
in interpreting correlation coefficients articulated by Schober et al., (2018), the results revealed a negligible but
non-significant correlation between ESM and teaching method (r = .001, p > .05), a negligible but non-significant
correlation between pre-test scores and teaching method (r = .005, p > .05), a negatively weak but significant cor-
relation between post-test and teaching method (r = ̶ .179, p < .05), a negatively strong but significant correlation
between pre-test and ESM (r = ̶ .819, p < .01), a negatively strong but significant correlation between post-test
scores and ESM (r = ̶ .841, p < .01), and a positively moderate but significant correlation between post-test and
pre-test scores (r = ̶ .635, p < .01).
Table 1
Correlations among Dependent and Independent Variables
Teaching method ESM Pre-test Post-test
Teaching method 1
ESM .001 1
Pre-test .005 ̶ .819** 1
Post-test ̶ .179* ̶ .841** .635** 1
** p < .01, two-tailed. * p < .05, two-tailed.

15
(pp. 10-20) ISSN 2538–7138 /Online/
To determine the main effect of teaching method and ESM on the pre-test scores of the students in FCI, pair-
wise comparisons were presented in Table 2. A non-significant difference between the pre-test mean scores of 5E
instructional model and traditional method has been observed. The hypothesis that teaching method has an effect
on the pre-test scores of the students was rejected. Furthermore, result revealed a significant difference between
pre-test mean score of high ESM and low ESM. This rejects the hypothesis that the students’ level of exposures to
science and mathematics has no effect on the pre-test mean scores of the students.
Table 2
Pre-test Pairwise Comparisons
Dependent Teaching method ESM

Variable (N = 178) (N = 178)
Learning Cycle Group Conventional Group High ESM Low ESM

groups
(n = 90) (n = 88) (n = 85) (n = 93)
M 5.02 5.03 6.00 4.05

SD 0.07 0.07 0.08 0.07
MD 0.01 1.95**
Note. ESM = exposure to science and math; n = number of participants in a group; N = total number of participants; M = mean;
MD = mean difference.

Moreover, to determine the main effect of teaching method and ESM on the post-test scores of the students
in FCI, pairwise comparisons were computed. The results revealed that both teaching method and ESM have a
main effect on the post-test scores of the students in FCI. Whereas the difference between the post-test scores of
5E instructional model and traditional groups was statistically significant, same with the post-test mean scores of
high ESM and low ESM groups. The hypothesis of non-significant effect of teaching method and ESM on the post-
test mean scores of the students in FCI has been rejected.
Table 3
Post-test Pairwise Comparisons
Dependent Teaching Method ESM

Variable (N = 178) (N = 178)
Learning Cycle group Conventional group High ESM Low ESM

groups
(n = 90) (n = 88) (n = 85) (n = 93)
M 6.10 5.61 7.06 4.65

SD 0.08 0.08 0.08 0.08
MD 0.50** 2.41**
Note. ESM = exposure to science and math; n = number of participants in a group; N = total number of participants; M = mean.
MD = mean difference.
The change in the pre-test and post-test mean scores of the four groups of students was presented in Table 4.
The group who has high ESM and was exposed to 5E instructional model gained the highest increase in the mean
score after the instructions while the group who has low ESM and was exposed to the traditional teaching method
gained the lowest increase in the mean score. The results also revealed a significant increase in the mean score of
those groups who were exposed to 5E instructional model, but no significant increase for the group who received
conventional method.
16
ISSN 2538–7138 /Online/ (pp. 10-20)
Table 4
Pre-test and Post-test Paired Sample Test
Pre-test Post-test
Teaching Method ESM n MD t-value
M SD M SD
High ESM 43 5.98 0.71 7.16 0.72 1.19 7.73**

5E instructional model
Low ESM 47 4.06 0.70 5.04 0.69 0.98 6.25**
High ESM 42 6.02 0.64 6.95 0.66 0.93 5.76**
Conventional
Low ESM 46 4.04 0.70 4.26 0.80 0.22 1.22
Total 178 4.98 1.19 5.80 1.43 0.82 9.60**
Note. ESM = Exposure to Science and Math; n = number of participants; M = mean; SD = standard deviation; MD = mean differ-
ence.
Discussion
The interaction effect of both teaching methods (5E instructional model and traditional) and ESM on the
conceptual learning of students in physics was evaluated in this research. To establish the validity of analysis of
covariance (ANCOVA) test in finding the main effect of both independent variables, homogeneity regression slope
test was conducted. Results revealed a non-significant difference on the regression slopes, indicating that AN-
COVA test is valid for testing the main effect of 5E instructional model and ESM on the FCI scores of the students.
The main effect of the teaching method on the FCI scores of the students was first studied. Results revealed that
the teaching methods have significantly contributed to the FCI scores of the students after the interventions (F =
21.42, p < .01, ɳ2 = .11). The significant effect indicates that the teaching methods have significantly contributed
to the conceptual understanding of the students in Newtonian mechanics. This main result was further analyzed
using pairwise comparison for both pre-test and post-test. Results revealed that the post-test mean score of the
5E instructional model group is significantly higher than the post-test mean score of the conventional group
(Table 3). The significant result suggests that the 5E instructional model is more effective compared to the con-
ventional method in improving students’ conceptual understanding in physics. The effectiveness of the 5E instruc-
tional model has been confirmed by the correlation coefficient between the teaching method and the post-test
results (r = ̶ .179, p < .05) in Table 1, indicating that the 5E instructional model group significantly exceeded the
conventional group in the post-test results. Furthermore, pairwise comparison also revealed that both teaching
methods have no main effect on pre-test scores of the students (Table 2), indicating that the significant difference
in the mean scores of the two groups was only observed after students were exposed to the interventions. The
effectiveness of 5E instructional model in the conceptual progress of students in physics has been also confirmed
in the pre-test-post-test mean score comparison in Table 4. Wherein, those students who received 5E instruc-
tional model method obtained the highest increase in post-test mean score as compared to those who received
the conventional way of teaching. The effectiveness of 5E learning model in students’ conceptual learning was also
observed in other studies. In the research conducted previously, the authors proved that 5E instructional model
is very effective in promoting students’ conceptual learning in cell division and reproduction (Arslan et al., 2015).
Wherein those students who received 5E instructional model significantly gained in their post-test score as com-
pared to those who received conventional teaching. The effect of ‘5E model’ on the conceptual learning of students
in physics suggests that this teaching model helped students activate conceptual learning in physics. During the
process, learners restructure new information in order to acquire an understanding of the present situations. This
occurs when learners engage in ‘inquiry processes activities’ (van Riesen et al., 2018). This instructional model,
which derived from the constructivist learning perspective, promotes knowledge construction, especially when
learners internalize the information acquired during learning. The effect of students’ exposure to science and math
on their conceptual learning was also evaluated. Results revealed that students’ EMS significantly contributed to
the FCI scores of the students after the interventions (F = 226.19, p < .01, ɳ2= .57).The result indicates that earlier
exposure to science and math courses contributed to the conceptual progress of students in Newtonian mechan-
17
(pp. 10-20) ISSN 2538–7138 /Online/
ics. The result was further analyzed using pairwise comparisons. Results revealed that the pre-test mean score of
the high ESM group significantly exceeded the pre-test mean score of the low ESM group (Table 2). Similarly, result
also revealed that the post-test mean score of the high ESM group significantly exceeded the post-test mean score
of the low ESM group (Table 3). These outcomes indicate that exposure to science and math has a significant
contribution to the conceptual understanding of students in physics. The main effect of ESM on both pre-test and
post-test scores of the students has been confirmed by the correlation coefficients between ESM and pre-test
scores (r = ̶ .819, p < .01) and between ESM and post-test scores (r = ̶ .841, p < .01). The correlations indicate that
the high ESM group significantly exceeded the low ESM group both in the pre-test and post-test mean scores.
Furthermore, the contribution effect of ESM has been also confirmed in the pre-test-post-test mean score com-
parison in Table 4, those groups who completed the highest number of courses in science and math obtained the
highest increase in the post-test mean score as compared to those groups who credited the least number of
courses in science and math. In a research that evaluated the relationship between students’ high school back-
ground and FCI performance, the authors proved that those students who took physics in the senior-level high
school scored significantly higher than those students who did not take senior-level high school physics course
(Harlow et al., 2014).This suggests that the students’ earlier exposure to physics courses is one of the factors that
contributed to their FCI scores. In the research conducted by McCullough (2002) among 300 undergraduate non-
physics students about the effect of math background on the performance of students in FCI results revealed that
those students who completed the highest number of math courses in the senior high school and early under-
graduate level obtained the highest mean score in the FCI test as compared to those students who credited the
lowest number of courses in math. Students’ early exposure to science and math courses contributed to the con-
ceptual progress of the students in physics. Wherein those students who have high exposure in science and math
courses perform better than those students who have low exposure in science and math courses. Learning expe-
riences from the past is considered the main source of schemas that students used when interpreting unfamiliar
situations (Graham et al., 2012). Schema, which consists clusters of related concepts, is useful when analyzing
problems in Newtonian mechanics. Aside from that, previous exposure to science and math courses promotes
learners’ problem-solving skill and critical thinking skill, which are important during the process of conceptual
change. Learning processes such as restructuring of knowledge are part of early learning experiences (Scot et al.,
2007).To determine how the interaction between independent variables contributed to the conceptual under-
standing of the students in physics, the interaction effect of both teaching methods and ESM on conceptual learn-
ing of the students was evaluated. The result revealed that both teaching method and ESM have a main effect on
the FCI test results (F = 7.52, p < .01, ɳ2= .04). The main effect indicates that both teaching method and ESM have
contributed to the conceptual understanding of students in Newtonian mechanics. The combinational effect of
the said factors in the FCI test scores can be further analyzed on the paired sample test in Table 4. Based from the
2 x 2 experimental design, the result revealed that the group who has high ESM and at the same time received 5E
instructional model gained the highest increase in the FCI test, while the group who has low ESM and at the same
time received conventional teaching gained the lowest increase in the FCI test. These results imply that higher
exposure to science and math courses contribute to students’ conceptual learning and become more effective if
the 5E instructional model is considered in teaching. Prior knowledge, which developed from previous learning
experiences, could serve as schema which students utilize when interpreting situations unfamiliar to them. How-
ever, the activation of a schema will become gradually if learners will not be engaged in learning activities that
involve observations, predictions, testing, application, and evaluation. These processes, embedded in the 5E in-
structional model, promote conceptual learning.
Conclusions and Implications
Relating students’ prior knowledge in the concepts they need to learn, engaging learners in inquiry-based
learning activities and helping students to apply and validate the concepts they learned are part of students’ learn-
ing development. These activities, inculcated in the 5E instructional model, are very important for the process of
conceptual change. However, many teaching practices are still relying on traditional way of teaching, that is more
teacher-centered rather than student-centered teaching. Important learning processes such as engaging, explor-
ing, explaining, elaborating and evaluating are often not emphasized during the learning process. However, even
though ‘5E learning model’ is considered, learners can still experience impediment in understanding the concepts
they need to learn if the knowledge and skills relevant to their present situation were not properly developed from
18
ISSN 2538–7138 /Online/ (pp. 10-20)
previous learning experiences. Same also when students’ exposure to science and math courses from previous
studies is very low, the essential skills and knowledge needed by students to undergo the process of conceptual
change become less available when they encounter questions unfamiliar to them. Whereas, the effectiveness of 5E
instructional model cannot be fully achieved if the important skills and knowledge needed to process conceptual
change were not properly developed from previous learning experiences. In this research the interaction effect of
both 5E leaning and previous learning experiences to students’ conceptual learning was studied. Results revealed
that both 5E learning model and previous exposures to science and math courses contributed to the conceptual
learning of the students. Those students who received 5E instructional model during the instructions significantly
performed better on the FCI test than those who were assigned to conventional teaching method. The effective-
ness of 5E instructional model suggests that the model promotes conceptual understanding, promotes students’
curiosity, encourages learners to engage in interactive discussion, and it encourages learners to apply the skills
and knowledge they learned. On the other hand, those students who credited the highest number of courses in
science and math in their previous studies significantly performed better compared to those who credited the
smallest number of courses science and math. The significant result suggests that previous learning experiences
serve as a source of knowledge which students restructure to develop new concepts. Furthermore, the interac-
tion effect of both teaching methods and learning experiences also contributed to the conceptual learning of
the students. Whereas, those students who received 5E instructional model and credited the highest number of
courses in science and math gained the highest increase in the mean score after the experiments compared to
the other groups. The outcome suggests that prior knowledge developed from previous learning experiences
plays an important role in the development of new knowledge and should be considered when using 5E learn-
ing in teaching. Students’ unproductive learning experiences from the past studies could be one factor why the
effectiveness of 5E instructional model in learning is not fully attained. Similarly, when knowledge construction
and skill development were not given emphasis during instructions, the knowledge learned from the past often
fails to transform into a new knowledge. The effective conceptual learning can be fully attained if two conditions
were satisfied: well established prior knowledge and high emphasis on knowledge construction during teaching.
Acknowledgements
The researcher would like to express his profound acknowledgement to all who participated in this research
activity.
References
Arslan, H. Ö., Geban, Ö., & Sağlam, N. (2015). Learning Cycle Model to foster conceptual understanding in cell division and
reproduction concepts. Journal of Baltic Science Education, 14(5), 670-684. http://www.scientiasocialis.lt/jbse/?q=node/462
Atkin, J. M., & Karplus, R. (1962). Discovery of invention? Science Teacher, 29(5), 45–51. https://doi.org/10.2307/24146536
Brown, D. E., & Hammer, D. (2008). Conceptual change in physics. In S. Vosniadou (Ed.), International handbook on conceptual
change (pp. 127-154). https://doi.org/10.1007/s11191-010-9283-6
Bybee, R., Taylor, J., Gardner, A., Scotter, P. V., Powell, J. C., Westbrook , A., & Landes, N. (2006). The
BSCS 5E instructional model: Origins, effectiveness, and applications (Full report). Colorado Springs: BSCS.
https://bscs.org/resources/reports/the-bscs-5e-instructional-model-origins-and-effectiveness/
Ceylan, E., & Geban, Ö. (2009). Facilitating conceptual change in understanding state of matter and solubility concepts by using 5E Learning
Cycle Model. Hacettepe University Journal of Education, 36, 41-50. http://www.efdergi.hacettepe.edu.tr/shw_artcl-1582.html
Cohen, L., Manion, L., & Morrison, K. (2018). Research method on education (8th ed.). Routledge.
Graham, T., Berry, J., & Rowlands, S. (2013). Are ‘misconceptions’ or alternative frameworks of force and motion spontaneous
or formed prior to instruction? International Journal of Mathematics Education in Science and Technology, 44(1), 84-103.
http://dx.doi.org/10.1080/0020739X.2012.703333
Hailikari, T., Katajavuori, N., & Lindblom-Ylanne, S. (2008). The relevance of prior knowledge in learning and instructional design.
American Journal of Pharmaceutical Education, 72(5), 113. https://doi.org/10.5688/aj7205113
Harlow, J. J. B., Harrison, D. M., & Meyertholen, A. (2014). Correlating student interest and high school preparation with learning
and performance in an introductory university physics course. Physical Review Special Topics-Physics Education Research,
10(1), Article 010112. https://doi.org/10.1103/PhysRevSTPER.10.010112
Hestenes, D., Wells, M., & Swackhamer, G. (1992). Force concept inventory. The Physics Teacher, 30, 141-158. https://doi.
org/10.1119/1.2343497
McCullough, L. (2002). Gender, math, and the FCI [Paper presentation]. In Physics Education Research Conference 2002. Boise, Idaho.
https://doi.org/10.1119/perc.2002.pr.013
Morales, M. P. E. (2017). Exploring indigenous game-based physics activities in pre-service physics teachers’ conceptual change
19
(pp. 10-20) ISSN 2538–7138 /Online/
and transformation of epistemic beliefs. Eurasia Journal of Mathematics, Science and Technology Education, 13(5), 1377-1409.
https://doi.org/10.12973/eurasia.2017.00676a
Ören. F. Ş., & Tezcan, R. (2009). The effectiveness of the learning cycle approach on learners’ attitude toward science in seventh
grade science classes of elementary school. Elementary Education Online, 8(1), 103-118. http://www.ilkogretim-online.org.
tr/index.php/io/article/view/1693/1529
Posner, G. J., Strike, K. A., Hewson, P. W., & Gertzog, W. F. (1982). Accommodation of a scientific conception: Toward a theory of
conceptual change. Science Education, 66(2), 211–227. https://doi.org/10.1002/sce.3730660207
Rodriguez, S., Allen, K., Harron, J., & Qadri, S. A. (2019). Making and the 5E learning cycle. The Science Teacher, 86(5), 48-55. https://
doi.org/10.2505/4/tst18_086_05_48
Rowland, S., Graham, T., Berry, J., & McWilliam, P. (2007). Conceptual change through the lens of Newtonian mechanics. Science
& Education, 16, 21-42. https://doi.org/10.1007/s11191-005-1339-7
Ruppert, J., Duncan, R. G., & Chinn, C. A. (2019). Disentangling the role of Domain-Specific Knowledge in students modeling.
Research in Science Education, 49(3), 921-948. https://doi.org/10.1007/s11165-017-9656-9
Schneider M., Vamvakoussi X., & Van Dooren W. (2012). Conceptual change. In: Seel N. M. (Eds), Encyclopedia of the sciences
of learning. Springer. https://doi.org/10.1007/978-1-4419-1428-6_352
Schober, P., Boer, C., & Schwarte, L. (2018). Correlation coefficients: Appropriate use and interpretation. Anesthesia & Analgesia,
126(5), 1763-1768. https://doi.org/10.1213/ANE.0000000000002864
Scott, P., Asoko, H., & Leach, J. (2007). Students conceptions and conceptual learning in science. In S. K. Abell & N. G. Lederman
(Eds.), Handbook of research on science education (pp. 31-56). https://doi.org/10.4324/9780203824696
Sotáková, I., Ganajová, M., & Babinčáková, M. (2020). Inquiry-based science education as a revision strategy. Journal of Baltic
Science Education, 19(3), 499-513. https://dx.doi.org/10.33225/jbse/20.19.499
van Riesen, S. A. N., Gijlers, H., Anjewierden, A. A., & de Jong, T. (2018). The influence of prior knowledge on experiment
design guidance in a science inquiry context. International Journal of Science Education, 40(11), 1327-1344.
https://doi.org/10.1080/09500693.2018.1477263
Vygotsky, L. S. (1978). Mind in society: The development of higher process. Harvard University Press. https://www.jstor.org/stable/j.ctvjf9vz4
Yaman, S., & Karaşah, Ş. (2018). Effects of learning cycle models on science success: A Meta-analysis. Journal of Baltic Science
Education, 17(1), 65-83. http://www.scientiasocialis.lt/jbse/?q=node/640
Received: August 24, 2020 Accepted: January 08, 2021
Cite as: Bahtaji, M. A. A. (2021). The role of math and science exposure on the effect of 5e instructional model in physics
conceptions. Journal of Baltic Science Education, 20(1), 10-20. https://doi.org/10.33225/jbse/21.20.10
Michael Allan A. Bahtaji PhD Science Education, Research Specialist, Technological

University of the Philippines, Ayala Boulevard Ermita Manila,
Philippines.
Website: http://www.tup.edu.ph
20
THE EFFECTS OF STEM
ACTIVITIES ON 6TH GRADE
STUDENTS’ CONCEPTUAL ISSN 1648-3898 /Print/
DEVELOPMENT OF SOUND ISSN 2538-7138 /Online/
Ayşegül Dedetürk, Abstract. Engineering design process-

oriented STEM activities are activities that
Aslı Saylan Kırmızıgül,
involve solving a real-world problem with
Hasan Kaya teamwork by using integrative engineering
design and engineering applications using
scientific and mathematical concepts and
21st century skills. The research aimed to
examine the effects of engineering design
process-oriented STEM activities, which
Introduction were developed for sound concept, on
the conceptual development of 6th grade
Societies are dominated and directed by ideas and products from sci- students. In the research, comparative case
ence and technology, and undoubtedly, the impact of science and technol- study design was used. The study group
ogy on life will continue to increase in the next years (Jenkins, 2002). Science, consisted of 40 students attending a middle
technology and innovation are highly interdependent. Our daily needs are school in Kayseri, Turkey. The teaching was
met with the information obtained through science, and technology applica- carried out by the activities in science cur-
tions are included in almost every step taken to make our lives easier. Societies riculum in the control group, while it was
that do not want to lag behind in scientific and technological innovations aim carried out by engineering design process-
to educate individuals who understand mathematics and science concepts oriented STEM activities in the experimen-
well and associate these concepts with daily life. Countries that want to tal group. Each student was individually
advance in science and technology are adopting new teaching approaches interviewed using 11 open-ended questions
that will make students competent in science and technology. In a similar in a verbal format. The mean of Kappa
manner, in Turkish science curriculum, one of the eight basic competences coefficients of the questions was 0.93. As
that students will need in their personal, social, academic and business life at a result of content analysis, 11 categories,
national and international level is determined as ‘mathematical competence including 21 themes and 81 codes, were
and basic competences in science/ technology’. In addition to this update, obtained. When the differences between
‘life skills’ and ‘engineering and design skills’ were added to science process pre- and post-interviews were examined, it
skills. Thus, students are expected to show performances such as designing was seen that conceptual understanding
a project, setting models and products, and advertising a product. In the levels of the students in experimental group
curriculum, which prioritizes the implementing and providing input to the were more positive than the students in
economy properties of science, the topics and learning outcomes are handled control group. This shows that engineering
with an approach that considers the production of technologies to meet the design process-oriented integrated STEM
needs of daily life (MoNE, 2018). activities have positive effects on students’
STEM education aims to integrate science, technology, engineering and conceptual understanding of sound.
mathematics with multidisciplinary, interdisciplinary and transdisciplinary
Keywords: case study, conceptual devel-
approaches (Vasquez et al., 2013). STEM education is thought to be critical
opment, engineering design, STEM
for the development of students’ 21st century skills (Topcu & Gokce, 2018).
Science curriculum aims to enable students to look at the problems from
the interdisciplinary point of view by integrating science with mathemat-
Ayşegül Dedetürk, Aslı Saylan
ics, technology and engineering (MoNE, 2018). In this context, engineering Kırmızıgül, Hasan Kaya
design-based STEM (science, technology, engineering, and mathematics) Erciyes University, Turkey
education, which enables students to gain a multidisciplinary point of view
and to achieve the competencies and performances aimed by the curriculum,
21
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
comes to the fore. Wicklein (2006) has emphasized that engineering design provides an ideal substructure for the
integration of mathematics, science and technology, and stated that STEM education should focus more on engi-
neering design method. Bybee (2010) has also stated that engineering should be more involved in STEM education.
One of the models for integrating STEM content and applications is Engineering Design Process-oriented
Integrated STEM Model (Savran Gencer et al., 2019). Five basic characteristics distinguish the integrated STEM
education (Moore et al., 2015, p. 24):
a) The content and practices of one or more anchor science and mathematics disciplines define some of
the primary learning goals;
b) The integrator is the engineering practices and engineering design of technologies as the context;
c) The engineering design or engineering practices related to relevant technologies requires the use of
scientific and mathematical concepts through design justification;
d) The development of 21st century skills is emphasized; and
e) The context of instruction requires solving a real-world problem or task through teamwork.
The problem is determined in the first step of engineering design process. In the second step, the research is
conducted for the problem. In the third step, the ideas are developed to solve the problem. In the fourth step, the
best idea(s) is chosen considering its material, practical and economic assessments. In the fifth step, a prototype for
the selected solution is created. In the sixth step, it is checked whether the prototype actually solved the problem,
and the selected material is suitable or not. In other words, the solutions are tested and evaluated. In the seventh
step, the prototype(s)/ solution(s) is presented, and feedback is received from other groups to determine whether
there are defects. In the last step, a final decision is made, and the process is completed (Hynes et al., 2011).
Research Problem
The main task of education is to enable students to acquire permanent and applicable knowledge. In this
context, international exams such as PISA (Programme for International Student Assessment) and TIMSS (Trends
in International Mathematics and Science Study) predict the extent to which students can use their knowledge
and skills in their daily lives rather than the degree to which they learn the subjects within the curriculum. Unfor-
tunately, the results of PISA and TIMSS tests have shown that Turkish students achieve lower results rather than
the students from many other countries (Martin et al., 2015; Organization for Economic Co-operation and Devel-
opment [OECD], 2019). Undoubtedly, classroom practices that students encounter in their schools are at the top
of the factors affecting their performance in these exams. Therefore, there is a need to improve science education
through new teaching approaches like STEM.
Science is a fear or concern for many students. Previous research has revealed that students’ success in physics
subjects is lower than the biology and chemistry subjects (Bahar & Polat, 2007; Gagić et al., 2019). As the reasons
for their difficulties in physics, students have stated that physics contains many abstract concepts, the concepts
cannot be related to the daily life, and teachers mostly use direct instruction (Gagić et al., 2019). Therefore, students
are often unmotivated to learn these concepts and they have difficulty in gaining meaningful learning. Sound is
one of the most difficult (Eshach, 2014; Sozen & Bolat, 2014; Timur et al., 2016) and important concepts in phys-
ics that can help students understand physical world around them (Eshach, 2014). Therefore, conducting further
studies on the conceptual understanding of sound is crucial. However, there is some research on the conceptual
understanding of sound (Yerdelen & Sungur, 2020).
It is known that engineering design process-oriented integrated STEM activities positively affect students’
achievement levels in science (Ercan & Sahin, 2015; Harwell et al., 2015; Wendell & Rogers, 2013). However, a limited
number of studies has examined the effects of these activities on students’ conceptual understanding of sound.
These research studies are generally at primary school level (Barnett et al., 2008; Wendell, 2011; Wendell et al., 2010;
Wendell & Lee, 2010). In these research studies, engineering design method has been used to teach the physics of
sound. In Wendell’s (2011) research, 3rd and 4th grade students from three classrooms have enacted the engineer-
ing design-based science curriculum regarding the science of sound topic. Teaching activities on the production,
propagation and characteristics of sound have been developed according to engineering design method and have
been applied. The pre- and post-interviews have been conducted to examine the changes in students’ ideas about
sound. As a result of the content analysis, their conceptual understanding of the three categories (production and
transmission of sound, and characteristics of the sound) have been divided into six themes. A significant increase
has been observed in students’ knowledge levels regarding the three categories. Being a limitation of Wendell’s
22
ISSN 1648–3898 /Print/ THE EFFECTS OF STEM ACTIVITIES ON 6TH GRADE STUDENTS’ CONCEPTUAL DEVELOPMENT
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
(2011) research, teaching the physics of sound to 3rd and 4th graders may be difficult since concrete operational
stage occurs between the ages of 7 and 11 years. The formal operational stage occurs from age 11 to adulthood.
In this stage, individuals have the ability to think in abstract ways. Therefore, teaching the physics of sound to 6th
grade students may be more appropriate. Lastly, throughout the research of Wendell (2011), STEM has never been
mentioned. However, the physics of sound, which is found abstract by students, is very suitable for engineering
design process-oriented integrated STEM activities. STEM education increases the development of not only social
and cognitive, but also psycho-motor aspects of the individuals. Moreover, STEM enables students to connect
learned theoretical knowledge with real life problems. Based on these facts, it is found necessary and important to
examine the effect of the use of engineering design process-oriented STEM activities on middle school students’
conceptual development of sound.
Research Aim and Research Questions
The aim of this research was to examine the effect of the use of integrated STEM activities, prepared in ac-
cordance with the engineering design process, in teaching sound subject on 6th grade students’ conceptual
understanding. In this context, research question is determined as ‘Does the teaching of sound concept through
engineering design-based STEM activities have an effect on the perceptions of 6th grade students?’ and the fol-
lowing sub-questions are created:
1) How do the control and experimental group students understand the sound subject?
2) What are the differences/ similarities between the pre- and post-interviews in terms of control group
students’ conceptual understanding of sound subject?
3) What are the differences/ similarities between the pre- and post-interviews in terms of experimental
group students’ conceptual understanding of sound subject?
Research Design
This research was a comparative case study that aimed to determine and examine comparatively the effects
of the engineering design-based STEM activities on 6th grade students’ conceptual understanding of sound con-
cept. Being one of the qualitative research designs, case study examines in depth the situation(s) that occur within
a certain period of time. Case studies, which aim to understand a specific situation in all its parts, often include
interview questions (Fraenkel et al., 2012). From this point of view, it was decided to conduct an oral achievement
test including open-ended questions in accordance with the purpose and design of the research. The achievement
test was conducted in the spring semester of the 2013-2014 academic year.
Sample
The study group, which was selected through convenience sampling, consisted of 6th grade students from a
middle school in Kayseri, Turkey. Due to the government’s strategy, the class size was decreased in recent years. The
average class size in middle schools located in Turkey is 26 (OECD, 2019). Consequently, 40 students participated
in this research. Each of the experimental (11 female, 9 male) and control group (10 female, 10 male) included 20
students. The experimental and control groups were determined by looking at the students’ 5th grade scores, 6th
grade fall semester grades, and scores on the achievement test, which include 19 multiple-choice and 14 open-
ended questions prepared by researchers (Dedeturk, Saylan Kirmizigul, & Kaya, 2020).
In the research, the principle of voluntary participation was taken into consideration and the students’ identi-
ties were kept confidential by using codes. Numbers were assigned in the order that each student was interviewed.
For instance, participant 1 refers to a student that was interviewed first on the interview timetable. All participants
were informed about the aim of the research and their right to leave the research at any time. Additionally, the par-
ent consent form that included detailed information about the voluntary and conﬁdential nature of participation
was signed by the parents of the participants. Also, the necessary permissions were obtained from the Ministry
of National Education to which the school is affiliated. All the rules stated in the “Higher Education Institutions
Scientific Research and Publication Ethics Directive” were followed in the research.
23
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
Instrument
In Turkey, the topic of sound had been included in the 3rd, 4th, 5th, 6th and 8th graders’ science curricula
before (MoNE, 2013). Currently, the physics of sound topic is included in the 3rd, 4th and 6th grades’ science cur-
ricula (MoNE, 2018). In 6th grade science curriculum, physics of sound has basic concepts such as propagation of
sound, reflection of sound, echo, absorption of sound, sound insulation and acoustic applications. There are 12
learning outcomes for the topic of sound in the science curriculum (Table 1). In order to determine the 6th grade
students’ conceptual understanding level about physics of sound, it was decided to develop a valid and reliable
instrument including the outcomes.
Achievement test can be of different forms like oral, written or practical test. An oral achievement test has been
developed in this study, as it enables a more in-depth investigation of students’ knowledge level and restatement
of unclear answers. In line with this purpose, firstly an item pool including 14 open-ended questions was prepared
based on the learning outcomes. Then, 11 questions were selected among them and an oral achievement test was
prepared. A table of specifications was prepared showing the distribution of the questions in terms of learning
outcomes and cognitive domains in Bloom’s taxonomy (Table 1).
Table 1
Distribution of the Questions in Terms of Learning Outcomes and Cognitive Domains
Cognitive domain in Bloom’s

Learning outcomes (MoNE, 2006) Question
Taxonomy
1. Notices that the sound is propagated in waves in all directions. 1 Analysis
2. Discovers that the sound is reflected when it encounters an obstacle. 2 Comprehension
3. States that an echo occurs as a result of sound reflection. 3 Comprehension

4. Gives examples of how s/he benefited from the reflection of sound in science and
4 Comprehension
technology.
5. Detect that the sound that encounters matter can be absorbed. 6 Analysis
6. Determines that sound loudness decreases through absorption. 6 Evaluation
7. Realizes that different materials absorb sound differently. 7 Evaluation

8. Realizes that the materials used in sound insulation and prevention of echo formation
8 Comprehension
absorb sound well.
9. Explains why a material medium is needed for propagation of sound through granular
5 Comprehension
structure of the medium.
10. Indicates that when the sound encounters matter; passing, absorption and reflection
9 Comprehension
may occur together at different rates, depending on the properties of the materials.
11. Gives examples of acoustic applications in the places like theater, concert hall and
10 Comprehension
historical buildings.
12. Develops and presents projects that can prevent echo formation in closed spaces. 11 Application
In order to ensure the content validity, a table of specifications and a rubric were prepared for the open-ended
questions and were checked by three science education experts. In the light of the feedback received from the
experts, the necessary adjustments were made in table of specifications, questions and rubric. Additionally, cor-
rections were made regarding the intelligibility of the questions and their suitability to the student level. The pilot
study was conducted with 35 sixth graders. The students were interviewed individually in a verbal format. These
students had learned the subject one month before the students in the study group. In order to ensure construct
validity, discrimination and difficulty indexes were calculated by applying item analysis to the data obtained from
35 students. According to Table 2, difficulty indexes of the questions are in the range of 0.50-0.69. Hence, all of the
questions are of medium difficulty. Since the discrimination indexes of all questions are greater than 0.40, they
are very good questions.
24
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
Table 2
The Difficulty (P) and Discrimination (R) Indexes of Interview Questions
Item P r Item P r Item P r
1 0.61 0.78 5 0.64 0.72 9 0.58 0.83
2 0.69 0.61 6 0.58 0.61 10 0.50 1.00
3 0.61 0.78 7 0.58 0.72 11 0.61 0.78
4 0.50 1.00 8 0.53 0.94
The 11-item achievement test, which was finalized as a result of the pilot study, was applied verbally and
individually to each student twice before and after the two-week implementations. Each face-to-face interview
lasted about 10 minutes and they were recorded with a voice recorder with the permission of the students.
Procedures
Following the pre-interview, sound subject was taught for 9 course hours over a 2-week time interval, in both
groups. During this period, all the activities used in the courses were prepared to meet 12 learning outcomes in
the curriculum and for each learning outcome, identified course hours in the curriculum were followed.
The research-based activities in the curriculum were applied and the instructions in the textbook were followed
in control group. Students sometimes worked individually and sometimes in groups. They usually listened to the
teacher during the courses and reached the result indirectly by experimenting with simple materials without using
any measuring device. Activities that do not include graph drawing, data interpretation, and robotic applications
are poor in terms of scientific process skills and they do not include life skills, and engineering and design skills.
The learning activities of control group are given in Table 3.
Table 3
Lesson Titles and Overviews Regarding Control Group’s Learning Activities
Lesson title Overview
1. Let’s observe the propagation The students put a 30 cm ruler on the end of the table, pressed and pulled their hands on the other end of
of sound the table and observed the vibration of the ruler, and listened to the sound of the ruler. Then they placed a
container of water under the ruler and observed the fluctuation in the water.
2. Sound is an energy. The students stretched the thin nylon bag onto the cylindrical can with the help of rubber band and poured
salt on it. They hit the metal tray that they positioned close to the cylinder with a wooden spoon and observed
that the salt vibrates as a result of the vibrations of the particles in the air. By this way, they tried to reach the
conclusion that sound is an energy.
3. Is the sound reflected? The students put the cotton in a glass jar and the alarm clock on it. They listened to the sound of the clock
from different distances and recorded these distances. Then they listened to the sound of clock by holding
the mirror tilted over the jar and recorded these distances. Students observed that sound can be heard from a
farther point when there is a reflector.
4. What feature is common? By examining the photographs given, students explained that the piano’s tail is directed towards the audience
and that the megaphone is conical, so that the sound was reflected and transmitted to the desired location.
5. Modeling of sound transmission In order to understand that the sound is propagated through the particles, the students were arranged in
groups of 6-8 people in a row facing the same direction. The hindmost student put her hand on the shoulder
of her friend in front of her; and he put his hand on the shoulder of his friend in front of him; and the shake
was delivered to the foremost student. By opening the gaps, the propagation of sound in liquid and gaseous
environments was modeled. In each case, students measured the time it took for the shake to reach from back
to front. Accordingly, they reached the conclusion that the speed of sound is slow in gaseous environments.
6. What is the American Indian The students explained the reason why the American Indian in the photograph tries to find out if there is a
doing? danger by resting the ear on the ground as the better transmission of sound in solids.
25
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
7. Which materials can absorb the Students listened to the sound after putting watch into the empty shoe box. Then they put newsprint, cloth
sound more? and styrofoam into the box, respectively, and listened to the sound of the watch. Students explained why the
loudness of sound was different when different materials were placed in the box.
8. I explain with what I have Students examined the places in the photographs and made explanations about the absorption and loudness
learned. of sound.
9. I am comparing sound and light Students filled in the blanks given in the paragraph about sound and light, and the table prepared for sound
and light.
In the experimental group, nine integrated STEM activities focusing on engineering design process developed
by the researchers were included. LEGO Mindstorms NXT education set, which includes LEGO-technics bricks, a
microprocessor, a software, sound, light, distance and touch sensors, and motors that control movement that allow
the students to develop robots on their own, and various materials (sponge, styrofoam, etc.) were used in these
activities. Before the implementation, the teachers received training on STEM activities and they were provided
with a guide including the important points, learning outcomes, duration of the activity and tools, technology and
techniques to be used in each activity. Since they did not have any knowledge and experience about engineering
design method, written and oral information were given to the students about subject. In STEM activities, which
are rich in scientific process, life, and engineering and design skills, students made original designs to solve prob-
lem, made measurements using sound sensors and used LEGO materials as well as simple materials in groups. In
interconnected activities, students collected data by making direct measurements, drawn and interpreted graphs,
and programmed robots. The activities performed in experimental group are given in Table 4.
Table 4
Lesson Titles and Overviews Regarding Experimental Group’s Learning Activities
1. How is sound propagated? Students estimated how sound was propagated. Then they made measurement, recorded
data and compared the results with their predictions to reach a scientific idea.
2. Does the sound get reflected? How? Students reasoned out whether sound was reflected and conducted experiments regarding
the reflection of sound using the given materials and recorded data. Students made a
minimal design about the reflection of sound.
3. How does an echo occur? Students made a design about echo formation. They tested and observed echo formation
through this design. They got an idea of how echo occurred due to their design.
4. What are the technologies produced using the reflec- Using LEGO, students made and tested a design that works owing to the reflection of
tion of sound in your daily life? In which fields does sound. Then they ran their designs with sound using LEGO programming.
science use them?
5. What is the most important condition for the propaga- Students designed using LEGO and other materials to investigate what was needed for
tion of sound? Why? propagation of sound. They wrote a hypothesis for this design and tested it, then they drew
a graph and reached a scientific idea.
6. Why is the loudness levels are different in front and Students made a design including a sound source and an obstacle. Based on this design,
at the back of the obstacle when the sound meets they made measurements and recorded the data, and concluded that the sound had been
an obstacle? Which side of the obstacle has a louder absorbed by the obstacle.
sound? Why?
7. Does each material has its own sound absorption Using data obtained from their design, students investigated which events occur when
rate? What are the properties of noise insulation materi- sound meets matter. They decided which materials should be used in sound insulation by
als? What happens when sound encounters a matter? grouping the materials they used in their designs.
8. What are the acoustic applications you notice in the By examining the given spaces, students discovered what kind of acoustic applications
places you go? could be made.
9. How can you solve the echo problems in closed Students developed and presented a project using the scientific ideas they had until that
spaces? time.
26
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
Figure 1 shows the robot design, programming and presentation stages of different groups belonging to the
activity-4 in the experimental group.
Figure 1
Robot Design, Programming and Presentation Stages of The Lesson-4 in Experimental Group
Data Analysis

Within the scope of the research, students’ answers were scored independently by two science education
experts using the rubric, and Kappa (κ) coefficient was calculated by looking at the consistency between the raters.
This coefficient, which is the proportion of observed agreement that exceeds what would be expected by chance
alone, was found as 0.93. Hence, agreement between the raters is almost perfect (Landis & Koch, 1977). Accordingly,
it was revealed that the rubric was understood by different raters in the same way with the high agreement rate.
The data obtained were transcribed and analyzed using summative content analysis. Although the summa-
tive content analysis involves counting and comparing the frequency of keywords or content and interpreting
the content (Hsieh & Shannon, 2005), the open response format of the qualitative items provides more in-depth
information in addition to the quantitative findings of the research. This method includes using codes, themes
and categories to organize content and arrive at a narrative description of findings (Fraenkel et al., 2012). The in-
terviews generated codes, the codes were grouped into themes, and the themes were grouped into a total of 11
categories. The themes reflected students’ perceptions and thoughts about sound, whereas the codes reflected
the concepts of sound in their minds.
Research Results
The answers to the interview questions were analyzed by using the rubric. According to Table 5, all students
in experimental group answered the questions 4, 5 and 8 correctly in the post-interview. Table 1 shows that these
questions refer to 4., 9. and 8. learning outcomes respectively. While the percentages of correct answers for these
questions by experimental group students were 0%, 10% and 5% respectively in pre-interview, all of these rates
were 100% in post-interview.
27
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
Table 5
Percentages of the Answers to the Interview Questions
% %
Question Answer Test Question Answer Test
Control Experimental Control Experimental
group Group group group
pre 5 10 pre 15 10
Correct Correct
post 20 70 post 20 95
Partially pre 55 50 Partially pre 35 45

1 7
correct post 55 25 correct post 55 5
pre 40 40 pre 50 45
Incorrect Incorrect
post 25 5 post 25 0
pre 25 30 pre 15 5
Correct Correct
post 30 90 post 15 100

2 8
pre 15 30 pre 50 65
Incorrect Incorrect
post 15 0 post 40 0
pre 15 0 pre 0 5
Correct Correct
post 30 90 post 15 75

3 9
pre 25 20 pre 45 50
Incorrect Incorrect
post 15 0 post 15 0
pre 10 0 pre 5 0
Correct Correct
post 15 100 post 25 95

4 10
pre 80 85 pre 60 70
Incorrect Incorrect
post 65 0 post 60 0
pre 25 10 pre 5 0
Correct Correct
post 35 100 post 10 90

5 11
pre 55 70 pre 30 55
Incorrect Incorrect
post 35 0 post 25 0
pre 10 5
Correct
post 30 80
Partially pre 80 80
6
correct post 50 20
pre 10 15
Incorrect
post 20 0
Table 5 shows that the percentage of correct answers increased for all questions for experimental group.
Accordingly, conceptual understanding level of the students in experimental group increased relatively more
28
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
than the control group students. In post-interview, experimental group students gave incorrect answer only to
the first question and this rate was 5%. For control group, while the percentage of correct answers of all questions
increased, the percentage of incorrect answers of question-6 also increased slightly. In terms of the differences
between pre- and post-interviews, the experimental group students’ conceptual changes in sound were much
more positive than control group students.
As a result of content analysis, 11 categories were obtained: 1) propagation of sound, 2) reflection of sound,
3) echo formation, 4) technological applications of sound reflection, 5) sound propagation-solid environment,
6) loudness of sound, 7) absorption of sound, 8) Materials used in sound insulation and acoustic applications, 9)
encounter of sound with matter, 10) acoustic applications, 11) echo problems in closed spaces. There were 21
themes and 81 codes belonging to these categories. The themes and codes for each category are given in Table 6.
Each percentage value belongs to a group of 20 students.
Table 6
Generated Categories, Themes and Codes
pre post
Category Theme Codes Group
(%) (%)
C 45 60
through waves
E 45 100
C 20 25
through vibrations
E 15 10
C 10 15
How sound propagates through reflection
E 5 0
C 15 0
Linear
E 10 0
C 15 30
No idea
1. Propagation of sound
E 25 0
C 20 60
in all directions (from source)
E 25 85
Direction of propagation of C 15 10
one way (forward)
the sound E 15 0
C 65 30
No idea
E 60 15
Waves formed by drops of water dripping into a puddle/ C 5 20
throwing something into a puddle E 10 70
C 5 5
Spreading of siren of ambulance/ oxygen everywhere
E 0 20
Analogy
C 5 5
Sea waves
E 10 5
C 85 70
No idea
E 80 5
C 60 85
Sound is reflected.
E 55 100
C 10 5
Sound is/ is not reflected Sound is not reflected.
2. Reflection of sound
E 5 0
C 30 10
No idea
E 40 0
C 20 10
By hitting an obstacle/ item/ wall
E 30 45
C 10 10
How sound is reflected By hitting a hard and smooth obstacle
E 0 15
C 70 80
No idea
E 70 40
29
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
pre post
(%) (%)
C 35 45
When the sound strikes
E 75 80
C 15 10
As a result of sound reflection
E 0 60
How an echo occurs
When the sound hits a material at least 17 meters away and C 5 5
3. Echo formation
returns E 0 10
C 30 20
No idea
E 5 5
C 55 50
Echoes in an empty room/ a closed area/ a bathroom
E 75 80
C 20 30
Example Echoes in a cave/ mountain
E 15 15
C 25 20
No idea
E 10 5
C 15 55
Sonar
E 15 100
C 10 50
Ultrasound device
4. Technological applications of sound reflection
E 5 95
Technological devices and
equipment Telephone/ tablet/ voice recorder/ loudspeaker/ head- C 5 15
phones/ hailer/ microphone/ computer
E 10 0
C 70 45
No idea
E 70 0
C 10 50
Field of medicine
E 5 95
C 15 30
Marine
E 10 100
C 15 30
Areas of usage Fishing
E 0 100
C 5 10
Mining
E 10 35
C 70 0
No idea
E 75 0
C 40 50
5. Sound propagation- solid environment
Granular materials
E 20 100
C 20 5
Speak loudly
Necessary condition for the E 20 0
propagation of sound C 10 5
Nongranular materials/ vacuum/ space
E 5 0
C 30 20
No idea
E 65 0
C 10 50
Reason Transmission of sound by vibration of the airborne particles
E 20 100
C 5 15
Example the alarm clock inside the deflated bell jar
E 10 85
C 40 45
in front of the obstacle
E 45 100
The side with the higher loud- C 15 5
behind the obstacle
ness of sound E 10 0
6. Loudness of sound
C 45 25
No idea
E 45 0
C 30 45
The sound is absorbed to some extent by the obstacle.
E 20 80
In front of the obstacle- rea- C 25 45
The sound source is in front of the obstacle.
son E 45 10
C 45 10
The sound is reflected/ came back by hitting the obstacle.
E 35 10
30
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
pre post
(%) (%)
C 20 35
The sponge absorbs the sound more.
E 25 60
C 5 10
Sponge is soft, while paper is hard.
E 5 60
C 20 20
7. Absorption of sound
Sponge is a better insulating material than paper.

E 25 35
The reason why the sponge C 5 10
transmits sound less than Sponge is rough/ porous, whereas paper is smooth.
E 5 20
paper C 5 5
The spaces between the particles of the sponge are
slightly more than that in the paper. E 5 0
C 20 0
The sponge is thicker than paper.
E 15 0
C 35 20
No idea
E 25 0
C 20 35
rough
E 10 90
8. Materials used in sound insulation and
C 20 20
Properties of the materials Soft
E 5 95
used in sound insulation and
C 20 5
acoustic applications
acoustic applications Smooth

E 10 0
C 50 40
No idea
E 55 5
C 40 55
to absorb sound
E 45 95
C 0 15
Reason to prevent echo formation
E 5 90
C 25 10
to provide a more uniform sound propagation
E 15 0
C 50 65
Reflection/ echo occurs
E 35 100
C 15 40
If the material is rough, the sound is absorbed.
E 30 80
C 0 15
If the material is soft, the sound is absorbed.
9. Encounter of sound with matter
E 0 85
The result of sound-matter C 0 20
If the material is smooth, the sound is reflected/ echoed.
interaction E 0 45
C 0 5
Sound and matter collide.
E 10 5
C 15 5
Sound is transmitted.
E 20 0
C 70 40
No idea
E 60 10
Shape of the material/ angle of incidence of the sound C 85 80
The reason why the interac- waves E 85 25
tion has different conse-
C 15 20
quences Type/ properties of matter
E 15 75
31
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
pre post
(%) (%)
C 15 5
Rough ceiling of the theater hall
E 0 55
C 15 10
Soft ceiling of the theater hall
10. Acoustic applications
E 0 40
C 0 10
Domes of different sizes in mosques
Examples of acoustic ap- E 10 30
plications C 0 0
Stepped seating arrangement in theater and cinema
E 0 35
C 5 0
Use of reflective plates in concert areas and theaters
E 5 5
C 50 60
No idea
E 40 0
C 10 40
Use of sound-absorbing materials
E 10 85
11. Echo problems in closed spaces
C 10 40
Narrowing the area
E 25 5
C 20 35
Increasing the number of the belongings in the room/ place
Solutions for the echo E 30 5
problem C 10 0
Speaking in a low voice
E 5 0
C 5 0
Covering the place with aluminum foil
E 5 0
C 15 10
No idea
E 25 0
Category-1 emerged from data analysis of the first interview question. As can be seen in Table 6, according
to the findings, the majority of the students thought that the sound is propagated in waves. As a result of teach-
ing practices, the number of students explaining sound propagation with this code increased in both groups. As
a remarkable finding, it was observed that all the students in experimental group had the knowledge that the
sound propagated waves during the post-interviews. Moreover, in pre-interview, more than half of the students
did not know the direction of sound propagation. However, in post-interviews, 85% of students in experimental
group and 60% of students in control group had the knowledge that the sound was propagated in all directions.
The findings revealed that almost all of the students could not make an analogy about the propagation of sound
in pre-interviews. In post-interviews, it was seen that only 20% of students in control group and 70% of students
in experimental group resembled the propagation of sound to the image that occurred when water dropped, or
stones were thrown into the pond.
Category-2 emerged from data analysis of the second question. While 60% of students in control group
gave the correct answer “Sound is reflected” to the first theme in pre-interview, this ratio increased to 85% in post-
interview. In parallel, 55% of students in experimental group gave the same answer in pre-interview and all of
them in post-interview. On the other hand, when the codes regarding how the sound is reflected were examined,
it was seen that the majority of the students in control group were abstainers in both pre- and post-interviews. In
post-interview, 40% of students in experimental group were abstainers, whereas 45% of them stated that sound
was reflected when it encountered an obstacle, material or wall.
Category-3 emerged from data analysis of the third question. The students who answered the first theme cor-
rectly stated that “Echo occurs as a result of sound reflection.” In both pre- and post-interviews, most of the students
stated that echo was caused by strike of sound. In post-interview, 10% of students in control group and 60% of
students in experimental group stated that the echo was the result of sound reflection. In the second theme, ap-
proximately half of both group of students gave examples of echo formation in an empty room, in a closed space
or in the bathroom.
Category-4 emerged from data analysis of the fourth question. Almost none of the students could give an
example to the tools produced by using the reflection of sound in pre-interview. In post-interview, all of the stu-
32
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
dents in experimental group gave sonar and 95% of them gave ultrasound device as an example. On the other
hand, almost half of the students in control group were able to give these examples. S8, a student who answered
this question correctly said that “In a sonar or an ultrasound, sound waves strike an object and return. Thus, we es-
timate the existence, location and shape of the object.” In post-interview, all students in experimental group gave
marine and fishing, 95% of them gave field of medicine and 35% gave mining as examples. In the control group,
the number of students who gave these examples is less. S16, who answered fourth question partially correct,
stated that “We find lost ships or mines with the device that sends sound waves into the water.” S27, who gave wrong
answer expressed her thoughts as “The technologies produced by using the reflection of sound are electronic devices
like telephone, tablet and voice recorder.”
Category-5 emerged from data analysis of the fifth question. S12, who answered this interview question cor-
rectly, gave the answer “The most important condition for the propagation of sound is the solid environment. Because
sound is produced through vibration of the air particles. Therefore, sound does not travel in space.” All of the students
in experimental group and half of the students in control group associated the propagation of sound with the
solid environment in post-interview. One of the students exemplified this situation as “We can’t hear the alarm
clock inside the deflated bell jar.”
Category-6 emerged from data analysis of the sixth question. S7 answered this interview question correctly
and stated that “Loudness of sound is higher in front of the obstacle. Because when it encounters an obstacle, the sound
is absorbed and the loudness decreases.” Also, S9 said that “Loudness of sound is higher in front of the obstacle. Because
sound strikes and comes back; that is reflected.” In the pre-interview, only 40% of students in control group and 45%
of students in experimental group thought that the sound loudness was higher in front of the obstacle, while this
rate increased to 45% and 80% in control and experimental groups, respectively in post-interview.
Category-7 emerged from data analysis of the seventh question. The students who answered this question
correctly, expressed their thoughts as “Since sponge is soft and rough, it absorbs the sound.” and “Since paper is smooth,
it absorbs the sound less.” S3 and S37, the students who gave a partially correct answer, stated that “Since sponge
is soft and paper is hard, the sponge transmits the sound less than the paper.” and “Since the sponge is an insulating
material, it transmits sound less.” respectively. When findings obtained in post-interviews are examined, it is seen
that the students in experimental group have more accurate information than students in control group.
Category-8 emerged from data analysis of the eighth question. S7 and S11 who answered this question cor-
rectly expressed their thoughts as “The materials used in sound insulation and acoustic applications are soft so that
the sound is absorbed.” and “Hard and smooth materials should be used to ensure the reflection of sound in acoustic
applications.” In post-interview, 90% of students in experimental group thought that the materials used should be
rough to ensure sound insulation, whereas only 35% of students in control group thought this. In addition, in the
post-interview, 95% of students in experimental group and 55% of the students in control group stated that it was
aimed to absorb sound in sound insulation.
Category-9 emerged from data analysis of the ninth question. The students who answered this interview ques-
tion correctly gave the answers like “Sound is absorbed, and loudness of sound is decreased when sound encounters a
soft material.” and “Sound is reflected, and echo may occur when sound encounters a smooth material.” The students
who gave a partially correct answer mostly stated that “Sound is transmitted or propagated when it encounters mat-
ter.” The students who answered this question wrong expressed their thoughts as ‘Echo occurs when the sound
encounters the matter.’ and “These events occur at different rates due to the shape or size of the material.” In the post-
interviews, 75% of the students in experimental group and 20% of the students in control group gave the correct
answer that the different results of this interaction stem from properties of the material.
Category-10 emerged from data analysis of the tenth question. S8 and S31, the students who answered this
question correctly stated that “With a stepped seating arrangement, the sound is heard better by echoing from stage
to the seats on hard and smooth surfaces.” and “The sound enters into 64 cubes in the dome of Süleymaniye Mosque
and is reflected equally and returns to us.” respectively. S5 gave a partially correct answer and said that “Sound in-
sulation prevents sound from going out”. S12 who gave the wrong answer expressed his thoughts as “If we cover
all the walls in the house with wallpaper, the wall reflects sound less since it became smooth.” In pre-interviews, the
students in experimental group gave examples of acoustic applications as rough (55%) and soft (40%) ceiling of
theater hall, domes of different sizes in mosques (30%), and use of reflective plates in concert areas and theaters
(5%). On the other hand, the students in control group gave examples of acoustic applications as soft (10%) and
rough (5%) ceiling of theater hall, and domes of different sizes in mosques (10%) in post-interviews. Additionally,
while almost half of the students in both groups could not give examples of acoustic applications in pre-interview;
33
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
60% of students in control group could not give any examples of acoustic applications in post-interview. In the
experimental group, no one left this question unanswered.
Category-11 emerged from data analysis of the last question. S37 answered this question correctly and stated
that “The use of soft and rough materials in closed area prevents sound transmission and causes sound absorption. Thus,
no echo occurs.” The students who gave completely correct answers consisted of 85% of experimental group and
40% of control group in post-interview. S13 gave a partially correct answer and expressed her thoughts as “If sound
insulation is made in a closed area, the echo does not occur.” S2 and S28 who gave wrong answers to this question
stated that “Echo formation is prevented by using aluminum foil indoors.” and “Echo does not occur if you speak in a
low voice in a closed area.” respectively.
Discussion
The results of the research provided an evidence on how engineering design process-oriented STEM activities
enhanced the students’ conceptual understanding levels on the physics of sound. As a result of content analysis, 11
categories including 21 themes and 81 codes were obtained in the research. It was seen that the conceptual under-
standing level of students in both experimental and control groups had been increased for each category. However,
when the findings were analyzed, it was seen that the conceptual understanding of the students in experimental
group has changed much more positively than the control group students. Accordingly, the students’ learning
about the topic of sound was more meaningful when they were able to interact with materials and make designs.
Wendell (2011) examined whether the 6-hour teaching activities developed in accordance with engineering
design method for the production, transmission and characteristics of sound caused the change in 3rd and 4th grade
students’ ideas about sound. Some conclusions were reached through the content analysis of the data obtained
from the interviews with the students. Students’ ideas about sound production were divided into six categories
in pre- and post-interviews. These categories were; additional physical evidence of process, vibration as process,
movement as cause, static characteristics, human action, no answer or invented component. Students’ ideas about
sound transmission were divided into six categories in the pre- and post-interviews. These categories were: Sound
transmitted via propagation of vibration, interactive role played by medium or matter, sound transmitted via action,
sound transmitted as substance, only mentions medium or matter, no explanation of transmission. In Wendell’s
(2011) research, after the activities using engineering design method, students’ thoughts about both production
and propagation of the sound increased in terms of percentage of correct answers. In parallel with these findings,
according to the results of the current research, when the differences between the pre- and post-interviews were
examined, it was seen that conceptual understanding levels of the students in experimental group were more posi-
tive than the students in control group. Different from the research of Wendell (2011), current research included
the activities that integrated the engineering design process with STEM, rather than just engineering design
method, and also the number of activities and much more. Lastly, in the current research, for most of the themes,
conceptual understanding level of the students in experimental group were increased much more than it was in
the research of Wendell (2011). This result is meaningful since individuals have the ability to think in abstract ways
after the age of 11 and the current research was conducted with 6th grade students with an average age of 12.
In the research of Gulhan and Sahin (2016), six engineering design process-oriented STEM activities regarding
the topics of light, living beings and electricity were carried out in the experimental group during 12 weeks. The ex-
perimental group activities were carried out in groups, and the engineering design process steps that were followed
were very similar to those developed by Hynes et al. (2011). The researchers concluded that STEM activities improved
5th grade students’ conceptual understanding. Many other research studies that concluded engineering design
process-oriented STEM activities positively affected students’ success (Doppelt et al., 2008; Ercan & Sahin, 2015; Gul-
han & Sahin, 2016; Wendell & Rogers, 2013). In these studies, written achievement tests consisting of multiple choice
and/ or open-ended questions were preferred as an instrument for determining the student achievement. On the
other hand, in the current research, open-ended questions were used verbally to examine the students’ conceptual
understanding in more detail. It was seen that the conceptual understanding of the students in experimental group
has changed much more positively than the control group students. This situation is parallel to the findings obtained
in the above-mentioned studies and shows that STEM activities are highly effective in increasing the success levels of
the 6th grade students. The estimated reasons for this situation were discussed separately below.
Owing to the designs they made, the students in experimental group reached at least one scientific idea
about sound at the end of each course. In the last lesson, the students developed a project by synthesizing all
34
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
the scientific knowledge they had gained on the subject thus, they reinforced their learning. It is thought that the
students’ testing the prototypes they developed and improving their prototypes in line with the results they had,
have a positive effect on their conceptual understanding. Gerlach (2012) stated that students need more than one
chance to be successful and engineering design process-oriented science activities give them that chance (Ercan
& Sahin, 2015).
It is known that the motivation of students working in collaborative groups increases and their success
increases accordingly (Sung & Hwang, 2013). Considering that control group students sometimes work individu-
ally and sometimes in groups during the learning process while experimental group students always make their
designs in groups to solve the problem, it is possible to conclude working in groups was effective in increasing
the level of conceptual understanding.
The students in experimental group learned the subject by conducting various experiments and solving the
problems in order to make the best design in the engineering design process. In this research, it is thought that
one of the reasons why the students in experimental group were more successful is that most of their time they
actively engaged in the experiments and designs in STEM activities since it is known that STEM promotes active
learning and collaboration. On the other hand, in control group activities, a large part of the time was used to
answer the questions in the workbook. Therefore, students in the control group were more passive throughout
the process. According to Dale’s (1969) Cone of Experience, the greater the number of sense organs used in the
learning process, the greater the learning and the more information is likely to be retained. Accordingly, in the
concretization of abstract concepts, the information obtained through the ways in which student is passive such as
reading, listening and watching remain insufficient, this information can only be concretized through experience.
In this sense, the fact that the students in the experimental group are more active in the learning process is an
important factor in their conceptual development. In the research, the results revealed in detail that the success
of the students who participated in integrated STEM activities is not a chance or coincidence.
Conclusions
The advantages of the use of STEM activities in learning different science topics are well known, but this
research focused on teaching physics of sound to the 6th grade students through engineering design process-
oriented integrated STEM activities, making it specific.
In accordance with the rapidly changing world, it is important to educate the new generation equipped with
21st-century skills and engineering and design skills to play an active role in scientific and economic developments.
In parallel with this, keeping up with the technology age is among the objectives of the science curriculum. In
order to overcome students’ lack of motivation for learning the topics of physics, which they perceive as difficult,
appropriate teaching approaches should be used. Engineering design process-oriented STEM is a teaching ap-
proach that leads to an increase in students’ motivation and achievement level in science.
Implications
This research was an example of STEM education by using engineering design-process. Engineering and design
skills have gained importance in the new science curriculum which aims to integrate science with mathematics,
technology and engineering. Therefore, it is suggested that teachers and researchers should apply engineering
design process-oriented STEM activities and examine the results of these applications according to many other
topics in the science curriculum. In this sense, it is thought that experimental activities developed within the scope
of this research will contribute to science education in terms of developing and using the integrated STEM activities.
In 2018 science curriculum the statement of ‘Education is given not only for knowing (thinking) but also for
feeling (emotion) and doing (action)’ is included. In this context, research studies should be conducted on the ef-
fects of STEM activities on affective characteristics of the students such as attitude, motivation and anxiety.
Limitations
In the current research, only one topic, physics of sound, was discussed. In addition to this, the research is
limited to the 6th grade students’ conceptual understanding. Lastly, the research was conducted in already-formed
classes rather than in randomly chosen groups.
35
OF SOUND
(pp. 21-37) ISSN 2538–7138 /Online/
References
Dedeturk, A., Saylan Kirmizigul, A., & Kaya, H. (2020). The effect of STEM activities on students’ achievement in “Sound” subject.
Pamukkale University Journal of Education, 49, 134-161.
Bahar, M., & Polat, M. (2007). The science topics perceived difficult by pupils at primary 6-8 classes: Diagnosing the problems
and remedy suggestions. Educational Sciences: Theory and Practice, 7(3), 1113-1129.
Barnett, G. M., Connolly, K., Jarvin, L., Rogers, C., Wendell, K. B. & Wright, C. (2008). Science through LEGO Engineering Curriculum
Units: Design a musical instrument: The Science of Sound. Curriculum Resources 2008-2009 Edition, Tufts University Center
for Engineering.
Bybee, R. W. (2010). Advancing STEM education: A 2020 vision. Technology and Engineering Teacher, 70(1), 30-35.
Dale, E. (1969). Audio-visual methods in teaching (3rd Ed). Holt, Rinehart & Winston.
Doppelt, Y., Mehalik, M. M., Schunn, C. D., Silk, E., & Krysinski, D. (2008). Engagement and achievements: A case study of design-
based learning in a science context. Journal of Technology Education, 19(2), 22-39.
Ercan, S., & Sahin, F. (2015). The usage of engineering practices in science education: Effects of design- based science learning
on students’ academic achievement. Necatibey Faculty of Education Electronic Journal of Science and Mathematics Education,
9(1), 128-164.
Eshach, H. (2014). Development of a student-centered instrument to assess middle school students’ conceptual understanding of
sound. Physical Review Special Topics-Physics Education Research, 10(1), 1-14. https://doi.org/10.1103/PhysRevSTPER.10.010102
Fraenkel, J. R., Wallen, N. E., & Hyun, H. (2012). How to design and evaluate research in education. McGraw Hill.
Gagić, Z. Z., Skuban, S. J., Radulović, B. N., Stojanović, M. M., & Gajić, O. (2019). The implementation of mind maps in
teaching physics: Educational efficiency and students’ involvement. Journal of Baltic Science Education, 18(1), 117-131.
https://dx.doi.org/10.33225/jbse/19.18.117
Gulhan, F. & Şahin, F. (2016). Fen-teknoloji-mühendislik-matematik entegrasyonunun (STEM) 5. sınıf öğrencilerinin kavramsal
anlamalarına ve mesleklerle ilgili görüşlerine etkisi [The effect of science-technology-engineering-mathematics
integration (STEM) on 5th grade students’ conceptual understanding and views on professions]. Pegem Yayıncılık, 283-302.
http://dx.doi.org/10.14527/9786053183563b2.019
Harwell, M., Moreno, M., Phillips, A., Guzey, S. S., Moore, T. J., & Roehrig, G. H. (2015). A study of STEM assessments in engineering,
science, and mathematics for elementary and middle school students. School Science and Mathematics, 115(2), 66-74.
https://doi.org/10.1111/ssm.12105
Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277-1288.
Hynes, M., Portsmore, M., Dare, E., Milto, E., Rogers, C., Hammer, D., & Carberry, A. (2011). Infusing engineering design into high
school STEM courses. National Center for Engineering and Technology Education. http://ncete.org
Jenkins, E. W. (Ed.). (2002). Innovations in science and technology education. The UNESCO Press.
Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159-174.
Martin, M. O., Mullis, I. V. S., Pierre, F., & Hooper, M. (2015). TIMSS 2015 International results in science. TIMSS & PIRLS International
Study Center. http://timssandpirls.bc.edu/timss2015/international-results
Ministry of National Education of Turkey [MoNE]. (2013). Science curriculum (3rd-8th grades). Ankara: Board of Education.
Ministry of National Education of Turkey [MoNE]. (2018). Science curriculum (3rd-8th grades). Ankara: Board of Education.
Moore, T. J., Johnson, C. C., Peters Burton, E. E., & Guzey, S. S. (2015). The need for a STEM road map. In C. C. Johnson, E. E. Peters
Burton, & T. J. Moore (Eds), STEM road map: A framework for integrated STEM education. Routledge.
Organisation for Economic Co-operation and Development [OECD]. (2019). Education at a glance 2019: OECD indicators. OECD
Publishing.
Organisation for Economic Cooperation and Development [OECD]. (2019). PISA 2018 results: What students know and can do
– Volume I. OECD Publishing.
Savran Gencer, A., Dogan, H., Bilen, K., & Can, B. (2019). Integrated STEM education models. Pamukkale University Journal of
Education, 45, 38-55.
Sozen, M. & Bolat, M. (2014). Determining the misconceptions of 11–18 age of students about sound velocity concept. Ondokuz
Mayis University Journal of Faculty of Education, 33(2), 505-523.
Sung, H. Y. & Hwang, G. J. (2013). A collaborative game-based learning approach to improving students’ learning performance
in science courses. Computers & Education, 63, 43-51.
Timur, B., Timur, S., Ozdemir, M., & Sen, C. (2016). Challenges encountered in the teaching of units in the primary science
course curriculum solution and recommendations. Journal of Theory and Practice in Education, 12(2), 389-402.
http://eku.comu.edu.tr/article/view/5000181692/5000161782
Topcu, M. S., & Gokce, A. (2018). STEM ve mühendislik [STEM and engineering]. Kırkıç, K. A., & Aydın, E. Merhaba STEM: Yenilikçi Bir
Öğretim Yaklaşımı [Hello STEM: An Innovative Teaching Approach] (p. 79-94). Eğitim Yayınevi.
Vasquez, J. A., Sneider, C., & Comer, M. (2013). STEM lesson essentials, grades 3-8: Integrating science, technology, engineering, and
mathematics. Heinemann.
We n d e l l, K . B. ( 2 0 1 1 ) . Science through engineering in elementar y school: Comparing three enactments
of an engineering-design-based curriculum on the science of sound (Doctoral dissertation). Tufts University.
http://www.proquest.com/products-services/dissertations
Wendell, K. B., & Lee, H. S. (2010). Elementary students’learning of materials science practices through instruction based on engineering
design tasks. Journal of Science Education and Technology, 19(6), 580-601. https://doi.org/10.1007/s10956-010-9225-8
36
OF SOUND
ISSN 2538–7138 /Online/ (pp. 21-37)
Wendell, K. B., Connolly, K., Wright, C., Jarvin, L., Rogers, C., Barnett, M., & Marulcu, İ. (2010). Incorporating engineering design into
elementary school science curricula [Paper presentation]. American Society for Engineering Education Annual Conference
& Exposition, Louisville.
Wendell, K. B., & Rogers, C. (2013). Engineering design-based science, science content performance, and science attitudes in
elementary school. Journal of Engineering Education, 102(4), 513-540. https://doi.org/10.1002/jee.20026
Wicklein, R. C. (2006). Five good reasons for engineering as the focus for technology education. The Technology Teacher, 65(7), 25-29.
Yerdelen, S., & Sungur, S. (2020). Pre-service science teachers’ conceptions of sound: The role of task Value beliefs. Science Education
International, 31(3), 295-303. https://eric.ed.gov/?id=EJ1268286
Received: June 04, 2020 Accepted: January 04, 2021
Cite as: Dedeturk, A., Saylan Kirmizigul, A., & Kaya, H. (2021). The effects of STEM activities on 6th grade students’ conceptual
development of sound. Journal of Baltic Science Education, 20(1), 21-37. https://doi.org/10.33225/jbse/21.20.021
Ayşegül Dedetürk MS, Institute of Educational Sciences, Erciyes University,

Yenidoğan, Turhan Baytop Sokak No:1, 38280 Talas/Kayseri,
Turkey.
Aslı Saylan Kırmızıgül PhD, Research Assistant, Department of Mathematics and Science
(Corresponding author) Education, Faculty of Education, Erciyes University, Kayseri,
Turkey.
Website: https://avesis.erciyes.edu.tr/aslisaylan
Hasan Kaya PhD, Professor, Faculty of Education, Department of Mathematics
and Science Education, Faculty of Education, Erciyes University,
Kayseri, Turkey.
37
DEVELOPING A SELF-
EFFICACY SCALE TOWARD
PHYSICS SUBJECTS FOR
ISSN 2538-7138 /Online/ LOWER-SECONDARY SCHOOL
STUDENTS
Abstract. Physics subjects or topics as a Mustafa Fidan,

part of science education enable students Meric Tuncel
to explore nature and natural events by
using physics laws and scientific methods.
In the lower-secondary school, physics is
a subfield within the science curriculum
developed holistically with the broad
field curriculum design unlike the subject-
Introduction
centred curriculum in the upper-secondary
education. As an important factor that
The term of self-efficacy is theoretically based on Social Learning Theory
directs the achievement, self-efficacy is
(Bandura, 1977). However, being formed in the light of cognitive processes,
one of the predecessors of these cognitive
self-efficacy addresses to Social Cognitive Learning Theory in later times (Sc-
and affective skills. Accordingly, most of
hunk & Parajes, 2009). Bandura (1997) stated self-efficacy as “beliefs in one’s
the scales which evaluate the students’
capabilities to organize and execute the course of action required to produce
cognitive or affective skills in the lower-sec-
given attainments”. According to Zimmerman (1995), self-efficacy includes
ondary school are related to science course
an individual’s judgements about performing a task or an assignment rather
in a general context. This research aimed to
than the physical or psychological characteristics owned by that individual. In
develop a valid and reliable scale measur-
other words, self-efficacy is not an observable skill that the individual owns.
ing the students’ self-efficacy beliefs toward
It is the individual’s personal beliefs, expectations, or judgements about the
physics subjects for lower-secondary
answers that he or she gives to the question “What can I do?” in relation to a
schools. The sample consists of 2737 stu-
specific field, task, or situation (Bandura, 1997; Klassen & Usher, 2010).
dents. The draft scale consisted of 52 items
Self-efficacy belief is related to an individual’s ability to discover and
and was applied to 1882 students in the
manage his or her own abilities. As a specific domain-oriented term rather
first stage. To construct validity of the scale,
than a global and stable structure, self-efficacy belief occurs with direct or
exploratory factor analysis was performed.
indirect experiences in a certain process (Bandura, 1997). Bandura (1997) has
The results showed that the scale consisted
underlined the four principal sources of self-efficacy as follows: (a) mastery
of single- factor with 28 items. To confirm
experiences (interpretations of actual performances, past accomplishments,
its factor structure, the scale was applied to
or failures etc.) (b) vicarious experiences (observation of others’ positive or
785 students. The research findings indicate
negative situations) (c) verbal persuasion (supportive messages, encourage-
that the scale is a valid and reliable instru-
ment of teachers etc.) (d) physiological and emotional situations (emotions,
ment to measure the self-efficacy beliefs
stress, happiness, concern etc.). The findings of some studies highlighted that
towards physics subjects.
these factors were power predictors of self-efficacy (Usher & Parajes, 2009).
Keywords: lower-secondary school, As a determinant of behaviour changes, self-efficacy belief also affects the
physics subjects, scale development, self- task choice, effort, persistence in achieving resilience to difficult situations
efficacy belief (Bandura, 1997).
In educational studies, self-efficacy is one of the critical variables in ex-
plaining the learning performance (Richardson et al., 2012; Schunk & Pajares,
Mustafa Fidan 2009). Accordingly, it has gained attraction from educators and researchers
Bartin University, Turkey over the past decades (Britner & Pajares, 2006; Usher & Pajares, 2009). Self-
Meric Tuncel
Bolu Abant Izzet Baysal University, Turkey efficacy has an important role on cognitive and affective outcomes such as
learning achievement, motivation, attitude, self-regulation, and learning goals
(Schunk & Pajares, 2009; Schunk & Usher, 2011). Actually, students with high
38
ISSN 1648–3898 /Print/ DEVELOPING A SELF-EFFICACY SCALE TOWARD PHYSICS SUBJECTS FOR LOWER-
ISSN 2538–7138 /Online/ (pp. 38-49)
self-efficacy make more efforts when they face with difficulties, they are eager to learn and they tend to continue
learning (Bandura, 1997). Conversely, there is considerable evidence that students with poor self-efficacy beliefs
have a low level of course achievement and ability to deal with difficulties and that their desire and tendency to
perform a task is low (Schunk, 1990). Therefore, students with low self-efficacy are less motivated to learn and do
not make much effort to perform the assigned tasks (Bandura, 1997).
On the other hand, it is still under discussion among some researchers whether the variable of self-efficacy
is peculiar to a general field or a more specific subject (Aypay, 2010). However, most of the researchers have em-
phasized that the concept of self-efficacy should be treated as being oriented to any assignments or tasks that
are peculiar to a certain field (Bandura, 1997; Klassen & Usher, 2010). This means that it is more like a discipline,
domain-specific or task-based content rather than a general construct. Therefore, students may have self-efficacy
towards sub-disciplines such as geometry in mathematics course, grammar in language course or tasks and more
specialized subjects like astronomy, anatomy in the field of science. In this context, the current study focused on
the self-efficacy beliefs towards physics subjects of science course at lower-secondary schools.
As a part of science education, physics in a broad sense is concerned about the matters such as the history
of science, movements and behaviours of nature, and earth and space sciences. It underlies other disciplines such
as biology, astronomy, geology, and chemistry and directs to new inventions or technologies (MoNE, 2018). While
students generally learn these subjects or concepts through physics curriculum in the secondary school, they learn
them within context of science course at the lower-secondary shool level. Moreover, partly because of its abstract
content or complexity, physics is critical for students to shape their knowledge, skills, and attitudes about science in
the lower-secondary schools (Tuncel & Fidan, 2018). In the literature, the research on physics teaching-learning as
a subfield of science education has started to increase over the past decades. Especially, the researchers have been
interested in cognitive and affective features that predict the learning achievement in physics education (Kapucu,
2017; Tezer & Asiksoy, 2015). Self-efficacy belief can be described as the individual’s belief in overcoming the tasks,
difficulties or situations related to the field of science education by evaluating his or her own abilities (OECD, 2017).
Indeed, students with high self-efficacy beliefs in physics are willing to take risks and perform difficult tasks.
Besides, they make more efforts to perform a behaviour related to learning and have higher levels of motivation
and self-confidence (Tezer & Asiksoy, 2015). Therefore, the high level of self-efficacy beliefs held by the learners in
physics may be an effective factor in the process of finding solutions to the problems they may face in the course.
In contrast, the low level of self-efficacy belief in physics negatively affects motivation, learning behaviour, and
expectations for future and performance (Lin et al., 2015).
Self-Efficacy Instruments in Physics
Recent studies have shown that self-efficacy regarding physics course is significantly correlated to students’
learning achievement and motivation (Alpaslan & Isik, 2016; Kapucu, 2017). This means that the perception of
self-efficacy in physics increases their learning performances and stimulates their actions. Considering the general
objectives of physics curriculum (MoNE, 2013), self-efficacy belief is an important psychological construct about
whether students can learn subjects and concepts in physics course or not. Also, it is included in the scope of sci-
ence literacy as well as the attitudes, beliefs, values, and motivational orientations (OECD, 2017).
There are several measurement tools developed for measuring the self-efficacy belief in physics. For instance,
Lin et al. (2015) attempted to develop the scale for physics self-efficacy belief of university students in a multi-
dimensional sense. They adapted the self-efficacy scale which was developed for science learning by Lin and Tsai
(2013) to physics learning. Its factors were as follows: conceptual understanding (structuring the physics concepts,
principles etc.), higher-order cognitive skills (problem solving, critical thinking, or scientific inquiry regarding
physics), practical work (performing laboratory activities regarding physics), everyday application (practicing
the physics concepts regarding daily life), science communication (discussing the physics subjects with others or
peers). Similarly, Alpaslan and Isik (2016) also adapted this scale developed by Lin and Tsai (2013) into Turkish to
assess the physics self-efficacy of university students in Turkey. Selcuk et al. (2018) developed physics self-efficacy
scale including two factors with 21 items for high school level. The first factor, self-efficacy belief towards physics
achievement, is associated with problem solving and remembering necessary formulas in physics lesson while the
second factor, self-efficacy belief towards the skill of using physics knowledge, is related to the ability to transfer
physics concepts or subjects into daily life. However, these scales were developed for grades from upper-secondary
school to higher education.
39
(pp. 38-49) ISSN 2538–7138 /Online/
On the other hand, the researchers integrally developed several instruments for measuring self-efficacy belief
in science education including physics in the lower-secondary school level (Aslam & Ali, 2017; Lin & Tsai, 2013; Tatar
et al., 2009). Some researchers have also used the self-efficacy as a sub-dimension of motivation on their scales
(Glynn et al., 2009; Tuan et al., 2005). These instruments focus on a general perspective oriented to the overall
course, field or discipline unlike specific subjects, themes, or tasks in the context of science discipline. In addition,
in the literature, most of the research on scale development has focused on high or undergraduate school level
rather than primary or middle school level. Specifically, the field of science education lacks an adequate instru-
ment that enables understanding students’ self-efficacy beliefs towards physics subjects. In recent years, the scale
development studies regarding the specific domain, theme or task features have begun to draw attention of sci-
ence researchers. For instance, Ilhan and Cicek (2017) have developed a self-efficacy scale for acid-base subjects
by focusing on a more specific subject in the field of science and examined students’ perceptions of self-efficacy in
chemistry. At the end of the research process, they approached the self-efficacy belief in two aspects as associating
it with daily life and as making scientific explanations about the subject. Similarly, in his research on prospective
teachers, Demirci (2017) has developed a self-efficacy belief scale for teaching astronomy subjects and discussed it
in three sub-dimensions as “student outcomes through astronomy teaching”, “astronomy teaching strategies”, and
“difficulty in teaching astronomy subjects”. Alternatively, there is a need to develop specific measurement tools to
make educationally valid and reliable inferences about the concept of self-efficacy that has such a critical role in
learning. Actually, these and similar measurement tools can provide important evidence for the factors affecting
the science achievements and affective characteristics of the students who have difficulty in learning.
Importantly, the impact of self-efficacy beliefs on students’ abilities to deal with the difficulties that they
experience in their learning of physics subjects and to resist to such situations is significant. Indeed, since the stu-
dents with high self-efficacy towards the field of physics are more aware of their own abilities, they can overcome
difficult, complex, and abstract issues more easily and their motivation for the course is also high (Yelgun, 2009). At
this point, examining the concept of self-efficacy with a more specific approach in such aspects as subject, theme,
sub-discipline, and task may contribute to making important inferences about the whole of science course. Even
though the students have high self-efficacy levels towards the sub-disciplines or a particular subject of science,
their general self-efficacy level towards the whole field of science may be low. For instance, a student with a high
self-efficacy about biology, chemistry or a specific subject can have low self-efficacy belief in science in the general
sense only because physics subjects are difficult, and they have negative tendencies towards physics. Moreover,
self-efficacy towards physics subjects not only may reveal an interrelation view with other factors affecting the
learning, but also may take on a mediation role between science learning and the conceptions of science.
In the light of all this information, the lower-secondary school science curriculum (learning areas, units) was
taken into consideration in the present research, and the focus was set on self-efficacy belief towards school physics
subjects in lower-secondary schools. In order to address these gaps in the field of science, this research focused on
the development of a valid and reliable scale to measure the self-efficacy belief level towards physics subjects in
science education. The scale may be a valuable tool for researchers or educators to assess the self-efficacy beliefs of
students at lower-secondary school level. As being one of the pioneer studies, this research may allow for educational
implications or directions regarding self-efficacy as a crucial predictor of learning achievement and motivation
in terms of science course. Therefore, the present research aimed to develop a valid and reliable scale measuring
the self-efficacy beliefs towards physics subjects within the context of science course for lower-secondary schools.
General Background
This scale development research was conducted by using the scanning model in quantitative methodology.
The scanning model refers to “a research approach describing an existing or a previous situation as it is” (Karasar,
2012, p. 79). Data is generally collected from a large population in this model. Ethical approval for this research
was obtained from Ethics Committee of the Bolu Abant Izzet Baysal University in Turkey. The official permission to
perform the research was also taken from the Directorate of National Education. This section outlines the detailed
information about the participants, procedures, data collection, and data analysis. The scale development process
proposed by Anderson (1988) was taken into consideration in this research.
40
ISSN 2538–7138 /Online/ (pp. 38-49)
Sampling
A total of 2737 students (aged from 11 to 16, M = 13.04, SD = 1.31) participated voluntarily in all phases of
the research process from the lower-secondary schools in Turkey. They are randomly selected. In the first stage,
the primary data were collected from 1918 students through Exploratory Factor Analysis (EFA) to determine the
construct validity of the draft scale. Data from 36 students were excluded from the scale because of their incor-
rect or missing responses. Accordingly, the pilot sample was composed of 1882 students (902 males, 980 females)
attending 6th (n = 599, 32.13%), 7th (n = 618, 32.84%), and 8th (n = 665, 35.33%) grades. MacCallum et al. (1996)
reported that a larger sample increases the power of the model being examined and contributes to more reliable
results regarding the measurement.
In the second stage, the secondary data were collected from 785 lower-secondary school students through
Confirmatory Factor Analysis (CFA) to test the factor structure of the scale on a different sample. Importantly, it is a
more appropriate approach that the sample for CFA may be different from EFA. Because of the fact that 21 students
responded to the scale items incorrectly, the data obtained from these students were excluded. Accordingly, the
sample two was composed of 764 students (366 males, 398 females) attending 6th (n = 204, 26.7%), 7th (n = 254,
33.2%), and 8th (n = 306, 40.1%) grades. In the final stage, 91 lower-secondary school students (41 females, 50
males) participated in the research of scale application on an independent sample to determine the test-retest
reliability and internal consistency. The participants in all stages were independent of each other.
Instrument and Procedures
1. Item development and preparation of the item pool
In the preliminary stage, the instruments were examined on self-efficacy in science education in detail along
with an extensive review of the literature on self-efficacy theories. The items related to self-efficacy towards the
concepts or units of physics particularly focus on the subject area of “physical events” included for lower-secondary
schools in the science curriculum. Firstly, an initial item pool was created with 52 items. Importantly, Bandura’s
(2006) criteria for developing the self-efficacy instrument were also taken into consideration to create the items.
For example, the items included the phrase “can do” instead of “will do” to carry a positive connotation rather than
negative.
2. Taking expert opinions
Prior to the pilot research of the scale, 14 experts with PhD degrees from “science education” (n = 6), “cur-
riculum and instruction” (n = 5), and “assessment and evaluation” (n = 3) departments at universities commented
on the scale items to ensure content (scope) and face validity. They judged the relevance of each item to the
theoretical framework of self-efficacy. In addition, physics (n = 3), science (n = 4), and Turkish language (n = 2)
teachers’ views were also applied with regard to the clarity of items. An assessment form, rated on a 3-point scale
(“essential”, “useful, but not essential”, “not necessary”) for each item, was sent to these teachers via e-mail. In line
with their feedback, the content validity ratio (CVR) of items was calculated as proposed by Lawshe (1975). This
value for each item was evaluated by taking .57 as a reference for the expert sample size (n = 14) according to the
CVR critical table prepared by Ayre and Scally (2014). The minimum expert size is supposed to be 5 to measure
the values of CVR (Ayre & Scally, 2014).
3. Item reduction and revision
A total of 12 items were excluded from the scale and some items were revised for trial application in terms
of conformity to the scale, expediency, spelling, and writing. Moreover, the criteria proposed by Oluwatayo (2012)
were taken into account to ensure face validity in the research. The experts evaluated the scale in the light of these
criteria (the clarity and unambiguity of items, the appropriateness of item difficulty level for the respondents, the
correct spelling of difficult words etc.). Although face validity is considered within the scope of content validity, it
is initially referenced as a validity type in the process of scale development.
41
(pp. 38-49) ISSN 2538–7138 /Online/
4. Draft scale for pilot research
The pilot draft scale was ultimately composed of 40 items, on a 5-point Likert scale from “Strongly Disagree”
(1) to “Strongly Agree” (5). There were no negative items in the scale. Before the pilot research, 12 students also
read the scale items in terms of comprehensibility. Their feedback indicated that the items of the scale were not
difficult to understand for pilot research.
5. Data collecting
The data were collected during the fall semester of 2017-2018 academic year. Prior to the application of the
scale, the students were informed about the purpose of this research. The students completed the scales in groups
at certain periods in the classroom setting. To complete the scale, the process lasted approximately 10-15 minutes.
Data Analysis
In this research, both EFA and CFA were conducted to reveal the construct validity of the scale. Prior to these
analyses, missing and outlier data, the normality of the data set, the number of the sample and its relevance were
controlled for these analyses. The normality of the data distribution was checked by using Kolmogorov-Smirnov test
(p > .05). The histogram graphics, Q-Q plots, and skewness and kurtosis values (+2, -2) were also examined (Kline,
2016). In addition, there is no consensus on the number of samples in the factor analysis studies. The sample size
should be at least 5-10 times higher than the number of items for scale development and contain at least 300 cases
(Tabachnick & Fidell, 2013). According to Kline (2016), although a sample of 200 people is enough for reliable find-
ings in factor analysis, still it may be more suitable to research in larger samples. Comrey and Lee (1992) expressed
that 100 as poor sample size, 300 as good and 1000 as excellent. In this context, the sample size is quite enough for
the reliability of the present research. After the items were tested for the multicollinearity problem, in other words
for the high-level correlation (r > .90) of item pairs, it was seen that there was not such a problem for the analysis.
In the first stage, the data obtained from the pilot application to determine the factor structure of self-
efficacy belief variable were analysed with EFA by using SPSS 21.0 software. The new latent factors are discovered
meaningfully by clustering the variables associated with each other through the EFA (Tabachnick & Fidell, 2013).
Importantly, Kline (2016) emphasized that the construct validity is an umbrella term that includes the other types
of validity. During the EFA, the principal component analysis was employed in this research. For the item analysis,
independent samples t test was also used to determine whether there is a significant discrimination between the
scores of lower and upper groups (27 % in each group). Next, item-total correlation coefficient (r) was calculated
Figure 1
The Process of the Scale Development in this Research
42
ISSN 2538–7138 /Online/ (pp. 38-49)
to determine item’s discrimination feature in this research. It means the relationship between each item and the
total scores of a scale (Buyukozturk, 2018).
In the second stage, CFA was performed on independent sample by using AMOS 22.0 software to confirm the
factor structure of the scale. Based on EFA results, CFA is mainly used to compare the models of the scale’s factor
structure, to test the scale’s model-data fit, and to decide the model with the best fit (Kline, 2016; Tabachnick & Fidell,
2013). Accordingly, Tucker Lewis Index (TLI ≥ .90), Comparative Fit Index (CFI ≥ .90), Normed-Fit Index (NFI ≥ .90),
Goodness Fit Index (GFI ≥ .90), (IFI ≥ .90), Root Mean Square Error of Approximation (RMSEA ≤ .08), and Standardized
Root Mean Square Residual (SRMR ≤ .08) values were assessed for data fit (Kline, 2016). Given the sensitivity to the
larger sample size, the chi-square was not taken into consideration in interpreting the analysis findings (Satorra &
Bentler, 2001). Instead of this value, chi-square/ degree of freedom (χ²/df) ratio was given for information purposes.
In the third stage, the internal consistency of Cronbach’s alpha coefficient (α) and Pearson correlation coefficient
(r) for re-test reliability were calculated for determining the reliability of the scale by using SPSS 21.0 software. The
respondents with higher scores have more belief of self-efficacy to perform the task in physics. After the analysis
of the data, we prepared the final version of the scale. Figure 1 shows the stages of the scale development in this
research.
Research Results
First Stage (Results of the Exploratory Factor Analysis)
Prior to performing the Exploratory Factor Analysis (EFA), Kaiser-Meyer-Olkin (KMO) was used to test the ad-
equacy of sample and Bartlett’s test of sphericity was applied to check the suitability of the data. According to the
analysis results, KMO index was .95 (KMO >.70) and Bartlett’s test of sphericity statistic for the correlation matrix
was significant (x2 = 24471.75; SD = 378; p < .05). These preliminary findings indicated that the data were suitable
for EFA (Tabachnick & Fidell, 2013).
The data set for 1882 cases was analysed in EFA. The EFA was initially performed with 40 items in the scale.
Based upon the principal component analysis method (Tabachnick & Fidell, 2013), the results showed that four
factors had an eigenvalue greater than 1, explaining 58.93% of the total variance. A total of 12 items, which had
factor loadings less than .40 and cut off points lower than .10 by overlapping in two factors, were excluded from
the scale and then EFA was again performed in this research (Stevens, 2009). The repeated factor analysis indicated
that 28 items with factor loadings ranging from .62 to .73 were distributed to a single factor, which accounted for
43.00% of the total variance in contrast to the first factor analysis. Cronbach’s alpha reliability coefficient (α) was
.93 with regard to the internal consistency of the scale.
In the factor analysis, the explained total variance values higher than 40% are acceptable for single-factor
scales (Cokluk et al., 2014). Also, the observation of the scree plot graphic indicated the scale had single-factor
structure as presented in Figure 2.
Figure 2
Scree-plot graphic of EFA
43
(pp. 38-49) ISSN 2538–7138 /Online/
Finally, it was seen that the scale consisted of 28 items with single factor. Based upon the total scores of the
students, the independent t test was used to determine the significant differences between lower (n = 508) and
upper (n = 508) groups (27% in each group). The t values for all items were significantly discriminative in the scale
consisting of 29 items (tmin = 29.87; tmax = 36.62; p < .001). According to t test results, it was determined that each
item had distinctiveness and these two groups were distinguished from each other.
The item-total correlations called discrimination index of the items were also calculated for item analysis
(Buyukozturk, 2018). Item-total correlation coefficients were ranging from .65 to .59 values (p < .001). The factor
loadings were ranging from .73 to .62 for these 28 items. These values are used to explain the relationship between
the factor and its items. Factor loadings above or equal to .45 are, as a good criterion, acceptable for including the
items in the scale (Buyukozturk, 2018). The items with a factor loading below this value were not considered within
the scope of the scale. In this regard, these findings are summarized in Table 1.
Table 1
Factor Loadings and Item-Total Correlations of the Items
Common
Factor
Items Variance r t
loadings
(h2)
i32. I can explain the basic concepts related to the subject of sound. .54 .73 .65* 35.26*
i31. I can offer solutions to reduce noise pollution. .53 .73 .65* 36.62*
i26. I can solve problems related to the subject of light. .53 .72 .64* 34.59*
i38. I can explain the differences between heat and temperature. .52 .72 .64* 35.97*
i29. I can give examples from daily life for light sources. .51 .72 .63* 35.25*
i28. I can do experiments about the propagation of light. .50 .71 .63* 32.95*
i2. I can use problem solving stages in questions about the subject of force. .49 .70 .63* 33.60*
i35. I can do experiments about the propagation of sound. .48 .69 .63* 33.93*
i15. I can relate basic concepts of electricity to everyday life. .47 .69 .63* 34.75*
i10. I can give examples from daily life with regard to energy types. .47 .69 .63* 32.43*
i18. I can find solutions to problems related to electricity in everyday life by means of what I
.46 .68 .62* 35.81*
have learned about electricity.
i24. I can do the assignments about the subject of electricity without difficulty. .45 .68 .62* 32.69*
i23. I can develop new ideas to benefit from sunlight for a long time. .45 .68 .62* 33.73*
i6. I can explain the relationship between force and movement with examples from daily life. .44 .67 .62* 31.85*
i1. I can do experiments about the effect of force on objects. .44 .66 .61* 35.09*
i25. I can explain the working principles of electrical appliances used in daily life. .44 .66 .61* 32.68*
i39. I can do experiments about the effect of heat and temperature on substances. .43 .64 .61* 33.27*
i34. I can tell about the development of audio technologies from past to present. .43 .64 .61* 31.67*
i37. I can relate what I have learned about heat and temperature to everyday life. .43 .64 .61* 32.64*
i19. I can solve problems related to electricity. .42 .64 .61* 33.74*
44
ISSN 2538–7138 /Online/ (pp. 38-49)
Common
Factor
Items Variance r t
loadings
(h2)
i27. I can offer solutions to reduce light pollution. .42 .63 .61* 35.16*
i30. I can prepare a project on lighting technologies. .42 .63 .61* 32.07*
i13. I can give examples from daily life about solid, liquid or gas pressure. .41 .63 .60* 31.68*
i4. I can explain the factors affecting the speed of moving vehicles. .41 .63 .60* 29.94*
i7. I can explain the working principles of tools that facilitate daily errands. .41 .63 .60* 30.39*
i5. I can solve problems related to speed. .41 .62 .60* 29.87*
i22. I can apply what I have learned about electricity saving in everyday life. .40 .62 .60* 32.31*
i11. I can do experiments about the interconversion of energies. .40 .62 .59* 31.10*
Eigen value (Total)=12.04

Explained total variance (%) = 43.00
Cronbach alpha = .92
*p < .001
With respect to the internal consistency coefficient, Cronbach’s alpha value was .87 for the single factor of
the scale. Based on this finding, it was seen that the results obtained from the data set (α > .80) were quite reliable
(Cohen et al., 2007).
Second Stage (Results of the Confirmatory Factor Analysis)
Following the EFA, the scale with single-factor model was tested by performing Confirmatory Factor Analysis
(CFA) to confirm it on a different sample. Hence, the data set for 764 cases obtained from an independent sample
was analysed in CFA.
The modification was made by taking into account the relationship between the errors of some items in the
suggested model (e.g., e1-e2). Based upon maximum likelihood method, CFA results indicated that the data fit sta-
tistics were at an acceptable level (χ²/df = 3.52, p < .001; TLI = .91; CFI = .91; NFI = .89; GFI = .89; IFI = .90; RMSEA =.05;
SRMR = .03). In this context, these results related to the independent sample indicated that single-factor structure
was suitable for the scale. Hence, none of the items was excluded from the scale.
With respect to the internal consistency coefficient, Cronbach’s alpha value was .84 for the single factor of
the scale. Based on this finding, it was seen that the results obtained from the data set (α >.80) were quite reliable
(Cohen et al., 2007).
Third Stage (The Results of Reliability Analysis)
In the final stage, both Cronbach’s alpha (α) value for internal consistency and Pearson correlation coefficient
(r) for re-test reliability were examined to determine the reliability of the scale. The data set for 91 cases obtained
from an independent sample was analysed for reliability. With regard to the internal consistency of the scale, the
results showed that Cronbach’s alpha reliability coefficient (α) was found to be .84 for the internal consistency of
the scale. In the related literature, if this coefficient is higher than .80, the scale is satisfactorily and highly reliable
(Cohen et al., 2007). The students in the sample (n = 91) answered the items of the scale again three weeks after
the previous application. The findings showed that there was a high level of relationship between the two applica-
tions (r = .87, p < .001). The high correlation coefficient showed that the scale was a stable measurement, and the
findings were consistent when the scale was applied at different times.
Finally, the scale consisted of 28 items with single factor. While the lowest score can be 28, the highest score is
140. From this point of view, the upper scores from the scale mean that the student has higher self-efficacy belief
45
(pp. 38-49) ISSN 2538–7138 /Online/
towards physics subjects in science education. On the contrary, lower scores mean that the student’s self-efficacy
belief is low.
Discussion
In this research, a valid and reliable scale was developed to measure the self-efficacy towards physics subjects
in science education by using data obtained from 2737 students at 6th, 7th, and 8th grades in a province located
in the north of Turkey. When the relevant literature was examined, it was observed that scale development stud-
ies were generally related to students’ self-efficacy beliefs towards science education at lower-secondary school
level (Aslam & Ali, 2017; Tatar et al., 2009). However, the number of the existing scales are insufficient to measure
the self-efficacy belief regarding more specified domains such as specific subjects, tasks, units, themes or learning
fields in science education. Moreover, the studies on these subjects were observed to have been conducted mainly
at higher education level and no self-efficacy scale towards specific fields or areas was encountered in science
education, particularly in physics at lower-secondary school level. Unlike the present research, it was seen that
these scales had a multi-factor structure. For example, “Acid-Base Self-Efficacy Perception Scale”, which included
a specific subject in the field of chemistry, was developed for prospective teachers by Ilhan and Cicek (2017). The
results showed that the explained total variance value of the two-factor scale consisting of 14 items (9 positive,
4 negative) was found to be 46.71% and the Cronbach’s alpha (α) value for internal consistency was .86, like the
findings of the reliability analysis in the research. Likewise, the scale entitled “Astronomy Subjects Teaching Self-
Efficacy Beliefs” was developed by Demirci (2017) and its final version consisted of 14 items with 3-factor structure
while the draft scale was composed of 41 items. With regard to biology, “Teaching Evolution Self-Efficacy Scale” was
developed for undergraduate students by Inan et al. (2018). It was composed of 11 items with two factors (efficacy
about evolution content knowledge, self-efficacy about teaching evolution).
On the other hand, there are few scales on self-efficacy belief toward physics. Most of them were concentrated
on upper-secondary school and higher education students, not lower-secondary or middle school students. In
addition, some researchers (see Alpaslan & Isik, 2016; Lin et al., 2015) adapted the self-efficacy scale which was
developed by Lin and Tsai (2013) for science learning to physics learning. Selcuk et al. (2018) investigated the
self-efficacy belief toward physics according to the factors of physics achievement and skills of using physics
knowledge in the upper-secondary school. However, while being centred upon the general structure of physics,
these scales neither focused on a task- or a subject-based construct nor were they at lower-secondary school level.
In Turkey, the physics subjects are covered by science course curricula at these grades. From a multidimensional
perspective, the previous scale development studies have offered a general or comprehensive framework in
science education. Likewise, there are also some studies exploring science self-efficacy with one-factor structure
as a component of motivation through a course- oriented approach (Glynn et al., 2009; Tuan et al., 2005; Pintrich
et al., 1991). To illustrate, Motivated Strategies for Learning Questionnaire (MSLQ) was developed by Pintrich
et al. (1991) has been frequently used to measure the self-efficacy -as a predictor of motivation- for science or
other disciplines in the literature. However, the single-factor structure in the current research is substantial; at
least for the following reasons, it differentiates from other studies. First, it may profoundly provide to subject-
based evidence in examining the cognitive or affective features of science. Second, it can transform into a
multi-dimensional structure together with variables regarding other sub-disciplines (the subjects of biology,
chemistry etc.) of science. Finally, it may specifically reveal the latent variables of science learning in terms of
physics subjects rather than a general perspective.
The final version of the scale consists of 28 items with single factor. Based on the findings of this research, the
several implications may be emphasized for science education. The scale can be used as a valid and reliable instru-
ment by educators or researchers to understand students’ self-efficacy towards physics subjects. Using these findings,
teachers may design learning environments by identifying the students with poor self-efficacy. For instance, they
can initially give students easier tasks for learning activities rather than difficult in physics education. Another way
to develop students’ self-efficacy belief is usage of instructional technologies or combination of these technologies
and several learning methods in physics teaching process (Fidan, 2018; Seifert & Tshuva-Albo, 2014). Moreover, the
scale can be utilized as a mediator variable in determining the effects of other latent factors in science learning.
46
ISSN 2538–7138 /Online/ (pp. 38-49)
In conclusion, the current research showed that the scale had good performance to measure self-efficacy
belief towards physics subjects for lower-secondary school in science education. Based on the relevant literature
reviews, 52 items were identified for a draft scale form by taking the experts’ views to ensure face and content
validity. After that, a pilot application of the scale was performed on the participants from the lower-secondary
schools. Firstly, EFA (n = 1918) was conducted for determining the factor structure of the self-efficacy scale. 12 items
were excluded from the scale owing to their poor factor loadings. The EFA findings demonstrated that the scale
consisted of 28 items with single factor. To determine the discrimination of the item, both item-total correlations
and the comparison between the scores of lower and upper groups (27% in each group) were examined.
In the second stage, the scale with single-factor model was tested to confirm it on a different sample by
conducting CFA (n = 785). CFA results showed that these fit indexes were acceptable and suitable for construct
validity. After all, both Cronbach’s alpha (α) value for internal consistency and Pearson correlation coefficient (r) for
re-test reliability (n = 91) were examined to determine the reliability of the scale. The Cronbach’s alpha reliability
coefficient (α) was .84 for the internal consistency of the scale and re-test reliability coefficient (r) was found as .87
for a three-week interval. The high reliability coefficients indicated that the reliability of the scale was at a satisfac-
tory level and it measured accurately.
The current research had some limitations. The first of these limitations was that the data were collected
from a province located in the north of Turkey. It is recommended for future studies that the scale can be applied
to a larger sample or different cultures including the students studying at different regions in Turkey. As another
limitation, the interrelationships among variables of science (attitude, motivation etc.) were not examined in this
study. As a matter of fact, physics is related to mathematics, natural science laws or rules, and the behaviours of
the universe. Hence, further correlational, predictive, or modelling studies should provide solid evidence about the
cause-effect relationship of variables and offer a general picture of science learning. It would be worth to explore
the relationships between physics self-efficacy belief and the sources of self-efficacy. Because the scale includes
topics and concepts of physics in detail, it can be used at higher grade levels aside from lower-secondary level. In
future studies, it can be examined whether self-efficacy differentiates in terms of demographic variables such as
student’s gender and class or not. Similar to the findings of this research, self-efficacy scales on specific-domain
subjects or tasks (biology subjects or themes, chemistry experiments etc.) can be developed in science education.
Moreover, the subject-based self-efficacy scales (force and movement, pressure, electricity etc.) may be created
separately in terms of physics subjects.
Acknowledgements
The research which supported by funding from the Bolu Abant Izzet Baysal University Scientific Research
Projects Committee in Turkey (Project Number: 2016.02.02.1101) was derived from first author’s PhD dissertation
conducted under the supervision of the second author.
References
Alpaslan, M., & Isik, H. (2016). Examination of the validity and the reliability of physics self-efficacy scale. Mustafa Kemal University
Journal of Social Sciences Institute, 13(33), 111-122.
Anderson, L. W. (1988). Attitudes and their measurement. In Keeves, J. P. (Ed.), Educational research, methodology and measurement.
An international handbook (pp. 421-426). Elsevier Science & Technology.
Aslam, S., & Ali, M. S. (2017). Effect of self-efficacy on students’ achievement in science: A case of secondary school students in
Pakistan. European Journal of Education Studies, 3(11), 219-234. https://doi.org/10.5281/zenodo.105029
Aypay, A. (2010). The adaptation study of general self-efficacy (GSE) scale to Turkish. Inonu University Journal of the Faculty of
Education, 11(2), 113-132.
Ayre, C., & Scally, A. J. (2014). Critical values for Lawshe’s content validity ratio: Revisiting the original methods of calculation. Measurement
and Evaluation in Counseling and Development, 47(1), 79-86. https://www.tandfonline.com/doi/full/10.1177/0748175613513808
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
Bandura, A. (1997). Self-efficacy: The exercise of control. Freeman.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares, & T. Urdan (Eds.), Adolescence and education, Vol. 5:
Self-efficacy beliefs of adolescents (pp. 307-337). Information Age.
47
(pp. 38-49) ISSN 2538–7138 /Online/
Britner, S. L., & Pajares, F. (2006). Sources of science self-efficacy beliefs of middle school students. Journal of Research in Science
Teaching, 43, 485-499.
Buyukozturk, S. (2018). Manual of data analysis for social sciences: Statistics, research pattern SPSS applications and interpretation
(24. ed.). Pegem Akademi.
Cohen, L. M., Manion, L., & Morrison, K. (2007). Research methods in education (6th ed.). Routledge.
Comrey, A., & Lee, H. (1992). A first course in factor analysis. Erlbaum.
Cokluk, O., Sekercioglu, G., & Buyukozturk, S. (2014). Multivariate statistic for social sciences: SPSS and LISREL applications (3rd ed.).
Pegem Akademi.
Demirci, F. (2017). Science teachers’ self-efficacy beliefs about astronomy subjects teaching: A mixed method research (Unpublished
master’s thesis). Ordu University, Ordu.
Fidan, M. (2018). The impact of problem-based science teaching assisted with augmented reality applications on academic achievement,
retention, attitude, and belief of self-efficacy (Unpublished PhD thesis). Bolu Abant Izzet Baysal University. https://tez.yok.
gov.tr/UlusalTezMerkezi
Glynn, S. M., Taasoobshirazi, G., & Brickman, P. (2009). Science motivation questionnaire: Construct validation with nonscience
majors. Journal of Research in Science Teaching, 46, 127-146.
Ilhan, N., & Cicek, Ö. (2017). The development of acid-base self-efficacy perception scale and the investigation of pre-service
science teachers’ self-efficacy perceptions on acid-base. Sakarya University Journal of Education, 7(1), 123-141.
Inan, S., Irez, S. Han-Tosunoglu, C., & Cakır, M. (2018). Teaching evolution self-efficacy scale: The development, validation, and
reliability study. Trakya University Journal of Education Faculty, 8(2), 241-260.
Kapucu, S. (2017). Predicting physics achievement: attitude towards physics, self-efficacy of learning physics, and mathematics
achievement. Asia-Pacific Forum on Science Learning and Teaching, 18(1), 1-22. https://www.eduhk.hk/apfslt/v18_issue1/
kapucu/page4.htm
Karasar, N. (2012). Bilimsel araştırma yöntemi [Scientific research methods]. Nobel.
Klassen, R. M., & Usher, E. L. (2010). Self-efficacy in educational settings: Recent research and emerging directions. In T. C. Urdan,
& S. A. Karabenick (Eds.), Advances in motivation and achievement: Vol. 16A. The decade ahead: Theoretical perspectives on
motivation and achievement (pp. 1-33). Emerald.
Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). The Guilford.
Lawshe, C. H. (1975). A quantitative approach to content validity. Personnel Psychology, 28(4), 563-575.
Lin, T.-J., & Tsai, C.-C. (2013). A multi-dimensional instrument for evaluating Taiwan high school students’ learning self-efficacy in
relation to their approaches to learning science. International Journal of Science and Mathematics Education, 11(6), 1275-1301.
Lin, T. C., Liang, J. C., & Tsai, C. C. (2015). Conceptions of memorizing and understanding in learning, and self-
efficacy held by university biology majors. International Journal of Science Education, 37(3), 446-468.
https://www.tandfonline.com/doi/abs/10.1080/09500693.2014.992057
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of sample size for covariance
structure models. Psychological Methods, 1, 130-149.
MoNE (2013). Physics curriculum for secondary education. http://mufredat.meb.gov.tr
MoNE (2018). Fen bilimleri dersi öğretim programı (ilkokul ve ortaokul 3, 4, 5, 6, 7 ve 8. sınıflar) [Science curriculum (elementary and
lower-secondary 3, 4, 5, 6, 7 and 8. grades)]. http://mufredat.meb.gov.tr
OECD (2017). PISA 2015 assessment and analytical framework: Science, reading, mathematic financial literacy and collaborative
problem solving (revised edition). PISA. http://dx.doi.org/10.1787/9789264281820-en
Oluwatayo, J. (2012). Validity and reliability issues in educational research. Journal of Educational and Social Research, 2, 391-400.
Pintrich, P. R., Smith, D. A. F., Garcia, T., & McKeachie, W. J. (1991). A manual for the use of the motivated strategies for learning
questionnaire (MSLQ). The University of Michigan, Ann Arbor, MI. https://files.eric.ed.gov/fulltext/ED338122.pdf
Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’ academic performance: A systematic
review and meta-analysis. Psychological Bulletin, 138(2), 353-387. https://doi.org/10.1037/a0026838
Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square statistic for moment structure analysis. Psychometrika, 66, 507-
514. https://link.springer.com/article/10.1007/BF02296192
Selcuk, G. S., Caliskan, S., & Demircioglu, S. (2018). Determining the applicability of the physics self-efficacy scale for high school
students. Turkish Studies, 13(4), 225-244.
Schunk, D. H. (1990). Goal setting and self-efficacy during self-regulated learning. Educational Psychologist, 25, 71-86.
Schunk, D. H., & Pajares, F. (2009). Self-efficacy theory. In K. R. Wentzel, & A. Wigfield (Eds.), Handbook of motivation at school (pp.
35-53). Routledge.
Schunk, D. H., & Usher, E. L. (2011). Assessing self-efficacy for self-regulated learning. In B. J. Zimmerman, & D. H. Schunk (Eds.),
Handbook of self-regulation of learning and performance (pp. 282-297). Routledge.
Seifert, T., & Tshuva-Albo, V. (2014, April). Teaching based augmented reality and smartphones to promote learning motivation
among middle school students. Proceedings of the 19th Annual TCC Worldwide Online Conference (pp. 111-121). The University
of Hawai’i Kapi’olani Community College & The University of Hawai’i at Manoa.
Stevens, J. (2009). Applied multivariate statistics for the social sciences. Routledge.
Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). Allyn and Bacon.
Tatar, E., Yildiz, N., Akpinar, E., & Ergin, Ö. (2009). Exploratory and confirmatory factor analysis of the metacognition scale for
primary school students. Educational Sciences: Theory & Practice, 9(3), 1591-1604.
48
ISSN 2538–7138 /Online/ (pp. 38-49)
Tezer M., & Asiksoy, G. (2015). Engineering students’ self-efficacy related to physics learning. Journal of Baltic Science Education,
14(3) 311-326. http://www.scientiasocialis.lt/jbse/files/pdf/vol14/311-326.Tezer_JBSE_Vol.14_No.3.pdf
Tuan, H., Chin, C., & Shieh, S. (2005). The development of a questionnaire to measure students’ motivation towards science learning.
International Journal of Science Education, 27, 639-654. https://www.tandfonline.com/doi/abs/10.1080/0950069042000323737
Tuncel, M., & Fidan, M. (2018). Subjects perceived as difficult in secondary science course and recommendations. Proceedings of
the 6th International Congress on Curriculum and Instruction. Kafkas University, Kars.
Usher, E. L., & Pajares, F. (2009). Sources of self-efficacy in mathematics: A validation study. Contemporary Educational Psychology,
34, 89-101.
Yelgun, A. (2009). Misconceptions and their forming reasons of 8th class students’ in primary schools about buoyant force. (Unpublished
master’s thesis). Ataturk University, Erzurum. https://tez.yok.gov.tr/UlusalTezMerkezi
Zimmerman, B. J. (1995). Self-efficacy and educational development. In A. Bandura (Ed.), Self-efficacy in changing societies (pp.
202-231). Cambridge. https://doi.org/10.1017/CBO9780511527692.009
Received: July 20, 2020 Accepted: January 06, 2021
Cite as: Fidan, M., & Tuncel, M. (2021). Developing a self-efficacy scale toward physics subjects for lower-secondary school
students. Journal of Baltic Science Education, 20(1), 38-49. https://doi.org/10.33225/jbse/21.20.38
Mustafa Fidan PhD, Assistant Professor, Department of Educational Sciences,

(Corresponding author) Faculty of Education, Bartin University, Bartin, 74100, Turkey.
Meric Tuncel PhD, Assistant Professor, Department of Educational Sciences,
Faculty of Education, Bolu Abant Izzet Baysal University, Bolu,
14280, Turkey.
49
IMPACT OF INSTRUCTION ON
SCIENCE PERFORMANCE:
ISSN 2538-7138 /Online/
LEARNING INITIATIVE AS A
MEDIATOR AND GENDER AS A
LIMITED MODERATOR
Abstract. This research explores whether

classroom life (CL), textbooks (TE), and Tao Jiang, Ji-gen Chen,
learning initiative (LI) are mediators
Yin-yin Wu
between instruction (P) and science
performance, as well as whether these
mechanisms are moderated by gender.
484 eighth grade students completed the
questionnaire with four subscales of P, LI,
TE, and CL. For the needs of triangulation
and complementarity, three focus group
interviews were conducted later. Based
on mediation analysis and multi-group
Introduction
structural equation modeling, it was found
that 1) the direct effects of P on LI, P on CL,
Asian science is experiencing rapid development. The same is true for
P on TE, and LI on science performance are
its science education. The 15 year old students in Singapore, Macao(China),
significant, while the other direct effects
Hong Kong(China), Beijing-Shanghai-Jiangsu-Guangdong(China), Beijing-
are insignificant; 2) comparing to the male
Shanghai-Jiangsu-Zhejiang(China), and Korea performed well in mathematics
group, the direct effect of P on LI in the fe-
and science tests (OECD, 2016; OECD, 2019a). Nevertheless, there are several
male group is larger; 3) characteristics hin-
deficiencies in Asian science education. The exam-centric education system
dering students’ science learning include:
in most of the Asian countries usually concentrates on content knowledge
the pace of a lesson is too fast, pictures and
and neglects of nurturing students’ innovative thinking, which would hamper
experiments are less in the textbook, and
the rise of Asian science (Lim, 2010). For example, comparing to American
top students and low proficiency students
college students, Chinese students gained much better scores on tests of
are uncooperative. Findings expose that
physics content knowledge but failed to maintain that advantage on tests
instruction significantly influence students’
of scientific reasoning (Bao et al., 2009). Meanwhile, it seems that some tra-
science performance, and this impact is
ditional advanced countries, such as the United States and the UK, also face
completely mediated by students’ learning
challenges in maintaining their leading advantages in science. Despite many
initiatives. The relation between instruction
years of standards-based reform, the US only had minimal improvements
and learning initiative is stronger in the
in its science education (“The Science of Education Reform,” 2006). The US
female group. Textbooks can be useless in
students’ mathematics and science achievements in the 2009 Program for
the context that instruction does not match
International Student Assessment (PISA) test were significantly behind the
students’ learning ability.
Asian participating countries and other developed nations (“Change the
Keywords: eighth-grade students, gender
Equation,” 2011), and this status was not meliorated in the 2015 and 2018
difference, mediation analysis, science
PISA tests, especially in the mathematics test (OECD, 2016; OECD, 2019a).
performance
Although the UK students’ science and mathematics performance in 2015 and
2018 PISA tests were better than the US students, the UK also needed to deal
with its students losing their interests in mathematics and science during the
Tao Jiang, Ji-gen Chen, Yin-yin Wu secondary school period (“Science Education Reforms in the UK,” 2012). On
Taizhou University, China account of there are some important factors in school environments influenc-
ing students’ science learning, this paper intends to depict a clear picture of
how these factors are integrated to affect a student’s science performance.
50
ISSN 1648–3898 /Print/ IMPACT OF INSTRUCTION ON SCIENCE PERFORMANCE: LEARNING INITIATIVE AS A
ISSN 2538–7138 /Online/ (pp. 50-66)
Factors Influencing Students’ Science Performance in School Context
Heredity, social-economic status, the quality of schooling, investment on per student by governments, and
cultural and social settings explain students’ science performance. This research has no intention to depict such a
big picture and only focus on exploring the mechanism of science performance production in the school context.
In school science, the stereotype is deeply entrenched that men in nature are more competitive than women
(Fine & Elgar, 2017). Poor media portrayals such as women are responsible for housekeeping, and a lack of role
models who are successful in a career may be responsible for that (Saujani, 2017). Therefore, the gender difference
in science learning is something related to women’s role definition, which human society imposes on them. Not dra-
matically, but gradually this stereotype has been changing. The increasing urbanization and well-educated parents
may play a big role in terms of this change (Burušić et al., 2019; Gupta, 2017). By early intervention, Saujani (2017)
succeeded in teaching middle school female students to write computer programs. Beaman et al. (2012) found
that the policy experiment of “Female Leadership” in India could promote girls’ educational attainment. Although
it was found that boys did better in mathematics in the 2003 PISA test (Machin & Pekkarinen, 2008), Guiso et al.
(2008) argued that this conclusion did not hold in countries with a more gender-equal culture. Coll et al. (2010)
also claimed that there was little gender difference in New Zealand students’ science performance in PISA 2006,
yet girls’ science performance was 17 scores higher than boys’ performance in Thailand. PISA 2018 results again
reported that ‘girls outperformed boys in science by two score points on average across OECD countries’ (OECD,
2019b). Nevertheless, by analyzing the “Trends in International Mathematics and Science Study” (TIMSS) 2015
data, Aşkın and Öz (2020) argued though girls outperformed boys in science in 5 Asian countries, the opposite
side was true for Georgia, Italy, Lithuania, and the United States. In a global view, girls have progressed in their
science performance, both boys and girls can be successful in learning science. Therefore, it is valuable to explore
the mechanism that accounts for boys’ and girls’ success in science classrooms.
In school science, instruction and learning, and their interactions influence students’ learning. Teachers who
accommodate their instruction to students’ learning levels could improve students’ test-scores (Kremer et al.,
2013). Also, strategies that prompt learner’s engagement in science tasks are repeatedly emphasized by scholars.
A large number of these strategies, such as enhancing teacher-student interactions (Allen et al., 2011), taking
notes through the mind-mapping method (Akinoglu & Yasar, 2007), active learning (Freeman et al., 2014), direct
instruction with hands-on and minds-on attributes (Cobern et al., 2010), inquiry-based reform (Sotáková et al.,
2020), integrating doing, reading, writing, and talking (Webb, 2010), making connections between student’s life
and subject matter knowledge (Hulleman & Harackiewicz, 2009), and using learning techniques (Dunlosky et al.,
2013) were reported to positively relate to students’ learning gains. Conversely, in classrooms where teachers did
most of the talking, student’s performance in tests was remarkably low (Setati et al., 2002). However, teachers
talking lot was a typical scene in mainland China’s science classrooms, but it did not prevent Chinese students
from getting good scores in PISA science tests. Then, it raises two important questions. First, by which means
does direct instruction affect students’ science learning? Second, which behaviors of direct instruction have
hindered their science learning from students’ viewpoints? However, previous research paid little attention to
the answers of these two questions.
In the school context, textbooks are essential resources (Oates, 2014; Wilkens, 2011). It supports teaching
and learning to make sure the pedagogy is structured (Reichenberg, 2016). For this purpose, it always includes
the necessary tools for learning (Hanbay, 2015), such as pictures, tables, and laboratory instructions. The utilization
levels of textbooks vary in countries due to their educational systems. Chinese teachers usually cover 100% subject
matter knowledge (SMK) in textbooks in their lessons, because direct instruction makes it feasible by controlling
time spent on a topic. While in the United States, less than 50% of high school science teachers, and less than
70% of high school mathematics teachers covered more than 75% SMK in textbooks in their lessons (Banilower
et al., 2013). In England, only 10% of teachers viewed textbooks as a basis for instruction, in contrast to 70% in
Singapore, and 95% in Finland (Oates, 2014). So, it seemed that a country’s textbooks utilization had some kind
of relationship with her PISA test score. But this relationship also could be a coincidence, it might be other factors
in the school context rather than textbooks, contributing to students’ test performance. Therefore, it is in urgent
need of investigating whether and how science textbooks influence students’ performance. If a textbook does
play as an irreplaceable role in students’ learning outcomes, teachers should value and maximize its utilization.
If a textbook does not play a big role in students’ learning, in consideration of the fact that it cost a lot of money
every year, there is a voice of replacing it with the low-cost electronic textbook (Robinson, 2011). However, it is hard
51
(pp. 50-66) ISSN 2538–7138 /Online/
to choose between the two arguments as past research has said little about the mechanism of science textbook
played in students’ performance with a quantitative method.
Some researchers made efforts to understand students’ classroom life (Brophy, 2006; Johnson et al., 2009).
Classrooms are shared social spaces where participants’ personal and institutional lives are weaved together (Gieve
& Miller, 2006). In institutional lives, teachers and students make efforts to resolve problems; in personal lives, they
speak to each other. The personal dimension highlights the emotional connections of participants, while the emo-
tional characteristics of classrooms influence students’ learning outcomes (Fraser, 1987). For example, mutual respect
between teachers and students prompted students’ engaging in tasks (Matsumura et al., 2008). For teachers, more
interactions fewer interventions improved their students’ achievements (Brophy, 2006; Djigic & Stojiljkovic, 2011).
For students’ community, more respect less public competition enhanced their learning outcomes (Pierce, 1994).
Recently, it was reported that school bullying has shaken the public’s trust in schools as a place of social learning
and development (Sachs & Mellor, 2005). Teachers also experienced “culture shock” and burnout in schools where
their students faced too much violence on the street (Rushton, 2000). If the minimum level of students’ safety is
unrealizable, there will be no authentic students’ engagement in tasks. Fortunately, conditions like that are rare,
students’ safety in schools is basically under control though it may have some kinds of discipline problems. In this
kind of situation, the roles students’ classroom lives played in forming their science performance need to be known.
However, little quantitative research has been done in this area.
Theoretical Framework and Research Questions
As said above, countries that heavily relied on textbooks (TE) performing well in PISA science tests. Meanwhile,
elements such as instruction (P), students’ learning initiative (LI), and classroom life (CL) were also responsible for
explaining students’ science performance. So far, the majority of existing research in this domain was a “simple
linear regression analysis” pattern because they only focused on a single influential factor of students’ science
performance. Findings that came from this kind of research are less convincing because they can be varied in the
context where more than one factor influenced student’s learning. However, up to now, little was known about the
holistic mechanism of P, in combination with LI, TE, and CL, functioning on students’ science performance. One of
the methods to explore this holistic picture is multiple linear regression (MLR) analysis. But it also has some deficien-
cies in exploring the complex mechanism of the foregoing factors exerting on students’ science performance. The
reason for it is the foregoing factors are both taken as covariates in MLR analysis. Therefore, although these factors
are taken into the statistic model simultaneously, they are not connected, but only connected to the outcome vari-
able respectively. It is hard to believe that these factors are independent of each other. Being components of the
instructional system, there must be some kinds of connections among them to make the system function well. A
reasonable holistic picture requires not only including these influential factors in the model simultaneously but
also showing their interactions. Therefore, a theoretical framework that provides a reasonable explanation for this
kind of interaction should be found first.
Since these factors focused on students’ science learning, the instructional design model (ID model) can be a
theoretical framework to integrate them. Instructional design is a paradigm that is related to encoding and decoding
the messages (Gagne et al., 2005; Khalil & Elkhider, 2016; Ledford & Sleeman, 2000, p.13). It exposes that teaching
and learning in classrooms is a process of message generation, flow, and assimilation. As far as assimilating the
message is concerned, Reigeluth (1999) put forward that there are four conditions: “what is to be learned, the nature
of learners, the learning environments and constraints.” Constraints are something like money and time teachers
owning to develop their instruction. As far as the condition “what is to be learned” is concerned, it relates to teach-
ers’ interaction with textbooks, thus could determine “what is to be learned” in a lesson. It is shown in figure 1 as
a2 path. For the condition “the nature of learners” is concerned, it means teachers’ understanding of their students’
characteristics and learning strategies, which in turn influence their instructional strategies, it is shown in figure
1 as a1 path. In terms of the condition “the learning environment,” it is built by the interactions of instruction and
students’ classroom life. It is shown in figure 1 as a3 path. To achieve learning goals, these three paths all should
point to the desired outcomes, which are bi path in figure 1. Based on systematic thinking, Dick et al. (2015) also
put forward the components leading to the desired outcomes were “the instructor, learners, materials, instructional
activities, delivery system, and learning and performance environments” (pp. 1-3). In this framework, the teacher,
students, and materials are the static component in the instructional system. Teachers’ adaptation of textbooks
generates abundant learning materials. The interactions between the teacher and students generate instructional
52
ISSN 2538–7138 /Online/ (pp. 50-66)
activities and the delivery system. Since the learning environments are students and teachers combining into a
collaborative group to solve problems (Dick et al., 2015, p.195), it is the interaction of instruction and students’
classroom life, as well as the interaction of instruction and learning. For a performance environment, it is behaviors
relating to students’ learning initiatives (Dick et al., 2015, p.214), which is ignited by instruction. Therefore, Dick et
al. (2015)’s framework also supports the statistical model shown in figure 1.
Figure 1
Statistical Diagram of the Mediation Effect of Instruction on Science Performance through Three Mediators
Learning initiative
(LI)
a1 b1
Textbooks (TE)
a2 b2
Instruction (P) c’ Performance
a3 Classroom life b3
(CL)
The key research questions were:

1. Do learning initiative (LI), textbooks (TE), and classroom life (CL) mediate the relation of instruction (P)
and students’ science performance?
2. Is gender a moderator of the following relations: instruction on learning initiative, learning initiative
on performance, instruction on textbooks, textbooks on performance, instruction on classroom life,
classroom life on performance, and instruction on performance?
3. Is gender a moderator of the three specific indirect effects and the total effect of instruction on the
performance shown in figure 1?
Answers to these questions meant to depict a clear picture of science performance production in the school
context.
General Background
In this research, the school life questionnaire (SLQ) was developed with the aids of two officers who came
from the city’s teaching research office. This office is affiliated to the city’s education bureau and responsible
for the quality of the city’s elementary and secondary education. The questionnaire selected items from several
classroom observation protocols (Sawada et al., 2002; Weaver et al., 2005; Weiss et al., 2004), then modified some
items based on the officers’ opinions to fit the local environment. In developing the original SLQ, the criteria of
specificity, clarity, and brevity were of utmost importance (Cowles & Nelson, 2015, p. 108; DeVellis, 2017, pp.
103-105; Dillman, 2009, p. 32; Fowler, 1995, p. 2). For specificity and clarity, no unfamiliar words or terms were
used in the SLQ’s item wordings. For the need for brevity, the words in the SLQ’s item were as little as possible.
The officers helped to arrange two focus group interviews in schools. It was a typical procedure in a pilot survey
to test the quality of items (Cowles & Nelson, 2015, p. 128; Fowler, 1995, pp. 104-105). Based on low secondary
school students’ feedback, some of the item wordings were revised to make its meaning as clear as possible. As
the original SLQ covered dozens of items, the interviewees also put forward their classmates would get bored to
answer so many questions. Therefore, the length of the original SLQ needs to be reduced. Then, the questionnaire
was applied to hundreds of lower secondary school students. According to the data gotten from the pre-test, the
“corrected item-scale correlation” and “Cronbach’s alpha if item deleted” values were used to evaluate the quality
of an item. The item which its corrected item-scale correlation was smaller than .4 and its Cronbach’s alpha if item
53
(pp. 50-66) ISSN 2538–7138 /Online/
deleted was higher than the scale’s Cronbach’s alpha, was deleted. In this procedure, the key items were identified
and constituted the final edition of SLQ. Then, 576 eighth-grade students completed the SLQ in December 2016.
Confirmatory factor analysis (CFA), and exploratory structural equation modeling (ESEM) confirmed SLQ’s four
factors’ structure and its composite reliability. Meanwhile, mediation analysis depicted a clear picture to explain
the complex mechanism of LI, TE, and CL played in the relationship between P and science performance. However,
some abnormalities were found hard to explain in the quantitative framework, such as why so many students who
satisfied with their science teachers’ instruction got poor performance in tests? It was a signal that depends on the
quantitative method itself was not enough to expose the complex mechanism of students’ science learning. For
the needs of triangulation and complementarity (Hesse-Biber, 2010, pp. 3-4), three focus group interviews were
conducted in December 2017. It allowed this research to make a thick description of students’ opinions on their
science learning initiative, classroom life, and textbook utilization.
Sample
This research selected participants from two public schools and one private school in Zhejiang province.
These schools had good representativeness to the city’s lower secondary schools. One public school is located in
downtown. The other public school is located in the urban-rural conjunctive region. Both of them recruit students
from the surrounding area. That private school is located downtown to recruit students from remote rural areas.
Students were free to choose whether or not to participate in this research.
Procedures and Instruments
In consideration of students who may hesitate to tell the truth in the survey, the survey was conducted by
student teachers. Lower secondary school students usually take student teachers as their elder sisters or elder
brothers. They believe in their student teachers would not hand over their responses to in-service science teachers.
In this light, they are more likely to tell their true feelings about school life, especially their opinions on the qual-
ity of their science teachers’ instruction. As soon as they finished the questionnaire with a real name, the student
teachers entered data in a spreadsheet and connected it to their science performance. In this semester, the four
chapters in the science textbook are chemistry, geography, biology, and physics topics, respectively. Students will
take an examination as soon as they finish one chapter. This educational system is named “the month test” and has
decades of history in Zhejiang province. Typically, a student’s scores on different topics are fluctuant, depending on
one test could not determine the student’s learning performance. The test papers, either used in the month tests
or the final examination, are arranged by schools. It is hard to guarantee the test invariance of these different test
papers. Since the three schools’ quality of teaching decreases in turn, the downtown public school is the best, and
the downtown private school is the worst, using the same test papers also would not help to this research. In that
case, students with the same learning initiatives may achieve very different scores due to the school’s quality of
teaching is different. That may result in misunderstanding the mechanism of science performance production in
the school context. Thus, students’ science performance was evaluated by in-service science teachers according to
their performance in the last six months’ science tests with a five-grade system. Compared to the original score in
one test, the five-grade rating was suitable to the SLQ and the different quality of teaching in schools. Eventually,
484 valid responses were obtained.
The research team conducted chi-square tests of independence with the questionnaire data to explore the
dependence of variables, also explored the mechanism of antecedent variables functioning on students’ perfor-
mance by the structural equation modeling approach. Some abnormalities were hard to explain in a quantitative
framework, such as why students who satisfied with their science teachers’ instruction got poor performance in
tests, why did so many students lose their confidence in learning science? Therefore, a qualitative approach would
benefit this research, as it is complementary to the quantitative method. The next year of the foregoing survey had
accomplished, two groups of student teachers came to the foregoing two public schools. In each group, there was
a team leader who was not only responsible for managing the entire team to make sure its members cooperating
well with school staffs, but also responsible for conducting focus group interviews. The interviews were oriented
to find students’ opinions on their teachers’ instruction, textbook, classroom life, and their learning initiative. Us-
ing focus group interviews rather than individual interviews is due to its advantages in saving time and igniting
participants to resonate with each other’s experiences (Auerbach & Silverstein, 2003, p. 17). Every team leader
54
ISSN 2538–7138 /Online/ (pp. 50-66)
organized several discussions in her team to discuss the semi-structured interview questions developed by the
research team. Then the research team held several discussions with two team leaders to reduce questions and
to simplify wordings of some questions, and at the same time clarifying the interview goal. The research team
purposively selected interviewees from the team leaders’ classes rather than other student teachers’ classes. In
this case, interviewees were more likely to tell their opinions as they were familiar with the moderator. One team
leader established two focus groups, each consisting of 14 eighth-grade students. One top student group in which
its member’s science performance was evaluated as level 4 or level 5, one low science proficiency student group
in which its member’s science performance was evaluated as level 1 through level 3. They were labeled as T1 and
L1 groups respectively. Another team leader established one focus group consisting of 6 ninth-grade students. It
was a low science proficiency student group and labeled as group L2. Every group had an equal number of male
and female students. Two team leaders held three interviews. T1 and L1 groups’ interviews lasted approximately
forty minutes, which was one lecture time. The L2 group’s interview cost 30 minutes. In total, three group interviews
were audiotaped and transcribed.
The SLQ has four subscales. One subscale is the independent variable P. The other three subscales in SLQ are
mediation variables LI, TE, and CL. All of them are 5-point subscale, and each subscale has 5 observed indicators/
items. Item wordings are detailed in table 1.
Table 1
The School Life Questionnaire (SLQ)
Items Wordings
Learning initiative (LI)

LI1 I am interested in science.
LI2 I have confidence in learning science.
LI3 I always pay attention to what my teacher is saying.
LI4 I have the enthusiasm to answer my teacher’s questions and attend science activities.
LI5 I use various strategies in learning science, such as preview, review, reflection, taking notes.
Instruction (P)
P1 My science teachers did not do the whole talking. They allow us to discuss and explore in the class.
P2 My science teachers frequently invited me to answer their questions.
P3 I hold the feelings appreciated and encouraged by my science teacher.
P4 My science teacher’s pedagogy draws my attention.
P5 My science teacher answers my questions on content knowledge that I don’t understand.
Textbook (TE)
TE1 The science textbook’s writing style draws my attention.
TE2 The wordings, symbols, tables, and pictures in the science textbook are easy to read and understand.
TE3 The science textbook highlights key content knowledge.
TE4 The depth of the science textbook fits my learning ability.
TE5 The breadth of the science textbook fits my learning ability.
Classroom life (CL)
CL1 I like my classmates.
CL2 My classmates keep discipline in the classroom.
CL3 My classmates cooperate well in learning science.
CL4 I played with my classmates happily after class.
CL5 My classmates and I feel proud of our class.
After the questionnaire survey, this research carried out a semi-structured interview later. It focused on dimen-
sions specified by the SLQ and with intentions to dig deeper into students’ perceptions of the influential factors
55
(pp. 50-66) ISSN 2538–7138 /Online/
of their science learning in the school context and to find reasons behind their perceptions. In the interviews, the
moderators asked students the following questions:
Q1. Are you interested in science? What are the reasons that you like/dislike it?
Q2. Do you have confidence in learning science? What is the reason for that?
Q3. Are you willing to answer questions, do hands-on and minds-on activities, and participate in group
discussion in science lessons? What is the reason for that?
Q4. To what extent your science learning gains are the results of your teacher’s instruction?
Q5. How are your feelings about the roles your classmates played in your science learning?
Q6. Are you interested in reading science textbooks? What is the reason for that?
Q7. What do you think is the main reason of students’ failure in science (question for top student group
only)? What do you think is the main reason of students’ success in learning science (question for low
science proficiency student group only)?
Following the questioning, the students were free to discuss any issue which was related to their science
learning.
Data Analysis

To verify the SLQ’s 4 factors structure, also for the needs of calculating the SLQ subscales’ reliability, the research
carried out confirmatory factor analysis (CFA) through Mplus 7.4. As far as a scale which subscales are concerned,
composite reliability (CR) is more suitable to be the estimator of a subscale’s reliability than Cronbach’s coefficient
alpha (Bentler, 2009; Raykov, 2004; Raykov & Grayson, 2003; Sijtsma, 2009). In this case, the SLQ subscales’ CR was
calculated based on Mplus’ output file of CFA. Meanwhile, the research conducted exploratory structural equation
modeling (ESEM) to further examine the SLQ’s factorial structure. Since the main purpose of this research was to
explore whether there were mechanisms of P via LI, TE, and CL to science performance respectively, the multiple
mediation analysis was performed to examine it. Besides, because it is of this research interests to explore whether
there was any gender difference in these mechanisms, the multi-group structural equation model was constructed
by Mplus 7.4 to answer it. Depending on the quantitative approach alone was not enough to explain the complex
mechanisms of P, LI, TE, and CL exerting on science performance. Semi-structured interviews were administrated for
the needs of complementing the quantitative approach. Then the interview transcripts were analyzed by a group
of five researchers. They were the authors, two team leaders of student teachers. The interviews were analyzed
according to the process of preparing, writing memos, coding, and presenting (Edmonds & Kennedy, 2017, pp.
321-331; Merriam & Tisdell, 2016, pp.196-199; Saldaña, 2013, pp. 41-52; Seidman, 2006, pp. 112-125). To ensure the
validity of data analysis, the norms of avoiding prejudice, cross-check, and using memos were adhered to (Edmonds
& Kennedy, 2017, p. 323; Merriam & Tisdell, 2016, p.208; Seidman, 2006, pp. 117-121). After reviewing the transcripts
or audio files, the members of the research team wrote memos to record reflections independently, and group
discussions were held to ensure no important messages were overlooked and the data were not contaminated by
one of the researcher’s biases or prejudice. The next steps were coding and presenting the reduced data accord-
ing to the research interests. For example, in the interests of understanding students’ classroom lives, the research
team wrote memos such as the low proficiency students never complained about the discipline problems, whereas
the top students complained about it lots. Then, they were coded as “uncooperative climate.” The findings of the
interviews were presented in the following sub-chapter of “Students’ Introspection on Their Science Learning.”
Research Results
Factorial Structure and Reliability of the SLQ

The CFA procedure was used to evaluate the extent to which the hypothesized factorial structure of the SLQ
was true. For this purpose, the cut-off criteria suggested by the literature were adopted. They are χ2/df<3 (Byrne
et al., 1989; Marsh & Hocevar, 1985), standardized root mean square residual (SRMR)<.08 (Hu & Bentler, 1999) or
SRMR<.10 (Marsh et al., 2005), root mean square error of approximation (RMSEA)<.06 (Hu & Bentler, 1999) or RM-
SEA<.08 (McDonald & Ho, 2002). The smaller the above-mentioned indices are, the better the model fits the data.
As far as Tucker-Lewis index (TLI) and comparative fit index (CFI) are concerned, TLI>.95 and CFI>.95 mean a close
56
ISSN 2538–7138 /Online/ (pp. 50-66)
model fit (Hu & Bentler, 1999; Marsh et al., 2005), while .90<TLI<.95 and .90<CFI<.95 mean a fair model fit (Brown,
2014, p. 87; Wang & Wang, 2012, pp. 18-19).
To get standardized factor loadings, the STANDARDIZED statement was specified in the OUTPUT command in
the Mplus program. SLQ subscales’ CR was calculated through the standardized factor loadings. However, the CFA
procedure restricted some factor loadings to zero, for example, P1 through P5, TE1 through TE5, and CL1 through
CL5 not loaded onto factor LI, often brought poor-fitting CFA solutions (Brown, 2014, p. 193) and overestimate fac-
tor correlations (Asparouhov & Muthén, 2009). In this case, ESEM was also conducted to provide more information
about the SLQ’s factorial structure. Table 2 shows the SLQ’s model fit indices.
Table 2
Model Fit Indices of the SLQ’s 4 Factors Structure
RMSEA
Model χ2 df χ2/df CFI TLI SRMR
Value CFit
CFA 455.13 164 2.778 .924 .912 .061 .005 .048

ESEM 265.163 116 2.286 .961 .936 .052 .366 .026
Note. N=484. CFit: test of close fit.
As table 2 shows, the CFA procedure proves that the SLQ’s 4 factors structure has an acceptable fit to the survey
data, whereas the ESEM solution gets a better fit to data. The reason for saying so was ESEM achieved better fit
indices, especially for CFit of RESEA was no longer significant (p=.366>.05). Depending on the fit indices only was
not enough to confirm the SLQ’s 4 factors structure. It was also necessary to examine whether the primary loadings
of the SLQ items were in accord with the prediction. Table 3 provides this kind of information.
Table 3
Items Factor Loadings and their z-test Results Based on ESEM
Factor 1 Factor 2 Factor 3 Factor 4

Items
Loading z-score Loading z-score Loading z-score Loading z-score
LI1 .726 13.637*** .030 .651 .108 1.901 -.056 -1.458

LI2 .755 15.465*** .026 .644 -.044 -1.264 .020 .700
LI3 .612 11.027*** .016 .300 -.039 -.912 .017 .435
LI4 .526 7.472*** -.072 -1.046 .255 3.484*** .133 2.212*
LI5 .370 6.863*** .121 1.935 .051 1.011 -.025 -.586
P1 .000 .011 .377 5.758*** .351 5.962*** .015 .379
P2 .065 1.471 .030 .748 .817 12.973*** -.035 -1.084
P3 -.077 -1.617 .033 .667 .787 11.574*** .010 .309
P4 .180 2.835* .454 5.847*** .346 4.826*** .088 1.708
P5 .203 3.184** .303 3.985*** .261 3.724*** .145 2.621**
TE1 -.021 -.538 .691 10.903*** .083 1.538 -.040 -.983
TE2 -.002 -.080 .691 13.514*** -.023 -.607 -.005 -.151
TE3 -.051 -1.322 .779 14.766*** -.034 -.962 .013 .440
TE4 .122 2.517* .604 10.606*** .035 .858 -.008 -.226
TE5 .138 2.654** .555 9.154*** .025 .574 .045 1.108
CL1 .035 1.050 .192 2.989** -.027 -.616 .556 10.621***
CL2 -.194 -2.780** -.047 -.85 .176 2.644** .643 10.862***
57
(pp. 50-66) ISSN 2538–7138 /Online/
Factor 1 Factor 2 Factor 3 Factor 4

Items
Loading z-score Loading z-score Loading z-score Loading z-score
CL3 .085 1.670 -.012 -.302 -.022 -.547 .853 15.607***

CL4 .078 1.438 .038 .589 .072 1.230 .653 10.992***
CL5 -.113 -1.937* .100 1.423 -.020 -.428 .763 12.385***
Note. N=484. * p<.05, ** p<.01, *** p<.001; z-score is determined by dividing the unstandardized parameter estimate by its
standard error (Brown, 2014, p. 128). Five largest loadings on every factor are in boldface.
As can be seen in table 3, items LI1 through LI5 have their largest loadings on factor 1, items TE1 through TE5
have their largest loadings on factor 2, and items CL1 through CL5 have their largest loadings on factor 4. There-
fore, factors 1, 2, and 4 represented the latent dimensions of LI, TE, and CL, respectively. However, items P1, P4,
and P5 had large and statistically significant loadings on TE. Since the five biggest loadings on factor 3 belonging
to items P1 through P5 and the wordings of items P1 through P5 were not similar to items TE1 through TE5, these
three cross-loadings were not a severe violation of the SLQ’s factorial structure. The cause of their cross-loadings
may root in teachers’ instruction would improve students’ perceptions of textbooks’ quality. The subscales’ CR were
computed and shown in table 4.
Table 4
Subscales’ Composite Reliability (N=484)
Composite Reliability
Subscales/Latent
Standardized loadings
dimensions
CR
LI .750, .794, .621, .559, .535 .790

P .670, .677, .602, .763, .663 .808
TE .698, .739, .769, .752, .710 .854
CL .692, .536, .765, .682, .658 .802
Table 4 shows all the latent dimensions have acceptable reliability (range of CR=.790 to .854). In summary,
tables 2 through 4 support the SLQ’s 4 factors structure. Since the SLQ was reliable and valid, it was suitable for
using the SLQ survey data to explore the holistic mechanism of P in combination with LI, TE, and CL, functioning
on students’ science performance.
Mediation Roles of Learning Initiative, Textbook, and Classroom Life

As figure 1 shows, ai estimates the effect of P on LI, TE, and CL, respectively; b1 estimates the effect of LI on
science performance holding P, TE, and CL constant. The same is true for b2 and b3. That will yield three specific
indirect effects. For example, a specific indirect effect of P on science performance through LI is a1b1. Therefore,
the total indirect effect is a1b1 + a2b2 + a3b3; the total effect of P on science performance is c = c’ + a1b1 + a2b2 + a3b3.
The result showed the parallel multiple mediation model depicted in figure 1 had acceptable fit indices. They
were χ2/df=2.87, RMSEA=.062, SRMR=.050, CFI=.916, and TLI=.901. The overall tests of significance of path coef-
ficients, indirect and direct effects are shown in table 5.
58
ISSN 2538–7138 /Online/ (pp. 50-66)
Table 5
Test of Significance of Path Coefficients, Specific Indirect Effects, Direct and Total Effect
95% CI
Standardized path
Parameter Product of path coefficients
coefficient
LL UL
a1 .676*** .603 .749

a2 .844 ***
.803 .894
Path coefficients
a3 .585*** .505 .698
b1 .557 ***
.448 .704
b2 -.137 -.275 .036
b3 -.003 -.129 .114
a 1b 1 .377*** .292 .509
Specific indirect effects
a2b2 -.116 -.233 .025
a 3b 3 -.002 -.078 .063
Direct effect c’ .003 -.150 .161
Total effect c .262*** .180 .359
Note. N=484. *** p<.001; CI=confidence interval; LL=lower limit, UL=upper limit.
According to table 5, all regression coefficients of ai path are large (a1=.676, a2=.844, a3=.585) and statistically
significant (p<.001). Two students that differ by one unit on P were estimated to differ by .676 units in their learning
initiative, .844 units in their perceptions of the textbook quality, .585 units in their perceptions of the classroom life,
respectively. The path coefficient of b1 was large (.557) and statistically significant (p<.001). It meant two students
that differ by one unit on LI were estimated to differ by .557 units on their science performance holding TE, CL,
and P constant. The latter two bi paths were negative and insignificant. For path TE→performance, two students
that differ by one unit on their perceptions of the textbook quality were estimated to differ by .137 units on their
science performance holding LI, CL, and P constant, with those who were more satisfied with the textbook had
worse performance. For path CL→performance, two students that differ by one unit on their perceptions of the
classroom life were estimated to differ by .003 units on their science performance holding LI, TE, and P constant,
with those who were more satisfied with the classroom life had worse performance.
This model had three specific indirect effects. The first indirect effect of P on performance was modeled through
LI, estimated as .377 and statistically significant. Students who were more satisfied with their teachers’ instruction
getting more learning initiative than those less satisfied with their teachers’ instruction (a1 =.676), which in turn
was positively related to the promotion of their science performance (b1 =.557). The second indirect effect of P on
performance was modeled through TE and statistically insignificant. Two students that differ by one unit on their
perceptions of teacher’s instruction were estimated to differ by .116 units on science performance, with those
more satisfied with their teacher’s instruction had worse performance (because b2 is negative). The third indirect
effect of P on performance was modeled through CL, estimated as -.002. Although two negative specific indirect
effects (P→TE→performance and P→CL→performance), which came from the negative path coefficients of b2
and b3, were statistically insignificant, they should not be ignored. The potential reasons for these negative effects
will be discussed later in more detail.
The direct effect of P on performance was c’ = .003. It estimates the amount by which two students that differ
by one unit on P differ on their performance holding all mediator constant. Since it was trivial (.003) and statistically
insignificant (p=.967), the effect of P on performance was completely mediated by the mediators. The total indirect
effect was positive, it was .259. The total effect of P on performance was .262. The total indirect effect accounted
for 98.8% of the effect of P on performance.
Testing Invariance of the Structural Path Coefficients and Indirect Effects across Female and Male Groups

For this research question 2, it required to test whether the three ai path coefficients, three bi path coefficients,
and the direct effect of P on performance was invariance among students in the female group and male group.
59
(pp. 50-66) ISSN 2538–7138 /Online/
For this research question 3, it required to test whether the indirect effect of P on performance through LI, TE, and
CL respectively, as well as the total effect of P on the performance, were invariant among students in the female
group and male group. If these path coefficients and effects are invariant between female and male groups, then
gender is not a moderator.
A baseline structural equation model (SEM) for male and female groups respectively needs to be established
first, so that the multi-group SEM modeling can be carried out later. Based on the information coming from modifi-
cation indices (MI) for fixed parameters specified in the baseline model, there was only one error covariance having
a MI larger than 10 in both male and female groups. In this case, the error covariance associated with items P3 and
P2 which had the largest MI (15.800 in the male group, 22.458 in the female group) was set free estimated in base-
line SEM models. It was found that the baseline models fit data well. The fit indices for the female group (n=218)
were χ2/df=1.281, RMSEA=.036, SRMR=.054, CFI=.945, TLI=.936. The close fit test did not reject the null hypothesis
of RMSEA<=.05 in this group (p=.965). The fit indices for the male group (n=266) were χ2/df =1.415, RMSEA=.039,
SRMR=.058, CFI=.927, TLI=.916. The close fit test also did not reject the null hypothesis of RMSEA<=.05 (p=.947).
Now that the baseline models fit data well, to test the invariance of structural path coefficients, indirect effects,
and the total effect across two groups, an unrestricted SEM model using male and female samples simultaneously
was established. Results of free estimated path coefficients, indirect effects, and total effect can be seen in table 6.
Table 6
Testing the Invariance of Structural Path Coefficients, Indirect Effects and Total Effect across Female and Male Groups
Standardized estimates Wald testa
Parameter Female group Male group χ2 p
Structural path coefficient

a1 .737*** .616*** 4.497* .034
a2 .881 ***
.837 ***
Not needed
a3 .620*** .603*** Not needed
b1 .676*** .509*** .609 .435
b2 .078 -.189 Not needed
b3 .061 -.090 Not needed
c ’
-.300 .113 2.182 .140
Indirect/Total effect
a 1b 1 .498*** .313*** 2.748 .097
a 2b 2 .069 -.158 1.503 .220
a 3b 3 .038 -.054 Not needed
c .305*** .214** .901 .343
Note. nfemale=218. nmale=266. * p<.05, ** p<.01, *** p<.001. a. df =1.
The unrestricted SEM model had an acceptable fit to data: χ2/df=1.410, RMSEA=.041, close-fit test p=.971,
SRMR=.066, CFI=.918, TLI=.913. All ai path coefficient, as well as b1, were positive and statistically significant in
both male and female groups. They showed the same tendencies of antecedent variables on consequent variables
(P→LI, P→TE, P→CL, LI→performance), but it seemed that the discrepancies of two path coefficients (i.e., a1, b1)
between male and female groups were large. The remaining three path coefficients showed different tendencies
across groups. For example, the path coefficient of c’, also known as the direct effect, was positive in the male group
and negative in the female group. The discrepancy of c’ across groups was also large. Then three restricted SEM
models were established to test the invariance of a1, b1, and c’ across groups. The MODEL TEST command in the
Mplus program was used to provide a Wald χ2 testing information with df =1. It can be seen in table 6, the path coef-
ficients of a1 in different groups are not identical to each other (χ2=4.497, p=0.034). It means P was more effective
in igniting girls’ LI than boys. In other words, the direct effect of P on LI was moderated by gender. The remaining
two Wald tests on the invariance of b1 and c’ across groups did not reject the null hypothesis: χ2=.609 (p =.435) for b1
60
ISSN 2538–7138 /Online/ (pp. 50-66)
and χ2=2.182 (p=.140) for c’. Since b1 and c’ were invariant across groups, it was not necessary to test the invariance
of a2, a3, b2, and b3 across groups, as they had a smaller discrepancy across male and female groups than b1 and c’.
As far as the indirect effects were concerned, the Wald test was used to examine the invariance of the specific
indirect effects that P on performance through LI, as well as P on performance through TE. To be specific, the MODEL
CONSTRAINT command set the indirect effect, accompanied by the MODEL TEST command to provide a Wald test.
Both the specific effect of P→LI→performance (χ2=2.748, p=.097) and the specific effect of P→TE→performance
(χ2=1.503, p=.220) were insignificant. There was not necessary to test the invariance of the remaining specific in-
direct effect (P→CL→performance) before accepting the null hypothesis, as this indirect effect had the smallest
discrepancy between male and female groups. Testing for the invariance of the total effect of P on performance also
resulted in accepting the null hypothesis (χ2=.901, p=.343). Although it was not the main purpose of this section,
it should not be overlooked that the total effect of P on performance in both groups was statistically significant.
Students’ Introspection on Their Science Learning

The negative path coefficients, such as b2, b3 in male students baseline SEM model and c’ in female students
baseline SEM model, as well as b2, b3 in the parallel mediation model, could be explained by the SLQ survey data.
A negative estimate reflected the inconsistency between the antecedent variable and the consequent variable.
For example, it was found that only 87 female students (39.9%, n=218) whose assessment levels on the factor P
were suitable to their levels of performance. The others thus contributed to explaining the cause of the negative
path coefficient of c’. However, depending on this technical analysis alone was not enough to answer this research
concerns. The following issues are also of this research interests. What characteristics deeply hide behind classroom
life and textbooks to impede them from acting as positive influential factors to students’ science performance?
What are the intrinsic defects in direct instruction to weaken its strength on students’ learning outcomes? Answers
to these questions are very important because it could complement some missing pieces to the holistic science
performance picture. The following statements covered some relevant information obtained from interviews.
Q1. All students were interested in biology because it had close relations with the human body and daily life. In the meantime,
all students were not interested in the difficult part of the science course. Low proficiency students were also interested in
observing or doing experiments. Students in group L2 were interested in chemistry because their teacher did many chemistry
experiments. They also indicated that physics was the most difficult part. Students in group L1 often got nervous in science
lessons because the science teacher was strict with them.
Q2. All top students had confidence in learning science. They said cleverness was the source of their confidence. Almost
all students in group L2, and 5 students (35.7%, n=14) in group L1, lost their confidence in learning science. 9 students in
group L1 (64.3%, n=14) had a little confidence in science because they could understand the easy part of the science course.
Q3. For the top students, the average time they stayed focus on science tasks was three-quarters of a class. Three-quarters
of them were not willing to answer questions as they may make mistakes. For the low proficiency students, the minimum
time they focused on science tasks was 5 minutes, with the whole lecture lasted 40 minutes. They were bored in the lecture
that the teacher did most of the talking and rarely did experiments. In this context, even top students complained that they
felt exhausted when the lesson was over. However, when there was an experiment, the low proficiency students were willing
to engage in it, even having discussions with their teachers, which they rarely did in other contexts. Compared to the top
students, the low proficiency students were more reluctant to answer questions for the same reason.
The first three interview questions focused on students’ learning initiatives. In sum, no matter what levels
they got in science performance, they were all fond of biology and not interested in the difficult part of the science
course. Their confidence in learning science was hurt badly by the difficult part of the science course. Cleverness was
the only factor put forward by the top students to account for their confidence in learning science. Meanwhile, the
low proficiency students were absent from the science lesson in most of the time unless there was an experiment.
Q4. Top students appreciated their teacher’s instruction lots. They acknowledged that if they had studied by themselves,
they would not have gained much in science. Students in group L1 agreed that the extent to which the science teacher
contributing to their studying was about half of all gains. Students in group L2 considered that the science teacher gave
little help to their learning. Though he taught well, the previous science teacher had taught bad, thus they did not prepare
well for the current studying.
Q5. Students in group L1 took top students as obstacles to their science learning. They felt upset for not as clever as the top
students. Students in group L2 hold the same attitudes toward the top students, they complained top students for caus-
ing the pace of the science lesson too fast. Moreover, the top students did not help them in learning science. Top students
appreciated each other for their endeavor to construct an atmosphere of studying hard. They complained that the low
61
(pp. 50-66) ISSN 2538–7138 /Online/
proficiency students chatting in the lecture always broke their thoughts. Besides, some of their classmates were chasing
and shouting during the break time, which impeded them to write the homework.
Q6. Students in group T1 and half of the students in group L1 were interested in reading the science textbook. The biological
knowledge, experiments, and pictures in the textbook attracted their attention. Yet all students in group L2 reported that
they were not interested in reading the textbook. Students in group T1 suggested to mark key SMK in the text and connect
SMK to real life to make it more comprehensible to them. Besides, they thought electromagnetic induction was too abstract
to understand. No matter how hard they had worked on this topic, they cannot resolve problems related to electromagnetic
induction. Students in group L1 put forward that textbooks should have more experiments and pictures to draw their attention.
Q7. The top students summarized three main reasons to account for the low proficiency students’ failure in a science course,
i.e., not smart, not following teachers’ instructions, and copying homework. By contrast, the main reasons proposed by
the low proficiency students to explain the top students’ success in science were learning well in previous science courses,
studying tirelessly, and accommodating to teacher’s pedagogy.
It can be seen from students’ responses to Q4 and Q7, students acknowledged that the science teacher con-
tributed greatly to the top students’ learning. By contrast, the instruction did not have much influence on the low
proficiency students’ learning. They were not suited to the teacher’s pedagogy, and the pace of the science lesson
was too fast to follow. Students’ responses to Q5 provided useful information on the CL dimension. It seemed that
the top students and the low proficiency students did not support each other in learning science, whereas acting
as an obstacle to impede another group’s students to improve their gains. The top students complained a lot about
the discipline issues in the classroom, while the low proficiency students said nothing about it. Students’ responses
to Q6 provided useful information on the TE dimension. The top students more frequently read the textbook than
the low proficiency students. Advice proposed by the top students related to the textbook was mainly about mak-
ing the text more convenient for them to understand the science content, whereas the low proficiency students
required more pictures and experiments in the text to ignite their interests in science.
Discussion
For this research question 1, it turns out that LI is the only significant mediator. Then the focus group in-
terviews were conducted to find the underlying causes for this mechanism. The top students reported that the
science teacher gave them lots of help in learning science. Meanwhile, the low proficiency students put forward
“adapting to the science teachers’ instruction” as one main reason to account for the top students’ success. Even the
low proficiency students could stay focus on science lessons where there was an experiment or having relations
with real life. Therefore, it seems the effect of P on performance would not come to fruition unless the instruction
ignites students’ learning initiative (LI). That is the primary mechanism of direct instruction affecting students’ sci-
ence learning. Direct instruction may not be necessary an “inferior pedagogy,” so long as it could ignite students’
learning initiatives.
Some strategies suggested by previous research again emphasized by students in this research, such as
relating science content with real-life (Hulleman & Harackiewicz, 2009) and accommodating teachers’ instruction
to students’ learning levels (Kremer et al., 2013). In students’ viewpoints, relating science content with real life as
well as teachers’ experimental demonstrations will make SMK more comprehensible. Students also complained
that they could never learn some difficult SMK, such as electromagnetic induction. Now that this kind of SMK is
beyond lower secondary school students’ learning ability, it may be reasonable to transfer them into the upper
secondary school science course. Moreover, although previous research had found that students learned less
with teachers talking more (Setati et al., 2002), this research found this argument may not be suitable for clever
students. However, much attention should be paid to low proficiency students. They do not adapt to the content-
heavy lessons where the pace is too fast to follow. As they do not understand science contents well, lots of them
are also not interested in reading textbooks.
In practice, some people argue that students’ success owes to their teachers little, as opposed to this opinion,
others believe instruction is the main reason account for students’ success. The research could not agree with these
opinions. The former only pays attention to the insignificant direct effect of c’ path. It may underestimate teach-
ers’ efficiency in prompting students’ learning initiatives that, in turn, improve students’ science performance. The
latter could not differentiate the specific indirect effect of a1b1 from the direct effect of c’, in other words, a1b1 is
misunderstood as c’. In that case, it may underestimate the importance of adapting instruction to students’ needs
to ignite students’ learning initiatives. Therefore, both of them are contrary to the mechanism that the effect of P
on performance is completely mediated by LI.
62
ISSN 2538–7138 /Online/ (pp. 50-66)
P has significant direct effects on TE and CL. It means P actually communicates with TE and CL, in this light, it is
reasonable for the mediation analysis to adopt ID models as its theoretic framework. However, the structural path
coefficients of b2 and b3 are insignificant, which also causes insignificant indirect effects of a2b2 and a3b3. It might
indicate that the science teachers’ instruction can compromise to TE and CL rather than producing fundamental
alterations in science content presentation and classroom life. In compromising with TE and CL, P could improve
students’ satisfaction with the textbook and the classroom life to some extent, and this leads to the significant
path coefficients of a2 and a3. Since the intrinsic defects of TE and CL have not been fundamentally changed, the
effects of TE on performance, and CL on performance, are destroyed. As far as the textbook was concerned, the
inherent flaws included not connecting to real life, lacking experiments and interesting pictures, having some
difficult content knowledge, and so on. As far as the classroom life was concerned, the low proficiency students
complained that the top students did not give them practical help in learning science, the top students complained
that chaos in classrooms always broke their thoughts. Therefore, students’ uncooperative attitude may cause an
insignificant direct effect of CL on performance.
For this research questions 2 and 3, only the relation of P on LI has gender difference. To be specific, the
instruction is more effective in promoting girls’ learning initiatives because the Wald test on the invariance of a1
across groups rejected the null hypothesis. But what caused this difference between groups? Previous studies had
found that the secular humanist society usually viewed women as responsible for housekeeping (Normile, 2006;
Saujani, 2017). This gender stereotype may explain girls’ tendencies of doing things quietly and gently. Compared
to boys, they are more likely to cooperate with their science teachers and therefore have a more positive reaction
to the science teachers’ instruction.
Conclusions
Through mediation analysis, this research shows instruction has a significant influence on students’ science
performance. The total effect of instruction on performance in both male and female groups are statistically sig-
nificant. Meanwhile, the impact of instruction on performance is completely mediated by students’ learning initia-
tives. It exposes that even in the context of direct instruction, teachers cannot play as authorities, but should be
facilitators. The higher instruction prompting students’ learning initiatives, the more it gains in desired outcomes.
Both boys and girls can succeed in the conditions that they have learning initiative. In terms of gender difference,
it was found by multi-group SEM, although girls are more sensitive to teachers’ pedagogy than boys, the strength
of instruction on performance is not significantly higher than boys.
This research emphasizes the impact of instruction on performance can be improved by the visualization
of science contents. Students required more pictures and experiments in textbooks, as well as more experiment
demonstrations in science lessons. Students believe in what they see in their eyes. This kind of materials and class-
room activities build an effective delivery system to make science concepts comprehensible. In this case, students
can stay focus on science tasks.
This research finds the impact of instruction on performance is impaired by students’ uncooperative class-
room life. Teachers’ beliefs about teaching shaped the classroom climate. In an exam-centric education system and
content-heavy classrooms, discipline problems exist to some degree. It is not only a signal of students’ revolt to
tedious instruction but also a signal of students’ dissatisfactions that the top students do not help them to learn
science. In this classroom context, students’ classroom life cannot play as a channel of communication between
instruction and performance. This research also finds the impact of instruction on performance is impaired by some
defects of textbooks. Those defects account for students losing their interests and confidence in learning science.
Further research can explore the mechanisms of instruction exerting on science performance in an inquiry teaching
context or explores whether family factors (e.g., parents’ support) moderate these mechanisms.
Acknowledgements
The authors would like to thank Ling-hui Liang and Chang-guan Li, the city’s educational officials, for their
help with developing the school life questionnaire, as well as the support of the National Natural Science Founda-
tion of China [Grant number 11975012].
63
(pp. 50-66) ISSN 2538–7138 /Online/
References
Akinoglu, O., & Yasar, Z. (2007). The effects of note taking in science education through the mind mapping technique on
students’ attitudes, academic achievement and concept learning. Journal of Baltic Science Education, 6(3), 34-43.
http://oaji.net/articles/2014/987-1404288606.pdf
Allen, J. P., Pianta, R. C., Gregory, A., Mikami, A. Y., & Lun, J. (2011). An interaction-based approach to enhancing secondary school
instruction and student achievement. Science, 333(6045), 1034-1037. https://doi.org/10.1126/science.1207998
Aşkın, O. E., & Öz, E. (2020). Cross-national comparisons of students’ science success based on gender variability: Evidence from
TIMSS. Journal of Baltic Science Education, 19(2), 186-200. https://doi.org/10.33225/jbse/20.19.186
Asparouhov, T., & Muthén, B. (2009). Exploratory structural equation modeling. Structural Equation Modeling, 16(3), 397–438.
https://doi.org/10.1080/10705510903008204
Auerbach, C. F., & Silverstein, L. B. (2003). Qualitative data: An introduction to coding and analysis. New York University Press.
Banilower, E. R., Smith, P. S., Weiss, I. R., Malzahn, K. A., Campbell, K. M., & Weis, A. M. (2013). Report of the 2012 national survey
of science and mathematics education. Horizon Research. http://www.horizon-research.com/2012nssme/wp-content/
uploads/2013/02/2012-NSSME-Full-Report1.pdf
Bao, L., Cai, T., Koenig, K., Fang, K., Han, J., Wang, J., Liu, Q., Ding, L., Cui, L. L., Luo, Y., Wang, Y. F., Li, L. M., & Wu, N. L. (2009). Learning
and scientific reasoning. Science, 323(5914), 586-587. https://doi.org/10.1126/science.1167740
Beaman, L., Duflo, E., Pande, R., & Topalova, P. (2012). Female leadership raises aspirations and educational attainment for girls:
A policy experiment in India. Science, 335(6068), 582-586. https://doi.org/10.1126/science.1212382
Bentler P. M. (2009). Alpha, dimension-free, and model-based internal consistency reliability. Psychometrika, 74(1), 137–143.
https://doi.org/10.1007/s11336-008-9100-1
Brophy, J. (2006). History of research on classroom management. In C. M. Evertson & C. S. Weinstein (Eds.), Handbook of classroom
management: Research, practice, and contemporary issues (pp. 17–43). Lawrence Erlbaum Associates Publishers.
Brown, T. A. (2014). Confirmatory factor analysis for applied research. Guilford Publications.
Burušić, J., Šimunović, M., & Šakić, M. (2019). Technology-based activities at home and STEM school achievement: The
moderating effects of student gender and parental education. Research in Science & Technological Education, 37(4), 1-22.
https://doi.org/10.1080/02635143.2019.1646717
Byrne, B. M., Shavelson, R. J., & Muthén, B. (1989). Testing for the equivalence of factor covariance and mean structures: The issue
of partial measurement invariance. Psychological Bulletin, 105(3), 456-466. https://doi.org/10.1037/0033-2909.105.3.456
Change the equation: Improving science and mathematics education in the US [Editorial]. (2011). Nature Cell Biology, 13(8), 875.
https://doi.org/10.1038/ncb2318
Cobern, W. W., Schuster, D., Adams, B., Applegate, B., Skjold, B., Undreiu, A., Loving, C. C., & Gobert, J. D. (2010). Experimental
comparison of inquiry and direct instruction in science. Research in Science & Technological Education, 28(1), 81-
96. https://doi.org/10.1080/02635140903513599
Coll, R. K., Dahsah, C., & Faikhamta, C. (2010). The influence of educational context on science learning: A cross-national analysis
of PISA. Research in Science & Technological Education, 28(1), 3-24. https://doi.org/10.1080/02635140903513532
Cowles, E. L., & Nelson, E. (2015). An introduction to survey research. Business Expert Press.
DeVellis, R. F. (2017). Scale development: Theory and applications (4th ed.). SAGE Publications.
Dick, W., Carey, L., & Carey, J. O. (2015). The systematic design of instruction (8th ed.). Pearson.
Dillman, D. A. (2009). Internet, mail and mixed-mode surveys: The tailored design method. John Wiley & Sons.
Djigic, G., & Stojiljkovic, S. (2011). Classroom management styles, classroom climate and school achievement. Procedia-Social
and Behavioral Sciences, 29, 819-828. https://doi.org/10.1016/j.sbspro.2011.11.310
Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students’ learning with effective
learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public
Interest, 14(1), 4-58. https://doi.org/10.1177/1529100612453266
Edmonds, W. A., & Kennedy, T. D. (2017). An applied guide to research designs: quantitative, qualitative, and mixed methods (3rd
ed.). SAGE Publications.
Fine, C., & Elgar, M. A. (2017). Promiscuous men, chaste women and other gender myths. Scientific American, 317(3), 32-37.
https://doi.org/10.1038/scientificamerican0917-32
Fowler, F. J., Jr. (1995). Improving survey questions: Design and evaluation. SAGE Publications.
Fraser, B. J. (1987). Classroom learning environments and effective schooling. Professional School Psychology, 2(1), 25-41.
https://doi.org/10.1037/h0090526
Freeman, S., Eddy, S. L., McDonough, M., Smith, M. K., Okoroafor, N., Jordt, H., & Wenderoth, M. P. (2014). Active learning increases
student performance in science, engineering, and mathematics Proceedings of the National Academy of Sciences, 111(23),
8410-8415. https://doi.org/10.1073/pnas.1319030111
Gagne, R. M., Wager, W. W., Golas, K. C., Keller, J. M., & Russell, J. D. (2005). Principles of instructional design. Performance Improvement,
44(2), 44-46. https://doi.org/10.1002/pfi.4140440211
Gieve, S., & Miller I.K. (2006). What do we mean by ‘quality of classroom life’? In S. Gieve & I. K. Miller (eds.), Understanding the
language classroom (pp. 18-46). Palgrave Macmillan. https://doi.org/10.1057/9780230523166_3
Guiso, L., Monte, F., Sapienza, P., & Zingales, L. (2008). Culture, gender, and math. Science, 320(5880), 1164-
1165. https://doi.org/10.1126/science.1154094
64
ISSN 2538–7138 /Online/ (pp. 50-66)
Gupta, M. D. (2017). Return of the missing daughters. Scientific American, 317(3), 80-85.
https://doi.org/10.1038/scientificamerican0917-80
Hanbay, O. (2015). Student survey on the textbook ‘DEUTSCH IST SPITZE! A1. 1’. International Journal of Language Academy, 3(1),
319-324. https://doi.org/10.18033/ijla.208
Hesse-Biber, S. N. (2010). Mixed methods research: Merging theory with practice. The Guilford Press.
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new
alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55. http://doi.org/10.1080/10705519909540118
Hulleman, C. S., & Harackiewicz, J. M. (2009). Promoting interest and performance in high school science classes. Science, 326(5958),
1410-1412. http://doi.org/10.1126/science.1177067
Johnson, B., Sullivan, A. M., & Williams. D. (2009). A one-eyed look at classroom life: Using new technologies to enrich classroom-
based research. Issues in Educational Research, 19(1), 34-47.
Khalil, M. K., & Elkhider, I. A. (2016). Applying learning theories and instructional design models for effective instruction. Advances
in Physiology Education, 40(2), 147-156. http://doi.org/10.1152/advan.00138.2015
Kremer, M., Brannen, C., & Glennerster, R. (2013). The challenge of education and learning in the developing world. Science,
340(6130), 297-300. http://doi.org/10.1126/science.1235350
Ledford, B. R., & Sleeman, P. J. (2000). Instructional design: A primer. Information Age Publishing.
Lim, W. K. (2010). Asian education must change to promote innovative thinking. Nature, 465(7295), 157. http://doi.org/10.1038/465157a
Machin, S., & Pekkarinen, T. (2008). Global sex differences in test score variability. Science, 322(5906), 1331-1332.
http://doi.org/10.1126/science.1162573
Marsh, H. W., Hau, K.-T., & Grayson, D. (2005). Goodness of fit in structural equation models. In A. Maydeu-Olivares & J. J. McArdle
(Eds.), Multivariate applications book series. Contemporary psychometrics: A festschrift for Roderick P. McDonald (pp. 275–340).
Lawrence Erlbaum Associates Publishers.
Marsh, H. W., & Hocevar, D. (1985). Application of confirmatory factor analysis to the study of self-concept: First- and higher order factor
models and their invariance across groups. Psychological Bulletin, 97(3), 562–582. https://doi.org/10.1037/0033-2909.97.3.562
Matsumura, L. C., Slater, S. C., & Crosson, A. (2008). Classroom climate, rigorous instruction and curriculum, and students’
interactions in urban middle schools. The Elementary School Journal, 108(4), 293-312. https://doi.org/10.1086/528973
McDonald, R. P., & Ho, M.-H. R. (2002). Principles and practice in reporting structural equation analyses. Psychological Methods,
7(1), 64-82. https://doi.org/10.1037/1082-989X.7.1.64
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research: A guide to design and implementation (4th ed.). John Wiley & Sons.
Normile, D. (2006). Getting women scientists back on the career track in Japan. Science, 311(5765), 1235-1236.
https://doi.org/10.1126/science.311.5765.1235
Oates, T. (2014). Why textbooks count: A policy paper. University of Cambridge. https://www.cambridgeassessment.org.uk/
Images/181744-why-textbooks-count-tim-oates.pdf
Organisation for Economic Co-Operation Development. (2016). PISA 2015 results in focus. OECD. https://doi.org/10.1787/aa9237e6-en
Organisation for Economic Co-Operation Development. (2019a). PISA 2018 results (volume I): What students know and can do.
OECD. https://doi.org/10.1787/5f07c754-en
Organisation for Economic Co-Operation Development. (2019b). PISA 2018 results (volume Ⅱ): Where all students can succeed.
OECD. https://doi.org/10.1787/b5fd1b8f-en
Pierce, C. (1994). Importance of classroom climate for at-risk learners. The Journal of Educational Research, 88(1), 37-42.
https://doi.org/10.1080/00220671.1994.9944832
Raykov, T. (2004). Behavioral scale reliability and measurement invariance evaluation using latent variable modeling. Behavior
Therapy, 35(2), 299-331. https://doi.org/10.1016/S0005-7894(04)80041-8
Raykov, T., & Grayson, D. A. (2003). A test for change of composite reliability in scale development. Multivariate Behavioral Research,
38(2), 143-159. https://doi.org/10.1207/S15327906MBR3802_1
Reichenberg, M. (2016). Explaining teachers’ use of textbooks. Journal of Educational Media, Memory, and Society, 8(2), 145-159.
https://doi.org/10.3167/jemms.2016.080208
Reigeluth, C. M. (1999). What is instructional-design theory and how is it changing? In C. M. Reigeluth (Ed.), Instructional-design
theories and models: A new paradigm of instructional theory (Vol.2, pp. 5-29). Lawrence Erlbaum Associates Publishers.
Robinson, S. (2011). Student use of a free online textbook. Academy of Educational Leadership Journal, 15(3), 1-10.
https://www.questia.com/library/journal/1G1-263157454/student-use-of-a-free-online-textbook
Rushton, S. P. (2000). Student teacher efficac y in inner- city schools. The Urban Review, 32(4), 365-383.
https://doi.org/10.1023/A:1026459809392
Sachs, J., & Mellor, L. (2005). ‘Child panic’, risk and child protection: An examination of policies from New South Wales and
Queensland. Journal of Education Policy, 20(2), 125-140. https://doi.org/10.1080/0268093052000341368
Saldaña, J. (2013). The coding manual for qualitative researchers (2nd ed.). SAGE Publications.
Saujani, R. (2017). Girl code. Scientific American, 317(3), 66-69. https://doi.org/10.1038/scientificamerican0917-66
Sawada, D., Piburn, M. D., Judson, E., Turley, J., Falconer, K., Benford, R., & Bloom, I. (2002). Measuring reform practices in science
and mathematics classrooms: The reformed teaching observation protocol. School Science and Mathematics, 102(6), 245-
253. https://doi.org/10.1111/j.1949-8594.2002.tb17883.x
Science education reforms in the UK [Editorial]. (2012). Nature Cell Biology, 14(10), 977. https://doi.org/10.1038/ncb2601
Seidman, I. E. (2006). Interviewing as qualitative research: A guide for researchers in education and the social sciences (3rd ed.).
Teachers College Press.
65
(pp. 50-66) ISSN 2538–7138 /Online/
Setati, M., Adler, J., Reed, Y., & Bapoo, A. (2002). Incomplete journeys: Code-switching and other language practices in
mathematics, science and English language classrooms in South Africa. Language and Education, 16(2), 128-149.
https://doi.org/10.1080/09500780208666824
Sijtsma, K. (2009). On the use, the misuse, and the very limited usefulness of Cronbach’s alpha. Psychometrika, 74(1), 107–120.
https://doi.org/10.1007/s11336-008-9101-0
Sotáková, I., Ganajová, M., & Babinčáková, M. (2020). Inquiry-based science education as a revision strategy. Journal of Baltic
Science Education, 19(3), 499-513. https://doi.org/10.33225/jbse/20.19.499
The science of education reform [Editorial]. (2006). Nature Neuroscience, 9(11), 1345. https://doi.org/10.1038/nn1106-1345
Wang, J. C., & Wang, X. Q. (2012). Structural equation modeling: Applications using Mplus. Higher Education Press; John Wiley &
Sons Commercial Service (Beijing).
Weaver, D., Dick, T., Higgins, K., Marrongelle, K., Foreman, L., & Miller, N. (2005). OMLI classroom observation protocol. RMC Research
Corporation. https://scaleupstem.ncsu.edu/OMLIClassroomObservationProtocols.pdf
Webb, P. (2010). Science education and literacy: Imperatives for the developed and developing world. Science, 328(5977), 448-
450. https://doi.org/10.1126/science.1182596
Weiss, I. R., Banilower, E. R., & Shimkus, E. S. (2004). Local systemic change through teacher enhancement: Year nine cross-site report.
Horizon Research. http://www.horizon-research.com/horizonresearchwp/wp-content/uploads/2013/04/cross-site03.pdf
Wilkens, H. J. (2011). Textbook approval systems and the program for international assessment (PISA) results: A preliminary
analysis. IARTEM e-Journal, 4(2), 63-74. http://biriwa.com/iartem/ejournal/volume4.2/papers/Paper4_Wilkens_
ApprovalSystemsandPISA_Final_Vol4No2.pdf
Received: September 14, 2020 Accepted: January 08, 2021
Cite as: Jiang, T., Chen, J.-G., & Wu, Y.-Y. (2021). Impact of instruction on science performance: Learning initiative as a mediator
and gender as a limited moderator. Journal of Baltic Science Education, 20(1), 50-66. https://doi.org/10.33225/jbse/21.20.50
Tao Jiang PhD, Professor, Department of Physics, School of Electronics &

(Corresponding author) Information Engineering, Taizhou University, 1139 Shifu avenue,
Taizhou 318000, China.
Ji-gen Chen PhD, Professor, Department of Materials Engineering, School of
Pharmaceutical and Materials Engineering, Taizhou University,
Taizhou 318000, China.
Yin-yin Wu PhD, Associate Professor, Department of Primary Education,
College of Teacher Education, Taizhou University, 605 Dongfang
avenue, Linhai 317000, China.
66
STUDENT, PARENT AND
TEACHER PERCEPTIONS
ON THE BEHAVIORAL ISSN 1648-3898 /Print/
CHARACTERISTICS OF ISSN 2538-7138 /Online/
SCIENTIFIC CREATIVITY
AND THE IMPLICATIONS
TO ENHANCE STUDENTS’
Abstract. Teachers’ and parents’ percep-
SCIENTIFIC CREATIVITY tions of scientific creativity are assumed
to be an important environmental factor
for scientific creativity, so this research
surveyed their perceptions of the behavio-
ral characteristics of scientific creativity and
Insun Lee, compared their perceptions to those of stu-
Jongwon Park dents. This is achieved with a list of behav-
ioral characteristics of creative physicists
during their growth period. For this survey,
48 science teachers, 112 parents, and 145
science gifted students participated. Out
of the 30 items of the list, they selected 10
Introduction items that were considered important indi-
cators to become creative scientists in the
There are many aspects to consider for the scientific creativity of stu- future, and they ranked them according to
dents, including cognitive, affective, attitude, and environmental influences. their importance. The results showed that
Of these, this research focuses on the environmental aspects that encour- the three groups all perceived ‘conducting
age scientific creativity. Since parents’ and teachers’ perception of scientific experiments, asking questions, thinking
creativity are parts of this environmental aspect, this research has examined logically to solve difficult problems, and
their perceptions on scientific creativity. sharing ideas’ as important for scientific
Many studies have emphasized the importance of the right environment creativity. For the items that were perceived
to encourage creativity1 (Cole et al., 1999; Csikszentmihalyi, 1996; Peter- to be less important, it was discussed why
Szarka, 2012; Sternberg & O’hara, 1999). For instance, Davies et al. (2013) these items might be necessary for scientific
conducted a literature review and summarized the conditions needed to creativity. Comparative result showed that
encourage creativity with eight items, such as the availability of appropri- parents gave more importance to learning-
ate materials, work outside the classroom/school, opportunities for peer related aspects, teachers to thinking-
collaboration, and so on. In more detail, de Souza Fleith (2000) proposed related aspects, and students to activities
a classroom environment that enhances creativity with 16 items in four such as making and experimenting. This
categories: teachers’ attitudes, strategies, activities and educational system. research showed the behavioral character-
Richardson and Mishra (2018) developed practical guidelines to build a cre- istics that should be encouraged to improve
ative learning environment consisting of 14 items in three categories: physical students’ scientific creativity at school and
environment, learning climate and learner engagement. These environmental at home.
conditions that have been suggested for creativity include a need for the Keywords: behavioral characteristics,
teacher to properly understand creativity. For example, teachers should be creativity perception, creative environment,
able to accept new ideas, support taking risks in the classroom (Richardson scientific creativity
& Mishra, 2018), and provide flexible directions and open-ended activities
(de Souza Fleith, 2000). Beghetto and Kaufman (2014) also discussed what
teachers should understand about creativity and what efforts should be Insun Lee
made to develop students’ creativity. Chungbuk National University, South Korea
Jongwon Park
Chonnam National University, South Korea
1 In this paper, general creativity is written as just ‘creativity’, and it is written as ‘scientific
creativity’ only in the case of scientific creativity.
67
CREATIVITY ISSN 2538–7138 /Online/
(pp. 67-79)
This research started with the assumption that the perception or awareness of scientific creativity by teachers,
parents and students plays an important role as an environment to encourage and guide the students’ scientific
creativity.
One interesting feature found in research of how teachers, parents, and students perceive creativity is that
there are often some misconceptions of creativity. According to Guncer and Oral (1993), students considered
non-compliant by Turkish teachers showed higher score on the TTCT creativity test, and Aljughaiman and Mueller-
Reynolds (2005) also said that students with tendencies that teachers disliked were often highly creative students.
Fryer and Collings (1991) noted that 70% of surveyed teachers in England and Wales regarded creativity as a rare
talent, and Jaba et al. (2009) also found that only 14% of surveyed college students saw creativity as a charac-
teristic that everyone had, and 78% of the students saw it as a characteristic that only a few had. Gralewski and
Karwowski (2018) said that some teachers perceived creative students as students who were self-controlled and
disciplined, not as students who could solve inventive or open problems well. Diakidoy and Kanari (1999) reported
that student-teachers believed creativity was only found in the arts, and creative products were new but not nec-
essarily appropriate or right. Aljughaiman and Mowrer-Reynolds (2005) also said that only 15% of the teachers
correlated divergent thinking with creativity. These findings indicate that people’s perceptions of creativity may
differ from those of experts.
Similarly, perceptions of the behavioral characteristics of scientific creativity may vary among teachers, parents
and students, and this difference in perception may serve as a hindrance to fostering students’ scientific creativity.
For example, if students think that activities such as ‘making’ are important for scientific creativity, but teachers
and parents think those activities are not very important, then the ‘making’ that students want to do may not be
supported by their teachers or parents.
Research Aim
The basic aim of this research is to examine and compare the perceptions of the behavioral characteristics of
scientific creativity by three groups: teachers, parents and students. The behavioral characteristics that are used
are obtained from a list of the behavioral characteristics of creative physicists (BCCP) developed by Shin and Park
(2020). This list consists of 30 items across eight categories, it is based on the actual behavioral instances of six
physicists during their growth period, and it showed a high reliability and validity by applying it to scientifically
gifted upper secondary school students. This list was developed based on the behavioral characteristics shown
by creative physicists in their childhoods, and it described observable behavioral characteristics. Therefore, it was
expected for parents, students, and teachers to be able to use the list easily to identify the behavioral characteristics
of students. The detailed aims of this research are:
•• To highlight the BCCP items that teachers, parents and students consider important.
•• To analyze the differences in the perception of behavioral characteristics of scientific creativity among
teachers, parents and students.
•• To discuss the perceptions that teachers and parents may need to change in order to encourage stu-
dents’ scientific creativity.
Research Background
While this research is about scientific creativity, previous findings on general creativity can also provide mean-
ingful implications. Research which examined teachers’ recognition of creativity (e.g., Bereczki & Karpati, 2018)
can be classified into three categories: perceptions of the nature of creativity, perceptions of the characteristics of
creative students and perceptions of the creative learning environment (Andiliou & Murphy, 2010).
First, regarding research on teachers’ perceptions of the nature of creativity, Fryer and Collings (1991) conducted
a survey of 1,028 teachers in England and Wales and reported that teachers perceived creativity as imagination,
original ideas, self-expression, and so on, and about 71% of teachers considered creativity as a ‘rare gift’. Aljughaiman
and Mowrer-Reynolds (2005) found that 88% of American teachers they surveyed associated creativity with original
ideas. Liu and Lin (2014) noted that the perception of scientific creativity from Taiwanese teachers could be put
into three categories: divergent thinking (e.g., innovation and imaginative), autonomy (e.g., adventurous and/or
non-conforming), as well as curiosity and interest.
Second, regarding research on teachers’ perceptions of the characteristics of creative students, Aljughaiman
68
ISSN 1648–3898 /Print/ STUDENT, PARENT AND TEACHER PERCEPTIONS ON THE BEHAVIORAL CHARACTERISTICS
ISSN 2538–7138 /Online/ CREATIVITY
(pp. 67-79)
and Mowrer-Reynolds (2005) reported 92% of respondents said creative students had the characteristic of think-
ing differently from others. Hoff and Carlsson (2011) found that teachers thought creative students also had other
characteristics, such as high levels of academic achievement, ability for cooperation, psychological well-being
and self-confidence. Karwowski (2010), who surveyed 630 teachers in Poland, said that teachers thought creative
students were more dynamic, intellectual, and excitable, but on the other hand they were also perceived to be
less agreeable and conscientious. Chan and Chan (1999) found that teachers in Hong Kong thought that creative
students asked a lot of questions, were imaginative, responsive, active, and had a high intellect.
Finally, according to research that examined teachers’ perception of a creative learning environment, 80%
of American teachers surveyed thought that creativity could be developed in school classrooms (Aljughaiman
& Mowrer-Reynolds, 2005). When Rubenstein et al. (2018) asked teachers what interfered with students’ creativ-
ity, 60% of the teachers’ responses were about external regulations, such as limited time, exams or standardized
curricula, 23% were about student characteristics, such as students not taking risks. Hartley and Plucker (2014)
reported that Chinese teachers believed that even routine classroom activities, such as group discussion or watch-
ing educational movies, could improve students’ creativity, and only 8% of them answered that multiple-choice
tests were not suitable to foster creativity.
Parents’ perceptions of creativity alongside teachers’ perceptions are important to create an environment in
which creativity can flourish (Walberg, 1988). Among few studies examined how parents perceive creativity, Singh
(1987) surveyed 260 Indian parents the relative importance of the personality traits of creative children and found
that parents did not prefer creative traits such as independent thinking and judgment, risk taking, and intuition.
Runco, Johnson and Bear (1993) had 16 teachers who were not parents and 29 parents who were not teachers
choose adjectives describing creative children. The result shows that 67% of the adjectives chosen by both groups
were identical, while the remaining 33% were chosen by only one of the groups, and this shows that there is a
difference in the perceptions created by the contrasting experiences of the two groups.
Students’ perceptions of creativity alongside teachers’ and parents’ perceptions are also important to create a
suitable environment for creativity. Jaba et al. (2009) conducted a research of 158 Romanian college students and
reported that students thought of creativity as solving problems in an original way, having many ideas, perform-
ing in a different way from others, making associations between seemingly different things, and so on, so we can
expect that students interested in developing creativity will consider some of these factors. Also, from students’
own assessments of creativity, we can learn about their perceptions of creativity. For example, Kreitler and Casa-
kin (2009) gave students the task of designing a small art museum and allowed them to evaluate whether their
designs were creative by themselves. The result showed that their evaluations of fluency, flexibility, and overall
creativity were consistent with expert’s evaluation. However, when Althuizen et al. (2010) compared the results of
a creative self-assessment for 120 Dutch university and graduate students with the results of a general creativity
test for them, there was only a low correlation (α=0.25) in originality and no correlation in the rest of the items of
creativity. Thus, in science, too, if students don’t recognize their scientific creativity correctly, they may not be able
to make correct efforts for the aspects of scientific creativity that should be developed.
As noted in previous literature reviews, there are many aspects to the perception of creativity, such as percep-
tions of the nature of creativity, perceptions of creative students or creative teachers, and the perception of creative
learning environments or creative learning strategies (Andiliou & Murphy, 2010).
This research examined the behavioral characteristics of scientific creativity that are evaluated by teachers,
parents, and students as an important aspect for scientifically creative students. With the assumption that these
perceptions are important to foster students’ scientific creativity, this research is meaningful in that fostering stu-
dents’ scientific creativity is one of the important goals of science education.
General Background
This research implemented a survey to collect data systematically on people’s preferences, thoughts and
behaviors. The questionnaire for the survey was developed using a list of behavioral characteristics of creative
physicists (BCCP) during their growth period (Shin & Park, 2020). A total of 305 participants who answered the
questionnaire were randomly selected among teachers, parents and students related to gifted science education
in Korea. When respondents participated in a university program related to gifted science education between
69
(pp. 67-79)
July and August 2019, they responded directly to the questionnaire on site. At that time, they chose 10 items from
the BCCP list that they thought were important for students to become creative scientists in the future, and they
ranked the selected items according to their level of importance. Their responses were quantitatively analyzed
using a statistical analysis to compare the differences between the three groups. The results were used to discuss
the behavioral characteristics parents and teachers should be concerned or interested in to encourage students’
scientific creativity.
Participants
In Korea, there are many centers for the gifted in science and mathematics. The center at Chungbuk National
University selects gifted students in the area of science and mathematics from elementary and middle schools,
and it then teaches them for about 100 hours a year. After a year of education, students who show excellent results
can receive two more years of education. Of the 160 first-year students at the center, 149 students were randomly
selected and agreed to participate in the survey. They responded to a questionnaire for about 20 minutes when
they participated in the university-provided education program.
Among the parents of the gifted science students in the center, 116 parents expressed their intention to
participate in the survey. They answered a questionnaire when they participated in a presentation in which their
children presented the results of the gifted education program.
The university to which the center belongs provided an in-service training program for science teachers for
gifted education in science. Among 191 science teachers that participated in the training program, 50 lower and
upper secondary school teachers expressed their willingness to answer the questionnaire. They have experience
teaching gifted students in science or are highly interested in gifted education in science, with an average educa-
tional background of 5.3 years.
Of the total 315 participants, 145 students, 112 parents and 48 teachers were included in the actual analysis,
excluding those who responded insincerely or omitted some responses in the survey. In general, to conduct a
statistical analysis, larger sample sizes are better. In a real situation, however, using the appropriate sample sizes
is also recommended. For example, according to Karada and Akta (2012), for an ANOVA analysis with 95% power,
α=0.05, and an effect size f=0.4, the estimated sample size for the three groups is 102, which is 34 per group. Since
the smallest teacher group’s sample size is 48, which is larger than 34, the size of three samples in this research
was judged to be sufficient for an ANOVA analysis.
Questionnaire
The behavioral characteristics of scientific creativity that teachers, parents, and students consider important
to become creative scientists is obtained from a list of behavioral characteristics of creative physicists during their
growth period (BCCP) (Shin & Park, 2020). The BCCP list was developed by extracting key behavioral characteristics
from the literature describing various episodes during the growth period of six creative physicists: Newton, Faraday,
Maxwell, Einstein, Heisenberg and Feynman. The BCCP list consists of 8 categories and 30 statements (items), as
shown in Table 1.
Table 1
The BCCP list (Shin & Park, 2020)
Category Code Statement
I. Making I.1 I like to make things.

I.2 I make things using what is available nearby.
I.3 I am used to using a machine or a device.
I.4 I have experience making new things or devices.
I.5 I make something complicated and accurate.
70
(pp. 67-79)
Category Code Statement
II. Inquiry/ II.1 I like experiment/inquiry and do it a lot.

Experiment
II.2 I have an experimental environment and have my own laboratory space.
II.3 I do scientific inquiry using what is available nearby.
II.4 I plan and make scientific inquiry deliberately and meticulously.
II.5 I do not hurry scientific inquiry
III. Task III.1 I prefer to learn by myself.
commitment
III.2 I like to solve difficult problems.
III.3 I have concentration.
III.4 I have experiences of solving problems by applying effort over a long time.
III.5 I have my own rules/standards and hate to lose.
IV. Curiosity/ IV.1 I am curious about natural phenomena.
Question
IV.2 I enjoy thinking deeply about natural phenomena.
IV.3 I am interested in profound/fundamental problems such as religion or philosophy.
IV.4 I like questions and ask many questions.
V. Reading/ V.1 I do a lot of reading.
Summary
V.2 I summarize my reading, exploring, studying and thinking.
V.3 I have a habit of taking notes in everyday life.
VI. Logic/ VI.1 I prefer logical thinking.
Insight
VI.2 I understand content easily (grasp the point well/have insight).
VI.3 I use/develop my own unusual methods.
VII. Artistic VII.1 I have a talent and interest in music.
interest/
VII.2 I like poetry and am good at poetic expression.
expression
VII.3 I understand or express things using visualization instead of words.
VIII. Communication VIII.1 I prefer discussion and debate.
VIII.2 I like to share ideas with a team.
This list has gained high validity based on a survey of experts and science teachers who had experience in
gifted education, and it was found to have high reliability by applying it to science upper secondary school stu-
dents who showed high interest and achievements in science and mathematics using the Cronbach alpha, Cohen’s
kappa, and correlations between two evaluators (Shin & Park, 2020).
The reasons why this list is chosen here is because it consists of statements describing behavioral charac-
teristics that can be easily observed by teachers, parents and students, and also because the list is appropriate
to judge students’ behaviors since it was developed based on actual behavioral episodes shown in the period of
childhood, not on the behavior of adults.
The questionnaire provided a list of 30 items of the BCCP and asked the participants to select 10 items that
they thought were important to become creative scientists, and to rank the 10 selected items from the 1st to 10th
according to their level of importance.
Some of the items in the list were modified and presented as more easily understandable expressions for
young students or parents. For example, item VI.1 “I prefer logical thinking” was modified to “I like logical think-
ing”, and item VI.2 “I understand content easily (grasp the point well/have insight)” was modified to “I understand
content easily and can grasp the point well”.
The BCCP list was developed by looking at six creative physicists, but this research introduced the items
simply as behavioral characteristics of creative ‘scientists’ because the subjects included young students, and as
such, the general expression ‘scientists’ was considered appropriate rather than giving detailed expressions such
as ‘physicists’ that could be misunderstood. Of course, the limitations of this research in this regard are discussed
in the Conclusions and Implications section.
71
(pp. 67-79)
Data Analysis
The rankings of the items selected by the respondents were converted into scores. That is, 10 scores were given
to the first ranked item, 9 scores were given to the second ranked item, and so on, until 1 score was given to the
10th ranked item. Using the converted scores, the total scores and the average values were obtained for each of the
three groups of teachers, parents and students, and for each of the 30 items and 8 categories in the BCCP list itself.
A one-way ANOVA was conducted to determine whether there were significant differences between the av-
erage of each group, each item, and each category. When a significant difference was detected, a Scheffé analysis
was conducted as a Post-Hoc test. To do this, the SPSS statistics program (version 25) was used.
Research Results
Comparison of the Eight Categories
Table 2 shows the results of averaging the importance perceived by the groups of students, parents and
teachers for each category of the BCCP.
Table 2
Three Groups’ Perceptions of the Importance of Each Category of the BCCP
Group ANOVA
Category S: Students P: Parents T: Teachers Scheffé test

(n=145) (n=112) (n=48) F p
M SD M SD M SD
I. Making 1.33 1.35 0.77 1.07 0.79 0.93 8.186 < .001** S > P**, S > T*
II. Inquiry/Experiment 2.27 1.29 1.51 1.16 1.55 1.14 14.057 < .001** S > P**, S > T**
III. Task commitment 2.40 1.36 2.89 1.52 2.30 1.50 4.516 .012* S < P*
IV. Curiosity/Question 1.69 1.58 2.16 1.70 3.78 1.84 28.260 < .001** S < T**, P < T**
V. Reading/Summary 2.29 1.94 2.69 1.94 1.46 1.83 6.889 .001** S > T*, P > T**
VI. Logic/Insight 2.09 1.87 2.30 1.66 2.32 1.67 0.584 .558 -
VII. Artistic interest/expression 0.17 0.63 0.49 0.94 0.17 0.46 6.696 .001** S < P**, P > T*
VIII. Communication 2.30 2.34 2.04 2.31 2.42 2.27 0.604 .547 -
*
p < .05, ** p < .01
Looking at Table 2, we can see that categories VI (Logic/Insight) and VIII (Communication) do not show any
significant differences in the importance perceived by the three groups. Since the average score for the importance
of each group in both categories was higher than the overall average of 1.84 (2.09–2.32 scores for category VI and
2.04–2.42 scores for category VIII), this means that all three groups recognized that both categories were important.
There are six categories with different scores in the three groups’ perceptions of importance. Among them, for
category I (Making) and II (Inquiry/Experiment), students perceived these as more important than parents and teach-
ers both, and for category V (Reading/Summary), students considered this more important than teachers. However,
the recognition of importance for category I (Making) was lower (0.77–1.33 scores) than the overall mean of 1.84.
On the other hand, parents or teachers gave a higher importance to category III, IV and VII compared to students.
That is, compared to students, categories III (Task commitment) and VII (Artistic interest/expression) were perceived
by parents as more important, and category IV (Curiosity/Question) was perceived by teachers as more important.
There were three categories where the perceptions of the importance of parents and teachers differed. That
is, category IV (Curiosity/Question) was perceived to be more important to teachers, while categories V (Reading/
Summary) and VII (Artistic interest/expression) were perceived as more important to parents. However, category
VII was very low in importance for all three groups (0.17–0.49 scores).
72
(pp. 67-79)
The following is a summary of the features analyzed above:

- Students, parents and teachers all value ‘thinking logically’ and ‘communicating with others’ as creative
behaviors.
- There are differences between groups, but all three groups consider ‘making’ and ‘artistic expression’ to
be relatively less important.
- Students think ‘active characteristics (i.e., making, inquiry/experiment)’ and reading/summary’ are more
important than their parents and/or teachers. However, for the aspects related to ‘learning (i.e., task
commitment)’ and ‘thinking (i.e., curiosity/question)’ and ‘artistic expression’, teachers or parents consider
it more important than students.
- When comparing perceptions of parents to those of teachers, teachers value ‘aspect related to thinking
(i.e., curiosity/question)’ more than parents, but parents think ‘learning aspect (i.e., reading/summary)’
and ‘artistic expression’ are more important than teachers.
Comparison of the 30 Statements (Items)
The BCCP list consists of 8 categories and 30 statements (Table 1). Table 3 shows the results of a comparison
between the three groups’ perceptions about the importance of each statement in the BCCP list.
Table 3
Three Groups’ Perceptions of the Importance of Each Statement in the BCCP
Group ANOVA
Category and Statement S: Students P: Parents T: Teachers Scheffé test

(n=145) (n=112) (n=48) F p
M SD M SD M SD
I.1 2.67 3.69 1.29 2.55 1.06 2.35 8.303 < .001** S > P**, S > T**
I.2 0.82 2.13 1.09 2.20 0.81 1.93 0.576 .563 -
I.
I.3 1.23 2.49 0.36 1.35 0.88 1.68 5.898 .003** S > P**
Making
I.4 0.96 2.26 0.98 2.28 0.98 2.01 0.004 .996 -
I.5 0.97 2.32 0.13 0.69 0.21 0.94 8.681 < .001** S > P**, S > T*
II.1 4.77 3.84 3.68 3.36 4.13 3.95 2.784 .063 -
II.2 0.64 1.72 0.08 0.59 0.06 0.43 7.769 .001** S > P**, S > T**
II.
II.3 1.30 2.35 1.29 2.39 1.04 1.95 0.247 .781 -
Inquiry/Experiment
II.4 3.32 3.53 1.91 2.92 1.75 2.76 7.974 < .001** S > P**, S > T*
II.5 1.30 2.62 0.61 1.75 0.79 1.71 3.257 .040* S > P*
III.1 2.12 2.97 3.93 3.61 1.83 3.17 11.954 < .001** S < P**, P > T**
III.2 2.21 3.22 2.57 3.37 2.54 3.35 0.444 .642 -
III.
III.3 5.74 3.72 4.43 3.77 3.88 3.56 6.359 .002** S > P*, S > T*
Task commitment
III.4 1.23 2.45 2.90 3.48 2.83 3.66 10.981 < .001** S < P**, S < T**
III.5 0.71 1.91 0.60 1.80 0.40 1.76 0.534 .587 -
IV.1 1.88 3.10 3.13 3.89 5.60 4.16 19.848 < .001** S < P*, S < T**, P < T**
IV. IV.2 0.92 2.22 1.82 3.18 4.88 4.19 32.18 < .001** S < T**, P < T**
Curiosity/Question IV.3 0.22 1.19 0.35 1.43 0.50 1.79 0.791 .454 -
IV.4 3.76 3.81 3.34 3.86 4.15 3.59 0.841 .432 -
V.1 3.98 3.69 4.29 4.04 2.08 3.18 6.125 .002* S > T*, P > T**
V.
V.2 1.82 2.98 3.13 3.79 1.38 2.72 7.077 .001* S < P**, P > T**
Reading/Summary
V.3 1.08 2.44 0.66 2.05 0.94 2.20 1.105 .333 -
73
(pp. 67-79)
Group ANOVA
Category and Statement S: Students P: Parents T: Teachers Scheffé test

(n=145) (n=112) (n=48) F p
M SD M SD M SD
VI.1 2.78 3.37 2.62 3.48 2.79 3.38 0.085 .919 -

VI.
VI.2 1.66 2.82 2.79 3.51 2.29 3.25 4.16 .017* S < P*
Logic/Insight
VI.3 1.83 2.98 1.50 2.64 1.88 3.10 0.512 .600 -
VII.1 0.16 0.95 0.31 0.98 0.10 0.47 1.286 .278 -
VII.
Artistic interest/expres- VII.2 0.06 0.59 0.09 0.59 0.00 0.00 0.463 .630 -
sion
VII.3 0.28 1.18 1.06 2.30 0.40 1.33 7.017 .001* S < P**
VIII. VIII.1 2.24 3.02 1.68 2.75 1.50 2.70 1.814 .165 -
Communication VIII.2 2.35 3.36 2.39 3.08 3.33 3.14 1.811 .165 -
* p < .05, ** p < .01
According to Table 3, fifteen statements (statement numbers I.2, I.4, II.1, II.3, III.2, III.5, IV.3, IV.4, V.3, VI.1, VI.3,
VII.1, VII.2, VIII.1 and VIII.2) do not show any significant difference between the three groups’ perceptions of their
importance to scientific creativity. Among these, three groups perceived statements II.1 (many inquiry/experiment,
average scores are 3.68-4.77), III.2 (solving difficult problems, average scores are 2.21-3.37), IV.4 (many questions,
average scores are 3.34-4.15), VI.1 (logical thinking, average scores are 2.62-2.79) and VIII.2 (sharing ideas, average
scores are 2.35-3.33) as more important than the overall average of 1.84.
Among statements showing no differences, statements VI.3 (developing unusual methods) and VIII.1 (discus-
sion and debate) showed values near the average (1.84), but the of rest the statements (I.2; making using what is
available nearby, I.4; making new things, II.3; experimenting using what is available nearby, III.5; developing my
own standards, IV.3; interest in fundamental problems, V.3; taking notes, VII.1; talent in music and VII.2; interest in
poetry) were below the average importance value of 1.84. Therefore, the following is a summary of the statements
that do not differ in perception between the three groups:
- The three groups all perceived ‘conducting many experiments, asking many questions, thinking logically
to solve difficult problems, and sharing ideas’ as important for scientific creativity.
- The three groups all considered ‘discussion/debate with their own unusual methods’ as averagely impor-
tant.
- Three groups all perceived ‘using what is available nearby to experiment or make something new, taking
notes with interest in fundamental problems, developing their own standards, and talent/interest in aesthetic
aspects (poetry and music)’ as less important compared to other statements in the BCCP list.
When comparing the perceptions of teachers or parents with that of students in Table 3, fifteen among thirty
statements showed differences.
First, the statements that students consider more important than their parents and/or teachers are statement
I.1 (like making), I.3 (using machine or device when making), I.5 (making something complicated and accurate), II.2
(having my own laboratory space), II.4 (conducting inquiry deliberately and meticulously), II.5 (conducting inquiry
without hurry), III.3 (concentrating) and V.1 (reading a lot).
Among these eight statements, six statements are from category I (Making) and II (Inquiry/Experiment), and
students’ perceptions about four particular statements, I.1 (like making, average score is 2.67), II.4 (conducting
inquiry deliberately and meticulously, average score is 3.32), III.3 (concentrating, average score is 5.74) and V.1
(reading a lot, average score is 3.98), are especially higher than the average value of 1.84.
Second, teachers or/and parents perceived the following statements to be more important compared to
students: statement III.1 (learning by myself ), III.4 (giving effort for a long time to solve problems), IV.1 (curiosity),
IV.2 (deep thinking), V.2 (summarizing), VI.2 (grasping the point), and VII.3 (visualizing of understanding).
Of these seven statements that teachers or parents considered important compared to students, teachers
74
(pp. 67-79)
and parents both perceived the statement III.4 (giving effort for a long time to solve problems) and IV.1 (curiosity)
as more important than students. And the statement that the teacher perceived as more important than students
was statement IV.2 (deep thinking), and the statements that the parent considered as more important than stu-
dents were statement III. 1 (learning by myself ), V.2 (summarizing), VI.2 (grasping the point) and VII.3 (visualizing
of understanding). Therefore, statements showing a difference between student’s perception and teachers’ or
parents’ perceptions can be summarized as follows:
- Students consider ‘reading a lot and making or experimenting accurately and deliberately using machines or
devices without hurrying but with concentration in their own laboratory space’ more important compared
to their teachers or/and parents.
- Parents and teachers both perceive ‘giving effort for a long time to solve problems and showing curiosity’
more important than students.
- Especially teachers consider ‘deep thinking’ a more important behavior compared to students.
- And parents perceive ‘summarizing and visualizing that they grasp the point of what they learned by
themselves’ to be more important than students.
As shown in Table 3, the following five statements show differences between parents’ and teachers’ percep-
tions: III.1, IV.1, IV.2, V.1 and V.2.
That is, parents consider statement III.1 (learning by myself ), V.1 (reading a lot), and V.2 (summarizing) more
important than teachers, and teachers perceive statement IV.1 (curiosity), IV.2 (deep thinking) more important than
parents. Therefore, this feature can be summarized as follows:
- Parents think the learning aspects (i.e. summarizing what students learn by themselves and reading many
books) are more important than teachers, on the other hand, teachers perceive thinking aspects (i.e.
deep thinking with curiosity) to be more important than parents.
Discussion
Of the 30 statements in the BCCP list, the three groups all perceived ‘conducting many experiments, asking
many questions, thinking logically to solve difficult problems, and sharing ideas’ as important for scientific creativity.
However, eight statements (using what is available nearby to experiment or make something new, taking notes
with interest in fundamental problems, developing their own standards, and talent/interest in poetry and music) were
considered less important than the average value for all three groups.
This result indicates that more attention is necessary to encourage these behavioral characteristics, which
were considered less important, for scientific creativity in schools and home. For example, ‘using what is available
nearby to experiment to make something new’ means that experiments whose methods, tools and results are not
prepared in advance are meaningful. In fact, school experiments provide experimental settings, tools and methods
designed to reduce errors as much as possible and to obtain the expected clear results. Therefore, even students
who have talent and interest in science have very few opportunities to conduct unprepared open experiments in
schools. As a result, when teaching students in a scientifically gifted center in a university, it has been observed
that many students have a lot of trouble conducting open inquiry. According to actual research procedures of
real scientists, they go through cyclic or non-linear procedures when conducting research (Park et al., 2009). That
is, scientists often go back to the previous stage of research, change their tentative hypotheses, obtain data re-
peatedly, and modify their research methods when results are not as expected, or they feel there are problems.
In this approach, there is also a non-linear process for missing certain steps or changing the path of the research.
Regarding this, Zion et al. (2004) observed that open inquiry could help students realize that the inquiry process
was not linear but was subject to change.
In addition, ‘developing their own standards’ is closely related to the aspect discussed above because this feature
appears in situations in which a pre-determined process and result are ignored. This approach also emphasizes
‘unusual thinking’ in creative endeavors (Park, 2004). According to the behavioral characteristics of creative physi-
cists in their growth period, Maxwell ignored formality when developing his own rules and procedures (Cropper,
2001/2007, p. 62), Feynman calculated it in his own original way, not just following prescribed methods of calculation
(Gleik, 2011, pp. 42-43), Einstein also proved the Pythagorean theorem in his own original way (Issacson, 2007, p. 17).
75
(pp. 67-79)
‘Interest in fundamental issues’ also relates to the aspects mentioned above in terms of the fact that fundamental
problems do not have pre-determined or typical solutions. In fact, Heisenberg (Hermann, 1976, p. 11), Newton
(Christianson, 1996, p. 22; Gleik, 2007, p. 25), and Einstein (Hoffman & Dukas, 1972, p. 24) are all said to have been
interested in Plato’s, Aristotle’s, or Kant’s philosophies since their childhood. Fundamental problems or philosophy
are difficult and complex subjects and, therefore, can often be inconclusive. Since schools don’t often deal with
topics that don’t have clear conclusions like these, teachers, parents and students all need to pay more attention
to this aspect to encourage scientific creativity.
It is also found that ‘taking notes’ needs to be encouraged to promote students’ scientific creativity. Regard-
ing this, Newton was so busy writing down many of his thoughts that it was difficult to find elegant handwriting
in his notes (Christianson, 1996, p. 23), and it is well-known that Faraday made good notes about the lectures he
heard (Russell, 1999, Location No. 235). The reason why ‘taking notes’ is important is because creative students are
apt to have many and various different ideas, and notes can help maintain and sustain them for a longer period
of time. In this approach, initial notes are modified and developed over time, depending on additional experience
and knowledge. In fact, even scientific knowledge was initially invented or proposed in an incomplete state and
was then developed in an evolutionary manner. For example, Lakatos (1994) described how the atomic model
evolved from Bohr’s initial simple and incomplete model to the more detailed and articulated model we know now.
For the two statements related to the aesthetic aspects (statements VII.1 and VII.2, interest and talent in music
or poetry), the mean value of the importance awareness was very low at 0.13 compared to the total average value
of 1.84. Of course, creativity is area- or content-dependent. That is, creativity in science has different features from
creativity in other fields or areas (Baer, 1991; Conti et al., 1996). However, there is something in common between
science and art in terms of creativity. In other words, just as it is important to express our own new feelings or ideas
differently in music or poetry, new scientific findings also need to be expressed through new languages, explana-
tions, or models (Shin & Park, 2020). In schools, many students think that scientific knowledge they are taught is
the truth, but the nature of science emphasizes the ‘tentativeness of scientific knowledge’ (Akerson et al., 2006;
Lederman et al., 2002). That is, scientific knowledge is subject to change. Therefore, in order to encourage scientific
creativity, science teachers, parents and students need to pay attention to how students express their own ideas
with their own language when suggesting new hypotheses to explain a phenomenon, drawing a new conclusion
about their scientific inquiry and so on.
According to the results, half of the 30 statements showed differences in the perception between the student
group and the parent and/or teacher groups. In this case, attention needs to be paid to these differences because
these differences may cause conflict between what students want and what they are encouraged to do. The main
feature of this difference is that while students mainly place more importance on active aspects (i.e., making and
experiment), teachers and/or parents mainly emphasized learning or thinking aspects (i.e., giving effort for a long
time and thinking deeply to solve problems with curiosity, summarizing and visualizing what they grasp the point in
the learning by themselves).
In fact, these two aspects (activity vs thinking/learning) are complementary elements in the behavioral char-
acteristics of scientific creativity. Therefore, if a student follows the guidance of the parent/teacher well and the
teachers/parents pay attention to the student’s attention, then these differences in perceptions are not important.
However, if teachers and parents only emphasize what they think is important, they may create an imbalance while
trying to encourage scientific creativity.
According to previous research (Shin & Park, 2020) that examined the behavioral characteristics of creative
physicists based on the literature that described various episodes during their growth period, about one-fourth of
210 episodes extracted from the literature analysis belonged to Category I (making) and II (inquiry/experiment).
This means that the activity aspect cannot be ignored and is an important part of scientific creativity. In fact, it is
not difficult to see that Newton, Maxwell, Einstein, and Feynman, well known as theoretical physicists, loved mak-
ing and experimenting a lot as children and showed talent for it. For example, Maxwell “loved apparatus and had
considerable manual dexterity” (Segre, 2012, Location No. 2857), Newton was also interested in special inventions
and machines (Christianson, 1996, p. 13), and Feynman is famous for making various devices as a child (Feynman
& Leighton, 1985, pp. 20-23). Therefore, teachers and parents need to understand the importance of the activity
aspects for scientific creativity and to encourage students’ behaviors related to it.
Finally, it is necessary to pay attention to the difference in perception between teachers and parents because
the two groups can give different guidance to the students, and as a result, students can also experience conflict.
Regarding this, parents thought learning aspects (summarizing what students learn by themselves and read many
76
(pp. 67-79)
books) were more important while teachers perceived thinking aspects (deep thinking with curiosity) to be more
important. It is anticipated that parents may have greater interest in aspects related to studying. Moreover, some
parents may like curiosity and deep thinking less because curious students may seem distracted by a wide range
of interests or often miss what they have to do by falling into deep contemplation.
However, Newton was distracted by seeing the sun light passing through the garden in front of his house
(Christianson, 1996, p. 14), and Maxwell was obsessed with observing beetles rather than playing with other boys
at school (Forbes & Mahon, 2014, p. 132). Einstein is also said to have liked daydreaming and meditation (Issacson,
2007, p. 11). Therefore, parents need to adopt an attitude of watching and waiting when students display curiosity
and deep thinking even when these behaviors do not quickly show clear results.
Based on the assumption that teachers’ or parents’ perceptions of scientific creativity play an important role
in creating a learning environment that affects students’ scientific creativity, the perceptions of teachers, parents,
and students of behavioral characteristics related to scientific creativity were analyzed. For this, a list of behavioral
characteristics regarding creative physicists during their growth period was used.
The results show that behavioral characteristics, such as making something new using unprepared materials
from nearby, taking notes and showing interest in fundamental problems, developing their own standards, and having
concern for aesthetics, are perceived as less important for all three groups. Nevertheless, it was discussed why such
characteristics are important for scientific creativity and why they should also be emphasized. When comparing
the perceptions of the three groups, an especially interesting finding is that the parent group provided greater
importance to aspects related to learning, teachers to aspects related to thinking, and students to aspects related
to activities such as making and experimenting. These differences in how the groups look at these three aspects
are not problematic if they are encouraged in a complementarily fashion. However, if each group emphasizes dif-
ferent aspects as being important, then it can cause conflicts for the students.
Based on the results of this research, future research can be meaningful. First, research is needed to develop
guidance materials of the behavioral characteristics that need to be encouraged for students’ scientific creativity. In
particular, it is necessary to encourage behavioral characteristics that all three groups perceive to be less important.
For this, real-life episodes or instances of behaviors that actual scientists have shown during their growth period
will be helpful. In addition, since some behaviors can be seen as abnormal behaviors in a negative way, such as
excessive meditation, logically arguing, trying to make their own rules, and asking too many questions, guidance
is also be needed to explain why these behaviors may appear in young creative physicists.
Second, this research assumes that the perception of creative behaviors by parents or teachers serves as
an important factor to create learning environments that affect how students’ scientific creativity is encouraged.
To check this assumption, further research is necessary to check whether there is a change in students’ scientific
creativity when there are differences between students’ perceptions and teachers’ and parents’ perceptions. That
is, it will be possible to examine whether students become more creative in science in the future when there are
fewer differences in the perceptions of students, teachers and parents, or what happens when creative behavioral
characteristics are emphasized in a balanced manner. This kind of research requires significant hard work and
observations over long periods of time, but this effort would be very meaningful.
Finally, it is necessary to point out limitations in this research. This research used a list of creative behaviors of
physicists during their growth period. However, when examining the perceptions of the participants, the behavioral
characteristics provided were simply introduced as behavioral characteristics of the scientists. Therefore, a list of
behavioral characteristics of ‘scientists’ that includes areas other than physics should be developed to obtain more
accurate comparisons of the perceptions of the participants. In addition to questionnaires, in-depth interviews will
also be helpful to examine the perceptions more accurately.
References
Akerson, V. L., Morrison, J. A., & McDuffie, A. R. (2006). One course is not enough: Preservice elementary teachers’ retention of
improved views of nature of science. Journal of Research in Science Teaching, 43(2), 194-213. https://doi.org/10.1002/tea.20099
Aljughaiman, A., & Mowrer-Reynolds, E. (2005). Teachers’ conceptions of creativity and creative students. The Journal of Creative
Behavior, 39(1), 17-34. https://doi.org/10.1002/j.2162-6057.2005.tb01247.x
Althuizen, N., Wierenga, B., & Rossiter, J. (2010). The validity of two brief measures of creative ability. Creativity Research Journal,
77
(pp. 67-79)
22(1), 53-61. https://doi.org/10.1080/10400410903579577

Andiliou, A., & Murphy, P. K. (2010). Examining variations among researchers’ and teachers’ conceptualizations of
creativity: A review and synthesis of contemporary research. Educational Research Review, 5(3), 201–219.
https://doi.org/10.1016/j.edurev.2010.07.003
Baer, J. (1991). Generality creativity across per formance domains. Creativity Research Journal, 4(1), 23-39.
https://doi.org/10.1080/10400419109534371
Beghetto, R. A., & Kaufman, J. C. (2014). Classroom contexts for creativity. High Ability Studies, 25(1), 53-69.
https://doi.org/10.1080/13598139.2014.905247
Bereczki, E. O., & Karpati, A. (2018). Teachers’ beliefs about creativity and its nurture: A systematic review of the recent research
literature. Educational Research Review, 23, 25-56. https://doi.org/10.1016/j.edurev.2017.10.003
Chan, D. W., & Chan, L. K. (1999). Implicit theories of creativity: Teachers’ perception of student characteristics in Hong Kong.
Creativity Research Journal, 12(3), 185-195. https://doi.org/10.1207/s15326934crj1203_3
Christianson, G. E. (1996). Isaac Newton and the scientific revolution. Oxford University Press.
Cole, D. G., Sugioka, H. L., & Yamagata-Ynch, L. C. (1999). Supportive classroom environments for creativity in higher education.
The Journal of Creative Behavior, 33(4), 277-293. https://doi.org/10.1002/j.2162-6057.1999.tb01407.x
Conti, R., Coon, H., & Amabile, T. M. (1996). Evidence to support the componential model of creativity: Secondary analysis of
three studies. Creativity Research Journal, 9(4), 358-389. https://doi.org/10.1207/s15326934crj0904_9
Cropper, W. H. (2007). Great Physicists: The life and times of leading physicists from Galileo to Hawking (H. Kim, Trans.). Science
Books (Original work published 2001).
Csikszentmihalyi, M. (1996). Creativity: Flow and the psychology of discovery and invention. Harper Perennial.
Davies, D., Jindal-Snape, D., Collier, C., Digby, R., Hay, P., & Howe, A. (2013). Creative learning environments in education-A
systematic literature review. Thinking Skills and Creativity, 8, 80-91. https://doi.org/10.1016/j.tsc.2012.07.004
de Souza Fleith, D. (2000). Teacher and student perceptions of creativity in the classroom environment. Roeper Review, 22(3),
148-153. https://doi.org/10.1080/02783190009554022
Diakidoy, I. A. N., & Kanari, E. (1999). Student teachers’ beliefs about creativity. British Educational Research Journal, 25(2), 225-243.
https://doi.org/10.1080/0141192990250206
Feynman, R. P., & Leighton, R. (1985). Classic Feynman: All the adventure of a curious character. W. W. Norton & Company.
Forbes, N., & Mahon, B. (2014). Faraday, Maxwell, and the electromagnetic field: How two men revolutionized physics. Prometheus
Books.
Fryer, M., & Collings, J. A. (1991). British teachers’ views of creativity. The Journal of Creative Behavior, 25(1), 75-81.
https://doi.org/10.1002/j.2162-6057.1991.tb01356.x
Gleik, J. (2011). Genius: The life and science of Richard Feynman [Kindle version]. https://amzn.to/3ncwBCF
Gleik, J. (2007). Issac Newton [Kindle version]. https://amzn.to/3lgGBsS
Gralewski, J., & Karwowski, M. (2018). Are teachers’ implicit theories of creativity related to the recognition of their students’
creativity? The Journal of Creative Behavior, 52(2), 156-167. https://doi.org/10.1002/jocb.140
Guncer, B., & Oral, G. (1993). Relationship between creativity and nonconformity to school discipline as perceived by teachers of
Turkish elementary school children, by controlling for their grade and sex. Journal of Instructional Psychology, 20(2), 208-214.
Hartley, K. A., & Plucker, J. A. (2014). Teacher use of creativity-enhancing activities in Chinese and American elementary classrooms.
Creativity Research Journal, 26(4), 389-399. https://doi.org/10.1080/10400419.2014.961771
Hermann, A. (1976). Werner Heisenberg, 1901-1976. Inter Nations.
Hoff, E. V., & Carlsson, I. (2011). Teachers are not always right: links between teacher ratings and students’ creativity scores, self-images
and self-ratings in school subjects. The Open Education Journal, 4(1), 120-129. https://doi.org/10.2174/1874920801104010120
Hoffman, B., & Dukas, H. (1972). Albert Einstein: Creator and rebel. The Viking Press.
Issacson, W. (2007). Einstein: His life and universe [Kindle version]. https://amzn.to/3d02WYx
Jaba, E., Roman, M. D., Pagliacci, M., Serban, D., Balan, C. B., & Asandului, M. (2009). Statistical evaluation of the students’ perception
of creativity. In International Conference on Education, Research and Innovation, ICERI 2008 Proceedings, Madrid, Spain.
http://dx.doi.org/10.2139/ssrn.1328296
Karadag, O., & Aktas, S. (2012). Optimal sample size determination for ANOVA designs. International Journal of Applied Mathematics
and Statistics, 25, 127-134.
Karwowski, M. (2010). Are creative students really welcome in the classrooms? Implicit theories of “good” and
“creative” student’s personality among polish teachers. Procedia-Social and Behavioral Sciences, 2(2), 1233-1237.
https://doi.org/10.1016/j.sbspro.2010.03.179
Kreitler, S., & Casakin, H. (2009). Self-perceived creativity: the perspective of design. European Journal of Psychological Assessment,
25(3), 194-203. https://doi.org/10.1027/1015-5759.25.3.194
Lakatos, I. (1994). Falsification and the methodology of scientific research programmes. In J. Worrall and G. Currie (eds.), The
methodology of scientific research programmes: Philosophical papers Vol. 1 (pp. 9-101). Cambridge University Press.
Lederman, N. G., Abd-El-Khalick, F., Bell, G. L., & Schwartz, R. S. (2002). Views of nature of science questionnaire: Toward valid and
meaningful assessment of learners’ conceptions of nature of science. Journal of Research in Science Teaching, 39, 497-521.
https://doi.org/10.1002/tea.10034
Liu, S. C., & Lin, H. S. (2014). Primary teachers’ beliefs about scientific creativity in the classroom context. International Journal of
Science Education, 36(10), 1551-1567. https://doi.org/10.1080/09500693.2013.868619
Park, J. (2004). A suggestion of cognitive model of scientific creativity. Journal of the Korean Association for the Science Education,
78
(pp. 67-79)
24(2), 375-386.
Park, J., Jang, K. A., & Kim, I. (2009). An analysis of the actual processes of physicists’ research and the implications for teaching
scientific inquiry in school. Research in Science Education, 39(1), 111-129. https://doi.org/10.1007/s11165-008-9079-8
Peter-Szarka, S. (2012). Creative climate as a means to promote creativity in the classroom. Electric Journal of Research in
Educational Psychology, 10(3), 1011-1031.
Richardson, C., & Mishra, P. (2018). Learning environments that support student creativity: Developing the SCALE. Thinking Skills
and Creativity, 27, 45-54. https://doi.org/10.1016/j.tsc.2017.11.004
Rubenstein, L. D., Ridgley, L. M., Callan, G. L., Karami, S., & Ehlinger, J. (2018). How teachers perceive factors that influence
creativity development: Applying a social cognitive theory perspective. Teaching and Teacher Education, 70, 100-110.
https://doi.org/10.1016/j.tate.2017.11.012
Runco, M. A., Johnson, D. J., & Bear, P. K. (1993). Parents’ and teachers’ implicit theories of children’s creativity. Child Study Journal,
23(2), 91-113.
Russell, C. A. (1999). Michael Faraday: Physics and faith [Kindle version]. https://amzn.to/30u3e51
Segre, E. (2012). From falling bodies to radio waves: Classical physicists and their discoveries [Kindle version]. https://amzn.to/3cZgwLF
Shin, W., & Park, J. (2020). Developing a list of behavioral characteristics of creative physicists during their growth period.
International Journal of Science and Mathematics Education. https://doi.org/10.1007/s10763-020-10082-w
Singh, R. P. (1987). Parental perception about creative children. Creative Child & Adult Quarterly, 12, 39-42.
Sternberg, R. J., & O’hara, L. A. (1999). Creativity and intelligence. In R. J. Sternberg (Ed.), Handbook of creativity (pp. 251–272).
Cambridge University press.
Walberg, H. J. (1988). Creativity and talent as learning. In Sternberg, R. J. (Ed.), The nature of creativity (pp. 340-361). Cambridge
University Press.
Zion, M., Slezak, M., Shapira, D., Link, E., Bashan, N., Brumer, M., Orian, T., Nussinowitz, R., Court, D., Agrest, B.,
Mendelovici, R., & Valanides, N. (2004). Dynamic, open inquiry in biology learning. Science Education, 88(5), 728-753.
https://doi.org/10.1002/sce.10145
Cite as: Lee, I., & Park, J. (2021). Student, parent and teacher perceptions on the behavioral characteristics of scientific
creativity and the implications to enhance students’ scientific creativity. Journal of Baltic Science Education, 20(1), 67-79.
Insun Lee PhD in Science Education, Assistant Professor, Department of Physics

Education, College of Education, Chungbuk National University,
28644, Cheongju, Republic of Korea.
ORCID: https://orcid.org/0000-0002-1623-857X
Jongwon Park PhD in Science Education, Professor, Department of Physics Education,
(Corresponding author) College of Education, Chonnam National University, 61186, Gwangju.
Republic of Korea.
79
THE CONCEPTUAL STRUCTURE
OF CHEMICAL EQUILIBRIUM IN
UPPER-SECONDARY SCHOOL
STUDENTS: EVIDENCE FROM
FACTOR ANALYSIS
Abstract. This research aimed to explore Yuhua Mai,

the conceptual structure of chemical equi- Yangyi Qian,
librium in upper-secondary school students Linshen Li,
using factor analysis. Research on chem-
Haihang Lan
istry education has shown that chemical
equilibrium is an important but difficult-to-
understand topic. Exploring the conceptual
structure of chemical equilibrium among
students will help chemistry researchers
and educators to conduct more targeted
teaching practices. Based on a survey of Introduction
chemistry research and teaching prac-
Concepts and their structure are fundamental and significant issues in
tice experts, a high-quality concept pool
cognitive psychology. As influential factors, they play critical roles in cognitive
composed of 24 relevant concepts was
processes. It is thus indispensable to explore and understand the conceptual
developed. Next, a survey involving a total
structures that students hold of science concepts (Taber, 2003).
of over 700 twelfth-grade students from Different techniques have been developed to determine and represent
five upper-secondary schools was con- the different forms of an individual’s conceptual structure (Jonassen, et al.,
ducted, and a factor analysis was utilized 1993). Similarity rating (or relatedness rating) is one of the bases of these
to determine the conceptual structure. The techniques, which requires individual to rate the degree of similarity (or the
results showed that a three-factor model degree of relatedness) between pairs of concepts (Goldsmith, et al., 1991;
and a five-factor model with 15 relevant Giamellaro, 2014; Neiles, et al., 2016). As an example, word association is a
concepts were all accepted as the concep- short-timed test to identify the conceptual structure individual has of cer-
tain concepts (Derman & Eilks, 2016; Geeslin, & Shavelson, 1975; Nakiboglu,
tual structure for students. The new form of
2008; Yener, et al., 2018). Additionally, using the matrix of proximity values,
conceptual structure in this research helps
multidimensional scaling (MDS) and the Pathfinder network algorithm are
understand the features and categories in
common methods of exploring global and local information regarding the
students’ latent organization of concepts. conceptual structures of individuals and groups (Giamellaro, 2014; Goldsmith,
Also, it may be revealed that factor analysis et al., 1991; Neiles, et al., 2016). The large sample of rating data can also be
can be utilized as an approach to exploring processed by other data analysis methods, such as factor analysis. However,
students’ conceptual structure. little research has reported on the use of factor analysis to determine the
Keywords: chemical equilibrium, chemis- latent structure of science concepts in students.
try education, conceptual structure, factor Research has consistently shown that chemical equilibrium has been
analysis considered as an important but difficult chemistry topic for students to
learn (Bergquist & Heikkinen, 1990; Finley, et al., 1982). To understand the
Yuhua Mai, Yangyi Qian learning process and outcome for this topic, and to select suitable teaching
South China Normal University, P. R. China methods, researchers have conducted various research studies on students’
Linshen Li understanding of chemical equilibrium. Students’ misconceptions (Akkus,
Shimen Senior High School, P. R. China
Haihang Lan et al., 2011; Barke, et al., 2009; Hackling & Garnett, 1985; Karpudewan, et
South China Normal University, P. R. Chin al., 2015; Özmen, 2008; Quílez, 2004; Quílez-Pardo & Solaz-Portolés, 1995),
mental models (Chiu, et al., 2002), and conceptual structures (Gorodetsky &
80
ISSN 1648–3898 /Print/ THE CONCEPTUAL STRUCTURE OF CHEMICAL EQUILIBRIUM IN UPPER-SECONDARY SCHOOL
ISSN 2538–7138 /Online/ (pp. 80-92)
Hoz, 1985; Gussarsky & Gorodetsky, 1988, 1990; Wilson, 1994, 1996) for chemical equilibrium have been identi-
fied and analyzed. A deeper understanding of students’ thinking regarding chemical equilibrium can be utilized
to develop more effective teaching methods to improve the classroom teaching (Akkuş, et al., 2003; Maia & Justi,
2009; Ollino, et al., 2018; van Driel, et al., 1998). Conducting research involving factor analysis would be beneficial
to fully understand the organization of concepts and to promote the development of teaching methods.
Conceptual Structure and Factor Analysis
Generally speaking, different theoretical perspectives have provided a diverse understanding of the conceptual
structure regarding the relation between concepts. This has made the conceptual structure of a polysemous term
with multiple meanings. In the classical view, concepts can be divided into different categories because those in the
same category have similar attributes or prototypes (Medin, 1989; Moss, et al., 2007; Rosch, 2002). The associated
set of attributes or the prototypes belonging to concepts can be considered as conceptual structures. Cognitive
psychologists have concerned about how the attributes or the prototypes construct the concepts.
On the other hand, the conceptual structure can be utilized to denote the relation between concepts in
declarative memory, such as a hierarchical organization of concepts, a propositional network, or other forms. In
the first form of conceptual structure, a significant concept and its relevant concepts are hierarchically organized
to form the categories and subcategories (Zimbardo, et al., 2017). In the second form of conceptual structure, the
combinations of concepts produce the propositional statements and semantic network. The concept map is a typi-
cal structural representation of the propositional network in the domain knowledge (Novak & Gowin, 1984; Soika
& Reiska, 2014). Moreover, the spreading activation theory has proposed that relevant concepts are distributed
around the core concepts in the conceptual structure, basing on the degree of relatedness and the frequency of
use (Collins & Loftus, 1975). Because conceptual structure represents how individual organize and understand the
concepts in own mind, researchers have referred to it as a cognitive structure, a knowledge structure, or a structural
knowledge (Geeslin & Shavelson, 1975; Goldsmith, et al., 1991; Jonassen, et al., 1993; Nakiboglu, 2008). Review-
ing students’ conceptual structures of science topics is essential in science learning and education (Taber, 2003).
Classification and composition are advised as the fundamental semantic relations between science concepts in
science texts (Unsworth, 2001).
Factor analysis shows the latent interrelationships (known as correlations) between the factors and variables
(Hair, et al., 2010; Schinka & Velicer, 2003). The variables are categorized into different factors basing on their high
correlations, sharing the common meaning of the belonging factor. The model composed of the factors and their
variables reveals how the individuals consider a certain topic at the group-level. Factor analysis includes exploratory
factor analysis (EFA) and confirmatory factor analysis (CFA). As a data-driven analysis, EFA extracts latent common
factors from the variables, reduces the number of variables, and obtains a simple model that represents the relation
between the factors and variables. In contrast, CFA is a theory-driven analysis, which tests, compares, and modifies
the hypothesized models from EFA or theoretical assumptions (Hair, et al., 2010; Harrington, 2009).
As a famous example, factor analysis has been utilized to discover personality traits from individuals’ rating of
trait terms that defined the personality components in personality psychology research. Cattell’s Sixteen Personal-
ity Factors and Five-Factor Model have been all extracted by using factor analysis (Dumont, 2010). Factor analysis
is an essential means of exploring the variety of personality structures (Cawley, et al., 2000; Kaufman, et al., 2019),
and is widely utilized in the psychometric evaluation of measures, construct validation, and for other purposes
(Schinka & Velicer, 2003).
The new form of conceptual structure for science concepts was expected to be based on the high correla-
tions between the categories of concepts (known as factors) and their relevant concepts and to represent a new
different facet of latent organization of the science concepts from students.
Students’ Conceptual Structure of Chemical Equilibrium
To understand how the concepts related to chemical equilibrium are organized and stored in the memory,
researchers have conducted research studies on the conceptual structure of chemical equilibrium in students
(Gorodetsky & Hoz, 1985; Gussarsky & Gorodetsky, 1988, 1990; Wilson, 1994, 1996). In the research conducted by
Wilson (1994), upper-secondary school chemistry students completed a concept mapping task individually, then
the concept maps were transformed into proximity matrices and analyzed by Pathfinder. The results indicated that
81
(pp. 80-92) ISSN 2538–7138 /Online/
different achievement groups had differences in important specific conceptual relations and hierarchical organiza-
tion. Using a similar research methodology for upper-secondary school chemistry students, Wilson (1996) found a
significant relation between academic achievement in chemistry and the characteristics of the conceptual structure
as represented in a concept map.
Besides, Gussarsky and Gorodetsky (1988) utilized constrained word associations to find the categories of
concepts in the conceptual structures of chemical equilibrium from upper-secondary school students with differ-
ent levels in chemistry. In their other research (Gussarsky & Gorodetsky, 1990), free word associations were utilized
to determine the respective conceptual structure of ‘equilibrium’ and ‘chemical equilibrium’ in upper-secondary
school students. The results indicated that the distribution of concepts in these two conceptual structures could
explain the misconceptions caused by the semantic similarity of these concepts. Furthermore, Gorodetsky and Hoz
(1985) employed latent partition analysis (LPA) to analyze students’ free-sorted concept results and thus to reveal
the ‘group cognitive structure’ of freshman engineering students. The latent categories of concepts from different
achievement groups were also extracted and compared.
It is worth noting that previous researchers have individually found and listed concepts from chemistry cur-
riculum materials (Gorodetsky & Hoz, 1985; Gussarsky & Gorodetsky, 1988; Wilson, 1994, 1996). Their concept pools
were thus easily influenced by personal subjective judgment. Besides, these concept pools included the concepts
in respect of some specific reactants, phase change, or salt solution. There were few concepts regarding the shift of
the chemical equilibrium state. Because the number and scope of concepts affect the organization and quality of a
conceptual structure, it is necessary to provide a high-quality concept pool when researching conceptual structure.
The analytical techniques utilized in previous research studies have represented different forms of conceptual
structure of chemical equilibrium in students. Factor analysis is a means to explore and provide a new form of
conceptual structure, though it has not been reported on the topic of chemical equilibrium. Moreover, the number
of latent factors, as well as the actual relation between concepts, was not yet clear and expected to be answered.
Research Questions
In this research, we attempted to utilize factor analysis to expand the diversity of conceptual structure.
The specific research questions that guided this research were as follows:
(1) What is the concept pool of chemical equilibrium acquired by university chemistry researchers and
upper-secondary school chemistry teachers?
(2) What is the conceptual structure of chemical equilibrium held by Chinese upper-secondary school
students, as shown by factor analysis?
General Background
This research was a quantitative-oriented survey, which involved a large number of chemistry research
and teaching practice experts, and upper-secondary school students in Guangzhou, China during the 2019-2020
academic year. In the first part of this research, the concept pool of chemical equilibrium was acquired through
a survey questionnaire to university chemistry researchers and upper-secondary school chemistry teachers. This
ensured that our list of relevant concepts made theoretical and practical sense. Second, the conceptual structure
for students was determined through EFA and CFA of survey data from students.
Participants
This research utilized three groups of participants, all from Guangzhou, the capital of Guangdong province
in China. Their demographic characteristics are listed in Table 1.
82
ISSN 2538–7138 /Online/ (pp. 80-92)
Table 1
Demographic Characteristics of Participants
Teaching experience
Age Gender
year
Group N
M SD M SD Male Female
Group 1 Upper-secondary school chemistry teachers 204 39.60 6.68 16.56 7.23 71 133
Group 2 University chemistry researchers 10 45.20 6.89 17.40 7.96 7 3
Group 3 Upper-secondary school students 602 16.74 0.49 - - 306 300
The concept pool must have a wide scope of concepts, which are on the subject of the origin and change
of chemical equilibrium. To make an appropriate concept pool of chemical equilibrium, 224 upper-secondary
school chemistry teachers and 25 university chemistry researchers were invited to participate in the concept pool
surveys, respectively.
In the survey of upper-secondary school chemistry teachers (Group 1), 204 teachers’ questionnaires were valid
(recovery rate: 91.1%). The participants were in a unique position to effectively judge the relatedness of concepts
related to chemical equilibrium from the perspective of upper-secondary school chemistry teaching practice.
In the survey of university chemistry researchers (Group 2), 10 researchers accomplished the survey (recov-
ery rate: 40.0%), comprising seven professors and three associate professors. Half had published research studies
regarding inorganic chemistry and others were experts in physical chemistry. This group was in an appropriate
position to give reasonable suggestions in respect of the concept pool of chemical equilibrium from the perspec-
tive of chemistry research.
After the concept pool was produced, a total of 705 twelfth-grade students from five upper-secondary schools
were invited to participate in the conceptual structure survey. The valid sample comprised 602 students (Group
3; recovery rate: 85.4%). There were 313 students from two ordinary upper-secondary schools studying chemistry
at a basic level, and 289 students from three key upper-secondary schools having a higher level of chemistry. All
students had studied chemical equilibrium in a chemistry elective module before participating in the survey. The
sample of upper-secondary school students was randomly split into two sub-samples (Group 3a and 3b) to conduct
EFA and CFA on different rating data sets.
Participants were invited via the writers and informed of the intent of this research. They agreed to accomplish
the survey.
Instrument and Procedures
The research instrument was a Chinese-language questionnaire about the concept pool of chemical
equilibrium. The questionnaire development had two stages. In the first stage, an initial concept pool of chemi-
cal equilibrium was established. First, the upper-secondary school chemistry curriculum materials in China were
analyzed to determine the concepts related to chemical equilibrium. The curriculum materials comprised the
chemistry curriculum standards of China (MOE, 2003, 2018), chemistry textbooks of a compulsory module and an
elective module (Song, 2007a, 2007b; Wang, 2007; Wang, et al., 2007; Wang, 2014a, 2014b), and a general college
entrance examination outline (National Education Examinations Authority, 2018). According to the content of the
concepts in chemistry, twenty-six concepts involving the concept learning of chemical equilibrium in chemistry
module were selected. The concepts about specific reactants, acid-base equilibrium, solubility products, and phase
equilibrium were not included in the result. Next, three upper-secondary school senior teachers and an associate
professor in physical chemistry inspected the result and recommended five concepts. Thirty-one concepts were
selected as an initial concept pool. The first draft of the questionnaire consisted of these concepts as items and
utilized a 7-point Likert scale to rate each item’s degree of relatedness to the concept ‘chemical equilibrium’, in
order to select the relevant concepts. The scale had seven labels: the most unrelated, more unrelated, unrelated,
not sure, related, more related, and the most related, which were assigned from 1 to 7 points respectively.
In the second stage, upper-secondary school chemistry teachers and university chemistry researchers (Groups
1 and 2) received a survey invitation E-mail sent by the writers and accomplished the survey online. They utilized
83
(pp. 80-92) ISSN 2538–7138 /Online/
the first draft of the questionnaire to provide the rating regarding the degree of relatedness. When the average
scores of both groups in an item were greater than 5, the concept corresponding to that item was accepted as a
relevant concept in the formal concept pool. As is shown in Table 2, 24 concepts were accepted and utilized as
items in the formal questionnaire. Using the same 7-point Likert scale as before, the formal questionnaire required
participants to rate the degree of relatedness between each item and the concept ‘chemical equilibrium’. The relat-
edness rating represented the participant’s judgement of the semantic distances between the relevant concepts
and the concept ‘chemical equilibrium’ in cognitive structure. Five upper-secondary school students randomly
selected participated in a small interview. Their statements indicated that they could distinguish the meaning of
concepts utilized as items.
Table 2
The Concept Pool of Chemical Equilibrium
Category Number Concepts
Chemical equilibrium constant, Limitation of chemical reaction, Establishment of the

Quantitative representation 5
equilibrium state, Degree of conversion at equilibrium, Reaction quotient
Le Chatelier’s Principle, Shift of equilibrium state, Direction of the shift in chemical
Shift of state 5 equilibrium, Position of equilibrium moved to the right, Position of equilibrium moved to
the left
Feature of equilibrium 4 Feature of chemical equilibrium, Dynamic equilibrium, Reversible process, Reversibility
Reversible reaction 3 Reversible reaction, Forward reaction, Reverse reaction
Reaction rate 3 Chemical reaction rate, Rate of the forward reaction, Rate of the reverse reaction
Condition 4 Condition, Temperature, Concentration, Pressure
After formal questionnaire development, a survey involving upper-secondary school students (Groups 3)
was conducted with a paper and pen. They took 15 minutes to complete the questionnaire in a chemistry class.
Data Analysis
Data analysis had two stages. First, descriptive statistics of SPSS 23.0 were utilized to calculate the average
score of items, based on the rating data from three groups of participants. Data checking found that the partici-
pants utilized 1 through 4 points in the rating, meaning they have utilized the full scale. Average scores of items
greater than 5, which were from upper-secondary school chemistry teachers and university chemistry research-
ers (Groups 1 and 2), were utilized to decide the corresponding concept accepted in the formal concept pool of
chemical equilibrium.
Secondly, factor analysis was conducted on the rating data obtained from the upper-secondary school
student survey, to determine the conceptual structure of chemical equilibrium. EFA was conducted on the rating
data of Group 3a. Principal axis factoring and Promax rotation method in SPSS 23.0 were utilized to extract the
initial factors and their model. To test and compare the models provided by EFA or theoretical assumption, CFA
was conducted on the rating data of Group 3b. The Robust Maximum Likelihood Estimation (MLR) of Mplus 7 was
utilized to estimate the model parameters.
Research Results
The Concept Pool of Chemical Equilibrium
Table 3 shows that the average scores given to five items by both upper-secondary school chemistry teach-
ers and university chemistry researchers are less than 5: ‘endothermic reaction’, ‘exothermic reaction’, ‘activation
energy’, ‘catalyst’ and ‘motionless’. Upper-secondary school teachers’ average score of the item ‘initial concentra-
tion’ is also less than 5. These six concepts were thus rejected as relevant concepts in the formal concept pool of
chemical equilibrium.
On the other hand, the average score of the item ‘equilibrium concentration’ is similar to that of the item ‘con-
84
ISSN 2538–7138 /Online/ (pp. 80-92)
centration’. Because ‘equilibrium concentration’ is a subordinate concept of ‘concentration’ and upper-secondary

school students easily confuse the two included in the same concept pool. After careful deliberation by the writers,
the concept ‘equilibrium concentration’ was not accepted as part of the concept pool.
Table 3
The Average Scores of Items in the Survey
Upper-secondary Upper-secondary
University researchers
teachers students
Item
M SD M SD M SD
Dynamic equilibrium 6.69 0.66 6.70 0.67 6.32 0.98

Chemical equilibrium constant 6.64 0.70 6.90 0.32 6.45 0.79
Reversible reaction 6.48 0.80 5.60 0.84 6.13 0.96
Limitation of chemical reaction 6.44 0.86 6.20 0.92 5.85 1.12
Le Chatelier’s Principle 6.43 0.97 6.60 0.97 6.42 0.91
Shift of equilibrium state 6.40 0.86 6.70 0.67 6.46 0.80
Reversibility 6.40 0.89 6.40 0.70 5.97 1.00
Degree of conversion at equilibrium 6.30 0.90 6.30 1.25 6.13 0.93
Feature of chemical equilibrium 6.27 0.94 6.20 1.14 6.16 0.99
Direction of the shift in chemical equilibrium 6.16 1.02 6.20 0.79 6.32 0.84
Establishment of the equilibrium state 6.14 1.16 6.00 0.94 5.91 1.11
Concentration 6.08 1.07 6.10 0.88 5.88 1.06
Equilibrium concentration 6.06 1.07 6.10 1.29 - -
Reversible process 6.03 1.16 5.40 1.07 5.89 1.07
Reaction quotient 6.03 1.18 6.10 1.10 5.46 1.22
Reverse reaction 6.01 1.11 5.90 0.74 5.89 1.08
Position of equilibrium moved to the right 5.96 1.14 6.10 0.88 6.00 1.03
Temperature 5.96 1.12 6.40 0.70 5.65 1.03
Position of equilibrium moved to the left 5.86 1.17 5.70 1.06 6.01 0.98
Condition 5.65 1.48 6.00 0.94 5.83 1.09
Forward reaction 5.53 1.20 5.40 1.17 5.72 1.10
Pressure 5.45 1.23 5.50 1.08 5.79 1.08
Chemical reaction rate 5.40 1.51 5.20 1.69 6.12 0.98
Rate of the reverse reaction 5.37 1.41 5.20 0.92 5.77 1.03
Rate of the forward reaction 5.19 1.43 5.10 1.52 5.71 1.04
Endothermic reaction 4.98 1.64 4.90 1.60 - -
Exothermic reaction 4.76 1.61 4.50 1.78 - -
Initial concentration 4.68 1.69 5.00 1.25 - -
Activation energy 4.44 1.74 4.50 1.72 - -
Catalyst 3.97 1.81 4.70 1.89 - -
Motionless 3.77 2.01 3.10 1.10 - -
In the end, twenty-four relevant concepts were accepted as a formal concept pool of chemical equilibrium.
Based on the classification and composition relation between those concepts in chemistry, the concept pool was
divided into six categories by the suggestions of chemistry research and teaching practice experts. The names and
components of the categories are shown in Table 2. This six-category model was considered as a scientific model
85
(pp. 80-92) ISSN 2538–7138 /Online/
regarding the domain knowledge of chemical equilibrium. Especially, the concept ‘feature of chemical equilibrium’
summarized by Chinese upper-secondary school chemistry teachers means some main features of chemical equi-
librium. This concept is widely utilized in Chinese upper-secondary school chemistry teaching practice so that it
is familiar to students. Therefore, it should be admitted to the concept pool.
The Conceptual Structure Provided by EFA
Table 3 depicts that the average scores of all items from upper-secondary school students are greater than 5,
meaning that students considered these corresponding concepts be related to the concept ‘chemical equilibrium’.
In the first round of EFA, the value of Bartlett’s test of sphericity was 3494.631 (p < 0.001) and the value of the
Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.930 > 0.70, which meant that the factor analysis
was good. Based on the Eigenvalue greater than 1, four factors that together explained 50.639% of the variance
were extracted. However, five items had low communalities with less than 0.40 and low primary factor loadings
with less than 0.40: ‘Le Chatelier’s Principle’, ‘reaction quotient’, ‘dynamic equilibrium’, ‘concentration’, and ‘feature of
chemical equilibrium’. They were thus eliminated according to the suggestion of ignoring the problematic variables
(Hair, et al., 2010) so that new factor models were respecified, respectively. Moreover, the item ‘establishment of
the equilibrium state’ having a high cross-loading, the items ‘rate of the forward reaction’, and ‘rate of the reverse
reaction’ having low primary factor loadings, were found in new factor solutions and also eliminated.
In the last ninth round of EFA, the value of Bartlett’s test of sphericity was 2197.521 (p < 0.001) and the value
of KMO was 0.918. A three-factor model as shown in Table 4 was extracted as an acceptable factor solution, which
had 16 concepts and accounted for 52.649% of the variance.
Table 4
The Three-factor Model Obtained from EFA
Number of Range of
Factor Items
items loadings
2) Direction of the shift in chemical equilibrium; 10) Pressure; 13) Chemical equilibrium constant;
1 7 0.501 - 0.801 14) Chemical reaction rate; 16) Shift of equilibrium state; 20) Limitation of chemical reaction; 21)
Degree of conversion at equilibrium
7) Condition; 8) Reversibility; 12) Reversible reaction; 15) Forward reaction; 17) Temperature; 18)
2 6 0.447 - 0.838
Reversible process
4) Position of equilibrium moved to the right; 5) Reverse reaction; 11) Position of equilibrium
3 3 0.485 - 0.904
moved to the left
Factor 1 mainly includes the concepts related to the quantitative representation and shift of chemical
equilibrium, which concepts represent the theoretical or actual state of chemical equilibrium. Factor 2 comprises
the concepts having the attribute of ‘reversibility’ or belonging to the ‘condition’ category, and the concept ‘forward
reaction’. Concepts in Factor 3 are related to moving equilibrium in different directions and ‘reverse reaction’. The
three-factor model with 16 relevant concepts represents a complex conceptual structure for students.
The Conceptual Structure (Three-Factor Model) Tested by CFA
The CFA aimed to test the fit between the data and the assumed model, comparing and determining an ap-
propriate and acceptable model. Kline’s (2005) recommendations were utilized as guidelines available for accept-
able model fit in this research: Root Mean Square Error of Approximation (RMSEA) < 0.08, Comparative Fit Index
(CFI) > 0.90, Standardized Root Mean Square Residual (SRMR) < 0.10 (Harrington, 2009). The value of Tucker-Lewis
index (TLI) greater than 0.90 was also utilized as a cut-off.
The three-factor model (Model 1) from EFA was tested with CFA firstly. The fit indices of Model 1 are presented
in Table 5. It was noted that the factor loading of the item ‘pressure’ was less than 0.40. The results indicated that
the model fit of Model 1 was not perfect enough so that Model 1 should be modified. It was an especially com-
mon result that the factor loadings of item ‘pressure’ was below 0.40 when this item was belonging to each factor.
Tabachnick and Fidell (2007) contended that factor loadings above 0.45 are fair and between 0.32 and 0.45 are
86
ISSN 2538–7138 /Online/ (pp. 80-92)
poor (Harrington, 2009). Item ‘pressure’ was eliminated based on the above factor loading criterion. As is revealed
in Table 5, fit indices of the modified three-factor model (Model 2) also suggested the writers modifying this model.
Table 5
The Fit Indices of Models of EFA and Theoretical Assumptions
90% confidence interval

Model S-B χ2 df RMSEA CFI TLI SRMR
for RMSEA
1 242.229* 101 0.068 [0.057, 0.079] 0.879 0.856 0.059

2 220.988* 87 0.071 [0.059, 0.083] 0.878 0.852 0.060
3 168.159* 85 0.057 [0.044, 0.069] 0.924 0.906 0.056
4 191.907* 80 0.068 [0.055, 0.080] 0.898 0.866 0.051
5 137.100* 78 0.050 [0.036, 0.063] 0.946 0.927 0.047
Note: * p < 0.001
The modification index provided by Mplus suggested the writers adding two error covariances between error
terms for item ‘position of equilibrium moved to the left’ and ‘reverse reaction’, and for item ‘reversibility’ and ‘forward
reaction’. Although researchers recommended the user to be aware of the model identification problems, they also
advised the user to take appropriate action on the correlated error terms if justified by theory (Hair, et al., 2010).
Because the reversible reaction is the basis for the shifted position of chemical equilibrium, the concept ‘position
of equilibrium moved to the left’ has a close relation with the concept ‘reverse reaction’. On the other hand, revers-
ibility is the essential feature of a reversible reaction that represents the forward reaction and the reverse reaction.
The concept ‘reversibility’ is closely linked to the concept ‘forward reaction’. Therefore, adding the within-construct
error covariances between error terms of these items was reasonable. The writers accepted the suggestions from
the modification index and derived the new modified three-factor model (Model 3). The model fit results showed
in Table 5 indicated that Model 3 could be considered to have a reasonably good fit (RMSEA = 0.057 < 0.08, CFI =
0.924 > 0.90, TLI = 0.906 > 0.90, SRMR = 0.056 < 0.10).
Furthermore, Model 2 (the nested model) was compared with Model 3 (the comparison model) in a chi-square
difference test. Difference test scaling correction (cd) was 0.781, and Satorra-Bentler scaled chi-square difference test
(TRd) was 97.766, p < 0.001. It indicated that the change to Model 2 resulted in a significant improvement in model fit.
Based on these results, as is shown in Figure 1, Model 3 with 3 factors and 15 relevant concepts was appropriate
and accepted as the conceptual structure for students. The writers named the factors according to the common
meaning of the concepts belonging to their factor. First, Factor 1 was named for ‘quantity and shift’, and the name
of the Factor 2 was ‘reversibility and condition’. Finally, Factor 3 was apt to be named ‘unidirectional shift’.
The Conceptual Structure (Five-Factor Model) Tested by CFA
In Model 3, Factor 1 and 2 could be considered to have two distinct parts in chemistry, respectively. For example,
the concepts of Factor 2 are regarding reversible feature or reaction condition. The writers tried to characterize
the conceptual structure for students with a more detailed model and tested its validity. We advised a new five-
factor model (Model 4) in which the Factor 1 and 2 for Model 3 were divided into respective two factors. In Model
4, Factor 1 includes items 13, 20, and 21, and Factor 2 is comprised of items 2, 14, and 16. Factor 3 consists of items
8, 12, 15, and 18, and Factor 4 comprises items 7 and 17. Other items 4, 5, and 11 are belonging to the factor 5. Fit
indices of Model 4 listed in Table 5 came close to the recommended levels.
The modification index also suggested adding some error covariances between error terms. The writers thus
followed the suggestions and set correlated error terms for the items as what has been done in the modification
of Model 3. The modified five-factor model (Model 5) is shown in Figure 2. Table 5 depicts that Model 5 could be
considered as another appropriate and acceptable model (RMSEA = 0.050 < 0.08, CFI = 0.946 > 0.90, TLI = 0.927 >
0.90, SRMR = 0.047 < 0.10). A Chi-square difference test between Model 4 as the nested model and Model 5 as the
comparison model was conducted. Difference test scaling correction (cd) was 0.678, and Satorra-Bentler scaled
chi-square difference test (TRd) was 117.964, p < 0.001. It represented that Model 5 made a significant improve-
ment in model fit compared with Model 4.
87
(pp. 80-92) ISSN 2538–7138 /Online/
Figure 1 Figure 2
Three-factor Model Five-factor Model
Overall, Model 5 with 5 factors and 15 relevant concepts was considered as the conceptual structure
for students. The names of the five factors were given as follows. First, the name of the Factor 1 was ‘quantitative
equilibrium’ and that of Factor 2 was ‘shift of state’. Second, writers called Factor 3 as ‘reversible feature’. Finally,
Factor 4 and 5 were yet to be named ‘condition’ and ‘unidirectional shift’, respectively.
In general, the model with a lower value of the Bayesian Information Criterion (BIC) means this model is better
fitting than other models. BIC values of models revealed that the model fit of Model 5 (BIC = 11595.839) was fitter
than that of Model 3 (BIC = 11600.467). It meant that Model 5 derived from Model 3 and 4 could better represent
the conceptual structure for students. Although Model 5 had a better structure, the conceptual structures that
comprised three-factor or five-factor were all reasonable and acceptable, based on the combination or separa-
tion of concepts. Following these factor analysis results, we can deeply understand these conceptual structures of
chemical equilibrium held by upper-secondary school students.
Discussion
This exploratory research utilized factor analysis to identify and examine the conceptual structure of chemical
equilibrium in upper-secondary school students, determining a three-factor model and a five-factor model. These
models represented the extracted relation between the relevant concepts. The analysis method adopted in this
research also studied a brand-new framework that could represent the conceptual structure for students. More
attention should be put on the research process and findings.
The Concept Pool
It is important to provide researchers and students with a high-quality concept pool having meaningful and
accurate concepts. In previous research studies (Gorodetsky & Hoz, 1985; Gussarsky & Gorodetsky, 1988; Wilson,
1994, 1996), concept pools were acquired not by the statistical results of large-sample surveys, but by researchers’
private judgments. To avoid the negative effects of personal subjective judgments, the present research was quan-
titatively oriented in selecting the relevant concepts to constitute a reasonable concept pool. As a result, twenty-
four relevant concepts represented chemistry research and teaching practice experts’ collective understanding.
88
ISSN 2538–7138 /Online/ (pp. 80-92)
On the other hand, some concepts in the concept pools did not have a direct relation to chemical equilibrium in
previous research studies. In contrast, the concepts utilized in this research are corresponding to different facets
of chemical equilibrium.
For the reasons, the concept pool developed by the writers is better justified than other concept pools in
previous research studies. Researchers not only can use it as a useful resource to reveal the conceptual structure for
upper-secondary school students but also can use it to design other cognitive tasks to acquire students’ different
understandings of chemical equilibrium in the future.
The Eliminated Concepts
Nine concepts were eliminated in the analysis process because the communalities and the factor loadings
of items corresponding to these concepts were small. In general, small communality of an item means that a
large amount of the variance in the item is not accounted for by the factor solution, indicating that the item has
a low correlation with whole extracted factors. Similarly, low factor loading of an item means the item has a low
correlation with the assumed factor provided by the factor solution. If an item is related to multiple factors at the
same time, this item is impossible to belong to only one factor and bring high correction to other factors. In other
words, factors only have high correlations with a finite number of items so that they cannot explain the eliminated
concepts and share their common connotation.
However, the eliminated concepts, including ‘Le Chatelier’s principle’, ‘reaction quotient’, ‘dynamic equilib-
rium’, and other relevant concepts, are important for upper-secondary school students and university students to
understand the regular pattern and performance of chemical equilibrium (Huddle & Pillay, 1996; MOE, 2003, 2018;
Mutlu & Şeşen, 2016; Quílez-Pardo & Solaz-Portolés, 1995). The concepts ‘Le Chatelier’s principle’, ‘concentration’,
and ‘pressure’ were also part of the conceptual structure in previous research studies (Gorodetsky & Hoz, 1985;
Gussarsky & Gorodetsky, 1988; Wilson, 1994, 1996). We assumed that eliminated concepts existed in students’
structural knowledge, but they had to be eliminated according to the criteria of factor analysis.
The Conceptual Structure
Regarded as the conceptual structures of chemical equilibrium, two appropriate and acceptable models had
a good fit between the assumption and students’ rating data in this research, respectively. The factor analysis re-
sults showed that the conceptual structures included only those concepts having close internal relation with each
other, and constituted a small-scale organization of concepts. Because the conceptual structures were derived and
examined from the rating data of students, they were expected to reflect the latent relation of relevant concepts
in students’ understanding at the group level.
In the five-factor model, the factors represent five essential categories of concepts, making theoretical sense
in chemistry. They are better corresponding to the physical quantity, common behavior, reversible feature, unidi-
rectional change, and influencing condition of chemical equilibrium. In contrast, the three-factor model is more
complicated than the five-factor model. The difference between these models is the way they combine concepts.
We supposed that the concepts might have some unexplored internal relations to make this general organization
of concepts available.
It is generally accepted that a reversible reaction consists of a forward reaction and a reverse reaction. The writ-
ers were aware of the item ‘forward reaction’ categorized to other items having the reversible feature and of the item
‘reverse reaction’ belonging to the factor ‘unidirectional shift’. The findings meant that the students might have their
unique understanding. The cause of these high interrelationships between the concepts remains to be further studied.
The factor model can be viewed as an analytical framework employed to understand chemical equilibrium.
But more evidence for the open-end survey is needed to identify whether students widely utilize this analytical
framework. Compared with the scientific model (shown in Table 2), the three-factor model and the five-factor
model are incomplete and simple. If the scientific model is considered as the learning goal, the students should
have a more comprehensive understanding of the relation between concepts. Teachers need to provide necessary
and supported teaching activities to develop students’ understanding of the concepts.
Basing on different concept pools and techniques, the conceptual structure obtained in this research is not
suitable to compare with those obtained in previous research studies. However, all the forms of conceptual struc-
ture are beneficial for researchers and educators to deeply understand the way students store domain knowledge.
89
(pp. 80-92) ISSN 2538–7138 /Online/
This research focused on the conceptual structure of chemical equilibrium for upper-secondary school students,
using factor analysis to determine the latent structure of relevant concepts. After twenty-four relevant concepts
were selected as a high-quality concept pool, a three-factor model and a five-factor model with 15 relevant con-
cepts were derived from the students’ rating. The new form of conceptual structure helps understand the features
and categories in students’ latent organization of concepts.
Based on the overall results, some implications are given. First, teachers can put more attention to students’
deeper understanding of the conceptual structure of chemical equilibrium. Specifically, teachers can conduct the
concept classification task to make students focus on the classification and composition relation between the
relevant concepts, and to identify which factor model in this research is utilized by students. Moreover, teachers
can mention the relation between the relevant concepts repeatedly during classroom teaching, and especially
emphasize the importance of the eliminated concepts in this research to understand chemical equilibrium. It is
beneficial that students fully understand the relation between the relevant concepts.
Second, further research studies regarding the conceptual structure of chemical equilibrium are needed on
conducting a misconception survey and problem-solving test. The conceptual structure is connected with the
student’s comprehension and misconceptions of chemistry. More research is conducive to know the interplay
among conceptual structure, misconception, and performance in problem-solving. Additional indicators of the
conceptual structure of chemical equilibrium that could be analyzed qualitatively may be sought.
Finally, as a practice example, this research demonstrated the use of factor analysis to provide the latent struc-
ture of science concepts for students. Factor analysis is intended as a complementary method to the conceptual
structure research. More research utilized factor analysis in the chemical equilibrium or other science concepts
can be conducted in the future.
References
Akkus, H., Kadayifci, H., & Atasoy, B. (2011). Development and application of a two-tier diagnostic test to assess secondary
students’ understanding of chemical equilibrium concepts. Journal of Baltic Science Education, 10(3), 146-155.
http://www.scientiasocialis.lt/jbse/?q=node/224
Akkuş, H., Kadayifçi, H., Atasoy, B., & Geban, Ö. (2003). Effectiveness of instruction based on the constructivist approach
on understanding chemical equilibrium concepts. Research in Science & Technological Education, 21(2), 209-227.
https://doi.org/10.1080/0263514032000127248
Barke, H.-D., Hazari, A., & Yitbarek, S. (2009). Misconceptions in chemistry: Addressing perceptions in chemical education. Springer.
Bergquist, W., & Heikkinen, H. (1990). Student ideas regarding chemical equilibrium: What written test answers do not reveal.
Journal of Chemical Education, 67(12), 1000-1003. https://doi.org/10.1021/ed067p1000
Cawley, M. J., Martin, J. E., & Johnson, J. A. (2000). A virtues approach to personality. Personality and Individual Differences, 28(5),
997-1013. https://doi.org/10.1016/S0191-8869(99)00207-X
Chiu, M.-H., Chou, C.-C., & Liu, C.-J. (2002). Dynamic processes of conceptual change: Analysis of constructing mental models of
chemical equilibrium. Journal of Research in Science Teaching, 39(8), 688-712. https://doi.org/10.1002/tea.10041
Collins, A. M., & Loftus, E. F. (1975). A spreading activation theory of semantic processing. Psychological Review, 82(6), 407-428.
https://doi.org/10.1037/0033-295X.82.6.407
Derman, A., & Eilks, I. (2016). Using a word association test for the assessment of high school students’ cognitive structures on
dissolution. Chemistry Education Research and Practice, 17(4), 902-913. https://doi.org/10.1039/C6RP00084C
Dumont, F. (2010). A history of personality psychology: Theory, science, and research from hellenism to the twenty-first century.
Cambridge University Press.
Finley, F. N., Stewart, J., & Yarroch, W. L. (1982). Teachers’ perceptions of important and difficult science content. Science Education,
66(4), 531-538. https://doi.org/10.1002/sce.3730660404
Geeslin, W. E., & Shavelson, R. J. (1975). Comparison of content structure and cognitive structure in high school students’ learning
of probability. Journal for Research in Mathematics Education, 6(2), 109-120. https://doi.org/10.2307/748612
Giamellaro, M. (2014). Primary contextualization of science learning through immersion in content-rich settings. International
Journal of Science Education, 36(17), 2848-2871. https://doi.org/10.1080/09500693.2014.937787
Goldsmith, T. E., Johnson, P. J., & Acton, W. H. (1991). Assessing structural knowledge. Journal of Educational Psychology, 83(1),
88-96. https://doi.org/10.1037/0022-0663.83.1.88
Gorodetsky, M., & Hoz, R. (1985). Changes in the group cognitive structure of some chemical equilibrium concepts following a
university course in general chemistry. Science Education, 69(2), 185-199. https://doi.org/10.1002/sce.3730690211
Gussarsky, E., & Gorodetsky, M. (1988). On the chemical equilibrium concept: Constrained word associations and conception.
Journal of Research in Science Teaching, 25(5), 319-333. https://doi.org/10.1002/tea.3660250502
90
ISSN 2538–7138 /Online/ (pp. 80-92)
Gussarsky, E., & Gorodetsky, M. (1990). On the concept “chemical equilibrium”: The associative framework. Journal of Research in
Science Teaching, 27(3), 197-204. https://doi.org/10.1002/tea.3660270303
Hackling, M. W., & Garnett, P. J. (1985). Misconceptions of chemical equilibrium. International Journal of Science Education, 7(2),
205-214. https://doi.org/10.1080/0140528850070211
Hair, Jr. J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis: A global perspective (7th ed). Pearson
Education.
Harrington, D. (2009). Confirmatory factor analysis. Oxford University Press.
H u d d l e , P. A . , & Pi l l a y, A . E . ( 1 9 9 6 ) . A n i n - d e p t h s t u d y o f m i s c o n c e p t i o n s i n s t o i c h i o m e t r y a n d
chemical equilibrium at a South African university. Journal of Research in Science Teaching, 33(1), 65-77.
https://doi.org/10.1002/(SICI)1098-2736(199601)33:1<65::AID-TEA4>3.0.CO;2-N
Jonassen, D. H., Beissner, K., & Yacci, M. (1993). Structural knowledge: Techniques for representing, conveying, and acquiring structural
knowledge. Lawrence Erlbaum Associates.
Karpudewan, M., Treagust, D. F., Mocerino, M., Won, M., & Chandrasegaran, A. L. (2015). Investigating high school students’
understanding of chemical equilibrium concepts. International Journal of Environmental and Science Education, 10(6), 845-
863. https://doi.org/10.12973/ijese.2015.280a
Kaufman, S. B., Yaden, D. B., Hyde, E., & Tsukayama, E. (2019). The light vs. dark triad of personality: Contrasting two very different
profiles of human nature. Frontiers in Psychology, 10, 467. https://doi.org/10.3389/fpsyg.2019.00467
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed). The Guilford Press.
Maia, P. F., & Justi, R. (2009). Learning of chemical equilibrium through modeling-based teaching. International Journal of Science
Education, 31(5), 603-630. https://doi.org/10.1080/09500690802538045
Medin, D. L. (1989). Concepts and conceptual structure. American Psychologist, 44(12), 1469-1481.
https://doi.org/10.1037/0003-066X.44.12.1469
MoE (Ministry of Education, PRC). (2003). Putong gaozhong huaxue kecheng biaozhun (Shiyan) [National standard of general
upper-secondary school chemistry curriculum (Experimental version)]. People’s Education Press.
MoE (Ministry of Education, PRC). (2018). Putong gaozhong huaxue kecheng biaozhun (2017 Nianban) [National standard of
general upper-secondary school chemistry curriculum (the 2017 version)]. People’s Education Press.
Moss, H. E., Tyler, L. K., & Taylor, K. I. (2007). Conceptual structure. In M. G. Gaskell (Ed.), The Oxford Handbook of Psycholinguistics.
Oxford University Press.
Mutlu, A., & Şeşen, B. A. (2016). Evaluating of pre-service science teachers’ understanding of general chemistry concepts by using
two tier diagnostic test. Journal of Baltic Science Education, 15(1), 79–96.
Nakiboglu, C. (2008). Using word associations for assessing non major science students’ knowledge structure before and after
general chemistry instruction: the case of atomic structure. Chemistry Education Research and Practice, 9(4), 309-322.
https://doi.org/10.1039/B818466F
National Education Examinations Authority. (2018). 2019 Nian putong gaodeng xuexiao zhaosheng quanguo tongyi kaoshi
dagang (Like) [National unified examination outline (Science) of general college entrance examination in 2019]. Higher
Education Press.
Neiles, K. Y., Todd, I., & Bunce, D. M. (2016). Establishing the validity of using network analysis software for measuring students’ mental
storage of chemistry concepts. Journal of Chemical Education, 93(5), 821-831. https://doi.org/10.1021/acs.jchemed.5b00748
Novak, J. D., & Gowin, D. R. (1984). Learning how to learn. Cambridge University Press.
Ollino, M., Aldoney, J., Domínguez, A. M., & Merino, C. (2018). A new multimedia application for teaching and learning chemical
equilibrium. Chemistry Education Research and Practice, 19(1), 364-374. https://doi.org/10.1039/C7RP00113D
Özmen, H. (2008). Determination of students’ alternative conceptions about chemical equilibrium: A review of research and the
case of Turkey. Chemistry Education Research and Practice, 9(3), 225-233. https://doi.org/10.1039/B812411F
Quílez, J. (2004). Changes in concentration and in partial pressure in chemical equilibria: students’ and teachers’ misunderstandings.
Chemistry Education Research and Practice, 5(3), 281-300. https://doi.org/10.1039/B3RP90033A
Quílez-Pardo, J., & Solaz-Portolés, J. J. (1995). Students’ and teachers’ misapplication of Le Chatelier’s principle:
Implications for the teaching of chemical equilibrium. Journal of Research in Science Teaching, 32(9), 939-957.
https://doi.org/10.1002/tea.3660320906
Rosch, E. (2002). Principles of categorization. In D. J. Levitin (Ed.), Foundations of cognitive psychology: Core readings. The MIT Press.
Schinka, J. A., & Velicer, W. F. (Eds.) (2003). Handbook of psychology, Vol. 2: Research methods in psychology. John Wiley & Sons.
Soika, K., & Reiska, P. (2014). Using concept mapping for assessment in science education. Journal of Baltic Science Education,
13(5), 662-673.
Song, X. Q. (Ed.) (2007a). Huaxue 2 [Chemistry 2] (3rd ed). People’s Education Press.
Song, X. Q. (Ed.) (2007b). Huaxue fanying yuanli [Principle of chemical reaction] (3rd ed). People’s Education Press.
Unsworth, L. (2001). Teaching multiliteracies across the curriculum: Changing contexts of text and image in classroom practice.
Open University Press.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed). Allyn and Bacon.
Taber, K. S. (2003). Modeling learners and learning in science education: Developing representations of concepts, conceptual structure
and conceptual change to inform teaching and research. Springer.
Van Driel, J. H., De Vos, W., Verloop, N., & Dekkers, H. (1998). Developing secondary students’ conceptions of chemical
reactions: The introduction of chemical equilibrium. International Journal of Science Education, 20(4), 379-392.
https://doi.org/10.1080/0950069980200401
91
(pp. 80-92) ISSN 2538–7138 /Online/
Wang, L. (Ed.) (2007). Huaxue 2 [Chemistry 2] (3rd ed). Shandong Science and Technology Press.
Wang, M. Z., Gao, P. L., & Wang, L. (Eds.) (2007). Huaxue fanying yuanli [Principle of chemical reaction] (3rd ed). Shandong Science
and Technology Press.
Wang, Z. H. (Ed.) (2014a). Huaxue 2 [Chemistry 2] (5th ed). Jiangsu Education Publishing House.
Wang, Z. H. (Ed.) (2014b). Huaxue fanying yuanli [Principle of chemical reaction] (5th ed). Jiangsu Education Publishing House.
Wilson, J. M. (1994). Network representations of knowledge about chemical equilibrium: Variations with achievement. Journal
of Research in Science Teaching, 31(10), 1133-1147. https://doi.org/10.1002/tea.3660311007
Wilson, J. (1996). Concept maps about chemical equilibrium and students’ achievement scores. Research in Science Education,
26(2), 169-185. https://doi.org/10.1007/BF02356430
Yener, Y., Bahar, M., Demir, N. S., Yılmaz, M., & Yener, D. (2018). The cognitive structure of pre-service science teachers about
energy metabolism. Journal of Baltic Science Education, 17(3), 462-473. http://www.scientiasocialis.lt/jbse/?q=node/674
Zimbardo, P. G., Johnson, R., & McCann, V. (2017). Psychology: Core concepts (8th ed). Pearson Education.
Cite as: Mai, Y., Qian, Y., Li, L., & Lan, H. (2021). The conceptual structure of chemical equilibrium in upper-secondary school
students: Evidence from factor analysis. Journal of Baltic Science Education, 20(1), 80-92. https://doi.org/10.33225/jbse/21.20.80
Yuhua Mai PhD Student, School of Psychology, South China Normal University,
No.55, West of Zhongshan Avenue, Tianhe District, Guangzhou, P.R.China.
Yangyi Qian PhD, Professor, School of chemistry, South China Normal University, No.
(Corresponding author) 378, West Waihuan Street, Higher Education Mega Center, Panyu District,
Guangzhou, P.R.China.
Linshen Li Master of Pedagogy, Teacher, Shimen Senior High School, No. 1,
Hongxing Street, Nanhai District, Foshan, P.R.China.
Haihang Lan Postgraduate Student, School of chemistry, South China Normal
University, No. 378, West Waihuan Street, Higher Education Mega Center,
Panyu District, Guangzhou City, P.R.China.
92
PEDAGOGICAL MODEL FOR
DECOLONISING, INDIGENISING
AND TRANSFORMING SCIENCE
EDUCATION CURRICULA: A
CASE OF SOUTH AFRICA
Abstract. In Africa, Science education
curricula have been instrumental in
promoting Western worldviews as being
universal. An educational transformation
Maxwell Jnr Opoku, and decolonisation of the school curriculum
Angela James is required. A focus on an African worldview
and an integration of the local context and
community-based information is neces-
sary for survival, i.e., Indigenous Knowledge
System (IKS). While IKS is enshrined in the
schooling curriculum, Educators experience
challenges with implementing it, because
the pedagogical strategies have not been
Introduction
clearly described. An in-depth qualitative
study was conducted with the Indigenous
In South Africa, since 1994 when democracy was gained, the curricula at Knowledge (IK) holders of the Zulu cultural
the school and tertiary levels were revised to be inclusive of local community- group and Senior High School (SHS) Science
based knowledge that is used by communities for their survival, Indigenous teachers to explore how IK on environ-
Knowledge. Indigenous Knowledge Systems (IKS), the system of knowledge mental sustainability could be taught in
that various societies developed before the introduction of modern science, South African science classrooms. The
is gaining greater recognition as policy, research, and teaching in the school research employed an interpretivist, multi-
and higher education levels. The national policy on the importance, the IKS site ethnographic, qualitative approach,
policy of 2004 and the current schooling curriculum - Curriculum Assessment and naturalistic research style. In-depth
Policy Statements (CAPS) for Natural, Physical and Life Sciences recognizes interviews were used to generate data
the teaching of IKS in schools. Several higher education institutions have from the purposively selected community
established IKS centres of excellence, notably – University of North West, persons. The thematically analysed findings
University of KwaZulu Natal, University of Limpopo etc (DBE, 2011; DST, were used to develop a culturally specific
2004; Kaya, 2013). These centres serve as hubs for the historical tracing and pedagogical model on how to teach IK in
growth of IKS in research, teaching and community engagement. Recent science classrooms: touring cultural places;
debates on decolonizing and transforming science education in South Africa demystifying indigenous practices and
have traversed from the relevance of indigenous epistemologies in formal perception; utilizing indigenous pedago-
education curricula to ‘how to teach’ indigenous knowledge in schools and gies; teaching wisdom behind indigenous
practices etc. The research recommends that
higher education institutions. Consequently, many scholarships advertising
future studies be conducted on applying the
for ways to rethink the curriculum in both the schooling and higher educa-
model in different geographical and cultural
tion sectors (Chikoko, 2016; De Beer & Petersen, 2016; Gumbo, 2016a, 2016b;
schooling contexts.
Philip Higgs, 2016; Kamwendo, 2016; Kaya & Seleti, 2014; Le Grange, 2016;
Keywords: context sensitive curricula,
Msila, 2016a, 2016b; Msila & Gumbo, 2016; Mudaly, 2018; Shava, 2016; Webb,
pedagogical strategies, South African cur-
2016). The integration of IKS in science education is hoped to counter the
riculum, Western worldview
Western-focused science content with contextually relevant knowledge
from communities in Africa. The access to knowledge for learners is broad-
ened and they can experience their contextual knowledge in the classroom
settings (De Beer & Petersen, 2016; Phillip Higgs, 2016). Moreover, teacher Maxwell Jnr Opoku,
Angela James
education programmes are to be grounded in the philosophical foundations University of KwaZulu-Natal, South Africa
of the indigenous culture of the local community and not to privilege only
westernised scientific knowledge (Aikenhead & Lima, 2009).
93
(pp. 93-107) ISSN 2538–7138 /Online/
Despite acknowledging the value of IK, most science educators hardly infuse IK in their teaching. Reasons given
include less exposure to IK in pre-service teacher education; under-developed IK pedagogical content knowledge;
inadequate understanding of the nature of science and IK; fear of teaching ‘false science’ (Cronje et al., 2015; De Beer
& Petersen, 2016; De Beer & Whitlock, 2009; Mothwa, 2011; Msila & Gumbo, 2016; Ogunniyi, 2007). In the quest for
Africanising, decolonization and transforming the educational system in South Africa, Le Grange (2002, p. 68) states
“a more important concern I wish to raise here is the failure of higher education policies after apartheid to provide
alternative frameworks for knowledge production to those provided by dominant Western knowledge system” In
October 2015, the South African Minister of Higher Education and Training (HET) called for a central African focused
(Africanization) of higher education and persuaded the institutions to shed off troubling aspect of apartheid, to
instead work for the decolonisation of the education curriculum (Le Grange, 2016). The curriculum therefore was
to be reconstructed and refashioned to reflect some of the African traditional and indigenous thoughts, ideas,
epistemologies, ontologies and axiologies. This is also a global concern and a central problem for the 21st century.
This research presents a construction of a pedagogical model for decolonising, indigenising and transforming sci-
ence education curricula, as informed from the cultural perspectives of indigenous environmental sustainability
(IES) IK-holders and science teachers. Govender et al. (2016, p. 182) argued that a ‘scholarly encounter with IKS
and IK holders could enhance the process and style with which academics ‘conduct research and conceptualise
education’ as IK holders could tremendously contribute towards the policy formulation and implementation for
addressing any gaps in moving from a solely western perspective to an integrated IKS and western perspective.
Purpose of the Research
The main purpose of this research was to explore and present a model for how indigenous environmental
sustainability could be taught in the science education curricula, as viewed from the IK holders and Science teachers
in South Africa and why they have the particular views on how to teach indigenous environmental sustainability.
The research is to contribute knowledge to the scholarship on recent call for africanisation, decolonisation, indi-
genisation, and transformation of the science education curricula (DBE, 2011; DST, 2004).
Literature Review
Education in South Africa and IKS
Many scholars highlight the status of the Western worldview in African educational systems since colonial
days to the democratic era. Castagno and Brayboy (2008) contended that Westernised science has no recognition
for other bodies of knowledge like IK and indigenous ontologies. Shava (2016) asserts that the Southern African
school agriculture syllabus is replete with foreign examples of fruit crops and captures insignificant examples of
indigenous plants. Odora Hoppers (2006) added that the colonialists disregarded and devalued the African be-
liefs and practices. Anwar (2011) states that the colonialist colluded to eliminate African IK and made Westernised
beliefs, practices and epistemologies predominate, thereby hegemonising black Africans formal education. Msila
and Gumbo (2016) contended that many Africans were influenced by Western ideologies, knowledge and practices
that they internalized to the disdain of their own beliefs and practices.
This is a major concern that calls for intensive scholarly research in African IKS. Furthermore, Msila and Gumbo
(2016) assert that there has been deliberate and gross neglect of African cultures and IKS in curricula. Meanwhile,
there are other ways of knowing by which students could explore their natural environment. For instance, the nature
of science and IK share many principles in common (Cronje, 2015; Fortuin, 2017). In recent times, the contestation of
denigration and subjugation of IK as well as issues of relevance of IK in school and higher education in the Republic
of South Africa is gradually giving way for issues of how the teaching of IK could happen. But science educators
and curriculum developers are confronted by various challenges in the quest to decolonise and indigenise the
South African science education curricula (Opoku & James, 2020b). Msila and Gumbo (2016, p. 10) posited that in
post-colonial Africa, education needs to be transformed through educational decolonisation and reconstruction,
as “knowledge is transformed, reconstructed and rewritten to celebrate differences, diversity, pluralism, multiplicity
and heterogeneity” without elevating one form of knowledge over another. Thus, the need for a pedagogy and
curricula that reflect African epistemic experience, through reclamation of indigenous African epistemologies in
curriculum is indispensable.
94
ISSN 1648–3898 /Print/ PEDAGOGICAL MODEL FOR DECOLONISING, INDIGENISING AND TRANSFORMING SCIENCE
ISSN 2538–7138 /Online/ (pp. 93-107)
Models for Indigenising Education Curricula
In the quest for the educational transformation and sustainable Africanisation of the curricula, several models
and or theories, approaches for indigenizing the education curricula have been proposed by some scholars:
Julius Nyerere’s philosophical model on indigenising curriculum (Kadenyi & Kariuki, 2011; Kassam, 1994; Na-
songo & Musungu, 2009); the Performance Partnership and Co-existence approach proposed by Turnbull (1997)
and expounded on by Le Grange (2002) to include creation of the knowledge space for practicable coexistence of
Western and African indigenous epistemologies in educational institutions; Adedeji (1981) Adaptation approach
and expounded on by Mokhoba (1999) to include adapting to the values and needs of the specific society; the
African Philosophy approach – renaissance of philosophies such as Afrocentricity, ubuntu, Africanism, African hu-
manism (Bennaars, 1990, 1998; Luthuli, 1982; Palermo, 1997); the Curriculum Engineering approach (Cross, 2004;
Shanyanana & Ndofirepi, 2015); the Separatist Development Approach (Le Grange, 2002); the Culture Product
Indigenization Process Model; Mazrui seven pillars of wisdom and model for indigenising the university curricu-
lum (Msila & Gumbo, 2016); the incorporationist, the separatist and the integrationist approaches to teaching IKS
(Naidoo & Vithal, 2014).
Theoretical Constructs
In recent years, the inestimable value of African IKS in addressing current environmental and social crisis has
been highlighted. For instance, Gyekye’s 1997 cultural revivalism African philosophy that contends for renaissance
of the hegemonised African indigenous practices and traditions is believed to be an essential tool for addressing
current economic, sociopolitical and environmental challenges. African minds are to be purged from colonialist
global north thought pattern (Ciaffa, 2008). However, cultural revivalism adherents have been criticised that their
philosophy rather hinders advances in science and technology in that the perspectives are old and less potent to
handling current African challenges (Famakinwa, 2012). Despite the criticism of cultural revivalism, several other
scholars believe that African traditions have depths of intellectual knowledge relevant for African societies. Con-
sequently, African epistemologies and cultural identity must be restored and re-appropriated considering that
many Africans have been incapacitated socially, economically, intellectually, religiously, politically, technologically
by the European colonialist. Africans are called upon to re-appropriate AIK (Dei, 2017; Hountondji, 2009; Kiti, 2013;
Le Grange, 2016).
This study is in line with the critical ontological and indigenous knowledge theories which contends that when
educators integrate IK into curriculum, it enhances learning and makes IK applicability very relevant to indigenous
people. This assertion is premised on the fact that learning is socially and culturally contextualized; an approach that
guides curriculum development in line with socio-cultural theories. Critical ontological theory contends against
curricula that is bereft of IKS. Rather, it contends for the emancipation on indigeneity and renaissance of indigenous
people’s confidence (Choy & Woodlock, 2006; McLaughlin & Whatman, 2008; Michie, 1999, 2002). Nhalevilo-Afonso
(2013) contended that it is imperative to look for legitimated theories to integrate IKS to remedy the practice of
teaching IKS in science classrooms, which is already disconnected from its own socio-cultural context. Consequently,
Nhalevilo-Afonso (2013) proposed a five stage subjective framework for interrogating the incorporation of IKS in
African school curricula, in which Nhalevilo-Afonso “reflects on framework curriculum changes and programmes
of research into cultural contextualization of science education and/or of IKS inclusion in school curricula”.
The stages include colonisation (“the period when IK had no recognition” p. 25); Decolonization (“the period
when awareness of the value of IK began to take centre stage in debates on curriculum policies” p. 26); Neo-colo-
nialism (“new kind of colonialism that undermine the cultural values through teaching IKS in a decontextualized,
expropriated and objectified manner” p. 27); Re-birth (“this interrogates the lenses through which IKS is commu-
nicated and the interest being served by specific IKS content and pedagogies and the way in which IKS” p. 29); and
Theorization (“the stage that aims to address ontological, axiological, and epistemological issues for including IKS
in school curricula” p. 31). In theorizing, Nhalevilo-Afonso (2013) posited that the crucial question for interrogation
in the theorisation stage is “How do we teach that knowledge – IKS”. The theorization stage question, according to
Nhalevilo-Afonso (2013, p. 31) “represents a particular and very important challenge for Africa”, one great challenge
being the “gap that might exist between theory and practice” (p. 32). The current research is situated in the Postco-
lonial theory which advocates for culturally relevant and practicable curricula in parts of the world that experienced
95
(pp. 93-107) ISSN 2538–7138 /Online/
colonialism, like South Africa. The Postcolonial theory contends against the superimposition of Western ideolo-
gies and practices that denigrate indigenous beliefs, cultures and practices (as subjugated knowledge, irrelevant
and unproven). The (African) philosophical perspectives discussed above undoubtedly ‘influence the philosophy,
direction, content, and pedagogy that undergird education systems’ (Mudaly, 2018, p. 51) and weaves well into
the social reconstructionist theory which contends against social inequality in educational curricular. Moreover,
adherents of the social reconstructionist theory perceive cultural factors to strongly frame human experiences and
therefore calls for a re-thinking of all facets of curriculum development and relevant pedagogical strategies such
as instructional planning, evaluation and methodologies to be employed (Schiro, 2015).
Research Design
This research utilized a qualitative research approach, and an ethnographic naturalistic research style, located
within the interpretivist paradigm. The research explored the views of the Zulu (South African) IK holders and Sci-
ence Educators on how to teach Indigenous Environmental Sustainability (IES) in formal educational institutions
and why they have these views. The research results were used to present a model of how to teach it. Creswell
(2017) indicates that a decision to adopt a certain paradigm and research design is predicated upon its suitability
to fulfilling purpose of study. The adoption of an ethnographic study approach was to explore the worldviews of
the Zulu IK holders and Science educators on the phenomena under study. It was naturalistic in the sense that, the
research was conducted in the participants’ natural environment and niche because the objective was to observe
and describe the research participants’ worldviews on the focus of the research in their particular contexts. The
design provided for the collection of rich, detailed qualitative data for description (Cohen et al., 2013; Olsen, 2011).
The interpretivist paradigm encapsulates interpretation of multiple realities and was deemed appropriate in this
study (Bertram & Christiansen, 2014; Quaye, 2007).
Research Context
The research was conducted among the Zulu cultural group in some local communities within the Zululand
District Municipality: Nongoma, Ulundi, uPhongolo and Nkandla (King Cetshwayo district), in the Province of
KwaZulu-Natal in 2018. The choice of the participants, the localities and guidance for getting access to participants
was informed partly by literature, the DSI-NRF Centre in IKS offices in UKZN and recommendations from other
participants in a snowball fashion. Zulu cultural practices are still relevant in these areas among others (Adeyemi,
2012; Creswell, 2017). The participants included Zulu IK-holders and science educators. The Zulu IK-holders whose
transcripts and narratives were used for the research were 21 people of African origin: three chiefs (ndunas), three
diviner-spiritualists (sangoma), three diviner-herbalists (inyanga), three elders (umdala); three youth (ubusha – the
isiZulu word for youth); three science educators from rural areas. Three science educators from township and city
areas were included. The participants were engaged in an in-depth conversational interview where there was dia-
logue. The number of participants was considered large enough to generate data for analysis and small enough
to work with considering the pure qualitative interpretive nature of the data (Cohen et al., 2013; Creswell, 2017;
Denzin et al., 2008; Miles & Huberman, 1994).
Data Gathering and Instrument
The key instrument for data generation were open-ended conversational interview questions, which permitted
participants to express their views without restriction and to talk at length, in a relaxed manner (Burgess-Limerick
& Burgess-Limerick, 1998; Denzin et al., 2008). Knowles and Cole (2008) contended that interpretivist researchers
majorly employ interviews in their quest to explore and describe people’s perceptions, worldviews and understand-
ings about a peculiar phenomenon. Kaya and Seleti (2014, p. 33) view the interview as an important data generation
instrument for exploring IK related phenomena and people’s attitudes and IKS beliefs. The interview questions were:
“How do you think you (the participant) and your cultural group’s understandings, perceptions and practices for the value
and care for nature (lands, plants, animals, waterbodies) should be taught. Why do you have these views?”
96
ISSN 2538–7138 /Online/ (pp. 93-107)
Data Analysis
The data was analysed inductively, using thematic content analysis based on themes, concepts and similar
features. Bamberg (2012) stated that during transcription, analysis has already begun. The transcript of each partici-
pant was read to identify patterns, regularities and commonalities in the responses of all other participants and then
organised in categories and colour-coded. The general categories were based on participants’ responses that were
connected to the social, cultural, economic, political, religious, technological, biophysical aspects of the phenomena
under study (how to teach indigenous environmental sustainability). Themes were then identified by sorting and
re-categorising the initial general categories into more in-depth and more specific categories. Moreover, untypi-
cal dissenting views were also identified in the process. Again, in the quest to enhance the theme identification,
the texts that had not been colour-coded (not linked to identified themes) were re-scrutinised in search for new
themes. All the themes identified in the entire analysis process were then discussed and supported with relevant
literature (Attride-Stirling, 2001; Bertram & Christiansen, 2014; Miles & Huberman, 1994).
Rigour of the Research
Rigour of the research was ensured with trustworthiness by employing multiple data generation sources and
triangulation (Rule & John, 2011). Triangulation was ensured through using the same conversational interview
questions for different categories of participants – IK holders and science educators (Creswell, 2017).
Limitations of the Research

The purposively selected participants had in-depth knowledge about the Zulu culture using a snowball
participant selection technique and did not prioritize gender nor specific age brackets in the selection process.
Consequently, the findings of this research could not be discussed on that basis. The research was conducted in
an African language, isiZulu language, which was translated to British English by a language translator and an
IKS expert in higher education research, whose mother tongue is isiZulu. There were instances where responses
were transliterated for want of exact word matches because the researcher did not speak the African language of
isiZulu. Some scholars assert that most researchers who deal with multi-lingual research studies are confronted
with interpretive dilemmas (Regmi et al., 2010; Vithal & Jansen, 2012).
Research Results
Themes were generated from the data and these include visiting IK-holders and Touring Cultural places within
local communities. The research findings are presented and discussed based on the identified and established
themes (see Table 1). Participant categories are coded (Chief – C1, C2, C3; Diviner-spiritualist -DS1, DS2, DS3;
Deviner-herbalist – DH1, DH2, DH3; Elder – E1, E2, E3; Youth – Y1, Y2, Y3; Teacher in rural area – Tr1, Tr2, Tr3; Teacher
in town/city – Tc1, Tc2, Tc3). Participant code highlighted in bold (example C1 for chief 1) shows the chief’s response
believed to elaborate on the corresponding theme. In instances where participants in the same category made
similar statements, one of the quotes was selected at random and presented to support the theme. In an attempt
to reduce the volume of data presented to support findings and for reasons of data saturation, only a single quote
is presented for each of the seven different participant categories.
97
(pp. 93-107) ISSN 2538–7138 /Online/
Table 1
How to Teach Indigenous Epistemologies (e.g., Environmental Sustainability) in Science Education Curricular
Categories of
Participants and
Themes Percentage (%)
Number
Visiting IK-holders and Touring Cultural Places within Local Communities All participants 100
Inviting IK-holders to School Science Classrooms to Teach as Experts of Indigenous

All participant 100
Epistemologies
Teaching the Indigenous Epistemologies (e.g., of IES) with Indigenous Language and
All participants 100
Indigenous Teaching Strategies
Conscientising Learners to Remember and Cherish Heroes of their Cultural Heritage All participants except
95
as a Decolonisation Measure Y3
Demystifying Mysteries Surrounding Indigenous Epistemologies and teaching the All the participants
90.5
Wisdom behind them except DS1, DH2
Developing Computer Programmes for Indigenous Epistemologies All participants 100
Students and Learners Wearing their Cultural Costumes, Patronising Indigenous All participants except
95
Products Y3
Mixing the Westernised knowledge on (e.g., Environmental Sustainability) with Indig- All participants except
95
enous Knowledge on (e.g., Environmental Sustainability) E3
Visiting IK-holders and Touring Cultural Places within Local Communities
Participants indicated that the cultural places in the local communities be toured and IK-holders visited.
These are captured in the following excerpts:
The chief (induna) said, “…the teachers and their learners can pay us a visit in the places (cultural villages, rural areas
etc) where they can easily interact with the knowledge holders…this will let the learners appreciate their cultural values
and easily embrace them…” (C1, C2, C3).
The elder (umdala) said, “…I suggest the school visit the cultural villages and spend some time with some of the experi-
enced IK-holders of the Zulu culture…because at a distance you may have misconceptions about the knowledge-holders
and some practices…” (E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…people normally have wrong perception about our work…because they
usually don’t come to see how we do our things…so the schools could organize themselves and come to visit us here and
we will gladly teach them what our ancestors say about the things we have in nature…”(DS1, DS2, DS3).
The diviner-herbalist (inyanga) said, “…it is important that the learners get to know what we do…so they do not feel
ashamed of us…I think the teachers and their learners could organize themselves on planned visit to meet us to teach
them what we actually do…how we care for the lands, waterbodies, plants, animals…which they will not hear in school…
”(DH1, DH2, DH3).
The youth (intsha) said, “…I do think that having an excursion to local places and teaching learners there will be the best
since this aspect has to do with the nature studies …they could even visit the cultural villages to appreciate the knowledge
of cultural people …” (Y1, Y2, Y3).
Teacher (uthisha) said, “…I do think that if the schools tour places where there are cultural practices like the ones related
to the natural environment…I am of the view that it will go a long way to enhance the teaching process…” (Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…tourism and excursion are ways of getting cultural knowledge of a people’s environment…
therefore if there is some frequent excursions and tour to some of the places where such cultural things are still relevant, it
will boost learners’ interest…” (Tr1, Tr2, Tr3).
Inviting IK-holders to School Science Classrooms to Teach as Experts of Indigenous Epistemologies
Participants asserted that their elders (IK-holders) to be invited to the academic institutions to help in the
teaching of their IK on environmental sustainability.
98
ISSN 2538–7138 /Online/ (pp. 93-107)
The chief (induna) said, “…knowledge holders should be exposed to the learners by calling upon them to come to the
schools regularly to teach them lots of these practices that helped us keep our environment…” (C1, C2, C3).
The elder (umdala) said, “…when we invite elders who have in-depth knowledge about our culture, especially those who
have high formal education background, it would boost the interest of the learners…”(E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…we have the knowledge of our ancestors and they keep talking to us to able
to help our community…the schools can invite us to come and teach the learners some of the things and it will make them
more interested in the cultural things…”(DS1, DS2, DS3).
The diviner-herbalist (inyanga) said, “…you have to use experts in everything you do so you can get the best results…
the teachers can organize regular programs and call us who have more knowledge in the traditional and cultural things
to teach the learners…”(DH1, DH2, DH3).
The youth (intsha) said that, “…I think that the schools can organize some forums where they could invite the either the
chiefs, sangomas, inyangas, elders and other traditional leaders to help in teaching the learners the cultural ways of caring
for nature…” (Y1, Y2, Y3).
Teacher (uthisha) said, “…one of the best pedagogical strategies is to get a resource person in any technical area that you
are teaching, and this will enhance the teaching and enable free flow of ideas, so I suggest the IK-holders of the Zulu cultural
be invited for such teaching…”(Tr1, Tr2, Tr3).
Teacher (uthisha) said, “… I will suggest that experts in the field, that is IK-holders of the amaZulu culture be invited to
assist in teaching the learners…”(Tr1, Tr2, Tr3)
Teaching Indigenous Epistemologies with Indigenous Language and Indigenous Teaching Strategies
Participants called for the use of their indigenous language and their indigenous teaching strategies or
pedagogies to guide the teaching of the cultural (IK) on Environmental Sustainability. The following are some
assertions made:
The chief (induna) said, “…because of politics we cannot use only our local language, isiZulu in our schools but it should be
used to teach our children…they should teach the learners the way we were taught by our elders and ancestors…” (C1, C2, C3).
The elder (umdala) said that, “…if the teachers could teach the learners the wisdom of our ancestors on how we protect
the nature and, in the process, use the isizulu language, it will make the learners love the cultural things you teach them…”
(E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…we should not throw away our language and use other people’s language
in our schools…it will make it easy and interesting…our own way of teaching (such as proverbs, stories, dreams from
ancestors) are what they (teachers) should use but they can add their own…it is our wisdom, knowledge they should
teach…”(DS1, DS2, DS3).
The inyanga (diviner-spiritualist) said, “…how can you teach someone their cultural practice without using their spoken
language...even some of the wild plants and wild animals we do not even know their English names, how can we tell you about
them with English…you should teach the things in our culture,…use same way we teach our people…” (DH1, DH2, DH3).
The youth (intsha) said, “…I believe that using the isiZulu language to teach will even make the teaching nice and interest-
ing…the elders they teach you everything whether spiritual things or physical things so that you will be cautious in your
actions to the natural environment…”(Y1, Y2, Y3).
Teacher (uthisha) said, “…in teaching aspects like this, using the local language will enhance it… the IK-holders would be
the best teachers…we as teachers should endeavor to learn the way the IK-holders teach using stories, proverbs, fairy tales,
dreams, taboos etcetera…we could even find our own way of incorporating other strategies…” (Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…the indigenous language of the people have all the deep things (the deep knowledges and
wisdom)…it must be the mode of teaching although we can intersperse with English but at least 80 percent of it must be
isiZulu…the teaching methods could be diversified…” (Tr1, Tr2, Tr3)
Conscientising Learners to Remember and Cherish Heroes of their Cultural Heritage as a Decolonisation Measure
Almost all participants (except Y2) reiterated the need for the academia to recognise the impact of the ancestors
and heroes of their cultural group (the Zulus) and show gratitude to them and remember them in their teaching.
The teachers and the learners are to decolonise their minds in such efforts. The following extracts expound this:
The chief (induna) said, “…teachers should make their learners love, cherish, appreciate what is their own…their culture
is their culture…they cannot take other people’s culture…and always remember what their elders and ancestors have
done for them to still have their culture intact and protected their natural resources (the lands, forests, water-bodies etc)
and made available for them…” (C1, C2, C3).
The elder (umdala) said that, “…when learners are made to know much about their cultural values, how the ancestors
have fought to keep their cultural values and natural resources…it rids their minds of the thought of having subservient
culture…because some of our learners sometimes feel the white child is better than them…” (E1, E2, E3).
99
(pp. 93-107) ISSN 2538–7138 /Online/
The diviner-spiritualist (sangoma) said, “…learners should know the value of their culture so that they will appreciate
us and the ancestors for great work they have done for us…they fought long and strong battles to keep what we still have
in nature now…we should give them due respect...” (DS1, DS2, DS3).
The diviner-herbalist (inyanga) said, “…teachers and learners should all become conscious of the fact that their culture
is their culture and their traditional leaders are their leaders…they cannot throw away their culture and leaders now and
those heroes…” (DH1, DH2, DH3).
The youth (intsha) said, “…everyone in our culture should understand that our ancestors have done a lot for us…those
in school should be confident of their culture and not feel the lifestyle of the Europeans, Americans, Asians are the only
acceptable ways of doing things …”(Y1, Y2, Y3).
Teacher (uthisha) said, “…appreciating one’s culture should also be a key area to consider in the in the teaching process…
because any disinterest will make learners not be attentive to what will be taught…teachers and their learners, should
recognize the efforts of the ancestors in ensuring the resilience of their culture and preservation of natural resources…”(Tr1,
Tr2, Tr3).
Teacher (uthisha) said, “…learners would need to fall in love with the cultural practices and accept that their culture is
their culture…the ancestors and the elders who ensured that our culture and natural resources are still preserved should
be commended…and remove from our minds any colonialist damaging thought about our culture…” (Tc1, Tc2, Tc3).
Demystifying Mysteries Surrounding Indigenous Epistemologies and Teaching the Wisdom behind them
Participants (90.5%) except for DS1 and DH2 indicated the need for teaching the wisdom behind their general
Indigenous practices and perception related to their cultural value and care for nature. The following are excerpts:
The chief (induna) said, “…it is not everything that you the teacher can tell the learners because they are young and some
of the things are spiritual…you will need the wisdom of the elders (IK-holders) to be able to explain such things…do not
put fear in them about the spirits in many of the natural things…just teach the knowledge and wisdom of placing value
and caring for the natural resources…” (C1, C2, C3).
The elder (umdala) said , “…our elders say a lot of things about nature that are spiritual… some scary though, but we
took it to heart because they were here on earth before us and might know better…we respected them…we didn’t enter
forests they said we shouldn’t enter…the spiritual part that the school teachers do not know about, they should invite the
elders (IK-holders) to come to the schools to teach…” (E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…there are lots of things that the teacher will not be able to explain very well
in school about our cultural practices unless the ancestors give them the knowledge and wisdom because there are a lot of
spiritual things that are connected with nature…teachers should not tell the learners anything that frightens them about
our natural resources though some of the spirits manifest in waterbodies, plants and animals…”(DS1, DS2, DS3).
The diviner-herbalist (inyanga) said, “…for instance some of the plant used for normal physical things can also be used
for other spiritual things…the teachers should therefore teach the physical aspect and benefits that they know and invite
us when it comes to the spiritual aspects…” (DH1, DH2, DH3).
The youth (intsha) said, “…our elders rarely tell you the reason behind some of the things they tell us not to do…they
sometimes tell you scary stuff about some plants, water-bodies, lands, forests, wildlife…I think this should be changed and
when teaching in schools the reasons behind things should be told the learners…” (Y1, Y2, Y3).
Teacher (uthisha) said, “…many of the cultural ways for taking care of nature are based on spiritual connectedness…
sometimes they are scary and can cause fear in the learners about some natural components…notwithstanding, the elders
have been able to use that to take care of many natural resources…”(Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…our cultural people’s way of caring for nature are mostly linked to spiritual attachment and such
can cause fear in learners…such aspects should be removed…we should find out what reasoning the elders had before
saying those spiritual things connected to some natural resources…” (Tr1, Tr2, Tr3)
Developing Computer Software Programmes for Indigenous Epistemologies
Participants called for the development of computer programmes or software or App for the Zulu indigenous
practices and perceptions such as their cultural values and care for nature. The following are some assertions made
by participants:
The chief (induna) said that, “…our generation now use computers and mobile phones…I suggest that the schools find
ways of developing special computer program that the learners can use to study the cultural peoples’ values and care which
they place on the lands, water-bodies, forest and other things in nature…” (C1, C2, C3).
The elder (umdala) indicated, “…learners are more interested in computer and phones than our traditions…and may be
the reason why they are not respecting the wisdom of the elders…some computer programmes could be made that contains
the wisdoms, practices, values and our cultural ways of caring for our natural environment…” (E1, E2, E3).
The diviner-spiritualist (sangoma) said that, “…now people believe more in computers, television and phones so they
100
ISSN 2538–7138 /Online/ (pp. 93-107)
do not even bother themselves about our culture values and what our ancestors say about our water-bodies, plants and
animals…so I think the school can create something for our cultural values with computer for our learners…” (DS1, DS2, DS3).
The diviner-herbalist (inyanga) said that, “…some people think that traditional healing is not good because we are not
using computers to heal and check the body of people as other doctors do…since learners are school where computers are
used, it will be better if government and the school leaders create some computer programme that has our cultural things
for them to learn…”(DH1, DH2, DH3).
The youth (intsha) said, “…we love computers games and use many google apps today…I think that if some kind of a
computer programme or a google play store app is developed that has all our the various cultural practices…it can make
the teaching very effective and interesting…” (Y1, Y2, Y3).
Teacher (uthisha) said, “…ICT is becoming a must in every classroom teaching and I am of the view that if a kind of a
computer software is developed for cultural value and care for nature it will be more attractive to the current generation
of learners…” (Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…some modernized interactive tool…could be a particular apps or computer software that has
detailed aspects of the cultural values and care for nature with details of the wisdoms of the elders…this may pique the
interest of the learners more…” (Tr1, Tr2, Tr3).
Learners Wearing their Cultural Costumes and Patronising Indigenous Products
Participants suggested that teachers encourage their students and learners to wear their cultural costumes
and regalia and eat indigenous foods to encourage the teaching of their indigenous practices and perceptions
such as their IES. Some participants made the following statements:
The chief (induna) said, “…our learners must be encouraged to wear our cultural clothing regularly and consume our
local foods often…” (C1, C2, C3).
The elder (umdala) indicated, “…the interest of our learners will be piqued in the teaching if they enjoy wearing our
traditional costumes and eat our local foods more…” (E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…our culture is endowed with many rich values…our traditional clothes are
wonderful and gorgeous…our foods are rich and nutritious and do not give people diseases like the fast foods…please let
our learners wear our cultural clothes and eat our foods…” (DS1, DS2, DS3).
The diviner-herbalist (inyanga) said, “…we have to inculcate our cultural things into our children…For instance our
children should be encouraged and allowed to wear our traditional clothes in school and eat local foods...” (DH1, DH2, DH3).
The youth (intsha) said, “…I think it will look trending, if we go to school with our traditional wear and enjoy our traditional
games and foods as well…” (Y1, Y2, Y3).
Teacher (uthisha) said, “…Cultural clothing could be worn especially on day of teaching our indigenous environmental
value and care for nature…learner should be encouraged to consume local foods…” (Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…It would be lovely if we keep creating the conscious of using and patronizing our local produced
items and wear our traditional clothes as formal dresses too…”(Tr1, Tr2, Tr3)”
Miixing the Westernised Knowledge (on Environmental Sustainability) with that of Indigenous Knowledge
The research participants called for infusion of their IES ethics with the Westernised ES. Participants articulated that:
The chief (nduna) said, “…the teachers should teach our cultural ways of caring for nature together with what is already
been studied in schools…” (C1, C2, C3).
The elder (umdala) said, “…the teachers should not only teach our local ways of taking care of the earth but mix with the
ones they teach in school…”(E1, E2, E3).
The diviner-spiritualist (sangoma) said, “…It will be good if our teachers add what we do in our culture to take care
of nature to the ones from other cultures that the schools teach so that our learners will know more…” (DS1, DS2, DS3).
The diviner-herbalist (inyanga) indicated that, “…the knowledge and wisdoms of our elders that we use in taking care
of the plants, animals, waterbodies, forests, lands should be taught together with the ones that the ones in the school and
not be thrown away…it’s good to have knowledge in many things…”(DH1, DH2, DH3).
The youth (intsha) said, “…these days even music when you blend with that of other foreign ones, they are unique and
lovely…so I suggest that the teachers combine how our cultural people care for nature with that of the westernized ones
in our schools…” (Y1, Y2, Y3).
Teacher (uthisha) said, “…an integrated approach to the teaching of the cultural environmental ethics with that of the
westernized ones will be better I believe…”(Tr1, Tr2, Tr3).
Teacher (uthisha) said, “…in my view we should find an appropriate strategy to interface the nature studies with that of
our local culture’s care and value for natural resources…”(Tr1, Tr2, Tr3).
101
(pp. 93-107) ISSN 2538–7138 /Online/
Discussion
The participants called on teachers to organise their learners and tour local communities with them
to the cultural places; engage with the local community members and most especially interact with elders
(IK-holders) to teach some of the Zulu people’s knowledge and wisdom regarding their natural environment.
Mosimege (2005) contended that cultural villages could provide educational purposes. Mawere (2015) as-
serted that integrating IKS in schools would promote the connection of learners with their culture. Khupe
(2014) indicated that learners could go to places like museums where they could learn about indigenous
knowledge. The wealth of (indigenous) knowledge of the IK-holders was appraised by the participants, and
reiterated that the schools invite them over to the schools to teach some of the Zulu cultural values and
care for the natural environment and the IK connected to their livelihood, in this research the environmental
sustainability.
Michie (1999) indicated a development of curricula in Canada where elders were called upon for guid-
ance and information. Simpson (2002) called for involvement of the elders in teaching the cultural curriculum,
while Hewson et al. (2009) contended for collaborating with IK holders and hopefully inviting them to the
classroom.
Participants were of the view that, for effective teaching of culturally related knowledge and indigenous
practices (for instance, IES) the teaching should be done in a similar fashion to that of their IK-holders with
a level of modification that fits with the formal schooling system. Most notable in participants’ expres-
sion was the use of the indigenous (isiZulu) language as medium of instruction in all such teachings, for
instance on their indigenous cultural values and care for nature. Kaya et al. (2016, p. 126) contended that
“African indigenous languages such as isiZulu contributes to the promotion of African scholarship, African
Renaissance and Africa-led globalization...and impart African cultural value systems in society, education
and socio-economic development…provide students access to African indigenous knowledge and value
systems”. Nhalevilo-Afonso (2013) called for teaching content relevant to the cultural setting of the learners.
The need for teachers and their learners making conscious effort to place great value on their heroes and on
their cultural ways and caring for nature was reiterated by the participants. The cultural group are to show
gratitude to the ancestors and heroes of the isiZulu culture for their bravery in fighting to keep their natural
resources, their geographical land etc., and making the isiZulu culture resilient. Moreover, for the teaching
of culturally relevant content to be effective, all and sundry within the academy ought to decolonise their
mind by rejecting all subjugated thoughts about the isiZulu culture, which was precipitated by the colonialist.
The general view of the participants, especially the teachers and the youth were that their IK-holders
usually make statements and tell stories about certain natural resources which are connected to spirits and
metaphysical beings which are scary. The participants called on teachers to dispel unexplained and frighten-
ing statements attached to indigenous teaching on natural resources. Research by Opoku and James (2020)
posited that teaching spiritual aspects of IK would pose a challenge to teachers as learners would demand
proof and experiments to prove some claims connected to the natural environment. Shava (2013) contends
that spiritual aspects are usually excluded from the curriculum because they do not conform to western ways
of education on basis of being holistic. The development of a computer programme or software or google
apps that has Zulus’ cultural values and care for nature was suggested by the participants. The software
could even be an interactive programme like google app game that could be put on phones to stir up the
interest of the learners and to bring a certain consciousness that the cultural value and care for nature is a
contemporary area of study. Moreover, the research participants reiterated that as a way of upholding their
cultural values and cultural heritage, the learners, including their teachers should be encouraged to wear their
traditional cultural costumes and regalia as well as patronize indigenous foods and locally produced products.
Doing this, is thought of to boost interest in teaching indigenous content in formal school setting. The
participants emphasised the need for infusing the Westernised ES with their IES contending that much as
their indigenous cultural values are to be upheld and highlighted, the already existing Westernised ES taught
in the schools should not be removed but be integrated with the IES. Several scholarships have highlighted
the need for integrating Western science with IKS (Bohensky et al., 2013; Mashoko, 2018; McPherson et al.,
2016; Verma et al., 2016).
102
ISSN 2538–7138 /Online/ (pp. 93-107)
Based on the research findings, the analysis and discussions, (See Figure 1) a pedagogical model and
guide for teachers, instructors and educators who have the quest to decolonizing, indigenising and trans-
forming school science curricula at both basic and higher education institution is developed.
Figure 1
MJ Opoku’ Culturally Pedagogical Model for Decolonising, Indigenising and Transforming Science Education Curricula
This research explored how indigenous environmental sustainability could be taught in the science
education curricula, as viewed from the IK holders and science teachers in South Africa and why they have
the particular views on how to teach indigenous environmental sustainability. The research is to contribute
knowledge to the scholarship on recent call for africanisation, decolonisation, indigenisation and transfor-
mation of the science education curricula. The research results indicate the need to contextualise formal
education processes for enhancing epistemological access for indigenous learners.
Based on the results of the research and the proposed pedagogical model, the cultural villages of the local
communities are to be toured to promote interaction between educators, learners, students and community
members especially the IK holders and to promote community engagement in general. When educators in-
teract with their learners, they should inspire to cherish the heroes and legends of their cultural heritage who
fought to keep their culture resilient. The IK holders need to be invited and brought into the formal science
classrooms to teach the IK on e.g., indigenous value and care for nature.
In the teaching process, the IK holders are to be encouraged to teach using their unique culturally rel-
evant pedagogical strategies, their local indigenous language and teach their indigenous epistemologies,
ontologies and axiologies. The content taught should be demystified and the wisdom behind the spiritual
connotations are to be taught. Learners should be encouraged to wear their cultural costume and regalia
and motivated to patronise foods and their cultural products. Moreover, a computer-based programme or
software or a google app containing all the local indigenous epistemologies, ontologies and axiologies should
be developed as learning programmes, games etc.
The indigenous content taught, should not be done in isolation to the already westernised science taught
in the schools but be infused and taught in an integrated manner but give more prominence and emphasis
on the indigenous epistemologies.
103
(pp. 93-107) ISSN 2538–7138 /Online/
The research recommends for further research into the development of resource materials for the diverse
cultural groups and the use of these materials in the respective settings. Teachers and education officials
require extensive interactions with the environmental ethics from an indigenous knowledge perspective
through working with IKS holders who have the wisdom and willingness to share their knowledge.
References
Adedeji, A. (1981). Indigenisation of African economies. In Ismailova, B. (2010). Curriculum reform in post-Soviet Kyrgyzstan:
Indigenization of the history curriculum. The Curriculum Journal, 15(3), 247-264. https://doi.org/10.1080/09585170412331311501
Adeyemi, S. (2012). Performing myths, ritualising modernity: Dancing for Nomkhubulwana and the reinvention of Zulu
tradition. In O. S. Omoera, S. Adeyemi, & B. Bineba (Eds.), A Gazelle of Savannal Sunday Ododo and the framing of techno-
cultural performance in Nigeria: Theatre, culture, literature, art (Vol. 2, pp. 435-445). Alpha Crownes Publishers Ltd.
https://doi.org/https://gala.gre.ac.uk/id/eprint/10255/1/Adeyemi-Performing_Myths.pdf
Aikenhead, G. S., & Lima, K. E. (2009). Science, culture and citizenship: Cross-cultural science education. Revista Brasileira de
Pesquisa em Educação em Ciências, 9(3), 1-15. https://periodicos.ufmg.br/index.php/rbpec/article/view/3998
Anwar, A. (2011). African indigenous knowledge systems - challenges and opportunities. Africa Insight, 40(4), 136-148.
https://journals.co.za/content/afrins/40/4/EJC17670
Attride-Stirling, J. (2001). Thematic networks: An analytic tool for qualitative research. Qualitative Research, 1(3), 385-405.
https://doi.org/10.1177/146879410100100307
B a m b e rg, M . ( 2 0 1 2 ) . N a r r a t i ve a n a l y s i s. APA handbook of research methods in psychology, 2, 7 7 - 9 4 .
https://doi.org/10.1002/9781405198431.wbeal0075
Bennaars, G. A. (1990). Towards a pedagogy of hope. WAJIBU, 5(2), 2-4. https://journals.co.za/content/wajibu/5/2/AJA10169717_1136
Bennaars, G. A. (1998). Schools in need of education: Towards an African pedagogy. In Itolondo, W. A. (2012). The role and status
of Christian religious education in the school curriculum in Kenya. Journal of Emerging Trends in Educational Research and
Policy Studies, 3(5), 721-729. https://journals.co.za/content/sl_jeteraps/3/5/EJC127684
Bertram, C., & Christiansen, I. (2014). Understanding research: An introduction to reading research. Van Schaik.
B o h e n s k y, E. L . , B u t l e r, J. R . A . , & D av i e s, J. ( 2 0 1 3 ) . I nte grat i n g i n d i g e n o u s e co l o gi c a l k n ow l e d g e
and science in natural resource management: Perspectives from Australia. Ecology and Society, 18(3), 1-6.
https://doi.org/http://dx.doi.org/10.5751/ES-05846-180320
Burgess-Limerick, T., & Burgess-Limerick, R. (1998). Conversational interviews and multiple-case research in psychology. Australian
Journal of Psychology, 50(2), 63-70. https://www.tandfonline.com/doi/abs/10.1080/00049539808257535
Castagno, A. E., & Brayboy, B. M. J. (2008). Culturally responsive schooling for indigenous youth: A review of the literature. Review
of Educational Research, 78(4), 941-993. https://doi.org/https://doi.org/10.3102/0034654308323036
Chikoko, V. (2016). Issues in Africanising higher education curricula. In V. Msila & M. T. Gumbo (Eds.), Africanising the curriculum:
Indigenous perspectives and theories. (pp. 71-82). African Sun Media https://books.google.co.za/books?id=Teq9DwAAQBA
J&dq=Africanising+the+curriculum:+Indigenous+perspectives+and+theories&lr=
Choy, S., & Woodlock, J. (2006). When tensions lead to change: meeting the training needs of indigenous learners and communities.
In Proceedings of the (Re) Contesting Indigenous Knowledge & Indigenous Studies Conference 2006 QUT, Oodgeroo Unit., 1-20.
https://eprints.qut.edu.au/7023/1/7023.pdf
Ciaffa, J. A. (2008). Tradition and modernity in postcolonial African philosophy. Humanitas, 21(1/2), 121-145.
https://css.cua.edu/wp-content/uploads/2018/01/ciaffa21-1.pdf
Cohen, L., Manion, L., & Morrison, K. (2013). Research methods in education (7th ed.). Routledge Falmer.
Creswell, J. W. (2017). Research design: Qualitative, quantitative, and mixed methods approaches. Sage Publications.
http://www.drbrambedkarcollege.ac.in/sites/default/files/research-design-ceil.pdf
Cronje, A. (2015). Epistemological border-crossing between Western science and indigenous knowledge and its implications for
teacher professional development. In de Beer, J., & Petersen, N. (2016). Decolonisation of the science curriculum: A different
perspective (# cookbook-labs-must-fall). http://uir.unisa.ac.za/bitstream/handle/10500/22869/Josef%20de%20Beer,%20
Neal%20Petersen.pdf?sequence=1
Cronje, A., De Beer, J., & Ankiewicz, P. (2015). The development and use of an instrument to investigate science teachers’ views
on indigenous knowledge. African Journal of Research in Mathematics, Science and Technology Education, 19(3), 319-332.
https://doi.org/ https://doi.org/10.1080/10288457.2015.1108567
Cross, M. (2004). Institutionalising campus diversity in South African higher education: Review of diversity scholarship and
diversity education. Higher education, 47(4), 387-410. https://link.springer.com/article/10.1023/B:HIGH.0000020854.04852.80
DBE. (2011). Department of Basic Education: Curriculum and Assessment Policy Statement Grade 10-12. Life Sciences.
https://www.education.gov.za/Portals/0/CD/National%20Curriculum%20Statements%20and%20Vocational/CAPS%20
FET%20_%20LIFE%20SCIENCES%20_%20GR%2010-12%20Web_2636.pdf?ver=2015-01-27-154429-123
De Beer, J., & Petersen, N. (2016). Decolonisation of the science curriculum: A different perspective (# Cookbook-Labs-Must-
Fall). http://uir.unisa.ac.za/bitstream/handle/10500/22869/Josef%20de%20Beer,%20Neal%20Petersen.pdf?sequence=1
De Beer, J., & Whitlock, E. (2009). Indigenous knowledge in the Life Sciences classroom: Put on your De Bono hats! The American
Biology Teacher, 71(4), 209-217.
104
ISSN 2538–7138 /Online/ (pp. 93-107)
Dei, G. J. (2017). An indigenous Africentric perspective on black leadership: The African scholar today. In Dei, G. J.
S. (2017). In Reframing Blackness and Black Solidarities through Anti-colonial and Decolonial Prisms Springer, 4.
https://www.thekalazys.com/press/pdf/files/Reframing%20Blackness%20and%20Black%20Solidarities%20Through%20
Anti-Colonial%20and%20Decolonial%20Prisms.pdf
Denzin, N. K., Lincoln, Y. S., & Smith, L. T. (2008). Handbook of critical and indigenous methodologies. Sage.
DST. (2004). Department of Science and Technology. 2004. Indigenous Knowledge Systems Policy. In Botha, L. R. (2012).
Using expansive learning to include indigenous knowledge. International Journal of Inclusive Education, 16(1), 57-70.
https://doi.org/https://doi.org/10.1080/13603110903560093
Famakinwa, J. (2012). Revisiting Kwame Gyekye’s critique of normative cultural relativism. Thought and Practice, 4(1), 25-41.
https://www.ajol.info/index.php/tp/article/view/81215
Fortuin, E. (2017). Crossing worldviews in the Life Sciences classroom: A case of indigenous teachers and learners in South Africa
Oslo and Akershus University College of Applied Sciences]. https://oda-hioa.archive.knowledgearc.net/bitstream/
handle/10642/5525/Fortuin.pdf?sequence=2&isAllowed=y
Govender, N., Mudaly, R., & James, A. (2016). Engaging indigenous knowledge holders in teaching preservice
teachers in IKS food production and practices: Implications for higher education. Alternation, 23(1), 180-207.
https://journals.ukzn.ac.za/index.php/soa/article/view/1345/1503
Gumbo, M. T. (2016a). Contesting technology education curriculum for the schooling of African learners in South Africa. In V.
Msila & M. T. Gumbo (Eds.), Africanising the curriculum: Indigenous perspectives and theories (pp. 97-120). African Sun Media.
Gumbo, M. T. (2016b). A model for indigenising the university curriculum: A quest for educational relevance. In V. Msila & M. T.
Gumbo (Eds.), Africanising the curriculum: Indigenous perspectives and theories (pp. 33-56). African Sun Press.
Hewson, M. G., Javu, M. T., & Holtman, L. B. (2009). The indigenous knowledge of African traditional health practitioners and the
South African science curriculum. African Journal of Research in Mathematics, Science and Technology Education, 13(1), 5-18.
https://doi.org/10.1080/10288457.2009.10740647
Higgs, P. (2016). The African Renaissance and the decolonisation of the curriculum. In V. Msila & M. Gumbo (Eds.), Africanising the
curriculum: Indigenous perspectives and theories (pp. 1-16). African Sun Media.
Higgs, P. (2016). The African renaissance and the transformation of the higher education curriculum in South Africa. Africa
Education Review, 13(1), 87-101. https://doi.org/10.1080/18146627.2016.1186370
Hountondji, P. (2009). Knowledge of Africa, knowledge by Africans: Two perspectives on African studies. Journal Revista Crítica
de Ciências Sociais, 1(1), 121-131. https://doi.org/10.4000/rccsar.174
Kadenyi, M., & Kariuki, M. (2011). Rethinking education for liberation and self-reliance: An examination of Nyerere’s and Plato’s
paradigm. In Tesha, M. W. (2018). Effectiveness of teaching and learning agricultural science subject in selected secondary
schools in Tanzania Sokoine University of Agriculture]. http://www.suaire.suanet.ac.tz:8080/xmlui/handle/123456789/2143
Kamwendo, G. H. (2016). Unpacking Africanisation of higher education curricula: Towards a framework. In V. Msila & T. Gumbo
(Eds.), Africanising the curriculum: Indigenous perspectives and theories (pp. 17-36). African Sun Media.
Kassam, Y. (1994). Julius Kambarage Nyerere. Prospects, 24(1-2), 247-259. https://link.springer.com/article/10.1007%2FBF02199019
Kaya, H. (2013). Integration of African indigenous knowledge systems into higher education in South Africa: Prospects and
challenges. Alternation, 20(1), 135-153. http://utlo.ukzn.ac.za/Files/Alternation%2020.1%20(2013).pdf#page=140
Kaya, H., Kamwendo, G. H., & Rushubirwa, L. (2016). African indigenous languages in higher education. Studies of Tribes and
Tribals, 14(2), 121-128. https://doi.org/https://doi.org/10.1080/0972639X.2016.11886739
Kaya, H., & Seleti, Y. N. (2014). African indigenous knowledge systems and relevance of higher education in South Africa.
International Education Journal: Comparative Perspectives, 12(1), 30-44. https://files.eric.ed.gov/fulltext/EJ1017665.pdf
Khupe, C. (2014). Indigenous knowledge and school science: Possibilities for integration. University of the Witwatersrand, Faculty
of Science, School of Science Education.
Kiti, P. (2013). Recentring Africa: From Paulin Hountonji to Meinrad Herbga. In Mudaly, R. (2018). Towards decolonising a module
in the pre-service science teacher education curriculum: The role of IKS in creating spaces for transforming the curriculum.
Journal of Education, 74, 47-66. https://doi.org/https://doi.org/10.17159/2520-9868/i74a04
Knowles, J. G., & Cole, A. L. (2008). Handbook of the arts in qualitative research: Perspectives, methodologies, examples, and issues. Sage.
Le Grange, L. (2002). Challenges for Higher Education Transformation in South Africa: Integrating the local and the
global. South African Journal of Higher Education, 16(1), 67-73. https://www.ingentaconnect.com/content/sabinet/
high/2002/00000016/00000001/art00011
Le Grange, L. (2016). Decolonising the university curriculum: Leading article. South African Journal of Higher Education, 30(2),
1-12. https://dx.doi.org/10.20853/30-2-709
Luthuli, P. (1982). An introduction to Black-oriented education in South Africa. In Alexander, G., & Mpisi, A. (2014). Issues influencing
black learners’ scholastic experiences in ex-model C further education and training schools. Journal of Sociology and Social
Anthropology, 5(3), 349-359. https://doi.org/ https://doi.org/10.1080/09766634.2014.11885639
Mashoko, D. (2018). Integrating indigenous knowledge of food preservation with school science teaching in Zimbabwe University of
the Witwatersrand]. https://pdfs.semanticscholar.org/5102/79318ee436d5964d0de1e5e318595cab3f19.pdf
Mawere, M. (2015). Indigenous Knowledge and Public Education in Sub-Saharan Africa. Africa Spectrum, 50(2), 57-71.
https://journals.sagepub.com/doi/full/10.1177/000203971505000203
McLaughlin, J. M., & Whatman, S. L. (2008). Embedding Indigenous perspectives in university teaching and learning:
Lessons learnt and possibilities of reforming/decolonising curriculum. Indigenous Education: Asia/Pacific, 123-146.
https://doi.org/https://eprints.qut.edu.au/27213/
105
(pp. 93-107) ISSN 2538–7138 /Online/
McPherson, J. M., Sammy, J., Sheppard, D. J., Mason, J. J., Brichieri-Colombi, T. A., & Moehrenschlager, A. (2016). Integrating
traditional knowledge when it appears to conflict with conservation: lessons from the discovery and protection of sitatunga
in Ghana. Ecology and Society, 21(1), 1-13. https://doi.org/ http://dx.doi.org/10.5751/ES-08089-210124
Michie, M. G. (1999). Where are indigenous peoples and their knowledge in the reforming of learning, curriculum and pedagogy Fifth
UNESCO-ACEID International Conference” Reforming learning, curriculum and pedagogy: Innovative visions for the new
century”. Bangkok, Thailand. http://members.ozemail.com.au/~mmichie/aceid.html
Michie, M. G. (2002). Why Indigenous science should be included in the school science curriculum. Australian Science Teachers
Journal, 48(2), 36. https://search.informit.com.au/documentSummary;dn=200206975;res=IELAPA
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. Sage.
Mokhoba, M. (1999). The African university and its challenges in transformation. No easy road: Transforming higher education
in South Africa. In Cross, M., Mhlanga, E., & Ojo, E. (2011). Emerging concept of internationalisation in South African higher
education: Conversations on local and global exposure at the University of the Witwatersrand (Wits). Journal of studies in
international education, 15(1), 75-92. https://journals.sagepub.com/doi/abs/10.1177/1028315309334635
Mosimege, M. (2005). National priorities in Indigenous Knowledge Systems: implications for research and curriculum development.
Indilinga African Journal of Indigenous Knowledge Systems, 4(1), 31-37. https://www.ingentaconnect.com/content/sabinet/
linga/2005/00000004/00000001/art0000
Mothwa, M. M. (2011). Teachers’ experiences of incorporating indigenous knowledge in the life sciences classroom University of
Johannesburg]. ujcontent.uj.ac.za
Msila, V. (2016a). Africanisation of education and the search for relevance and context: Indigenous perspectives and theories. African
Sun Press.
Msila, V. (2016b). Revival of the university: Rethinking teacher education in Africa. In Africanising the curriculum: Indigenous
perspectives and theories (pp. 195-205). African Sun Media.
Msila, V., & Gumbo, M. T. (2016). Africanising the curriculum: Indigenous perspectives and theories. African Sun Media.
Mudaly, R. (2018).Towards decolonising a module in the pre-service science teacher education curriculum:The role of IKS in creating spaces
for transforming the curriculum. Journal of Education, 74, 47-66. https://doi.org/https://doi.org/10.17159/2520-9868/i74a04
Naidoo, P. D., & Vithal, R. (2014). Teacher approaches to introducing indigenous knowledge in school science
classrooms. African Journal of Research in Mathematics, Science and Technology Education, 18(3), 253-263.
https://doi.org/https://doi.org/10.1080/10288457.2014.956407
Nasongo, J., W, & Musungu, L. L. (2009). The implications of Nyerere’s theory of education to contemporary education in Kenya.
Educational Research and Reviews, 4(4), 111. https://www.tandfonline.com/doi/abs/10.1080/10288457.2014.956407
Nhalevilo-Afonso, E. Z. d. F. (2013). Rethinking the History of Inclusion of IKS in School Curricula: Endeavouring to Legitimate the
Subject. International Journal of Science and Mathematics Education, 11(1), 23-42. https://doi.org/10.1007/s10763-012-9382-8
Odora Hoppers, C. A. (2006). Centre for African renaissance studies, the academy, the state and civil society: Methodological
implications of transdisciplinarity and the African perspective. International Journal of African Renaissance Studies - Multi-,
Inter- and Transdisciplinarity, 1(1), 33-52. https://doi.org/https://doi.org/10.1080/18186870608529705
Ogunniyi, M. B. (2007). Teachers’ Stances and Practical Arguments Regarding a Science-Indigenous Knowledge Curriculum: Part 1
[Article]. International Journal of Science Education, 29(8), 963-986. https://doi.org/https://doi.org/10.1080/09500690600931020
Olsen, W. (2011). Data Collection: Key debates and methods in social research. Sage.
Opoku, M. J., & James, A. (2020). Challenges of Teaching Akan (Ghana) Culturally-specific Environmental Ethics in Senior High
Schools: Voices of Akan IK-holders and Biology Teachers. Southern African Journal of Environmental Education, SAJEE, 36(2020),
33-49. https://doi.org/10.4314/sajee.v36i1.4
Palermo, J. (1997). Reading Asante’s Myth of Afrocentricity: An ideological critique. In Janetius, S. T., Workie, B., & Mini, T. C. (2012).
Indigenising knowledge base for consequential education in ethiopia–a conceptual model. Insight on Africa, 4(2), 103-119.
https://doi.org/https://doi.org/10.1177/0975087814411151
Quaye, S. J. (2007). Voice of the researcher: Extending the limits of what counts as research. Journal of Research Practice, 3(1), 3.
http://jrp.icaap.org/index.php/jrp/article/view/60/129
Regmi, K., Naidoo, J., & Pilkington, P. (2010). Understanding the processes of translation and transliteration in qualitative research.
International Journal of Qualitative Methods, 9(1), 16-26. https://doi.org/https://doi.org/10.1177/160940691000900103
Rule, P., & John, V. (2011). Your guide to case study research. Van Schaik Pretoria.
Schiro, M. S. (2015). Curriculum theory: Conflicting visions and enduring concerns (2nd ed.). Sage. https://www.sagepub.com/sites/
default/files/upm-binaries/16446_SchiroWebExtensionActivities.pdf
Shanyanana, R. N., & Ndofirepi, A. P. (2015). Contestation in the 21st century African university: Local and global challenges and
opportunities. Mediterranean Journal of Social Sciences, 6(1 S1), 53. http://www.richtmann.org/journal/index.php/mjss/
article/view/5508/5312
Shava, S. (2013). The representation of Indigenous knowledges. In International handbook of research on environmental education
(pp. 384-393). Routledge.
Shava, S. (2016). The application/role of indigenous knowledges in transforming the formal education curriculum: Cases from
Southern Africa. In V. Msila & M. T. Gumbo (Eds.), Africanising the curriculum: Indigenous perspectives and theories (pp. 121-
138). SUN Press.
Simpson, L. (2002). Indigenous environmental education for cultural survival. Canadian Journal of Environmental Education (CJEE),
7(1), 13-25.https://cjee.lakeheadu.ca/article/view/271/160
106
ISSN 2538–7138 /Online/ (pp. 93-107)
Turnbull, D. (1997). Reframing science and other local k nowledge traditions. Futures, 29(6), 551-562.
https://doi.org/10.1016/S0016-3287(97)00030-X
Verma, P., Vaughan, K., Martin, K., Pulitano, E., Garrett, J., & Piirto, D. D. (2016). Integrating Indigenous Knowledge and Western
Science into Forestry, Natural Resources, and Environmental Programs [Article]. Journal of Forestry, 114(6), 648-655.
https://doi.org/https://doi.org/10.5849/jof.15-090
Vithal, R., & Jansen, J. (2012). Designing your first research proposal: a manual for researchers in education and the social sciences.
Juta and Company.
Webb, P. (2016). Indigenous knowledge and science education: What knowledge, whose knowledge and how do we bridge the
gap. In V. Msila & M. T. Gumbo (Eds.), Africanising the curriculum: Indigenous perspectives and theories (pp. 177-194). Sun Press.
Received: June 24, 2020 Accepted: January 15, 2021
Cite as: Opoku, M. J, & James, A. (2021). Pedagogical model for decolonising, indigenising and transforming science education
curricula: A case of South Africa. Journal of Baltic Science Education, 20(1), 93-107. https://doi.org/10.33225/jbse/21.20.93
Maxwell Jnr Opoku PhD, Lecturer, UKZN, Edgewood Campus, Pinetown, Republic
(Corresponding author) of South Africa.
Website: http://ukzn.ac.za
Angela A James PhD, Senior Lecturer, UKZN, Edgewood Campus, Pinetown,
Republic of South Africa.
Website: http://ukzn.ac.za
107
THE PREDICTORS OF NINTH
GRADE STUDENTS’ ATTITUDES
TOWARDS PLANTS
Abstract. As plants are vital to the survival Meryem Selvi,

of the planet, developing positive attitude Emel Çelepçıkay İslam
towards them is important for protection of
plants, biodiversity, and the environment.
The aim of this research was to determine
the predictors of ninth grade students’ atti-
tude towards plants. In the study, a descrip- Introduction
tive correlational design was used. The data
were collected from 773 upper-secondary Plants play a major role in maintaining the balance of life on earth, not
students selected through stratified- only for humans but for all living things. However, despite the improvements
in quality of life that have been achieved through developments in technol-
purposive sampling. The “Plant Recognition
ogy and industry, the need for more residential areas due to the growth in
Test” and “Plant Attitude Scale” were used
world population have threatened the number of green spaces. This reduction
as data collection tools. Multiple regression
of green spaces in cities limits the interaction people have with nature and
analysis was performed to determine the prevents them from knowing more about it. It has been shown that plant
way different variables predicted attitude recognition levels have decreased today compared to those of the past (Ju &
towards plants. The results of research indi- Kim, 2011). A positive correlation exists between the efforts individuals make
cated that the linear relationship between towards protecting nature and being responsible stewards and the extent
predictive variables and attitude towards to which nature is recognized by people. The first and most important step
plants explained 14% of the total variance towards achieving better recognition levels is to have people experience
in attitude towards plants at a significant nature. In other words, living in harmony with nature helps individuals to
level. Among the variables, gender, time know more about it.
spent in a garden, involvement in indoor The attitudes and perceptions individuals have about the environment
have been reported to be largely shaped by their attitudes and perceptions
or outdoor cultivation, experience with
about experience with living organisms (Tunnicliffe & Reis, 2000). Attitude is
picking flowers, fruits, and/or vegetables,
defined as “the pre-disposition of a mental, emotional and behavioural reac-
and plant recognition level had significant
tion organized by the individual based on their experience, knowledge, emo-
contributions to predicting variance in at-
tion and motivation towards any object, social subject or event” (İnceoğlu,
titude towards plants, and the recognition 2010, p. 13). A positive attitude generally translates into positive behaviour
of plants was determined to be the most towards the object of attitude (Aydın, 2002), as the individual who has a posi-
important predictor relative to all the other tive attitude towards an object will tend to behave positively towards that
variables. object, sympathize with it, support and help it (Taylor et al., 2007). Therefore,
Keywords: descriptive correlational the positive attitude individuals have towards plants shapes their behaviours
design, plant recognition, secondary in a positive way and contributes to the protection of plants, biodiversity, and
education the environment. Current and future generations can have a better under-
standing of the meaning of biodiversity and its relationship to the natural
environment by gaining awareness about plant species (Lindemann-Matthies,
Meryem Selvi 2006), as the interest and care shown to organisms have been reported to
Gazi University, Turkey increase as students learn more about them (Lindemann-Matthies, 2005).
Emel Çelepçıkay İslam
Sakarya Yönder Schools, Turkey
Prokop et al. (2016) reported that the use of live plants in botanical lessons,
considering their fruits, colours and toxicity, was effective in improving stu-
dents’ knowledge and attitudes. It was also stated that the use of toxic plants
108
ISSN 1648–3898 /Print/ THE PREDICTORS OF NINTH GRADE STUDENTS’ ATTITUDES TOWARDS PLANTS
(pp. 108-118)
in botany is effective in increasing interest in plants as it creates a perception of survival in individuals (Prokop &
Fančovičova, 2018), also survival-relevant information and fruit coloration enhance the retention of knowledge
(Prokop & Fančovičova, 2014). Using some interesting examples of useful plants (Pany, 2014) and interesting fea-
tures of plants (Strgar, 2007) can increase students’ interest in plants.
Research Focus
While science education literature includes studies examining students’ knowledge and understanding of
plants, these studies tend to focus heavily on plant growth, development, nutrition and photosynthesis (Barman et
al., 2003; Barman et al., 2006; Fančovičova & Prokop, 2010; Köse et al., 2006; Köse et al., 2009). In research conducted
with students that aimed to examine their ability to name and recognize plants, students were shown to have
limited knowledge and ability on this subject (Barman et al., 2003; Barman et al., 2006; Bebbington, 2005; Gatt et
al., 2007; Patrick & Tunnicliffe, 2011; Tunnicliffe & Reiss, 2000; Türkmen et al., 2003). In addition, many studies have
shown that children and adults are more interested in animals than in plants, and that their perceptions and at-
titudes towards plants are generally indifferent (Fančovičova & Prokop, 2010, 2011a; Schussler & Olzak, 2008; Strgar,
2008; Tunnicliffe, 2001; Tunnicliffe & Reis, 2000; Wandersee, 1986; Wandersee & Schussler, 1999) as “plant blindness”,
meaning that plants are given secondary importance, and that individuals have a lack of awareness about plants
and are unable to comprehend the importance of plants on earth and to evaluate the unique biological features
of life forms belonging to the world of plants. In other words, plant blindness refers to the inability of individuals
to recognize plants in their daily lives and to take care of them and an overall lack of focus on plants (Wandersee
et al., 2006). Parsley (2020) proposed the term “plant awareness disparity” instead of the term “plant blindness”. She
stated that the term plant awareness disparity is more appropriate because it is not a disability metaphor, it does
not mean that people have never seen plants, and this situation can be corrected by education. Also, the concept
of disparity means that people do not notice the plants around them as often as animals. It has been stated that
the disparity in this term leads to individuals who do not have an interest and positive attitude towards plants
and do not appreciate the importance of plants. This is especially concerning considering that “plants are vital to
the survival of the planet, in terms of ecology, biodiversity and climate” (McCormick & Tijan, 2010, p.1021). The
protection of plant diversity can only be achieved by loving and recognizing plants, and by developing a positive
attitude towards them (Allen, 2003).
In the literature, there are a few studies examining the variables governing the students’ attitudes towards
plants. Lohr and Pearson-Mims (2005) found that both passive and active interactions with plants in childhood
were effective factors in the positive attitudes of individuals towards trees. Krosnick et al. also (2018) stated that the
experience of growing their own plants increased the students’ interest in plants. Fancovicova and Prokop (2010)
found that students living in garden houses had more positive attitudes towards plants, but there was no differ-
ence in terms of gender. Also, Strgar (2008) and Kaplan and Topsakal (2013), revealed that students’ gender does
not affect attitudes towards plants. However, Lohr and Pearson-Mims (2005) and Özel et al. (2013) concluded that
female students’ attitude towards plants is more positive in terms of gender. In another research (Özarslan, 2019) it
was determined that the general attitude scores of gifted and talented students towards plants were significantly
higher than the attitude scores of the students with normal development. By determining the dynamics of attitude
towards a specific subject, it is possible to gain a better understanding of how attitudes function and to observe
the attitude change processes; with this information, the behaviours of individuals can be altered through the
influence of their attitudes (Baysal & Tekarslan, 1996). Therefore, it is important to determine the factors that affect
the attitude towards plants. Previous studies have emphasized different variables that affect the attitude towards
plants. In this research, the variables that are stated to be effective on attitudes towards plants in the literature
were considered and it was examined how much these variables together predicted the attitude towards plants.
Research Aim and Research Question
The aim of research was to determine the predictors of ninth grade students’ attitudes towards plants. To
achieve this, present study sought the answer to the following question: Do gender, the presence of a home garden,
the presence of a green field around the home, experience with indoor or outdoor cultivation, spending time in a
green field and/or picking flowers, fruits, and vegetables, grades in school, and plant recognition level significantly
predict ninth grade students’ attitudes towards plants?
109
(pp. 108-118)
General Background
A descriptive correlational design was used to identify the relation between variables affecting the attitudes
towards plants. Descriptive correlational designs describe variables and investigate nature of relations between
and among variables (Sousa et al., 2007). The research was carried out during the spring semester of the 2015/2016
academic year.
Sample
The research sample included 773 ninth-grade students from three different socio-economic regions of Ankara
who were selected through stratified-purposeful sampling. Bujang et al. (2017) reported that “minimum sample size
of 500 or more will yield reliable and valid sample estimates for the intended population” for performing multiple
regression analysis. Therefore, all 9th grade students in selected schools were included in the sample, considering
that there may be missing data. Prospective research participants were informed about the research before data
collection. The principle of voluntary participation was taken into consideration. Participants identity was protected
using anonymity and confidentiality. The distribution of the students by grade and gender is given in Table 1.
Table 1
The Students’ Distribution by School and Gender
Gender
School Total
Female Male
School A (High grades) 104 81 185

School B (Average grades) 170 117 287
School C (Low grades) 154 147 301
Total 428 345 773
Instruments and Procedures
Personal Information Form
The aim behind the use of the personal information form was to collect information about the independent
variables considered to be related to the attitude towards plants. Lohr and Pearson-Mims (2005) found that both
passive and active interactions with plants in childhood were effective factors in the positive attitudes of individu-
als towards trees. As in the stated research to determine the students’ passive and active interactions with plants,
questions about the presence of a home garden, a green field around the home, experience with indoor or outdoor
cultivation, spending time in a green field, and picking flowers, fruits and/or vegetables were included on this form.
For all the questions on this form, a 4-point Likert-type scale, with response options of “Often”, “Sometimes”, “Rarely”
and “Never”, was used. For the analysis of data “Often” was coded as 4, “Sometimes” was coded as 3, “Rarely” was
coded as 2 and “Never” was coded as 1.
Plant Attitude Scale
The Turkish version of the Plant Attitude Scale (PAS), originally developed by Fancovicova and Prokop (2010),
with 29 items arranged under four factors, ‘interest’, ‘importance’, ‘urban trees’ and ‘utilization’, was used in this
study to understand the students’ attitudes towards plants. The scale was adapted by Selvi (2012) for use in the
Turkish culture. The alpha internal consistency coefficients calculated for the scores obtained from the 28-item
Turkish version were .76 for the interest factor, .75 for the importance factor, .58 for the urban trees factor, and .59
for the utilization factor.
110
(pp. 108-118)
The alpha internal consistency coefficients calculated for the scores obtained from this study are .81 for the
interest factor of the scale, .73 for the importance factor, .78 for the urban trees factor, and .61 for the utilization factor.
Plant Recognition Test
For the plant recognition test, three faculty members who are experts in seed plants were consulted for their
opinions. The plants that the students were familiar with and most likely to have encountered in their environ-
ment were selected. Rather than being interesting or having a distinctive feature, the criteria for the selected 22
plants were plants that students would regularly see in gardens, parks or streets around them. The test included
two pictures of each of the first 12 plants, with one of the pictures being presented at a wide angle to allow all the
parts to be seen clearly, and the other being presented with its fruit, flower and leaf. For the other ten plants, one
picture was considered to be sufficient for recognition in order to increase the intelligibility of the pictures. Special
attention was given to making the pictures as clear and intelligible as possible. A space was provided under each
of the pictures for the names of the plants to be written. Plants included in the test are given in Table 2.
Table 2
Plants Included in Plant Recognition Test and Percentage of Plants Recognized
Percentage of plants
Latin Name English Name
recognized
Rosa sp. Rose 97.5
Matricaria sp Daisy 95.8
Morus alba Mulberry tree 82.6
Eleagnus sp. Elaeagnus 74.1
Parthenocissus tricuspidata Ivy 45.2
Viola sp. Violet 43.3
Taraxacum sp. Dandelion 40.8
Pinus sp Pine tree 37
Populus sp. Poplar tree 35.5
Salix sp. Willow tree 34.3
Aesculus hippocastanum Horse chestnut 33.8
Syringa vulgaris Lilac 29.8
Lonicera etrusca Honeysuckle 23.8
Platanus sp. Plane tree 14.6
Robinia pseudoacacia Locust 10.4
Cercis siliquastrum Redbud 6.7
Pyracantha sp. Scarlet firethorn 4
Cupressus sempervirens Cypress 2.9
Ligustrum sp. Prie 1,2
Cedrus pinaceae Cedar tree 0.4
Senecio sp. Senecio 0.1
Xanthium spinosum Cocklebur 0
111
(pp. 108-118)
The prepared test was organized in line with the opinions of expert faculty members, and a pilot study was
conducted with 32 ninth-grade students to confirm the face validity of the test. In this pilot study, the students
were asked about the intelligibility of the pictures, and all of them indicated that the pictures were clear and
understandable.
Data Collection
The data collection tools were administered to students during their lessons, and it took approximately 25
minutes for each of the students to complete them.
Data Analysis
The extent to which the data collected about (a) gender, (b) presence of a home garden, (c) green field around
the home, (d) experience with indoor or outdoor cultivation, (e) spending time in a green field, (f ) picking flowers,
fruits, vegetables, (g) school grades and (h) the plant recognition level predicted the attitude towards plants was
analysed by multiple regression analysis. In this research, Pearson correlations between all variables were calculated
to understand the structure of the correlation between variables.
Multiple regression analysis was performed to determine the degree to which each variable contributed to
the attitude towards plants. For multiple regression analysis, normality was examined by measures of central ten-
dency, Skewness-Kurtosis values and graphical interpretation, and they all showed normal distribution. Variance
inflation factor (VIF) values were examined to determine whether the predictor variables had multicollinearity
among themselves. The VIF values of the predictor variables should be lower than 5 (Büyüköztürk, 2012). In this
research it was found that the VIF values for the predictor variables ranged between 1.05-1.81, and there was no
collinearity on the regression model.
In order to provide a detailed view of the strengths of the relationships between the dependent variable
(attitude towards plants) and predictors (gender, presence of a home garden, a green field around the home,
experience with indoor or outdoor cultivation, spending time in a green field, experience picking flowers, fruits,
and/or vegetables, plant recognition level, and school grades), the standard input procedure (enter method) was
used, where independent variables are entered and evaluated as a block in one step. The analyses of data were
carried out using SPSS version 22.0.
Data Management
For the multiple regression analysis, two dummy variables were created to determine the relation between
attitude towards plants and the students’ grades in school. In the first dummy variable, students with high school
grades were encoded as 1 and the others as 0. In the second dummy variable, students with low school grades
were encoded as 1 and the others as 0.
Research Results
Pearson correlations were calculated to determine whether there was a significant relationship among the
variables of the study. The results of Pearson correlations are shown below in Table 3.
Table 3
Inter-Correlations for Attitudes towards Plants and Other Variables
Variables 1 2 3 4 5 6 7 8 9 10
1. Attitudes towards plants -
2. Gender .084* -
112
(pp. 108-118)
Variables 1 2 3 4 5 6 7 8 9 10
3. Presence of home .102* .034 -

garden
4. Green area around the .084* .103** .278** -

home
5. Spending time in green .191** .087* .067 .141** -

area
6. Indoor or outdoor .230** .012 .086* .049 .001** -

cultivation
7. Picking flowers, vegeta- .228** .084* .038 .015 .212** .386** -

bles and fruits
8. Plant recognition .268** .106** .085* .125** .083* .099** .088* -
9. High school grades .130** .014 .016 .083* .009 .039 .063 .478** -
compared to average and
low grades–DV 1
-
10. Low school grades .110** .072* .016 .010 .020 .024 .062 .579** .448**
compared to high and
average grades-DV 2
* p< .05 ** p< .01
As shown in Table 3, the attitude towards plants is significantly correlated with gender, presence of home
garden, a green field around the home, spending time in a green field, indoor or outdoor cultivation, picking flow-
ers, fruits, and/or vegetables, plant recognition, high school grades compared to average and low grades (Dummy
variable 1) and low school grades compared to high and average grades (Dummy variable 2). All of the correlation
values were determined to be positive.
Results of the data analysis showed that the female students had more positive attitudes towards plants than
those of the male students. Furthermore, the students who had home gardens, had a green field around their home,
engaged in indoor or outdoor cultivation, spent more time in a green field, and had the experience of picking
flowers, fruits and/or vegetables had attitudes towards plants that were more positive. Similarly, the attitudes of
the students who could recognize the plants in their neighbourhoods were more positive, and the students with
high school grades had more positive attitudes towards plants than those with average and low grades (Dummy
variable 1), while the attitudes of those with high and average school grades were more positive than those with
low grades (Dummy variable 2).
In Table 4 below the results of the multiple regression analysis show that the linear relationship between the
predictive variables and attitude towards plants significantly explained 14% of the total variance (adjusted R2 =
.14), in the attitude towards plants (F (9.773) = 12.102, p <0.05).
Table 4
B and Beta Correlation Coefficients and Significance Levels of the Variables
B Standard Error β t p
Gender -2.303 .879 .100 2.621 .009
Presence of home garden -.984 .892 .043 1.103 .270
113
(pp. 108-118)
B Standard Error β t p
Green area around the home

-1.002 1.251 .032 .801 .423
Spending time in a green

1.664 .561 .116 2.968 .003
field
Indoor or outdoor cultivation 1.461 .478 .125 3.058 .002
Picking flowers, vegetables

1.849 .559 .137 3.307 .001
and fruits
Plant recognition .679 .171 .199 3.970 .0001
High school grades com-

pared to average and low .868 1.222 .031 .710 .478
grades –DV 1
Low school grades com-

pared to
.654 1.112 .028 .588 .557
high and average grades
-DV 2
As shown in Table 4, only gender, spending time in a green field, indoor or outdoor cultivation, picking flow-
ers, fruits, and/or vegetables, and plant recognition have a significant contribution to predicting the variance in
attitude towards plants. Standard β values on Table 4 show that the recognition of plants is a relatively important
predictor of attitude towards plants.
Although Pearson correlations revealed that having a home garden and a green field around home, high
school grades compared to average and low grades (Dummy variable 1), and low school grades compared to
high and average grades (Dummy variable 2) had significant relationships with the attitudes towards plants,
when entering the other variables into the regression model, these did not significantly contribute to explaining
the variance in attitude towards plants. This could have resulted from the correlations of other variables with the
plant recognition level.
When the plant recognition percentages of the students given in Table 2 are examined, it is seen that their plant
recognition levels are low. On average 7.6 plants recognized correctly out of 22 plants (x̄ = 7.6, SE =. 12). Over 70%
of students recognized rose, daisy, mulberry and elaeagnus. Other plants are not recognized by more than half of
the students. There is no plant recognized by all students. The cocklebur was not recognized by any of the students.
Discussion
In this research, the predictors for ninth-grade students’ attitudes towards plants and the degree of simultane-
ous relationship between attitude towards plants and other variables, including gender, the presence of a home
garden, having a green field around the home, indoor or outdoor cultivation, spending time in a green field, picking
flowers, fruits, and/or vegetables, plant recognition, and school grades, were examined.
The results of the research reveal that female students have more positive attitudes towards plants than those
of male students. Similarly, Lohr and Pearson-Mims (2005) and Özel et al. (2013) report that females’ attitudes towards
plants are more positive than those of males. However, in contrast to these, Fancovicova and Prokop (2011a) and
Kaplan and Topsakal (2013) and Strgar (2008) concluded that gender is not effective in students’ attitudes towards
plants. The result in this research that the females have more positive attitudes towards plants than males can
be attributed to the interest in the plant as mentioned in some studies (Gatt et al. 2007; Schussler & Olzak 2008).
In the study, the attitudes of the students who had a home garden and a green field around their home,
engaged in indoor or outdoor cultivation, spent more time in a green field, had the experience of picking flowers,
fruits and/or vegetables, and knew more about the plants in their close environment, are more positive. In addi-
114
(pp. 108-118)
tion, it is observed that the students with high school grades have more positive attitudes towards plants than
those with average and low grades, and that students with average and high school grades have more positive
attitudes towards plants than those with low grades. Furthermore, it was determined that only gender, spending
time in a green field, indoor or outdoor cultivation, picking flowers, fruits, and/or vegetables, and plant recognition
level have a significant contribution to predicting variance in attitude towards plants. This finding is in line with
the research results reported by Lohr and Pearson-Mims (2005), who state that individuals’ first-hand experiences
in the past affect their attitude towards plants. They specifically examine the effect of childhood experiences, in
terms of children’s active and passive relationships with nature, on attitudes towards trees in adulthood, and report
that while both passive and active interactions with plants in childhood have an impact on the individuals’ positive
attitudes towards plants, being in an active relationship with nature through gardening or picking vegetables has
a more significant effect on individuals’ positive attitude towards plants during adulthood. In addition to this, the
same study states that being raised in an urban environment, one with such features as skyscrapers, has a negative
effect on attitudes. Similarly, Fancovicova and Prokop (2011a) reported that students with a home garden have
more positive attitudes towards plants.
In the present study, it has been determined that the predictive variables explained only 14% of the total vari-
ance in attitude towards plants, with most of the variance (86%) remaining unclear, meaning that different variables
must be explaining the remaining variance. As attitude has cognitive, affective, and behavioural dimensions the
strength of these components differs according to the attitudes informing them. Students’ attitudes towards plants
could be related to their environmental attitudes, considering that a person’s attitudes and perceptions about the
environment are basically shaped by their attitudes, experiences and perceptions about living organisms (Tun-
nicliffe & Reis, 2000). On the other hand, since attitudes have not only affective but also cognitive components, it
has been suggested that it would be easier for children to develop positive attitudes towards plants by teaching
them meaningful ideas about the subject (Hadzigeorgiou et al., 2011).
The recognition of plants, one of the variables discussed in this study, has been found to be a more impor-
tant predictor of attitude towards plants than that of other variables. This could be attributed to the fact that
when children are taught about nature and how to be responsible individuals and to protect the environment,
they must also, whether indirectly or directly, be informed about the common organisms in their environment
(Bebbington, 2005). However, results from the students’ plant recognition tests showed that the students’ levels
of recognizing plants were low in general. This finding is in line with those reported by research (Bebbington,
2005; Civelek, 2012; Haymana Ulucanlı, 2009; Fančovičová & Prokop, 2011a; Kaasinen, 2019; Lückmann & Menzel,
2014; Mercan & Köseoğlu, 2019; Yüce & Doğru, 2018; Wyner & Doherty 2019) reveals that the level of participants’
plant recognition is low. Considering the characteristics of the plants that are highly recognized by the students
in the present research, it is seen that these plants have either flowers used as ornamental plants or edible fruits.
Similarly, Fančovičová & Prokop (2011b) found that children recognize better about plants with edible fruit. More
recognition of these plants may be due to the students’ experiences and memories for them (Kaasinen, 2019) or the
characteristics of the plants such as colourful flowers or outstanding fruits (Tunnicliffe, 2001). It has been reported
that the characteristics of plants such as smell, colour, pattern, scale and floristic features are effective in attracting
the attention of individuals (Sanders, 2007). Some plants are less recognized by students. Inability of students to
recognize these plants may be due to the fact that as stated by Kaasinen (2019) they do not have distinctive features
or have no special meaning for students. Lückmann and Menzel (2014) also noted, poor knowledge about plants
may be derived from the lack of interest in plants among young people, since in this period, social interaction is
more important. In addition, the reduction of urban green spaces limits the opportunities to experience nature
and to learn more about the environment, for both adults and children (Ju & Kim 2011). The decreased contact
that children and young people have with nature also affect their knowledge about nature (White, 2004). For this
reason, it has been argued that children today have less information about the natural environment than that of
their own parents (O’Brien, 2010). As mentioned above, both the lack of interest in plants and the limitation of their
experiences with nature can be said to be effective in individuals’ attitudes towards plants. Studies have shown
that teachers can increase the students’ interests and improve their attitudes towards plants by showing the plants
from new perspectives and using appropriate methods and field knowledge (Çil, 2015, 2016; İri & Çil, 2020; Strgar,
2007). It has been further shown that students’ interest in living organisms will increase and they will care about
them more as they get to know the organisms (Lindemann-Matthies, 2005).
115
(pp. 108-118)
Limitations
The present research revealed that the predictive variables explained a small part of the total variance in
attitude towards plants. This low explanatory power is a limitation of this research. This suggests that there are
other unexplored variables that contribute to the attitude towards plants. It is known that many factors play a
role in the formation and development of attitudes, such as personality, family, education, socialization, economic
status and social class, mental and informational factors, physiological factors, peers, group membership, experi-
ence, living environment and place, cultural effects, profession, personal motivation, special interests, and mass
media. Further research can examine other variables, in addition to those discussed in this study, that may affect
attitudes towards plants.
In this research it was found that the active interactions individuals have with plants, such as spending time
in a green field, indoor or outdoor cultivation, picking flowers, fruits, and vegetables, and plant recognition level
significantly contributed to predicting the variance in attitude towards plants. It could be argued that students,
from primary school to university, have not had the opportunity to gain awareness and develop positive attitudes
towards plants because lessons are not designed to familiarize students with the plants around them, nor do they
provide opportunities for interaction with the environment. Therefore, in order to realize the importance of the
natural environment, plants and biodiversity, individuals, starting at an early age, should be provided first-hand
experience with plants and educational activities should be developed to facilitate this experience.
Given that the results from this study showed the level of recognizing plants to be a relatively more important
predictor of attitude towards plants, teachers should be trained in developing this skill in students, as studies have
shown that teachers can generate greater engagement and interest from students by presenting plants from new
perspectives and with appropriate methods.
References
Allen, W. (2003). Plant blindness. BioScience, 53(10), 926-926. https://doi.org/10.1641/0006-3568(2003)053[0926:PB]2.0.CO;2

Aydın, O. (2002). Tutumlar. [Attitudes.] E. Özkalp (Ed.), Davranış bilimleri, [Behavioral sciences.] (s.279- 295) Açık Öğretim Fakültesi
Yayınları
Barman, C., Stein, M., Barman, N., & McNair, S. (2003). Assessing students’ ideas about plants. Science and Children, 10(1), 25-29.
https://www.jstor.org/stable/4451935
Barman, C., Stein, M., McNair, S., & Barman, N. (2006). Students’ ideas about plants and plant growth. American Biology Teacher,
68(2), 73-79. https://doi.org/10.2307/4451935
Baysal, A. C., & Tekarslan, E. (1996). Davranış bilimleri. [Behavioral sciences.] Avcıol Basım-Yayın.
Bebbington, A. (2005). The ability of a-level students to name plants. Journal of Biological Education, 39(2), 63-67.
https://doi.org/10.1080/00219266.2005.9655963
Bujang, M. A., Sa’at, N., & Sidik, T. M. I. T. A. B. (2017). Determination of minimum sample size requirement for multiple linear
regression and analysis of covariance based on experimental and non-experimental studies. Epidemiology, Biostatistics
and Public Health, 14(3). https://www.researchgate.net/publication/320165481
Büyüköztürk, Ş. (2012). Sosyal bilimler için veri analiz el kitabı. [ Data analysis handbook for social sciences.] (17. Baskı). Pegem
Akademi.
Civelek, S. (2012). Ortaöğretim 9. sınıf öğrencilerinin yakın çevrelerindeki bitkileri tanıma düzeyleri: [Trabzon ili örneği. Secondary school
9th grade students’ level of the plant identification in their environment:A case from Trabzon.] Yüksek Lisans Tezi, Karadeniz
Teknik Üniversitesi, Eğitim Bilimleri Enstitüsü, Trabzon.
Çil, E. (2015). Integrating botany with chemistry & art to improve elementary school children’s Awareness of Plants. The American
Biology Teacher, 77(5), 348-355. https://doi.org/10.1525/abt.2015.77.5.5
Çil, E. (2016). Instructional integration of disciplines for promoting children’s positive attitudes towards plants. Journal of Biological
Education, 50(4), 366-383. https://doi.org/10.1080/00219266.2015.1117512
Fančovičova, J., & Prokop (2010). Development and initial psychometric Assessment of the plant attitude questionnaire. Journal
of Science Education Technology, 19, 415-421. https://doi.org/10.1007/s10956-010-9207-x
Fančovičova, J., & Prokop, P. (2011a). Plants have a chance: Outdoor ̌ educational programmes alter students’ knowledge and
attitudes towards plants. Environmental Education Research, 17, 537–551. https://doi.org/10.1080/13504622.2010.545874
Fančovičová, J., & Prokop, P. (2011b). Children’s ability to recognise toxic and non-toxic fruits. Eurasia Journal of Mathematics, 7(2),
115–120. https://doi.org/10.12973/ejmste/75186
Gatt, S., Tunnicliffe, S. D., Borg, K., & Lautier, K. (2007). Young Maltese children’s ideas about plants. Journal of Biological
116
(pp. 108-118)
Hadzigeorgiou, Y., Prevezanou, B., Kabouropoulou, M., & Konsolas, M. (2011). Teaching about the importance of trees: A study
with young children. Environmental Education Research, 17(4), 519-536. https://doi.org/10.1080/13504622.2010.549938
Haymana Ulucanlı, F. (2009). İlköğretim öğrencilerinin yakın çevrelerindeki bitkileri tanıma düzeyleri: Bolu ili örneği. [Recognition
levels of primary school students the plants in their immediate surroundings:Bolu example.] Yüksek Lisans Tezi, Abant İzzet
Baysal Üniversitesi, Sosyal Bilimler Enstitüsü, Bolu.
İnceoğlu, M. (2010). Tutum algı iletişim. [Attitude perception communication.] (5. Baskı). Beykent Üniversitesi Yayınevi.
İri, F. G., & Çil, E. (2020). Attitudes toward plants: Comparing the impact of instruction through writing & through a botanical
garden trip. The American Biology Teacher, 82(4), 218–226. https://doi.org/10.1525/abt.2020.82.4.218
Ju, E. J., & Kim, J. G. (2011). Using soil seed banks for ecological education in primary school. Journal of Biological Education, 45(2),
93-101. https://doi.org/10.1080/00219266.2010.546010
Kaasinen, A. (2019). Plant species recognition skills in Finnish students and teachers. Education Sciences, 9(2), 85-97.
https://doi.org/10.3390/educsci9020085
Kaplan, K., & Topsakal, U. (2013). Primary school students’ attitudes toward plants. Procedia - Social and Behavioral Sciences 89,
598 – 606. https://doi.org/10.1016/j.sbspro.2013.08.901
Köse, S., Ayas, A., & Uşak, M. (2006). The effect of conceptual change texts instructions on overcoming prospective science
teachers’ misconceptions of photosynthesis and respiration in plants. International Journal of Environmental and Science
Education, 1(1), 78–103. http://www.ijese.net/makale_indir/IJESE_1572_article_583986272d610.pdf
Köse, S., Uşak, M., & Bahar, M. (2009). A cross-age study of students’ understanding and their misconceptions about plant
nutrition. Didactica Slovenica-Pedagoska Obzorja, 24(1), 109-122. https://www.researchgate.net/publication/268924740
Krosnick, S. E., Baker, J. C., & Moore, K. R. (2018). The pet plant project: Treating plant blindness by making plants personal. The
American Biology Teacher, 80(5), 339-345. https://doi.org/10.1525/ abt.2018.80.5.339
Lindemann-Matthies, P. (2005). ‘Loveable’ mammals and ‘lifeless’ plants: how children’s interest in common local organisms
can be enhanced through observation of nature. International Journal of Science Education, 27 (6), 655-677.
https://doi.org/10.1080/09500690500038116
Lindemann-Matthies, P. (2006). Investigating nature on the way to school: Responses to an educational programme by teachers
and their pupils. International Journal of Science Education, 28(8), 895–918. https://doi.org/10.1080/10670560500438396
Lohr, V. I., & Pearson-Mims, C. H. (2005). Children’s active and passive interactions with plants influence their attitudes and actions
toward trees and gardening as adults. HortTechnology, 15(3), 472-476. https://doi.org/10.21273/HORTTECH.15.3.0472
Lückmann, K., & Menzel, S. (2014). Herbs versus trees: Influences on teenagers’ knowledge of plant species. Journal of Biological
McCormick, S., & Tjian, R. (2010) A new focus on plant sciences. Science, 330, 1021. doi: 10.1126/science.1198153.
Mercan, G., & Köseoğlu, P. (2019). Biyoloji öğretmen adaylarının yakın çevrelerindeki ağaçları tanıma düzeyleri: Ankara ili örneği.
[Biology teacher candidates’ identification levels of trees in their immediate surroundings: Ankara example.] Yüzüncü Yıl
Üniversitesi Eğitim Fakültesi Dergisi, 16(1), 538-560. http://dx.doi.org/10.23891/efdyyu.2019.135
O’Brien, C. M. (2010). Do they really “Know Nothing”? An inquiry into ethnobotanical knowledge of students in Arizona,
USA. Ethnobotany Research and Applications, 8, 35-47. http://ethnobotanyjournal.org/index.php/era/article/view/405/259
Özarslan, M. (2019). Üstün zekâlı ve yetenekli öğrenciler ile normal gelişim gösteren öğrencilerin bitkilere yönelik tutumlarının
karşılaştırılması. [Comparison of the plant attitudes of gifted and talented students and those of normal development.]
Elementary Education Online, 18(4). doi:10.17051/ilkonline.2019.631812
Özel, M., Sürücü, A. & Bilen, K. (2013). İlkögretim öğrencilerinin bitkilere yönelik tutumları. [Primary school students’ attitudes
toward plants.] Pamukkale Üniversitesi Eğitim Fakültesi Dergisi, 34, 119–132. https://doi.org/10.9779/PUJE588
Pany, P. (2014). Students’ interest in useful plants: a potential key to counteract plant blindness. Plant Science Bulletin, 60, 18–27.
https://www.researchgate.net/publication/301287584_
Patrick, P., & Tunnicliffe, S. D. (2011). What plants and animals do early childhood and primary students’ name? Where do they
see them? Journal of Science Education and Technology, 20(5), 630-642. https://doi.org/10.1007/s10956-011-9290-7
Prokop, P., & Fančovičová, J. (2014). Seeing coloured fruits: Utilization of the theory of adaptive memory in teaching botany.
Journal of Biological Education 48(3), 127–132. doi:10.1080/00219266.2013.837407
Prokop, P., & Fančovičová, F. (2018). The perception of toxic and nontoxic plants by children and adolescents
with regard to gender: implications for teaching botany, Journal of Biological Education, 53(4), 463-473.
https://doi.org/10.1080/00219266.2018.1501405
Prokop, P., Majerčíková, D., & Vyoralová, Z. (2016). The use of realia versus powerpoint presentations on botany lessons. Journal
of Baltic Science Education, 15(1), 18–27. http://journals.indexcopernicus.com/abstract.php?icid=1196713
Sousa, V. D., Driessnack, M., & Mendes, I. A. C. (2007). An overview of research designs relevant to nursing: Part 1: Quantitative research
designs. Revista Latino-Americana de Enfermagem, 15(3), 502-507. https://doi.org/10.1590/S0104-11692007000300022
Selvi, M. (2012). Adaptation into Turkish of the plant attitude questionnaire. Journal of Baltic Science Education, 11(2), 175-183.
http://www.scientiasocialis.lt/jbse/?q=node/253
Schussler, E.E., & Olzak L (2008). It’s not easy being green: Student recall of plant and animal images. Journal of Biological Education,
42, 112–119. https://doi.org/10.1080/00219266.2008.9656123
Strgar, J. (2007). Increasing the interest of students in plants. Journal of Biological Education, 42(1), 19-23.
https://doi.org/10.1080/00219266.2007.9656102
Strgar, J. (2008). How are age and gender related to attitude toward plants and animals? Acta Biologica Slovenica. 51(1), 33-38.
http://bijh-s.zrc-sazu.si/ABS/SI/ABS/Cont/51_1/ABS_51_1_Strgar_2008_33-38.pdf
117
(pp. 108-118)
Taylor S. E., Peplau, L. A., & Sears, D. O. (2007). Sosyal psikoloji, [Social psychology] (Çev: Ali Dönmez), İmge Kitabevi.
Topsakal, U. U., & Oversby, J. (2012). Turkish student teachers’ ideas about diagrams of a flower and a plant cell. Journal of Biological
Tunnicliffe, S. D. (2001). Talking about plants - comments of primary school groups looking at plant exhibits in a botanical garden.
Journal of Biological Education, 36, 27-34. https://doi.org/10.1080/00219266.2001.9655792
Tunnicliffe, S. D., & Reiss, M. J. (2000). Building a model of the environment: How do children see plants? Journal of Biological
Türkmen, L., Dikmenli, M., & Çardak, O. (2003). İlköğretim öğrencilerinin bitkiler hakkındaki alternatif kavramları. [Primary
school students’ alternative conceptions about plants.] Afyon Kocatepe Üniversitesi Sosyal Bilimler Dergisi, 5(2), 53-70.
https://acikerisim.aku.edu.tr/xmlui/bitstream/handle/11630/3239/53-70.pdf?sequence=1&isAllowed=y
Wandersee, J. (1986). Plants or animals – Which do junior high school students prefer to study? Journal of Research in Science
Teaching, 23(5), 415-426. https://doi.org/10.1002/tea.3660230504
Wandersee, J. H., Clary, R. M., & Guzman, S. M. (2006). A writing template, for probing students’ botanical sense of place. The
American Biology Teacher, 68(7), 419-422. https://doi.org/10.2307/4452030
Wandersee, J., & Schussler, E. E. (1999). Preventing plant blindness. The American Biology Teacher. 61(2), 82-
87. https://doi.org/10.2307/4450624
Wandersee, J., & Schussler, E. E. (2001). Toward a theory of plant blindness. Plant Science Bulletin 47(1), 2-12.
https://botany.org/PlantScienceBulletin/psb-2001-47-1.php#Toward%20a%20Theory%20of%20Plant
White, R. (2004). Interaction with nature during the middle years: It’s importance in children’s development and nature’s future.
http://www.whitehutchinson.com/children/articles/nature.shtml.
Wyner, Y., & Doherty, J. H. (2019). Seeing the trees: what urban middle school students notice about the street trees that surround
them. Journal of Biological Education, 1-23. https://doi.org/10.1080/00219266.2019.1667407
Yüce, M., & Doğru M. (2018). Fen bilimleri öğretmen adaylarının yakın çevresindeki bitkileri tanıma düzeyleri. [ Level of
plant recognition in immediate vicinity of science teacher candidates.] Ihlara Eğitim Araştırmaları Dergisi, 3(1), 15-35.
https://dergipark.org.tr/tr/pub/ihead/issue/36890/349808
Received: September 17, 2020 Accepted: January 12, 2021
Cite as: Selvi, M., & Çelepçıkay İslam, E. (2021). The predictors of ninth grade students’ attitudes towards plants. Journal of
Baltic Science Education, 20(1), 108-118. https://doi.org/10.33225/jbse/21.20.108
Meryem Selvi PhD, Associate Professor, Department of Biology Education, Gazi

(Corresponding author) University, Faculty of Education, 06500, Ankara, Turkey.
Website: https://avesis.gazi.edu.tr/meryema
Emel Çelepçıkay İslam MSc, Teacher, Sakarya Yönder Schools, Sakarya, Turkey.
118
GRADE 6 & 9 STUDENT AND
TEACHER PERCEPTIONS OF
TEACHING AND LEARNING ISSN 1648-3898 /Print/
APPROACHES IN RELATION ISSN 2538-7138 /Online/
TO STUDENT PERCEIVED
INTEREST/ENJOYMENT
TOWARDS SCIENCE LEARNING
Abstract. Student perceptions of science
teaching could be expected to differ in
grade 6 compared with teaching in grade
9, as could student interest in science
Moonika Teppo, learning. To compare Estonian grades 6
and 9, lower secondary school student and
Regina Soobard,
science teacher perceptions of teaching/
Miia Rannikmäe learning approaches as well as student
perceived interests/enjoyment on science
learning, an instrument was developed
taken into account 18 different teaching
approaches that could take place in science
lessons. An analytical, exploratory struc-
tural equation modelling (ESEM) approach
supported 4-factor models differentiating
Introduction
between teacher-centred and constructivist
approaches for both teachers and students,
In recent years, research in science education has focused on how to each having acceptable model fits. Based
teach science effectively in a way that initiates and increases student inter- on outcomes, a regression model was
est and motivation towards science learning. This focus has been stressed, developed associating student interest to
because there is a tendency for declining interest, motivation and attitudes learn science with the frequency of teach-
among lower secondary school students (Osborne et al., 2003; Potvin & ing and learning approaches explored.
Hasni, 2014). Research has shown that the teaching approach can also play Results indicated that approaches associ-
an important role. In teacher-centred approaches, both interest and student ated with teacher-centred approaches
learning are shown to be limited, whereas for student-centred approaches, were those most frequently undertaken in
science classes when compared with those
where students actively participate in the learning process (e.g., inquiry-based
seen as social constructivist and student-
learning, hands-on activities) these have been shown to positively promote
centred, both among science teachers and
student interest (e.g., Potvin et al., 2017; Wang et al., 2015). Furthermore, students. Regression analysis showed that
research studies have shown that student interest towards science learn- there was a relationship between student
ing is influenced by the manner in which science learning is initiated and interest/enjoyment towards science sub-
presented to students (Hasni & Potvin, 2015; Häussler & Hoffmann, 2000; jects and the frequency, in which different
Teppo et al., 2017). teaching and learning approaches took
The Teaching and Learning International Survey (TALIS) (OECD, 2019) place in the classroom.
indicated that Estonian teachers were only using cognitive activation ap- Keywords: exploratory structural equa-
proaches (e.g., students working in groups, undertaking critical thinking, or tion modelling, regression analysis, social
solving complex tasks) to a relatively low extent, although this was depen- constructivism, student interest/enjoyment,
teaching-learning approaches
dent on the school subject involved. For example, science and mathematics
teachers used such approaches less than other teachers. Not surprising,
therefore, Estonian PISA 2015 (OECD, 2016) results indicated that, in science
lessons, students rarely undertook practical activities and had few opportuni- Moonika Teppo, Regina Soobard,
Miia Rannikmäe
ties to plan experiments, compared with the OECD average. These findings University of Tartu, Estonia
pointed to a predominance of teacher-centred approaches in the teaching
of science subjects.
119
SCIENCE LEARNING ISSN 2538–7138 /Online/
(pp. 119-133)
The Estonian curriculum has tried to promote competence-based teaching (National Curriculum for Basic
School, 2014) and has long advocated a student-centred approach. Nevertheless, teachers have pointed to ob-
stacles, such as a cognitive, factual dominance in external assessment methods and a lack of teaching time and
resources. Not surprising, such perceived obstacles can be expected to hinder the promotion of student-centred
approaches as an effective teaching and learning process in science classes.
Learning Approaches in Science Teaching
Student learning of science in schools and the role of the teaching and learning environment have been
research topics for decades. Generally, studies on the approach to teaching and learning have been character-
ised through researching student activities and practices conducted in science lessons. According to Anthony’s
hierarchical model (Richards & Rodgers, 2001) an approach is a wide conception, referring to the philosophy
(theory) related to how teaching and learning occurs, while method of teaching is taken to be procedural, in-
dicating a systematic way of teaching that forms an implementable technique so as to help learners learn and
achieve their goals. For example, researchers have undertaken investigations of approaches in science teaching
and learning using a variety of terminology, e.g., teaching methods (Hasni & Potvin, 2015; Kousa et al., 2018),
teaching and learning activities (Hampden-Thompson & Bennett, 2013), teaching and learning practices (Eb-
enezer & Zoller, 1993). More specifically, Juuti et al. (2010) have used the term teaching method as a synonym
for an instructional method/model/strategy, student activity, or classroom practices that is designed to help
students achieve learning goals.
A Social Constructivism Approach to Teaching
Teaching and learning can be generally organised into two main categories: a teacher-centred approach vs.
a student-centred approach, which are often used as synonyms for traditional vs. constructivist forms of teaching.
A teacher-centred approach is recognised as teaching in which the teacher is dominant and plays the leading role.
The teacher behaves as an instructor, frequently using science textbooks for communicating new information (e.g.,
giving a lecture), while students passively receive knowledge or work individually using workbooks on independent
tasks, with student learning measured through tests with assessment viewed as separate from teaching (Arends,
2012; Concept to Classrooms, 2020; McLeod, 2019). Student centred learning is much more diverse, expecting the
student to play an active role and aiming at developing learning autonomy and independence (Jones, 2007) while
striving towards developing skills and practices that enable lifelong learning (Young & Paterson, 2007).
The central idea of constructivist teaching is that people actively construct or make their own knowledge and
build this on previous learning experiences (Arends, 1998; Elliott, et al., 2000). Social constructivism, as a sub-set
of constructivism, emphasises that learning is dependent on interactions with others, i.e., learning is socially and
culturally constructed in an active way, focusing on the learner as part of a social group, and learning is determined
by the complex interplay between a learner’s existing knowledge, the social context and the problem, or situation
to be solved (Taber, 2011). This suggests learning, within the classroom, is actively promoted in a social constructiv-
ist way through collaboration among students and between students and teachers.
The social constructivist learning theory (Palincsar, 1998) underpins a variety of student-centred teaching
approaches and techniques, which contrast with a traditional view of education, whereby knowledge is simply
transmitted passively by teachers to students. In the social constructivist approach, the teacher role in the learn-
ing process is to be a collaborator, advisor and educator, creating and facilitating a collaborative problem-solving
environment, in which students become active participants in their own learning (Good & Lavigne, 2018; Lalley &
Miller, 2007). From this perspective, the teacher organises the educational process to support students and makes
use of a variety of teaching methods, guiding student to meet learning objectives (Sharples et al., 2016).
In science classes, teachers do not always apply only a teacher-centred approach, or a student-centred approach
and hence the teaching can include elements from both approaches. Nevertheless, the teacher has an important
role and responsibility, both to increase student interest in learning and also to develop student competencies as
stipulated in the curriculum and to do so through selecting the most relevant learning approaches in any given
situation (Ebenezer & Zoller, 1993).
120
ISSN 1648–3898 /Print/ GRADE 6 & 9 STUDENT AND TEACHER PERCEPTIONS OF TEACHING AND LEARNING
ISSN 2538–7138 /Online/ SCIENCE LEARNING
(pp. 119-133)
Teaching and Learning Approaches and Student Interest/Enjoyment
Numerous studies have researched different teaching and learning approaches on student interest and
motivation towards science learning (e.g., Ebenezer & Zoller, 1993; Hampden-Thompson & Bennett, 2013; Hasni &
Potvin, 2015; Juuti et al., 2001; Kousa et al., 2018; Sturm & Bogner, 2008). For instance, both stimuli, such as the use
of puzzles, challenging tasks, or well-organised texts, or simply providing students with choices in the way and
what they learn, have been used to increase situational interest, focusing attention and, as such, increasing levels
of learning (Hidi, 1990; Hidi & Renninger, 2006). Additionally, Bergin (1999) has suggested that hands-on activities,
novel learning stimuli, social interaction via group work, modelling of experts, etc. have been shown to increase
student interest/enjoyment. Furthermore, research by Hampden-Thompson and Bennett (2013) has indicated
that 15-year-old student motivation and enjoyment of science has increased the more activities are related with
interactions (debates, discussions, explanations), hands-on activities (doing experiments, drawing conclusions
based on experiment) and applications in science (relevance of science concepts to student lives).
Juuti et al. (2001) have used a two-fold approach to explore how grade 9 students have evaluated teaching
methods used in science teaching - in the current practice and preferred (desired) situations. They have also ex-
plored student perceptions of school science (e.g., interest, difficulty, importance of science and technology for
society, becoming a scientist, etc.), in addition to relationships between student demographic backgrounds (girl-
boy, number of books at home and academic performance), perceptions of school science and teaching methods.
Their findings have shown that interested students perceived school science to be relevant in their everyday life
and wished to undertake more creative activities, such as brainstorming and project work. In addition, their results
have identified gender differences related with teaching methods, in that boys seemed to be more satisfied with
the current teaching, while girls preferred group discussions, debates and group projects.
Hasni and Potvin (2015) have shown, by exploring grade 5-11 student interest towards science and technology
aspects, including a general interest towards science and technology education (e.g., school science is fun, boring),
plus preferences for relevant teaching methods that students have tended to learn science through spending more
time doing exercises, such as using handout or workbooks, etc. As a result, they have noted a moderate to weak
positive correlation between general interest towards science and technology and different teaching methods. In
addition, their results have indicated that students express a desire to participate when the teaching methods are
seen as enabling students to be active i.e., make observations, experimenting, participating in debates, etc. and
showed lower preference for putting forward explanations, using textbooks, etc.
Research Aim and Research Questions
Taking into consideration the declining student interest towards science learning and the teacher-centred
teaching environment, the aim of the current research was to explore the perceptions of Estonian science teach-
ers and students towards teaching and learning approaches being implemented in science classes, associating
this with student interest (in terms of interest/enjoyment) towards science and the teaching/learning approaches
being used. Specifically, this research focused to answer the following research questions:
1. What perceptions do students and science teachers hold regarding teaching and learning approaches
used in science classes and how do such perceptions differ from teaching in grade 6 and grade 9?
2. What is the effect of teaching and learning approaches in grades 6 and 9 on student interest/enjoyment
towards learning in science subjects?
Current Research
The initial focus of this research was the development of a general model interrelating teaching approaches.
The model was constructed, based on a cross-model of teaching initiated by Bundsgaard (2009), but adapted
based on the Estonian science learning environment (National Curriculum for Basic School, 2014). The horizontal
axis distinguished between a science content approach and social constructivist teaching, while the vertical axis
identified teacher-centred vs. student-centred teaching.
121
(pp. 119-133)
Figure 1
Teacher versus Student centred Learning Model Distinguishing between Science Content and Social Constructivism Approaches
(Adapted from Bundsgaard, 2009)
The research focused on gathering data from both students and teachers from two perspectives – perceptions
about using teaching and learning approaches and student perceived interest/enjoyment towards science learn-
ing (Table 1), taking into account the divisions within the theoretical model of teaching and learning approaches
presented in Figure 1.
Sample
The student and teacher samples were derived as part of a large-scale research project focusing on student
and teacher perceptions of teaching and learning approaches, carried out in science lessons associated with student
interest towards science learning. The student sample for the study was formed from a representative sample of
the Estonian 6 and 9 grade population. The school location and number of students were taken into account for
sampling (for detailed description of the sampling process see Pedaste et al., 2017). The sample comprised of a
total of 3521 students from 6th grade (N = 2673, average age 12.6) and 9th grade (N = 848, average age 15.6). In ad-
dition, for comparative purposes, data were also collected from science teachers (N = 205) from the same schools,
related to the teaching and learning approaches used.
Information about the research project (e.g., purpose, use of data, expected outcomes) were provided for
students, their parents, science teachers and schools representatives. At the beginning of the research, written per-
mission from parents were asked related to their students’ participation. Only students who had received parental
approval were included in the research. Science teachers voluntarily participated in the research.
Instrument Development
To undertake this research a questionnaire was developed enabling data gathering with respect to (a) percep-
tions about using teaching and learning approaches and (b) perceived student interest/enjoyment towards science
learning (Table 1). Taking into account the theoretical model of teaching and learning approaches presented in
Figure 1, the measurement tool included 18 teaching and learning approaches chosen and modified based on
earlier research (Ebenezer & Zoller, 1993; Holbrook & Rannikmae, 2014; Juuti et. al, 2001; Pedaste et al., 2015). The
122
(pp. 119-133)
interest/enjoyment section comprised 3 items appropriately worded to be used in a science class, or separately in
biology, geography, physics and chemistry classes.
Table 1
Overview of the Instrument Used for Measuring Perceptions of Grade 6 and 9 Students and Science Teachers on Teaching and
Learning Approaches and Students’ Perceived Interest/Enjoyment towards Science Learning
Measures Grade 6 students Grade 9 students Science teachers
I. Teaching and learning approaches

science ˅ - ˅
biology - ˅ ˅
geography - ˅ ˅
physics - ˅ ˅
chemistry - ˅ ˅
II. Interest/enjoyment in
science ˅ -
biology - ˅ Not measured
geography - ˅
physics - ˅
chemistry - ˅
Note. ˅ - measured in corresponding grade.
Both teachers and students in grade 6 and grade 9 were asked to evaluate the frequency in which certain
teaching and learning approaches occurred in science lessons using a 3-point scale (1 - never; 2 - sometimes, 3 -
often) e.g., “New content is presented by the teacher as a lecture”, “Individual work with textbook or workbook”,
“Student plans and carries out the experiment”. In addition, teachers were asked to evaluate all given approaches
based on how they liked using them in their teaching, using a 3-point scale (1 - less; 2 - same, 3 - more).
The items determining student interest/enjoyment were taken from one of the sub-scales in the Intrinsic Moti-
vation Inventory Instrument (Deci & Ryan, 2016) which was adapted to consist of the three following items: “I enjoy
studying science”, “Science learning is very interesting” and “Science learning is fun”. These items were included in
the science questionnaire for grade 6 and in each of the biology, geography, physics, chemistry questionnaires for
grade 9. All items in the interest/enjoyment section were presented in a 5-point Likert scale (1-disagree, 2-rather
disagree, 3-neutral, 4-rather agree, 5-agree).
Reliability and Validity
The internal consistency of the instrument measures was determined based on Cronbach alpha. Acceptable
values were found - .81 for teachers and .84 for students within the teaching and learning approach component
and .95 for the interest/enjoyment sub-scale among students.
The instrument was piloted with grade 6 and 9 students before the main study was carried out to increase
the validity of the instrument. Based on student feedback, minor changes in wording were made.
Content validity was used to evaluate whether the instrument represented the teaching and learning ap-
proaches and was considered familiar to students in both grades and practiced by science teachers. To achieve
content validity, three experts evaluated the instrument’s accuracy (usability, relevance) within the Estonian science
learning environment. Separate interviews with a science education professor, researcher and science teacher were
undertaken to validate teaching and learning approaches suitability for lower secondary school students. Finally,
the validation was based on expert consensus (agreement rate was 85%).
Construct validity of the instrument and its sections was measured using factorial analysis, specifically using
an exploratory structural equational modelling approach (ESEM) among both students and teachers. The corre-
sponding statistics were presented in results section.
123
(pp. 119-133)
Data Collection
In Estonian lower secondary schools, science is taught as an integrated subject from grades 1-7, while biology
and geography are separate additions in grade 7, followed by the addition of physics and chemistry and removal
of integrated science from grade 8. Thus, while grade 6 students were asked to indicate their interest/enjoyment
and teaching and learning approaches in science, grade 9 students provided data for just one of the different
science subjects (each science subject was evaluated by a randomly selection of approximately 25% students for
each subject in the whole sample).
The data were gathered as part of the project “Smart technologies and digital literacy in promoting a change
of learning” (2015-2020) using two sections of a larger instrument. Students completed the questionnaire elec-
tronically, either in a school computer class, or with tablets. Similarly, teachers answered to an electronic form,
which was sent by e-mail to all science teachers who taught at the same schools where the student survey was
conducted, asking teachers to participate in the study.
Data Analysis

Descriptive statistics (frequencies, mean, mode, standard deviation) were used to describe perceived frequen-
cies of teaching and learning approaches and interest towards science learning. The Independent Samples t Test
was used to compare the means of grade 6 and 9 students in order to determine whether there was statistical
evidence of difference between groups. Descriptive statistics and comparative analyses were conducted using
SPSS Statistics 26.
To reduce the dataset and assess the factor structure of the teaching and learning approaches, an ESEM
(exploratory structural equation modelling) approach was used. ESEM integrated confirmatory and exploratory
factor analyses (CFA and EFA) as a preferable method to test for model fit, being robust for non-normality of data.
Based on Asparouhov and Muthén (2009), the selection of the optimal number of factors was made using model
fit information. Well-established indices and criteria were used to assess the goodness of fit of the measurement
models, based on CFI and TLI ranging from 0 to 1, with higher values indicating a better fit, and with values above
.90 usually seen as associated with a good model fit (Hair et al., 2010). However, a value of .95 was taken as the
expected standard for both CFI and TLI (Hu & Bentler, 1999).
For RMSEA, it was accepted that lower values indicated a better fit, whereas acceptable values were taken to
be between .03 - .08, at a 95% confidence level, depending on the sample size (Hair et al., 2010). All models were
tested using Mplus 8.4 software (Muthén & Muthén, 1998–2017) with maximum likelihood robustness (MLR),
currently considered to be the most accurate estimator where there was no normal distribution of items with
categorical data. The Geomin oblique rotation was used as the default.
After determining the factor structure for the teaching and learning approaches, the predictive ability of dif-
ferent independent factors (teaching and learning approaches) on interest/enjoyment sub-scale (as dependent
variable) towards science learning was investigated separately for grade 6 and 9, using linear regression analysis.
Assumptions associated with linear relationship, multivariate normality, no or little multi-collinearity, no autocor-
relation, and homoscedasticity were checked and found to be valid for use.
Research Results

Grade 6 and 9 Student Perceptions about Teaching and Learning Approaches
As shown in table 2, the perceptions by grade 6 and 9 students about the use of different teaching and learning
approaches had similar tendencies. Both grades of students perceived lecturing, asking questions and individual
work with textbook, or workbook as the most often used approaches in science lessons (51-79% of the student
opinions depending on the approach). On the other hand, role play (plus brainstorming in grade 9) were indicated
as the least used approaches (59% grade 6 students and 71% of grade 9 students indicated a never response). These
results were supported by mode values, correspondingly ‘3’ for the value that appeared most often and ‘1’ for the
least often. Variability (SD) in student answers ranged from 0.49 to 0.70 indicating quite moderate divergence in
responses. The teaching “new content is presented by the teacher as a lecture” had the least variability in student
124
(pp. 119-133)
answers. It was also the most often used approach in science lessons as perceived by both grade 6 and 9 students.
At the same time, role-play, as the least used approach, had greater variability in student responses.
Table 2
Items for which Students’ Expressed their Perceptions about Usage of Different Teaching and Learning Approaches Conducted
in Their Science Classes
Grade 6 students (N=2673) Grade 9 students (N=848)
Distribution of Distribution of
Items responses (%) responses (%)
M Mo SD M Mo SD
Some- Some-
Never Often Never Often
times times
1. New content presented by the teacher

2 19 79 2.78 3 0.45 3 21 76 2.74 3 0.49
as a lecture
2. Class discussion 4 38 58 2.55 3 0.57 10 40 50 2.40 3 0.66
3. Students ask questions 3 33 64 2.60 3 0.55 8 42 50 2.44 3 0.62
4. Teacher asks questions 2 28 70 2.68 3 0.51 4 32 64 2.60 3 0.57
5. Student individual work with textbook/

3 36 61 2.59 3 0.54 5 44 51 2.46 3 0.59
workbook
6. Students work in small groups 6 62 32 2.26 2 0.56 21 58 21 2.00 2 0.65
7. Role-play 59 36 5 1.47 1 0.60 71 25 4 1.34 1 0.56
8. Have debate during the lesson 33 53 14 1.81 2 0.65 38 47 15 1.77 2 0.69
9. Brainstorming 40 46 14 1.73 2 0.68 47 40 13 1.66 1 0.70
10. Compiling the poster or presentation 22 57 21 1.99 2 0.65 37 49 14 1.77 2 0.68
11. Teacher gives feedback on student

5 47 48 2.42 2 0.59 11 46 43 2.32 2 0.66
work
12. Formulation of hypothesis or research
26 57 17 1.91 2 0.65 31 52 17 1.86 2 0.68
questions
13. Student plans and carries out the
32 56 12 1.79 2 0.63 44 46 10 1.67 2 0.66
experiment
14. Teacher undertakes a demonstration 28 58 14 1.86 2 0.63 36 49 15 1.79 2 0.68
15. Students make conclusions based on

23 55 22 1.99 2 0.67 31 49 20 1.88 2 0.71
experiments
16. Students seek information from differ-
7 55 38 2.30 2 0.60 16 57 27 2.10 2 0.65
ent sources
17. Solving everyday-life related science
33 53 14 1.81 2 0.66 20 58 22 2.02 2 0.65
problems
18. Socio-scientific decision making 33 53 14 1.82 2 0.66 23 57 20 1.98 2 0.66
Note. M – Mean; Mo – Mode; SD – standard deviation. Items are presented in the order in which they appeared in the student
questionnaire.
125
(pp. 119-133)
Grade 6 and 9 Student Perceived Interest/Enjoyment towards Science Learning
Results indicating student interest/enjoyment towards science learning were as shown in Table 3. Learning
science in grade 6 and learning biology in grade 9 were perceived as the most interesting subjects for students
(agreement rate is close to 50%), while studying chemistry and physics in grade 9 was generally uninteresting, not
enjoyable, or being indicated as less fun for a third of the students. Across all items, more than a third of students
did not express a clear opinion towards science learning, i.e., they had not revealed their polarity about their inter-
est towards science learning. Results indicated large standard deviations across items showing higher variability
in student answers (1.04 < SD < 1.21) compared with teaching and learning approaches.
Table 3
Comparison of Grade 6 and 9 Student Perceived Interest/Enjoyment towards Science Learning
Distribution of responses (%)

Items/grades M SD
Agree Neutral Disagree
Science, grade 6 (N=2673)

I enjoy studying science at school 3.45 1.12 48.4 34.8 16.8
Learning science is very interesting 3.49 1.12 50.0 33.6 16.4
Science learning is fun 3.40 1.11 44.7 37.7 17.6
Biology, grade 9 (N=220)
I enjoy studying biology at school 3.45 1.11 47.7 35.9 16.4
Learning science is very interesting 3.52 1.09 50.9 34.5 14.6
Biology learning is fun 3.47 1.09 49.5 34.1 16.4
Geography, grade 9 (N=216)
I enjoy studying geography at school 3.19 1.04 34.7 43.5 21.8
Learning geography is very interesting 3.23 1.06 38.4 39.8 21.8
Geography learning is fun 3.16 1.09 33.8 42.6 23.6
Physics, grade 9 (N=192)
I enjoy studying physics at school 3.20 1.20 43.5 28.5 28.0
Learning physics is very interesting 3.17 1.20 40.9 30.6 28.5
Physics learning is fun 3.11 1.20 37.3 31.6 31.1
Chemistry, grade 9 (N=220)
I enjoy studying chemistry at school 3.06 1.13 34.1 39.1 26.8
Learning chemistry is very interesting 3.11 1.18 35.9 37.3 26.8
Chemistry learning is fun 3.12 1.21 38.6 34.1 27.3
Note. M – Mean; SD – standard deviation. Mean values are presented on 5-point scale.
Science Teacher Perceptions about Teaching and Learning Approaches
Science teachers were asked to express their perceptions about teaching and learning approaches to include
in their teaching (less, same, more) comparing with the current situation (never, sometimes, often) in their science
classes (Table 4). The results showed that role-play, debating, brainstorming, formulation of hypothesis or research
questions, planning experiments and teacher undertaking demonstrations were the approaches teachers wished
to include more often, compared with their current situation. Undertaking the rest of the approaches were desired
with the same, or less frequency. Similar to the outcomes from students, teachers indicated that teacher-directed
approaches (teaching new content, asking questions, giving feedback to students) were most frequently used in
science classes, although teachers expressed a wish to use them significantly less than in their current practice.
126
(pp. 119-133)
Table 4
Items for which Science Teachers (N = 205) Expressed their Current and Preferred Perceptions about Usage of Teaching and
Learning Approaches
Current situation Preferred situation
Distribution of Distribution of
Items responses (%) responses (%)
M Mo SD M Mo SD
Some-
Never Often Less Same More
times
1. New content is presented by the

2 15 83 2.82 3 0.42 23 72 5 1.82 2 0.50
teacher as a lecture
2. Class discussion 2 25 73 2.72 3 0.48 1 68 32 2.31 2 0.48
3. Students ask questions 4 28 67 2.63 3 0.57 1 56 44 2.43 2 0.51
4. Teacher asks questions 1 18 82 2.81 3 0.40 6 85 9 2.02 2 0.40
5. Individual student work with textbook/

18 45 37 2.19 2 0.72 10 81 9 2.00 2 0.44
workbook
6. Students work in small groups 5 59 36 2.31 2 0.57 1 75 24 2.23 2 0.45
7. Role-play 59 37 4 1.45 1 0.57 2 52 47 2.45 2 0.53
8. Have debate during the lesson 47 46 7 1.60 1 0.62 1 51 49 2.48 2 0.51
9. Brainstorming 39 46 15 1.76 2 0.70 1 61 39 2.38 2 0.50
10. Compiling the poster or presentation 17 52 31 2.14 2 0.68 1 76 23 2.23 2 0.43
11. The teacher gives feedback on

2 12 86 2.84 3 0.41 1 79 21 2.20 2 0.41
student work
12. Formulation of hypothesis or research
28 55 17 1.89 2 0.67 2 53 45 2.42 2 0.54
questions
13. Students plans and carry out the
37 52 11 1.75 2 0.65 1 42 57 2.57 3 0.51
experiment
14. Teacher undertakes demonstrations 24 57 19 1.95 2 0.66 1 57 42 2.41 2 0.51
15. Students make conclusions based on

17 36 47 2.31 3 0.74 1 66 33 2.32 2 0.50
experiments
16. Students seek information from differ-
3 37 61 2.58 3 0.55 1 70 29 2.29 2 0.47
ent sources
17. Solving everyday-life related science
6 49 44 2.38 2 0.60 1 66 34 2.33 2 0.48
problems
18. Socio-scientific decision making 9 52 39 2.30 2 0.62 1 63 36 2.35 2 0.50
Note. M – Mean; Mo – Mode; SD – standard deviation. Items are presented in the order in which they appeared in the teacher
questionnaire.
127
(pp. 119-133)
Exploratory Structural Equation Modelling
Exploratory structural equation modelling (ESEM) was carried out separately within the student (as the factor pat-
tern of 6th and 9th grade students was the same) and teacher data. Different factor models (2-4) were tested, reaching
acceptable fit indices for solutions for four factors among students and teachers. The four-factor solution was the easiest
to interpret related to item content and factor loadings higher than .30. However, since, two items in the student and
teacher analysis (student individual work and teacher giving feedback to student work), plus, in addition (new content
presented by the teacher as a lecture) within the teacher analysis, had loadings < .3, these were excluded, and factor
analysis was carried out with the remaining 16 items among students and 15 items among teachers. As shown in Table
5, the fit statistics for the final four-factor model for both students and teachers indicated acceptable fit indices.
Table 5
Goodness-of-fit Statistics for Four-Factor Model
4-factor models χ2 df p CFI TLI RMSEA
Students (16 items) 688.67 62 < .001 .96 .93 .05
Teachers (15 items) 79.68 51 .006 .96 .93 .05
These final four factors were meaningfully named as: teacher-centred, cooperative approaches, experimental
approaches, plus solving problems and decision making. The number of items per factors varied from two to seven.
Standardised loadings between the factors ranged from .23 to .64. The internal consistency of all four factors had
acceptable values (Cronbach alpha ranged between .70 - .86).
The factor “teacher-centred” included four approaches in the student model (new content presented by the
teacher as a lecture, class discussion, students ask questions and teacher asks questions i.e., items 1-4) and the same
(minus new content presented by the teacher as a lecture i.e., item 1) in the teacher model. As these covered teaching
approaches mostly initiated, or carried out by the science teacher, this factor was classified as the teacher-centred
approach. The largest factor included seven items, related with different cooperative approaches (items 6-10, 14, 16),
mostly collaborative in nature and requiring active participation by all students, was interpreted as a constructivist
approach. The “experimental approaches” factor was formed from activities related to experimenting – formulate
research questions/ hypothesis, carrying out experiments and making conclusions (items 12, 13, 15), indicating a
student-centred and constructivist approach. The factor “solving problems & decision making” included two highly
correlated items (solving everyday-life related science problems and socio-scientific decision making i.e., items 17, 18),
which students practice in science lessons and thus could be seen as a further constructivism way of teaching/learning.
Correlation Analysis between the Factors among Science Teachers
Correlation analysis showed positive relationships between the 4 factors among science teachers, regarding to the
current situation (Table 6). Weak positive correlations (r < .30) (Cohen et al., 2007) were identified among teacher-centred
approaches compared with cooperative, experimental and problem solving/decision making approaches. Moderate
correlations (r > 0.4) appeared among cooperative, experimental and problem solving/decision making approaches.
Table 6
Descriptive Statistics and Correlations between Factors
Factors M SD 1 2 3 4
1. Teacher-centred approaches 2.72 .36 1.00

2. Cooperative approaches 1.96 .37 .30** 1.00
3. Experimental approaches 1.98 .53 .23** .46** 1.00
4. Problem solving and decision making 2.34 .55 .29** .44** .45** 1.00
Notes. Spearman’s rank correlation coefficient (rho) was used for analysis.
** p < .001
128
(pp. 119-133)
Grade Differences in Student Perceptions of Teaching and Learning Approaches
Table 7 compared grade 6 and 9 grade student differences in their perceptions according to fours factors.
Statistically significant grade differences were indicated across all factors, in a way that teacher-centred, coopera-
tive and experimental approaches were given higher mean scores by grade 6 students compared with students
in grade 9. However, for problem solving/decision making approaches, grade 9 were shown to give a higher mean
value compared with grade 6. Among teacher-centred and cooperative approaches students had less variability
in their answers than within the experimental and problem solving/decision making approaches.
Table 7
Grade differences in Student Perceptions of Teaching and Learning Approaches
Grade 6
Grade 9 (N=848) 95% CI
(N=2673)
Factors t MD SE
M SD M SD LL UL
Teacher-centred approaches 2.65 .38 2.54 .43 7.08* .11 .02 .08 .14
Cooperative approaches 1.92 .38 1.78 .47 9.22* .14 .02 .11 .17
Experimental approaches 1.90 .51 1.80 .55 4.74* .10 .02 .06 .14
Problem solving and decision making 1.82 .62 2.01 .61 -7.63* -.18 .02 -.23 -.14
Note. M – mean; SD – standard deviation; SE – standard error; MD – mean difference; CI – confidence interval; LL – lower limit; UL
= upper limit.
*p < .001
Associations between Teaching and Learning Approaches and

Student Interest/Enjoyment towards Science Learning
The regression coefficients shown in table 8 for the interest/enjoyment of science (as dependent variable)
using a linear regression approach explored the potential predictors among the four teaching and learning factors
(as independent variables). Five models were estimated in total - Model 1 for grade 6 students and Models 2-5 for
grade 9 students (Table 8). In all models, maximum variance inflation (VIF) was established such that there were
no multi-collinearity problems.
Among grade 6 students, results indicated that there was a positive and significant association between in-
creased frequency of teaching and learning approaches for all factors, especially those related to teacher-centred
approaches (β = .22). The increase in the frequency in approaches (the more teacher-centred activities practiced
in science classes) resulted in higher student interest/enjoyment in science learning.
In grade 9, student responses towards interest to learn science were given separately with respect to each of
the separate science subject. Nevertheless, similar to grade 6 outcomes, results showed a tendency for there to
be a positive association between increased frequency of teaching and learning approaches among all models,
but especially related with teacher-centred approaches in models 3 (geography) and 5 (chemistry). In models
2 (biology) and 4 (physics), the problem solving/decision making factor was the strongest and most significant
predictor of interest/enjoyment (accordingly β = .26 and β = .34), indicating that the more everyday-life problem
solving/decision-making activities were carried out, the more students were interested in learning biology and
physics at school. Interest/enjoyment towards the four science subjects in grade 9 involving cooperative ap-
proaches did not significantly associate with each other, as was the case with experimental approaches among
biology and chemistry subjects.
129
(pp. 119-133)
Table 8
Standardised Coefficients (β values) for Teaching and Learning Approach Variables on Student Interest/Enjoyment
Model 1 – Model 2 - Model 3 - Model 4 - Model 5 -

Variables science biology geography physics chemistry
Grade 6 Grade 9
Teaching and learning approaches
Teacher-centred approaches .22** .17* .28** .06 .34**
Cooperative approaches .07* .04 .05 -.00 .09
Experimental approaches .09** .04 .15* .17* .11
Problem solving and decision making .08** .26* .15* .34** .16*
Sample size (N) 2673 220 216 192 220
Adjusted R2 .10 .15 .19 .20 .24
Max VIF 1.43 2.20 1.69 1.42 1.41
* p < .05, ** p < .001
Discussion
The current research explored science teacher and grade 6 and 9 student perceptions of using different teaching
and learning approaches in science class and the relationship with student interest/enjoyment towards science learning.
The results indicated that both students’ and teachers’ perceived teacher-centred approaches were more frequently used
in science classes than those student-centred (cooperative and experimental), both in grade 6 and grade 9. Specifically,
students indicated that lecturing, asking questions and class discussions were approaches most often implemented
compared with role-play, debate, brainstorming and carrying out experiments/making conclusions. These findings were
similar to findings by Juuti et al. (2009) and Hasni and Potvin (2015), who also showed teacher-centred methods were
perceived to be most often used in science classes, even though students expressed a desire to participate more. These
findings were also supported by the TALIS international study (OECD, 2019), which concluded that teacher classroom
practices were often teacher-centred, students passively adopting knowledge from the teacher and, of much concern,
students were not perceived to be cognitively active. Somewhat more contradictory results were put forward by
Hampden-Thompson and Bennett (2013), whose findings showed that the majority of 15-year-old students perceived
investigations occurring in most, or all, science lessons, but only about half the students reported interactions, in terms
of hands-on activities and applications in science, occurring in most, or all, lessons. Nevertheless, as their results were
based on a PISA 2006 UK survey (Hampden-Thompson & Bennett, 2013) and taking into consideration limited activities, it
was not actually possible to draw parallels, for example, with perceptions of teacher-centred activities (e.g., lecturing) as
included in the current research. From a teacher perspective, the current research findings showed that science teachers
did desire to use more student-centred (collaborative and experimental-related) approaches, although unfortunately,
based on responses related to the current situation in the classroom, it seemed the current practice was the opposite
i.e., science teachers most often used teacher-centred approaches.
Results from the current research indicated a relationship between student interest towards science subjects and
the frequency of different teaching and learning approaches being used in science lessons. Specifically, the results
pointed to the teaching and learning approaches being positively associated with student interest/enjoyment to learn
science subjects at school, but not to the same degree. For example, both for grade 6 student interest in science and
grade 9 student interest in geography and chemistry, learning was mostly predicted by teacher-centred approaches
(e.g., lecturing, asking questions), while grade 9 student interest towards biology and physics was triggered by ev-
eryday life problem solving and decision-making approaches. These results were consistent with study by Hasni and
Potvin (2015) who found also positive (moderate to weak) correlations between general interest in S&T at school and
teaching methods factors (e.g. desire to spend more time exposed to traditional teaching methods, degree of student
involvement in developing the inquiry process, desire to spend more time exposed to teaching practices based on open
investigation), indicating that some teaching methods were better predictors of general interest in S&T than others. For
example, hands-on activities appeared to have little effect on interest comparing with preference for open investigation
methods and for traditional teaching methods.
130
(pp. 119-133)
Noting that teaching and learning could take on many formats, an analysis of the current research outcomes
showed four factors associated with the teaching and learning approaches, distinguishing between teacher-centred
(1 factor), or constructivist approaches (3 factors). The first factor related to teacher-centred approaches (new content
presented by the teacher as a lecture, teacher asks questions) and indicated these approaches were more often car-
ried out, as perceived by both by students and teachers, compared with other approaches. Of the 3 factors identified
with social constructivist approaches, two could be categorised as being more collaborative in nature (i.e., factors
identified as group learning and experimenting). Approaches within these factors focused on being student-centred,
including activities that students usually implemented in a social group (brainstorming, debates, carrying out experi-
ments, formulating research questions or hypotheses, etc.) and perceived to be carried out in science lessons, at most,
only sometimes. The factor, identified as less collaborative, related to problem solving/decision making and indicated
a potential controversial outcome, arising, perhaps, through student and teacher lack of familiarity with these terms
when applied to student-centred, or teacher-centred operations. Results of correlation analysis based on teacher data
(related with the current situation) indicated that the four factors had a positive weak relationship with each other, rising
to a more moderate correlation among cooperative, experimental and problem solving/decision making factors. This
showed a significant relationship for the use of a variety of student-centred approaches for those teachers adopting a
more student- centred approach in their teaching.
The results exhibited in Figure 1 indicated that teachings in science classrooms were heavily teacher-centred with
very little emphasis on social constructivist approaches. At the same time, teachers wished to use more social construc-
tivist approaches in the classroom, such as those involving cooperative and experimental teaching in an inquiry setting.
However, a constructivist approach in teaching and learning required different behaviours by both teachers and students,
as well as different classroom management and assessment compared with teacher-centred approaches (Arends, 2012).
More consideration needed to be placed on social constructivist ways of teaching and learning, whereby the teacher’s
role was seen as initiating and maintaining interest/enjoyment towards science learning through the incorporation of
cooperative and experimental activities, taking into account the students’ prior knowledge (Good & Lavigne, 2018).
Conclusions
The current research points to a general perception of science teaching and learning in schools to be teacher-centred,
involving direct teaching of whole groups of students, and providing little time for collaborative and experimental activi-
ties, even though international surveys highlight the importance of student-centred teaching approaches.
The results from this research demonstrate that while science teachers indicate a desire to use more cooperative
and experimental related approaches, perceptions of current teaching by grade 6 and 9 students and also science teach-
ers relate to teacher-centred approaches. Nevertheless, the results also indicate grade differences in that science classes
perceived by grade 6 students tend towards being a mixture of teacher centred, cooperative and experimental, while
problem solving/decision making approaches are perceived to be implemented more in grade 9 classrooms.
The analysis indicates that there is a positive association between both grade 6 and 9 student interest towards science
subjects and the frequency with which different teaching and learning approaches take place in the classroom having
subject differences. Teacher-centred approaches tend to predict grade 6 student interests in science, as well as grade 9
student interests in biology, geography and chemistry, although an orientation towards everyday-life related problem
solving/decision making is seen to have a positive effect in biology and physics.
Recommendations
Noting that the manner in which science is taught, the curriculum presented, and the classroom approaches are
conducted needs to take into account student perspectives if student interest is to be meaningfully promoted, teach-
ers need to be encouraged to put more emphasis into the use of social constructivist approaches in science teaching.
Limitations
The following limitations need to be taken into consideration.

1. The research instrument includes a limited range of teaching and learning approaches and thus the
results of the study cannot be generalised to other approaches used in school science lessons.
2. The study sample covered lower secondary school students (grade 6 and 9); therefore, the results are
valid only for these age groups.
131
(pp. 119-133)
3. The purpose of the current study does not include examining casual, nor longitudinal effects of the
perceptions of teaching and learning approaches used in science classrooms on student interests or
motivation. This can be considered as one aspect for future research.
Acknowledgements
This study was funded by the Estonian Research Council through an institutional research funding project
“Smart technologies and digital literacy in promoting a change of learning” (Grant Agreement No. IUT34-6).
References
Arends, R. I. (2012). Learning to teach (9th ed.). McGraw-Hill.

Asparouhov, T., & Muthén, B. (2009). Exploratory structural equation modelling. Structural Equation Modeling: A Multidisciplinary
Journal, 16(3), 397–438. https://doi.org/10.1080/10705510903008204
Bergin, D. A. (1999). Influences on classroom interest. Educational Psychologist, 34(2), 87–98. https://doi.org/10.1207/s15326985ep3402_2
Bundsgaard, J. (2009). Krydsmodel for undervisningstilrettelæggelse [Cross model for teaching organization]. Dansk
Universitetspædagogisk Tidsskrift, 4(7), 10–17. https://tidsskrift.dk/dut/article/view/5592
Concept to Classrooms. (2020, August 6). What is constructivism and how does this theory differ from traditional ideas about teaching
and learning? https://www.thirteen.org/edonline/concept2class/constructivism/index_sub1.html
Cohen, L., Manion, L., & Morrison, K. (2007). Research methods in education (6th ed.). Routledge.
Deci, E. L., & Ryan, R. M. (2016, January 8). Intrinsic Motivation Inventory (IMI). https://selfdeterminationtheory.org/intrinsic-motivation-
inventory/
Ebenezer, J. V., & Zoller, U. (1993). Grade 10 students’ perceptions of and attitudes towards science teaching and school science.
Journal of Research in Science Teaching, 30(2), 175–186. https://doi.org/10.1002/tea.3660300205
Elliott, S. N., Kratochwill, T. R., Littlefield Cook, J., & Travers, J. (2000). Educational psychology: Effective teaching, effective learning (3rd
ed.). McGraw-Hill College.
Good, T. L., & Lavigne, A. L. (2018). Looking in the classrooms (11th ed.). Routledge.
Hair Jr, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate Data Analysis (7th ed.). Pearson Education.
Hampden-Thompson, G., & Bennett, J. (2013). Science teaching and learning activities and students’ engagement in science.
International Journal of Science Education, 35(8), 1325–1343. https://doi.org/10.1080/09500693.2011.608093
Hasni, A., & Potvin, P. (2015). Student’s interest in science and technology and its relationships with teaching methods, family context and
self-efficacy. International Journal of Environmental & Science Education, 10(3), 337–366. https://doi.org/10.12973/ijese.2015.249a
Hidi, S. (1990). Interest and its contribution as a mental resource for learning. Review of Educational Research, 60(4), 549–572.
https://doi.org/10.3102/00346543060004549
Hidi, S., & Renninger, K. A. (2006). The four-phase model of interest development. Educational Psychologist, 41(2), 111−127.
https://doi.org/10.1207/s15326985ep4102_4
Holbrook, J., & Rannikmae, M. (2014). The philosophy and approach on which the PROFILES project is based. CEPS Journal, 4(1), 9–29.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives.
Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118
Häussler, P., & Hoffmann, L. (2000). A curricular frame for physics education: Development, comparison with
students’ interests, and impact on students’ achievement and self-concept. Science Education, 84(6), 689−705.
https://doi.org/10.1002/1098-237X(200011)84:6<689::AID-SCE1>3.0.CO;2-L
Jones, L. (2007). The student-centered classroom. Cambridge University Press.
Juuti, K., Lavonen, J., Uitto, A., Byman, R., & Meisalo, V. (2010). Science teaching methods preferred by grade 9 students’ in Finland.
International Journal of Science and Mathematics Education, 8(4), 611−632. https://doi.org/10.1007/s10763-009-9177-8
Kousa, P., Kavonius, R., & Aksela, M. (2018). Low-achieving students’ attitudes towards learning chemistry and chemistry teaching
methods. Chemistry Education Research and Practice, 19, 431–441. https://doi.org/10.1039/C7RP00226B
Lalley, J. P., & Miller, R. H. (2007). The learning pyramid: Does it point teachers in the right direction? Education, 128(1), 64–79. https://
www.impudent.org.uk/wordpress/wp-content/uploads/2015/03/Lalley-Miller-TheLearningPyramid-Education-200709-.pdf
McLeod, S. A. (2019, July 17). Constructivism as a theory for teaching and learning. Simply Psychology.
https://www.simplypsychology.org/constructivism.html
Muthén, L. K., & Muthén, B. O. (1998–2017). Mplus user’s guide: Statistical analysis with latent variables (8th ed.). Muthén & Muthén.
https://www.statmodel.com/HTML_UG/introV8.htm
National Curriculum for Basic Schools. (2014). Riigi Teataja I. https://www.riigiteataja.ee/en/eli/524092014014/consolide
OECD. (2016). PISA 2015 Results ( Volume I): Excellence and Equity in Education, PISA, OECD Publishing.
https://doi.org/10.1787/9789264266490-en
OECD. (2019). TALIS 2018 Results (Volume I): Teachers and school leaders as lifelong learners, TALIS, OECD Publishing.
https://doi.org/10.1787/1d0bc92a-en.
Osborne, J., Simon, S., & Collins, S. (2003). Attitudes towards science: A review of the literature and its implications. International
Journal of Science Education, 25(9), 1049−1079. https://doi.org/10.1080/0950069032000032199
Palincsar, A. S. (1998). Social constructivist perspectives on teaching and learning. Annual Review of Psychology, 49, 345–375.
https://doi.org/10.1146/annurev.psych.49.1.345
132
(pp. 119-133)
Pedaste, M., Mäeots, M., Siiman L. A., de Jong, T., van Riesen, S. A. N., Kamp, E. T., Manoli, C. C., Zachariac, Z. C., & Tsourlidaki,
E. (2015). Phases of inquiry-based learning: Definitions and the inquiry cycle. Educational Research Review, 14, 47−61.
http://dx.doi.org/10.1016/j.edurev.2015.02.003
Pedaste, M., Must, O., Leijen, Ä., Mäeots, M., Siiman, L., Kori, K., & Adov, L. (2017). Profiles of students who use mobile
devices for the purposes of learning science and mathematics [Nutiseadmete kasutamise profiilid loodusainete
ja matemaatika õppimise kontekstis]. Eesti Haridusteaduste Ajakiri, Estonian Journal of Education, 5(1), 99–129.
https://ojs.utlib.ee/index.php/EHA/article/view/eha.2017.5.1.04/8460
Potvin, P., & Hasni, A. (2014). Interest, motivation and attitude towards science and technology at K-12 levels: A systematic review of
12 years of educational research. Studies in Science Education, 50(1), 85–129. https://doi.org/10.1080/03057267.2014.881626
Potvin, P., Hasni, A., & Sy, O. (2017). Using inquiry-based interventions to improve secondary students’ interest in science and technology.
European Journal of Science and Mathematics Education, 5(3), 262–270. http://scimath.net/articles/53/533.pdf
Richards, J., & Rodgers, T. (2001). Approaches and methods in language teaching (2nd ed., Cambridge Language Teaching Library).
Cambridge University Press. https://doi.org/10.1017/CBO9780511667305
Sharples, M., de Roock , R., Ferguson, R., Gaved, M., Herodotou, C., Koh, E., Kukulska-Hulme, A., Looi, C-K, McAndrew, P., Rienties, B.,
Weller, M., & Wong, L. H. (2016). Innovating Pedagogy 2016: Open University Innovation Report 5. The Open University.
Sturm, H., & Bogner, F. X. (2008). Student-oriented versus teacher-centred: The effect of learning at workstations about birds
and bird flight on cognitive achievement and motivation. International Journal of Science Education, 30(7), 941–959.
https://doi.org/10.1080/09500690701313995
Taber, K. S. (2011). Constructivism as educational theory: Contingency in learning, and optimally guided instruction. In J. Hassaskhah
(Ed.), Educational Theory (pp. 39-61). Nova Science Publisher.
Teppo, M., Semilarski, H., Soobard, R. & Rannikmäe, M. (2017). 9. klassi õpilaste huvi eri kontekstis esitatud loodusteaduslike teemade
õppimise vastu ja motivatsioon õppida loodusteadusi [Grade nine students’ learning interests towards science topics presented
in different contexts and their motivation to learn science]. Eesti Haridusteaduste Ajakiri. Estonian Journal of Education, 5(1),
130–170. https://doi.org/10.12697/eha.2017.5.1.05
Wang, P-H., Wub, P-L., Yu, K-W., & Lin, Y-X. (2015). Influence of implementing inquiry-based instruction on science learning motivation and
interest: A perspective of comparison. Social and Behavioral Sciences, 174, 1292–1299. https://doi.org/10.1016/j.sbspro.2015.01.750
Young, L. E., & Paterson, B. L. (2007). Teaching nursing: Developing a student-centered learning environment. Lippincott Williams & Wilkins.
Received: August 24, 2020 Accepted: January 10, 2021
Cite as: Teppo, M., Soobard, R., & Rannikmäe, M. (2021). Grade 6 & 9 student and teacher perceptions of teaching and learning
approaches in relation to student perceived interest/enjoyment towards science learning. Journal of Baltic Science Education,
20(1), 119-133. https://doi.org/10.33225/jbse/21.20.119
Moonika Teppo MSc, PhD Student, University of Tartu, Faculty of Science and
(Corresponding author) Technology, Institute of Ecology and Earth Sciences, Centre of
Science Education, Vanemuise 46, Tartu 51003, Estonia.
Website: https://www.ut.ee/et
Regina Soobard PhD, Research Fellow of Science Education, University of Tartu,
Faculty of Science and Technology, Institute of Ecology and Earth
Sciences, Centre of Science Education, Vanemuise 46, Tartu 51003,
Estonia.
Miia Rannikmäe PhD, Professor of Science Education, University of Tartu, Faculty of
Science and Technology, Institute of Ecology and Earth Sciences,
Centre of Science Education, Vanemuise 46, Tartu 51003, Estonia.
133
EFFECT OF A STEM-ORIENTED
COURSE ON STUDENTS’
MARINE SCIENCE MOTIVATION,
Abstract. In recent years, marine science

education has received increasing attention
around the world, but the integration of
Liang-Ting Tsai,
STEM education into the marine curricu-
lum is rarely applied to lower-secondary
Cheng-Chieh Chang,
schools. The purpose of this research was to Hao-Ti Cheng
determine the effect of a teaching module
on the topic of marine science, developed
using a STEM approach, on the motivation,
interest, and marine science achievements. Introduction
The 9th-grade students from two classes in
a Taiwanese public school were recruited With the rapid advancement of technology worldwide, new knowledge
in this study. Students in the experimental societies are being formed. In such societies, knowledge workers are expected
group were taught using the STEM course. to be able to organise and analyse complicated knowledge and information
Students in the control group were taught and subsequently innovate on that basis, with the end goal being economic
in a traditional didactic teaching course. development. On the Partnership for 21st Century Skills website, 21st century
Pre- and post-tests were used to determine skills, otherwise known as the 4 C’s, are critical thinking, creative thinking,
marine science achievement, interest, and communication, and collaboration (Qian & Clark, 2016; Van Laar et al., 2017).
learning motivation. The experimental Governments hope that students and other citizens relentlessly pursue high-
group gained significantly superior scores level thinking and problem-solving skills. As such, conventional education
in three out of four dimensions towards methods must be adjusted to suit this new knowledge era.
motivation and two out of three dimensions In 1986, to promote national development and define requirements for
interest. A significant difference was found nurturing talent, the National Science Foundation proposed a curriculum
between the experimental group and the composed of science, technology, engineering, and mathematics (STEM)
control group, with the former showing and focused on linking knowledge with practical situations (National Science
increased marine science achievement over Board Task Committee on Undergraduate Science and Engineering Educa-
the latter in higher-order thinking; however, tion, 1986). STEM curricula are designed to integrate science, technology,
no significant differences were with respect engineering, and mathematics to encourage students to use technology and
to cognition. It is concluded that involving scientific thinking to solve problems and connect acquired knowledge to life
STEM-based implementations in teaching, experiences (Mohtar et al., 2019). In addition, through nurturing communica-
the subject “Marine Science” increases the tion, teamwork, and practical skills; independent thinking; and the ability to
success rate and attitude towards the subject innovate and create in students such curricula, to ensure students can apply
of students. related knowledge successfully upon entering the workforce.
Keywords: learning interest, learning moti- STEM curricula focus on imparting in students the ability to adapt to
vation, marine science achievement, STEM rapidly changing societies and providing them with the requisite knowl-
edge, attitude, and skills to solve real-world problems. Dabney et al. (2012)
contended that interdisciplinary discussions and activities enrich STEM cur-
Liang-Ting Tsai ricula and enhance students’ learning interest (Raju & Clayson, 2010; Tindall
Tzu Chi University, Taiwan & Hamil, 2004). Harrel (2010) considered STEM education, which involves
Cheng-Chieh Chang, Hao-Ti Cheng
National Taiwan Ocean University, Taiwan interdisciplinary teaching, to be more suitable for instruction in natural sci-
ence and perceptive topics compared with conventional teaching because
solutions to the problems encountered in daily life often involve the appli-
134
ISSN 1648–3898 /Print/ EFFECT OF A STEM-ORIENTED COURSE ON STUDENTS’ MARINE SCIENCE MOTIVATION,
ISSN 2538–7138 /Online/ (pp. 134-145)
cation of knowledge from more than one academic field. According to Dewaters and Powers (2006), students are
generally satisfied with having discussions in STEM courses; such discussions enable them to resolve issues they
encounter in daily life. Moreover, according to Bingölbali et al. (2007), the interdisciplinary STEM teaching process
affects students profoundly, reinforcing their positive attitudes towards STEM curricula and related career choices.
STEM education involves learning topics related to technology and engineering, which, according to Cunningham
and Lachapelle (2016), heightens students’ creativity and higher-level thinking, stimulates the integration of an
entire STEM curriculum, and improves students’ learning motivation and outcomes—the last being a product of
interdisciplinary discussions. Moore et al. (2015) indicated that STEM education, which involves engineering-based
thinking, enhances students’ key 21st century skills, learning interest, and learning outcomes. Cantrell et al. (2006)
reported that incorporating engineering-based thinking in a STEM curriculum improves students’ scientific learn-
ing and comprehensive analytical skills. Schnittka and Bell (2011) reported that employing engineering scenarios
to explain scientific concepts and knowledge reinforces students’ conceptual understanding. Through a teaching
experiment, Ortiz et al. (2015) demonstrated that building a Lego robot design incorporated with mathematical
and engineering concepts considerably increased students’ knowledge of related concepts as well as their learn-
ing retention and transfer.
Lumsden (1994) defined learning motivation as a learner’s voluntary intention to learn. Learners differ in
their learning behaviours and outcomes because they have unique personalities and learning goals. Although
psychologists from various academic disciplines disagree about the definition of learning motivation, they agree
that it relates to learning behaviours. Learning motivation comprises internal motivation (e.g., preferences, feelings,
and interests) and external motivation (e.g., rewards, praise, and punishment avoidance). Keller (1987) established
the attention, relevance, confidence, and satisfaction (ARCS) model, which aims to improve learning motivation.
Interest and motivation are internal factors that trigger specific behaviours. Although the behaviours triggered
by motivation are aimed at a specific goal, motivation does not necessarily lead to satisfaction because goals may
not always be achieved. Subject-specific learning interests can be divided into those related to subject content
and specific learning activities (Häussler et al., 1998). Learning motivations and interests affect learning outcomes
profoundly (Fan, 2011; Pintrich, 2000; Pintrich & De Groot, 1990; Solak & Cakir, 2015; Tasgin & Coskun, 2018; Tuan et
al., 2005). Studies have revealed that favourable learning motivation can improve learning efficiency, performance,
and outcomes (Afzal et al., 2010; Berger & Karabenick, 2011; Ismail et al., 2010; Maliqi & Borincaj-Cruss, 2015; Pokay
& Blumenfled, 1990).
Taiwan is surrounded by ocean and possesses abundant ocean resources. However, pollution of ocean waters
and the severity of ocean-related environmental problems have continued to increase. Inappropriate handling of
oceanic resources and helplessness in the face of marine disasters indicate that Taiwanese citizens have generally
received insufficient ocean education (Chang, 2015; Tsai et al., 2019). In the face of globalisation, the supply of tal-
ent for ocean-related industries does not meet the demand, a problem which is exacerbated by students having
low willingness to seek occupations in marine industries. Therefore, the Taiwanese government has increasingly
focused on ocean education in recent years under an aggressive policy to quickly nurture ocean-related talents
who can contribute to national development (Ministry of Education, 2017). This study focused on applying STEM
education to ocean-related curricula and to highlighting the problem of marine debris (trash in the ocean). The
objective was to help students explore marine science in an interdisciplinary manner, collect and analyse scientific
data, and use technology and engineering design techniques to support their thinking and actions. As an outcome,
students were expected to be able to create devices that solve or mitigate the problem of marine debris.
Research Aim and Questions
The purpose of this research was to determine how incorporating STEM education into marine science edu-
cation would affect the learning motivation, interest, and achievement of lower-secondary schools students. The
following research goals were proposed:
1. To explore the effects of a STEM-oriented marine science course on the learning motivation of junior
high school students.
2. To explore the effects of a STEM-oriented marine science course on the learning interest of junior high
school students.
3. To explore the effects of a STEM-oriented marine science course on the learning achievement of junior
high school students.
135
(pp. 134-145) ISSN 2538–7138 /Online/
Research Design
One experimental group and one control group were used. The experimental group received teaching in a
STEM-oriented marine science course, whereas the control group received teaching in employing didactic teaching.
Because of class size restrictions imposed to maintain normal classroom operations during the semester, random
sampling of participants and strict control of experimental scenarios were not possible. Purposive sampling was
used to select two classes, with one to serve as the experimental group and the other as the control group. The
experimental group received STEM course teaching that integrated beliefs, principles, and methods related to STEM
into marine science, with a focus on marine debris (National Science Board Task Committee on Undergraduate
Science and Engineering Education, 1986; Qian & Clark, 2016; Van Laar et al., 2017). The control group students
received conventional didactic teaching. The study period was 6 weeks; two classes were held each week from
February to April 2018. The students underwent Pre-test that assessed their learning motivation, interests, and
achievements. Immediately after the end of the course, the same three topics were evaluated through a Post-test.
Participants
The participants were 9th-grade students from two classes at a public lower-secondary school in Taiwan.
Upon entering lower-secondary school (seventh grade), these students were placed in classes according to their
academic level through the spiral arrangement method; for this reason, their academic levels were approximately
equivalent. The experimental group consisted of 20 students (9 boys and 11 girls), and the control group con-
sisted of 27 students (13 boys and 14 girls). All the students were approximately 16 years old. The two groups of
participating students first filled out a consent form to indicate their agreement to take part in this research, after
which they then completed the questionnaires anonymously.
Education Approach
STEM Course
Biweekly classes were held for 6 weeks (12 classes in total), and each class lasted 45 minutes. Through questions,
answers, discussions, and exercises, educators guided students to develop good communication skills and to use
their acquired knowledge to solve problems and complete hands-on learning activities. The planned education
process is presented in Table 1.
Table 1
Learning Progress in the STEM-Oriented Marine Science Course
Week Segment Activities
1 The importance of the ocean

2 Ocean currents and circulation Ocean current simulation experiment
3 Trash in the ocean
4 Trash belts in the ocean
5 The effects of trash on ocean ecosystems
6 Protecting the ocean and removing ocean trash Creation of an ocean vacuum cleaner
The curriculum was based on the perspective of Trowbridge and Bybee (1990) and the 5E (engage, explore,
explain, elaborate, and evaluate) instructional model they proposed in the Biological Science Curriculum Study. The
aim of using this teaching strategy was to design educational content that can help students develop their spirit of
exploration. The learning activities in this study encompassed, for example, the topics of scientific principles and
136
ISSN 2538–7138 /Online/ (pp. 134-145)
the use of technology, engineering design, mathematical statistics, and chart analysis. The aim of the integrated
STEM-oriented curriculum was to consolidate scientific knowledge and skills through hands-on activities so that
students could apply scientific knowledge to real-life problem solving. Each STEM component is presented in Table 2.
Table 2
STEM Knowledge Analysis and Assessment for the Marine Science Curriculum
Subject Teaching concepts
1. Water circulation
2. Oceans and climate adjustments
3. Ocean currents, planetary wind systems, circulation, and wind speeds
4. Material decomposition
5. Buoyancy
Science ( S ) 6. Types and sources of ocean trash
7. Understanding the effects of ocean trash decomposition
8. Biomagnification
9. Discussing methods for reducing plastic product usage and preventing waste from becoming ocean trash
10. Writing down the thinking process and explaining it (scientific method)
11. Principles of an ocean trash cleaning device
1. Principles of ocean current power generation, tidal power generation, and energy development
2. Designing an experiment
3. Designing a project that solves a problem
Technology ( T )
4. Designing and simulating a small-scale ocean current model
5. Learning about plastic waste treatment methods
6. Designing a device for removing ocean trash
1. Processes of ocean current power generation, tidal power generation, and energy development
2. Creating a device
Engineering ( E )
3. Thinking of strategies for preventing or removing ocean trash
4. Creating a device for removing ocean trash
1. Calculating the amount and volume of ocean trash
2. Collecting and analysing data
Math ( M ) 3. Presenting data in charts and graphs and explaining the results
4. Converting data into percentages and explaining them
5. Creating an easy-to-understand presentation of data by using a chart
Didactic Teaching Course
The didactic teaching course in this research was prepared by researchers and contained identical conditions
to those of the STEM course in terms of science content, teaching materials, and pace of teaching. This 6-week
course (12 classes, 45 minutes each) was based on group learning. The teaching materials given to students and
the education progress markers were identical to those in the experimental group. The content of the course is
presented as in Table 1, but no activities are included. This group is teacher-centred, with students sitting quietly
and listening attentively, only passively accepting the teacher’s efforts to convey the marine science knowledge
in Table 1.
Instruments

The instruments used in this study included a marine science learning motivation scale, a marine science
learning interest scale, and a marine science learning achievement test.
137
(pp. 134-145) ISSN 2538–7138 /Online/
Marine Science Learning Motivation Scale
Keller’s ARCS model (Keller, 1987) was the foundation for developing this study’s learning motivation scale. For
this 5-point Likert-type scale, students could select one option for each question (strongly agree, agree, neither agree
nor disagree, disagree, or strongly disagree). The scale had four subscales, namely attention, relevance, confidence,
and satisfaction, and 16 questions in total were asked (Lin et al., 2019). Although Lin et al. (2019) demonstrated
that this scale has good reliability and validity, its internal consistency and construct validity were again verified
in this study.
Cronbach’s α was used to measure the internal consistency of the scale and subscales. The α values for atten-
tion, relevance, confidence, and satisfaction were respectively .96, .93, .92, and .96, and the total scale score was
.98. All values were above .9. The confirmatory factor analysis (CFA) results are presented in Table 3. The subscale
of emotional reactions towards the ocean contained three items. It had an average variance extracted of .78 and
component reliability of .91. The acknowledgement of marine science’s importance subscale contained four items,
and it had an average variance extracted of .82 and component reliability of .95. The actions taken to learn marine
science subscale contained four items (average variance extracted: .75, component reliability: .92). The average
variance extracted values of these three subscales satisfied the requirement that they be >.5. Moreover, the three
subscales had favourable component reliability scores. The learning motivation and learning interest scales thus
had appropriate internal consistency and construct validity.
Marine Science Learning Interest Scale
Marine science learning interest was assessed using a 5-point Likert-type scale (strongly agree, agree, neither
agree nor disagree, disagree, or strongly disagree). The scale had three subscales, namely emotional reactions towards
the ocean, acknowledgement of marine science’s importance, and actions taken to learn marine science; 11 ques-
tions were asked (Lin et al., 2019). The marine science learning interest and marine science learning motivation
scales were both reviewed by four experts to evaluate their content validity and remove unsuitable questions.
These scales were pretested together before educational instruction commenced. The Pre-test was administered
by test administration committee members, all of whom had undergone standardised test training.
CFA was used to evaluate the construct validity of the scales. The results are presented in Table 3. The atten-
tion subscale of the marine science learning motivation scale contained five items (average variance extracted:
.82, component reliability: .96). The relevance subscale contained three items (average variance extracted: .82,
component reliability: .93). The confidence subscale contained three items (average variance extracted: .80, com-
ponent reliability: .92). The satisfaction subscale contained five items (average variance extracted: .83, component
reliability: .96). The average variance extracted values of all the four subscales satisfied the requirement that they
be >.5. Moreover, the four subscales had good component reliability scores. Generally, a component reliability
above .6 is considered appropriate and a reliability above .5 is acceptable (Bagozzi & Yi, 1988; Diamantopoulos &
Siguaw, 2000; Kline, 1998).
Table 3
Research Variables and Subscales as Well as Average Variance Extracted and Component Reliability Values
Variance
Variables Subscale Number of items Component reliability
extracted
Attention 5 .82 .96
Marine science learn- Relevance 3 .82 .93

ing motivation
Confidence 3 .80 .92
Satisfaction 5 .83 .96
Emotional feeling toward the ocean 3 .78 .91
Marine science learn-
Acknowledgment of marine science 4 .83 .95
ing interest
Actions taken toward marine science 4 .75 .92
138
ISSN 2538–7138 /Online/ (pp. 134-145)
Marine Science Learning Achievement Test
The marine science learning achievement test consisted of 15 single-choice items. The cognitive processes
of memory (four items), understanding (four items), and higher-level thinking (seven items) were analysed. To
ensure the test had content validity, it was designed by the authors and three marine science experts, who were
also secondary school teachers. The measurement framework for this test was based on ocean literacy and knowl-
edge, especially the seven basic principles of ocean literacy set forth by the National Marine Educators Association
(NMEA, 2010) and National Oceanic and Atmospheric Administration (NOAA, 2013). The seven basic principles of
ocean literacy are as follows: (1) The Earth has one large ocean with numerous features; (2) the ocean and ocean life
shape the features of the Earth; (3) the ocean has a major influence on weather and climate; (4) the ocean makes
the Earth habitable; (5) the ocean supports diverse life and ecosystems; (6) the ocean and humans are inextricably
interconnected; and (7) the ocean is largely unexplored (NMEA, 2010).
The Pre-test and Post-test results for the learning achievement test in the current study were analysed using
the Kuder–Richardson Formula 20 (KR-20). The KR-20 estimates reliability according to the consistency of all items.
The KR-20 results for the pilot study, Pre-test, and Post-test were .67, .62, and .66, respectively, indicating that this
instrument had appropriate internal consistency.
Statistical Analyses
Cronbach’s α and CFA were employed to evaluate the reliability and construct validity of the instruments.
Mean and standard deviation were used to describe learning motivation, interest, and achievement. A one-way
analysis of covariance (ANCOVA) was used to evaluate differences between the pre-test and post-test scores for
learning motivation, interest, and achievement.
Research Results
Comparison of Learning Motivation Between Groups
Descriptive statistics for the learning motivation of students in the experimental and control groups before
and after marine science course instruction are presented in Table 4. The results in Table 4 indicate that the experi-
mental group attained higher scores than the control group for all subscales and the total test.
Table 4
Descriptive Statistics for Learning Motivation, Interest, and Achievement
Number Pre-test Post-test

Scales Group
of items M SD M SD
Experimental 15.75 2.34 18.85 2.87
Attention 5
Control 15.48 3.72 17.52 4.77
Experimental 13.00 2.10 15.10 2.13
Relevance 3
Control 11.56 4.10 13.96 3.40
Experimental 10.95 1.61 12.05 1.61
Learning Motiva- Confidence 3
tion Control 9.52 3.43 10.85 2.93
Experimental 15.05 3.50 17.35 3.45
Satisfaction 5
Control 14.11 5.66 15.93 4.90
Experimental 54.75 7.99 63.35 7.96
Total score 16
Control 50.67 15.70 58.26 14.90
139
(pp. 134-145) ISSN 2538–7138 /Online/
Number Pre-test Post-test

Scales Group
of items M SD M SD
Emotional feeling Experimental 9.90 1.62 10.3 1.53
3
toward ocean Control 9.48 2.49 9.81 1.82
Acknowledgment of Experimental 13.95 2.74 15.5 2.33
4
marine science Control 12.59 4.37 14.07 3.46
Learning Interest
Actions taken toward Experimental 13.05 2.74 13.95 2.24
4
marine science Control 11.30 3.88 13.15 3.11
Experimental 36.90 5.03 39.75 4.54
Total score 11
Control 33.37 10.20 37.04 7.30
Experimental 3.10 .96 3.20 1.06
Memory 4
Control 2.19 1.18 3.37 .97
Experimental 2.85 1.09 2.90 .72
Understanding 4
Control 2.37 .88 3.07 .92
Achievement
Experimental 4.40 1.64 4.80 1.65
Higher-level 7
Control 3.74 1.87 3.59 1.85
Experimental 10.35 2.82 10.90 2.83
Total score 15
Control 8.30 2.77 10.04 2.65
To compare learning motivation scores between experimental and control groups, a one-way ANCOVA was
conducted for data analysis; the results are presented in Table 5. A significant difference (F = 4.22, p = .046 [<.05])
in attention scores was observed between the experimental and control groups. A significant difference was also
identified in relevance (F = 5.77, p = .021 [<.05]) and satisfaction (F = 5.77, p = .021 [<.05]) scores. The total scores
were also significantly different between the groups (F = 5.87, p = .020 [<.05]). The results indicated that the
experimental group significantly outperformed the control group in attention, relevance, satisfaction, and total
scores. This indicated that the STEM-oriented marine science course for lower-secondary school students positively
affected their learning motivation.
Table 5
ANCOVA Results for Learning Motivation, Interest, and Achievement
Type III Average

Scales Source sum of df sum of F p
square square
Between 82.26 1 82.26 4.22 .046*

Attention
Within 858.55 44 19.51
Between 55.09 1 55.09 5.77 .021*
Relevance
Within 420.03 44 9.55
Between 25.02 1 25.02 3.75 .059
Learning motivation Confidence
Within 293.39 44 6.67
Between 128.11 1 128.11 5.77 .021*
Satisfaction
Within 977.24 44 22.21
Between 1140.39 1 1140.39 5.87 .020*
Total score
Within 8545.90 44 194.23
140
ISSN 2538–7138 /Online/ (pp. 134-145)
Type III Average

Scales Source sum of df sum of F p
square square
Emotional feeling toward Between 31.18 1 31.18 7.43 .009**

marine science Within 184.52 44 4.19
Acknowledgment of marine Between 48.52 1 48.52 4.71 .035*
science Within 453.13 44 10.30
Learning interest
Between 34.94 1 34.94 3.71 .061
Actions taken toward marine
science Within 414.97 44 9.43
Between 339.31 1 339.31 6.32 .016*

Total score
Within 2362.46 44 53.69
Between .31 1 .31 .32 .576
Memory
Within 43.25 44 .98
Between .14 1 .14 .19 .663
Understanding
Within 31.04 44 .71
Achievement
Between 35.77 1 35.77 10.82 .002**
Higher-level
Within 145.41 44 3.31
Between 33.88 1 33.88 4.32 .044*
Total score
Within 345.22 44 7.85
*p < .05, **p < .01, ***p < .001.
The results indicated that with respect to total scores, the effect of the STEM-oriented course on high school
student’s motivation to learn marine science was significant. For the learning motivation scale, a significant
between-group difference was observed in attention subscale scores. Those taking the STEM-oriented course be-
lieved that marine science courses are highly attractive, possibly because in STEM education, real-life scenarios are
often simulated that force students to think independently and conduct group discussions, resulting in increased
opportunities for hands-on learning. Such opportunities can increase student engagement.
Comparison of Learning Interest Between Groups
The results in Table 4 indicate that the experimental group scored higher than the control group in the sub-
scales of emotional reactions towards the ocean, acknowledgement of marine science’s importance, and the actions
taken to learn towards advancing marine science in the learning interest pre-test and post-test. The experimental
group also attained a higher total score than that of the control group.
The one-way ANCOVA results are presented in Table 5, and they indicate a significant difference in learning inter-
est between groups with respect to emotional reactions towards the ocean (F = 7.43, p = .009 [<.01]) and acknowl-
edgement of marine science’s importance (F = 4.71, p = .035 [<.05]). Actions taken to learn marine science subscale
scores were not significantly different (F = 3.71, p = .061 [>.05]) between the groups. The total score was significantly
different (F = 6.32, p = .016 [<.05]) between the groups. The experimental group outperformed the control group in
the emotional reactions towards marine science, acknowledgement of marine science’s importance and total scores,
implying that the STEM-oriented marine science course had positive effects on the learning interest of students.
The effect of the STEM-oriented course on the total score for learning interest was significant. Incorporating
the engineering thinking component of STEM into education is beneficial for students’ development of key 21st
century skills and can improve their interest and achievement in STEM fields.
Significant differences were noted between student group scores in the emotional reactions towards marine
science subscale. Scores of the STEM-oriented education group revealed a significantly increased enjoyment of
and interest in marine science. STEM education may have led to these increases through the incorporation of
knowledge and skills required in daily life into the curriculum. Therefore, these students did not view marine sci-
ence irrelevant or boring. Through hands-on activities, students came to understand that they could contribute
to protecting the ocean through their actions.
141
(pp. 134-145) ISSN 2538–7138 /Online/
With respect to acknowledging marine science’s importance, significant differences were observed between
the groups. Students in the STEM education group believed they could effectively obtain knowledge and skills
related to marine science from the curriculum. STEM education emphasises interdisciplinary curricula and ground-
ing coursework in daily life situations to demonstrate the practical value of course content. Problems encountered
in daily life are often relevant to two or more disciplines. If education is strictly siloed in disciplines, students may
encounter difficulty learning and ultimately lose interest in a course. The interdisciplinary nature of STEM educa-
tion can help students overcome such difficulties. STEM education is especially appropriate for education related
to natural sciences and the sensory world. When education has an entertaining approach, students can be guided
to develop an interest in their courses and may acquire an in-depth understanding of the relevance and practi-
cality of the material. Subsequently, they may have an increased willingness to learn and ability to absorb more
knowledge. Compared with the didactic teaching course group, the STEM education group demonstrated greater
marine science learning interest with respect to emotional reactions towards marine science and acknowledge-
ment of marine science’s importance. If educators can guide students to engage in learning not strictly oriented
to test-taking and increase their interest in marine science, more students may experience sustained, autonomous
learning of marine science, which may increase their marine science–related activities.
Comparison of Learning Achievement Between Groups
The results in Table 4 indicate that the experimental group had higher scores in all subscales (memory, un-
derstanding, and higher-level orientation) and a higher total score than that of the control group. The one-way
ANCOVA results are presented in Table 5, and they indicate no significance between-group differences in memory
(F = 0.32, p = .576 [>.05]) and understanding (F = 0.19, p = .663 [>.05]). However, the experimental group obtained
higher scores for higher-level orientation (F = 10.82, p = .002 [<.01]) and higher total scores (F = 4.32, p = .044 [<.05])
than did the control group.
Discussion
The main aim of this research was to determine the effect of incorporating STEM aspects into marine science
education on the learning motivation, interest, and performance of lower-secondary school students. With respect
to learning motivation, students in the STEM education group scored better in attention, relevance, and satisfaction
than did the control students. Confidence was the only subscale in which significant differences were not evident.
Students in the STEM education group may not have been familiar with the tools used or lacked the confidence
required to devise possible approaches. Skills learning was not reflected in the test scores. Therefore, no increase in
confidence was observed in the learning motivation of students in the STEM education group. If students undergo
long-term development that results in having the confidence to express their thoughts and propose ideas, and
if they can be given the opportunity to familiarise themselves with related tools, they may develop a high level
of confidence in their execution ability. Therefore, teachers seeking to increase students’ marine science learning
motivation should consider incorporating STEM education into their teaching approach.
This result agrees with the conclusion of Cunningham and Lachapelle (2016). Gaining learning experiences
related to technology and engineering promotes creativity and higher-level thinking and can encourage the con-
solidation of STEM subjects. In interdisciplinary discussion scenarios, such an approach can also improve learning
motivation and performance (Cunningham & Lachapelle, 2016). In student views on marine science’s relevance,
significant between-group differences were observed. Compared with the students in the control group, the stu-
dents in the STEM education group had a stronger belief that marine science was highly relevant to them. STEM
education emphasises connecting knowledge and real life through the incorporation of current events and real-life
examples into a curriculum. Students can identify problems and propose relevant solutions. This finding accords
with the conclusion of Dewaters and Powers (2006). In their study, students were satisfied with interdisciplinary
discussion of STEM themes and believed that their courses and experiences were beneficial for solving problems
encountered in daily life (Dewaters & Powers, 2006). With respect to course satisfaction, the groups exhibited sig-
nificant differences. Students in the STEM education group stated that the marine science curriculum gave them
a greater sense of achievement. STEM education guides students to first identify problems and then design and
manufacture products that can solve these problems. Students are engaged throughout the process, and the tan-
gible finished products can give students a sense of accomplishment. Therefore, students in the STEM education
142
ISSN 2538–7138 /Online/ (pp. 134-145)
group had greater satisfaction with their course than the control group had with theirs.
The STEM education group scored higher than the control group did in two learning interest subscales, namely
emotional reactions towards marine science and acknowledgement of marine science’s importance. However,
no significant difference was observed in the actions taken to learn marine science subscale. Possible reasons for
this result are that both groups of students did not have an in-depth understanding of the practical utility of their
marine science education and future marine science developments, or that they held certain preconceptions
and believed that such an education would not benefit them in the future. Furthermore, marine science is not
a primary test subject in Taiwan. Therefore, as neither group would be tested on the subject (and thus, neither
group perceived a benefit in continued study), the scores related to actions to continue the pursuit of knowledge
and skills related to marine science were not significantly different between the two groups of students. Overall,
the effect of STEM education on students’ learning interest in marine science courses was superior to that of the
didactic teaching course. Therefore, instructors who hope to increase learning interest in marine science should
consider incorporating STEM education into related curricula. This recommendation concurs with those of Dabney
et al. (2012), Raju and Clayson (2010), and Tindall and Hamil (2004). When students participate in STEM learning,
interdisciplinary discussions and learning activities give STEM themes greater meaning and increase students’
learning interest in STEM fields.
Regarding learning achievement, students in the STEM education group had higher scores than did control
group students in higher-level orientation; however, no significant differences were observed in memory and
understanding. The didactic teaching course group had superior performance to the STEM group in higher-level
orientation, possibly because STEM education involves using exploratory teaching methods to stimulate student
thinking as well as using hands-on activities to link life experiences to learning and increase students’ evaluation
ability. The results indicated that after STEM education was incorporated into teaching, its effect on the learning
achievement of junior high school students was significant, as indicated by total scores. These findings accord with
those of Cunningham and Lachappelle (2016), English (2015), and Moundridou and Kaniglonou (2008). The results
for higher-level learning achievement (including application, analysis, combination, and evaluation) between
groups were significantly different. This result is in agreement with that of Cunningham and Lachapelle (2016),
who posited that participation in learning activities related to technology and engineering promotes creativity
and higher-level thinking.
The purpose of this research was to determine the effect of a teaching module on the topic of marine science,
developed using a STEM approach, on the motivation, interest, and marine science achievements. With respect
to learning motivation, students in the STEM education group scored better in attention, relevance, and satisfac-
tion than did the control students. The STEM education group also scored higher than the control group did in
emotional reactions towards marine science and acknowledgement of marine science’s importance subscales of
learning interest. Regarding learning achievement, students in the STEM education group had higher scores than
did control group students in higher-level orientation.
From the conclusions of this research, some suggestions can be provided for further educational curriculum.
For learning motivation, students in the STEM education group scored higher in attention, relevance, and satisfaction
than did students in the control group. The confidence subscale was the only one in which significant differences
were not observed. Possibly, students in the STEM education group were not familiar with related tools or lacked
the confidence required to express their ideas and propose solutions. In addition, they may have felt that the skills
they learned would not be reflected in test scores. Therefore, in terms of learning motivation, students in the STEM
education group had no increase in confidence. Instructors who hope to increase the learning motivation of students
with respect to marine science should consider incorporating STEM education into their teaching approaches.
For learning interest, students in the STEM education group scored higher than did the control students on
two subscales, namely emotional reactions towards marine science and acknowledgement of marine science’s
importance. Possibly, both student groups did not have an in-depth understanding of the practical use of their
marine science education and how they might develop marine science knowledge. Furthermore, they may have
held preconceptions and believed that a marine science education would not benefit their future lives. Overall,
teachers who hope to increase learning interest in marine science should consider incorporating STEM education
into such curricula.
143
(pp. 134-145) ISSN 2538–7138 /Online/
With respect to learning achievement, students in the STEM education group scored higher than did the
control students in higher-level orientation. This may have been because STEM educators use exploratory teach-
ing methods to stimulate student thinking and use hands-on activities to link learning to relevant life experiences
and increase the evaluation ability of students.
Acknowledgements
The authors would like to thank the lower-secondary school teachers and the students from Taiwan for their
substantial assistance in the participation of the research.
References
Afzal, H., Ali, I., Khan, M. A., & Hamid, K. (2010). A study of university students’ motivation and its relationship with their academic
performance. International Journal of Business and Management, 5(4), 80–88. https://doi.org/10.5539/ijbm.v5n4p80
Bagozzi, R. & Yi, Y. (1988). On the evaluation of Structural Equation Models. Journal of the Academy of Marketing Sciences, 16, 74-
94. https://doi.org/10.1007/BF02723327
Berger, J. L., & Karabenick, S. A. (2011). Motivat ion and students’ use of learning strategies: Evidence of unidirectional effect s
in mathematics classrooms. Learning and Instruction, 21(3), 416–428. https://doi.org/10.1016/j.learninstruc.2010.06.002
Bingölbali, E., Monaghan, J., & Roper, T. (2007). Engineering students’ conceptions of the derivative and some implications for
their mathematical education. International Journal of Mathematical Education in Science and Technology, 38(6), 763–777.
https://doi.org/10.1080/00207390701453579
Cantrell, P., Pekcan, G., Itani, A., & Velasquez‐Bryant, N. (2006). The effects of engineering modules on student learning in middle school
science classrooms. Journal of Engineering Education, 95(4), 301–309. https://doi.org/10.1002/j.2168-9830.2006.tb00905.x
Chang, C. C. (2015). A study on the influential factors of marine science knowledge and capability indicators for junior high
school students. Education Journal, 43(2), 173-196.
Cunningham, C. M., & Lachapelle, C. P. (2016). Designing engineering experiences to engage all students. Journal of the International
Society for Design and Development in Education, 3(9), 1–26. https://www.educationaldesigner.org/ed/volume3/issue9/
article31/pdf/ed_3_9_cunningham_16.pdf
Dabney, K. P., Tai, R. H., Almarode, J. T., Miller-Friedmann, J. L., Sonnert, G., Sadler, P. M., & Hazari, Z. (2012). Out-of-school time
science activities and their association with career interest in STEM. International Journal of Science Education, Part B, 2(1),
63–79. https://doi.org/10.1080/21548455.2011.629455
Dewaters, J., & Powers, S. E. (2006). Improving science and energy literacy through project- based K-12 outreach efforts that use
energy and environmental themes. Paper presented at 2006 Annual Conference & Exposition, Chicago, Illinois.
Diamantopoulos, A., & Siguaw, J. A. (2000). Introducing LISREL: A guide for the uninitiated. Sage Publications.
Fan, W. (2011). Social influences, school motivation and gender differences: An application of the expectancy-value theory.
Educational Psychology, 31(2), 157–175. https://doi.org/10.1080/01443410.2010.536525
Harrel, P. E. (2010). Teaching an integrated science curriculum: Linking teacher knowledge and teaching assignments. Issues in
Teacher Education, 19(1), 145–165.
Häussler, P., Hoffman, L., Langeheine, R., Rost, J., & Sievers, K. (1998). A typology of students’ interest in physics and
the distribution of gender and age within each type. International Journal of Science Education, 20(2), 223–238.
https://doi.org/10.1080/0950069980200207
Ismail, A., Hasan, A., & Sulaiman, A. Z. (2010). Supervisor’s role as an antecedent of training transfer and motivation to learn in
training programs. Acta Universitatis Danubius, 6(2), 18–37.
Keller, J. M. (1987). Development and use of the ARCS model of instructional design. Journal of Instructional Development, 10(3),
2–10.
Kline, R. B. (1998). Principles and practice of structural equation modeling. Guilford Press.
Lumsden, L. S. (1994). Student motivation to learn. ERIC Digest, 92, 43–58.
Maliqi, A., & Borincaj-Cruss, I. (2015). The influence of teachers on increasing student’s motivation to the Ismail Qemaili high
school in the city of Kamenica, Kosovo. Psychology, 6(8), 915–921. https://doi.org/10.4236/psych.2015.68089
Ministry of Education (2017). Marine Education Policy White Paper. Retrieved from https://ws.moe.edu.tw/001/Upload/3/
relfile/6315/55805/40f900df-a70b-4e1f-847f-e53ed94b4c77.pdf
Mohtar, L. E., Halim, L., Rahman, N. A., Maat, S. M., Iksan, Z. H., & Osman, K. (2019). A model of interest in STEM careers among
secondary school students. Journal of Baltic Science Education, 18(3), 404–416. https://doi.org/10.33225/jbse/19.18.404
Moore, T. J., Tank, K. M., Glancy, A. W., & Kersten, J. A. (2015). NGSS and the landscape of engineering in K‐12 state science standards.
Journal of Research in Science Teaching, 52(3), 296–318. https://doi.org/10.1002/tea.21199
Moundridou, M., & Kaniglonou, A. (2008). Using LEGO MINDSTORMS as an instructional aid in technical and vocational secondary
education: Experiences from an empirical case study. In P. Dillenbourg & M. Specht (Eds.), Times of convergence: Technologies
across learning contexts (pp. 312–321). Springer.
National Science Board Task Committee on Undergraduate Science and Engineering Education (1986). Undergraduate Science,
Mathematics, and Engineering Education, Washington, DC: National Science Board.
144
ISSN 2538–7138 /Online/ (pp. 134-145)
NMEA (2010). NMEA special report #3: The ocean literacy campaign. Retrieved from https://www.marine-ed.org/ocean-literacy/
special-report
Ortiz, A. M., Bos, B., & Smith, S. (2015). The power of educational robotics as an integrated STEM learning experience in teacher
preparation programs. Journal of College Science Teaching, 44(5), 42-47. https://doi.org/10.2505/4/jcst15_044_05_42
Pintrich, P. R. (2000). Multiple golas, multiple pathways: The role of goal orientation in learning and achievement. Journal of
Educational Psychology, 92(3), 544–555. https://doi.org/10.1037/0022-0663.92.3.544
Pokay, P., & Blumenfeld, P. C. (1990). Predicting achievement early and late in the semester: The role of motivation and use of
learning strategies. Journal of Educational Psychology, 82(1), 41-50. https://doi.org/10.1037/0022-0663.82.1.41
Qian, M., & Clark, K. R. (2016). Game-based Learning and 21st century skills: A review of recent research. Computers in Human
Behavior, 63, 50–58. https://doi.org/10.1016/j.chb.2016.05.023
Raju, P. K., & Clayson, A. (2010). The future of STEM education: An analysis of two national reports. Journal of STEM Education:
Innovations and Research, 11(5/6), 25.
Schnittka, C., & Bell, R. (2011). Engineering design and conceptual change in science: Addressing thermal energy and heat transfer
in eighth grade. International Journal of Science Education, 33(13), 1861–1887. https://doi.org/10.1080/09500693.2010.529177
Solak, E., & Cakir, R. (2015). Language learning strategies of language e-Learners in Turkey. E-Learning and Digital Media, 12(1),
107–120. https://doi.org/10.1177/2042753014558384
Tasgin, A., & Coskun, G. Ă. (2018). The relationship between academic motivations and university students’ attitudes towards
learning. International Journal of Instruction, 11(4), 935–950.
Tindall, T., & Hamil, B. (2004). Gender disparity in science education: The causes, consequences, and solutions. Education, 125(2),
282–296.
Trowbridge, L. W., & Bybee, R. W. (1990). Becoming a secondary school science teacher (5th ed.). Merrill.
Tsai, L. T., Lin., Y. L., & Chang, C. C. (2019). An assessment of factors related to ocean literacy based on gender-invariance measurement.
International Journal of Environmental Research and Public Health. 16(19), 3672; https://doi.org/10.3390/ijerph16193672
Tuan, H. L., Chin, C. C., & Shieh, S. H. (2005). The development of a questionnaire to measure students’ motivation towards science
learning. International Journal of Science Education, 27(6), 639–654. https://doi.org/10.1080/0950069042000323737
Van Laar, E., Van Deursen, A. J., Van Dijk, J. A., & De Haan, J. (2017). The relation between 21st-century skills and digital skills: A
systematic literature review. Computers in Human Behavior, 72, 577–588. https://doi.org/10.1016/j.chb.2017.03.010
Received: October 08, 2020 Accepted: January 20, 2021
Cite as: Tsai, L.-T., Chang, C.-C., & Cheng, H.-T. (2021). Effect of a stem-oriented course on students’ marine science motivation,
interest, and achievements. Journal of Baltic Science Education, 20(1), 134-145. https://doi.org/10.33225/jbse/21.20.134
Liang-Ting Tsai Assistant Professor, Institute of Education & Center of Teacher

Education, Tzu Chi University, 701 Zhongyang Rd., Sec. 3, Hualien
97004, Taiwan (R.O.C).
Cheng-Chieh Chang Professor, Director of Taiwan Marine Education Center, Institute of
(Corresponding author) Education & Center of Teacher Education, National Taiwan Ocean
University, Taiwan. (R.O.C)
Hao-Ti Cheng Master, Taipei Municipal Dong-Hu Junior High School Teacher,
Institute of Education, National Taiwan Ocean University, No.2,
Beining Rd., Jhongjheng District, Keelung City 202, Taiwan (R.O.C)
145
ANALYSIS OF TWO-TIER
QUESTION SCORING METHODS:
A CASE STUDY ON THE
LAWSON’S CLASSROOM TEST
OF SCIENTIFIC REASONING
Abstract: The Lawson’s Classroom Test of

Shao-Na Zhou,
Scientific Reasoning (LCTSR) is a popular
Qiao-Yi Liu,
instrument that measures the development
Kathleen Koenig,
of students’ scientific reasoning skills. The
Qiu-ye Li-Yang Xiao,
instrument has a two-tier question design,
Lei Bao
which has led to multiple ways of scor-
ing and interpretation. In this research, a
method of pattern analysis was proposed
and applied to analyze two-tier item pairs
on the subskill of Control-of-Variable (COV)
of LCTSR. The data were collected from Introduction
students in grade 4 through college in both
In science, technology, engineering, and mathematics (STEM) education,
the United States and China. Students’ scientific reasoning has gained increasing attention recently from science
response patterns to two combined educators and researchers. In general, the scope of scientific reasoning en-
item pairs were analyzed and compared compasses various thinking and reasoning skills involved in inquiry, experi-
against students at different grade levels mentation, evidence evaluation, inference, and argumentation that support
and reasoning development levels. Six the formation and modification of concepts and theories about the natural
and social world (Zimmerman, 2007). It plays a crucial role in the development
performance levels were established based
of students’ creative thinking and problem -solving skills, which are necessary
on students’ response patterns, serving as when managing real-world situations in professions beyond the classroom
indicators of COV reasoning development (iSTAR Assessment, 2010; Zhou et al., 2016). In K-12 education, it has been
levels. With the new method, a relation was reported that effort in improving scientific reasoning skills has a profound
obtained between students’ level of COV impact on students’ academic performance (Adey & Shayer, 1990), and a
skills and grade level, as well as their level positive correlation has been reported between students’ scientific reason-
ing abilities and course achievement (Coletta & Phillips, 2005). Furthermore,
of overall reasoning development. It can
it has been shown that the level of students’ scientific reasoning skills is a
provide useful information on the possible better indicator of their success in biology learning, comparing with their
developmental levels of students’ reason- prior knowledge (Johnson & Lawson, 1998). These results prompt the need
ing skills. for students to develop a strong set of scientific reasoning skills, alongside a
Keywords: control-of-variable, grade level, solid foundation of content knowledge.
pattern analysis, scientific reasoning A number of studies have been conducted to determine the mechanism
of acquiring scientific reasoning skills, as well as effective teaching strategies
Shao-Na Zhou that aid students in learning, retaining, and transferring these skills (Dunbar
South China Normal University, China
Qiao-Yi Liu & Klahr, 2012; Joep, et al., 2019). In a review on scientific reasoning, Zimmer-
The Ohio State University, USA man claimed that investigation skills can interact with science knowledge
Kathleen Koenig to create a relationship that promotes scientific reasoning development
University of Cincinnati, USA
Qiu-ye Li-Yang Xiao (Zimmerman, 2007). It has also been reported that scientific reasoning re-
South China Normal University, China quires a complex set of cognitive skills, the development of which follows
Lei Bao a prolonged path, and students’ performance varies as they progress along
The Ohio State University, USA
this path (Zimmerman, 2007).
146
ISSN 1648–3898 /Print/ ANALYSIS OF TWO-TIER QUESTION SCORING METHODS: A CASE STUDY ON THE LAWSON’S
ISSN 2538–7138 /Online/ (pp. 146-159)
In order to quantitatively measure the development of students’ scientific reasoning skills, several assessment
instruments have been developed and employed by education researchers, such as the group assessment of logi-
cal thinking test (GALT), the test of logical thinking (TOLT), and the Lawson’s Classroom Test of Scientific Reason-
ing (LCTSR) (Lawson, 1978). The LCTSR is a popular assessment instrument that investigates students’ scientific
reasoning skills from primary school to university (Bao et al., 2009), and the validity of its current version has been
analyzed in detail (Xiao et al., 2018). Nevertheless, multiple scoring methods of the two-tier LCTSR test have been
proposed, and generated different interpretations (Xiao et al., 2018). More specifically, two-tier multiple-choice item
pairs with different difficulties are generally not distinguished. However, this dimension of difficulty can provide
insights regarding students’ scientific reasoning progress. This work details a cross-sectional study using data from
students in both the United States and China on the correlation between students’ answer patterns and their level
of scientific reasoning for different grade levels and difficulty levels, which further extends the application of LCTSR
on measuring the development of students’ scientific reasoning skills.
Literature Review on Lawson’s Classroom Test of Scientific Reasoning (LCTSR)
The development of LCTSR
Historically, the Piagetian tasks were considered to be a standard method of measuring students’ scientific
reasoning skills, which, however, are time-consuming and required experienced interviewers, special materials, and
equipment (Goldschmid, 1967; Lawson & Blake, 1976; Lawson et al., 1975). In 1978, Lawson designed an assessment
instrument that measures students’ level of scientific reasoning development, called the Lawson’s classroom test of
formal reasoning (CTFR-78). The paper and pencil style of the CTFR-78 addressed the need for a reliable, convenient
assessment tool that would be more practical for classroom use, compared to the Piagetian tasks.
A paper and pencil test, compared to clinical interview tasks, not only requires the ability to read and
write, but also provides little motivation for the test takers from the materials or equipment being used, since
it is not as personal or relaxed. Taking these challenges into consideration, Lawson aimed to strike a balance
between the convenience of paper and pencil tests and the interaction of interview tasks in CTFR-78. CTFR-78
involves an instructor performing a demonstration in front of a class, after which the instructor would pose
a question to the entire class, and the students would mark their answers in their test booklets. The booklets
contain the questions followed by several answer choices. For each of the test items, students had to choose
the correct answer and provide a reasonable explanation in order to receive credit for that item, forming the
two-tier test design. In order to establish the validity of the test, Lawson administered CTFR-78 to 513 students
from 8th through 10th grade, and selected 72 of them to participate in clinical interviews with Piagetian tasks
that reflected the three established levels of reasoning (concrete reasoning, transitional reasoning, and formal
reasoning) (Lawson, 1978). After comparing the test scores with students’ response to interview tasks, Lawson
found that the results from CTFR-78 and the clinical interviews had a good agreement, while CRFR-78 might
have a tendency to underestimate students’ scientific reasoning ability slightly. The validity of CTFR-78 was
further established by other researchers (Pratt & Hacker, 1984; Stefanich et al., 1983), with item analysis and
principle-components analysis.
In 2000, building on previous work, Lawson developed an improved version of the assessment instrument,
named Lawson’s Classroom Test of Scientific Reasoning (LCTSR). It is a two-tier, multiple-choice test with 24 items
(Lawson, 2000). A two-tier multiple-choice item pair contains a question with some possible answer choices, fol-
lowed by another question proving some possible reasons for the response to the previous question. All the answer
choices were designed based on previous studies on student misconceptions with free response tests, interviews,
and relevant literature (Treagust, 1995).
The scoring methods of LCTSR
In accordance with the LCTSR design, both questions in a two-tier, multiple-choice item pair must be correct
in order for the students to receive credit (Lawson, 2000). According to Lawson’s method, getting both questions
wrong in an item pair indicates the lowest level of scientific reasoning, and getting both correct indicates the high-
est, but the level is not distinguished when getting only one of the questions correct (only the answer, or only the
reasoning) (Lawson, 2000; Treagust, 1995).
147
(pp. 146-159) ISSN 2538–7138 /Online/
However, two latent traits are embedded in the two-tier multiple-choice item pairs: answering the question
is for knowing the result in the first tier, and choosing the reason is for explaining the reason in the second tier
(Tsai & Chou, 2002). A number of studies have shown that explaining the reason represents a higher skill than
answering the question correctly using the classical test theory (CTT) (Bayrak, 2013; Caleon & Subramaniam,
2009; 2010; Chang et al., 2007; Xiao et al., 2018). This implies that students may know the answer before they
have developed the capability of explaining the reason for the answer. Therefore, there exist intermediate levels
of understanding that correspond to different response patterns. For a two-tier, multiple-choice item pair, there
are four different response patterns of correct and incorrect answers. Among these, “00” represents getting both
answer and reasoning incorrect as the lowest level of student performance on the two-tier item pair, while “11”
represents getting both answer and reasoning correct as the highest level. For the two intermediate patterns,
“01” corresponds to incorrect answer with correct reasoning, which is often interpreted as guessing. On the
other hand, “10”, which corresponds to correct answer with incorrect reasoning may represent a higher level of
understanding than guessing.
In general, three scoring methods are commonly employed for evaluating two-tier multiple-choice items
in the previous studies: individual scoring method, pair scoring method, and partial credit scoring method.
Individual scoring method treats two questions in a two-tier item pair as individuals and assigns score for each
tier question independently (Chang et al., 2007; Chu et al., 2009). Pair scoring methods treats two questions in a
two-tier item pair as a combined entity, and assigns credit only for answering both questions correctly, zero point
for all other response patterns (Bayrak, 2013; Chandrasegaran et al., 2007; Lin, 2004). Following the assumption
that students may know the answer to a question before they can fully articulate the reasoning based on their
response, partial credit scoring method may assign 2 points for the pattern of “11”, 1 point for the pattern of
“10”, and 0 point for patterns of “00” and “01” (Xiao et al., 2018), or assign 3 points for the pattern of “11”, 2 point
for the pattern of “10”, 1 point for the pattern of “01” and 0 point for the pattern of “00” (Satriana et al., 2018;
Xiao et al., 2018). Xiao et al. (2018) pointed out that individual scoring method rewards the intermediate levels,
but it may also assign credit for guessing, while pair scoring method avoids guessing rewards, but ignores the
possible intermediate learning stages by underlining the relationship of knowing the answer and explaining
the reason. Rasch analysis was then used to explore different scoring methods on the data of LCTSR, and the
results confirmed that for the partial credit scoring method, the pattern “10” represents a higher level of scientific
reasoning than the pattern “01”, which should be treated as guessing.
Research Questions
Among the previous studies on analyzing two-tier, multiple-choice items, some were limited to analyzing
individual two-tier, multiple-choice item pair (Chang et al., 2007; Satriana et al., 2018; Xiao et al., 2018), while others
treated different two-tier, multiple-choice item pairs as independent combinations, assigning scores for individual
items pairs, and simply adding them together for a total score of all items (Lawson, 2000; Luo et al., 2020). However,
the possible inclusion of several item pairs that probe the same aspect of scientific reasoning yet with varying dif-
ficulties is ignored. Neither examining the item pairs individually nor simply adding all their scores fully explores
this dimension of item difficulty.
From the theory of cognitive development (Inhelder & Piaget, 1958; Piaget, 1971), the development of rea-
soning follows a process that evolves from simple to complex (Watson, 1975). Students are likely to understand
simple phenomena, concepts, and laws of the nature first before some of the more difficult ones that are in the
same domain. Similarly, students are likely to gain the ability to solve simple problems first before complicated
problems in the same domain. Therefore, students’ performance on item pairs of varying difficulty in the same
domain can provide important information on student’s scientific reasoning development.
Building on previous works, this research aimed to establish a method of pattern analysis for assessing two-
tier item pairs in LCTSR. The method of pattern analysis was chosen for the research, as it is a common technique
in data mining, where processes, algorithms, and mechanisms are investigated to retrieve potential knowledge
from data collections (Norton, 1999). In this case, pattern analysis is an intuitive and straightforward method
to investigate students’ performance on item pairs of different difficult levels. Through the pattern analysis
in student responses, the finer details of students’ scientific reasoning development levels can be probed by
examining a cross-section of all the student data with different grade levels. Specifically, this research aimed to
answer three research questions:
148
ISSN 2538–7138 /Online/ (pp. 146-159)
1. Do intermediate response patterns represent different reasoning levels in terms of grade levels?
2. Do intermediate response patterns represent different reasoning levels in terms of students’ overall
development?
3. Are combined response patterns to item pairs of different difficulty good indicators of reasoning de-
velopment levels?
The Difficulty Characterization of Two-tier Item Pairs in LCTSR
When LCTSR was first designed, the items typically fell into three levels of difficulty: concrete reasoning,
transitional reasoning, and formal reasoning (Lawson, 1978). For instance, for the test items designed to measure
student reasoning on the subskill of Control-of-Variable (COV), two item pairs have been shown to have signifi-
cant difference in difficulty (Lawson, 2000). These two item pairs are presented in Figure 1, and the individual
items are labeled as P1 (pendulum answer), P2 (pendulum reasoning), F1 (flies in a tube answer), and F2 (flies
in a tube reasoning).
The two responses for P1 and P2 are listed as the first pair, while those for F1 and F2 are listed as the second
pair. Each pair has an answer item and a reason item. The truth value of the answer is the first digit in each pair,
and that of the reasoning is the second digit. As a basic indicator of difficulty, the percent correct was calculated
from the performance among students from middle school (grade 6-7), high school (grade 9-10), and college, us-
ing individual score method for these four individual items, and the results are shown in Table 1. As can be seen
in Table 1, it is evident that students from all grade levels performed better on the first item pair (P1 and P2) than
the second (F1 and F2), thus it can be assumed that the first item pair (P1 and P2) is easier than the second (F1 and
F2). The present research focuses on students’ performance on these four items (P1, P2, F1, and F2) when analyzing
student reasoning on the subskill of Control-of-Variable (COV) of LCTSR.
Table 1
Percent Correct on the Two Item Pairs in LCTSR across Different Grade Levels
Percent correct (%)

Difficulty
Item Context
level
Grades 6-7 Grades 9-10 College
P1 Pendulum – answer 34 66 79
1st item pair Easy
P2 Pendulum – reasoning 29 57 78
F1 Flies – answer 16 29 50
2nd item pair Difficult
F2 Flies – reasoning 16 17 30
Sample
Since students’ scientific reasoning development levels were important to the present research, the partici-
pants were constituted as students in different grade levels. One part of the participants consisted of students
in grade 4-12 in both the United States and China, and the other part included college first-year students from a
large Midwestern university in the United States. The student responses from the United States and China were
not distinguished in the analysis, since it has been shown that scientific reasoning levels of the two sets of popula-
tion are effectively similar (Bao et al., 2009). Therefore, there were a total number of 10707 students enrolled into
the present research with the number distribution of participating students shown in Table 2. Specifically, in the
primary level, there were 336 students from grade four, 547 students from grade five and 588 students from grade
six. In the junior school level, there were 868 students from grade seven, 606 students from grade eight, and 1489
students from grade night. In the secondary school level, there were 1520 students from grade ten, 2083 students
from grade eleven, and 847 students from grade twelve. In addition, 1823 college first-year students participated
in the present research.
149
(pp. 146-159) ISSN 2538–7138 /Online/
Table 2
The Number Distribution of Participating Students
Grade Level 4 5 6 7 8 9 10 11 12 College
N 336 547 588 868 606 1489 1520 2083 847 1823
Figure 1
Two-Tier, Multiple-Choice Item Pairs from LCTSR Used in this Research
P1 F1
P2 F2
Notes: P1:pendulum answer, easy; P2: pendulum reasoning, easy; F1: flies in a tube answer, difficult; F2: flies in a tube reasoning,
difficult.
Data Collection
All students volunteered to take the LCTSR. For the administration of the test, students in the United States
used the English version, while those in China used the Chinese version. To ensure the consistency between the
two versions, the test was carefully translated and evaluated by a group of six faculty members who are proficient
in both languages. All students were given sufficient time to finish the test. Students from lower grade levels took
about 45 to 50 minutes for the test, while college students needed about 30 minutes. Prior to the formal imple-
mentation of the test, a sample test was conducted with both primary students and college freshmen to ensure
that the time provided was appropriate.
Data Analysis
This research focused on students’ performance on the four chosen items (P1, P2, F1, and F2) from LCTSR when
analyzing student reasoning on COV. In each item pair, students choose an answer to the question in the first tier,
and choose a reason of explanation in the second tier. Responses are coded using “0” for an incorrect answer and
“1” for a correct answer. Thus, the code 00-00 means all responses are incorrect, while a code of 11-11 means all
150
ISSN 2538–7138 /Online/ (pp. 146-159)
responses are correct. A code of 11-10 means that the student provides correct responses to P1, P2, and F1, but
an incorrect response to F2. Since each of the four items can be answered correctly or incorrectly, there are a total
of 16 possible response patterns: 00-00, 00-01, 01-00, 01-01, 00-10, 01-10, 00-11, 01-11, 10-00, 10-01, 11-00, 11-01,
10-10, 10-11, 11-10, and 11-11.
As previous research studies have suggested that explaining the reason represents a higher level of reasoning
than answering the question correctly (Bayrak, 2013; Caleon & Subramaniam, 2009; 2010; Chang et al., 2007; Xiao
et al., 2018), students are likely to have the ability to provide a correct answer before they can clearly articulate
correct reasoning. Therefore, students’ intermediate levels of scientific reasoning may be expressed within the
different response patterns on the four items chosen from LCTSR.
Another factor that can be examined to indicate students’ level of scientific reasoning development is the
item difficulty. Since P1 and P2 are considered to be easier than F1 and F2, students who can answer F1 and F2
correctly are considered to have a higher level of reasoning.
Based on the above assumptions, the 16 response patterns were analyzed and matched with different levels
of reasoning. According to Piaget’s cognitive development theory, children’s cognitive development can be divided
into four stages by age group (Inhelder & Piaget, 1958; Piaget, 1971): sensorimotor stage (ages 0-2), preoperational
stage (ages 2-7), concrete operational stage (ages 7-12), and formal operational stage (ages 12 to adult). Therefore,
a higher level of a particular subskill is more likely to be observed among students from higher grade levels and
higher overall scientific reasoning skill levels. This assumption serves as the basis of the student performance
analysis across different grade levels and overall scientific reasoning development levels.
Research Results
Intermediate Response Patterns Change with Students’ Grade Levels
Student response patterns of the four COV items from grade 4 through college were collected and organized.
Since the first item pair (P1 and P2) is easier than the second (F1 and F2), four intermediate response patterns are
compared first, which include 01-00, 10-00, 11-01, and 11-10. The percentage of these four patterns at different
grades are listed in Table 3.
Table 3
Student Performance on P1, P2, F1, and F2 from Grade 4 through College
Patterns
Grade Levels
01-00 10-00 11-01 11-10
(%) (%) (%) (%)
4 9.5 6.5 0.6 0.3

5 6.4 9.0 0.9 1.1
6 6.0 10.7 2.4 1.5
7 5.1 8.2 2.0 3.1
8 2.8 17.0 1.5 10.1
9 3.8 12.0 3.2 9.9
10 4.6 9.5 2.9 12.8
11 3.5 8.5 4.6 14.5
12 1.8 4.4 4.1 23.1
College 0.4 1.6 4.4 21.5
The results indicated that as the grade level increases, the percentage of student responses of “01-00” decreases
gradually. As for student response of “10-00”, the percentage starts rising from grade 4, peaks in grade 8, and then
falls gradually until college. From grade 4 to grade 8, the increase in the “10-00” pattern suggests an improving
cognitive development with grade levels. The decease after grade 8 is due to a larger portion of the students
151
(pp. 146-159) ISSN 2538–7138 /Online/
moving to higher developmental level (i.e. answering both P1 and P2 correctly), which is an indication of a major
learning shift, and a critical point in students’ scientific reasoning development.
In the case where the students answer the first pair correctly, the percentage of student responses of “11-01”
is relatively small and independent of grade levels. This is another evidence that “01” does not accurately reflect
students’ level of scientific reasoning development, and is likely a result of guessing. As for student response of “11-
10”, the percentage rises steadily as the grade level increases, from 0.3% in grade 4 to 21.5% in college. It indicates
“10” as some level of reasoning development, which is higher than “00” or “01”.
Putting all these together, this cross-sectional study of all students from grade 4 through college confirms
that student responses of “11-10” and “10-00” represent higher levels of scientific reasoning development than
“11-01” and “01-00”, respectively.
Intermediate Response Patterns Change with Students’ Overall Development
Student response patterns of the four COV items were analyzed based on their overall reasoning develop-
ment levels. In order to eliminate any variation due to the grade level, three subsets of the whole sample were
chosen based on the grade level, with at least two-grade-level difference between each group: grades 67, grades
9- 10, and college.
Within each group, students were divided among three different overall reasoning development levels based
on their total scores on the remaining 20 items of the LCTSR, excluding the four chosen COV items. These levels were
defined based on Lawson’s original research (Lawson, 1978), which included bottom 30% for concrete reasoning,
medium 40% for transitional reasoning, and top 30% for formal reasoning. In this research, these levels translated
approximately into score cutoffs (out of 20 points) at 0 to 11 for bottom 30%, 12 to 16 for the intermediate 40%,
and 17 to 20 for the top 30%. The final percentages and the number of students in each development level for
different populations are shown in Table 4.
Table 4
Responses to Lawson Test Items P1, P2, F1, and F2 from Grades 6-7, 9-10, and College
Patterns
Percentage of
ranking Score N
(%) 01-00 10-00 11-01 11-10
(%) (%) (%) (%)
Low 30 0-5 433 6.2 7.8 1.6 0.9
Grades 6-7 Mid 42 6-9 616 5.8 8.6 1.6 1.1
High 28 10-19 404 4.0 11.6 3.5 6.2
Low 33 0-8 991 7.8 11.3 2.4 3.1
Grades 9-10 Mid 40 9-13 1211 2.6 11.8 3.5 10.2
High 27 14-20 804 2.2 8.4 3.2 23.1
Low 29 0-11 523 1.1 3.4 4.2 9.2
College Mid 40 12-16 724 0.1 1.4 6.2 20.8
High 31 17-20 573 0.2 0.2 2.3 33.6

Notes: Low (%) means the bottom part of students; Mid (%) means the intermediate part of students; and High (%) means the
top part of students.
Similar to the previous analysis on grade levels, the intermediate response patterns on the two pairs of COV
questions were examined. The results are also included in Table 4, which indicate that the percentage of student
responses of “01-00” and “11-01” are relatively low, across all grade levels. Within each group, these percentages
are not significant, and independent of the overall reasoning development level. Both of these two observations
152
ISSN 2538–7138 /Online/ (pp. 146-159)
suggest that the “01” type response does not represent a meaningful level of reasoning development, and is likely
due to guessing.
On the other hand, the percentage of student responses of “10-00” and “11-10” are relatively high, across all
grade levels. Moreover, within each group, as the level of overall reasoning increases, the percentage of student
responses of “11-10” increases accordingly, contrary to that of “11-01”. This relation is more pronounced in grades
9-10 and college. These observations also well explain that the “10” response indicates a meaningful level of sci-
entific reasoning, while “01” response indicates guessing.
Putting all these together, the two pattern analysis on student responses of different grade levels and overall
reasoning development suggest that a correct answer with incorrect reasoning indicates an intermediate level
of reasoning development, whereas an incorrect answer with correct reasoning is likely a result of guessing. As a
result, among the 16 answer patterns to the two-tier item pairs of LCTSR, the pattern “10-00” and “11-10” represent
a higher level of scientific reasoning development that “01-00” and “11-01”, respectively.
Combined Response Patterns to Item Pairs of Different Difficulty as Good Indicators of

Reasoning Development Levels
As discussed previously, the response patterns of “01” and “10” represent different levels of development. Ac-
cordingly, students’ development in COV can be divided into different levels based on their responses to the four
chosen COV items. The pattern “00-00” represents the lowest level, whereas “11-11” represents the highest. The
intermediate levels are ordered based on the rules shown below:
1. Students are able to provide a correct answer before they can provide the correct reasoning for the
same item;
2. Students are able to answer the easy items correctly before they can answer the difficult ones correctly;
3. The “01” response (correct answer, incorrect reasoning) is likely a result of guessing.
Following these considerations, all 16 possible response patterns are grouped into six levels.
Level 1 (00-00) includes students giving all incorrect answers, and is the lowest level of reasoning development.
Level 2 (0x-xx) includes students giving an incorrect response to P1, which expands into 7 patterns: “00-01”,
“01-00”, “01-01”, “00-10”, “01-10”, “00-11”, and “01-11”. If a student cannot answer P1 correctly, according to the rules,
it is likely that any other correct answers from other items are due to guessing. This level also includes the students
giving incorrect responses to both P1 and P2 incorrect but correct one(s) to either or both of F1 and F2. Since the
first item pair (P1 and P2) is easier than the second (F1 and F2), these responses are also considered as the results
of guessing. Although level 2 is mostly comprised of guessing responses, it is still considered as a level higher than
level 1, because a correctly guessed response may indicate some level of reasoning, however little or implicit. For
instance, a student can eliminate some answer choices, and make an informed guess. On the other hand, the “00”
response could indicate a misconception, which should be separated from guessing.
Level 3 (10-0x) includes students giving correct responses to P1, incorrect to P2, and incorrect to F1: namely,
“10-00” and “10-01”. The “10-01” response is included in this level, as the correct response of F2 is likely due to
guessing. This level is considered higher than level 2 because the response to P1 is correct.
Level 4 (11-0x, 10-1x) includes students giving correct responses to both P1 and P2 or P1 and F1: “11-00”, “11-
01”, “10-10” and “10-11”. These responses are grouped together, as it is unclear which of these responses indicates
a higher level of reasoning. The “10-11” response is included in this level because there is a possibility that students
miss P2 while still fully understand F1 and F2. Also, it is unlikely that a student would guess both F1 and F2 correctly.
Level 5 (11-10) includes students giving correct responses to P1, P2, and F1, and incorrect responses to F2:
“11-10”. It means that students fully understand the easier item pair (P1 and P2), and are in midway to understand
the more difficult item pair.
Level 6 (11-11) includes students giving correct responses to all items, which represents the highest level of
development.
153
(pp. 146-159) ISSN 2538–7138 /Online/
Table 5
Percentage of Response Patterns from Students in Different Developmental Stages of Scientific Reasoning
Stages of reasoning Response Patterns

development
1 2 3 4 5 6
Overall Row Sum

00-00 0x-xx 10-0x 11-0x, 10-1x 11-10 11-11 (%)
Stages Score
(%) (%) (%) (%) (%) (%)
(%)
1 0.0 46.5 38.2 8.7 6.0 0.5 0.2 100.0
2 10.0 42.6 30.0 11.7 13.5 1.5 0.6 100.0
3 20.0 36.4 23.1 11.9 24.2 3.0 1.4 100.0
4 30.0 29.5 16.2 12.5 35.4 4.6 1.8 100.0
5 40.0 20.8 13.3 11.5 41.1 8.5 4.8 100.0
6 50.0 14.0 12.8 10.5 43.0 11.9 7.9 100.0
7 60.0 9.5 8.0 9.7 46.5 15.4 10.8 100.0
8 70.0 3.9 5.9 5.7 41.3 24.7 18.5 100.0
9 80.0 1.5 5.2 3.2 33.0 35.1 22.0 100.0
10 90.0 1.9 4.8 1.7 26.6 31.2 33.8 100.0
11 100.0 1.0 5.4 3.4 11.3 35.3 43.6 100.0
In order to explore the development of COV skills, all students were divided into 11 developmental stages
according to the percentage of correct answers of the remaining 20 items in LCTSR. For each of the developmental
stages, the percentages of students responding with the different patterns are calculated and given in Table 5. The
changes of the patterns across different developmental stages are plotted in Figure 2.
The results show that as expected level 1 patterns start with high percentages in the low-end of reasoning
development and decrease steadily as the stage of reasoning development increases. Similarly, the level 2 patterns
also decrease as the reasoning development stage increases. This is also expected since the level 2 responses repre-
sent mostly the results of guessing, which should decrease as students’ reasoning skills develop into higher stages.
The level 3 patterns are relatively steady for students from the low-end through the middle of the reason-
ing development stages. These students usually can respond to P1 (the first answer) correctly but have incorrect
answers to P2 (the reasoning), confirming that answer often precedes reasoning. On the other hand, students in
higher reasoning stages have less level 3 patterns, which suggests that students at higher reasoning stages have
developed both “knowing” and “reasoning”.
The level 4 patterns first increase then decrease with the developmental stage. The results show that more
and more students are able to answer P1 and P2 correctly as their reasoning kills develop, which accounts for the
initial increase. Meanwhile students at higher overall reasoning stages begin to answer F1 and F2 correctly, which
accounts for the decrease.
The level 5 patterns also increase with the reasoning development stage and somewhat plateaued at the
highest stages. Meanwhile the level 6 patterns increase steadily with the developmental stage. The increases of
both patterns are more dramatic after stage 7, which suggest that a thorough understanding of COV skills are
achieved when students develop into the formal reason stage (top 30%).
154
ISSN 2538–7138 /Online/ (pp. 146-159)
Figure 2
Percentage of Students From Low-End to High-End Overall Scientific Reasoning Abilities at the Six Performance Levels On P1,
P2, F1, And F2 Of LCTSR
Low High
The results in Figure 2 illustrate the trends of how the different response patterns change with students’
stages of overall reasoning development. Among the six levels of response patterns, the most interesting one is
level 4, which shows a clear concave down shape with a peak at around 60% of the overall reasoning score. The
remaining five levels are mostly monotonically increasing or decreasing. This suggests that level 4 patterns can be
an important transitional stage of learning that students may first develop into and then evolve beyond. Therefore,
assessment of the level 4 patterns may provide useful indicators of substantive cognitive transitions during the
progression of learning.
Figure 3
Distribution of the Six Levels of Response Patterns across Different Grades
The same data can also be analyzed in terms of grade level. Student responses are combined into 5 groups
based on their grade levels: grades 4-5, 6-7, 8-10, 11-12, and college. The percentages of different levels of re-
sponse patterns for different grade groups are plotted in Figure 3, which shows that the higher level of response
patterns are more popular among advanced students. These results are consistent with those presented in
155
(pp. 146-159) ISSN 2538–7138 /Online/
Table 5 and Figure 2. In particular, the level 4 patterns show a more pronounced peak among middle school
to college students (grades 8-12 and college), which indicates that many of students in these grades are in the
transitional stage of developing COV skills. This result also corroborates with previous analysis showing that
the level 4 patterns can be indicators of significant learning transitions.
Discussion
In this research, a method of combined pattern analysis was proposed for evaluating two-tier item pairs
on students’ COV skills in LCTSR. To be able to measure students’ COV skills in finer detail, combined patterns of
two item pairs responses were analyzed to show how they correspond to students both in different grade levels
and in various stages of overall reasoning development. A few key results were uncovered in this analysis. The
essential goal of the research was to identify if getting just the answer or just the reasoning correct indicated
different skill levels, and which represented a higher skill level. On the one hand, the research of how inter-
mediate response patterns represent different reasoning levels in terms of grade levels supports that the “10”
response indicates a higher level of scientific reasoning development than the “01” response by cross-sectional
analysis of data from grade 4 through college. The result that the percentage of student responses of “01-00”
decreases gradually as the grade level increases agrees with Piaget’s cognitive development theory, inferring
that a higher level on a specific subskill development is more likely to be observed among students from
higher grade levels (Inhelder & Piaget, 1958; Piaget, 1971). On the other hand, detailed analysis of data from
students in different overall reasoning development levels also suggests that a correct answer with incorrect
reasoning indicates an intermediate level of reasoning development, whereas an incorrect answer with correct
reasoning does not represent a meaningful level of reasoning development and is likely a result of guessing.
These results support previous research studies that students are able to provide the correct answer before
they can provide the correct reasoning (Bayrak, 2013; Caleon & Subramaniam, 2009; 2010; Chang et al., 2007;
Xiao et al., 2018). This is a result that is often overlooked by traditional two-tier scoring methods, in which the
“01” and “10” responses are not explicitly distinguished. The traditional scoring of the Lawson Test allows for
only two levels of performance, both the answer and reasoning need to be correct or no credit is given. It has
been proved in the present research that the traditional scoring method does not accurately reflect the pos-
sible levels of student understanding. Students who get just the answer correct should be at a higher level
of understanding than those who get just the reasoning correct or who get both incorrect. It is believed that
students’ skill levels should be identified at a finer grain size. This leads to a step-function in the scoring of a
particular individual question or two-tier question.
Based on the previous work, sixteen combined response patterns to the two-tier item pairs of LCTSR were
ordered and six performance levels were established based on some proven rules. Aside from Piaget’s cogni-
tive development theory that students are able to answer the easy items correctly before they can answer the
difficult ones correctly (Inhelder & Piaget, 1958; Piaget, 1971), the rules also included that students are able to
provide a correct answer before they can provide the correct reasoning, and an incorrect answer with correct
reasoning is likely a result of guessing. From the results, student performance resembles a developmental
process from low to high skill levels, with level 1 in the low-end and level 6 in the high-end of reasoning devel-
opment. With the method of pattern analysis, a relation was obtained between students’ COV skill, grade level,
and overall reasoning development. It can be found how the trends of the different response patterns change
with students’ grade levels and their stages of overall reasoning development. Particularly, level 4 patterns can
be an important transitional stage of learning that students may first develop into and then evolve beyond.
The level 4 patterns with a pronounced peak indicate that many of students among middle school to college
students are in the transitional stage of developing COV skills. It suggests that the level 4 patterns provide use-
ful indicators of significant learning transitions. As the data have shown, reasoning skills develop slowly, and
there is an intermediate level that traditional scoring does not recognize. The present result is highly valuable,
indicating that pattern analysis could reflect students’ ability more accurately than the traditional scoring of
two-tier questions, contribute to better data analysis, and provide with a simple way to track learning progress.
The research strives to strike a balance between the complexity of the data analysis and the simplicity
of grading a multiple-choice assessment instrument. Through the above efforts, students can perceive that
they still retain some valuable thinking even though they may not solve problems correctly in all aspects. This
156
ISSN 2538–7138 /Online/ (pp. 146-159)
would provide motivation and a sense of achievement for students in the process of developing cognitive
progression. The results also enable teachers and students to intuitively realize the unique meaning between
“knowing” and “reasoning”. As in Moraes et al. (2020)’ proposal of using Problem Based Learning (PBL) method
to cultivate students’ scientific literacy and science concepts, the teachers first consider how students arrive at
the conclusions to the questions and then how the systemization of learning can be achieved. Even students’
scientific reasoning were demonstrated to be improved at high, moderate or low levels for different dimensions
(Erlina et al., 2018), pattern analysis could be served as a technique to deeply excavate student performance
of the intermediate level difference for each dimension.
Conclusions
The present research first conducted a cross-sectional analysis of data from grade 4 through college in
both the United States and China to explore how intermediate response patterns represent different reasoning
levels in terms of both students’ grade levels and overall reasoning development levels. The results showed that
a correct answer with incorrect reasoning indicates an intermediate level of reasoning development, whereas
an incorrect answer with correct reasoning is likely a result of guessing. Based on the combined response pat-
terns, six performance levels for the two-tier item pairs were established to reflect students’ developmental
process from low to high skill levels. In particular, the level 4 patterns provide useful indicators of substantial
cognitive transitions during the learning progression.
The pattern analysis method provides a new way to exploit the information embedded within the combined
response patterns of two two-tier item pairs of different difficulties. The analysis outcomes are consistent with
Piaget’s cognitive development theory that a higher level of a particular reasoning skill development is more
likely to be observed among students from higher grade levels and higher overall scientific reasoning develop-
ment levels. The pattern analysis is an easily accessible method and provides a straightforward interpretation
of the data. Therefore, this approach can provide a supplemental means to standard statistical tools, which
can reveal useful information to enrich the interpretation of assessment data regarding students’ learning and
development. Future work with this pattern analysis method can specialize in additional item pairs in LCTSR,
as well as other scientific reasoning instruments for their potential power in defining the scientific reasoning
learning progression.
Acknowledgements
The research is supported in part by 2020 Guangzhou Philosophy and Social Science Planning Fund of P.R.
China under the Grant No. 2020GZQN20, and by the National Science Foundation Award DUE-1712238. Any
opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and
do not necessarily reflect the views of the funding agencies.
References
Adey, P., & Shayer, M. (1990). Accelerating the development of formal thinking in middle and high school students. Journal
of Research in Science Teaching, 27(3), 267-285. https://doi.org/10.1002/tea.3660270309
Bao, L., Cai, T., Koenig, K., Fang, K., Han, J., Wang, J., Liu, Q., Ding, L., Cui, L., Luo, Y., Wang, Y., Li, L., & Wu, N. (2009). Learning
and scientific reasoning. Science, 323(5914), 586-587. https://doi.org/10.1126/science.1167740
Bayrak, B. K. (2013). Using two-tier test to identify primary students’ conceptual understanding and alternative conceptions
in acid base. Mevlana International Journal of Education, 3(2), 19-26.
Caleon, I. S., & Subramaniam, R. (2009). Do students know what they know and what they don’t know? Using a four-tier
diagnostic test to assess the nature of students’ alternative conceptions. Research in Science Education, 40(3), 313-337.
https://doi.org/10.1007/s11165-009-9122-4
Caleon, I. S., & Subramaniam, R. (2010). Development and application of a three-tier diagnostic test to assess
secondary students’ understanding of waves. International Journal of Science Education, 32(7), 939-961.
https://doi.org/10.1080/09500690902890130
Chandrasegaran, A. L., Treagust, D. F., & Mocerino, M. (2007). The development of a two-tier multiple-choice diagnostic
instrument for evaluating secondary school students’ ability to describe and explain chemical reactions using multiple
157
(pp. 146-159) ISSN 2538–7138 /Online/
levels of representation. Chemistry Education Research & Practice, 8(3), 293-307. https://doi.org/10.1039/B7RP90006F
Chang, H.-P., Chen, J.-Y., Guo, C.-J., Chen, C.-C., Chang, C.-Y., & Lin, S.-H., et al. (2007). Investigating primary and secondary
students’ learning of physics concepts in Taiwan. International Journal of Science Education, 29(4), 465-482.
https://doi.org/10.1080/09500690601073210
Chu, H.-E., Treagust, D. F., & Chandrasegaran, A.L. (2009). A stratified study of students’ understanding of basic optics concepts
in different contexts using two-tier multiple-choice items. Research in Science & Technological Education, 27(3), 253-265.
https://doi.org/10.1080/02635140903162553
Coletta, V. P., & Phillips, J. A. (2005). Interpreting FCI scores: Normalized gain, reinstruction scores, and scientific reasoning
ability. American Journal of Physics, 73(12), 1172-1179. https://doi.org/10.1119/1.2117109
Dunbar, K., & Klahr, D. (2012). Scientific thinking and reasoning. In K. Holyoak & R. Morrison (Eds.), The Oxford handbook of
thinking and reasoning (pp. 701–718). Oxford University Press.
Erlina, N., Susantini, E., Wasis, W., & Pandiangan, P. (2018). The Effectiveness of evidence-based reasoning in inquiry-
based physics teaching to increase students’ scientific reasoning. Journal of Baltic Science Education, 17(6), 972-985.
Goldschmid, M. L. (1967). Different types of conservation and non-conservation and their relation to age, sex, IQ, MA, and
vocabulary. Child Development, 38(4), 1229-1246. https://doi.org/10.2307/1127120
Inhelder, B., & Piaget. J. (1958). The grow of logical thinking. Basic Books.
iSTAR Assessment. (2010). iSTAR Assessment: Inquiry for scientific thinking and reasoning. http://www.istarassessment.org/
Joep, V. D. G., Eva, V. D. S., Gijsel, M., & Segers, E. (2019). A combined approach to strengthen children’s scientific thinking:
Direct instruction on scientific reasoning and training of teacher’s verbal support. International Journal of Science
Johnson, M. A., & Lawson, A. E. (1998). What are the relative effects of reasoning ability and prior knowledge on
biology achievement in expository and inquiry classes? Journal of Research in Science Teaching, 35(1), 89-103.
https://doi.org/10.1002/(SICI)1098-2736(199801)35:13.0.CO;2-J
Lawson, A. E. (1978). The development and validation of a classroom test of formal reasoning. Journal of Research in Science
Teaching, 15(1), 11-24.
Lawson, A. E., (2000). Classroom test of scientific reasoning: Multiple choice version. Based on a. E. Lawson, “development
and validation of the classroom test of formal reasoning”. Journal of Research in Science Teaching, 5(1), 11–24.
Lawson, A. E., & Blake, A. J. D. (1976). The factor structure of some Piagetian tasks. Journal of Research in Science Teaching,
13(5), 461-466.
Lawson, A. E., Nordland, F. H., & Kahle, J. B. (1975). Levels of intellectual development and reading ability in disadvantaged
students and the teaching of science. Science Education, 59(1), 113-126.
Lin, S. W. (2004). Development and application of a two-tier diagnostic test for high school students’ understanding of
flowering plant growth and development. International Journal of Science and Mathematics Education, 2(2), 175-199.
https://doi.org/10.1007/s10763-004-6484-y
Luo, M., Wang, Z., Sun, D., Wan, Z., & Zhu, L. (2020). Evaluating scientific reasoning ability: The design and validation of an
assessment with a focus on reasoning and the use of evidence. Journal of Baltic Science Education, 19(2), 261-275.
Moraes, J. V., Castellar, S. M. V., Castellar, S. V., & Castellar, S. V. (2010). Scientific literacy, problem- based learning and
citizenship: A suggestion for geography studies teaching. Problems of Education in the 21st Century, 19, 119-127.
http://www.scientiasocialis.lt/pec/node/364
Norton, M. J. (1999). Knowledge discovery in databases. Library Trends, 48(1), 9-21.
Piaget, J. (1971). The theory of stages in cognitive development. In D. R. Green, M. P. Ford, & G. B. Flamer (Eds.), Measurement
and Piaget. McGraw-Hill.
Pratt, C., & Hacker, R. G. (1984). Is Lawson’s Classroom Test of Formal Reasoning Valid? Educational and Psychological
Measurement, 44(2), 441-448. https://doi.org/10.1177/0013164484442025
Satriana, T., Yamtinah, S., Ashadi, & Indriyanti, N. Y. (2018). Student’s profile of misconception in chemical equilibrium. Journal
of Physics: Conference Series, 1097, 012066. https://doi.org/10.1088/1742-6596/1097/1/012066
Stefanich, G. P., Unruh, R. D., Perry, B., & Phillips, G. (1983). Convergent validity of group tests of cognitive development.
Journal of Research in Science Teaching, 20(6), 557-563.
Treagust, D. F. (1995). Diagnostic assessment of students’ science knowledge. In S. M. Glynn & R. Duit (Eds.), Learning science
in the schools: Research reforming practice (pp. 327-346). Lawrence Erlbaum Associates.
Tsai, C. C., & Chou, C. (2002). Diagnosing students’ alternative conceptions in science. Journal of Computer Assisted Learning,
18(2), 157-165.
Watson, M. S. (1975). A developmental study of empathy: Egocentrism to sociocentrism or simple to complex reasoning.
Cognitive Development, 15, 1-14. https://files.eric.ed.gov/fulltext/ED114179.pdf
Xiao, Y., Han, J., Koenig, K., Xiong, J., & Bao. L. (2018). Multilevel Rasch modeling of two-tier multiple-choice test: A case
study using Lawson’s classroom test of scientific reasoning. Physical Review Physics Education Research, 14, 020104.
https://doi.org/10.1103/PhysRevPhysEducRes.14.020104
158
ISSN 2538–7138 /Online/ (pp. 146-159)
Zhou, S., Han, J., Koenig, K., Raplinger, A., Pi, Y., Li, D., Fu, Z., & Bao, L. (2016). Assessment of scientific reasoning: The effects of
task context, data, and design on student reasoning in control of variables. Thinking Skills and Creativity, 19, 175-187.
https://doi.org/10.1016/j.tsc.2015.11.004
Zimmerman, C. (2007). The development of scientific thinking skills in elementary and middle school. Developmental Review,
27(2), 172-223. https://doi.org/10.1016/j.dr.2006.12.001
Cite as: Zhou, S.-N., Liu, Q.-Y., Koenig, K., Li, Q.-Y., Xiao, Y., & Bao, L. (2021). Analysis of two-tier question scoring methods:
A case study on the Lawson’s classroom test of scientific reasoning. Journal of Baltic Science Education, 20(1), 146-159.
Shao-Na Zhou PhD, Associate Professor, School of Physics and Telecommunication

Engineering, South China Normal University, Guangzhou 510006,
China.
Qiao-Yi Liu PhD, Department of Physics, The Ohio State University, Columbus,
OH 43210, USA.
mail: [email protected]
Kathleen Koenig Associate Professor, Department of Physics, University of Cincinnati,
Cincinnati, OH 45221, USA.
Qiu-ye Li MSc, School of Physics and Telecommunication Engineering, South
China Normal University, Guangzhou 510006, China.
Yang Xiao PhD, School of Physics and Telecommunication Engineering, South
China Normal University, Guangzhou 510006, China.
mail: [email protected]
Lei Bao PhD, Professor, Department of Physics, The Ohio State University,
(Corresponding author) Columbus, OH 43210, USA.
159
INFORMATION
FOR CONTRIBUTORS
EDITORIAL AND REVIEW PROCEDURES
Journal of Baltic Science Education (JBSE) publishes original scientific research articles in the field of Natural Science Education and
related areas for all educational levels in the Baltic and other countries. It is possible to publish special (thematic) issues of JBSE. The
papers should be submitted and will be published in English. JBSE will promote to establish contacts between researchers and practical
educators both in the Baltic countries and countries around.
The authors of the manuscripts are responsible for the scientific content and novelty of the research materials. Articles, published
before in other international journals or papers’ collections will not be accepted for publication in JBSE.
As a publication that represents a variety of cross-disciplinary interests, both theoretical and practical, the JBSE invites manuscripts on
a wide range of topics, especially in the following areas:
• Didactics of natural sciences. • Philosophical, political, economical and social aspects
• Theory and practice in natural science teacher of natural science education.
education. • The supplementary natural science education.
• Integrated natural science education. • ICT in natural science education.
• Natural science and technological literacy. • The standardisation of natural science education etc.
• General and professional natural science education.
MANUSCRIPTS GUIDELINES
The structure of the research paper presented to the Journal of Baltic Science Education should be as follows: abstract - short report of the
investigation; introduction inc. aim and subject of the research; research methodologies and methods; results of the research incl. discussion;
conclusions; list of references in APA style.
The papers should be submitted in English. If English is a second language for the author, please consider having the manuscript
proof read and edited before submitting. The preliminary text of the article can be sent as a.doc file in the attachment by e-mail: mail.
[email protected]
The text must be elaborated in Word for Windows, using 12 point Times New Roman letters. An article should not exceed 7-10 A4 pages,
included figures, tables and bibliography. Publishing of longer articles should be negotiated separately. Texts margins: top and bottom 20mm,
left - 25mm, right - 20mm. The title: capital letters, 14pt, bold; space between the title and the author’s name is one line interval. Author’s name
and surname: small letters, 12pt, bold. Under the name, institution: 11 pt, italics; space between the title and the text: 1 line interval. Abstract
– about 100-150 words - precedes the text. The text: 12pt Single or Auto spacing, in one column. Key words: no more than five words. The
language must be clear and accurate. The authors have to present the results, propositions and conclusions in a form that can suit scientists
from different countries.
Titles of the tables and figures: 11 pt, small letters. Space between figures or tables and the text: 1 line interval. Introduction, titles of
chapters and subchapters: 12pt, bold, small letters. Numbers: Arabic, subchapters numbered by two figures (1.1, 1.2, etc.). Figures, tables and
captions should be inserted within the manuscript at their appropriate locations. Diagrams and graphs should be provided as finished black
and white line artwork or electronic images. When there are a number of illustrations, the author should endeavour to reduce the amount of
text to accommodate the illustrations in the limited space available for any article.
References in the text should be presented in brackets (Knox, 1988; Martin, 1995). If necessary, the page can be indicated: (Martin,
1995, p.48). The list of references should be presented after the text. The Words List of References: 11pt, bold, small letters. The references
should be listed in full at the end of the paper in the following standard form:
For books: Saxe, G.B. (1991). Cultural and Cognitive Development: Studies in Mathematical Understanding. Hillsdale, NJ: Erlbaum.
For articles: Bekerian, D.A. (1993). In Search of the Typical Eyewitness. American Psychologist, 48, 574-576.
For chapters within books: Bjork, R.A. (1989). Retrieval Inhibition as an Adaptive Mechanism in Human Memory. In: H.L. Roediger III & F.I.M.
Craik (Eds.), Varieties of Memory & Consciousness (pp. 309-330). Hillsdale, NJ: Erlbaum.
On a separate page, author - related data should be presented in English: name, surname, degree and academic title, institution, full
correspondence address in the clearest and most complete form /ordinary post and e-mail addresses /, position (to ensure anonymity in the
review process). The author (authors) should confirm in writing, that the manuscript has not been published in other journal or
handed over (transferred) to other journal for publication.
EDITORIAL AND REVIEW PROCEDURES

Manuscripts will be sent anonymously to reviewers with expertise in the appropriate area. All manuscripts will be rewieved by two
experts before JBSE’s accept them for publication. This process usually takes about two months. The journal co-editors will make minor
editorial changes; major changes will be made by the author(s) prior to publication if necessary. JBSE’s redaction will sent to author(s)
only one correcture which must be sent back within 2 weeks. JBSE will not review submissions previously published elsewhere through
print or electronic medium.
Manuscripts submitted to the JBSE cannot be returned to authors. Authors should be sure to keep a copy for themselves. Authors’
signatures should be at the end of the paper and its second checked proofs.
Manuscripts, editorial correspondence (and other correspondence for subscription and exchange), and any questions should be
sent to editor-in-chief or to regional redactors.
Journal`s requirements for the authors are available online: http://www.scientiasocialis.lt/jbse/files/JBSE_requirements_2019.pdf
160
JBSE
SMC „SCIENTIA EDUCOLOGICA“
The Associated Member of Lithuanian Scientific Society, European Society for the History of Science
(ESHS) and International Council of Association for Science Education /ICASE/,
SCIENTIA SOCIALIS Ltd
A member of the Lithuanian Publishers Association & Crossref
E-mail: [email protected] Website: http://www.scientiasocialis.lt
Dear colleagues,
We would like to invite articles for an international blind peer-review
scientific journal (Twenty third CFP)
Problems of Psychology in the 21st Century

ISSN 2029-8587 (Print) ISSN 2538-7197 (Online)
http://www.scientiasocialis.lt/ppc/ 2021: Vol. 15, No. 1; No. 2
Papers submitted to PPC_23CFP_2021 should be original work and

substantively different from papers that have been previously published or are
under review in a journal or another peer-reviewed conference. We invite
submission of papers describing innovative research on all aspects of psychology
and related areas. Submitted papers will be assessed based on their novelty,
technical quality, potential impact, and clarity of writing.
Responsible
E-mail: [email protected] Publication
Peer Reviewed
The call is open till 25 May 2021. Open Access
Global Readership
Application form + declaration + full paper: 25 May 2021
EBSCO http://www.scientiasocialis.lt/ppc/files/PPC_guidelines_2018.pdf
INDEX We advise you to use a template for paper preparation. You can download it
COPERNICUS from our website: http://www.scientiasocialis.lt/ppc/node/6 (template for paper,
ANVUR Word format). Theoretical papers usually follow an argumentative pattern and
CNKI are organised around the solution of a problem.
OAJI
Ulrich's Periodicals
Last but not least
Directory
MIAR
International journal „Problems of Pscyhology in the 21st Century“ is a high quality
ROAD, scientific journal which publishes original research on all areas of psychological
Europub, sciences. The Journal welcomes the submission of manuscripts that meet the general
DOI, criteria of significance and scientific excellence. Submitted papers will be assessed
Crossref, based on their novelty, scientific and technical quality, potential impact, and clarity of
Scilit etc. writing. All submissions will be checked by iThenticate before sending to reviewers.
The publisher and journal have a policy of “Zero Tolerance on the Plagiarism”.
PPC is an open access academic journal published semiannually.
Information related to APC can be found at:

http://www.scientiasocialis.lt/ppc/node/node/11
Sincerely yours, Editorial Board
161
GAMTAMOKSLINIS UGDYMAS / NATURAL SCIENCE

JBSE
EDUCATION
ISSN 1648-939X /Print/, ISSN 2669-1140 /Online/
Dear colleagues,
GAMTAMOKSLINIS UGDYMAS / NATURAL SCIENCE EDUCATION – is a
periodical, peer reviewed, scientific-methodical journal, issued by the SMC „Scientia
Educologica“ in cooperation with Scientia Socialis Ltd. It is an international journal,
wherein the scientific and methodical/applied articles published in Lithuanian, English and
Russian languages. This journal is intended for the teachers of general education schools,
the lecturers of higher educational institutions and all, who are interested in the problems
of natural science education.
The GU/NSE journal welcomes the submission of manuscripts that meet the general
criteria of scientific and methodical (practical/applied) papers.
GAMTAMOKSLINIS UGDYMAS / NATURAL SCIENCE EDUCATION

ISSN 1648-939X /Print/, ISSN 2669-1140 /Online/
http://gu.puslapiai.lt/GUwww/indeks.htm
http://gu.puslapiai.lt/GUwww/Indeksavimas.htm
http://oaji.net/journal-detail.html?number=514
Responsible This journal is abstracted / listed / indexed / cited in:
COPERNICUS INDEX , LIST OF SCIENCE EDUCATION
Publication JOURNALS, JOURNALS OF INTEREST TO CHEMICAL
Peer Reviewed EDUCATORS, SKYLIGHT, OAJI, WebQualis (QUALIS/CAPES ),
MIAR, EuroPub, QOAM, ESJI, Crossref.
Open Access
Applied research/practical/methodical work. This type of submission is best suited for
practical/didactical work and reports, as well as position papers raising original and
provocative theoretical or practical discourses and questions (small-scale research, applied
research, didactical/methodical papers, case studies, best educational practices etc.). Each
submission is carefully reviewed by two independent reviewers and ranked based on:
quality of preparation, relevance to the educational community, didactical quality,
originality, and importance of the contribution.
Instruction for authors and other details are available on the journal`s website at:
http://gu.puslapiai.lt/GUwww/Bendra_informacija.htm
GU/NSE is an Open Access journal accessible for free on the Internet. Papers must be
submitted on the understanding that they have not been published elsewhere and are not
currently under consideration by another publisher. Optimal paper`s size: 8/12 pages.
Partial article processing charges are: 5-8 EUR per one A4 page.
For contacts, questions and papers submission: [email protected]
Sincerely yours, Editorial Board
162
JBSE
ŠVIETIMAS: POLITIKA, VADYBA, KOKYBĖ
Education Policy, Management and Quality
ISSN 2029-1922 /Print/, ISSN 2669-1159 /Online/
Dear colleagues,
EDUCATION POLICY, MANAGEMENT AND QUALITY – is a periodical, peer

reviewed, scientific-methodical journal, issued by the SMC „Scientia Educologica“ in
cooperation with Scientia Socialis Ltd. It is an international journal, wherein the scientific
and methodical/applied articles published in Lithuanian, English and Russian languages.
The EPMQ journal welcomes the submission of manuscripts that meet the general criteria
of scientific and methodical (practical/applied) papers.
“Education Policy, Management and Quality”

http://gu.puslapiai.lt/SVIETIMAS_PVK_en.htm
http://gu.puslapiai.lt/SVIETIMAS_PVK_indeksavimas_eng.htm
http://oaji.net/journal-detail.html?number=513
This journal is indexed / abstracted / listed / cited / in:

Responsible
INDEX COPERNICUS, OAJI, MIAR, ESJI, Ulrich’s
Publication Periodicals Directory, QUALIS/CAPES, Academic
Peer Reviewed Resource Index, EuroPub, WCOSJ, QOAM, Crossref.
Open Access
This type of submission is best suited for practical/didactical work and reports, as well as
position papers raising original and provocative theoretical or practical discourses and
questions (small-scale research, applied research, didactical/methodical papers, case
studies, best educational practices etc.). Each submission is carefully reviewed by two
independent reviewers and ranked based on: quality of preparation, relevance to the
educational community, didactical quality, originality, and importance of the contribution.
Instruction for authors and other details are available on the journal`s website at:
http://gu.puslapiai.lt/Informacija-autoriams_Info-authors.pdf
EPMQ is an Open Access journal accessible for free on the Internet. Papers must be
submitted on the understanding that they have not been published elsewhere and are not
currently under consideration by another publisher. Optimal paper`s size: 8/12 pages.
Partial article processing charges are: 5-8 EUR per one A4 page.
For contacts, questions and papers submission: [email protected]
Sincerely yours,
Editorial Board
163
Compiled by: Vincentas Lamanauskas

Linguistic Editor: Ilona Ratkevičienė
Cover design by: Jurgina Jankauskienė
Layout design by : Linas Janonis

15 February 2021. Publishing in Quires 10,5. Edition 90
Publisher Scientia Socialis Ltd.,

Donelaicio Street 29, LT-78115 Siauliai, Lithuania
Phone: +370 687 95668
http://www.scientiasocialis.lt

Printing Šiauliai printing house
9A P. Lukšio Street
LT-76207 Šiauliai, Lithuania
Phone: +370 41 500 333.
Fax: +370 41 500 336
https://siauliuspaustuve.lt/

Journal of Baltic Science Education, Vol. 20, No. 1, 2021

Uploaded by

Copyright:

Available Formats

Journal of Baltic Science Education, Vol. 20, No. 1, 2021

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Baltic Science Education, Vol. 20, No. 1, 2021

Uploaded by

Copyright:

Available Formats

Vol.20, No.

Published since 2002 Address:

ISSN 1648–3898 (Print) © Scientia Socialis Ltd., a member of the Lithuanian

ISSN 2538–7138 /Online/

THE CONCEPT OF INERTIA: AN INTERDISCIPLINARY APPROACH

THE EFFECTS OF STEM ACTIVITIES ON 6TH GRADE STUDENTS’ CONCEPTUAL DEVELOPMENT

DEVELOPING A SELF-EFFICACY SCALE TOWARD PHYSICS SUBJECTS FOR LOWER-

IMPACT OF INSTRUCTION ON SCIENCE PERFORMANCE: LEARNING INITIATIVE AS A

STUDENT, PARENT AND TEACHER PERCEPTIONS ON THE BEHAVIORAL CHARACTERISTICS

THE CONCEPTUAL STRUCTURE OF CHEMICAL EQUILIBRIUM IN UPPER-SECONDARY SCHOOL

PEDAGOGICAL MODEL FOR DECOLONISING, INDIGENISING AND TRANSFORMING SCIENCE

THE PREDICTORS OF NINTH GRADE STUDENTS’ ATTITUDES TOWARDS PLANTS

GRADE 6 & 9 STUDENT AND TEACHER PERCEPTIONS OF TEACHING AND LEARNING

EFFECT OF A STEM-ORIENTED COURSE ON STUDENTS’ MARINE SCIENCE MOTIVATION,

ANALYSIS OF TWO-TIER QUESTION SCORING METHODS: A CASE STUDY ON THE LAWSON’S

INFORMATION FOR CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

PROBLEMS OF PSYCHOLOGY IN THE 21ST CENTURY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

NATURAL SCIENCE EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

EDUCATION POLICY, MANAGEMENT AND QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

THE CONCEPT OF INERTIA: AN INTERDISCIPLINARY APPROACH

Relative motion water/bucket Water’s surface Inertial forces

Received: January, 10, 2021 Accepted: February 02, 2021

Abstract. 5E instructional model is Michael Allan A. Bahtaji

Exposure to Science and Math Courses

5E instructional model method Conventional method

Exposure to science and math High ESM Group 1 Group 3

Teaching method ESM Pre-test Post-test

** p < .01, two-tailed. * p < .05, two-tailed.

Dependent Teaching method ESM

Learning Cycle Group Conventional Group High ESM Low ESM

M 5.02 5.03 6.00 4.05

Dependent Teaching Method ESM

Learning Cycle group Conventional group High ESM Low ESM

M 6.10 5.61 7.06 4.65

High ESM 43 5.98 0.71 7.16 0.72 1.19 7.73**

Conclusions and Implications

Received: August 24, 2020 Accepted: January 08, 2021

Michael Allan A. Bahtaji PhD Science Education, Research Specialist, Technological

DEVELOPMENT OF SOUND ISSN 2538-7138 /Online/

Ayşegül Dedetürk, Abstract. Engineering design process-

Research Aim and Research Questions

Cognitive domain in Bloom’s

1. Notices that the sound is propagated in waves in all directions. 1 Analysis

2. Discovers that the sound is reflected when it encounters an obstacle. 2 Comprehension

3. States that an echo occurs as a result of sound reflection. 3 Comprehension

6. Determines that sound loudness decreases through absorption. 6 Evaluation

7. Realizes that different materials absorb sound differently. 7 Evaluation

Item P r Item P r Item P r

1 0.61 0.78 5 0.64 0.72 9 0.58 0.83

2 0.69 0.61 6 0.58 0.61 10 0.50 1.00

3 0.61 0.78 7 0.58 0.72 11 0.61 0.78

4 0.50 1.00 8 0.53 0.94

Lesson title Overview

Lesson title Overview

Lesson title Overview

Partially pre 55 50 Partially pre 35 45

Partially pre 60 40 Partially pre 35 30

Partially pre 60 80 Partially pre 55 45

Partially pre 10 15 Partially pre 35 30

Partially pre 20 20 Partially pre 65 45