IMPLICATIONS OF TEACHERS’ BELIEFS

ABOUT MATHEMATICS FOR CLASSROOM

AND TEACHER EDUCATION REFORM
Catherine P. Vistro-Yu
Department of Mathematics, Ateneo de Manila University, Quezon City

Abstract
This article reports partial results from a study that investigated secondary school
mathematics teachers’ beliefs and attitudes about mathematics. Using Ernest’s (1988) model, it
describes what teachers believe about the nature of mathematics and about the teaching and
learning of mathematics consistent with the instrumentalist, Platonist, and problem solving views
of mathematics. A 47-item Mathematics Attitude Survey and a demographic questionnaire were
administered to 57 high school mathematics teachers. Data analyses used the non-parametric tests
of Friedman, Mann-Whitney, and Kruskal-Wallis and a non-parametric correlation test. Results
revealed that teachers hold strong beliefs about mathematics that are consistent with the problem
solving view. There is a significant difference in the intensity with which public and private
school teachers hold the instrumentalist view of mathematics. There is also a significant difference
in the intensity with which relatively low- and high-scorers in the Licensure Examination for
Teachers hold the instrumentalist view. Implications on how mathematics teacher education
programs can be improved are discussed.
Keywords: teachers’ beliefs, mathematics for classroom, teacher education reform, instrumentalist,
Platonist, problem solving

Introduction

Ernest noted that the practice of teaching mathematics depends on a number of key elements
(1):
• The teacher’s mental contents or schemas, particularly the system of beliefs concerning
mathematics and its teaching and learning;
• The social context of the teaching situation, particularly the constraints and opportunities it
provides; and
• The teacher’s level of thought processes and reflection.

These factors determine the autonomy of the mathematics teacher and subsequently, also the
outcome of teaching innovations, like problem solving that depends on teacher autonomy for their
successful implementation (1). The present study addresses the first key element – the teacher’s
mental contents or schemas including the system of beliefs concerning mathematics and its teaching
and learning. Teachers’ mental contents or schemas include a) mathematics content knowledge, b)
beliefs and attitudes displayed towards an object or a group of objects, and c) belief systems about the
nature of mathematics and the teaching and learning of it. Although mathematics content
knowledge is important, it is not enough to determine teachers’ readiness or capacity to shift to more
desirable teaching approaches. Teaching reforms cannot take place unless teachers’ deeply held
beliefs about mathematics and its teaching and learning change (1).

Theoretical Framework

Teachers’ conceptions of mathematics are almost always interpreted in light of a philosophy

or several philosophies of mathematics because these constitute their views, attitudes, beliefs, and
preferences about the nature of mathematics as well as the teaching and learning of mathematics.
Ernest distinguished three possible conceptions of mathematics (1, p.2) that relate significantly to a
philosophy of mathematics (2):

First of all, there is the instrumentalist view that mathematics is an accumulation of

facts, rules and skills to be used in the pursuance of some external end. Thus, mathematics is a set
of unrelated but utilitarian rules and facts.
Secondly, there is the Platonist view of mathematics as a static, but unified body of
certain knowledge. Mathematics is discovered not created.
Thirdly, there is the problem solving view of mathematics as a dynamic, continually
expanding field of human creation and invention, a cultural product. Mathematics is a process of
enquiry and coming to know, not a finished product, for its results remain open to revision.

Ernest further associated the above views to corresponding models of teaching in the aspects
of the teacher’s role, the intended outcome of teaching, and the teacher’s use of curricular materials
(1). All these further reflect a corresponding learning model implicitly adhered to by the teacher.
These aspects are embedded in a diagram (cf. 1, p.3) illustrating the specific relationships between
teachers’ beliefs about the nature of mathematics and their models of teaching and learning. Table 1
shows a likely set of significant associations and implications of one’s view about mathematics based
on Ernest’s model (1). This matrix, where the instrumentalist view of mathematics is at the bottom,
appears to model a hierarchy.
As shown by Table 1, a teacher who follows the instrumentalist view of mathematics will
tend to take an instructor’s role in teaching where the main objective is for students to master the
skills needed in mathematics. Performance is important because that is how an instructor can
determine if mastery has been achieved. To ensure this, teachers of this view would strictly follow the
prescribed curriculum that is broken down into a hierarchy of specific skills to be learned. The basis
of knowledge here is rules, not necessarily with understanding. There is an implicit belief that the
curriculum and the corresponding instructional materials offer the best formula for mastering skills;
thus, instructions would tend to be very rigid.
However, teachers who follow the Platonist view of mathematics would tend to take on an
explainer’s role in teaching. They tend to lecture and explain concepts, focusing on mathematical
content. They emphasize students’ understanding of ideas and processes, particularly students’
understanding of the logical relationships of mathematical concepts. The objective of instruction is
for students to have a unified concept of mathematics and a consistency of ideas.
The problem-solving view of mathematics is the highest in the hierarchy because such a view
encourages learning by active construction of one’s knowledge. Teachers who subscribe to this view
see themselves as facilitators of learning. As such, they would prefer to construct or develop their
own materials that would suit the needs and interests of their students. Their instructional objective
is to develop more confident and better problem-solvers. This view of mathematics encourages
creativity and multiple approaches to learning a concept or skill.
Table 1. Teachers’ views about mathematics and their implications based on Ernest’s model (1, pp. 2-4).
Related Research

Various studies on mathematics teachers’ beliefs and conceptions have consistently

confirmed one major idea: that mathematics teachers’ beliefs and conceptions, particularly about the
nature of mathematics and about the teaching and learning of mathematics, have an impact on the
type of mathematics instruction they deliver in the classroom. Thompson identified several studies in
mathematics education (3-12) that have indicated that beliefs about mathematics and its teaching
play a significant role in shaping the teachers’ patterns of instructional behavior (2). In particular,
Thompson concluded that the relationship between teachers’ conceptions and their instructional
decisions and behavior is a complex one (8). In her study of three junior high school teachers, she
found that the quality and level of these teachers’ reflection about their beliefs contributed to the
presence or lack of congruence between their beliefs and their instructional behavior, making the said
relationship complex. Additional conceptions about their students and the social and emotional
make-up of the classroom also affect their instructional behavior patterns, perhaps, much more for
some teachers. These conceptions likewise affect teachers’ view about mathematics and its teaching.
Many studies (13-15) have strongly suggested that teachers’ cognitive knowledge of
mathematics directly influence what they do in the classroom and ultimately what their students
learn (16). A critical study of the literature related to teachers’ cognitive knowledge of mathematics,
also reveals that the linkage between teachers’ mathematical knowledge and their instructional
practices is complex. Fennema & Franke proposed a working framework that provides a more
complete picture of how teachers’ mathematical knowledge influences their pedagogical practice (16).
Rather than focusing on teachers’ mathematical knowledge alone, the framework illustrated how
teachers’ mathematical knowledge along with teachers’ knowledge of the learning contexts, of their
students’ understanding of mathematics, and of specific pedagogical techniques, including
mathematics teachers’ beliefs contribute to the way they teach mathematics. For instance, a
classroom teacher knows that a teaching methodology would not elicit the same response from two
different classes principally because of the varying personalities that make up the two classes.
Therefore, the teacher would try to use different methodologies that would suit the classes.
Thompson also reported varying degrees of consistency between mathematics teachers’
espoused beliefs about mathematics and their instructional practices based on studies done by other
researchers (2). For example, (6) reported that strong relationships exist between the knowledge base
of novice teachers and their instructional practice. McGalliard (10) observed a high degree of
consistency between the mathematics conceptions of four senior high school teachers and their
instructional practices in geometry (2). In the same study, McGalliard also reported inconsistencies
between teachers’ professed beliefs and their instructional practice. This cautions researchers in
interpreting results of studies involving teachers’ professed beliefs. It further suggests that meaningful
studies on teachers’ beliefs should include observable data on teachers’ teaching practices.

Purpose of the study

The purpose of the study was to investigate high school mathematics teachers’ beliefs about
the nature of mathematics and about the teaching and learning of mathematics. Following Ernest’s
framework (1), the study sought to identify the beliefs that teachers hold consistent with the
instrumentalist, Platonist, and the problem solving views of mathematics and to determine the
intensity with which they hold these beliefs. Related to this is the importance of the relationship
between the beliefs held and certain grouping variables used to describe the sample of teachers for the
study. In sum, the study attempted to answer the following questions:

1. What do high school mathematics teachers believe about the nature of mathematics and
about the teaching and learning of mathematics?
2. Do teachers hold much more strongly beliefs that are consistent with one type of view of
mathematics according to Ernest’s framework?
 3. Do certain subgroups of teachers hold much more strongly beliefs consistent with the types
of view of mathematics according to Ernest’s framework than other subgroups do?
4. Is there a relationship between any of the grouping variables used to describe the sample of
teachers for the study and the types of view of mathematics according to Ernest’s framework?

Method

Participants. Fifty-seven high school mathematics teachers participated in the study. Of these, 26
teach in public schools spread over 14 regions in the country. The remaining 31 teach in private
sectarian schools that belong to one geographical cluster in Metro Manila. Tables 2a and 2b
summarize the demographic variables used to describe the teachers.

Table 2a. Mean and standard deviation of age, years of teaching, and Licensure Exam score
Score in
Years of
Age Licensure
Teaching
Examination
Number of teachers who a
53 51 37
reported (n)
Mean (x) 32.6 9.4 78.9%
Standard Deviation (σ) 7.4 6.8 4.3
a
One teacher simply reported “passed” for this item.

Table 2b. Frequency and percentage for gender, post-baccalaureate studies, awards, non-teaching jobs,
and in-service training
Variables Frequency %
GENDER
Female 37 70
Male 16 30
Total 53 100
POST-BACCALAUREATE STUDIES
Yes 4 8
No 47 92
Total 51 100
HONORS/AWARDS RECEIVED
Yes 5 10
No 46 90
Total 51 100
NON-TEACHING JOB EXPERIENCE
Yes 9 17
No 43 83
Total 52 100
IN-SERVICE TRAINING
Yes 25 48
No 27 52
Total 52 100

Instrument. In addition to the demographic questionnaire, the main instrument used to determine
teachers’ beliefs about mathematics was a 47-item Mathematics Attitude Survey containing a mixture
of belief and attitude statements. This instrument was used in the Second International Mathematics
Study (SIMS) that was completed in 1982 as part of the effort to learn about the 12-year-old
participants from a cross-cultural sample of 18 countries (17). The same instrument was used to
obtain information about middle school mathematics teachers in a NSF-funded Middle School
Mathematics Project at the University of Georgia in 1989 (18). Thirteen of the items in the survey
have been classified as pertaining to beliefs about the nature of mathematics while 12 of these refer to
beliefs about the teaching and learning of mathematics. Two items were classified as pertaining to
both. The rest of the items are a combination of belief and attitude statements pertaining to other
aspects of mathematics, which will not be discussed in this paper. The classification followed the one
used for SIMS. The survey uses a 5-scale response: strongly disagree, disagree, undecided, agree, and
strongly agree. Numerical values of 1, 2, 3, 4, and 5, respectively, were assigned to the responses to
facilitate quantitative analyses.

Data Collection and Analyses. The two instruments were administered to the private school teachers
at the beginning of the aforementioned half-day in-service workshop while they were administered to
the public school teachers at the end of a class session.
The 27 items in the Mathematics Attitude Survey were classified further according to the
instrumentalist (I), Platonist (P), and problem-solving (PS) views of mathematics. The researcher
and a mathematics education professor who had done extensive studies on the philosophy of
mathematics education independently classified the items. Initially, the two agreed on the
classification of only 25 items. A 100% agreement was achieved after some discussion. It is
important to note, however, that this categorization forced one to place items in one and only one
category if only to simplify the analysis. The reality is that some of these items may be classified in
two or three categories. For example, the item on logic may be considered important in the problem
solving view. However, because the Platonist view emphasizes logic, it is classified under the
Platonist view of mathematics. Furthermore, since a hierarchical structure has been assumed,
classifying the item at a lower level in the hierarchy does not diminish its importance in the higher
level. Table 3 indicates the classified items.

Table 3. Classification of items according to the type of belief and the type of view of mathematics
Instrument- Problem-
Platonist
alist solving
Nature of Items 5, 7, 8, Items 3, 6, Items 1, 2, 4,
Mathematics 27, 40 12, 14 9, 10, 11
Teaching and Items 27, 40, Item 44 Items 13, 23,
Learning of 43 34, 36, 37,
Mathematics 41, 42, 45,
46, 47

For every item, a weighted average was obtained based on the responses of all 57 teachers. A
weighted average of 3.5 and above is considered as manifesting the belief statement. A weighted
average of 2.5 and below is considered as not manifesting the belief statement. A weighted average
between 2.5 and 3.5 is considered as manifesting indecision about the belief statement. For every
teacher, one score for each of the three views of mathematics was obtained by taking the average of
the responses to the items classified under each type of view. Thus, associated with each teacher are
three scores. Hence, there are 57 scores reported for each type of view of mathematics. The mean of
the 57 scores for each type of view was then obtained.
Each of the demographic variables type of school, age, gender, years of teaching experience, post-
baccalaureate studies, awards, in-service training, non-teaching job, and Licensure Examination score was
assigned a nominal value according to the subgroups determined by the researcher. Some of the
subgroups were natural groupings (e.g. female vs. male, private vs. public school, or have vs. have
not) while the rest were ascertained based either on the available data or on the researcher’s
perception of critical periods or points in the life of or about a teacher. For example, the following
were grouped based on perceived critical periods or points: AGE – (x < 30, 30 ≤ x < 40, x > 40) and
YEARS OF TEACHING – (1 ≤ x ≤ 3, 4 ≤ x ≤ 7, 8 ≤ x ≤ 15, x > 15).
Since the variables involved were discrete, distribution-free (non-parametric) tests were used
to check for significant differences between the means, and to determine significant correlations
between variables (19). The following non-parametric tests were used from the SPSS software: the
Friedman Test, the Mann-Whitney Test, and the Kruskal-Wallis Test for comparison of mean ranks.
Finally, correlations were obtained using the Spearman ρ coefficient. A non-parametric test for the
Spearman ρ coefficient was also used from the SPSS software to check for any significant correlation
between each demographic variable and each type of view of mathematics.

Results

Beliefs about the Nature of Mathematics. The weighted averages showed that teachers believe in the
following aspects of mathematics:
• The importance of rules
• The importance of logic
• The importance of proofs
• That mathematics can be solved in different ways
• That mathematics offers opportunities for creativity
• That mathematics is a dynamic field
• That mathematics allows for the use of trial and error in solving problems
The weighted averages also showed that teachers do not seem to believe in the following
aspects of mathematics:
• That mathematics is made up of unrelated topics
• That new discoveries are not being made in mathematics
It appears that teachers are undecided about the following aspects of mathematics:
• The role of memorization in mathematics.
• That there is little place for originality in mathematics
• That problems can be solved without using rules
Beliefs about the Teaching and Learning of Mathematics. The weighted averages showed that
teachers believe in the following with respect to the teaching and learning of mathematics:
• The importance of memorization in learning mathematics
• The role of tests in learning mathematics
• The importance of teachers’ explanations in the learning of mathematics
• The need for estimation skills
• The importance of the problem-solving process
• The value of using manipulative objects to learn mathematics
• The role of play
• The role of problem solving in the teaching of mathematics
• The value of computers in teaching mathematics
• The value of group work
• The value of writing about mathematics
There are no belief statements about the teaching and learning of mathematics that the
teachers do not believe in or are undecided about.

Comparison Tests on the Type of View of Mathematics. To determine whether teachers tend toward
one type of view of mathematics more than the other two, the Friedman Test was used. In this test, a
teacher’s scores for each of the three views are converted to ranks and a mean rank for each of the
three views is then obtained (19). Table 4 shows the results from this test.

Table 4. Mean, standard deviation, and mean rank of each type of view of mathematics based on
Friedman Test
Instrument- Platonist Problem p-value
alist Solving
Nature of X = 3.20; X = 3.57; X = 3.87;
mathematics σ = 0.67 σ = 0.48 σ = 0.54
Teaching and X = 3.53; X = 4.02; X = 4.39;
learning of σ = 0.81 σ = 0.94 σ = 0.30
mathematics
Overall X = 3.36; X = 3.66; X = 4.20;
σ = 0.61 σ = 0.45 σ = 0.32
Mean Rank 1.40 1.75 2.84 0.000*
*significant at p < 0.02

The p-value obtained from this test is 0.000, which is significant (p < 0.02). Hence, it
implies that teachers significantly tend to hold much more strongly beliefs consistent with the
problem solving view (highest mean rank) than those consistent with the instrumentalist and the
Platonist views of mathematics.

Comparison Tests on Subgroups within Variables. To determine whether subgroups within each
variable differ in the way they view mathematics, the Mann-Whitney Test for 2-independent samples
and the Kruskal-Wallis Test for k-independent samples (k > 2) were performed. Both tests compare
the mean ranks of the subgroups. The Mann-Whitney Test is the non-parametric equivalent of the
t-test while the Kruskal-Wallis Test is the non-parametric equivalent of the one-way ANOVA (20).
Table 5 shows that the tests yielded only 3 significant results. These are for the variables
school and Licensure Exam Score with the instrumentalist view and for the variable years of teaching
with the problem solving view.

Table 5. Significant test results for 3 variables.

Variable Sub-group N Mean Test p-value
Rank (2-tailed)
School (for Public 26 34.94 Mann- .013*
Instrumentalist Whitney
view) Private 31 24.02
Licensure Passed to 22 23.18 Mann- .016*
Exam Score 79.5% Whitney
(for X > 79.5% 16 14.44
Instrumentalist
view)
Years of 1 ≤ x ≤ 3 5 18.10 Kruskal- .024*
Teaching (for Wallis
Problem 4≤x≤7 20 23.40
Solving view) 8 ≤ x ≤ 15 19 33.82
x > 15 7 17.86
*significant at p < 0.05 level
 Results from Table 5 can then be stated as follows:
1. Public school teachers significantly tend to hold much more strongly beliefs consistent with the
instrumentalist view of mathematics compared to private school teachers.
2. Teachers who obtained relatively low scores in the Licensure Examination significantly tend to
hold much more strongly beliefs consistent with the instrumentalist view of mathematics
compared to teachers who obtained relatively high scores in the said exam.
3. There is a significant difference in the intensity with which different subgroups for the variable
years of teaching hold beliefs consistent with the problem solving view.

Discussion

If beliefs are to be interpreted as expressions of values, then teachers may value many of the
important aspects of mathematics, foremost of which are logic, proof, and the dynamic aspect of
mathematics. Rules seem to play an important role in mathematics. However, research has shown
that placing too much emphasis on rules may limit higher-order thinking skills, such as creative
thinking (21) and cognitive elaboration (22). Results seem to indicate that teachers generally agree
that mathematics is a coherent body of knowledge and a dynamic field of endeavor.
On the average, teachers seem to be undecided about the role of memorization in
mathematics, the place of originality in mathematics, and the tolerance for problem solving without
rules. The second point appears to be inconsistent with the beliefs that mathematics is a good field
for creative people and that mathematics is a dynamic field. A possible explanation for this lies in the
way teachers perceive mathematics – first, as a school subject and second, as a non-school endeavor.
Research studies have shown these two separate perceptions to exist among children (23) and,
possibly, among practicing teachers (8).
The teachers’ beliefs indicate that they value multiple approaches to the teaching and
learning of mathematics, and that these approaches are wide-range. They include the use of
technology, writing, play, group work, and manipulative objects for the high school level. These
beliefs indicate that teachers have very positive views about using creative and fun approaches to
learning mathematics.

Comparisons and Correlations. The Friedman Test showed that teachers significantly tend to hold
much more strongly the problem solving view of mathematics compared to the instrumentalist and
Platonist views. This is encouraging because the problem solving view is the highest in the hierarchy
of Ernest’s framework. This is a good starting point and, if they persist in these beliefs and carry
them on to actual practice, genuine reforms in mathematics teaching can begin to take shape.
The Mann-Whitney Test, the Kruskal-Wallis Test, and the non-parametric correlation test
for Spearman ρ coefficient indicate crucial areas that need further investigation. The Mann-Whitney
Test results supported by the correlation test showed that public school teachers significantly tend to
hold more strongly the instrumentalist view of mathematics compared to private school teachers.
This result might partly explain why public schools have not been successful in mathematics, as
revealed by results of the Third International Mathematics and Science Study-Repeat (TIMSS-R)
that included 114 public schools out of a total sample of 150 schools all over the Philippines (24).
Teachers who hold more strongly the instrumentalist view tend to teach mathematics as a set of
unrelated topics, focusing more on rules and skills (procedural understanding), rather than on a deep
conceptual understanding of the subject matter. Previous studies on teachers (25) have shown that
procedural understanding without conceptual understanding of mathematics does not last.
Procedural understanding encourages a fragmented type of learning mathematics. Conceptual
understanding, on the other hand, promotes connectivity and networking of concepts, which is much
more useful and needed in mathematics.
The result bolsters the need for training programs aimed at developing deeper understanding
of mathematical concepts among public school teachers. One can deduce from here that part of
helping public school teachers is to help them “weaken” their instrumentalist beliefs about
mathematics, it being the lowest level in the hierarchy. The aim is to help them learn to teach
mathematics much more effectively and correctly, thereby leading them to turn their “problem
solving” beliefs into meaningful actions.
The Mann-Whitney Test results, again supported by the correlation test, also showed that
teachers with relatively low scores in the Licensure Examination significantly hold more strongly
beliefs consistent with the instrumentalist view compared to those who got relatively higher scores in
the same exam.
The results have implications for pre-service teacher education programs. Studies by Ibe (26-
28) have shown that mean scores obtained by prospective high school mathematics teachers in the
Professional Board Examination for Teachers (PBET) (given from 1978 to 1995) and the Licensure
Examination (given beginning in 1996), both in the General Education and Major Subject
components of the exams, fall below the desired score of 50%. This pattern of performance in the
exams reflects the level of competence of mathematics teachers from the cognitive standpoint. The
results from the current study add to these data. Performance in the Licensure Exam also reflects the
affective tendencies of teachers, in this case through their espoused beliefs. Because relatively low-
scorers in the Licensure Exam tend to be more instrumentalist in the way they view mathematics,
they would be less effective teachers of mathematics. Research has shown, e.g. (1), that the
instrumentalist approach may fail to engage students in genuine learning. Teacher educators must
continue to be vigilant and ensure that teacher preparation programs do address the cognitive and
affective needs of future teachers.
The Kruskal-Wallis Test calls attention to seemingly critical periods in teachers’ professional
lives. By looking at the mean ranks of the four subgroups, the result from this test may be
interpreted in the following way: teachers who have taught for 8 to 15 years tend most toward the
problem solving view followed by teachers who have taught for 4 to 7 years. There does not seem to
be a significant difference between the remaining two subgroups. This may mean that it is the first
two subgroups of teachers who have the potential to initiate reforms toward a more problem solving
approach to teaching.
One possible explanation for this is the fact that these are teachers who have been in the
profession beginning in the last half of the 1980s to the first two-thirds of the 1990s. Those years
comprised a vibrant period of reform efforts and innovations in mathematics education locally (the
advent of the SEDP curriculum and the nationwide Engineering and Science Education Project) and
internationally (the launching of the NCTM Curriculum and Evaluation Standards in 1989 and the
Third International Mathematics and Science Study in 1995). These would be teachers who may
have been exposed to these efforts, not necessarily through in-service training (a significance test on
the data reveal no significant differences involving in-service training and years of teaching variables),
but, perhaps, through involvements in the reform efforts themselves. These results need further
investigation. What is it about the 8-15 years and 4-7 years of teaching experience that caused the
significant difference? What other explanations can be offered? What alternative groupings can be
used?

Conclusions and Recommendations

The study showed that high school mathematics teachers hold certain beliefs and views about
mathematics, which have both desirable and undesirable consequences in the way they teach in the
classroom. These beliefs cannot be ignored because of the impact they have on classroom teaching
(8).
This study also showed that there are variables that significantly relate to the intensity with
which teachers view mathematics from the instrumentalist perspective. There is a need, therefore, to
pay attention to the variables type of school and Licensure Examination score. For instance, in the
Licensure Exam, the cut-off point of 80% was based on the available data – the scores ranged from
71% to 87%. Would the test yield the same result if a different kind of grouping were used? Further
studies on this variable have to be done.
 The study has also shown that teachers significantly tend to hold more strongly the problem
solving view of mathematics over the other two types of views. Therefore, teacher educators should
assist and support teachers in concretizing these beliefs by undertaking genuine reforms at both the
pre-service and in-service education levels. Teacher education programs must pay attention to both
affective and cognitive issues that surround prospective and practicing teachers.
The study has its limitations that have to be noted. First, the sample of 57 teachers may be
too small especially since non-parametric methods were used for the analyses. Replicate studies over a
larger sample of teachers are highly recommended to confirm the results that have been obtained. In
the Mann-Whitney Test, one pair of variables yielded a (2-tailed) p-value of 0.052, which is slightly
over the largest acceptable p-value of 0.10. One cannot help but wonder if a larger sample would
have yielded significant results.
Second, the Kruskal-Wallis test performed on the years of teaching variable in relation to the
problem solving view offered limited information. Perhaps, more detailed investigations using pairs
of subgroups within the same variable or tests using different subgroups would be very enlightening.
Third, Ernest’s framework (1) is not the only perspective that can be used to investigate
teachers’ beliefs. Ernest’s categorizations alone can be too limiting for the complex nature of our
school system. The political, socioeconomic, and socio-cultural factors that greatly affect schools and
students may not be addressed by Ernest’s framework. There is a need, therefore, to use other
frameworks that might enrich the data that have been gathered from this study. An alternative
framework may be a cross between Ernest’s levels (1) and Green’s (29) three dimensions of belief
systems that describe the way beliefs relate to one another in the system. Green used the dimensions
of primary and derivative beliefs, central or peripheral beliefs, and clustered beliefs.
Fourth, what have been looked at are general views and group results. It would be
interesting to know what certain individuals really believe in and how consistent they are in their
expression of beliefs across different contexts. Case studies of individual teachers are recommended
to follow up on the results here. Future studies can also move toward more general affective issues
influencing the teacher.
Fifth, it must be remembered that what have been gathered are teachers’ espoused beliefs,
which may be translated differently into action. Consistent with McGalliard’s recommendation (10),
studies on how teachers translate these beliefs into their teaching are also recommended. In what
ways can we help teachers so that their actions are consistent with their espoused beliefs? Studies that
answer this question are very useful in the teacher education field.

References
1. P. Ernest, “The impact of beliefs on the teaching of mathematics”, paper prepared for ICME VI,
Budapest, Hungary, July 1988, pp.1-4.
2. A. G. Thompson, “Teachers’ beliefs and conceptions: A synthesis of the research”, in Handbook
of Research on Mathematics Teaching and Learning, D. A. Grouws, Ed., (Macmillan, New York,
1992), pp. 127 - 146.
3. B. J. Dougherty, “Influences of teacher cognitive/conceptual levels on problem-solving
instruction”, in G. Booker et al., Eds., Proceedings of the Fourth International Conference on the
Psychology of Mathematics Education, (International Group for the Psychology of Mathematics
Education, Oaxtepec, Mexico, 1990), pp. 119 – 126.
4. R. Marks, Those who appreciate: The mathematician as secondary teacher. A case study of Joe, a
beginning mathematics teacher, (Stanford University School of Education, Stanford, CA, 1987).
5. R. Kesler, Jr., doctoral dissertation, University of Georgia, Athens (1985).
6. R. Steinberg, J. Haymore, R. Marks, paper presented at the annual meeting of the American
Educational Research Association, San Francisco, April 1985.
7. C. E. Grant, doctoral dissertation, University of North Dakota (1984); Dissertation Abstracts
International, 46, DA8507627 (1984).
8. A. Thompson, Educational Studies in Mathematics 15, 105 – 127 (1984).
9. S. Lerman, International Journal of Mathematical Education in Science and Technology 14 (1),
69 –66 (1983).
10. W. A. McGalliard, Jr., doctoral dissertation, University of Georgia (1982); Dissertation Abstracts
International, 44, 1364A (1983).
11. T. Kuhs, doctoral dissertation, Michigan State University, East Lansing (1980).
12. J. C. Shroyer, paper presented at the annual meeting of the American Educational Research
Association, Toronto (1978, March).
13. D. L. Ball, Unlearning to teach mathematics (Issue Paper 88-1), (Michigan State University,
National Center for Research on Teacher Education, East Lansing, 1988).
14. T. R. Post, G. Harel, M. J., Behr, R. Lesh, “Intermediate teachers’ knowledge of rational number
concepts”, in Integrating research on teaching and learning mathematics, E. Fennema, T. P.
Carpenter, S. J. Lamon, Eds., (SUNY Press, Albany, NY, 1991), pp. 177 – 198.
15. S. I. Brown, T. J. Cooney, D. Jones, “Mathematics teacher education”, in Handbook of Research
on Teacher Education, W. R. Houston (Ed.), (Macmillan, New York, 1990), pp. 639 – 656.
16. E. Fennema, M. L. Franke, “Teachers’ knowledge and its impact”, in Handbook of Research on
Mathematics Teaching and Learning, D. A. Grouws, Ed., (Macmillan, New York, 1992), pp. 147
– 164.
17. D. F. Robitaille, R. A. Garden, Eds., The IEA Study of Mathematics II: Contexts and Outcomes
of School Mathematics, (Pergamon Press, Oxford, 1989).
18. D.E. Barnes, T. J. Cooney, I. Coronel, C. P. Vistro, “Evaluation of the Middle School Project”,
(The University of Georgia, Atlanta, GA, 1989).
19. M. Hollander, D. A. Wolfe, Nonparametric Statistical Methods, (John Wiley & Sons, NY, 1973).
20. M. F. Triola, Elementary Statistics, (Addison-Wesley Publishing Company, Massachusetts, ed. 6,
1995).
21. P. Zeitz, The Art and Craft of Problem Solving, (John Wiley & Sons, NY, 1999).
22. E. A. Silver, “Knowledge organization and mathematical problem solving”, in Mathematical
Problem Solving Issues in Research, F. K. Lester, J. Garofalo, Eds., Franklin Institute Press,
Pennsylvania, 1982), pp.14 – 24.
23. D. W. Carraher, T. N. Carraher, A. D. Schliemann, British Journal of Developmental Psychology
3, 21-29 (1985).
24. F. Brawner, et al., TIMSS-R Philippine Report Volume 2: Mathematics. (Department of
Education, Culture, & Sports (DECS), Department of Science and Technology – Science
Education Institute (DOST-SEI), & University of the Philippines National Institute of Science
and Mathematics Education (UP-NISMED), Manila, 2000).
25. C. P. Vistro, doctoral dissertation, University of Georgia (1991); Dissertation Abstracts
International, 52, 6, 2059A, DA9133542 (1991).
26. M. D. Ibe, “The First Licensure Examination for Teachers (LET): Implications for Teacher
Education Initiatives”, in Proceedings of the MATHTED ’97 Conference, (De La Salle University,
Manila, 1997), pp. 41 – 57.
27. M. D. Ibe, “The Scenario in Mathematics Teacher Education. In the Proceedings of the
Conference”, in Strengthening Collaboration in Mathematics Teacher Education, (Ateneo de
Manila University, Quezon City, 1995), pp. 11 – 29.
28. M. D. Ibe, “An Analysis of Examinee Performance in the 1991 PBET”, (University of the
Philippines, Diliman, 1991).
29. T. F. Green, The Activities of Teaching, (McGraw-Hill, NY, 1971).

Acknowledgements
This paper was taken from a research funded by the Celestino M. Dizon Endowed Professorial
Chair and the Senator Gil J. Puyat Endowed Professorial Chair, Ateneo de Manila University, 2000 –
2001. The author also wishes to acknowledge Ms. Josephine Chua of the Ateneo Mathematics
Department for her invaluable assistance and insights.