MODULE 

V: RESEARCH
Intended Learning Outcomes
At the end of this module, the learner should be able to :
1. identify the different research approaches;
2. describe the nature  of quantitative research, its characteristics and 
          strengths;
3. understand the concepts of  standardized instruments, validity and 
           reliability, and random   sampling;
4. differentiate the kinds of research designs;
5. explain the concept of action research;
6. identify and explain the various data gathering techniques

The Selection of a Research Approach

Research approaches are plans and the procedures for research that span the steps from broad
assumptions to detailed methods of data collection, analysis, and interpretation. The selection of
a research approach is also based on the 
1. nature of the research problem or issue being addressed 
2. the researchers’ personal experiences
3. and the audiences for the study.

THE THREE APPROACHES TO RESEARCH

The three research approaches are advanced: (a) qualitative, (b) quantitative, and (c)
mixed methods.

Qualitative research is an approach for exploring and understanding the meaning individuals or

groups ascribe to a social or human problem. 

Qualitative research 

The process of research involves 
emerging questions and procedures
data typically collected in the participant’s setting
data analysis inductively building from particulars to general themes
and the researcher making interpretations of the meaning of the data.

Qualitative research 

The final written report has a flexible structure. 

Those who engage in this form of inquiry support a way of looking at research that honors 
an inductive style
a focus on individual meaning
and the importance of rendering the complexity of a situation.

Quantitative Research
An approach for testing theories by examining relationship among variables.
These variables, in turn, can be measured, typically on instruments, so that numbered data can be
analyzed using statistical procedures. 

Quantitative Research
The final written report has a set structure consisting of 
Introduction
literature and theory
Methods
Results
and discussion.

Quantitative Research
Like qualitative researchers, those who engage in this form of inquiry 
have assumptions about testing theories deductively
building in protections against bias
controlling for alternative explanations
and being able to generalize and replicate the findings
Mixed methods research 
is an approach to inquiry involving collecting both quantitative and qualitative data, integrating
the two forms of data, and using distinct designs that may involve philosophical assumptions
and theoretical frameworks. 
The core assumption of this form of inquiry is that the combination  of qualitative
and quantitative  approaches provides a more complete understanding of a research problem than
either approach alone.

Quantitative Research
A survey of research designs shows that the researcher has a variety of designs to choose from in
attempting to research works. 
One can approach it quantitatively or qualitatively. 
The descriptive research design is one among the most commonly used by many researchers.

Quantitative research designs use numbers in stating generalization about a given problem

or inquiry in contrast to qualitative research that hardly uses statistical treatment in
stating generalizations. 
These numbers are the results of objective scales of measurements of the units of analysis
called variables. 
Research findings are subjected to statistical treatment to determine significant relationships or
differences between variables, the
results of which are the bases for generalization about performance. 

Characteristics of Quantitative Research
Methods or procedures of data gathering include like age, gender, educational status, among
others, that can call for measurable characteristics of the population.

Standardized instruments guide data collection, thus, ensuring the

accuracy, reliability and validity of data.

Figures, tables or graphs showcase summarized data collected 

in order to show trends, relationships or differences among variables. 
In sum, the charts and tables allow you to see the evidence collected.

A large population yields more reliable data, but principles of random sampling must be strictly
followed to prevent researcher’s bias.

Quantitative methods can be repeated to verify findings in another setting,

thus reinforcing validity of findings.

Quantitative research puts emphasis on proof, rather than discovery.

Basic Concepts
What are Psychological Tests?

Psychological Test is an objective and standardized measure of a

sample of behavior (Anastasi and Urbina, 1997).

Psychological Test is a measurement instrument with  three defining characteristics:

1. A psychological test is a sample behavior.

2. The sample is obtained under standardized conditions.
3. There are established rules for scoring, or for obtaining quantitative
          (numeric) information from the behavior sample (Murphy & 
          Davidshofer, 1998).

Standardized refers to established rules for administration, scoring, and interpretation.

Use of Behavioral Samples has several implications:

Psychological test is not an exhaustive measurement of all possible behaviors that could be used
in measuring a particular attribute.
The quality of a test is largely determined by the representatives of this sample.

Type of Tests. Most psychological tests are sorted into 

                         three general categories.
1.  Tests of performance – subject performs some specific task such as writing an essay,
answering multiple choice items, or mentally rotating images presented on a computer   screen.

2. Behavior Observations – tests that involve observations of the subject’s behavior  within

a particular context.

3. Self-reports – subject describes her / his feelings, attitudes, beliefs, interests, and the  like.  

Validity and Reliability are one of the technical attributes of a good test.

Type of Tests. Most psychological tests are sorted into 

                         three general categories.
1.  Tests of performance – subject performs some specific task such as writing an essay,
answering multiple choice items, or mentally rotating images presented on a computer   screen.

2. Behavior Observations – tests that involve observations of the subject’s behavior  within

a particular context.

3. Self-reports – subject describes her / his feelings, attitudes, beliefs, interests, and the  like.  

Validity and Reliability are one of the technical attributes of a good test.

VALIDITY
Validity – 
is “the degree to which a certain inference from a test is appropriate or meaningful.” 
the extent to which a test does the job described of it, the evidence may be either empirical or
logical. 
is the extent to which a test measures what is supposed to measure

Types of Validity
Content – to compare whether the test items match the set of goals and objectives
Criterion : Concurrent – to determine whether there is a relationship between a test and an
immediate criterion measure
Criterion : Predictive – to determine whether there is a relationship between a test and a criterion
measure to be obtained.
Construct – to determine whether there a construct exists and to understand the traits or concepts
that make up the set of scores or items.

RELIABILITY
refers to the degree to which test scores are consistent, dependable, or repeatable; it is a function
of the degree to which test scores are free from errors of measurement. (Drummond,2000).
refers to the consistency of test scores obtained by the same persons when they are
reexamined with the same test on different occasions, or with different sets of equivalent items,
or under other variable examining conditions (Anastasi & Urbina, 1997)

Sampling Methods
When you conduct research about a group of people, it’s rarely possible to collect data
from every person in that group. Instead, you select a sample. The sample is the group of
individuals who will actually participate in the research.

To draw valid conclusions from your results, you have to carefully decide how you will select
a sample that is representative of the group as a whole. There are two types
of sampling methods:

Probability sampling involves random selection, allowing you to make statistical inferences

about the whole group.
Non-probability sampling involves non-random selection based on convenience or other
criteria, allowing you to easily collect initial data.
You should clearly explain how you selected your sample in the methodology section of your
paper or thesis.

Population vs Sample
The population is the entire group that you want to draw conclusions 
about.

The sample is the specific group of individuals that you will collect data 

from.

The population can be defined in terms of geographical location, age, income, 

and many other characteristics. It can be very broad or quite narrow: maybe 
you want to make inferences about the whole adult population of your 
country; maybe your research focuses on customers of a certain company, 
patients with a specific health condition, or students in a single school.

It is important to carefully define your target population according to the 

purpose and practicalities of your project.

If the population is very large, demographically mixed, and geographically 

dispersed, it might be difficult to gain access to a representative sample.
Sampling frame
The sampling frame is the actual list of individuals that the sample will be drawn from. 
Ideally, it should include the entire target population (and nobody who is not part of that
population).
Sample size
The number of individuals in your sample depends on the size of the population, and on how
precisely you want the results to represent the population as a whole.
You can use a sample size calculator to determine how big your sample should be. In general,
the larger the sample size, the more accurately and confidently you can make inferences about
the whole population.

Probability sampling methods

Probability sampling means that every member of the population has a chance of being selected.

It is mainly used in quantitative research. If you want to produce results that are representative
of the whole population, you need to use a probability sampling technique.

Four main types of probability sample.

1. Random sampling
2. Systematic sampling
3. Stratified sampling
4. Cluster sampling

1.  Random sampling
 
As the name suggests, random sampling literally means selection of the sample randomly from a
population, without any specific conditions. This may be done by selecting the sample from a
list, such as a directory, or physically at the location of the survey. If you want to ensure that
a particular household does not get selected more than once, you can remove it from the list. This
type of sampling is called simple random sampling without replacement. If you choose not to
remove duplicate households from the list, you would do a simple random sampling
with replacement. 

In a simple random sample, every member of the population has an equal chance of being

selected. Your sampling frame should include the whole population.
To conduct this type of sampling, you can use tools like random number generators or other
techniques that are based entirely on chance.

Example
You want to select a simple random sample of 100 employees of Company X. You assign a
number to every employee in the company database from 1 to 1000, and use a random number
generator to select 100 numbers.

2. Systematic sampling

Systematic sampling is similar to simple random sampling, but it is usually slightly easier to

conduct. Every member of the population is listed with a number, but instead of randomly
generating numbers, individuals are chosen at regular intervals.

Example

All employees of the company are listed in alphabetical order. From the first 10 numbers,
you randomly select a starting point: number 6. From number 6 onwards, every 10th person on
the list is selected (6, 16, 26, 36, and so on), and you end up with a sample of 100 people.

3. Stratified sampling
This sampling method is appropriate when the population has mixed characteristics, and you
want to ensure that every characteristic is proportionally represented in the sample.
You divide the population into subgroups (called strata) based on the relevant characteristic (e.g.
gender, age range, income bracket, job role).
From the overall proportions of the population, you calculate how many people should be
sampled from each subgroup. Then you use random or systematic sampling to select a sample
from each subgroup.

Example
The company has 800 female employees and 200 male employees. You want to ensure that the
sample reflects the gender balance of the company, so you sort the population into two
strata based on gender. Then you use random sampling on each group, selecting 80 women and
20 men, which gives you a representative sample of 100 people.
4. Cluster sampling

Cluster sampling also involves dividing the population into subgroups, but each subgroup should
have similar characteristics to the whole sample. Instead of sampling individuals from
each subgroup, you randomly select entire subgroups.

If it is practically possible, you might include every individual from each sampled cluster. If the
clusters themselves are large, you can also sample individuals from within each cluster using one
of the techniques above.

This method is good for dealing with large and dispersed populations, but there is more risk of
error in the sample, as there could be substantial differences between clusters. It’s difficult
to guarantee that the sampled clusters are really representative of the whole population.

Example

The company has offices in 10 cities across the country (all with roughly the same number of
employees in similar roles). You don’t have the capacity to travel to every office to collect your
data, so you use random sampling to select 3 offices – these are your clusters.

Non-probability sampling methods

In a non-probability sample, individuals are selected based on non-random criteria, and not every
individual has a chance of being included.
This type of sample is easier and cheaper to access, but it has a higher risk of sampling bias, and
you can’t use it to make valid statistical inferences about the whole population.

Non-probability sampling methods

Non-probability sampling techniques are often appropriate for exploratory
and qualitative research. 
In these types of research, the aim is not to test a hypothesis about a broad population, but to
develop an initial understanding of a small or under-researched population

1. Convenience sampling
A convenience sample simply includes the individuals who happen to be most accessible to the
researcher.

This is an easy and inexpensive way to gather initial data, but there is no way to tell if the sample
is representative of the population, so it can’t produce generalizable results.

Example

You are researching opinions about student support services in your university, so after each of
your classes, you ask your fellow students to complete a survey on the topic. This is a convenient
way to gather data, but as you only surveyed students taking the same classes as you at the same
level, the sample is not representative of all the students at your university

2. Voluntary response sampling

Similar to a convenience sample, a voluntary response sample is mainly based on ease of access.
Instead of the researcher choosing participants and directly contacting them, people volunteer
themselves (e.g. by responding to a public online survey).
Voluntary response samples are always at least somewhat biased, as some people will inherently
be more likely to volunteer than others.

Example
You send out the survey to all students at your university and a lot of students decide to complete
it. This can certainly give you some insight into the topic, but the people who responded are
more likely to be those who have strong opinions about the student support services, so you can’t
be sure that their opinions are representative of all students.

3. Purposive sampling

This type of sampling involves the researcher using their judgment to select a sample that is most
useful to the purposes of the research.

It is often used in qualitative research, where the researcher wants to gain detailed knowledge

about a specific phenomenon rather than make statistical inferences. An effective
purposive sample must have clear criteria and rationale for inclusion.

Example
You want to know more about the opinions and experiences of disabled students at your
university, so you purposefully select a number of students with different support needs in order
to gather a varied range of data on their experiences with student services.

4. Snowball sampling
If the population is hard to access, snowball sampling can be used to recruit participants via
other participants. The number of people you have access to “snowballs” as you get in contact
with more people.

Example
You are researching experiences of homelessness in your city. Since there is no list of all
homeless people in the city, probability sampling isn’t possible. You meet one person who
agrees to participate in the research, and she puts you in contact with other homeless people that
she knows in the area.

 Research Designs

The researcher not only selects a qualitative, quantitative, or mixed methods study to conduct;
the inquirer also decides on a type of study within these three choices. Research
designs are types of inquiry within qualitative, quantitative, and mixed methods approaches that
provide specific direction for procedures in a research design. Others have called them strategies
of inquiry (Denzin  & Lincoln, 2011).

Researchers that are done with high level of quality and integrity provide us with valuable
information about child and adolescent development. To be able to conduct quality research, it is
important that you know various research designs and the different data-gathering techniques
used by developmental researchers. 

Case Study

An in-depth look at an individual

It provides information about an individual’s fears, hopes, fantasies, traumatic experiences,

upbringing, family relationships, health, and anything that helps a psychologist understand that
person’s development (Santrock,2002)

Need to exercise caution when generalizing from the information, the subject of a case study is

unique, with a genetic make-up and experiences no one else shares, involves judgements of
unknown reliability, in that usually no check is made to see if other psychologists agree
with other observations (Santrock, 2002)

Correlational  Study
  A research design that determines associations 
 Useful because the more strongly two events are correlated, the more we can predict
one from the other.
 Because correlational research does not involve the manipulation of factors, it is not
a dependable way to isolate cause (Kantowitz, et. Al, 2001 cited by Santrock, 2002)

Experimental 
A research design that determines cause-and-effect relationships. The experimental method
involves manipulating one variable to determine if changes in one variable cause changes in
another variable. This method relies on controlled methods, random assignment and
the manipulation of variables to test a hypothesis. 

The only true reliable method of establishing cause and effect.

Experimental research is limited to what is observable, testable and 
                      manipulable.

 Failure to achieve randomization may limit the extent to which the study sample
is representative of the parent population and, with it, generalisability of the findings of
the study.

 Experimentation with humans is subject to a number of external influences that may

dilute the study results (Donnan, 2000). 

 A further limitation of experimental research is that subjects may change their behavior

or respond in a specific manner simply because of awareness of being observed.

 Naturalistic Observation 
      
       A research design that focuses on children’s experiences in natural settings.

    This does not involve any intervention or manipulation on the part of the researcher. This
technique involves observing subjects in their natural environment. This type of research is often
utilized in situations where conducting lab research is unrealistic, cost-prohibitive or would
unduly affect the subject’s’  behavior.

    One of the advantages of this type of research is that it allows the researcher to directly
observe the subject in a natural setting.
     
The disadvantages of naturalistic observation include the fact that it can be difficult to determine
the exact cause of a behavior and the experimenter cannot control outside variables.

Longitudinal
  This research design studies and follows through a single group over a period of time. The
same individuals are studied over a period of time, usually several years.

Allows them to record and monitor developmental trends.

They are expensive and time-consuming.

The longer the study lasts, the more subjects drop out – they move, get sick, lose interest, etc.
Subjects can bias the outcome of a study, because those who remain may be dissimilar to those
who drop out.

Cross – sectional
A research strategy in which individuals of different ages are compared at one time
Allows them to record and monitor developmental trends. The researcher does not have to wait
for the individuals to grow up or become older.
It gives no information about how individuals change or about the stability of
their characteristics (Santrock,2002)

Sequential 
This is the combined cross – sectional and longitudinal approaches to learn about life –
span development (Schaie, 1993 cited by Santrock, 2002). This starts with a cross-sectional
study that includes individuals of different ages. A number of months or years after the initial
assessment, the same individuals are tested again-this is the longitudinal aspect of the design. At
this later time, a new group of subjects is assessed of each grade level.
Allows them to record and monitor developmental trends. It provides information that is
impossible to obtain from cross-sectional or longitudinal approaches alone (Santrock, 2002).
It is complex, expensive, and time consuming.

ACTION RESEARCH 
Action research has been defined in a number of ways. Lewin (1946) introduced the term “action
research” which refers to it as “a way of generating knowledge about a social system while, at
the same time, attempting to change it.”  It may be conducted by a person in a particular situation
or the persons in that setting who have a major role in the design of the investigation. (Martella,
Nelson, Marchand – Martella, 1999)

Characteristics of Action Research

To solve problems in real life situations. The primary goal of Action Research is to  identify
problems and solutions to real life situations.

Action Research is context bound. The context is the real situation where the problem is
identified. The problem is solved first and then a theory is attempted to fit unlike most
quantitative research where the research attempt to verify a theory.

The goal is to generate data and make interpretation with the intent to make changes in their
practice.

Participants of action research are part of the process because they are part of the context being
investigated, therefore they are also being investigated. 

Dissemination of the solutions to problems that are obtained by action research have to be

disseminated so that others may benefit from it,

Process   Involved in Action Research

1. Diagnosis of the Problem

2. Planning Action Steps
3. Implementing
4. Evaluating of Outcome

Methods Used in Action Research

Quantitative Methods  - are most useful when comparisons are made between two large groups.
Data gathering techniques include survey, questionnaire, tests, rating scales, etc.

Qualitative Methods – participant observation, interviews, focus group discussion can be used

to gain specific information. Daily journal can also be written and analyzed. 
Parts of an Action Research Report
Title Page – should contain, the title, action researcher’s name, affiliation and other information
as may be required
Abstract – the abstract is a summary of the study and is typically a single paragraph of about
150 words. It contains information about the topic, what was done, and the findings.
Introduction – The rationale of the study which explains the problems situation looking for a
solution, the scientific question, hypotheses if any.
Method – this section describes the procedures used to conduct the investigation in
detail. Participants of the study, description of the setting, the dependent variable, measurement
tools, procedure, and analysis of data.
Results – the findings of the study are presented in this section. Graphs, tables can also be
presented to substantiate the findings

Discussion – this is the most important part of the paper where the findings are interpreted.
The situation discussed in the introduction must be revisited. Weaknesses of the current study
can be explained in the findings. Such weaknesses can be pointed out as things that can be
avoided in future studies, implications of the findings for future research should be explicated.
Conclusion – inferences about the findings are presented in this section.
Recommendations – implementation of the solution can be endorsed in this section as well as
other research possibilities as suggested by the study.
References – list of literature – books and related studies cited in the study are included.

Preparation of Action Research

Step 1 Identification

What is the area or situation you want to improve?

What evidence/s do you have to justify that the area of 
focus is a problem?
What factors serve as driving and / or restraining forces?

 Step 2 Collect Information

 What are your research question/s.

Do you have an intervention or an innovation?
Who are the members of the action research group?
What negotiations have to be undertaken?
What are your resources?

Step 3          Data Collection 

 State your question for data source 1, data source 2, data 

source 3..

 Step 4         Action Chart Template

 Summary of findings to research questions

 Recommended action targeted to findings
 Who is responsible for action?
 Who needs to be consulted or informed?
 Who will monitor or collect data?
 Timetable
 Resources

Data Gathering Techniques 

1. Observation – Observations can be made in either laboratories, or materialistic settings. In
naturalistic observation, behavior is observed in the real world like classrooms, home in
neighborhood.

2. Physiological Measures – Certain indicators of children’s development such as, among

others, heart rate, hormonal levels, bone growth, body weight, and brain activity are measured.

3. Standardized Tests – these are prepared test that assess individuals’ performance in different

domains. These tests are administered in a consistent manner.

4. Interviews and Questionnaires -  Involve asking the participants to provide information

about themselves based on the interview or questionnaire given by the researcher. 

Gathering of data may be conducted through a printed questionnaire, over the telephone, by mail,
in person, or on-line.
Information is obtained by utilizing standardized procedures so that every participant is asked the
same questions in the same manner. It entails asking participants for information in some
structured format.

5. Life-History Records – these are records of information about a lifetime chronology of

events and activities. They often involve a combination of data records on education, work,
family, and residence. These include public records or historical documents or interviews with
respondent.