Block-1 Introduction To Research Methods in Psychology
Block-1 Introduction To Research Methods in Psychology
RESEARCH
Structure
1.0 Introduction
1.1 Objectives
1.2 Definition and Meaning of Research
1.3 Criteria of Good Research
1.4 Objectives of Good Research
1.5 Qualities of a Good Research
1.6 Research Process: Basic Considerations
1.6.1 Context of Discovery
1.6.1.1 Role of Theories, Hypotheses and Paradigms in Psychological Research
1.6.1.2 Research Biases
1.6.2 Context of Justification
1.6.2.1 Scientific Attitudes and Values Associated with Research Process
1.6.2.2 Objectivity Safeguard in Research Process
1.7 Steps in Research Process
1.7.1 Step-I: Identification of the Problem
1.7.2 Step-II: Formulating a Hypothesis
1.7.3 Step-III: Identifying, Manipulating and Controlling Variables
1.7.4 Step-IV: Formulating a Research Design
1.7.4.1 Constructing Device for Observation and Measurement
1.7.4.2 Sample Selection and Data Collection
1.7.5 Step-V: Data Analysis and their Interpretation
1.7.6 Step-VI: Drawing Conclusion
1.7.7 Step-VII: Preparation of Report and Publication
1.8 Importance of Research in Psychology
1.9 Let Us Sum Up
1.10 Unit End Questions
1.11 Glossary
1.12 Suggested Readings and References
1.0 INTRODUCTION
Let us start our journey into the realm of human mind. A scientific quest for
understanding will be the foremost in our journey. We want to know why we
think, feel and behave as we do. What makes each of us different from all other
people? Why we do often behave as alike in some situations. Psychologists, as
scientists, answer these questions systematically, develop the principles to explain
them and use those principles to solve various problems. They are actively engaged
in process of doing research.
Research is a process through which new knowledge is discovered. A theory,
such as a theory of motivation, or development, or learning, for example, helps
5
Introduction to Research us to organise this new information into a coherent body, a set of related ideas
Methods in Psychology
that explain events that have occurred and predict events that may happen.
Conducting research requires to follow a sequence of steps. The exact sequence
and steps vary somewhat with the type of research. The steps vary slightly by
whether a study involves a quantitative or a qualitative approach and data.
This unit attempts to acquaint you with the nature and relevance of research.
This is followed by the qualities of a good research. Further, you will find the
process of psychological research within the context of discovery (getting and
idea) and context of justification (Testing hypothesis). This unit will provide a
quick glance at the whole process of research to acquaint you with the various
tasks you faced to undertake to carry out your research. It will give your some
idea of what the research journey involved. Finally, this unit will cover the total
spectrum of research and endeavor starting from the problem through to writing
a research report and its publication.
1.1 OBJECTIVES
After reading this unit, you will be able to:
• Define research;
• Describe the meaning of research;
• Explain the relevance of research;
• Describe the process of conducting research;
• Elucidate how to evaluate a research;
• Analyse how to maintain objectivity and minimize bias in psychological
research;
• Explain the role of theory, hypothesis and paradigm in psychological
research;
• Enlist the various steps needed for conducting a research;
• Explain how theoretical knowledge can be further applied to undertake a
research; and
• Analyse the importance of each steps involved in research process.
Thus, the term research refers to the systematic method consisting of enunciating
the problem, formulating a hypothesis, collecting the facts or data, analysing the
facts and reaching certain conclusions either in the form of solution (s) towards
the concerned problem or in certain generalisations for some theoretical
formulation.
The research report should be as much as possible frank enough to gauge effects
of the findings.
Data analysis in the research report should be adequate to reveal its significance
and the method of analysis employed be appropriate and
7
Introduction to Research 1) To gain familiarity with a phenomenon or to achieve new insights into it
Methods in Psychology
(studies with this object in view are termed as exploratory research studies);
2) To portray accurately the characteristics of a particular individual, situation
or group (studies with this object in view are known as descriptive research
studies);
3) To determine the frequency with which something occurs or with which it is
associated with something else (studies with this object in view are known
as diagnostic research studies)’
4) To test a hypothesis of a causal relationship between variables (such studies
are known as hypothesis- testing research studies/experimental studies).
Thus, research is the fountain of knowledge for the sake of knowledge and an
important source of providing guidelines for solving different business, personal,
profession governmental and social problems. It is a sort of formal training which
enables one to understand the new developments in one’s field in a better way.
CONSIDERATIONS
Before we examine what researchers have found in the major areas of psychology,
we need to identify the ways psychologists gather data about behaviour and mental
processes. Recall that psychology is the scientific study of behaviour and mental
functioning of individuals. It is scientific because it uses the principles and
practices of the scientific method.
Psychological research process can be divided into two major categories that
usually occur in sequence that is (i) getting an idea [context of discovery] and
then (ii) testing it (context of justification).
The value of a theory is often measured in terms of the new ideas, or hypotheses,
that can be derived from it and tested. A hypothesis is a tentative and testable
explanation of the relationship between two or more events or variables. A
variable is any factor that changes, or varies, in size or quality. To illustrate this
mood may be a variable, since people’s moods may vary from one situation to
another. Test performance is another variable, since a person’s score may vary
from one test to the next.
Finally, our understanding of a complex process is also aided by using the correct
paradigm. A paradigm is a model of the functions and interrelationships of a
process, a “way of thinking” about the world and how to study it. Entire field of
knowledge, including psychology, can change directions when new paradigm
challenges existing ones. When paradigms shift, revolutions of knowledge usually
9
Introduction to Research follow (Kuhn, 1970).Before a new theory, hypothesis, or paradigm makes a
Methods in Psychology
difference in science, it has to undergo an “ordeal of proof.” Most often this
happens when researchers publish (i.e. make public) their findings, and other
scholars investigate whether they find the same patterns in their own data. This
process of publication and communication moves scientific research into the
public eye, where ideas are tested and proven.
Finally, placebo biases operate when people strongly want to believe a treatment
is successful. For example, many people may claim to feel better after taking a
placebo such as a sugar pill. In those cases where the outcome involves a subjective
judgment about results, that is, how well a person feels well or whether the pain
has been reduced or relieved, the desire for a drug or therapeutic method to work
may be enough to achieve the desired result.
Since subjectivity must be minimized in the data collection and analysis phases
of scientific research, procedural safeguards are used to increase objectivity.
These safeguards begin with keeping complete records of observations and data
analyses in a form that other researchers can understand and evaluate. As a result,
most scientific reports are written in a similar form and published by organisations
of scientists. These reports communicate ideas to the entire scientific community
and open those ideas to criticism. A second safeguard is standardisation.
Standardisation means using uniform, consistent procedures in all phases of
data collection. All subjects should receive the same instructions and be treated
in the same way. By applying a standard treatment for all participants in the
course of study, researchers ensure they will have the same basic experience.A
third safeguard involves standardising the meaning of concepts, known as
operationalisation. An operational definition of a concept defines that concept
in terms of how it is measured or what operations produce it. Researchers must
also safeguard objectivity by avoiding bias. As explained earlier, bias from external
influences, personal beliefs, observers’ perspectives, and human expectations
can all distort data. Researchers use various control procedures to avoid such
biases and test hypotheses in ways that are fair and error-free.
Self Assessment Questions
1) An investigator comes up with a new idea or a different
way of thinking is known as context of discovery (True/False)
2) Use of uniform consistent procedure in all faces of data
collection is known as standardisation. (True/False)
3) Scientific knowledge is not based on empirical evidences. (True/False)
4) Psychologists should not maintain objectivity by avoiding
biases. (True/False)
5) Psychological researches should be replicable (True/False)
6) Operational definition of a concept is not necessary in
scientific research. (True/False)
Answers: (1) T, (2) T, (3) F, (4) F, (5) T, (6) F
11
Introduction to Research 1.7.1 Step-I: Identification of the Problem
Methods in Psychology
The first and most important step for identifying a problem is asking a question
or identifying a need that arises as a result of curiosity and to which it become
necessary to find an answer. The psychological studies are focused on one or
many of the following kinds of questions:
What are the events that cause or determine a given behaviour or response?
What is the nature of behaviour or action (i.e., its structure) and how it is linked
with other actions and behaviours?
What are the relationships of internal psychological processes with behavioural
phenomenon?
The research question determines the direction of study and researchers have to
struggle a lot in identifying and articulating the same. Essentially two steps are
involved in formulating the research problem, viz, understanding the problem
thoroughly, and rephrasing the same into meaningful terms. The main function
of formulating a research problem is to decide what you want to find out about.
It is extremely important to evaluate the research problem in the light of funds,
time and expertise and knowledge available at your disposal. It is equally
important to identify any gaps in your knowledge of relevant disciplines, such as
statistics required for analysis.
Once the question has been asked, that next step is to identify the factors that
have to be examined to answer the question. Such factors might range from the
most simple, such as a child’s age or socioeconomic status, to more complicated
measures such as the effects of violent cartoons on a child’s behaviour. The
factors may be age of the child, degree of violence in programs, emotional arousal,
facial expression, family communication patterns etc. Besides this, following
factors should be considered by the researcher for identifying in research problem:
Have not been investigated before;
Will contribute to the understanding of your question;
Are available to investigate;
Lead to another question!
For identifying a good solvable problem, the investigator undertakes the review
of literature. A body of prior work related to a research problem is referred to as
literature. Scientific research includes a review of the relevant literature. When a
researcher reviews the previous researches in related fields, he becomes familiar
with several known and unknowns. Therefore one obvious advantage of review
of the literature is that it helps to eliminate duplication of what has already been
done and provide guidance and suggestions for further research. The main purpose
of review of the literature is fourfold. First it gives an idea about the variables
which have been found to be conceptually and practically important and
unimportant in the related field. Thus the review of literature helps in discovering
and selecting variables relevant for the given study. Second the review of the
literature provides as estimate of the previous work and provides an opportunity
for the meaningful extension of the previous work.
Third, a review of the literature helps the researcher in systemising the expanding
12
and growing body of knowledge. This facilitates in drawing useful conclusions
regarding the variables under study and provides a meaningful way of their Basic Process/Concept in
Research
subsequent applications. Fourth, a review of the literature also helps in redefining
the variables and determining the meanings and relationships among them so
that the researcher can build up a case as well as a context for further investigation
that has merit and applicability. There are different sources of review of the
literature such as journals, books, abstracts, indexes and periodicals. If you are
not sure what journals and other resources to examine for research idea, you
should know that the computer search engine PsycINFO is a very effective tool
for going through the technical literature. The print companion to PsycINFO is
psychological abstracts and both of theses contain abstracts of articles from almost
all journals that publish psychological research. If you find an abstract of interest,
you can then read the entire articles for additional information.
Practical gains of psychological research are many, yet include discoveries such
as improved methods of treating psychologically disordered people, better designs
of vehicles to make them easier and safe to use, and new ways of enhancing the
performance and happiness of workers.
1.11 GLOSSARY
Empirical Investigation : Research that relies on sensory experience and
observation as research data.
Context of Discovery : Initial phase of research during which an
investigator comes up with a new idea or a
different way of thinking about phenomena.
Theory : A body of interrelated principles used to explain
or predict some psychological phenomenon.
Hypothesis : A tentative and testable explanation of the
relationship between two or more events or
variables.
Paradigm : A model of the functions and interrelationships
of a process; a “way of thinking” about the world
and how to study it.
Context of Justification : Second phase of research, in which results are
tested and prepared for useful communication
with the scientific community.
Standardisation : The use of uniform consistent procedures in all
phases of data collection.
Operational definition : Definition of a concept in terms of how the
concept is measured or what operations produce
it.
Placebo Control : A control strategy where researchers compare
those who received actual treatment with those
who received only attention or a “dummy drug.”
17
Introduction to Research Variable : something that can occur with different values
Methods in Psychology
and can be measured.
Independent Variable : a variable that represents the hypothesised
“cause” that is precisely controlled by the
experimenter and independent of what the
participant does.
Dependent Variable : a variable that represents the hypothesised
“effect “whose values ultimately depend on the
values of independent variable.
References
Grinnell, Richard Jr (ed.) 1988, Social Work Research and Evaluation (3rd edition)
Itasca, Illinois, F.E. Peacock Publishers.
Kuhn, T.S. (1970). The Structure of Scientific Revolutions (2nd edition) Chicago
: University of Chicago Press.
18
Basic Process/Concept in
UNIT 2 RELIABILITY AND VALIDITY Research
Structure
2.0 Introduction
2.1 Objectives
2.2 Reliability
2.3 Methods of Estimating Reliability
2.3.1 External Consistency Procedures
2.3.1.1 Test Re-tests Reliability
2.3.1.2 Parallel forms Reliability
2.3.2 Internal Consistency Procedures
2.3.2.1 Split Half Reliability
2.3.2.2 Kudar-Richardson Estimate of Reliability
2.3.2.3 Cronbach’s Alfa (α)
2.4 Comparison of Reliability Estimators
2.5 Validity
2.6 Types of Validity
2.6.1 Content Validity
2.6.2 Criterion Related Validity
2.6.2.1 Concurrent Validity
2.6.2.2 Predictive Validity
2.6.3 Construct Validity
2.6.3.1 Convergent Validity
2.6.3.2 Discriminant Validity
2.6.4 Face Validity
2.6.5 Internal Validity
2.6.5.1 Threats to Internal Validity
2.6.6 External Validity
2.6.6.1 Threats to External Validity
2.7 Let Us Sum Up
2.8 Unit End Questions
2.9 Glossary
2.10 Suggested Readings and References
2.0 INTRODUCTION
Most research is designed to draw the conclusion about the cause and effect
relationship among the variables. The goal of the research remains to develop a
theory that explains the relationship found among variables. This unit mainly
concerns about various problems that can threaten the reliability and validity of
conclusions drawn by the researcher.
There are two goals of research design;
1) Obtain information relevant to the purposes of the study.
2) Collect this information with maximal reliability and validity. 19
Introduction to Research How can a researcher be sure that the data gathering instrument being used will
Methods in Psychology
measure what it is supposed to measure and will do this in a consistent manner?
This is a question that can only be answered by examining the definitions for
and methods of establishing the validity and reliability of a research instrument.
Reliability and validity are central issues in all measurement. Both concern
connecting measures to constructs. Reliability and validity are salient because
constructs are often ambiguous, diffused and not directly observable. Perfect
reliability and validity are virtually very difficult to achieve. These two very
important aspects of research design will be discussed in this unit. All researchers
want their measures to be reliable and valid. Both ideas help to establish the
truthfulness, credibility, or believability of findings. This unit will be discussed
in two parts. First part covers the concept of reliability and the definitions of
reliability. This is followed by various methods of establishing reliability of a
research instrument of this unit. Second part of this unit discusses the concept of
validity in research. You will familiarise with the various types of validity. Finally,
some problems that constitute threats to validity are described.
2.1 OBJECTIVES
After reading this unit, you will be able to:
• Define reliability;
• Describe the various methods of calculating reliability;
• Explain how test retest reliability is accessed;
• Differentiate between tests of reliability;
• Define validity;
• Describe various methods of validity;
• Identify the problems that constitute threats to internal external validity;
and
• Differentiate between internal and external validity.
2.2 RELIABILITY
Meaning of Reliability
The idea behind reliability is that any significant results must be repeatable.
Other researchers must be able to perform exactly the same experiment, under
same conditions and generate the same results. This will vindicate the findings
and ensure that all researchers will accept the hypothesis. Without this replication
of statistically significant results, experiment and research have not fulfilled all
of the requirements of testability. This prerequisite is essential to a hypothesis
establishing itself as an accepted scientific truth. For example, if you are
performing a time critical experiment, you will be using some type of stopwatch.
Generally, it is reasonable to assume that the instruments are reliable and will
keep true and accurate time. However, scientists take measurements many times,
to minimize the chances of malfunction and maintain validity and reliability. At
the other extreme, any experiment that uses human judgment is always going to
come under question. Human judgment can vary as individual observer may rate
20
things differently depending upon time of day and current mood. This means Reliability and Validity
(External and Internal)
that such experiments are more difficult to repeat and are inherently less reliable.
Reliability is a necessary ingredient for determining the overall validity of a
scientific experiment and enhancing the strength of the results.
According to Stodola and Stordahl (1972), the reliability of a test can be defined
as the correlation between two or more sets of scores of equivalent tests from the
same group of individuals.
The reliability of test is also defined from another angle. Whenever we measure
something, measurement involves some kind of measure. Error of measurement
is generally between true scores and the observed score. However, in psychological
term, word error does not imply the mistake has been made. In other words,
error in psychological testing implies that there is always some inaccuracy in
measurement. Hence, goal of psychological measurement remains to find out
the magnitude of such error and develop ways to minimize them.
The most frequently used method to find the reliability of a test is by repeating
the same test on same sample, on two different time periods. The reliability
coefficient in this case would be the correlation between the score obtained by
the same person on two administrations of the test.
21
Introduction to Research Test-Retest reliability is estimated, when same test is administered on same
Methods in Psychology
sample. Therefore, if refers to the consistency of a test among on two different
time periods different administrations. The assumption behind this approach is
that there will be no substantial changes in the measurement of the construct in
question, upon administration on separate occasions. The time gap that is given
between measures is of critical value, the shorter the time gap, higher the
correlation value and vice versa. If the test is reliable, the scores that are attained
on first administration should be more or less equal to those obtained on second
time also. The relationship between the two administrations should be highly
positive.
Parallel forms reliability compares two equivalent forms of a test that measure
the same attribute. The two forms use different items. However, the rules used to
select items of a particular difficulty level are the same. When two forms of the
test are available, one can compare performance on one form versus the other.
Sometimes the two forms are administered to the same group of people on the
same day.
The method of parallel forms provides one of the most rigorous assessments of
reliability commonly in use. Unfortunately the use of parallel forms occurs in
practice less often than is desirable. Often test developers find it burdensome to
22
develop two forms of the same test, and practical constraints make it difficult to Reliability and Validity
(External and Internal)
retest the same group of individuals. Instead many test developers prefer to base
their estimate or reliability on a single form of a test.
In practice, psychologists do not always have two forms of a test. More often
they have only one test form and must estimate the reliability for this single
group of items. You can asses the different sources of variation within a single
test in many ways. One method is to evaluate the internal consistency of the test
by dividing it into subcomponents.
N Σ σ2j
rα = ——— 1 – ———
N–1 σ2
In a situation that involves use of two forms as alternate measure of the same
thing, parallel forms estimator is best suited. However, this and the internal
consistency measures of reliability have constrains, i.e. one has to have multiple
items engineered to measure same construct.
Cronbach’s Alpha is useful in case, where lots of items are present. The test-
retest reliability is mostly employed in case of experimental and quasi-
experimental designs. This also depends upon string of availability of a control
group that is measured on two different occasions and until post-test is done,
one does not have information about reliability. Accordingly, each one of the
above mentioned estimators will give a different value for reliability. Generally,
test-retest and inter-rater reliability estimates will be lower in value as compared
to parallel forms and internal consistency due to involvement in measurement at
different times or with different raters.
24
Reliability and Validity
Self Assessment Questions (External and Internal)
1) Internal Consistency Concerns whether the various items on
a test are measure the same thing. T/F
2) Memory effect / carry over effect is possible in parallel
form method. T/F
3) K.R. Formula is applied in which each test item is scored
0 or 1. T/F
4) Scores from the two halves of a test are correlated with
one another in split half reliability. T/F
5) Spearman Brown formula is used for adjusting split half
correlation T/F
Answer: 1) T, 2) F, 3) T, 4) T, 5) T
2.5 VALIDITY
As you know that the merit of the psychological test is determine first by its
reliability but then ultimately by its validity. Validity refers to the degree to which
a test measures, what it claims to measure. It is very necessary for a test to be
valid for its proper administration and interpretation.
According to Standard for Educational and Psychological testing (AERA, APA
& NCME 1985, 1999); a test is valid to the extent that inferences drawn from it
are appropriate, meaningful and useful.
According to Cronbach (1951) validity is the extent to which a test measures
what it purports to measure.
According to Freeman (1971) an index of validity shows the degree to which a
test measures what it purports to measure when compared with accepted criteria.
According to Anastasi (1988) the validity of a test concerns what the test measures
and how well it does so.
The above definitions pointed out that for determining the validity of the test,
the test must be compared with some ideal independent measures or criteria.
The correlation coefficients computed between the test and an ideal criterion is
known as the validity coefficients. Independent criteria refer to some measure of
the trait or group of the traits (out side the test) that the test itself claims to
measure.
A test has content validity inbuilt in it. Items are selected in accordance with
their compliance with the requirements of the test after a careful examination of
the subject area.
26
ii) A test that has construct validity should measure what it intends to measure Reliability and Validity
(External and Internal)
but not measure theoretically unrelated constructs. For example, a test of
musical aptitude should not require too much reading ability.
iii) A test should prove useful in predicting results related to the theoretical
concepts it is measuring. For example, a test of musical ability should
predict who will benefit from taking music lessons, should differentiate
groups who have chosen music as a career from those who haven’t should
relate to other tests of musical ability and so on.
There are two types of construct validity— ‘convergent validity’ and ‘divergent
validity’ (or discriminant validity).
These include (i) confounding, (ii) selection bias, (iii) history, (iv) maturation,
(v) repeated testing ,(vi) instrument change, (vii) regression toward the mean,
(viii) mortality, (ix) diffusion, (x) compensatory rivalry, (xi) experimenter bias.
i) Confounding: Confounding error that occurs when the effects of two
variables in an experiment cannot be separated, resulting in a confused
interpretation of the results. Confounding is one of the biggest threat to
validity in experimentation. The problem of confounding is particurly acute
in research in which the experimenter cannot control the independent
variable. When participants are selected according to presence or absence
of a condition, subject variable can affect the results. Where a false
relationship cannot be avoided, a rival hypothesis may be developed to the
original cause and inference hypotheses.
ii) Selection bias: Any bias in selecting a group can undermine internal validity.
Selection bias indicates the problem that occurs as a result of its existence
at the pre-test differences between groups, may interact with the independent
variable and thus influence the observed outcome and creates problems;
examples would be gender, personality, mental capabilities, and physical
abilities, motivation level and willingness to participate.
If at the time of selection, an uneven number of subjects to be tested have
similar subject-related variables, there could be a threat to the internal
validity, for instance, if two groups are formed i.e. experimental and control
group, the subjects in the two groups are different with regards to independent
variable but alike in one or more subject related variables. It would then be
difficult for the researcher to identify if the difference between in the groups
is the result of independent variable or subject related variable as well as
randomisation of group assignment. It is not possible always as some
significant variables may go unnoticed.
iii) History: Events outside the experiment or between repeated measures of
dependent variables may influence participants’ responses, attitudes and
behaviour during process of experiment, like; natural disasters, political changes
etc. In this condition, it becomes impossible to determine whether change in
dependent variable is caused by independent variable or historical event.
iv) Maturation: Usually, it happens that subjects change during the course of
an experiment or between measurements. For instance, in longitudinal studies
young kids might grow up as a result of their experience, abilities or attitudes
which are intended to be measured. Permanent changes [such as physical
growth] and temporary changes [like fatigue and illness] may alter the way
a subject would react to the independent variable. Thus, researcher may
have trouble in ascertaining if the difference is caused by time or other
variables.
v) Repeated testing: Participants may be driven to bias owing to repeated testing.
Participants may remember correct answers or may be conditioned as a
result of incessant administration of the test. Moreover, it also causes
possibility of threat to internal validity.
vi) Instrument change: If any instrument is replaced/changed during process of
experiment, then it may affect the internal validity as alternative explanation
28 easily available.
vii) Regression toward the mean: During the experiment, if subjects are selected Reliability and Validity
(External and Internal)
on the basis of extreme scores, then there are chances of occurrence of such
an error. For example, when subjects with minimum mathematical abilities
are chosen, at the end of the study if there is any improvement chances are
that it would be due to regression towards the mean and not due to
effectiveness of the course.
viii) Mortality: It should be kept in mind that there may be some participants
who may have dropped out of the study before its completion. If dropping
out of participants leads to relevant bias between groups, alternative
explanation is possible that account for the observed differences.
ix) Diffusion: It might be observed that there will be a lack of differences between
experimental and control groups if treatment effects spread from treatment
groups to control groups. This, however, does not mean that, independent
variable will have no effect or that there would not be a no relationship
between dependent and independent variable.
x) Compensatory rivalry/resentful demoralisation: There will be a change in
the behaviour of the subject if the control groups alter as a result of the
study. For instance, control group participants may work extra hard to see
that expected superiority of the experimental group is not demonstrated.
Again, this does not imply that the independent variable created no effect or
that there would be no relationship between dependent and independent
variable. Vice-versa, changes in the dependent variable may only be effected
due to a demoralised control group, working less hard or demotivated.
xi) Experimenter bias: Experimenter bias happens while experimenters, without
any intention or reluctance, behave differently to the participants of control
and experimental groups, that in turn, affect the results of the experiment.
Experimental bias can be reduced by keeping the experimenter from knowing
the condition in the experiment or its purpose and by standardising the
procedure as much as possible.
External validity lacks from the fact that experiments using human participants
often employ small samples collected from a particular geographic location or
with idiosyncratic features (e.g. volunteers). Because of this, it cannot be made
sure that the conclusions drawn about cause-effect-relationships are actually
applicable to the people in other geographic locations or in the absence of these
features.
The validity of a test is the degree to which it measures what it claims to measure.
A test is valid to the extent that inferences made from it are appropriate,
meaningful, and useful. There are various kinds of validity – content validity
30
determine by the degree to which the question, task or items on a test are Reliability and Validity
(External and Internal)
representative of the universe of behaviour the test was designed to sample. A
test has face validity if it looks valid to test users, examiners, and especially the
examinees. Criterion-related validity is demonstrated when a test is effective in
predicting performance on an appropriate outcome measure. An investigation
has internal validity is a cause-effect relationship actually exists between the
independent and dependent variables. Confounding occurs when the effects of
two independent variables in an experiment cannot be separately evaluated.
External validity concerns whether the results of the research can be generalised
to another situation: different subjects, settings, times, and so forth. Treats to the
internal validity of an experiment include events outside the laboratory,
maturation, effects of testing, regression effect, selection and mortality. Threats
to external validity include problems arising from generalising to other subjects,
other times, or other settings. Experimenter bias can be reduced by keeping the
experiment from knowing he conditions in the experiment or its purpose and by
standardising procedure as much as possible.
2.9 GLOSSARY
Concurrent validity : a type of criterion-related validity in which the
criterion measures are obtained at approximately
the same time as the test scores.
Confounding : error that occurs when the effects of two variables
in an experiment cannot be separated, resulting in
a confused interpretation of he results.
31
Introduction to Research Construct : a theoretical, tangible quality or trait in which
Methods in Psychology
individuals differ.
Construct validity : the property of a test that the measurements
actually measure the constructs they are designed
to measure, but no others.
Content validity : idea that a test should sample the range of
behaviour represented by the theoretical concept
being tested.
Criterion validity : idea that a test should correlate with other
measures of the same theoretical construct.
Cronback alpha : an index of reliability that may be thought of as
the mean of all possible split-half co-efficient,
corrected by the Spearman-Brown formula.
External validity : how well the findings of an experiment generalise
to other situations or populations.
Inter observer reliability : the typical degree of agreement between scores.
Internal consistency: the degree to which the
various items on a test are measures of the same
thing.
Internal validity : extent to which a study provides evidence of a
cause-effect relationship betwen the independent
and dependent variables.
Kuder-Richardson : an index of reliability that is relevant to the special
formula 20 case where each test item is scored 0 or 1 (example,
right or wrong)
Maturation : a source of error in an experiment related to the
amount of time between measurements.
Regression effect : regression effect tendency of subjects with
extreme score on a first measure to score closer to
the mean on a second testing.
Reliability : the property of consistency of a measurement that
gives the same result on different occasions.
Spearman-Brown formula: a formula for adjusting split-half correlations so
that they reflect the full length of a scale.
Split-half reliability : a form of reliability in which scores from the two
halves of a test (e.g. even items versus odd items)
are correlated with one another; the correlation is
then adjusted for test length.
Test – retest reliability : the degree to which the same test score would be
obtained on another occasion.
Validity : of a measurement the property of a measurement
that tests what it is supposed to test.
32
Reliability and Validity
2.10 SUGGESTED READINGS AND REFERENCES (External and Internal)
References
Guilford, J.P. (1954). Psychometric Methods. New Delhi: Tata McGraw Hill.
Kaiser, H.F., & Michael, W.B. (1975). Domain Validity and Gernalisability.
Educational and Psychological Measurement, 35, 31-35.
McBurney, D.H. & White, T. L. (2007) Research Methods, New Delhi; Akash
Press.
Novick, M.R., & Lewis, C. (1967). Coefficient Alpha and the Reliability of
Composite Measurements. Psychometrika, 32, 1-13.
33
Introduction to Research
Methods in Psychology UNIT 3 VARIABLES AND CONSTRUCTS
Structure
3.0 Introduction
3.1 Objectives
3.2 Meaning of Variables
3.3 Types of Variables
3.3.1 Stimulus, Organism and Response Variables
3.3.2 Independent and Dependent Variables
3.3.3 Extraneous and Confounded Variables
3.3.4 Active and Attribute Variables
3.3.5 Quantitative and Categorical Variables
3.3.6 Continuous Variables and Discrete Variables
3.4 Constructs
3.5 Types of Constructs
3.5.1 Intervening Variables
3.5.2 Hypothetical Constructs
3.6 Let Us Sum Up
3.7 Unit End Questions
3.8 Glossary
3.9 Suggested Readings and References
3.0 INTRODUCTION
In the process of formulating a research problem there are two important
considerations; the use of constructs/concepts and the construction of hypotheses.
Constructs/concepts are highly subjective as their understanding varies from
person to person and therefore, as such, may not be measurable. In a research
study, it is important that the concepts used should be operationalised in
measurable terms so that the extent of variation in respondents understanding is
reduced if not eliminated. Knowledge about constructs and variables are very
important to understand conceptual clarity and quantitative accuracy as they
provide the ‘fine tuning’ to research.
This unit attempts to acquaint you with the term variables and constructs which
are used by the psychologists in gaining knowledge about the behaviour and
mental processes. It begins with definition of variables then you will find the
details about the types of variables along with the examples. Further, you will be
exposed to the nature of the scientific concept or construct and the way in which
behavioural scientist travel from the construct level to observation level. Finally,
types of constructs are described.
3.1 OBJECTIVES
After reading this unit, you will be able to:
• Define the term variable and construct;
34
• Discuss the types of variable and constructs; Variables and Constructs
35
Introduction to Research Suppose a theory reveals that increasing anxiety will increase the affiliation
Methods in Psychology
motive. To test out this theory, you may take the theoretical concepts of anxiety
and affiliation motive and relate them to variables by means of operational
definitions. The theory is an abstract statement. .For example, anxiety can be
measured by the Anxiety Scale and affiliation by how close people sit to each
other in the experiment. These two measures constitute the variables of the study.
The scores on the variables of anxiety and distance apart are related to one another
as test of the hypothesis. The relationship between the variable is taken as
providing support for or against the particular theory that generated the
experiment.
Dependent upon the mode of manipulation, some experts have tried to divide
the independent variable into ‘Type E’ independent variable and ‘Type S’
independent variable (D’Amato, 1970). Type E independent variable is one of
which is directly or experimentally manipulated by the experimental and type S
36
independent variable is one which is manipulated through the process of selection Variables and Constructs
only. For example the experimenter wants to study the effect of noise upon the
task performance in an industry. Here the IV (Independent Variable) is the noise
and the DV(Dependent Variable) is the task performance. He may manipulate
the noise by dividing into three categories — continuous noise, intermittent noise
and no noise and examine its effect upon the task performance. Here the noise is
being directly manipulated by the experimenter and hence, it constitutes the
example of Type-E independent variable. Suppose, for the time being, that the
experimenter is interested in answering the question: Is the rate of production
dependent upon the age of the workers? Age is here the independent variable.
For investigating this problem, the experimenter will have to select groups of
workers on the basis of their age in a way by which he can get an appropriate
representation from different age groups ranging from say, 16 to 55 years.
Subsequently, he will compare the rate of production obtained by each age group
and finally, conclude whether or not age is a factor in enhancement of the
performance .Hence this constitutes the examples of S-independent variables.
Here, you can examine the relationship between independent and dependent
variables. The relationship is that of dependence. One variable depends upon
the other. Suppose you find a relationship between meaningfulness of the learning
material and speed of learning. Speed of learning then depends upon
meaningfulness; the greater the meaningfulness, the faster the learning. The speed
of learning is, therefore, called dependent variable; meaningfulness is independent
variable. Similarly, rest between work periods is independent variables; output
of work is dependent variable. Sudden noise is independent variable; change in
breathing is dependent variable. In an experiment one discovers and confirms a
relationship between an independent variable and a dependent variable.
Self Assessment Questions
Identify the independent and dependent variables?
1) As the number of trials increases error decreases.
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
37
Introduction to Research
Methods in Psychology 2) Intensity of lights affects the task performance.
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
3) Healthy parent child relationship facilitated emotional adjustment of
the children.
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
4) Teaching methods affects learning.
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
38
Confounding variables is one that varies with the independent variable. While Variables and Constructs
doing a study if we are not careful then two variables may get combined so that
the effect of one cannot be separated from the effect of other. This is known as
confounding. For instance, if you conducted a study of the effect of television
viewing on perception of violence and the experimental group contained only
adolescents, whereas the control group only adults, the age of participants would
be confounded with the independent variable under study. Confounding makes
the conclusions of the study doubtful. It is ,therefore, necessary that effort should
be made to unconfound the variables.
To explain these variables let us take one example. Suppose you want to study
the relationship between smoking and coronary heart disease. You assume that
affecting this relationship, such as a number of cigarettes or the amount of tobacco
smoked every day; the duration of smoking; the age of the smoker; dietary habits;
and the amount of exercise undertaken by the individuals. All of these factors
may affect the extent to which smoking might cause coronary heart disease.
These variables may either increase or decrease the magnitude of the relationship.
In this example, the extent of smoking is the independent variable, coronary
heart disease is the dependent variable and all the variables that might affect this
relationship, either positively or negatively, are extraneous variables.
Extraneous variables
When a variable can have only one value or category, for example taxi, tree and
water, it is known as a constant variables. When a variable can have only two
categories as in yes/no, good/bad and rich/poor, it is known as dichotomous
variables. When variables can be divided into more than two categories, for
example: religion (Christian, Muslim, Hindu); political parties (Labor, Liberal,
Democrat); and attitudes (strongly favorable, favorable, uncertain, unfavorable,
strongly unfavorable), it s called a polytomous variable.
Self Assessment Questions
1) The event manipulated or altered by the experimenter is termed as ......
........................................................
2) The ....................... is the response that the person or animal makes.
3) Any and all other variables that may mask the relationship between
independent variable and dependent variables are known as ...............
4) Manipulated variables are called ............... and measures variable are
called .......................................
5) .............................. is one that varies in amount whereas ....................
varies in kind.
Answer: (1) Independent variables (2) Dependent variables (3) Extraneous
variables (4) Active variables, Attribute variables (5) Quantitative
variables & Categorical variables.
Activity -2
Identity Types of Variables
A researcher wants to administer an intelligence test to 30 college students.
After collecting information on subjects’ age, sex, height, weight, political
preference, career goals, and socioeconomic status, the researcher administers
and attitude survey on current issues to all 30 subjects.
Required:
Identify examples of the following types of variables in the paragraph and
the scales by which they would be measured:
a) discrete
b) continuous
c) categorical
d) quantitative
3.4 CONSTRUCTS
The terms ‘concept’ and ‘construct’ have similar meanings. Yet, there is an
important distinction. A concept may be defined as any describable regularity of
real or imagined events or objects (Bourne, Ekstrand, & Dominowski, 1971). A
concept is a set of features connected by some rule (Hulse, Egeth, Deese 1980)
Concepts are building block of thinking. They allow us to organise knowledge
in systematic ways. Concept represents objectives of activities, ideas or living
organism. Concept also represents properties, abstraction and relations between
the features. For example – ‘achievement’. It is an abstraction formed from the
observation of certain behaviours of children. These behaviours are associated
with the mastery or “learning” of school tasks – reading words, doing arithmetic
problems, drawing pictures, and so on. The various observed behaviours are put
together and expressed in a word – ‘achievement’. ‘Intelligence’, ‘aggressiveness,’
‘conformity’, and ‘honesty’ are all concepts used to express varieties of human
behaviour of interest to behavioural scientists. Researcher often invents or constructs
new concepts for special scientific purposes; such concepts are called constructs.
Thus, you may well explain construct as concept. It has the added meaning i.e
invented or adopted for a special scientific purpose. For example, “Intelligence”
is a concept, an abstraction from intelligent and nonintelligent behaviours. But,
as a scientific construct, “intelligence” means both more and less than a concept.
It means that scientists consciously and systematically use it in two ways. One, it
41
Introduction to Research enters into theoretical schemes and related in various ways to other constructs.
Methods in Psychology
In this sense, school achievement may also be, in part, a function of intelligence
and motivation. Two, “intelligence” is so defined and specified that it can be
observed and measured. We can make observations of the intelligence of children
by administering X intelligence test to them, or we can ask teachers to tell us the
relative degrees of intelligence of their pupils.
Constructs are created and used for a wide variety of reasons, but generally have
two common characteristics. First, the construct is a part of a theoretical
framework and is related in various ways to other constructs. Second, a construct
usually operationally defined so as to allow its observation and measurement.
An example of a commonly employed psychological construct would be
reinforcement. On a theoretical level, reinforcement can be, and is, related to
other constructs such as drive, motivation, association, and habit strength.
K = knowledge
AC = amount of conditioning
IQ = intelligence
AC could be defined as the number of reinforced trials a person receives and IQ
as that person’s score on a standard intelligence test. K could be defined as being
a function of AC and IQ. Therefore, AC and IQ are hypothetical constructs (they
describe something real and are defined directly by the operations that established
them or by which they were measured). On the other hand, K is an intervening
variables (it has no meaning of its own, but only summarizes or stands for other
constructs). However, if K were defined as the number of correct solutions a
person achieved on the “knowledge test,” then K would also be a hypothetical
construct.
Self Assessment Questions
Which of the following constructs (as used) would be classified as an
Intervening Variable(IV) or a Hypothetical Construct (HC)?
IV HC
1) Thinking is the mental activity leading to problem ( ) ( )
solving.
2) Arousal is he increase in neural activity in the lower ( ) ( )
brain stem following stimulation.
3) A reinforcement is something that makes you want to ( ) ( )
repeat the behaviour that was rewarded.
4) The id is the deepest part of the psyche and ( ) ( )
motivates our “base” desires.
43
Introduction to Research
Methods in Psychology 3.7 UNIT END QUESTIONS
1) Defined variable and discuss the various kinds of variable.
2) Explain Intervening variables and Hypothetical constructs in your own words.
3) Differentiate between variables and constructs. Giving suitable example
elucidate the different types of variables.
4) Write short notes on any two:
i) Independent & dependent variables.
ii) Quantitative & categorical variables.
iii) Active & attribute variables.
5) Explain the nature of extraneous variable with suitable examples.
3.8 GLOSSARY
Variable : A variable is a property that taken as different
values.
Independent variables : The condition manipulated or selected by the
experimenter to determine its effect on
behaviour.
Dependent variables : A measure of the subject’s behaviour that
reflects that independent variable’s effects.
Quantitative variable : One that varies in amount.
Categorical variable : One that varies in kind.
Continuous variable : One that falls along a continuum and is not
lifted to a certain number of values.
Discrete variable : One that that falls into separate bins with no
intermediate values possible.
Active variables : Manipulated variables are active variables.
Attribute variables : Measured variables are attribute variables.
Constructs : Is a concept, used for scientific purpose, is a
part of theoretical framework.
Intervening variables : Is a construct which is utilised as a summary
term for a group of other constructs.
Hypothetical constructs : Is a theoretical term which is employed to
describe something real.
Bourne, L.E., Ekstrand, B., & Dominowski, R. 1971. The Psychology of Thinking.
Englewood Cliffs, N.J.: Prentice-Hall.
Postman , L. and Egan, J.P. (1949). Experimental Psychology. New York; Harper
& Row
45
Introduction to Research
Methods in Psychology UNIT 4 HYPOTHESIS FORMULATION AND
SAMPLING
Structure
4.0 Introduction
4.1 Objectives
4.2 Meaning and Characteristics of Hypothesis
4.3 Formulation of Hypothesis
4.4 Possible Difficulties in Formulation of a Good Hypothesis
4.5 Types of Hypotheses
4.5.1 Null Hypothesis
4.5.2 Alternative Hypothesis
4.6 Errors in Testing a Hypothesis
4.7 Importance of Hypothesis Formulation
4.8 Sampling
4.8.1 Definition of Sampling
4.8.2 Sampling Terminology
4.8.3 Purpose of Sampling
4.9 Sampling Methods
4.9.1 Non Probability Sampling
4.9.2 Probability Sampling
4.10 Importance of Sampling
4.11 Let Us Sum Up
4.12 Unit and Questions
4.13 Glossary
4.14 Suggested Readings and References
4.0 INTRODUCTION
Scientific process or all empirical sciences are recognised by two inter-related
concepts, namely; (a) context of discovery (getting an idea) and (b) context of
justification (testing and results). Hypotheses are the mechanism and container
of knowledge moving from the unknown to known. These elements form
techniques and testing ground for scientific discovery. Hypotheses are tentative
explanation and potential answer to a problem. Hypothesis gives the direction
and helps the researcher interpret data. In this unit, you will be familiarised with
the term hypothesis and its characteristics. It is, then, followed by the hypothesis
formulation and types of hypothesis. Errors in hypothesis testing are also
highlighted.
Further, In order to test the hypothesis, researcher rarely collects data on entire
population owing to high cost and dynamic nature of the individual in population.
Therefore, they collect data from a subset of individual – a sample - and make
the inferences about entire population. This leads us to what we should know
about the population and sample. So, researcher plans sample design and uses
46
various method of sampling. This unit will acquaint you with the meaning of Hypothesis Formulation
and Sampling
sampling and basic terminology which is used in sampling design.
Now, it will move to purpose of sampling. And finally, various probability and
non-probability sampling methods along with advantages and disadvantages are
described.
4.1 OBJECTIVES
After reading this unit, you will be able to:
• Define and describe hypothesis and its characteristics;
• explain formulation of hypothesis;
• Enumerate the possible difficulties in formulating hypothesis;
• Explain types of hypotheses;
• Identify in hypotheses testing;
• Define sampling;
• Explain the purpose of sampling; and
• Analyse various probability and non-probability sampling methods.
By stating a specific hypothesis, the researcher narrows the focus of the data
collection effort and is able to design a data collection procedure which is aimed
at testing the plausibility of the hypothesis as a possible statement of the
relationship between the terms of the research problem.
It is, therefore, always useful to have a clear idea and vision about the hypothesis.
It is essential for the research question as the researcher intents to verify, as it
will direct and greatly help to interpretation of the results.
Hypothesis plays a key role in formulating and guiding any study. The hypotheses
are generally derived from earlier research findings, existing theories and personal
observations and experience. For instance, you are interested in knowing the
effect of reward on learning. You have analysed the past research and found that
two variables are positively related. You need to convert this idea in terms of a
testable statement. At this point you may develop the following hypothesis.
Those who are rewarded shall require lesser number of trails to learn the lesson
than those who are not rewarded.
A researcher should consider certain points while formulating a hypothesis:
i) Expected relationship or differences between the variables.
ii) Operational definition of variable.
iii) Hypotheses are formulated following the review of literature
The literature leads a researcher to expect a certain relationship.
Hypotheses are the statement that is assumed to be true for the purpose of testing
its validity.
49
Introduction to Research
Methods in Psychology 4.5 TYPES OF HYPOTHESES
As explained earlier, any assumption that you seek to validate through
investigation is called hypotheses. Hence theoretically, there should be one type
of hypotheses on the basis of the investigation that is, research hypothesis.
However, because of the conventions in scientific enquiries and wording used in
the constructions of the hypothesis, Hypotheses can be classified into several
types, like; universal hypotheses, existential hypotheses, conceptual hypotheses
etc. Broadly, there are two categories of the hypothesis:
i) Null hypothesis
ii) Alternative hypothesis
Researchers usually can not make direct observation of every individual in the
population under study. Instead, they collect data from a subset of individuals- a
sample – and use those observations to make inferences about the entire
population.
Sampling unit: Each individual or case that becomes the basis for selecting a
sample is called sampling unit or sampling elements.
Sampling frame: The list of people from which the sample is taken. It should be
comprehensive, complete and up-to-date. Examples of sampling frame: Electoral
Register; Postcode Address File; telephone book.
Self Assessment Questions (Fill in the blanks)
1) Any identifiable and well specified group of individual is known as
.............................................
2) List of all the units of the population is called ............................
3) Purposes of sampling is to derive the desired information about the
population at the minimum ..................... and maximum ....................
4) The way the researcher selects the sample is known as .....................
5) ........................... is the miniature picture of entire group.
Answers: (1) population, (2) sampling frame, (3) cost, reliability,
(4) sampling design, (5) sample.
For example, an investigator may take student of class X into research plan
because the class teacher of the class happens to be his / her friend. This illustrates
accidental or convenience sampling.
Quota sampling ensures that some differences are in the sample. In haphazard
sampling, all those interviewed might be of the same age, sex, or background.
But, once the quota sampler fixes the categories and number of cases in each
category, he or she uses haphazard or convenience sampling. Nothing prevents
the researcher from selecting people who act friendly or who want to interviewed.
Quota sampling methods are not appropriate when the interviewers choose who
they like (within above criteria) and may therefore select those who are easiest
to interview, so, sampling bias can take place. Because not using the random
method, it is impossible to estimate the accuracy. Despite these limitations, quota
sampling is a popular method among non-probability methods of sampling,
because it enables the researcher to introduce a few controls into his research
plan and this methods of sampling are more convenient and less costly then
many other methods of sampling.
For studying attitude toward any national issue, a sample of journalists, teacher
and legislators may he taken as an example of purposive sampling because they
can more reasonably be expected to represent the correct attitude than other
class of people residing in country.
Purposes sampling is some what less costly, more readily accessible, more
convenient and select only those individual that are relevant to research design.
v) Systematic sampling
Systematic sampling is another method of non-probability sampling plan, though
the label ‘systematic’ is somewhat misleading in the sense that all probability
sampling methods are also systematic sampling methods. Due to this, it often
sounds that systematic sampling should be included under one category of
probability sampling, but in reality this is not the case.
Despite these advantages, systematic sampling ignores all persons between every
ninth element chosen. Then it is not a probability sampling plan. In Systematic
sampling there is a chance to happen the sampling error if the list is arranged in
a particular order.
Activity
Make a list of some research studies where some of the non probability
methods could be used. Also justify the choice of particular sampling method
you have selected for the study.
A blindfolded person, then, may be asked to pick up one slip. Here, the probability
of each slip being selected is 1-40. Suppose that after selecting the slip and
noting the name written on the slip, he again returns it to the box. In this case, the
probability of the second slip being selected is again 1/40. But if he does not
return the first slip to the box, the probability of the second slip becomes 1/39.
When an element of the population is returned to the population after being
selected, it is called sampling with replacement and when it is not returned, it is
called sampling without replacement.
Thus random sampling may be defined as one in which all possible combinations
of samples of fixed size have an equal probability of being selected.
57
Introduction to Research Advantages of simple random sampling are:
Methods in Psychology
1) Each person has equal chance as any other of being selected in the sample.
2) Simple random sampling serves as a foundation against which other methods
are sometimes evaluated.
3) It is most suitable where population is relatively small and where sampling
frame is complete and up-to-date.
4) As the sample size increases, it becomes more representative of universe.
5) This method is least costly and easily assessable of accuracy.
Despite these advantages, some of the disadvantages are:
1) Complete and up-to-date catalogued universe is necessary.
2) Large sample size is required to establish the reliability.
3) When the geographical dispersion is so wider therefore study of sample
item has larger cost and greater time.
4) Unskilled and untrained investigator may cause wrong results.
Activity
In a class of 140 students, select a simple random sample of size 20 students
with replacement technique. Also mention the probability of each one of
140 students being included in the sample.
Having divided the population into two or more strata, which are considered to
be homogeneous internally, a simple random sample for the desired number is
taken from each population stratum. Thus, in stratified random sampling the
stratification of population is the first requirement.
There can be many reasons for stratification in a population.
Two of them are:
1) Stratification tends to increases the precision in estimating the attributes of
the whole population.
2) Stratification gives some convenience in sampling. When the population is
divided into several units, a person or group of person may be deputed to
supervise the sampling survey in each unit.
Advantages of stratified Random Sampling are:
1) Stratified sampling is more representative of the population because
formation of stratum and random selection of item from each stratum make
it hard to exclude in strata of the universe and increases the sample’s
representation to the population or universe.
58
2) It is more precise and avoids the bias to great extent. Hypothesis Formulation
and Sampling
3) It saves time and cost of data collection since the sample size can be less in
the method.
Despite these advantages, some of the disadvantages of stratified sampling are:
1) Improper stratification may cause wrong results.
2) Greater geographical concentration may result in heavy cost and more time.
3) Trained investigators are required for stratification.
iii) Cluster sampling
A type of random sample that uses multiple stages and is often used to cover
wide geographic areas in which aggregated units are randomly selected and then
sample are drawn from the sampled aggregated units or cluster
For example, if the investigator wanted to survey some aspect of 3rd grade
elementary school going children. First, a random sample of number of states
from the country would be selected. Next, within each selected state, a random
selection of certain number of districts would be made. Then within district a
random selection of certain number of elementary schools would be made. Finally
within each elementary school, a certain number of children would be randomly
selected. Because each level is randomly sampled, the final sample becomes
random. However, selection of samples is done to different stages. This is also
called multi stage sampling.
This sampling method is more flexible than the other methods. Sub-divisions at
the second stage unit needs be carried out only those unit selected in the first
stage. Despite these merits, this sampling method is less accurate than a sample,
containing the same number of the units in single stage samples.
Self Assessment Questions
1) Non probability sampling is one which there is way of assessing the
probability of the element or group of element of population, being
included in the sample. T/F
2) Simple random sampling is the core technique and attaches equal
probability to each unit of the population to be selected. T/F
3) Cluster sampling method sometimes known as multi stage sampling
method. T/F
4) Snowball technique is a probability sampling method. T/F
5) Stratified sampling is more representative for the population than other
methods. T/F
Answer: (1) F, (2) T, (3) T, (4) F, (5) T.
The three main advantage of sampling are that cost in lowest, data collection is
faster, and since the data set is smaller, it is possible to ensure homogeneity and
to improve the accuracy and quality of data (Ader, Mellenbergh & Hard (2008)
Researchers rarely survey the entire population for two reasons: The cost is too
high, and the population is dynamic in that the individual making up the
population may change over time. Sampling methods are of two types i.e. Non
probability and probability sampling methods. Probability sampling methods
are those in which some probability to each unit of the population to be included
in the sample and this is more representative. Three different probability sampling
method are discussed as simple random sampling, stratified random sampling
and cluster / multi stage sampling. The other non probability sampling methods
discussed are convenience sampling, Quota sampling, Purposive sampling,
Snowball sampling and systematic sampling. These methods are also used but
lack the representative character of samples.
4.13 GLOSSARY
Hypothesis : A tentative and testable statement of a potential
relationship between two or more variables.
Null hypothesis : The hypothesis that is of no scientific interest;
sometimes the hypothesis of no difference.
Alternative hypothesis : Statistical term for research hypothesis that
specifies values that researcher believes to hold
true.
Population : It is the aggregate from which a sample is drawn.
In statistics, it refers to any specified collection of
objects, people, organisation etc.
Population size : It is the total number of units present in the
population.
Sampling units : They are members of the population.
Sampling frame : It is the list of all the units of population.
Sampling design : It is a definite plan for obtaining a sample from a
given population.
Sample size : It is the total number of units in the sample.
Simple random sample : It is a sample in which each unit of the population
has an equal chance of being selected in the
sample.