Beruflich Dokumente
Kultur Dokumente
OBJECTIVES:
Introduction
1
This unit deals with the question, why conduct educational research? This is
followed by a description of different types of research, paradigms of social
research and basic elements of research. Finally, the unit deals briefly with each
major stage in the research process: the problem, research design, measurements
design, data collection, and report writing. Key concepts in social research are
discussed.
Methods of acquiring knowledge can be grouped into two main types: non-
scientific and scientific ways of knowing.
Non - scientific methods are based on faith, accepting things at phase value. When
people rely on knowledge that they have not questioned or tested, they are using
non - scientific methods way of knowing. There are four ways of non - scientific
methods knowing. These are briefly described below.
Authority
People often cling to their beliefs in the face of clearly conflicting facts. They even
go to the extent of inferring “new” knowledge from beliefs that may be false.
2
Many people tend to believe things simply because most people in their society
assume them to be true.
Method of Intuition
This method is based on reason. Proponents of this method argue that human
beings have the ability to reason and reach truth, and that a priori propositions
agree with reason and not necessarily with experience.
The word science is derived from the Latin noun, Scientia (meaning knowledge)
and the verb, soire (meaning “to know). The scientific method of acquiring
knowledge, also known as scientific research is systematic investigation of a
question, phenomenon, and/or problem following some principle.
3
It is replicable and transmittable; this means anyoe placed in the same
situation can observe the same event and reasoning.
It is public
It is cumulative and self- correcting.
It is cyclic. This means that it starts logically with a problem and ends with a
tentative empirical generalization.
4
knowledge about what happens in the social sector. E.g. the spread of
HIV/Aids among the married couples.
Prediction
This is the ability to predict or anticipate a phenomenon that will
occur at a certain time from information available at an earlier time.
Control/Improvement of the Social Phenomena
This is concerned with intervention programmes i.e.
e.g. all the diabetes patients in Thika district., all teenagers who are addicted to
drugs in Nairobi west district.
A researcher first defines the population to which he/she wants to generalize the
results.
For some studies the number of a given population may be large or scattered
a wide geographical area, thus a lot of time, money and personnel may be
required to locate all the members in order to select a representative sample.
Hence researchers draw the sample from an “accessible population”.
Sample; this a smaller group obtained from the accessible population. Each
member or case in the sample is referred to as the subject or respondent or
interviewee.
Data; this is all the information a researcher gathers for his/her study.
There are two types of data recognized by researchers
a) Primary data; this is the information e researcher obtains from the field
i.e. from the respondents.
b) Secondary data; this is the information a researcher obtains from articles,
books etc
Data may also be classified as qualitative (words, phrases) and
quantitative (numerical).
Parameters; these like variables are characteristics that are measurable
and can assume different values in the population.
The difference between a parameter and a variable is that a parameter refers
to a population characteristic while a variable is related to a characteristic of
the sample drawn from the population.
7
Statistics; this is the science of organizing, describing and analyzing
quantitative data. Statistics is also defined as the indices which are derived
data through statistical procedures, e.g. mean, median, standard deviation,
correlation coefficients etc.
Descriptive statistics; these are indices that describe a given sample. E.g.
of descriptive statistics are measures of central tendency (mean, median,
mode), measures of dispersion (range, standard deviation, variance),
distributions (percentages, frequencies) and relationships (correlations).
Objectives; this any kind of desired end or condition. Thus objectives in
research studies is refer to specific aspects of the phenomenon under study
that the researcher desires to bring out at the end of the research study. E.g. a
researcher may want to carry out a study on the factors that contribute to
malnutrition of children in Kenya.
E.g. the purpose of this study is to establish the reasons as to why teenagers
get easily addicted to narcotics
8
The purpose of this study is to compare the effects of teaching methods in
physics, lecturing and the practical approach.
Hypothesis; it is a tentative answer to a research problem. In qualitative
research a hypothesis is expressed in terms of relationship between two
variables.
Summary
In this unit we have identified and explained the different methods of acquiring
knowledge. These include: daily experiences, authority, tenacity, intuition,
supernatural revelation, and the scientific method. Certain fundamental
FURTHER READING
9
Cohen, L. Marion, L. and Morrison, k. (2000). Research Methods in Education.
London: Routledge Falmer.
INTERNET
TOPIC TWO
OBJECTIVES:
After reading this unit you should be able to:
10
There are three methods of collecting qualitative date
11
Main Features of Qualitative Research Methods
ACTIVITY 1:
13
TYPES OF RESEARCH
1. Classification by Purpose.
Basic Research
Applied Research
Action Research
14
Action research is not concerned with whether the results are
generalized to any other setting since its major goal is seek a solution
to a given problem.
Evaluation Research
15
Types of Evaluation Research
a) Needs Assessment
b) Formative Evaluation
c) Summative Evaluation
This type of evaluation occurs after the programme has been fully
implemented/developed. It is conducted to evaluate how worthwhile the
final programme has been, especially as compared to other similar
programmes.
16
ACTIVITY 2:
1. When is formative and summative evaluation research carried out?
2. Classification by Method.
Descriptive Research
Collecting data
Causal-comparative
17
This kind of research attempts to identify the relationships between
variables. It determines the reason or cause for the current status of the
phenomenon under study.
E.g. a researcher may observe that boys are more prone to drug
addiction than girls and set out to identify the major factors which
have led to this difference.
18
a) It allows a comparison of groups without having to
manipulate the independent variables
ACTIVITY 3
1. Explain the causal-comparative research design
Correlation Methods
Problem statement
Selection of subjects
Data collection
Data analysis
20
E.g. a correlation can be computed between weight and
level of education even though this relationship is
meaningless.
ACTIVITY 4:
1. Briefly describe correlation design
Survey Research
21
Information collected through surveys may be used for various
purposes e.g. market researchers use survey to evaluate products
acceptance and use.
Problem statement
Defining objectives
Selecting a sample
Preparing instruments
22
Data collection
Data analysis
Limitations of Survey
ACTIVITY 5
Steps Involved
23
Historical Research
24
Official records which may include legal records, legal
instruments such as contracts, wills, court decisions etc.
There are many social issues and problems which are better understood or
studied in the light of past experiences, e.g.
25
Collect and classify resource materials, determining facts
by internal and external criticism
1. External Criticism
2. Internal Criticism
ACTIVITY 6
Differentiate between primary and secondary sources of
information in historical research.
26
Observational Research
Sample selection
a) Non-participant observation
b) Naturalistic observation.
27
Since certain kinds of behavior can only be observed as they occur
naturally, the researcher does not control or manipulate the subjects or
the environment in any way. The purpose of naturalistic observation is to
record and study behavior as it naturally occurs.
c) Participant observation
ACTIVITY 7:
28
d) Ethnography
This method involves very intensive data collection. The data collected
on many variables are collected over an extended period of time in a
natural setting. The rationale behind using ethnography is the belief that
behavior is greatly influence by the environment in which it occurs.
Advantages
Disadvantages
29
It is very costly because a lot of time is needed to understand the
environment being studied.
e) Cases Studies
30
f) Content analysis
Definition
Purpose
1. Decide on the unit of analysis. These are individual units about which
descriptive and explanatory statements are to be made.
3. Coding – This involves logical grouping of the data. This enables the
researcher to apply quantitative or qualitative methods of data
analysis.
4. Data analysis.
31
Advantages of content analysis
2. Errors which arise during the study are easier to detect and correct
Disadvantages
Summary
In this second unit of our module, we have identified and explained the
characteristics and uses of both qualitative and quantitative research. We
have also explained, by use of appropriate examples, the different types
of research such as Basic research, applied research. action research and
Evaluation. Lastly, we looked at the classification of research by method;
such include Descriptive, Causal-comparative, Correlation, Survey,
Historical, Observation, Ethnography, and Case study.
32
SELF EVALUATION QUESTIONS
FURTHER READING
INTERNET
TOPIC THREE
OBJECTIVES:
4. Data collection
Collection of data using a variety of research instrument
5. Data analysis
a) Data organization
b) Data processing
c) Quantitative and qualitative data analysis.
34
6. Generalization
a) Interpretation of data
b) Conclusion and recommendations
c) Production of the research report
The process is diagrammatically presented in figure 1 below.
Figure1: The Main Stages of the Research Process.
Problem
Research Design
Generalization
Measurement Design
Data Analysis
Data Collection
Before conducting any research, you should decide on the major problem area that
interest you. Normally this will be your field of specialty. Examples of such areas
are:
Diabetic patients
The challenges of bringing up a physically/mentally challenged child
35
Challenges children from dysfunctional homes
Next, you should select a specific topic for research. This should be done after
thorough review of literature on the general area. After identifying the topic,
narrow it down to a specific research problem.
The final activity in this stage is definition of the research problem. A problem is a
difficulty to which a solution or answer is being sought in the study. The first step
in defining a research problem is to ask yourself the question, “What do I want to
find out?”
Some characteristics of a good research problem are:
It should be one which can be investigated thorough the collection and
analysis of data.
The problem can be stated clearly and concisely.
It generates a number of more specific research questions.
36
It relates to an academic discipline such as psychology, education,
science etc.
It has a basis in the research literature.
It should have potential significance. It should be important to you
and other people as well, i.e. it should be one whose solution will
contribute to improvement of the society.
It should be one that you can carry out within the available timeframe
and budget.
Sufficient data should be available or can be obtained to address it.
It must be one that you can adequately investigate given your research
skills, available time and other resources.
Following the statement of the research problem and research questions, the null
hypotheses are stated. A well defined hypothesis should have the following
characteristics:
It must be testable with available techniques.
It should be possible to reject by collecting and analyzing data.
It should conjecture upon a relationship between two or more
variables
It should be stated clearly and ambiguously.
38
Since it is extremely difficult to unequivocal support for hypothesis, the
researcher instead attempts to test and disapprove its negation. The negative
version of no significant difference between means scores of two or more
groups is called null hypothesis and is symbolized by H O. The null
hypothesis is usually an expression of no significant relationship between
the variables. It suggest that the observed differences have occurred due to
chance variation and thus are not real differences. The other hypothesis is
the alternative or research hypothesis symbolized by H1.It states that a
difference or relationship exists between the variables.
The following are examples of null hypothesis:
H O: There is no significant difference between the attitudes of parents
towards physically/mentally challenged children when they are categorized
by
Sex
Socioeconomic background
Location of school (rural/urban)
H O: there is no significant difference between the attitudes of women of
different social settings towards abuse in marriages.
Conceptual framework / theoretical framework
In this section, you should provide a theoretical explanation of the research design
used and describe the concepts, variable and theories underlying you study and
relationship among various variables. This section is naturally followed by a
definition of the terms used.
Example
Title: effective of family life education programme on students’ knowledge and
attitudes towards responsible parenthood.
Conceptual Framework
39
The independent variables in this study age, gender, marital status, place of
residence (urban /rural). Knowledge of quality of life issues and attitude towards
responsible parenthood will be the dependent variables.
Students Gender
Student’s Age
Student’
Student’ Marital AttitudesTowards Students’
Status Responsible Knowledge of
Parenthood Quality of Life
issues
Students place of
birth
Determining the Research Design
Students’ Extent of
Attendance in Class
Research design refers to the procedure used by the researcher to select the sample,
administer the instruments and analyze the data. Your choice of research design is
influenced by the questions you intend to answer.
Measurement Design
The fourth stage of the research process involves the construction of research
instruments. Research instruments are the tools you will use to collect the data
necessary to answer your research questions and to test your hypothesis.
Collecting the Data
During this stage data is collected using a variety of methods including visits to
Education institutions, mailing questionnaires to respondents and content analysis
of Curriculum documents.
40
Once data have been collected, they should be analyzed and interpreted. Specify
the techniques you will use to analyze data and the interpretative operations you
will perform on analyzed data.
Generalization
During this stage you should draw and state conclusion on the basis of the findings
of the study. The conclusion you make should be stated in terms of the original
research questions and hypothesis. Generalizations are statements of the
relationship between two or more concepts or variables. The following statements
are generalizations:
The parents and the society need to be educated and sensitized on how to
handle the physically/mentally challenged in homes and society.
If parents and society treat the physically/mentally challenged like normal
people and not like outcasts then it encourages them to live normal lives
RESEARCH PROPOSAL
Research Proposal Development
A research proposal is a plan that indicates the strategy the researcher intends to
use in search of answers to the main research question/ problem and specific
research questions.
41
The actual format of the research proposal, as well as its contents, can vary from
study to study. However, a research proposal should consist of the following
components in the order in which they are listed here.
1.0. Introduction.
1.1. Background of the problem
1.2. Statement of the problem
1.3. Research questions and hypothesis
1.4. Significance of the study
1.5. Delimitations and limitations of the study
1.6. Conceptual / theoretical framework
1.7. Operational definitions of terms
Reference
Appendices
Introduction
42
The purpose of this section is to place the research in context. The section should
be as brief as possible. You should indicate that you are aware of related studies on
the topic and the extent to which those studies do not address the problem of your
study. The section includes a brief review of previous research in the field and a
statement of the questions which previous research studies leave unanswered.
Definition of Terms
All the variables that will be manipulated must be defined. In addition, terms
which may convey different meanings different people should be defined.
Researcher Design
The research approach you have selected and why you have selected it. Some
examples of research design are experimental design, survey design and
naturalistic design.
Target Population
A population is any group of institution, people or object that has at least one
characteristic in common. Examples of a population are women in marginalized
areas, challenges of single parenthood, psychological challenges of the children in
the IDP, the challenges of having a physically challenged child. The target
population should be explicitly and equivocally defined, otherwise statement about
the target population after the analysis of data will not be trustworthy.
Data analysis
45
Indicate the statistical you will use to analyze data. These might be a simple as
frequencies and as complex as two-way analysis of variance. You should ensure
that the statistical procedure you choose are appropriate to the research questions
being asked and the hypothesis you intend to test.
Schedule
Provide a schedule for the various stages of your study.
Budget
Estimate how much the research will cost and provide a budget indicating the
amount or money you will spend on each item.
SUMMARY
In this unit, we have discussed the six main stages in the research process. The first
among those stages is the identification, selection and definition of the research
problem. This stage includes the background, statement of the problem, research
questions, objectives, hypothesis and the theoretical/conceptual frameworks.
Literature review helps in identifying the knowledge gaps that the research desires
to fill in regard to the research problem. The second stage is the selection of the
research design and methodology including sampling procedures. Third is
measurement design which entails identification of data collection procedures and
construction of research instruments. Fourth is data collection while fifth is data
analysis which includes data organization and processing. The last stage in the
research process is generalization which entails interpretation of data, conclusion,
recommendations, and production of the data report.
46
SELF EVALUATION QUESTIONS
Briefly explain four criteria for a good scientific hypothesis
8marks
Briefly explain the six stages in the research process 12marks
Make a sequential list of the topics usually covered in a research
proposal. 20marks
Define the following terms:
a) Variable
b) Research Problem
c) Hypothesis
Describe the main stages of the research process
Five problems from the research literature are given below. Study
them carefully and construct one or two hypothesis based on them.
a) Do teachers’ comments cause improvement in student
discipline?
b) How does organizational climate affect behavioral
development?
How does the socio-economic stat
47
FURTHER READING
Cohen, L. Marion, L. and Morrison, k. (2000). Research Methods in Education.
London: Routledge Falmer.
INTERNET
TOPIC FOUR
OBJECTIVES:
After studying this unit, you should be able to:
Research Design
There are several types of research designs. These include experimental designs,
quasi-experimental designs, survey designs, correlation, case study, and
naturalistic designs.
1. Experimental Designs
These designs are used to study cause and effect relationships. The true
experimental design is considered the most useful design to demonstrate
programme impact if there is random selection of participating units and random
assignments of treatment and control conditions.
49
The programme is implemented and post-test administered to the group which
received the treatment as well as the control group. Sometimes a pre-test is
administered before the treatment. Finally, differences between the group that
received the treatment and the one that did not are compared to determine the
impact of the programme.
Note:
ACTIVITY 9:
Choose a topic in your area of interest, describe the research design you will
use in this study
2. Survey Designs
50
Survey is used to gather systematically factual information by interviewing or
administering a questionnaire to a sample of individuals. It is the most frequently
used method of collecting information about peoples attitudes, opinions, habits or
any of the variety of social issues.
They are an efficient method of collecting descriptive data regarding:
The characteristics of the population;
Current practices, conditions or needs;
Preliminary information for generating research questions.
When using a survey design, you should follow the following procedures:
Identity the population of interest
Clarify the objectives of the study
Develop a questionnaire and field test it
Select a relevant sample
Administer the questionnaire to members of the sample by mail or in
person
Tabulate the results in a descriptive fashion (i.e. reporting means,
frequencies percentages).
3. Correlation Designs
51
are better indicators of students’ future performance in mathematics or
physics.
5. Evaluation Design
Evaluation research takes two aspects, depending on the stage of the project
at which it is introduced.
The first one is the formative evaluation whose main purpose is to collect
data about projects while they are being developed. The results can be used
by the project developers to modify the project if need be or to stop the
project altogether.
52
The second one is summative evaluation which occurs after the programme
has been fully implemented/developed. It is conducted to evaluate how
worthwhile the final programme has been, especially as compared to other
similar programmes.
6. Naturalistic Design
The naturalistic design grew out of the need to study phenomena as they naturally
occur in the field. The goal of a naturalistic study is to understand the phenomena
being observed. A major feature of this design has been the use of the human
instrument (the observer) to collect, filter and organize incoming data. The design
differs from others because it evolves during the course of the study.
ACTIVITY 10:
State the main stages you should follow when using the survey design.
Discuss the naturalistic design
53
Population
Population refers to all people who posses the characteristics of interest. For
example, to find out whether arts graduate teachers are most likely to become head
teachers than science graduate teachers, the population is all graduate teachers of
arts and science.
Sample
Sampling
Sample size
The optimum sample size is directly related to the type of research you are
undertaking. For different types of research, “rule of thumb” can be used to
determine the appropriate sample size.
Rules of Thumb
54
Quantitative Research Rules of Thumb
Borg and gall (1989) recommend the following sample sizes fro different kinds of
research;
55
Qualitative Research Rules of the Thumb
56
Representativeness of Samples
1. Non-Probability Sampling
This method involves choosing items from the population without using a random
sampling technique. Elements in the target population have an unknown chance of
being selected into the sample. A non-probability sample is based on subjective
judgement and is biased in the sensethat some members of the target population
have more chances of being selected than others. There are three main types of
non-probability sampling: quota, judgement and convenience.
a) Quota Sampling
57
Quota sampling derives its name from the practices of assigning quotas or
proportions of kinds of people to interview. In this technique, sample members are
drawn from various target population strata e.g. untrained teachers, graduate
teachers, P1 teachers, etc.
b) Purposive/Judgement Sampling
c) Convenience Sampling
ACTIVITY 11
2. Probability sampling
58
Probability sampling is a method of drawing a portion of a population so that each
member of the target population has a known and non-zero chance of being
selected into the sample. There are many ways in which a probability sample may
be drawn from a population. Some of these ways are described below.
A simple random sample is obtained when each member of the target population is
assigned a consecutive number and then selected randomly according to the
assigned number. To use this method, it is necessary to have a complete list of the
target population.
This technique consists of selecting every nth sampling unit from a list of all
members of the population. Suppose the researcher wants to select a sample of 50
students from a list of 500 students. To use this method, she first divides the
population by the number needed for the sample (500/50 = 10). She then selects a
number smaller than 10. Then, starting with the number (e.g. 6), she selects every
tenth name from the list of students.
59
Randomly select from within each chosen school, individuals from the
list of all individuals of the defined type.
d) Stratified Sample
Stratification ensures that different groups of the population are represented in the
sample. The population is divided into strata such as boys and girls, schools, rural
and urban schools etc, from which random samples are drawn. In this procedure,
the target population is first stratified into a number of categories. Thus the strata
may be based upon grades, e.g. 200 grade 5 pupils, 100 grade 6 pupils and 50
grade 7 pupils. For each of these grades, pupils are selected at random. The number
selected from each grade should be proportional to the number of pupils in each
grade.
e) Cluster Sampling
This is sampling in which groups, not individuals, are randomly selected. For
example, instead of randomly selecting Form III pupils, you could randomly select
Form III classrooms and use all the students in each classroom.
60
SUMMARY
61
FURTHER READING
INTERNET
OBJECTIVES:
After studying this unit, you should be able to:
Introduction
After you have designed your study, and in particular chosen the sample, the next
step is to design research instruments and then collect information for the sample.
This unit describes the instruments and methods for collecting data from the
sample.
Content
Questionnaire
A questionnaire is a carefully designed instrument (written, typed or printed)
for collecting data directly from people. A typical questionnaire consists of
questions and statements. Two types of questions are normally asked:
closed-ended and open-ended questions. Closed-ended questions are
structured in such a way that the respondent is provided with a list of
responses from which to select an appropriate answer. The main advantages
of close-ended questions are:
a) They are easy to process
b) They can be transferred directly into a computer format
Interviews
“An interview is a conservation in which one person, the interviewer
seeks response for a particular purpose from another person, the
interviewee”
64
The interview is one of the most used techniques of obtaining information. It
is a way of obtaining data about a person by asking him rather than by
watching him behave. A personal interview helps the evaluator to measure
what a person knows (knowledge) and what he likes and dislikes (values and
preferences). The information obtained can be transformed into a number of
quantitative data by using attitude scaling or rating – scaling techniques.
65
Interviewing requires skilled personnel.
Quantitative: An education media specialist asks teachers how many radio lessons
they listen to.
ACTIVITY 12:
66
Information questions
Demographic Questions
These are questions which seek background information about the respondent, for
example, sex, age, level of education, marital status, occupation, religious
affiliation, place of residence, etc.
Information questions
These questions seek to find out the respondent’s knowledge of an area of concern
to the evaluator.
Examples: a. When was the secondary school at which you teach established?
b. How many children dropped out of school last year at your school?
Self-Perception Questions
There are questions that try to determine the respondent’s evaluation of his/her
behaviour in relations to others and also his evaluation of others.
67
Examples: a. how often do you use teaching aids?
b. Are teachers in your district conversant with authentic assessment
methods?
Response Formats
Unstructured Format
Example: Qualitative open-ended question: What do you like best about your
jobs?
Quantitative open-ended question: What is your highest level of
education?
These are filled in response items. The response is open-ended but only a short
answer is expected. They are easier to code than unrestricted free response items.
These questions give the respondent choice from which to select an answer. There
are four categories of structured response items: checklist, inventory type, ranking
type and scaling or rating type.
68
SELF-EVALUATION QUESTIONS
Example: The following are some of the reasons students are given for studying
education. Tick any one of them, which you think represents the most
important reason why you are studying education?
i) The course is easy [ ]
ii) Teachers are marketable [ ]
iii) Teachers are very well paid [ ]
iv) It was difficult to get admission to another course [ ]
v) My parents wanted me to become a teacher [ ]
69
Inventory: Inventories are used when respondents are required.
Example: some statements below are true, others are untrue. Decide whether
each one is true or untrue and write true or untrue on the right-hand
side of each statement as appropriate.
1. School is enjoyable ________________
2. I think going to school is a waste of time ________________
3. It is important for you to do well at school
________________
Format
1. Title
Write the title of the study followed by the title of the questionnaire.
72
Example: Evaluation of the effectiveness of School Broadcasting in Kenya
Questionnaire for Teachers.
2. Purpose
State the purpose of the study.
3. Instructions
Give brief general instructions, including instructions for returning the
instrument. Do not give instructions for completing individual sections. This
should appear at the beginning of each section.
4. Cover letter
Seek permission from the respondent for participation in the study. Informed
consent may be included in the cover letter in the questionnaire.
5. Organize the questionnaire in sections according to the type of questions
asked. Normally, section 1 consists of demographic questions.
Questionnaire Format
a) A questionnaire should have a suitable title.
b) Instructions should be clear and unambiguous.
c) The questionnaire should be attractive and brief and as easy to respond to as
possible.
73
d) Arrange the questionnaire in content sub-sections, e.g. background data,
objectives, implementation of the curriculum, etc.
e) No item should be included which does not directly relate to the objectives
of the study.
f) Divide the questionnaire into meaningful components
g) The first section is personal data. The major consideration is variables that
influence what you are investigating. The question you should ask is, what
are the major variables that are likely to influence people’s responses?
h) Include brief, clear instructions printed in bold type.
i) The questionnaire should start with simple factual questions, so that the
person completing it gets off to a good start.
j) Avoid negative items
k) Avoid biased items and terms.
l) Do not use one item to select multiple answers
m) Open-ended general questions should be at the end to allow expression of
points which the respondent thinks important.
n) Questionnaire may include attitude scales, rating and check lists provided
they are brief and straightforward.
The following are some of the issues clarified during the pilot study:
1. Whether there are flaws and ambiguities
74
2. The feasibility of the proposed procedure for coding responses
3. Weather or not the prospective respondents are available and accessible
4. Weather the intended respondents posses the information being sought and
are willing to participate in the study
5. Weather the intended respondents will understand the questions
6. weather the procedures for administering the questionnaire are appropriate
Revising Questionnaires
Administer the questionnaire and find out how it works. This you can see from:
People’s willingness to respond
Queries
Level of understanding or misunderstanding of the items
Reactions to specific items
75
b) Interview respondents and non- respondents
Administering Questionnaires
There are four main methods of administering questionnaires to respondents. These
methods are briefly described below.
Captive groups
A captive group has been defined as “an assembly of people that the researcher has
enough control over to allow for completion of questionnaires” (Kelly, 1999). In
addition, a captive group may be obtained by administering the questionnaire to
students in each class. This method has a very high response rate.
In-person Interview
In this method, the interviewer reads the questionnaire verbatim and records the
answers either mechanically or in writing.
Phone Administration
Mail Administration
This is an expensive way to cover a large geographical area. However, this method
has a low response rate.
76
Procedure for interviewing
77
SELF-EVALUATION QUESTIONS
Observation Techniques
One way of obtaining about progress and outcomes of a program or project is to
directly observe selected aspects of its development and implication as they occur.
Participant – Observation
In this method, the data gather becomes a participant in the study, observes and
records information. Since he is a member of the group, the other members tend to
act naturally. This way, the data gather is able to obtain information that he could
not have obtained were to use an interview schedule, a questionnaire or other
methods.
Advantages
It provides reliable information.
Disadvantages
It is expensive and time consuming.
It could be subject to bias.
78
Examples of Qualitative and Quantitative Use
Qualitative: Teachers are asked to record their observation for one-week regarding
their prescriptions of how the school is run by the head teacher.
Qualitative: A teacher is told to record the time when different school activities
start and end.
Direct observation
Disadvantages
SELF-EVALUATION QUESTIONS
Distinguish between participant observation and
direct observation. 2 marks
Briefly describe three situations that may be served by
observation data. 8 marks
Briefly explain five disadvantages of direct observation. 10
marks
80
Choose a topic and a goal for questionnaire. Develop a four-item questionnaire
with one item of each of the following types:
a) Open ended
b) Multiple choice
c) Likert
d) Ranking (10 marks)
Explain why direct observation would be the most appropriate method of investigating
the change in behavior in indisciplined teenagers. (10 marks)
Reliability
Reliability of an instrument is the consistency in producing a reliable result. It
focuses on the degree to which empirical indicators are consistent across two or
more attempts to measure a theoretical concept.
1. Test-retest Method
This is a technique of assessing the reliability of a research instrument. It involves
the administration of thsube same instrument twice to the same subjects. there is
usually a two weeks lapse between the first and the second administration.
81
The test-rest method involves the following steps:
Reliability
Reliability is a measure of the degree to which a research instrument yields
consistent results or data after repeated trials. Reliability in research is influenced
by random error. As random error increases, reliability decreases. Random error is
the deviation from a true measurement due to factors that have not effectively been
addressed by the researcher. Errors may arise from inaccurate coding, ambiguous
instructions to the subject, interviewers fatigue, interviewee’s fatigue, interviewer
biased etc.
Random error will always exist regardless of the procedures used in a study.
Let us assume that the researcher want to determine the effectiveness of different
diets in controlling weight.
Subjects are randomly selected and assigned to three groups. The groups feed on
three different diets and each subject’s weight is taken regularly. If the researcher
uses a weighing scale sometimes overestimates the subject’s weight while at one
time, your scale underestimates their weights, and the data from the study will not
be reliable. Random error, from the faulty scale, is operation her to make the data
not reliable. Another example of how random error can lead unreliable data would
82
be in measuring a patient’s blood pressure. A patient with high blood pressure will
be shown to have consistently high blood pressure readings if the measuring
instrument is reliable.
However repeated readings would not be the same: they would be distributed
around the true blood pressure level of the patient. This deviation above and below
the true blood pressure level would be due to factors such as joy, anger , anxiety
etc. the smaller the deviation the more reliable the data. It is important to realize
that research instruments yield data that have two components, the true value or
score of error components,
The error component of the data reflects the limitations of the instruments. Strictly
speaking, there are three types’ random errors that arise at the time of data
collection.
These are:-
Error due to the inaccuracy of the instrument: error due to the inaccuracy scoring
by the research and not explained error.
These three types of errors combine to produce inconsistencies in the
measurements, which ultimately affect the reliability of the data collected.
The research process attempts to minimize random error and hence increase the
reliability of the data collected. In research study, a reliability coefficient can be
computed to indicate how reliable data are. A coefficient of 0.80 or more implies
that there is high degree of reliability of the data.
The test-retest technique
The test-retest method of assessing reliability of data involves administering the
same instrument twice to the same group of subjects. There is usually a time lapse
between the first test and the second test. The test-rest method involves the
following steps:
5. Select an appropriate group of subjects
83
6. Administer the test to the subjects
7. Keeping all initial conditions constant, administer the same test to the same
subjects. One can wait one to four weeks before administering the test a
second time.
8. Correlate the stores from both testing period.
The correlation coefficient obtained is referred to as “the coefficient of reliability
or stability,” if the coefficient is high, the instrument is said to yield data that have
high test-retest reliability”
The disadvantages of this of assessing reliability of data are that subjects may be
sensitized by the first testing or they may tend to remember their responses during
the second testing. If that happens, the coefficient may be artificially high.
The other difficulty with this method is establishing a reasonable period between
the two testing sessions. If the period between the tests is too long, say one year,
subjects may have changed or other extraneous factors may interfere with the
variable being measured. The computed coefficient of stability in such cases may
be artificially low or high.
86
The use of the (K-R) 20 formula is assessing internal consistency of an instrument
is based on the split-half reliabilities of data from all time required to compute a
reliability coefficient in other methods. Its application also results in a more
conservative estimate of reliability; the estimated coefficient of reliability of data is
always lower. To avoid erroneous conclusion, it is always better to underestimate
reliability of data than to overestimate it.
The K-R 20 formula is as follows;
KR20 = (K) (S2-£s2)
_____________
(S2) (K-1)
Where:
KR 20 = reliability coefficient of internal consistency
K = Number of items used to measure the concept
S2 = variance of all scores
s2 variance of individual items
a high coefficient implies that items correlate highly among themselves i.e. there
is consistency among the items in measuring the concept of interest. This is
sometimes referred to as homogeneity of data. There is perhaps a slight overlap
here with validity of data, which is discussed next,
Validity
Validity is the accuracy and meaningfulness of inferences, which are based on the
research results. In other words, validity is the degree to which results obtained
from the analysis of the data actually represent the phenomenon under study.
Validity, therefore, has to do with how accurately the data obtained in the study
represents the variables of the study.
87
if such data is true reflection of the variable s, then inferences based on such data
will be accurate and meaningful, for example , accurate inference could be made
about people’s driving experience if data regarding the number of accident each
driver has caused over a period of ten years are obtained. However, we cannot use
such data to make inferences about the drivers’ ability to operate other machinery
such as lathe machines. It should be noted that validity is a matter of degree: no
data can have perfect validity.
Validity is largely determined by the presence or absence of systematic error in
data.
Systematic error, also referred to as non random error, has a consistent boosting
effect on the measuring instrument. For example, if a researcher uses a faulty scale
to measuring instrument. For example, if a researcher uses a faulty scale to
measure the weight of rats, the scale may overestimate the weights of the rats by a
few grams, such as error is non- random: the obtained readings are either always
above or always below the true scores i.e. Always in one direction.
89
SUMMARY
In this unit, we have discussed the whole question of data collection. Six
major and widely used instruments of data collection were identified and
discussed at length. These are: questionnaire, interview schedule,
observation schedule, focus group discussion guidelins, content anslysis
guide, and tests. We went on to explain the characteristics, use and
advantages of these data collection instruments. For instance, we noted
that there are different kinds of questions in a questionnaires such as
demographic, opinion, attitude, self-perception, and information
questions. Questionnaires can also be structured, semi-structured or
unstructured. We also distinguished between participatory and non-
participatory observation and explained the advantages and
disadvantages of each. The manner of administering different data
collection instruments was explained and the unit ended with an
explanation of the significance of reliability and validity on research.
90
SELF EVALUATION QUESTIONS
What is the importance of editing research instruments?
Explain the reasons for piloting a questionnaire.
Giving appropriate examples discuss steps followed in constructing
the following instruments:
i. Questionnaire
ii. Checklist
iii. Interview Guide
iv. Discuss the methods of establish reliability of measurement,
outlining the weakness and strengths of each method.
v. Using appropriate examples, discuss the factors that affect
internal validity.
91
FURTHER READING
INTERNET
TOPIC SIX
OBJECTIVES:
Written Report
The final stage in the research process is the preparation of research report
Writing a report
An overview
A research report is a question-answering or problem-solving document. A
research report informs the reader of the problem the researcher initially set out to
investigate, the method of investigation used and the researcher’s findings. The
report should present data fully and adequately, it should include accurate
interpretation of the analyses of such data and should relate findings back to the
objectives, hypotheses or research questions. A research report must therefore be
formal, precise and economical. The report must also be consistent, with an orderly
flow of ideas from the beginning of the document to the end. In addition, various
sections of the document, such as main headings and sub-headings, must he well
92
co-coordinated, often in a linear progression. For guidance, it is important that a
writer consults the relevant report writing manuals, for example, the American
Psychological Association Manual, donors’ formats and institutional formats.
A well-written research report comprises a number of distinct sections or
components. Table 11 gives a checklist of the various components that a researcher
should include in a research report. Each section is then briefly described.
Title
The title of the research report should be short and simple yet informative. The title
serves two main functions: (1) It gives the reader some initial information about
the area of research the report has covered and it is therefore a pointer to the
content of the report. The title should also give an indication of the major variables
of the research, the target population and if possible the geographical location of
the research. (2) It is used for indexing once the document becomes official. Other
researchers interested in reading your research report may use the title to search for
the report in a library or documentation centre. A misleading title may therefore
make it difficult for other researchers or persons interested in reading your report
to locate it easily, however good the study may be.
Dedication
Some authors of research reports dedicate their work to a person or per sons whom
they deem special in their lives. Such persons may be a spouse, children, parents or
very dear friends. Dedications are especially common in masters thesis or doctoral
dissertations. A dedication is basically a statement such as “to my children: Juma,
Fatuma and Odhiambo”. Sometimes, the author may give a reason for the
dedication, for example. “to Osike, for her years of friendship.”
Abstract
An abstract contains the meat of the research report. It is therefore a reflection of
the salient findings of the research study. An abstract should be placed at the front
of the report preferably after the dedication, but on a separate page. A good
abstract presents the reader with the purpose of the study, the population studied,
94
the major results and conclusion of the study. An abstract should not be more than
half a page long. For an example of an abstract, see Appendix I.
Table of contents
A table of contents is like a map that guides the readers in locating various sections
of the research report. A table of contents contains the chapter headings, main
headings and subheadings and the corresponding page of’ each in the body of the
document. A table of contents helps the reader to locate the desired sections
quickly and easily.
List of tables
In some research reports, abbreviations and acronyms are used, especially when a
word or a group of words is repeated several times in the document. An
abbreviation is a short form of a word. An example is Dr. for “Doctor” or hrs. for
“hours”. An acronym is a contraction formed by taking the first letter of several
words. Examples of acronyms are FAQ for “Food and Agriculture Organization”
or GOK for “Government of Kenya”.
The researcher should give a list of all abbreviations and acronyms used in a report
and explain in full what each abbreviation or acronym stands for. Abbreviations
and acronyms are not appropriate in formal writing and should be used only when
necessary. It is always advisable to write the word or words in full. The following
are pointers on the use of abbreviations.
•When using acronyms, write in full the first time that the phrase is used.
Thereafter, use the acronym.
Do not abbreviate military, religious and political titles.
Abbreviate units of measurement only when they are used often in a report.
For example “km” for kilometres or “yrs.” for years.
Use only those abbreviations that your audience will understand.
Do not abbreviate days of the week or months.
96
Abbreviate time designations only when they are used with actual time, for
example, 4.30 p.m.
Executive summary
An executive summary is a short version or summary or the research report. The
summary provides readers with a concise and accurate picture of the entire
document. The executive summary is an economical way of communicating one’s
research report. It saves time and energy for readers who do not have time to read
the entire report. In this regard, the executive summary must provide the salient
points from all the major components of the research report.
A good executive summary begins with a clear and brief statement of time
problem. Details from the body of the report are then presented in brief and the
summary ends with conclusions and recommendations. The use of bullets and
bolded subheadings helps to make the summary reader-friendly. Researchers
should avoid use of technical terms in the executive summary. Therefore in
meaning, as well as in style, an executive summary should stand alone as a self-
contained and easy-to-understand message. Readers should only have to read the
entire report for a closer view but definitely not to make sense out of one’s
summary.
Introduction
The introduction to a research report has several components that include the
background of the study, the problem statement, objectives, hypotheses or research
questions and limitations of the study. The major role of the introduction section is
to outline the gap or gaps that exist in the area of study and present the rationale of
the study. This leads to the statement of the problem. The researcher should quote
existing studies and theories to support the problem statement. Since there is a
97
section in the research report that deals with previous studies, literature cited in this
part of the document should he limited to the most relevant and current.
Literature review
Review of related literature is very important because it provides the needed
support to the researcher’s rationale for undertaking research in a certain area. A
literature review is a researcher’s critique of findings from other studies done in
related areas. The critique should include an assessment of the methodologies used
in these studies, theoretical or conceptual frameworks and the relationships or
differences between the researcher’s study and the studies reviewed.
The most critical aspect of reviewing the literature on the subject is the clear and
systematic analysis of existing studies that a researcher must engage in to identify
the missing gaps in a specific area of study and hence justify the need for another
study. This exercise also assists the researcher in making a strong case for the
study’s objectives, research questions or hypotheses. Literature review should be
relevant to the topic. How broad the review ought to be depends on the degree to
which a specific area has been studied. All the studies referred to in the literature
review section must be well referenced.
Methodology
The methodology section gives details regarding the procedures used in conducting
the study. Pertinent issues discussed in this section include the population, sample
and sampling techniques, the research design, a description of instruments or tools
used to collect data, the measurement of variables and the techniques to be used in
analyzing data. If hypotheses are to he tested, the researcher must state the level of
significance to be used in such tests.
The most important point to bear in mind when writing the methods’ section of
98
your report is that the details given regarding the procedures used in conducting the
study should be precise enough to enable other researchers to replicate your study.
The basic principle in presenting results is to give all the evidence relevant to the
research objectives and questions, if any. In quantitative research, the researcher is
sometimes expected to test hypotheses. In such cases, the researcher should
address all the hypotheses posed in the introduction chapter and give the results of
such tests, stating whether each hypothesis is supported by the data or not. It
should be borne in mind that a statistical test is done on the null hypothesis and the
research hypothesis is supported by the data only when the null hypothesis is
rejected at the significance level initially stated by the researcher in the
methodology section.
When writing the results and discussion section of the report, it is advisable to start
with a short introduction that describes the general procedures followed in
analyzing the data. It is also important to state the objectives or the hypotheses of
the study. The following are pointers on writing the results and discussion section
of a research report:
In this section of the report, the researcher has the opportunity to summarize the
study undertaken, to conclude and make recommendations based on the study
findings. The summary reminds and informs the reader about the purpose of the
study, the process used to collect and analyze data and the major findings of the
study. A summary must reflect as accurately as possible the body of one’s report.
As Lannon2 puts it, the conclusion “culminates” your research report. The
conclusion is of utmost importance to your readers because it answers the
questions that sparked the collection and analysis of the data in the first place. In
concluding your report, it is important to discuss the practical application and
implications of your findings in the real world. Although you are supposed to have
101
interpreted evidence at every step of your analysis, your conclusions pull the
strands together in a broader perspective and indicate possible action, where
appropriate.
Such implications are usually targeted to stakeholders. For example, a study on
primary school drop-outs should discuss the implications of the study to education
policy makers (e.g., is there a need to change policies?), teachers (e.g. is there a
need to change teaching methods or the curriculum?) and students (e.g., is there ii
need to provide students with counseling services and interventions to boost their
morale?).
Your summary and Conclusions should lead logically to your recommendations.
Recommendations must be consistent with the purpose of the study, its objectives,
the evidence presented by the data and the interpretations given. Recomrnendations
should be practical and achievable.
For example. It is not practical to recommend that all students in public
universities be provided with computers to make them computer literate because
such a recommendation would require a lot of money, which is simply unavailable.
102
initiated in secondary schools to reduce dropout rate and frequent student
strikes or riots in Kenya
References
In writing a research report, a researcher relies heavily on published or unpublished
documents that bear some relationship to the researcher’s area of study. The
researcher should give credit to authors of any work or materials quoted from or
referred to by listing them.
There are several formats or styles of referencing documents used in a research
report. The American Psychological Association Manual is a good example of a
referencing style.
Acknowledgements
The research process takes a lot of time, money and energy. During the study, a
researcher may require help from other individuals such as professional colleagues,
respondents, research assistants, etc. The researcher may also seek materials or
information relating to the stated area of research front institutions or originations.
It is therefore only a matter of courtesy for the researcher to acknowledge the help
received from various individuals, institutions or organizations. Because it ma he
impossible to name all the people that contribute in one way or another to the
research, it is important to mention only a few individuals by name to keep the
acknowledgement section as short as possible.
Appendices
The appendix Section of the report contains information that the researcher does
not deem necessary to include in the body of the report. Most of the time, such
information makes the report unnecessarily long without adding much value to the
content of the document However, the reader may use such information as
reference material to clarify some statements appearing in the body of the report.
103
Instruments used in collecting data, statistical tables from data analyses
correspondence related to the study such as letters of respondents’ consent to
participate in the study and letters of approval to conduct research arc examples of
the type of information that may be included in appendices It is advisable to
include a copy of the research permit and the terms of reference, if any.
Further Readings
104
Mertens, D.M., (1998). Research Methods in Education and Psychology.
Thousand Oaks: Sage Publications.
Orodho, A. J (2003) statistics made user friendly for Education and Social
Science Research. Masola Publishers
Mt Kenya University
SCHOOL OF EDUCATION
DEPARTMENT OF PSYCHOLOGY
105
UNIT CODE: CCU 213
Mt Kenya University
SCHOOL OF EDUCATION
DEPARTMENT OF PSYCHOLOGY
106
Time: 2 Hours
1
i) a) Briefly explain the following;
ii) Variable 2marks
iii) Data 2marks
iv) Objectives
2marks
v) Problem 2marks
b) Name the two approaches of research design and give two features of each approach. 8marks
e) Differentiate between;
3. Describe how a survey design could be used in a study of your choice. 20marks
5. You have been sponsored to undertake research on HIV/AIDS patients at all district hospitals
in Kenya.
b) Discuss five (5) ethical considerations you would bear in mind. 10marks
107
108