SOURCES OF DATA IN RESEARCH

• Primary Source of Data

• Primary data refer to materials collected as
first hand information (fresh data from its
origin). The process includes administering a
questionnaire, conducting an interview,
observation, or at best, combinations thereof.

1
 Sources of Data
• Primary Source of Data
• Primary data refer to materials collected as first hand information
(fresh data from its origin). The process includes administering a
questionnaire, conducting an interview, observation, or at best,
combinations thereof.
 
• Secondary Source of Data
• Secondary sources contain information compiled by others on the
subject matter and contained in various documents that may
include books, magazines, journals, periodicals, companies’ annual
reports, bulletins, articles, newspapers, government documents
and government directorates.

2
 Strategies for Ensuring Quality

These will involve preventing:

• Deviation from the sampling procedure
set out in the proposal.
• Bias in observation of measurement
made.
• Variation in criteria for measurement of
(or) for categorizing answer.

3
The data - collection process that may help ensure quality include:

• Preparation of a fieldwork manual for the

research team as a whole.
• Selecting research assistants required with care.
• Training of research assistants.
• Pre-testing research instrument and research
procedure.
• Arranging for ongoing supervision of research
assistants.

4
 Contents of Data Processing and Analysis Plan

A data processing and analysis plan

should include:
a. Sorting data
b. Performing quality control checks
c. Data processing
d. Data analysis

5
RESEARCH PROBLEM AND OBJECTIVE
Sources of Research Problem
The following are the various sources of
research problems:
- Experience
- Literature
- Theoretical sources
- Contemporary issues
- Consultation with subject experts

6
 Process of Problem Identification

The procedure for problem identification

in research includes:
- identify a general problem area;
- review the existing literature for gaps;
- break up the identified area; then
- select the one appropriate to your
circumstance.

7
 Qualities of a Good Research Problem

In writing a problem statement, certain features must be put

in place for it to be valid and acceptable. These features
include:
- It must be breakable into dependent and independent
variables.
- It must be recast-able into question form, unambiguously.
- It must be liable to empirical analysis and testing.
- Its solution must either correct or impact positively on
people/society.
- It must be solvable within the context of the researcher’s
intellect.
- It should show the relationship that exists between two or
more variables.
8
 Considerations in Choosing a Research Problem
• In choosing a problem statement, certain factors have to come
into play. To have a valid problem statement, an evaluation of
the following questions and issues must be made.
• (i) Background: This requires selecting a topic that has to do
with the availability of preparatory materials before the
research work and writing can commence.
• (ii) Interest: This involves evaluating your interest in the
problem area.
• (iii) Workability: Will it be easy to evaluate the problem?
• (iv) Practical Value: What contribution will the solution give to
the society
• (v) Involvement: Evaluate your ability to handle the Stress
and inconveniences that may be involved.
9
 Meaning of Research Objective

• The objective of a research is the totality of what the study

stands to achieve. A good research topic must be capable
of being summarized to what is to be achieved by the
study. Objective must be stated under two categories of;
• General or Main Objective and Specific Objectives.
• Note that, Specific Objectives are the subsets of the
General Objective. Therefore in formulating the research
objective, this distinction must be borne in mind.
• Once the objective already stated is wrong, no other
structure built on it can be right and this is why the
objective of the study has been given a prominent place in
this text.
10
 Methods of Stating Research Objectives
• Research Objectives should be stated in such a way that:
- All aspects of the problem are taken care of.
- They are realistic, considering local environmental conditions.
- Action verbs that are specific enough to be evaluated are used.
- Operational terms are to be adopted to state phases.
• Forms of Research Objective
- The study objective is usually presented in two related forms by
coining one out of the other. An example will suffice.
- Let’s take a study captioned “The factors militating against
effective budget performance in Nigeria”. Now to fashion out
the two forms of objectives (General and Specific) from the
above topic we have the following:

11
 The General Objective

• The General Objective will be to examine

the factors affecting budget performance
in Nigeria.
• Note that the General Objective is a
mere re-statement of the research topic
but in a more captivating manner that is
amenable to statistical evaluation.

12
 The Specific Objectives

• To be able to fashion out relevant Specific Objectives from the above stated
General Objective, there is the need to know those factors militating against
budget performance in Nigeria.
• It is not in doubt that we have so many factors affecting budget
performance in Nigeria. Therefore, to guard against any unwieldiness,
there is the need to limit the coverage of the study to achievable standard.
For this reason, let us now assume that out of the numerous factors
militating against budget performance, the study now identifies (and want
to examine) the following prominent ones.
• (i) Lack of accountability
• (ii) Political instability
• (iii) Bad governance
• Each of the above identified factors will now be converted to a Specific
Objective. To accomplish this, it is necessary to note the use of
conventionally active words (verbs) such as: examine; assess; identify and
evaluate but not determine) to activate them into the required objective
status. This is presented below in that sequential order. 13
 The Specific Objectives of the study are to:
• (i) assess the effects of lack of accountability on
budget performance in Nigeria;
• (ii) identify evaluate the influence of political
instability on budget performance in Nigeria; and
• (ii) examine the impact of bad governance on
budget performance in Nigeria.
Noteworthy is the fact that, by the time all the
various enlisted factors have been individually
examined and the results obtained are put together,
it will be discovered that, the required information
on the General Objectives will automatically emerge.
14
 Hypothesis
• Hypothesis is a conjectural statement of the
relationship between two or more variables.
Hypothesis is in declarative sentence form and
relates either generally or specifically, to variables.
• It is a testable statement which subsequently
needs to be either accepted or rejected. This
definition presupposes that:
(1) hypothesis is made up of variables;
(2) the variables may or may not be related; and
(3) there is the likelihood of a subsisting
relationship between or among the variables.
15
 Formulating Research Hypothesis

In formulating any research hypothesis, the researcher

should bear the following fundamental principles in
mind.
• (i) It must be stated in declarative terms, relating the
variables to each other.
• (ii) It must be related to the research problem.
• (iii)It must be logical, simple worded, and contain few
concepts at a time.
• (iv) It should not contradict well-established principles
and laws.
• (v) It must be stated in a way that will make it
amenable to verification, rejection or acceptability.
16
 Attributes of a Good Hypothesis

(i) The statement of hypothesis must be in clear,

precise, correct and simple language.
(ii) Hypothesis must be operationally defined
(iii) Hypothesis must not contradict already known
and established facts on ground.
(iv) Hypothesis needs not be of global significance.
(v) Hypothesis should be testable and measurable.
(vi) Hypothesis should be void of vague constructs
or ambiguous language.
(viii) The selected hypothesis should be amenable
to testing 17
 Types and Forms of Hypotheses
Types of Hypothesis
• There are two types of hypothesis, namely;
i. Directional hypotheses (one tail) and
ii. Non-Directional (Null) hypotheses (two tail).
Forms of Stating Hypothesis
• A hypothesis is usually stated in two forms viz: (i) Null
Hypothesis and (ii) Alternate Hypothesis.
• Null Hypothesis: A null hypothesis is a negative statement
which states that there is no true difference between two
mean populations and the difference found between sample
means is therefore accidental and so unimportant Ho.
• Alternative Hypothesis: This is the opposite of the null
hypothesis. Also known as the positive hypothesis, it is
represented by the symbol HA.
18
 Other Classifications of Research Hypotheses

• (a) Simple Hypothesis: This shows the relationship

that exists between two variables (dependent and
independent variables).
• (b) Complex Hypothesis: This is a hypothesis that,
although, contains dependent and independent
variables, it has more of independent variables.
• (c) Pseudo Hypothesis: This is a hypothesis that
contains not less than two variables. It is possible to
see a pseudo hypothesis with one variable statement.
• (d) Conjoint Hypothesis: This is a hypothesis that
states situations which may lead to the actualization
of the predicted relationship among the variables.
19
 General Procedure for Hypothesis Testing
Several steps are involved in testing hypothesis. These steps have
been highlighted below:
Select a suitable level of Significance
• In the course of testing hypothesis, certain errors are likely to be
made. This will definitely affect the conclusion that will be drawn
from the outcome. These errors are categorised into:
• Type I error
• Type II error
Type I Error (Alpha error): Occurs when a null hypothesis that is
supposed to be accepted is rejected due to a mistake made
before arriving at the sample result.
Type II Error (Beta error): occurs when a null hypothesis that
ought to have been rejected is accepted due to a mistake made
in the course of processing the data. 20
 Functions of Research Hypothesis

Hypothesis is an indispensable tool in any research work.

It is basically used to enlighten the researcher on the
problem and highlighting the direction in which the
solution to the problem can be found.
The relevance of hypothesis in any research work
includes:
• (i) It relates the variables being studied
• (ii) It delimits the scope of the research problem
• (iii) It contributes to knowledge
• (iv) It saves time and effort
• (v) Acts as a framework for achieving the objective of the
research
• (vi) It provides a basis for empirical analysis 21
 Formulation of a Feasible Hypothesis

The following points must be borne in mind in order to

formulate a feasible hypothesis.
(a) It should not use people’s names.
(b) It should not exceed a particular period i.e., it
should have a time frame.
(c) It should avoid the usage of ‘conducive like verbs’
i.e could, might, would, etc.
(d) It should not make conclusion or value judgment.
(e) It should relate at least two variables.
(f) It should expressly state the relationship between
the variables.
22
 Decision Options on Types of Errors

Decision Options No oncoming vehicle There is an oncoming Vehicle

Decide to cross (a) Correct Decision (b) Wrong Decision the road
Type I Error

Decide not to cross (c) Wrong Decision (d) Correct Decision

the road Type II Error

23
 Degrees of Freedom

• This is the number of observation that is free to vary

within a constant parameter. To explain this concept,
suppose you are asked to list any four numbers that will
sum up to twenty. You are free to list any three
numbers, but for the last (fourth) number you must
make sure that the number makes for a total of twenty.
If the first three numbers were I, 8 and 5, then the
fourth number must be 6. In that wise you have four
numbers to list out and one restriction, so the degree of
freedom is four minus one which equals three.

24
 RESEARCH VARIABLES

Meaning of Variables
• These are properties being studied. It
is a symbol to which numerals are
assigned. A variable is a characteristic
of a person, object or phenomenon
that can take on different values e.g.
age, weight, distance between home
and libraries, monthly income, etc. 25
 Dependent Variable

Dependent Variable
• This is the element of relationship under
investigation, which is observed but not controlled
by the researcher. The dependent variable (itself not
liable to manipulation) is expected to be influenced
by the manipulation of the independent variable.

• For example in a beauty context, the dependent

variable (beauty) is influenced by those independent
variables such as height, size, colour and statistics.
26
 Independent Variable
Independent Variable
• This is the characteristic or attribute of a research object which
can be manipulated i.e. it is susceptible to treatment and
measurement by the researcher so as to determine its
influence on, or relationship with a dependent or criterion
variable.
Intervening Variable
• Just as the name connotes, it is a type of variable which
interferes between the Dependent and the Independent
Variables even though, it cannot be measured directly. It is
believed theoretically to affect directly, the Dependent
Variable. Examples of Intervening Variables (in the case of the
study on Beauty as the Dependent Variable) are: boredom,
fatigue, hunger, anxiety, hatred, etc. 27
 Scales for Measuring Variables
1. Continuous scale: This is made up of a
continuum of measurement such as pound,
weight in kg, etc.
2. Ordinal scale: This is used to measure
variables that can be categorized and the
categories can be ranked in increasing or
decreasing order eg age; class performance
etc.

28
 TYPES OF VARIABLES

TYPES OF VARIABLES
Background Variables
• These variables are often related to a number of
Independent Variables. They include such qualities as
educational qualification; gender; marital status and, etc.
 
Other Types of Variables
• Variable is a condition or characteristic or attribute which
the researcher controls or observes by way of manipulation.
Variable can take on various forms as highlighted below.
• Numerical variables: These are variables that can be
expressed in numbers.
29
 TYPES OF VARIABLES – contd 1

• Categorical variables: These are variables that can be

expressed in categories.
• Discrete Variable: This is a variable is of the nature
that takes on countable (integral) values such as
amount of balance in the bank account, number of
oranges in a basket, etc.
• Random Variables: When a variable can assume an
unpredictable value at any point in time such that the
value cannot be predetermined, it is said to be of
random nature. Example is the outcome of an ongoing
football match, the outcome of a coin tossed, etc.

30
 TYPES OF VARIABLES – contd 2

• Continuous Variable: A variable is said to be

continuous if it can assume any possible value within
a specified range or interval such as length, weight,
distance, age, etc.
• Quantitative Variable: Any variable that has the
quality of being counted or measured directly is said
to be a quantitative variable e.g. number of cattle in
a herd, number of policemen on a road.
• Qualitative Variable: When a variable possesses an
observable characteristic or attribute, it is said to be
a qualitative variable. Examples is in beauty (very
beautiful, not beautiful, very ugly) etc. 31
 Mediator and Moderator Variable

Mediator variable
• This is the middle variable / "middleman" between an
independent variable (IV) and a dependent variable (DV). 
Objective of the mediator variable is to explain the
relationship between IV & DV e.g. IV is not directly
influencing DV but rather IV is indirectly influencing DV
through mediator variable. Independent variable -->
Mediator variable --> Dependent variable.
• For example, salary (IV) is positively influencing education
(mediator variable) and then education is positively
influencing health-screening expenses (DV).  When the
effect of education is removed, the relationship between
salary and health-screening disappears.
32
 Moderator Variable
• Moderator variable is a third party variable that
modify the relationship between an independent
variable (IV) and a dependent variable (DV). 
Objective of the moderator variable is to
measure the strength of the relationship
between the IV & DV.  For example, if age is a
moderator variable between salary (IV) and
health-screening expenses (DV), then
relationship between salary & health-screening
can be stronger for older men and less strong for
younger men.
33
 Mediator vs Moderator Variables

• A mediating variable (or mediator) explains the

process through which two variables are related,
while a moderating variable (or moderator) affects
the strength and direction of that relationship.
• Including mediators and moderators in your
research helps you go beyond studying a simple
relationship between two variables for a fuller
picture of the real world. These variables are
important to consider when studying complex 
correlational or causal relationships between
variables.
34
 What’s the difference?

• You can think of a mediator as a go-between for two variables. For

example, sleep quality (an independent variable) can affect academic
achievement (a dependent variable) through the mediator of alertness.
In a mediation relationship, you can draw an arrow from an
independent variable to a mediator and then from the mediator to the
dependent variable.
• In contrast, a moderator is something that acts upon the relationship
between two variables and changes its direction or strength. For
example, mental health status may moderate the relationship between
sleep quality and academic achievement: the relationship might be
stronger for people without diagnosed mental health conditions than
for people with them.
• In a moderation relationship, you can draw an arrow from the
moderator to the relationship between an independent and dependent
variable. 35
 Mediating variables
• A mediator is a way in which an independent variable
impacts a dependent variable. It’s part of the causal
pathway of an effect, and it tells you how or why an effect
takes place.
If something is a mediator:
• It’s caused by the independent variable.
• It influences the dependent variable
• When it’s taken into account, the statistical correlation
between the independent and dependent variables is
higher than when it isn’t considered.
• Mediation analysis is a way of statistically testing whether
a variable is a mediator using linear regression analyses
 or ANOVAs. 36
 Moderating variables

• A moderator influences the level, direction, or

presence of a relationship between variables. It
shows you for whom, when, or under what
circumstances a relationship will hold.
• Moderators usually help you judge the 
external validity of your study by identifying the
limitations of when the relationship between
variables holds. For example, while social media
use can predict levels of loneliness, this
relationship may be stronger for adolescents
than for older adults. Age is a moderator here.
37
 Measurement in Research

• Measurement is the act or process or science of finding the size,

quantity or degree of the objects or items to be measured.
However, measurement in research is defined as the act of
assigning numerical values to such objects, items or events that is
being measured, according to laid down rules or procedures. Note
that, any attribute that is non-abstract (i.e. can be quantified)
would easily qualify as a measurable attribute, while, any attribute
that is abstract such as influence, love, etc cannot be measured on
any scale objectively.

• In order for a measurement scale to be good, it must possess

either of these or a combination of any two or all of the attributes
of a reliable measurement scale. These attributes are three viz:
• an origin; an order and a distance
38
 Attributes of Measurement Scale
  (i) Origin – This marks the specific beginning of a
scale. It can be zero or any other number in so far
as it is accepted as being the starting point.
(ii) Order – This serves as the end product of a
good measurement because it enables the
variables or objects of the study to be put in their
right positions on the scale.
(iii) Distance – This has to do with relative
positions of objects or variables on a given scale.
• Any measurement scale that does not possess at
least one of these three attributes should be
regarded as a weak measurement scale.
39
 MEASUREMENT SCALES

There are four basic measurement scales in use which are:

Nominal Scales
• These are scales that have no origin, no order and no distance. It is
indeed the weakest form of scale. The permissible statistics for their
evaluation are: percentages, mode, binomial test and chi- square test.
• Good examples of nominal variables are found in gender (male or
female) religion (Christianity or Islam), Marital status (married, single,
divorced or widow), Age (youth, adult or aged) etc.
Ordinal Scales
• They are those with order, but no origin and no distance. They are
better than Nominal in that they possess at least one of the three
attributes hitherto stated. Examples are: Weight, Height, Positions in
exam, Percentile, Social class, Educational qualification, etc. The
permissible statistics for their evaluation are percentiles, median,
spearman rank interval scales.
40
 MEASUREMENT SCALES - contd

Interval Scales
• In this case the scales possess both order and distance but without
specific origin unless such is specifically assigned unto them. By the rule
of the game their formulation are always founded on equal intervals
between the numbers. A good example of this type of scale is a
thermometer in which zero point is not indicative of there being no
temperature but rather that the temperature is at zero which by itself is
a temperature.
Ratio Scales
• This measures equal intervals with time (absolute) zero. It has the three
attributes or components of a good measurement scale (i.e. origin,
order and distance) hence it is the most qualitative of all the
measurement scales. It is the most widely used scale of measurement
in accounting and finance. Examples include: Time, Height, Distance,
Weight, and Temperature. The permissible statistics under ratio are
geometric mean, harmonic mean and coefficient of variation.
41