PART 4

CONCEPTS, VARIABLES AND HYPOTHESES

1. CONCEPT DEFINITION
2. VARIABLE DEFINITION
3. TYPES OF VARIABLES
4. RESEARCH PROPOSITIONS
5. HYPOTHESES:
HYPOTHESES DEVELOPMENT
RESEARCH AND NULL HYPOTHESES
6. MODELS: TYPES OF MODELS
7. RESEARCH ASSUMPTIONS
8. DELIMITATIONS
 4
1 3
2 CLARIFICATION
OF CONCEPTS
REVIEW OF
IDENTIFICATION OF FOPRMULATION OF
RELATED
RESEARCH ISSUE/QUESTIONS RESEARCH PROBLEMS IDENTIFICATION
LITERATURE
OF VARIABLES

STATEMENT OF
HYPOTHESIS

EXPLORATORY
SELECTION OF SURVEYS
5 RESEARCH
DESIGN DESCRIPTIVE
CASE STUDIES

LABORATORY
CAUSAL EXPERIMENTS

FIELD
EXPERIMENTS

PARTICIPANT
OBSERVATION
NON-PARTICIPANT
SELECTION OF PRIMARY
6 DATA COLLECTION DATA
QUESTIONNAIRES
TECHNIQUES

INTERVIEWS
SECONDARY DIRECT
DATA COMMUNICATION
PROJECTIVE
CLARIFICATION
OF CONCEPTS

IDENTIFICATION
OF VARIABLES

STATEMENT
OF HYPOTHESES
 WHAT IS A CONCEPT?
A Concept is …………..

- An idea, A notion

- That which exists in the mind as the product of

careful mental activity

- Is an abstraction from observed events

- A word that represents the similarities or

common aspects of events.
 WHAT IS A VARIABLE?

A variable is a qualitative or quantitative

entity that can vary or take on different
values.

In research, variables are things that are

measured and represent the concepts
studied.
TYPES OF VARIABLES

INDEPENDENT

DEPENDENT
 INDEPENDENT VARIABLES
2 TYPES:

1. Explanatory/Experimental Variables
- variables/Factors which are hypothesized to cause
certain effect upon the Dependent Variables

2. Extraneous/Moderating/Intervening Variables
- the distorting variables which tend to
invalidate or weaken the hypothesis that the
Explanatory Variables cause a change in the
Dependent Variables

TEST UNITS: is the entity acted upon by the

Independent Variable. The test unit
varies with the type of experiment.
 DEPENDENT VARIABLES
- The quantity that is affected by the Independent Variable
- Various Dependent Variables used in different disciplines

DISCIPLINE TYPICAL DEPENDENT

VARIABLE
Agriculture Bushels per acre
Psychology Attitude of people
Education IQ of students
biology Growth of plants
Engineering Strength of materials
Medicine Physiological measures
Management Performance of employees
Marketing Sales performance
Human Resources Labour productivity
Management
 RESEARCH PROPOSITIONS

A proposition states that “Every concept

either has a certain property” or “it stands in
a certain relationship with other concepts”.

Propositions are statements concerning

“Relationships among concepts” and that at
the explanatory level, a proposition is the
“Logical linkage among concepts”.
 STATEMENT OF HYPOTHESES
- “If you don’t know what you are looking for, then
you won’t find it”.

- Hypotheses statements state what you are

looking for.

- They anticipate the possible answers to the

research question.

- An important role of an hypothesis is to suggest

variables to be included.
- Hypotheses should be simple, precise
statements. They should not contain multiple
possibilities that make the task of accepting or
rejecting them difficult.

- Hypothesis should be based on statement that

can be quantified.

- They are statements that a researcher sets out to

accept or reject based on the evidence from the
data collected.

- The hypotheses should flow from the objectives.

 RESEARCH HYPOTHESES

The research Hypotheses states the

“expectation of the researcher in positive
terms”.

It identifies the variables or conditions

which, in causal relationship, will be
advanced to account for the results and
is often derived from theory.
 NULL HYPOTHESES
The Null Hypotheses which are the hypotheses of
“No Relationship or No Difference” is the one
actually tested.

It is an arbitrary convention hypothesizing that

any relation or difference in the findings is due to
chance or sampling error, and puts this
supposition to a probability test.

Theoretically, it is an hypothesis set up for

possible rejection and though the degree of
relationship or margin of difference need not be
zero, it frequently is.
 HYPOTHESES DEVELOPMENT

How does the researcher generate

hypotheses?

The answer is that “whatever information

is available is used to speculate on
which answers to the Research
Questions are possible and which are
likely”.
THREE MAIN SOURCES OF INFORMATION
TO DEVELOP HYPOTHESES

1. Draw upon previous research effort

2. Theory from other disciplines (marketing,

economics, psychology, sociology.)

3. On-hand experience with the problem

 A FRAMEWORK FOR HYPOTHESIS
DEVELOPMENT

Theory

Research Research Past Statement of Research

Objectives Question Experience Hypothesis Design

Other
Disciplines
 STATING THE NULL HYPOTHESIS
Four Alternatives
(using the example of the difference between two
sample means)

1. There is no difference between the mean of two populations from

which the two samples were drawn at random.

2. The two means in the two populations from which the samples were
respectively drawn at random are equal.

3. The mean of the sampling distribution of differences between an

infinite number of pairs of means randomly drawn samples equals
zero.

4. Any differences between two sample means drawn randomly from

their respective populations is due to sampling error - that is, such
differences can readily be attributed to chance.
 ACCEPTING OR REJECTING THE NULL
HYPOTHESES - WHAT IT MEANS
Testing the null hypothesis results in one of two
outcomes:

1. ACCEPTING (Failing to Reject) the Null Hypothesis as

TRUE, in which case it is concluded that any
differences in the result are:

a) Not statistically significant, therefore are

probably due to
b) Sampling Error or chance.

Accepting the Null Hypothesis also means that the

corresponding Research Hypothesis is not supported
or disconfirmed.
2. Rejecting the null hypothesis as FALSE, in
which case it is concluded that the
differences in the results are:

Statistically Significant, therefore are

probably due to
Some determining factors or
conditions, other than chance.

Rejecting the Null Hypothesis also means

the corresponding Research Hypothesis has
survived a test of disconfirmatory and, in
that sense, is supported.
More precisely, the differences in the data
reasonably cannot be attributed to chance
and the causal factors stated in the research
hypothesis may possibly explain the
differences, but not necessarily so as long as
there are plausible rival hypothesis also to
account for the data.

Obviously, the more tests of disconfirmatory

a given Research Hypothesis can survive, the
stronger is the evidence in its support.
 THEORETICAL FRAMEWORK
The Scientific approach to Social
Scientific investigation consists of:

1. A logical consistent theoretical

framework, and

2. A process of statistical measurement

and estimation/inference.
CONCEPTUAL MODEL/FRAMEWORK
DEFINITION: Simplified representation of a more
complex process or condition. It may be qualitative or
quantitative; normative or descriptive, and may range
from simple to very complex.

STATIC MODEL - Serves mainly to aid the

understanding of the user.

DYNAMIC MODEL - Permits manipulation of its

components to test alternative actions or predict future
conditions, or it may synthesize a whole from an
incomplete set of component parts.
 TYPES OF MODELS

1. QUALITATIVE MODEL
Commonly used in the analysis and prediction of
behaviour e.g. Personality characteristics and
Leadership behaviour.

2. QUANTITATIVE MODEL
Have the capability of providing the user with numerical
outputs. These might be measurement of customer
flows and expenditures, forecasts of return on
investment etc.
The Quantitative Models are limited in the range of
situations in the social science fields for which they
can be constructed.

This is due to:

- the difficulty in quantifying many of the
variables relating to behaviour
- the fact that statistical data of the industry
have not been thoroughly and systematically
collected and maintained
- the historical data often form the basis for
designing and/or evaluating a model.
 OTHER MODELS
1. PHYSICAL AND ECONOMIC IMPACT MODELS
Provide estimates of the size, and composition of activity – measure
changes in size etc. in order to evaluate promotional efforts and/or
assess changes in the demand of different market segments.

2. INPUT/OUTPUT MODELS OF AN ECONOMY

To estimate impacts of activities e.g. how a dollar spent by tourist is
redistributed.

3. BEHAVIOURAL MODELS
Motivational model e.g Maslow

4. TRANSACTIONAL ANALYSIS MODEL

To explain the underlying reason why people decide to travel and the
process which promotional efforts must undertake in order to
motivate a decision.
 PURPOSE OF MODEL
- The research process starts with a conceptual model
and its properties.
- The model is used at 3 different stages in the
research process:

1. Selecting significant problems when gaps in the

theory make exploratory necessary, or when
theoretical propositions require further testing.

2. Selecting appropriate empirical methods for the

research design.

3. After obtaining the empirical findings, the

interpretation with reference to the larger conceptual
scheme.
 …..Cont’d

- Conceptual Model is a heuristic devise serving

to guide the formulation and solution of
problems.

- Working Model: A tentative way of construing

a particular set of phenomena.

- Wholly Pragmatic Model: Helps in explaining

and making understandable already observed
behaviour and also suggest new behaviour to
look for.
 THEORETICAL
UNDERPINNING
- Special demand to rely heavily upon the
conceptual model since many available
research tools were originally designed
for use, not in the social science, but in
other disciplines e.g. psychology,
geography, sociology.

- Conceptual appropriateness

- Empirical accuracy
 RESEARCH ASSUMPTIONS

- Assumptions reduce the potential

criticism of the research

- Helps to isolate factors being tested

- Assumptions are a set of conditions that

the researcher presume are true and
valid as they act on the condition or
situation being studied.
 DELIMITATIONS
- Delimitations are acceptable limits
- Restrictions placed on the study to make it
doable
- Example:
-The sample taken from the population
-Delimit the study to certain population
only
-Due to cost and time factor
- May relate to people and things
 ASSIGNMENT 3

1. DEVELOPMENT OF CONCEPTUAL
FRAMEWORK

2. DEFINITIONS OF VARIABLES

3. FORMULATION OF RESEARCH
PROPOSITIONS/HYPOTHESES

4. DETERMINE THE THEORETICAL

UNDERPINNING