Types of research variables

Dr. Abdelhamid Kotb

 What is theory?
And Why is theory important?

2
 What is theory? The term theory is used to refer to ‘a
systematic body of knowledge grounded in empirical evidence
which can be used for explanatory or predictive purposes’
(Saunders et al. 2015: 37).
 Theory is a statement of relationships between constructs,
which cannot be observed directly (e.g., centralization,
satisfaction, or culture).
 The primary goal of a theory is to answer the questions of how,
when, and why. With explicit restrictions regarding space and
time.

3
 Whetten (1989) identified that theory is composed of four
elements, related to
1. What are the concepts or variables that the theory
examines?
2. How are these concepts or variables related?
3. Why are these concepts or variables related? This is the
critical element in a theory because it explains the
reasons for relationships between the concepts or
variables.
4. Who does this theory apply to; where does this theory
apply; when does this theory apply?

4
 A theoretical framework is a collection of theories and models
from the literature.
◦ It is a fundamental part of most research studies and
underpins the research questions.

◦ However, these can also be suggested by empirical evidence

(from an exploratory study, for example), from which you
subsequently develop a theory and construct propositions to
test.

5
Theory of Reasoned Action and Planned Behavior

6
How is theory developed?
 How theory is developed also provides a crucial reason for
recognizing relevant theory when writing your research
question and objectives.
 Your research project will be designed to test a theory or to
develop a theory.
 Where you wish to adopt a clear theoretical position that you will
test through the collection of data, your research project will be
theory‐driven and you will be using a deductive approach.
 Where you wish to explore a topic and develop a theoretical
explanation as the data are collected and analyzed, your research
project will be data‐driven and you will be adopting an inductive
approach.

7
 Why is theory important?
Theory published in the literature may inform your proposed
research question in several ways.
1) It will help you to formulate a research question that should
lead to a theoretical explanation, rather than just a descriptive
answer.
2) It will allow you to find out whether others have asked similar
questions to the question you propose.
3) Where you find that a similar research question to yours has
been addressed in the literature,


8
 Why is theory important?
4) you will be able to learn about the context within which it was
explored and how the research was conducted.
5) This may help to focus your question to provide you with a set
of variables to test, or concepts to explore, to determine
whether, how and why they are related in the context of your
own research project.
6) it is crucial to be able to explain how variables or concepts are
related and why they are related.

9
Why is theory important?
7) It will be important to discuss how the results of your research
relate to theory, to be able to assess that theory in the context
of your work and to demonstrate the theoretical contribution,
no matter how limited, of your research.

10
Types of research variables

1
1
 Constructs (also called latent / unobserved variables): measure
concepts that are abstract, complex, and cannot be directly
observed by means of (multiple) items. Constructs are
represented in path models as circles or ovals (e.g.,
centralization, satisfaction, or culture). .
 Indicator (observed variable): single variable used in
conjunction with one or more other variables to form a
composite measure.
 Independent variable (IV) : an independent variable is a
predictor variable that influences the values of a dependent
variable (also called antecedent, or predictor variables).

12
 Dependent variable (DV): A dependent variable is a variable
whose values are influenced by one or more independent
variables ; they are the outcomes or results of the influence of
the independent variables (other names for dependent
variables are outcome).
 Constructs that only act as independent variables are generally
referred to as exogenous latent variables and are on the
extreme left side of the structural model.
◦ Exogenous latent variables only have arrows that point out of them
and never have arrows pointing into them.

13
 Endogenous latent variables: constructs considered as
dependent in a structural model (i.e., those that have an arrow
pointing into it) and are on the right side of the structural
model.
 Constructs that operate as both independent and dependent
variables in a model also are considered endogenous and, if
they are part of a model, appear in the middle of the diagram.

14
The model has two exogenous constructs‐corporate social responsibility (CSOR) and
attractiveness (AITR)‐and one endogenous construct, which is competence (COMP). Each of these
constructs is measured by means of multiple indicators. For instance, the endogenous construct
COMP has three measured indicator variables, comp_1 to comp_3.

15
Path models are made up of two elements
 the structural model (also called the inner model in PLS‐SEM),
which describes the relationships between the latent variables,
 the measurement models, which describe the relationships
between the latent variables and their measures (i.e., their
indicators).
 When a structural model is being developed, two primary issues
need to be considered: the sequence of the constructs and the
relationships between them.
 Both issues are critical to the concept of modeling because they
represent the hypotheses and their relationship to the theory
being tested.

16
17
 The sequence of the constructs in a structural model is based on
theory, logic, or practical experiences observed by the
researcher.
 Determining the sequence of the constructs is seldom an easy
task because contradictory theoretical perspectives can lead to
different sequencing of latent variables.
◦ For example, some researchers assume that customer satisfaction
predicts corporate reputation, while others argue that corporate
reputation predicts customer satisfaction.
◦ Theory and logic should always determine the sequence of
constructs in a conceptual model, but when the literature is
inconsistent or unclear, researchers must use their best judgment
to determine the sequence.

18
 The mediating variable (MV) is the means by which the
independent variable brings about changes on a dependent
variable and helps explain the relationship between them (IV →
MV → DV).
 Mediation is often used to provide a more accurate
explanation for the effect the antecedent has on the
dependent variable.
 the most common application of mediation is to "explain" why a
relationship between an exogenous and endogenous construct
exists.

19
 For example, Intelligence leads to increased performance ‐ but
not in all cases, as not all intelligent people are high performers.
 So, some other variables are needed to explain the reason for
the inconsistent relationship between IV and DV.
 This other variable is called a mediator. In this example, work
effectiveness, may be a good mediator.
 Thus, the direct relationship between intelligence and
performance is better explained through the mediator of work
effectiveness.
 The logic is, even if you are intelligent, if you don't work
smarter, then you won't perform well.

20
In the diagram, the intervening process (mediating effect) is modeled as
satisfaction. If a respondent perceives a company to be highly reputable,
this assessment may lead to higher satisfaction levels and ultimately to
increased loyalty.

21
 A moderating variable (Interaction) is a variable that affects the
direction and/or the strength of the relationship between the
dependent and independent variables.
 interaction effects are the joint effects of two predictor
variables in addition to the individual main effects.
 Interactions enable more precise explanation of effects by
providing a method for explaining not only how X effects Y, but
also under what circumstances the effect of X on Y changes
depending on the moderating variable of Z.

22
 For example, income has been shown to significantly affect the
strength of the relationship between customer satisfaction and
customer loyalty (Homburg & Giering, 2001). In that context,
income serves as a moderator variable on the sa sfac on →
loyalty relationship, as shown in Exhibit 2.3.

Specifically, the relationship

between satisfaction and
loyalty has been shown to be
weaker for people with high
income than for people with
low income.

23
24