Data Analysis in Business

Lecture 1
 Variables
• A variable is anything that can take on differing or
varying values.
Type Of Variables:
1. Independent Variable: A variable that influences
the dependant or criterion variable and accounts
for (or explains) its variance
2. Dependant Variable: The variable of primary
interest to the study, also known as criterion
variable.
Difference Between Independent and Dependant
Variable

Independent Variable Dependant Variable

 Presumed cause  Presumed effect

 Stimulus  Response
 Predicted from…  Predicted to…
 Antecedent  Consequence
 Manipulated  Measured Outcome
 Predictor  Criterion
 Types of Variables
3. Moderating Variable: .It is a second independent variable
that is included because it is believed to have a significant
contributory or contingent effect on the originally stated
IV-DV relationship
When we position a given variable in one hypothesis as a
moderator we:
a. Recognize it as an important impactor
b. Which we are interested in treating as a subsidiary IV
c. As one upon which the function of one major IV is
contingent
 Types of Variables
• Example of MV:
I. A promotion campaign (IV) brings
increased savings (DV)
i.e.;
Promotion
campaign Increase in Savings
 Types of Variables
But the effect is more pronounced among small
a/c savers,
Therefore,
Increase in
Promotion
Savings
Campaign

I DV
V Small a/c
Savers

MV
 Types of Variables
• Whether a variable is independant or
moderating variable depends often on the
way we pose the hypothesis. For example if
we are studying the impact of size of a/c on
various saver actions we might well treat
“small a/c savers” as an IV rather than MV
 Types of Variables
II. Example of Moderating Variable:
The introduction of a four-day week (IV) leads to
an increase in productivity (DV)
i.e.;

Four- Day Increased

Week Productivit
y
 Types of Variables
Especially among young workers (MV)
Therefore,

Four- Day Increased

Week Productivity

IV DV

Young
Workers M
V
 Types of Variables
If you were interested in studying the impact of
length of work week, you would make the length of
week as IV. But if you were focusing on the
relationship of age of workers and productivity you
might use work week as a moderating variable
 Types of Variables
4. Intervening Variable: (IVV)
That factor which theoretically affects the observed
phenomenon but cannot be seen, measured or
manipulated, its effect must be inferred from the effects
of IV & MV on the observed phenomenon.
Example: A promotion campaign (IV) brings increased
savings (DV) By increasing motivation to save (IVV) but
the effect is more pronounced in small a/c savers (MV)
 Types of Variables
• Example

Promotion Increase in Increased

campaign motivation savings
IV IVV DV
Small
a/c savers

MV
 Types of Variables
• Explanation: An intervening variable is one that
surface between the time their impact is felt on it.
There is a time dimension to the intervening
variable. The IVV surfaces as a function of the IV
operating in any situation and helps to
conceptualize and explain the influence of IV on the
DV.
• Moreover it is easy to differentiate among an IV,
IVV and a MV
Contd.
 Types of Variables
• The IV helps to explain the variance in the DV. The
intervening variable surfaces at a time as a function
of IV which also helps us to conceptualize the
relationship between the IV, DV and MV has a
contingent effect on the relationship between two
variables.
 Other Types of Variables
5. Discrete / Classificatory/ Categorical Variables : Any
variable that has a limited number of distinct
values
Which consists of only two variables i.e.; male and
female
6. Continuous Variable: Any variable that has an
infinite number of values
Which is continuous i.e.; the age of the person,
sales, profitability
7. Extraneous Variable: Which has a cause and effect
relationship
 A Research Problem
• Defining a research problem is the fuel that
drives the scientific process, and is the
foundation of any research method and
experimental design, from true experiment to
case study.
 Steps to Defining Research
Problems
Area
a Narrow
N SSources Evaluate Statement

Balance

Format

Feasible Understood Researchable

 Steps to Defining Research Problems
Area: Decide on the general area of study or investigation
• General area of investigation :
• Sustain your interest & stimulate your imagination
• Within your range of competencies
• Manageable in size
• Potential to make a contribution to body of knowledge
• Based on obtainable data
• Demonstrate your independent mastery of both the
subject and method
 Steps to Defining Research Problems
• General area of investigation, continued
• “My study is about…….” or “the purpose of …….”
• Make it twelve words or less if possible
• Becomes a working title for your research.
• Is it researchable?
• Example –My study is about the effect of size and color of
screen icons on user perceptions
 Steps to Defining Research
Problems
Narrow: Narrow the general topic down
• To a specific statement of the research problem
• Use a single paradigm if possible
• Difficulty --the topic & research question must be
formulated before you have a thorough
understanding of research
 Steps to Defining Research
Problems
• Narrow the general topic down
• Literature review usually limited at this point
• Must make wise choices about what to investigate,
study, explore
• Is the topic better suited to a qualitative or
quantitative paradigm?
• Nature of the problem
– 􀂍Previously studied, much literature –quantitative
• Exploratory study, lacking theory base --qualitative
 Steps to Defining Research
Problems
Sources: Understand sources from which you define the
problem
• Experience
􀂍Experts that you know
􀂍Deductions from theory
􀂍Readily available problem
􀂍Review of literature
􀂍Limits of sources
 Steps to Defining Research
Problems
Evaluate: Evaluate the potential of the problem
Important enough to merit investigation or study?
Does it meet criteria?
Criteria:
􀂍Will findings make a contribution to body of knowledge?
􀂍Will findings make a difference for others?
􀂍Lead to definition of new problems or other research?
􀂍Really researchable?
􀂍Knowledge & experience in the problem area?
􀂍Information or data available to you?
􀂍Complete in the allotted time frame?
􀂍Simple enough for your first study?
 Steps to Defining Research
Problems
• Statement: A good problem statement
• Clarify exactly what you want to determine or solve
• Scope limited to a specific question; sub-questions
• Operationally defines key terms
Operational definition (quantitative study)
􀂍Defines the variables operationally
􀂍Defines a concept in terms of the operations or processes that
will be used to measure or manipulate the concept
􀂍Tentative definition (qualitative study)
􀂍Emerge from data collection
􀂍Not usually included in a list of definitions but is/are
tentative pending visiting the field setting to gather info
 Steps to Defining Research
Problems
• Balance: Balance between general & specific
in problem statement
􀂍Avoid trivial problems that are meaningless
􀂍Broad enough to be significant according to the
criteria you establish
􀂍Specific enough to be feasible for the research
situation
 Steps to Defining Research
Problems
• Format: Format of problem statement –how
you state the problem
􀂍Question –implies relationship between two or
more variables
􀂍Statement –describes the scope of your work
􀂍Hypothesis --relationships
􀂍Objective –achieve, measure
 Steps to Defining Research
Problems
• Researchable: Problem stated in a way that it is
researchable
􀂍Is research into the “question” possible?
• Feasible Understood: Clear & feasible problem
statement
􀂍Can it be understood by others?
􀂍Can you describe it concisely, clearly?
􀂍Do you demonstrate understanding of the area
being investigated, studied?
 Distinguish Between Researchable and Non
• ResearchableResearchable Problems
problems imply the possibility of
empirical investigation:
a. What are the achievement and social skill differences between children
attending an academically or socially oriented pre-school program?
b. What is the relationship between teachers’ knowledge of assessment
methods and their use of them?
 Distinguish Between Researchable and Non
Researchable Problems

Non-researchable problems include explanations

of how to do something, vague propositions, and
value-based concerns:
a. Is democracy a good form of government?
b. Should values clarification be taught in public
schools?
c. Can crime be prevented?
d. Should physical education classes be dropped from
the high school curriculum?