The research design refers to the overall strategy that you choose to integrate the different components of the study in a coherent and logical way, thereby, ensuring you will effectively address the research problem; it constitutes the blueprint for the collection, measurement, and analysis of data.
3. What is the study about?
Why is the study being made?
Where will the study be carried
out?
What type of data is
required?
Where can the required data be
found?
What periods of time will
study include?
What will be the sample
design?
How will the data be
analyzed?
In what style will the report be
prepared?
What technique of data
collection will be used?
4. Research Design
Sampling Design
Observational
Design
Statistical Design
Operational
Design
Deals with
methods of
selecting sample
Conditions under
which observations
to be made
How the data will
be analyzed
Techniques which
will be used to
carry out designs
CATEGORIZATION OF RESEARCH DESIGN
5. FEATURES OF RESEARCH DESIGN
It’s a plan specifying sources and types of information
Strategy used to gather and analyze the data
Also includes time and cost of budget
6. FEATURES OF GOOD RESEARCH
DESIGN
Flexible
Appropriate
Efficient
Economical
Minimizes bias and maximizes reliability of data
Gives smallest experimental error
Yield maximum information
Considers different aspects within limited
resources
13. EXTRANEOUS VARIABLE
Independent variables that are not related to the
purpose of the study but may affect the dependent
variable are called as extraneous variables.
16. CONTROL
Fertilizer X Everything same except fertilizer
Control is used to minimize the effect of extraneous independent variables.
Measure
Effect
17. RESEARCH HYPOTHESIS
Hypothesis is a predictive statement that relates an
independent variable to a dependent variable.
The fertilizer will significantly improve productivity of plants.
There will be significant difference between IQ of males and females.
The training session will improve performance of workers.
EXAMPLE
21. EXPERIMENTAL AND CONTROL GROUPS
USUAL STUDY PROGRAM SPECIAL STUDY PROGRAM
CONTROL GROUP EXPERIMENTAL GROUP
Group is exposed to usual condition Group is exposed to special condition
23. EXPERIMENT
Process of examining truth is called as
‘Experiment’
Absolute Experiment
Comparative
Experiment
Effect of fertilizer X on crop
yield
Impact of fertilizer X
compared to Y on crop
yield
27. Simplest and most loosely structured.
Objective is to explore and obtain clarity about problem
situation.
Discovery of ideas and insights.
Flexible in approach.
Mostly involve qualitative investigation.
Sample size is not strictly representative.
EXPLORATORY
28. TECHNIQUES FOR CONDUCTING EXPLORATORY RESEARCH:
1. Secondary Resource Analysis:
Previously collected findings
Easy to collect and less expensive
Points out that proposed research is redundant and already made
2. Structured and unstructured observations:
Exploring problem through observations
Example: Customer behavior for a window display of a shop
EXPLORATORY
29. TECHNIQUES FOR CONDUCTING EXPLORATORY RESEARCH:
3. Expert Opinion Survey:
No previous information is available
Seek help from expert
4. Focus Group Discussion
Discussion with individuals associated with problem under study
Carefully selected small set of individuals
EXPLORATORY
32. ASSESSMENT CRITERIA
1. Quality of discussion
2. Variety of points discussed
3. Number of students discussing in each group
4. Discipline and decorum maintained by the group
5. Speaking and listening ability of group
33. You are a business manager with the ITC
group of hotels. You receive a customer
satisfaction report on your international
hotels from the research agency to which
you had outsourced the work. What or
how will you evaluate the quality of work
done in the study?
34. Concerned with describing the characteristics of a particular individual,
group, situation, problem etc.
More structured and formal in nature than exploratory
EXAMPLE:
DISCRIPTIVE
Businessman wants to design advertising and promotion campaign
for high end watches
Descriptive Research
Who? What? When? Where? Why? How?
35. Narration of facts
Characteristics concerning individual, group or situation
Social research comes under this category
DISCRIPTIVE
39. DISCRIPTIVE
Cross-
Sectional
Selection of current subdivision of population and study the nature of
variables.
Carried out at a single moment in time
Applicable for specific period.
EXAMPLE:
Attitude/perception of Americans towards Asian-Americans, pre- and post- 9/11
40. DISCRIPTIVE
Cross-
Sectional
Situations in which population is not homogenous
(Religion, Gender, rural-urban etc.)
Multiple cross-sectional
studies
May be carried out at
same moment in time
May be carried out at
different time interval
Cohort Analysis
called as
called as
examples
Attitude of adult Vs teenage Americans
towards Asians post event
Predicting election results
41. DISCRIPTIVE
Longitudi
nal
Group of
Consumers
Study of this group over a stretched period of time
(purchase frequency, amount, behavior, pattern, types of products
purchased, medium of purchase etc.)
A study of single sample of the identified population
that is studied over a stretched period of time
43. EXPERIMENTAL
SALES MANAGER
(Pepsico India.)
Sales Personnel
3 Month Training
20% increase in sales
Conclusion by
manager
Training programme
is effective
Sales force from other
territories should also be
sent for same training
Decision
IS IT A GOOD DECISION ???
44. EXPERIMENTAL
Sales manager is trying to infer that sales training has caused
increase in sales.
i.e. Training is a Causal Variable and Sales growth is an Effect
Variable
::: This may not be true because :::
It may be caused due to:
Reduction in price of soft drink
Strike at the competitor’s plant
Increase in price of competitor’s product
Reduction in quality of competing product
Weather conditions etc.
45. EXPERIMENTAL
WHAT IS EXPERIMENT?
“It is used to infer causality.”
In an experiment, researcher manipulated one or more causal
variables (independent variables) and measure its effect on
dependent variables.
Relation is probabilistic in nature.
Impossible to prove causality.
Can only infer cause and effect relationship.
46. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
Pre-experimental
Quasi Experimental
True-experimental
Statistical
No Randomization
Randomization
47. RANDOMIZATION
Suppose you have five chocolates bars and total 8 friends to distribute these 5
chocolates to. Now how you are going to do this so the whole distribution process is
with a minimum of bias?
You may write down names of each of your friends on a separate small piece of paper,
fold all small pieces of papers so no one know what name is on any paper. Then you ask
someone to pick 5 names and give chocolates to first 5 names.
This will remove the bias without hurting any of your friend's feelings.
This is called as Randomization.
48. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
Treatment (X) and measurement of effect (O) on
dependent variable
Measurement before (O1) and after (O2) the
Treatment (X)
Group 1 – X O1
Group 2 – O2
O1 X O2
X O1
O = Measurement
X = Treatment
Do not use randomization
procedure to control
extraneous variable.
51. NORMAL DISTRIBUTION
Data can be "distributed" (spread out) in different ways.
There are many cases where the data tends to be around a central value with no bias left
or right, and it gets close to a "Normal Distribution"
52. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
O1 O2 O3 O4 X O5 O6 O7 O8
Experimental Group : O1 O2 O3 O4 X O5 O6 O7 O8
Control Group : O1 O2 O3 O4 O5 O6 O7 O8
O = Measurement
X = Treatment
Do not use randomization
procedure to control
extraneous variable.
53. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
R = Randomization
O = Measurement
X = Treatment
Researcher is able to
eliminate effect of
extraneous variables from
experimental and control
group.
Experimental Group : R O1 X O2
Control Group : O3 O4
Experimental Group : R X O1
Control Group : O2
Experimental Group 1 : R O1 X O2
Control Group 1 : O3 O4
Experimental Group 2 : R X O1
Control Group 2 : O2
54. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
It allows for statistical
control and analysis of
external variable.
Used to investigate effect of one independent variable
on dependent variable.
Independent Variable should be nominal scale
Example:
Measurement of sales for different price level
Price Level : Low Medium High
Stores : A B C
Use ANOVA technique to measure effect
55. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
Limitation of Completely Randomized Design: Extraneous Variables are
assumed to be constant
Using Randomized Blocks to minimize influence of one extraneous
variable
56. 1 2 3 4 5 6 7 8 9 10 11 12
LOW MEDIUM HIGH
SMALL SIZE
1 2 3 4 5 6 7 8 9 10 11 12
LOW MEDIUM HIGH
MEDIUM SIZE
1 2 3 4 5 6 7 8 9 10 11 12
LOW MEDIUM HIGH
LARGE SIZE
STORES
PRICE
STORES
PRICE
STORES
PRICE
57. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
Used to separate influence of 2 Extraneous Variables
Example: Sales Would be influenced by 2 extraneous variables.
Store Size
Packaging
I II III
1 (Small) X1 X2 X3
2 (Medium) X2 X3 X1
3 (Large) X3 X1 X2 Here, X1=Low price, X2=Medium, X3=High
58. Experimental
Design
Pre-
Experimental
One Shot Case
Study
One Group Pre
test-post test
Static Group
Quasi-
Experimental
Time Series
Multiple Time
Series
True-
Experimental
Pre-test-post-
test Control
Group
Post-test-only
Control Group
Solomon Four
Group
Statistical
Completely
Randomized
Randomized
Blocks
Latin Square
Factorial
Used to measure effect of 2 or more independent variable.
Allows interaction between 2 variables.
Ex. An individual may like mango as well as ice cream but may not like
Mango Ice Cream
Price
Stores
Small (B1) Big (B2)
Low (A1) A1B1 A1B2
Medium (A2) A2B1 A2B2
High (A3) A3B1 A3B2
60. • Research Design is a plan specifying sources and types of information
• Categories of research design – Sampling, Observational, Statistical,
operational
• A variable is defined as anything that has a quantity or quality that varies.
• Independent Variable influences change in dependent variable.
• Extraneous Variables - Variables that are not related to the purpose of the
study but may affect the dependent variable
• Control is used to minimize the effect of extraneous independent variables.
61. • Hypothesis is a predictive statement that relates an independent variable to a
dependent variable.
• Research in which independent variable is manipulated – Experimental
• Research in which independent variable is not manipulated – Non
Experimental
• Conditions under which groups are put is called as Treatments
• Process of examining truth is called as ‘Experiment’
• Exploratory Design - Objective is to explore and obtain clarity about problem
situation.
62. • Descriptive Design - Concerned with describing the characteristics of a
particular individual, group, situation, problem etc.
• Cross Sectional - Selection of current subdivision of population and
study the nature of variables.
• Longitudinal - Study of a group over a stretched period of time
• Experiment - It is used to infer causality
• Experimental Design – Pre experimental, Quasi experimental, True
Experimental, Statistical
Editor's Notes
This presentation demonstrates the new capabilities of PowerPoint and it is best viewed in Slide Show. These slides are designed to give you great ideas for the presentations you’ll create in PowerPoint 2010!
For more sample templates, click the File tab, and then on the New tab, click Sample Templates.