100% found this document useful (1 vote)
196 views40 pages

Research Design: - Framework/Blueprint For Research Plan of - Specifies The Details. - Components (Specification Of)

This document outlines various research designs and methods. It discusses exploratory research design which aims to provide insight and understanding of problems through methods like surveys of secondary data, experience surveys, qualitative research, and pilot surveys. Descriptive research design aims to describe characteristics and relationships and can be classified as longitudinal, cross-sectional, or causal. Experimental designs allow control and manipulation of variables to determine causal relationships and include types like after-only, before-after, and Solomon's four-group designs.

Uploaded by

Priyanka Jain
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
196 views40 pages

Research Design: - Framework/Blueprint For Research Plan of - Specifies The Details. - Components (Specification Of)

This document outlines various research designs and methods. It discusses exploratory research design which aims to provide insight and understanding of problems through methods like surveys of secondary data, experience surveys, qualitative research, and pilot surveys. Descriptive research design aims to describe characteristics and relationships and can be classified as longitudinal, cross-sectional, or causal. Experimental designs allow control and manipulation of variables to determine causal relationships and include types like after-only, before-after, and Solomon's four-group designs.

Uploaded by

Priyanka Jain
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 40

RESEARCH DESIGN

• Framework/Blueprint for research plan of


action, giving a general statement of the
methods to be used.
• Specifies the details.
• Components (Specification of):
– Type of Design
– Information needed
– Measurement & Scaling Techniques
– Questionnaire construction & pretesting
– Sampling process & size
– Data Analysis Plan
CLASSIFICATION
• EXPLORATORY RESEARCH
– To provide insight into, & an
understanding of, the problem (s) faced.
• CONCLUSIVE RESEARCH
– To assist in determining, evaluating, &
selecting the best course of action to take
in a given situation.
– DESCRIPTIVE RESEARCH
– CAUSAL RESERACH
EXPLORATORY RESEARCH
DESIGN
• To explore or search through the problem/
situation for sharper focus of situation
• PURPOSES:
– Identification of problem
– Precise formulation of the problem
– Identify alternative courses of action
– Develop hypothesis
– Isolate key variables & their relationships
• Not a conclusive study so design is flexible &
informal
EXPLORATORY RESEARCH
DESIGN: METHODS
• SURVEY OF SECONDARY DATA
• EXPERIENCE SURVEY
• QUALITATIVE RESEARCH
• PILOT SURVEY
EXPLORATORY RESEARCH DESIGN:
SURVEY OF SECONDARY DATA
• Existing Data or Literature
• Located & collected rapidly, easily & inexpensively
• A prerequisite to the collection of primary data
• May not provide solution but provide direction
• CLASSIFICATION:
– Internal
• Ready to Use
• Requires Further Processing
– External
• Published Material
• Computerized Databases
• Syndicated Services
EXPERIENCE SURVEY
• Obtaining information from Experienced &
Knowledgeable persons
• Judgement Sampling + Referral Sampling
• Unstructured & Informal
QUALITATIVE RESERACH
• An unstructured exploratory research
methodology based on small sample that
provides insights & understandings
• Quantitative research must be preceded by
appropriate qualitative research
• Not possible/desirable to use structured/formal
method:
– People are unwilling/unable to answer
– Unable to provide accurate answer to
questions that tap their subconscious
– Values, emotional drives, motivations are
disguised
• Not conclusive & should not be generalized.
QUALITATIVE RESEARCH:
CLASSIFICATION
• DIRECT APPROACH (NON-DISGUISED):
– FOCUS GROUPS
– DEPTH INTERVIEWS
• INDIRECT APPROACH (DISGUISED)
PROJECTIVE TECHNIQUES:
– ASSOCIATION TECHNIQUES
– COMPLETION TECHNIQUES
– CONSTRUCTION TECHNIQUES
– EXPRESSIVE TECHNIQUES
FOCUS GROUP INTERVIEWS
• Joint informal interview in the form of free
flowing group-discussions
• To gain insight by listening to people from
target group
• Most important qualitative research
technique
• Group size- 8 to 12, Group composition-
homogeneous, prescreened, Time duration- 1
to 3 hrs
• Role of Moderator
DEPTH INTERVIEWS
• Unstructured direct, personal interview, on
one to one basis
• Single respondent is probed to uncover
underlying motivations, beliefs, attitudes &
feelings on a problem/situation.
• Interviewing Techniques:
– Laddering: Issue characteristics User characteristics
– Hidden Issue Questioning
– Symbolic Analysis: Comparing with opposites.
PROJECTIVE TECHNIQUES
• Unstructured, indirect form of questioning,
encouraging respondents to project their
underlying motivations, beliefs, attitudes etc
• Respondents are asked to interpret the behavior of
others rather than own
• Classification:
– Association Techniques: Word Association
– Completion Techniques: Sentence Completion,
Story Completion
– Construction Techniques: Picture Response,
Cartoon Tests
– Expressive Techniques: Role Playing, Third-
Person Technique
PILOT STUDY
• Small scale exploratory research survey
• Uses Sampling but rigid standards are not
applied.
• Primary data, Qualitative Analysis.
• To identify and eliminate potential problems
in research administration
DESCRIPTIVE RESEARCH
DESIGN
• To know the characteristics of certain groups
or situations
• To facilitate description & inference building
about parameters & their relationships.
• Uncover causal relationship but not
established
• Assumes prior knowledge about the problem
situation
• Structured & formal design
• Methods: Quantitative Analysis of Secondary
data, Surveys, Panels, Observational Study
DESCRIPTIVE RESEARCH
DESIGN: CLASSIFICATION
• LONGITUDINAL RESEARCH / PANEL
RESEARCH DESIGN
• CROSS-SECTIONAL RESEARCH
DESIGN
– SINGLE CROSS-SECTIONAL DESIGN
– MULTIPLE CROSS-SECTIONAL
DESIGN
• COHORT ANALYSIS
LONGITUDINAL RESEARCH
• Based on PANEL DATA & PANEL METHOD
• Continual or periodic collection of
information from fixed sample of
respondents
• Related to repeated measurement of the
same variable (s)
• To detect changes because of repeated
measurements
• OMNIBUS PANEL:
– Information collected from fixed sample
may vary over time or may vary between
individuals
LONGITUDINAL RESEARCH
(Cross-Sectional Data)
Brand Period 1 Survey Period 2 Survey
Purchased
Brand A 200 200

Brand B 300 300

Brand C 500 500

Total 1000 1000


LONGITUDINAL RESEARCH
(Longitudinal Data: Brand-Switching Matrix)
Brand Purchased Brand Purchased in Period 2
in Period 1
Brand A Brand B Brand C Total

Brand A 100 50 50 200

Brand B 25 100 175 300

Brand C 75 150 275 500

Total 200 300 500 1000


CROSS-SECTIONAL DESIGN
• Aimed at taking one time stock of the
situation
• Collection of information only once
• Most frequently used Descriptive
Design.
• SINGLE CROSS-SECTIONAL DESIGN:
– One sample of respondents &
information is obtained from the sample
only once
CROSS-SECTIONAL DESIGN
• MULTIPLE CROSS-SECTIONAL DESIGN:
– Two or more samples of respondents &
information is obtained from each sample
only once (Information at different times)
– Allows comparison at the aggregate level
but not at the individual level
– COHORT ANALYSIS:
• Series of surveys at appropriate time
interval from the groups of respondents
(Cohorts) who experience the same
event within the same time interval
COHORT ANALYSIS:EXAMPLE
(Consumption of Soft Drink by various Age Groups)

Age 1980 1990 2000 2010

10-19 53 63 73 81

20-29 45 61 76 76

30-39 34 47 68 71

40-49 23 41 59 68

50+ 18 29 50 52
CAUSALITY
• The basis of classification in between
Experimental & Non-Experimental designs
• Conditions for making Causal Statement:
– Concomitant Variation
– Time Sequence
– Absence of Extraneous Factor (s)
• Control of Extraneous Variables:
– Physical Control
– Randomization
– Experimental Design
EXPERIMENTATION
• Manipulation of variable (s) and control
of other variables and examining their
effect
ELEMENTS
INPUTS OUTPUT
EXPLANATORY & DEPENDENT
TEST UNITS
EXTRANEOUS VARIABLE
VARIABLES
VALIDITY OF EXPERIMENT
• INTERNAL VALIDITY
• EXTERNAL VALIDITY
• Factors affecting Validity:
– HISTORY
– MATURATION
– TESTING EFFECT
– INTRUMENTATION
– SELECTION BIAS
– TEST UNIT MORTALITY
EXPERIMENTAL DESIGN
• Stronger & more reliable basis for the
existence of Causal Relationship
• All extraneous variables are eliminated
through the Use of Randomization & Control
Group (s)
TYPES OF EXPERIMENTAL DESIGN
 AFTER ONLY WITH ONE CONTROL GROUP
 BEFORE-AFTER WITH ONE CONTROL GROUP
 EX-POST FACTO DESIGN
 SIMULATED BEFORE AFTER DESIGN
 SOLOMON’S FOUR GROUPS DESIGN
 MULTIPLE TIME SERIES DESIGN
 C.R.D.
 R.B.D.
 L.S.D.
 FACTORIAL DESIGN
AFTER-ONLY WITH ONE
CONTROL GROUP
EXPERIMENTAL GROUP: R X O1
CONTROL GROUP : R O2

• Treatment Effect = O1 - O2
• Very widely used in marketing
• Control over ‘Testing’ & ‘Instrument Effect’
• No pre- observations so large samples are
required
BEFORE-AFTER WITH ONE
CONTROL GROUP
EXPERIMENTAL GROUP: R O1 X O2
CONTROL GROUP : R O3 O4

• TREATMENT EFFECT = (O2 O1) (O4 O3)


• Treatment Effect contains ‘Testing Effect’
• Effective relative comparison
EX-POST FACTO DESIGN
EXPERIMENTAL GROUP: X R O1
CONTROL GROUP : R O2

• TREATMENT EFFECT = O1 O2
• Used in Advt. Study & study of Commercial
Media
• No ‘Testing Effect’ & more realistic study
SIMULATED BEFORE-AFTER
DESIGN
EXPERIMENTAL GROUP: R X O2
CONTROL GROUP : R O1

• TREATMENT EFFECT = O2 O1
• No ‘ Testing Effect ’
• Used in Advt. Research
SOLOMON’S FOUR – GROUP
DESIGN
EXPERIMENTAL GROUP 1 : R O1 X O2
CONTROL GROUP 1 :R O3 O4
EXPERIMENTAL GROUP 2 : R X O5
CONTROL GROUP 2 :R O6

• Also known as FOUR GROUP SIX STUDY


DESIGN
SOLOMON’S FOUR – GROUP
DESIGN
• Combination of ‘BEFORE-AFTER WITH ONE
CONTROL GROUP’ & ‘AFTER-ONLY WITH ONE
CONTROL GROUP’ Designs. So also known as ‘THE
IDEAL CONTROL EXPERIMENT’
• Not very commonly used because of high cost &
high efforts
• TREATMENT EFFECT = O5 O6
• TESTING EFFECT = (O2 O1) (O4 O1 + O3)
2
• EXTRANEOUS EFFECT = O6 ( O 1 + O3 )
2
MULTIPLE TIME SERIES DESIGN
EXPERIMENTAL GR: O1 O2 O3…..X…. O4 O5 O6
CONTROL GR : O1 O2 O3…. ...…. O4 O5 O6

• Extension of BEFORE-AFTER WITH ONE


CONTROL GROUP DESIGN
• Also known as ‘CONTINUOUS DIARY PANEL
DESIGN
• Better analysis of trend
• Higher cost.
MANIPULATION / CONTROL OF MORE
THAN ONE LEVELS OF TREATMENTS

COMPLETELY RANDOMIZED
DESIGN (C.R.D.)

• To study the effect of different levels/


categories (Treatment) of explanatory
variable on dependent variable
• Treatment are assigned completely at
random to experimental units
COMPLETELY RANDOMIZED
DESIGN (C.R.D.)
• Uses only the principles of Replication
& Randomization but ignores Local
Control
• Useful only in small preliminary
experiments where experimental units
are homogeneous
• Analysis - ONE WAY CLASSIFICATION
RANDOMIZED BLOCK DESIGN
(R.B.D.)
• It is possible to separate out the effect
of one extraneous factor
• All three principles: RANDOMIZATION,
REPLICATION & LOCAL CONTROL are
used.
• There are ‘m’ treatments, each being
replicated ‘n’ times. So, N= m.n
experimental units & units are not
homogeneous.
RANDOMIZED BLOCK DESIGN
(R.B.D.)
• R.B.D. consists of two steps :
STEP 1
• The experimental units are allocated to
‘n’ homogeneous BLOCKS, number of
units within a block = number of
treatments.
STEP 2
• Assign treatments at random to units of
a block.
RANDOMIZED BLOCK DESIGN
(R.B.D.)
• By restricting randomization to a block
control over extraneous factor (error) is
obtained.
• Most popular but block size ( no. of
treatments ) should be small o.w. blocks
may not be internally homogeneous.
• Analysis: TWO WAY CLASSIFICATION
of the data & may be used to examine
different levels of two factors.
LATIN SQUARE DESIGN (LSD)
• Principle of LOCAL CONTROL is used
by grouping the units in two ways:
Rows & Columns
• Each way corresponding to a source of
variation among the units.
• It allows the control & measurement of
effect of two extraneous variables on
dependent variable.
LATIN SQUARE DESIGN (LSD)
• Condition : Two extraneous variables
have same number of categories as that
of treatments. So number of treatments
= number of replications = m
• Total number of experimental units = m2
( arranged in m rows & m columns,
according to two sources of variation )
LATIN SQUARE DESIGN (LSD)

• The m treatments are allotted to these m2


units at random s.t. each treatment occurs
once & only once in each row & each
column. This is incomplete 3 – way layout
C1 C2 C3 C4 C5
Ex. R1 A E D C B
R2 D B A E C
R3 B A C D E
R4 C D E B A
R5 E C B A D

You might also like