Discriminant Analysis,

Factor Analysis, Cluster

Analysis, Multidimensional
scaling and Conjoint
analysis
Discriminant Analysis

A technique for analyzing

marketing research data when
the criterion/dependent variable
is categorical and the
predictor/independent variable
are interval in nature
Objectives of discriminant
analysis
1. Development of discriminant functions
(linear combinations of the predictor
which will best discriminate between the
categories of the criterion)
2. Examination whether significant
differences exist among the group
3. Determination of which predictor
variables contribute to most of the
intergroup differences
4. Classification of cases to one of the
groups based on the values of the
predictor variables
5. Evaluation of the accuracy of
classification
Similarities and Differences
among ANOVA, Regression and
Discriminant Analysis
 Similarities:
 Number of dependent variables - 1
 Number of independent variables -
Multiple
 Differences
 Nature of dependent variable:
Metric, Metric, Categorical
 Nature of independent variable:
categorical, metric, metric
Steps involved in Discriminant
Analysis
1. Formulate the problem
2. Estimate the discriminate
function coefficient
3. Determine the significance of
the discrimination function
4. Interpret the results
5. Assess the validity of
discriminant analysis
Marketing Research Applications

 In term of demographic characteristics, how

do customers who exhibit store loyalty differ
from those who do not
 Do heavy, medium and light users of soft
drink differ in terms of their consumption of
frozen food
 What psychographic characteristics help
differentiate between price-sensitive and
non-price sensitive buyers of groceries
 Does the various market segment differ in
their media consumption habits
 In terms of lifestyles, what are the differences
between heavy patrons of general stores and
Mall stores
Factor analysis
A class of procedures primarily used for data
reduction and summarization
It is use in the following circumstances
1. To identify underlying dimensions/factors
that explain correlations among a set of
variables
2. To identify a new, smaller set of
uncorrelated variables to replace the
original set of correlated variables in
subsequent multivariate analysis
3. To identify a smaller set of salient variables
from a larger set for use in subsequent
multivariate analysis
Steps involved in factor
analysis
1. Formulate the problem
2. Construct the correlation matrix
3. Determine the method of factor
analysis
4. Determine the number of factors
5. Rotate the factors
6. Interpret the factors
1. Calculate the factor scores
2. Select the surrogate variables
7. Determine the model fit
Marketing Research
Application
 Identifying the underlying variables on which to group
the customers ( new car buyers may be grouped
based on the relative emphasis they place on
economy, convenience, performance, comfort and
luxury.)
 Determining the brand attributes that influence
consumer choice. (toothpaste brands might be
evaluated in terms of protection against cavities,
whiteness of teeth, taste, fresh breath and price)
 Understand the media consumption habits of the target
market. The users of ready to eat food may be heavy
viewers of cable TV, see a lot more movies, and listen
to western music
 Identifying the characteristics of price-sensitive
consumers (these consumers might be methodical,
economy minded and home centered)
Cluster Analysis
 A class of techniques used to classify
objects/cases into relatively homogeneous
groups called clusters.
 Objects in each cluster tend to be similar to
each other and dissimilar to other cluster
 A.k.a. classification analysis/numerical
taxonomy
 Both cluster analysis and discriminant
analysis is used for classification but in
cluster analysis there is no a priori
information about the group/cluster
membership
Steps involved in cluster
analysis
 Formulate the problem
 Select a distance measure
 Select a clustering procedure
 Decide on the number of cluster
 Interpret and profile cluster
 Assess the validity of clustering
Marketing Research
Application
 Segmenting the market
 Understanding buyer behaviour
 Identifying new product
opportunities
 Selecting test markets
 Reducing data
Multidimensional scaling
 A class of procedures for representing
perceptions and preferences of
respondents spatially by means of a
visual display.
 MDS has been used in marketing to
identify:
1. The number and nature of dimensions
consumer use to perceive different
brands in the market place
2. The positioning of current brands on
these dimensions
3. The positioning of consumers’ ideal
brand on these dimensions
Steps in MDS

 Formulate the problem

 Obtain input data
 Select an MDS procedure
 Decide on the number of
dimensions
 Label the dimensions and
interpret the configurations
 Assess reliability and validity
Marketing Research
Application
1. Image measurement
2. Market segmentation
3. New product development
4. Assessing advertising
effectiveness
5. Pricing analysis
6. Channel decisions
7. Attitude scale construction
Conjoint Analysis
 A technique that attempts to
determine the relative importance
consumers attach to salient
attributes and the utilities they attach
to the levels of attributes
 It seeks to develop the part-
worth/utility functions describing the
utility consumers attach to the levels
of each attribute.
 It is complementary to MDS
Steps involved in Conjoint
Analysis
 Formulate the problem
 Construct the stimuli
 Decide on the form of input data
 Select a conjoint analysis
procedure
 Interpret the results
 Assess reliability and validity
Market Research Application
 Determining the relative importance
of attributes in the consumer choice
process
 Estimate market share of brands that
differ in attribute levels
 Determining the composition of most
preferred brand
 Segmenting the market based on
similarity of preferences for attribute
levels