Session 17 & 18 - Group D - Operations - Factor & Discriminant Analysis

Comparison of Factor Analysis and Discriminant Analysis

Definition

Factor analysis is essentially a data reduction technique in which the objective
is to represent the original pool of attributes in terms of a smaller number of underlying dimensions or factors. After the factors have been identified, the brands' ratings on these factors are used to position the brands in perceptual space.

Discriminant analysis also requires that respondents provide attribute ratings. Like factor analysis, an objective of this method is to reduce the number of attributes to a smaller number of underlying dimensions. However, unlike factor analysis, discriminant analysis focuses on attributes which show differences between brands. 
This approach will tend to ignore attribute ratings which show a large variation within brands; these latter dimensions will not be "significant" in discriminant analysis. In discriminant analysis, sets of observations represent different "groups." As its name suggests, discriminant analysis will identify those underlying dimensions which are most useful in discriminating among groups. 

Factor analysis and discriminant analysis are both data reduction techniques, and both are useful in condensing a large number of product attribute measures into a smaller set of meaningful underlying dimensions. 

The key differences between factor analysis and discriminant analysis are

• Discriminant analysis defines dimensions which show maximum differences between groups, tending to ignore those dimensions of brands which show large variation across all consumers. For this reason, discriminant analysis is likely to generate fewer dimensions than factor analysis, because factor analysis will include those dimensions that account for a significant amount of the total variation even when such dimensions do not show variation across groups. This may or may not be desirable, depending on the objective at hand. 

• The maximum possible number of dimensions in factor analysis is the number of attributes on which ratings have been obtained. In discriminant analysis, the maximum possible number of dimensions is given by the lesser of p - 1 and m - 1, where p is the number of brands and m is the number of attributes on which these brands are evaluated. Thus, it is important that m and p be sufficiently large. As previously mentioned, too few brands leads to the possibility that important discriminating dimensions may be overlooked. In circumstances with a limited number of brands and/or attributes factor analysis is preferred. 

In summary 

• Discriminant analysis should be generally preferred over factor analysis when objective dimensions, as opposed to evaluative dimensions, are of interest. 
• Factor analysis and discriminant analysis may be used as complementary techniques to highlight those dimensions which differ substantially in the level of perceptual agreement among consumers.
• Factor analysis is generally preferred when attribute ratings are available on very few brands. 

Ritesh Mehta

14158

Group D - Operations
(September 15th Class)