Beyond parsimony: principal components analysis as a tool for data exploration
Pages 17

Parsimonious description of large and complex sets of data is the main reason for using principal components analysis and I have employed this argument to justify use of the technique above. A good indication that this has become a standard argument for using a factor analysis is that in the standard software package for social science statistical analyses (SPSS), the family of factor analytic techniques is reached by clicking on the ‘Data Reduction’ button. However, in the classic social science text on factor analysis by Rummel (1970), a much wider range of uses is discussed. Rummel identifies ten different ‘design goals’ for these techniques (p. 182), one of which is to ‘explore’. This is the approach I discuss here, but, alas, exploratory analysis is not followed up at all in his otherwise comprehensive text. I attempt an initial rectification of this oversight here.