COMPARATIVE STUDY OF PRINCIPAL COMPONENTS AND FACTOR ANALYTIC TECHNIQUES

Principal Components and factor analytic techniques take large number of variables and reduce them to much smaller number of coherent subsets such that variables within a subset are related to one another but independent to those in other subsets. These methods summarize the pattern of correlation between observed variables. In this paper, principal components and factor analytic techniques are compared using data from Nigerian Consumption Pattern 2009/2010. The results revealed that factor analytic techniques preserve correlation more than principal components, while on the other hand, principal components preserve variance more than factor analytic techniques. We therefore conclude that factor analysis should be used when interest is placed on making statements about the factors that are responsible for a set of observed responses, and principal component analysis should be used when interest is based on performing data reduction.