Methods for Factor Analysis

STATISTICA includes several procedures for analyzing factors. Although many of the available statistics overlap, each is best suited for particular applications.

Confirmatory factor analysis. STATISTICA also includes options for general Structural Equation Modeling (SEPATH). The procedures available in that module allow you to test specific hypotheses about the factor structure for a set of variables, in one or several samples (e.g., you can compare factor structures across samples). The Examples section of SEPATH discusses several examples of such analyses.

Correspondence analysis. Correspondence Analysis is a descriptive/exploratory technique designed to analyze two-way and multi-way tables containing some measure of correspondence between the rows and columns. The results provide information which is similar in nature to those produced by factor analysis techniques, and they allow one to explore the structure of categorical variables included in the table. For more information regarding these methods, refer to the Introductory Overview for Correspondence Analysis.

Factor Analysis. Factor Analysis is an exploratory technique designed 1) to reduce the number of variables and 2) to detect structure in the relationships between variables, that is to classify variables. Therefore, factor analysis is applied as a data reduction or (exploratory) structure detection method (the term factor analysis was first introduced by Thurstone, 1931, although similar techniques were used by Spearman as early as 1904 in his classic research on the nature of intelligence). For more information regarding these methods, refer to the Introductory Overviews for Factor Analysis.

General Partial Least Squares Models (GPLS). These methods are an implementation of partial least squares (GPLS) techniques. GPLS allow you to extract factors (components) from a data set that includes one or more predictor variables, and one or more dependent (response) variables. GPLS is particularly suited for problems involving very many predictor variables (and possibly dependent variables), but relatively few cases.