ABSTRACT
This chapter discusses the procedure that are generally used when, a survey of the literature is being conducted or if one is evaluating whether the factors found are also the simple structure factors for a new study. Confirmatory factor analyses can be used not only to test past findings in a new sample but also to factor several samples simultaneously to obtain the hypothesized structures. Separate factor analyses are particularly appropriate if prediction or causation can be assumed from one domain to another on the basis of external information. Confirmatory factor-analytic procedures are a recent innovation that provides more elegant solutions for the problem of relating factors when the complete data are available for one set of individuals. The factor pattern or other information would be used to set up the hypotheses and that would then be tested by confirmatory maximum likelihood factor analysis.
