Abstract
FACTREL relates factors from two studies with the same variables using Kaiser, Hunka, and Bianchini's (1971) method. Both test and factor vectors for the two sets are placed in a common space, then one is rotated to maximize the sum of the row cross-products of loadings between the sets, with the resulting cosines between factor axes serving as the primary measures. Cosines between corresponding tests are also computed, as is their average. This method of relating factors is potentially useful in meta-analytic studies when only the factor loadings (and in oblique cases, factor correlations) are available in the published literature. Written in FORTRAN-77, FACTREL is a general, interactive program that handles both orthogonal and oblique cases.
Get full access to this article
View all access options for this article.