A Study Of A Measure Of Sampling Adequacy For Factor-Analytic Correlation Matrices

Abstract

Kaiser's Measure of Sampling Adequacy (MSA) for factor-analytic correlation matrices is studied for several levels each of p, the number of variables, q, the number of factors, and rfl, the root-mean-square off-diagonal correlation. The major influence for MSA is p, in agreement with theory; the joint main effect influences of p, q, and rfl to the total SSs remains greater than 84% under various choices of the levels of p, q, and rfl.