A particularly interesting and relatively well-executed study conducted by Guadagnoli & Velicer (1988) investigated the effects of sample size under a series of different conditions. The authors used Monte Carlo methods to examine the effects of four criteria—1) the number of factors (3, 6, 9, 18), 2) the number of items (36, 72, 108, and 144), 3) the average item loadings (.40, .60, or .80), and 4) the number of subjects (Ns of 50, 100, 150, 200, 300, 500, and 1000)—on the stability of factor patterns in EFA. In this data, each item loaded on only one factor, all nonzero item loadings were of equal size, and each factor contained an equal number of variables. Their study represents one of the few studies ...