Equivalency and reliability of vectors of g-loadings across different methods of estimation and sample sizes

We examine the variability in g-loadings with particular focus on the differential stability (i.e., rank order inconsistencies) of the vector of loadings due to sample size and method of estimation. Our results, based on 1600 factor-analyses, show that different methods of estimation require different sample sizes to achieve the same level of stability in loadings. However, all methods examined yielded reliable vectors when sample sizes of N = 400 were used. Some methods yielded reliable vectors with sample sizes of N = 200. Such results provide guidelines for choosing an appropriate sample size when g-loadings are of interest.