Hi, I'm running EFA analyses in Mplus as well as R. Using exactly the same input correlation matrix, ML estimation, and GEOMIN rotation (GEOMINQ in R, allowing orthogonal solution), and either or not Kaiser normalization, I get a different EFA solution. Do you maybe have an idea what causes this? I assume its a difference in default settings but I cant seem to find what it is.
The Geomin rotation depends on a small epsilon value. This is from the Mplus manual
"The GEOMIN epsilon (Browne, 2001) default setting varies as a function of the number of factors. With two factors, it is .0001. With three factors, it is .001. With four or more factors, it is .01. The default can be overridden using the GEOMIN option."