gorkabr posted on Monday, January 17, 2005 - 7:04 am
I am working on the factor analysis of a diagnostic trauma measurement, which consists of 53 items - binary/dichotamous response. I am trying to locate any information or sources in reference to appropriate sample sizes for EFA and CFA using Robust weighted least squares. Any info will be GREATLY appreciated!
You should email email@example.com for paper 75 which discusses sample size for smaller models. You could do a simulation for a model your size to determine the sample size needed for your model. It depends on the item distribution.