0

I'm trying to do exploratory factor analysis (EFA) and it seems that all functions (in MATLAB, python or R) standardise the data matrix first. I was wondering if there is a way to perform factor analysis over the unstandardised matrix and thus obtain unstandardised scores.

iditbela
  • 117
  • 7
  • You mean that the analysis is performed on the correlation and not the covariance matrix, right? – chl Oct 03 '20 at 18:06
  • Well, partially, I think. Since I want it to not subtract the mean as well. Does it make sense? – iditbela Oct 03 '20 at 18:39
  • 1
    Factor analysis on raw "as is" data is possible but is unusual to do. It is called FA based on [MSCP](https://stats.stackexchange.com/a/22520/3277) matrix. One reason to do it is ratio scale variables and you want to depart from the native 0 point instead of the mean point. But note, magnitude (SS of deviations) from native 0 is not "variability". Check also https://stats.stackexchange.com/q/22329/3277. – ttnphns Oct 03 '20 at 18:50
  • 1
    "and thus obtain unstandardised scores" This is a very very different topic from undertaking the whole FA on raw data. You can do traditional FA and then de-standardize factor scores. – ttnphns Oct 03 '20 at 19:03
  • @ttnphns, thanks for pointing out the difference between what I asked and what I eventually want to achieve. It would be great if you can help me with the latter. As far as I understand, it is not possible to obtain the de-standardized factors after performing traditional FA. – iditbela Oct 04 '20 at 08:40

0 Answers0