0

I am using R to do EFA, I tried all the methods to make RMSEA of EFA low, but still around 0.14, and TLI still around 0.6
I am not sure is that possible to make mydata RMSEA<.05 and TLI>.95
I am asking experts help to analysis mydata, thanks!

Chris Kelly
  • 99
  • 7
  • What are you trying to accomplish by this analysis? – T.E.G. Aug 06 '17 at 04:44
  • I want to use EFA to get the good factors and then delete some bad items. – Chris Kelly Aug 06 '17 at 06:15
  • Could you provide more information about the variables/data and what have you tried so far? As far as I see from the linked data file, all your variables are binary. Why do you think that factor analysis is an appropriate method for the analysis of such data? – T.E.G. Aug 06 '17 at 06:25
  • Factor analysis is the good way to determine nature and number of latent variables.you have 60 variables and 1000+ obs, it's enough for EFA.It's better if you set 4 factors. – Chris Kelly Aug 06 '17 at 06:38
  • It's problematic to use factor analysis on binary variables. Please see the related questions for various approaches [here](https://stats.stackexchange.com/questions/16331/doing-principal-component-analysis-or-factor-analysis-on-binary-data-using-spss) and [here](https://stats.stackexchange.com/questions/215404/is-there-factor-analysis-or-pca-for-ordinal-or-binary-data). – T.E.G. Aug 06 '17 at 06:40
  • Except FA, any other method to understand the factor structure? – Chris Kelly Aug 06 '17 at 07:14

0 Answers0