0

I have a model with one latent variable (two observed variables A and B). I think variable A is contributing to the majority of the entire model's significance, and variable B might contribute less or none. I wonder whether there is a way to compare statistically the contribution of observed variables A and B to this model. I am using the Lavaan package in R so any packages in R will be more useful. Also I tried SEM models with only A or B included, but it seems that variable A itself doesn't make the model significant (P= 0.1), and if only with variable B then P=0.8. My hypothesis is that variable A and B both will contribute the model but A's contribution is bigger. Any suggestions will be appreciated. Thanks very much, Betty

Betty
  • 1
  • Your question would probably be better answered on Cross Validated. This isn't really a coding issue so much is it is a question about how to perform SEM. – Shawn Hemelstrand Oct 17 '22 at 22:19

0 Answers0