Hi everyone,
I have conducted a SEM analysis in AMOS then replicated it in RStudio. While I obtained identical estimates and model fit statistics, the p-values differ between software programs (see indirect effects in the attached output). Could you please explain me what kind of p-values are reported and which one of these should I report in my paper?
Best,
Robin
It most likely depends on what approximation is used to compute the p-value I know some of the guys that do R, so I would put my money there. Best, D. Booth
Robin Wollast, if one of the software uses bootstrapped standard error and the other not, this may also lead to a difference between p-values. Please check if this is the case.
Dear Robin Wollast ,
in an extension of David Eugene Booth s and Mehmet Mehmetoglu s answers, you should definitely consult the vignettes and help pages of your SEM-function in R, on what exactly is done. In Rstudio you can find these via the "help" or via the "packages" tab in the lower right pane.
Best,
Sven
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling