Hello professor, I have a question for running SEM. I've run an EFA on the entire dataset (N=347) and used the obtained simple structure to run the CFA for the entire model (7 constructs). The RMSEA and SRMR are good but both the CFI and TLI is just shy of 0.9. I tried to look up this issue but found no simple answer as to what action should be taken before running the analysis for SEM.
Thank you for your video. I have one question, please. When we do Moderation and Mediation analysis using process macro, do you think we still need to run CFA? Thank you in advance
You would only use moderation and mediation lavaan style if you wanted to or needed to because you had latent variables - if you can do everything in regression, that's easier. I did this example for my class to help them see the parallel's between path analyses and regular regression.
Struggling through Multivariate Stats and this helps a ton. Thank you. Thank you. Thank you.
Thanks for the kind words!
This is a great video. It helps a lot to fully understand CFA. Thank you Dr. Buchanan
Thanks for the kind words!
Hello professor, I have a question for running SEM. I've run an EFA on the entire dataset (N=347) and used the obtained simple structure to run the CFA for the entire model (7 constructs). The RMSEA and SRMR are good but both the CFI and TLI is just shy of 0.9. I tried to look up this issue but found no simple answer as to what action should be taken before running the analysis for SEM.
I would suggest that running an EFA and CFA on the same data is just doing the same analysis twice and should not be done. I would pick one.
@@StatisticsofDOOM Ok. In case the CFA doesn't produce good fit, what is the best approach to take?
@@ALI_B usually you check the modification indices to see where potentially bad fit can be found.
You are Great Dr. Buchanan! thanks a lot!
Glad to be of help!
you are the best miss, thanks a lot!
Thank you!
Super video
thanks so much for amazing videos! I learned so much from them!!!
Thank you for your video.
I have one question, please.
When we do Moderation and Mediation analysis using process macro, do you think we still need to run CFA?
Thank you in advance
You would only use moderation and mediation lavaan style if you wanted to or needed to because you had latent variables - if you can do everything in regression, that's easier. I did this example for my class to help them see the parallel's between path analyses and regular regression.
Remember: All caps when you spell the girl name :)