Meta-analyzing bootstrapped indirect ... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 K. N. posted on Friday, February 15, 2019 - 1:29 pm
I’d like to calculate the boostrapped indirect effect of X -> M1 -> M2 -> Y across a number of studies, and then meta-analyze these indirect effects. Is it possible to do this in MPlus? If so, I'd appreciate any advice/resources on how to best do this. Thank you!
 Bengt O. Muthen posted on Friday, February 15, 2019 - 5:29 pm
This is possible. For meta analysis in Mplus, see our Short Course Topic 9 (note: the 6/1/11 version of T9), slides 150 and on.
 K. N. posted on Thursday, March 21, 2019 - 9:50 am
Thank you! I've reviewed the resource and have another question. Would you recommend that I bootstrap the meta-analysis, too, or is that not necessary/possible?
 Bengt O. Muthen posted on Friday, March 22, 2019 - 9:50 am
Bootstrapping might be good for the indirect effects to get non-symmetric CIs. That can be done with ML in the single-level approach (two-level bootstrapping is not available). Or you can use the two-level approach with Bayes which also gives non-symmetric CIS.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: