Integrative Data Analysis in SEM PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Johannes Bauer posted on Wednesday, April 09, 2014 - 11:52 pm
I want to do a SEM by simultaneously analyzing data from two studies. The goal is to get combined structural parameter estimates (like in a mini meta-analysis of individual participant data). Both studies have an identical instrumentation, but differ in the sampling (both have complex survey samples; study two has more clusters).

In the literature on integrative data analysis/meta-analysis I found two methods for this purpose:
* Cooper and Patall (2009) mention that study can be used as a stratification variable.
* Curran & Hussong (2009) suggest including study membership as a categorical predictor as well as its interactions with other predictors in the model. This seems impractical in SEM because - to my knowledge - interactions between manifest and latent variables can only be modeled in multi-group SEM which would not provide pooled estimates over both groups.

So my questions are:
(1) Is it sufficient/defensible to use the stratification approach for the discussed purpose?
(2) Is there a better way to get the pooled estimates?

Many thanks
Johannes
 Linda K. Muthen posted on Thursday, April 10, 2014 - 10:19 am
Why don't you want to do a multiple group analysis? Also, the XWITH option can be used for an interaction between an observed and latent variable.
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: