Multi-group path analysis with two di... PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Ahyoung Song posted on Monday, February 24, 2014 - 7:52 am

I have two different data sets (men and women) which were collected different settings. Each data has its own sampling weight. I attempted to merge two data sets to test the moderating effect of gender, which could be created by merging data. But due to complexity of sampling weights, it does not seem to be feasible to merge data sets. A statistician suggested me to try multi-group analysis to test the moderating effect of gender.

Here are my questions.
1) Does multi-group approach allow to use two different data sets?
2) Do you think it is possible to apply sampling weights for each data separately when attempting multi-group analysis?

Thank you so much,
 Linda K. Muthen posted on Tuesday, February 25, 2014 - 1:19 pm
You can combine the data sets and create a grouping variable to do multiple group analysis. The weights will be used correctly for each group.
 Amanda Pollitt posted on Wednesday, February 03, 2016 - 7:48 am
As a followup, do you suggest creating a single weight variable that is a combination of the two weights in each dataset? Or can Mplus handle more than one sampling weight?
 Linda K. Muthen posted on Wednesday, February 03, 2016 - 2:18 pm
As I understand it, you have two data sets and each has a sampling weight. If you put them together with males first and females second, for example, the column with the weight variable will contain the weight for males first and females after that. Use that single variable with the WEIGHT option.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message