Calculating within-cluster variance PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Eva Muenkhoff posted on Thursday, December 19, 2013 - 2:35 am

In addition to the cluster mean, I would like to calculate the (within-)cluster variance or standard deviation for some of my level 1-IVs. My aim is to use the (within-)cluster variance as IV on level 2, thereby representing the diversity/heterogeneity of the variable on the group level (similar to Ahearne/Lam/Hayati/Kraus 2013, p. 44). Is this possible with Mplus?

Thank you!
 Bengt O. Muthen posted on Thursday, December 19, 2013 - 1:24 pm
Not automatically in a single step, but you can do TYPE=TWOLEVEL BASIC to get the variance and then enter it as a level-2 IV.
 Michael Covell posted on Tuesday, August 27, 2019 - 4:07 pm
Is it possible to compute a separate variance for each cluster? This would help in computing rwg.
 Bengt O. Muthen posted on Wednesday, August 28, 2019 - 3:06 pm
UG ex 9.28 shows how a random variance can be handled in Mplus.

You can also get the estimated values in all clusters for any random effect by using the FSCORES option (see the UG index).
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