RMESA, Information criteria ML-CFA PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 smeme posted on Sunday, June 19, 2016 - 7:18 am
With ML estimator how are RMSEA, and BIC computed for CFA with two-level, three-level and cross-classified models?

1. RMSEA with a grouping variable:

RMSEA = sqrt((chi-square/(n*d)) - (1/n))*sqrt(g)

a. what is n = nclusters or n = total level-1 observations?

b. What is g with a within- level or between-level grouping variable?

2. BIC = -2LL + np*log(n)

a. what is n = nclusters or n = total level-1 observations?

b. What about missing data?
 Tihomir Asparouhov posted on Monday, June 20, 2016 - 4:57 pm
RMSEA is not computed with cross-classified models in Mplus

1a. Find in the output "Number of observations" i.e. n = total level-1 observations

1b. G is a between-level grouping variable (always) and is just the number of groups.

2a. n = total level-1 observations, find in the output "Number of observations"

2b. If an observation has missing values on all dependent variables, it is not counted in n.
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