Intercept vector in twolevel factor a... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Student 09 posted on Tuesday, December 30, 2008 - 12:19 am
Hello

I have a question regarding the intercepts of the observed indicators on the within-level in a twolevel factor analysis. Is it correct that the intercepts must be considered to equal zero - because on the within-level, the data are pooled and group mean centered?

best
Jan
 Linda K. Muthen posted on Tuesday, December 30, 2008 - 8:07 am
Yes.
 Student 09 posted on Thursday, January 08, 2009 - 3:37 am
Hello, here is a follow-up question.

I wonder whether it is possible to examine if the intercepts (tau’s) of the observed indicators of a within-level factor vary across between-level units?

I am struggling with this question, because if in twolevel factor analyses the vector of intercepts of the within-level factors is assumed to be zero - does this imply that one can't test for varying intercepts?

My question is motivated by the idea that varying indicator intercepts between contexts could be seen to indicate contextual-level differential item functioning. Would you agree with this view?
 Linda K. Muthen posted on Thursday, January 08, 2009 - 8:03 am
Random intercepts appear on the between level. If the variances of the random intercepts are significant, they vary across clusters.
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: