|
 |
EFA with egocentric network data |
 
|
Message/Author |
|
|
I'm a newbie and have a very basic question about the proper data format for EFA. I'm using data with variables that represent survey items describing different domains of social support between individuals (egos) and their egocentric social network members (alters). The data is formatted long -- each ego has from 1 to 30 alters. I'd like to perform EFA to help explain covariation among the social support items by mapping them to latent construct(s). Is long format appropriate for the EFA in Mplus? Appreciate your help and patience. |
|
|
Perhaps you want to do 2-level EFA. This is described in the UG. See ex 4.5 and 4.6. The data are on our website so you can see how they are arranged. |
|
Back to top |
|
|