Jenna Ward posted on Tuesday, April 12, 2016 - 8:17 pm
I'm new to CFA and am trying to complete an analysis for my dissertation.
I have a model with 8 observed variables and two factors. 6 of the variables load onto one factor, with 2 cross loading on the other factor. So, there are 6 indicators for the first factor and 4 for the second, with 2 of those indicators loading on both factors. The factors are correlated.
I am trying to interpret the R-squared for the cross loading indicators. Why doesn't the sum of the squared factor loadings add up to the R squared for a cross loading indicator?
I have a CFA model in MPlus where I want the same indicator to load on more than one factor (income) as it seems both theoretically relevant and loaded in an earlier EFA model. I am getting good goodness of fit statistics and the correlations between factors is quite low but I wanted to check whether I need to let MPlus know that the variable is loading to more than one factor. I thought there may be something similar to adding the correlated variance between two indicators within one factor. Is there any additional syntax I need to add to my model?