Marike Deutz posted on Wednesday, November 25, 2015 - 6:14 am
Dear Linda and Bengt Muthén,
I want to save factor scores of bifactor models in multiple waves across time, and I have been told to save them in tandem in order to be able to use them as input for GMM.
I use binary indicators and WSLMV estimation using Delta parameterization. It makes more sense to me to specify same-factor correlations across time (stability) using the ON command, but only WITH is possible with WSLMV+Delta. Is that correct and does it matter for the factor scores I intend to save if I use WITH instead of ON?