Extracting tech3

[Dani Schulze]

Dear Michael and Joshua,

unfortunately I cant parse tech3 output from my files.

readModels() yields either an empty list or an error (<simpleError in extractTech3(outfiletext, fileInfo, curfile): Object 'fileInfo' not found>), depending on the model.

The covariance matrices are present in Mplus, of course.

My setup:

• Tested with Mplus 7 and 8.4
• Tested simple CFA models (-> empty list) as well as a sort of experimental moderated non-linear factor models (-> error). Those include some model constraints. If you need input on the latter, see Bauer (2017) or I could attach the output.

Things I looked up:

• https://groups.google.com/g/mplusautomation/c/K49IPkhV7FE/m/Ksm1AYB8AjcJ, I couldnt get the suggestion with SAVEDATA to work.
• directly messing around with extractTech3() didnt help either.

I would be gratefull if you have a quick idea what's going wrong (or what I am doing wrong). Otherwise I will go down the path of writing something up on my own.

Thanks!

References:

Bauer, D. J. (2017). A more general model for testing measurement invariance and differential item functioning. Psychological methods, 22(3), 507.