Can I use the unstandardised latent variable scores as they are provided in the IPMA for hypothesis testing (t-test)?

[ElNicolosi]

Hello everyone,

hope you can help me with this as I am stuck in my data analysis.

I know I can obtain unstandardized latent variable scores from the IPMA. I also read online lots of articles discussing Hypothesis testing through the latent factor scores. My questions are:

I need to carry out a t-test to test differences between two latent variables of two different groups. Can I use the Latent variable scores as they are provided by the IPMA for each observation or do I need to compute any sort of adjustment? Meaning, can I copy and paste those LVS in SPSS for all the observation and then carry out my T-test? Although I read several posts and articles on the matter I could not find any clear indication on whether I could simply copy and paste those scores or whether I need to adjust them for the number of items etc.

Also, providing that I am comparing two different groups form two different population, would be good to check first for measurement invariance?

Hope you can help me with the above cause I am really stuck and do not know how to proceed further.

Thank you!

[jmbecker]

First, yes you should check for measurement invariance.

Second, when you test for measurement invariance, in step 3 you get a permutation estimate of the mean differences which is similar to the t-test that you want to calculate. Hence, you might just use this (assuming that you have measurement invariance in step 2).

Third, generally, if you have measurement invariance in step 2 you could also copy the unstandardized LV scores from the IPMA on each group and use a t-test. However, you are then leaving the realm of nonparametric testing. One of the main reasons for why people use PLS. This brings you back to second.