Dear Dr. Becker,
I would very much want to know how values in the column entitled Mean - Original Difference ( GROUP_1 - GROUP_2 ) from the MICOM procedure (step 3) are calculated.
As I understand from the paper Testing measurement invariance of composites using partial least squares for a given construct (let’s say C1) we have to subtract the average of the construct scores of the observations in the second group from the average of the construct scores of the observations in the first group. In my case, the reported value by in the MICOM results is 0.161.
As I wanted to double-check this results, I ran the IPM analysis for both groups in order to get the unstandardized scores for the latent variables. For the same construct (C1), the average of the LV scores in group one is 5.400 and in group 2 the average reaches 5.202, resulting in a difference of 0.198 which is not equal to the one reported in Step 3 of the MICOM procedure (0.161).
Please let me know what I am doing wrong or how I could get to the same mean difference as in the permutation output.
Frequently asked questions about PLS path modeling.
1 post • Page 1 of 1