Interpretation of MGA and MICOM results
Posted: Fri Sep 08, 2023 1:54 pm
Dear All,
I have observed heterogeneity in my data. So I separated my data into two data groups based on the categorical moderator. I ran MICOM and I have achieved configural and compositional invariance. However, I do not have equality of composite means and variances. Thus my measurement model shows partial invariance and hence I can proceed to MGA. After this, I checked for MGA and none of the paths show any significant difference across the two groups, establishing structural invariance of my model.
I am confused how to interpret this case. What does it mean to have partial invariance in the measurement model but complete invariance in the structural model? How to proceed further? Any guidance would be much appreciated. Thanks.
Reference:
Henseler, Jörg, Christian M. Ringle, and Marko Sarstedt. "Testing measurement invariance of composites using partial least squares." International marketing review 33.3 (2016): 405-431.
I have observed heterogeneity in my data. So I separated my data into two data groups based on the categorical moderator. I ran MICOM and I have achieved configural and compositional invariance. However, I do not have equality of composite means and variances. Thus my measurement model shows partial invariance and hence I can proceed to MGA. After this, I checked for MGA and none of the paths show any significant difference across the two groups, establishing structural invariance of my model.
I am confused how to interpret this case. What does it mean to have partial invariance in the measurement model but complete invariance in the structural model? How to proceed further? Any guidance would be much appreciated. Thanks.
Reference:
Henseler, Jörg, Christian M. Ringle, and Marko Sarstedt. "Testing measurement invariance of composites using partial least squares." International marketing review 33.3 (2016): 405-431.