procedure to test partial invariance in measurement model across groups

If the measurement model consists of multiple constructs

• examine each of the constructs in the model for invariance. A separate model is estimated for each construct. In each model (for each construct), the factor loadings associated with the construct are constrained to be equal across groups, while the loadings associated with the other constructs are not. The chi-square fit statistic of each model (for each construct) is used to calculate chi-square difference. If chi-square difference is significant, then at least one of the items within the construct is non-invariant.