Another advantageous characteristic of MIMIC DIF models is that testing for
measurement and item invariance can be (using Wald tests instead of nested-model deviance
tests) a one-step process, rather than the multi-step process required by other latent-variable
invariance testing methods (i.e. invariance testing in multiple-group CFA). The comparative
simplicity in this regard is all the more apparent when categorical data are introduced, where
correct implementation of multi-group CFA invariance testing requires additional knowledge
(proper specification of item threshold constraints, etc.) that might make the process more
difficult (and consequently perhaps less appealing) to some researchers