Abstract
Morin, A. J. S., Meyer, J. P., Creusier, J., & Biétry, F. (2016). Multiple-group analysis of similarity in latent profile solutions. Organizational Research Methods, 19(2), 231-254. doi:10.1177/1094428115621148
Morin, Meyer, Creusier, and Biétry (2016) proposed a methodological framework to systematically investigate the extent to which latent profile solutions generalize to different groups or samples of participants according to a sequence of profile similarity tests: (a) configural similarity (the same number of profiles based on the same indicators), (b) structural similarity (the same within-profile means), (c) dispersion similarity (the same within-profile variances), (d) distributional similarity (the same marginal probabilities, or relative size of the profiles). Following these initial steps, and starting from the most “similar” model, they further proposed to test the similarity of associations between profiles and covariates through tests of (e) predictive similarity (relations between predictors and profile membership), (f) explanatory similarity (relations between profile membership and outcomes). Despite the fact that their contribution was limited to the multiple group context, they also proposed, as part of their online supplements, a similar methodology and relevant syntax for longitudinal tests of profile similarity using latent transition analyses (LTA). However, the approach described for LTA tests of distributional similarity described in these online supplements is suboptimal, and has been recently optimized in a webnote prepared by Morin and Litalien (2017). This webnote should be consulted by anyone thinking to transpose Morin et al.’s (2016) approach to the LTA context.
Morin, A. J. S., & Litalien, D. (2017). Webnote: Longitudinal tests of profile similarity and latent transition analyses. Montreal, Canada: Substantive-Methodological Synergy Research Laboratory. Available at http://smslabstats.weebly.com/uploads/1/0/0/6/100647486/la_distributional_similarity_v02.pdf