The Limits of the Multivariant Growth Model in SEM

The MLM is exceedingly well suited in the ways that we've described primarily for a single outcome variable. You can indeed do a multi-variant growth model in SEM and at least in my eyes, the guy who cracked the code on this was Bud McCallum. But here's the problem. The current level of knowledge is your multivariate models stop there. If you want to look at how does time one math predict time to grit and time one grit predict time to math in the presence of those trajectories, the multi level model is not able to do that. That would be kind of interesting to explore.