In statistics, latent variable models relate an unobservable variable or construct to a set of directly measured indicator variables. Through the model parameters, the unobserved latent variables influence the directly measured indicator variables in a causal way. OIT’s Research Computing Support (RCS) team can assist you with latent variable models, including confirmatory factor analysis, item response theory, latent profile analysis, latent class analysis, and structural equation modeling. These analyses are generally performed in Amos, Mplus, or R.

To schedule a one-on-one consultation with an OIT statistician for help with latent variable models, submit a Research request.