B1211
Title: A two-step estimator for multilevel latent class analysis
Authors: Roberto Di Mari - Universita' di Catania, Dipartimento di Economia e Impresa (Italy) [presenting]
Abstract: The goal is to review the recent contribution to the two-step estimation of multilevel latent class models with covariates. The general design of the estimator is as follows. The measurement model for observed items is estimated in its first step, and in the second step, covariates are added to the model, keeping the measurement model parameters fixed. The model identification is discussed, and an Expectation Maximization algorithm is derived to implement the estimator efficiently. The resulting computer programs are openly available as an R package (multilevLCA) which can be downloaded from CRAN. By means of an extensive simulation study, it is shown that (i) this approach performs similarly to existing stepwise estimators for multilevel LCA but with much-reduced computing time, and (ii) it yields approximately unbiased parameter estimates with a negligible loss of efficiency compared to the simultaneous (one-step) estimator. The proposal is illustrated with a cross-national analysis of predictors of citizenship norms.
