Correcting for Nonresponse in Latent Class Analysis
Rita R. Kolb, C. Mitchell Dayton · Multivariate Behavioral Research · 1996
Monte Carlo methods were used to evaluate an EM algorithm for correction for missing data in latent class analysis. Bias in parameter estimates was assessed under various assumptions concerning the mechanism for missingness including cases where missingness was not at random. Findings suggest practical limits for the utility of the EM algorithm in terms of sample size and nonresponse rate.