Stepwise Latent Class Analysis in the Presence of Missing Values on the Class Indicators

While latent class (LC) modeling using bias-adjusted stepwise approaches has become widely popular, little is known on how these methods are affected by missing values. Using synthetic data sets, we illustrate under which conditions values introduce biases in the estimates of relationship between membership and auxiliary variables. We apply three-step LC analysis with both modal proportional assignments, as well recently proposed two-step method.Our results show that yields unbiased parameter long MAR assumption holds step-one model. When this does not hold because covariates omitted from model, each some bias, but bias much larger assignments. The amount deviation MAR, proportion values, separation classes.