Course Text: Pages 1-7
The basic idea underlying traditional Latent Class (LC) models is that responses to variables (called indicators) come from persons who belong to one of K distinct mutually exclusive and exhaustive populations called latent classes. The word ‘latent’ means ‘hidden’ which underscores the fact that we do not know which cases belong to which latent classes nor do we know how many classes there are — two goals of the latent class analysis.
In LC we assume that respondents in a given latent class are homogeneous with respect to certain model parameters that characterize their responses.