Latent Class Analysis with partially cross-classified data.

Title: Latent Class Analysis with partially cross-classified data.
Identifier: AAI9020812
identifier: 9020812
Creator: Stewart, William N.
Contributor: Adviser: David Rindskopf
Date: 1990
Language: English
Publisher: City University of New York.
Subject: Education, Educational Psychology
Abstract: Latent Class Analysis (LCA) separates observations in a fully-observed cross-classified table into two or more latent classes bases upon the assumption of conditional independence of the observed variables within each latent class. Frequently, a substantial number of observations are missing categorization with respect to one or more of the observed variables. We consider an alternate parameterization of the usual maximum-likelihood method for LCA in which multiple partially-crossclassified tables, that have some observed variables either Missing Completely at Random (MCAR) or Missing at Random (MAR), can be used with a completely cross-classified table or by themselves to derive the latent classes. Two specific populations are modelled and the substantial gain in using partially cross-classified tables in LCA is demonstrated.
Type: dissertation
Source: PQT Legacy CUNY.xlsx
degree: Ph.D.