The Enumeration Accuracy of Cross-Validation Indices in Latent Class Analysis

• In Event: Marriott Roundtable Session 29
  In Roundtable Session: Advances in Clustering and Classification

Abstract

A crucial issue when estimating mixture models is selecting the model with the correct number of classes underlying the data, commonly referred to class enumeration. Various statistical indices are available that may aid researchers during this process. Cross-validation methods have been seldom used for class enumeration. The purpose of this simulation study is to compare the performance of traditionally used single sample enumeration indices with the performance of cross-validation indices when selecting the correct latent class model. Various conditions will be manipulated, including sample size, class separation, mixing proportions, and number of latent classes. It is hoped that this paper will inform applied researchers about the utility of cross-validation indices for class enumeration.

Authors

• Tiffany Ann Whittaker, The University of Texas at Austin

• Joseph E Miller, The University of Texas