Abstract
Unobserved heterogeneity in data can pose a challenge in statistical analysis. One alternative to addressing this is latent class analysis (LCA), a useful technique to find groups of individuals who share similar data patterns and, subsequently, to determine the relationship of these patterns with certain variables of interest. The objective of this paper is to present a general introduction to LCA and to describe the operations required to calculate a model of this type using the multilevLCA package in R. Before presenting the tutorial, (a) recent research that has used LCA to study phenomena in various disciplines is reviewed; (b) some differences between LCA and other classification techniques are established; and (c) the sequence of actions involved in executing the LCA is indicated, starting with aspects such as study design, required sample size and data configuration, including the evaluation of different solutions, until selecting and interpreting a final model. It is hoped that this work will serve as an instructional resource for implementing LCA and may contribute to promoting its use among Spanish-speaking and Latin American researchers.

This publication has 58 references indexed in Scilit: