A Latent Variable Modeling Approach To Understanding Attitudinal And Behavioral Ethnic Identity In Hispanic Consumer Behavior: A Structured Abstract

Abstract

Ethnic identity has been shown to play an important role in Hispanic consumer behavior research. The measurement and statistical treatment of ethnic identity measures, however, have been mixed across studies. First, the operationalization of ethnic identity has varied with some studies defining the construct by behaviors (e.g., language use) and others using theoretically-based measures. Second, different research questions and statistical treatments of ethnic identity as a predictor have been applied. Ethnic identity measures have, for instance, been used to create groups of self-identified Hispanics and non-Hispanics so that comparisons in consumer behaviors could be done. Studies later relied on within-group comparisons of behaviors between Hispanics reporting relatively high ethnic identity levels to those reporting relatively low levels. In these examples, ethnic identity was essentially treated as a categorical variable. The present study uses structural equation modeling to define ethnic identity by two latent, continuous dimensions and statistically treating them as such. Comparisons to past approaches to the measurement and statistical analysis of ethnic identity suggest a latent variable modeling approach may aid in understanding the relationship between ethnic identity and Hispanic consumer behavior. The goals of this study are 1) to evaluate a single measure of ethnic identity as a predictor of HCB and 2) to compare results from the former to predictions of HCB using distinct but related aspects of ethnic identity, namely attitudinal and behavioral.