Quiénes Somos: Ethnic Self-Identification for Second-Generation Latin American Immigrants and Its Implications for the U.S. Census

Abstract

The U.S. Census Bureau has recently proposed combining race and ethnicity into one question, potentially allowing people who identify as Hispanic/Latine to select this option as their sole ethno-racial identity. This thesis seeks to contribute to explore the possible ramifications of this choice by exploring the factors that affect how second-generation Latin American immigrants self-identify. The literature finds links between self-identity and location, sex, length of residence, citizenship status, national origin, preferred language, parental ethnic identity, family cohesion, parental arrival in the United States, mother occupation, father occupation, parental educational attainment, parental wealth, demographics of school attended, close friendships with other individuals whose parents are from abroad, and experience with discrimination. I used subsetted Latin American data from the Children of Immigrants Longitudinal Survey (CILS) to construct multinomial logistic regression models that test relationships between these factors and self-identity and changes in self-identity. The results confirm that all of these factors are influential in affecting self-identity, albeit their impact varies across model specifications. Overall, I find that there are a multitude of factors that shape self-identity and changes in self-identity. Identity is in constant evolution and U.S. Census attempts to demarcate U.S. social constructions of race and ethnicity fall flat in capturing the wide ranges of ethnic and racial identities that Latin American descended individuals identify with.