Measures of Racial Identity and Their Implications for Political Behavior

• In Event: Multidimensionality in Measurement and the Study of Racial and Ethnic Boundaries

Abstract

The U.S. population is more diverse and multiracial than it has ever been. Standard measures of ethnoracial identity use crude categories that obscure the tremendous heterogeneity within and hybridity between categories. These well-documented limitations (Guluma and Saperstein 2022, Howell and Emerson 2017, Shiao 2019) result in an incomplete picture of the political attitudes and behavior associated with members identified by such categorization. In this paper, we articulate theoretical grounds for adopting a multidimensional approach to measurement and present more complete empirical study of the ethnoracial identity-to-politics link. Specifically, we propose a move away from singular and superordinate racial categories (Roth 2016) and compare multiple measures: self-identification on a survey, race identified as most central, ascribed race by others, self-identified skin color, identified race of mother and father, and a relativelynew Point Allocation continuous measure of ethnoracial self-identification (Lee 2009). We then present our comparative analysis of these measures using data from the 2020 Collaborative Multiracial Post-Election Survey. Specifically, the paper explores the following animating questions: 1) How does the standard race question map on to other racial identity measures? 2) How does the size of an ethnoracial population shift depending on the racial dimension used? 3) Are the observed patterns on racial political attitudes consistent when we examine them across different ethnoracial dimensions? 4) What is the best-fitting operationalization of ethnoracial categorization across a series of racial political attitudes?

Authors

• Alexander Agadjanian, University of California, Berkeley

• Lauren D. Davenport, Stanford University

• Taeku Lee, Harvard University