So What “Should” We Use? Evaluating the Impact of Five Racial Measures on Markers of Social Inequality

Abstract

In recent years, researchers have increasingly noted the malleability of racial boundaries across time, context, and life course. Although this research has advanced our knowledge of the maintenance and perceptions of racial groups, it has introduced a new question: If we are attempting to best capture the actual variation in racial inequality, how should we operationalize race? Using the 2006 wave of the Portraits of American Life Study, a national-level, in-home survey with extensive race measures and oversamples of Blacks, Hispanics, and Asians, the authors identify five ways that race can be and to varying degrees is operationalized: census, combined race/ethnic, pentagon, triracial, and skin tone measures. Using the Vuong non-nested model tests, the authors compare the effectiveness of these five measurements in predicting three measures of social inequality: household income, education, and self-rated health. The authors find that overall, Hollinger’s ethnoracial pentagon is best able to capture existing inequality. Thus, for scholars attempting to understand variation in contemporary racial inequality, this research suggests that scholars should use five monoracial categories: White, Black, Hispanic, Native American and Asian.