Spatially Based Rules for Reducing Multiple–Race into Single–Race Data

Abstract

There is a discord between the categorization of mixed–race data in spatial studies, which has become more complex as the mixed–race population increases. We offer an efficient, spatially based method for assigning mixed–race respondents into single–race categories. The present study examined diversity within 25 Metropolitan Statistical Areas in the United States to develop this racial bridging method. We identify prescriptions for each two–race category based on average diversity experiences and similarity scores derived from census tract data. The results show the following category assignments: (1) Black–Asians to Black, (2) White–others to White, (3) Asian–others to Asian, (4) White–Blacks to other, (5) White–Asians to White (if Asian >3.0 percent), (6) White–Asians to Asian (if Asian <3.0 percent), (7) Black–Asians to other (if Black >8.5 percent), and (8) Black–Asians to Black (if Black <8.5 percent). We argue that the proposed method is appropriate for all race–based studies using spatially relevant theoretical constructs such as segregation and gentrification.