This research examines patterns and processes of racial/ethnic intermixing in 2000, and its change from 1990-2000 in 49 mid-to-large sized continental metropolitan statistical areas (MSAs), at the Census-tract scale in two MSAs, Columbus and Milwaukee, and the underlying reasons. In doing so, it also measures the degree of intermixing observed between racial/ethnic groups and at nested scales of geography, as the Theil Entropy Index, the measure used for intermixing in this study, allows scalar and group-based decomposition. In contrary to the classical frameworks of Assimilation, Stratification or Resurgent Ethnicity, this study uses the newly published Market-Led Pluralism (M-LP) of Brown and Chung (2008), and a household and multi-city perspective to explain contemporary patterns in US urban settings.This work is accomplished in three phases. The first two phases focus at MSA and Census-tract scales, using Census-based secondary data analysis. The third phase uses primary data collected from selected households and communities in Columbus and Milwaukee. Racial/ethnic categories considered here include Caucasian, African-American, American-Indian, Asian, and Hispanic. For Phases I and II, the Entropy Index and Change in Entropy Index serve as the dependent variables in regression analyses, wherein independent variables include demographic, socio-economic, and built-environment characteristics. In the third phase, primary data is collected through neighborhood reconnaissance, household surveys, open-ended interviews, and focus groups in the two MSAs. This phase, besides fleshing out the interaction between housing market elements and households that may impact residential choices, tests and expands the M-LP from a household and a multi-city perspective.
Major findings suggest that all MSAs have become more intermixed in 2000. Factors relating positively with intermixing in 2000 include a larger share of recently arrived foreign-born population, a larger share of recently built housing, a higher share of movers (i.e., those who lived in a different home in 1995), a higher share of population with better educational achievements in 1990, and a higher share of those employed in managerial/professional jobs in 1990 and 2000. Concerning change in mixing, it seems that the places with low-to-medium mixing in 1990 encountered the biggest gains, suggesting a community norm hypothesis, and there were others that maintained a status quo of their 1990 intermixing levels, suggesting a community inertia hypothesis.
Considering the role of housing market elements through household perspectives, it seems that while there is substantial decline in discriminatory practices, it still occurs in subtle ways as realtors tend to impose their ideas on consumers by making assumptions about their choices based upon skin color, but more so when consumers are not self educated about the market norms and are easily swayed by realtors. Very often, people are steered into bad mortgages through aggressive lending practices. Concerning developers/builders, Market-Led Pluralism fairs well from household and community perspectives. Class is increasingly becoming more important in explaining segregation, though the role of culture is more prominent among African-Americans where even middle to higher SES homeowners have difficulties in buying homes in primarily Caucasian neighborhoods.