Social distance and social change: how neighborhoods change over time

Two key theoretical themes guided my exploration of neighborhood change. First, I utilized the classic sociological notion of social distance in testing its determinants and viewing its effect on neighborhood change over time. I measured social distance in various manners: 1) racial/ethnic differences, 2) a composite of several characteristics (including racial/ethnic, socio-economic, and demographic), or 3) the consolidated inequality created by difference along both racial/ethnic and socio-economic differences simultaneously. Second, I built an explicit micro-level theory of household residential mobility decisions to explain the generation of the structural characteristics that theories posit cause neighborhood crime. I found that social distance had important implications for neighborhoods. Using a multi-level, longitudinal sample of the American Housing Survey (AHS) I found that individual-level social distance along multiple characteristics helps explain neighborhood satisfaction: this suggests the importance of focusing on the fit of the household with the neighborhood. Dynamic analyses using this same sample showed that racial/ethnic heterogeneity explains crime rates four years later. Fixed effects analyses using a sample of census tracts in eleven cities found that changing ethnic heterogeneity over the decade is positively related to changing crime rates. These same fixed effects analyses showed that increasing inequality between African-Americans and whites is positively related to the change in various official crime rates. My theoretical model helped explain the change in neighborhood structural characteristics. Using the AHS sample, I found that perceived crime in a block increases general residential mobility. This theoretical model also predicted and found that the presence of more homeowners on a block reduces perceived crime four years later in dynamic models. While residential instability had no effect on crime four years later in cross-lagged models, more vacant units in the block are positively related to perceived crime four years later. This suggests a possible manner in which residential mobility may affect neighborhood crime rates. I also found using the AHS sample that higher levels of block perceived crime creates racial/ethnic residential transformation by increasing the likelihood that African-Americans and Latinos will move into the block, and reducing the likelihood that whites will move in