"Neighborhood matters": assessment of neighborhood social processes

Abstract

Neighborhoods are important contexts for understanding development and behavior, but cost and difficulty have challenged attempts to develop measures of neighborhood social processes at the neighborhood level. This article reports the development, reliability, and validity of Neighborhood Matters, a collection of instruments assessing three aspects of neighborhood social processes, namely, norms (five subscales), informal social control (six subscales and total scale), social connection (two subscales), as well as individual scales for assessing neighborhood change, neighborhood resources, and neighborhood problems. Six hundred six residents of Chicago, chosen at random from 30 neighborhoods (defined by US Census tracts), completed the measures. Neighborhoods were selected randomly from pools that balanced poverty and predominant (African-American vs. Latino Hispanic) ethnicity. Within each neighborhood 20 individuals were selected at random, balanced by age (18-24 vs. 30+) and gender. Scaling and item analysis permitted reduction of the number of items in each scale. All subscales had individual-level internal consistency in excess of .7. Generalizability theory analysis using random effects regression models found significant shared variance at the neighborhood level for three norms subscales, four informal social control subscales, both social connection subscales, and the neighborhood change, resources and problems scales. Validity analyses found significant associations between neighborhood-level scores on multiple Neighborhood Matters scales and neighborhood levels of violent, property, and drug-related crime. Discussion focuses on potential applications of the Neighborhood Matters scales in community research.