Mapping urban revitalization: using GIS spatial analysis to evaluate a new housing policy.

This longitudinal, multimethod study uses geographical information system (GIS) software to evaluate the community-wide impact of a neighborhood revitalization project. Unsystematic visual examination and analysis of GIS maps are offered as a complementary tool to quantitative analysis and one that is much more compelling, meaningful, and effective in presentation to community and nonscientific professional audiences. The centerpiece of the intervention was the development of a new, middle-class housing subdivision in an area that was declining physically and economically. This represents three major urban/housing policy directions: (1) the emphasis on home ownership for working-class families, (2) the deconcentration of poverty through development of mixed-income neighborhoods, and (3) the clean up and redevelopment of contaminated, former industrial brownfields. Resident survey responses, objective environmental assessment observations, and building permit data were collected, geocoded at the address level, and aggregated to the block level on 60 street blocks in the older neighborhoods surrounding the new housing in two waves: during site clearing and housing construction (Time 1: 1993-95) and three years post-completion (Time 2: 1998-99). Variables mapped include (a) Time 1-2 change in self-reported home repairs and improvements, (b) change in the assessed physical condition of yards and exteriors of 925 individual residential properties, (c) change in residents' home pride, and (d) a city archive of building permits at Time 2. Physical conditions improved overall in the neighborhood, but spatial analysis of the maps suggest that the spillover effects, if any, of the new housing were geographically limited and included unintended negative psychological consequences. Results argue for greater use of GIS and the street block level in community research and of psychological and behavioral variables in planning research and decisions.