![[ Visit Client Website ]](images/banner.gif)
Background: Maps of STD rates produced by the San Francisco Department of Public Health have used neighborhood definitions that were based on an analysis of socioeconomic data from the US Census. This has allowed us to account for similarities and differences in neighborhood composition when comparing STD rates across the City. This analysis was done more than thirty years ago, however, and many neighborhoods have changed substantially since then. Our objective was to recreate the original analysis to examine how neighborhoods have changed, to create a more current set of neighborhood definitions, and to see if those updated definitions better explained differences in STD rates.
Methods: Census tract data from the American Community Survey Five-Year Estimate data was used to define neighborhoods, including 65 variables relating to socio-economic status, racial/ethnic background, employment, housing, and families. A Principal Component Analysis was run to consolidate the variables into independent, hypothetical factors. These factors were used to refine the existing neighborhood definitions. ANOVA methods were used to examine whether these new definitions better explained geographical STD trends.
Results: Differences in variables available prevented us from testing the fit of the original model from the 1980 census data. Eight factors emerged from the Principal Component Analysis of the ACS data; together these explained 76.7% of the variation between the census tracts. Thirteen (13) neighborhood definitions were added or altered, bringing the new neighborhood count to 45. Results from ANOVA indicated that these neighborhood definitions explained geographical trends better than the existing map.
Conclusions: Our empirical examination of socioeconomic data on census tracts confirmed that there have been major demographic shifts in the city neighborhoods over the past thirty years. Our new map will present a better picture of geographic trends in STDs.