Xingyou Zhang, Katherine KauferChristoffel, Maryann Mason and Lin Liu. (2006). Identification of contrastive and comparable school neighborhoods for childhood obesity and physical activity research. International Journal of Health Geographic, 5, 14-23.

This paper studied the effects of neighborhood environmental profiles such as land use, traffic, crime, and census tract to childhood obesity.

The use of a geographic information system (GIS) which incorporates graphical features with tabular data in order to assess real-world problems, and cluster analysis would enable us to define neighborhood characteristics in which they can be used to select neighborhoods for community-based health study.

The study was aimed to answer the following questions:

1. How to choose representative communities and neighborhood?

2. What are important neighborhood characteristics to be considered?

3. What are the effects of school neighborhood social and build environments on child obesity?

To answer to the above questions, geoprocessing tools within a GIS environment has been used to integrate data such as land use, traffic, crime, and census to generate objective social and physical environment measures for school districts. Using this method Chicago public elementary school districts of 412 schools was divided to eight school neighborhoods. The effects of characteristics of these eight cluster for childhood obesity has been studied.

Since the method discussed on this paper allow objective characterization of neighborhoods, it rely heavily in identification and data availability of such characteristics as well as health outcomes in which in most cases these data are not available so it requires a separate research to collect such data. Another limitation of this method is the assumption that the long list of neighborhood variables are independent from each other which practically is not true and considering the correlations between the variables having large number of variables makes the study to be very complex.