Comparing proximity measures of exposure to geostatistical estimates in environmental justice research

This paper tests the validity of proximity as an estimate for environmental health hazard exposure, and suggests how it may be used as an indicator in future environmental health and justice research. Using geostatistics and geographic information systems, air pollution monitoring data in Hamilton, Canada are interpolated to obtain local estimates of total suspended particulates. These estimates are used address the following questions: How does the distribution of proximity to health hazards compare with monitored air pollution data? Does the use of proximity rather than air pollution data significantly change the substantive conclusions of environmental injustice in models with sociodemographic data? The results show that proximity measures can be useful indicators if flexibly applied. Guidelines for future applications are discussed.