Geocoding Errors in Assessing Children's Exposure to Air Pollution at Schools
Author Information
Author(s): Paul A. Zandbergen, Joseph W. Green
Primary Institution: Department of Geography, University of New Mexico; Department of Geography, University of South Florida
Hypothesis
How does positional error in geocoding affect the analysis of children's exposure to traffic-related air pollution at school locations?
Conclusion
The study found that typical geocoding methods introduce significant positional errors that can lead to unreliable assessments of children's exposure to air pollution near schools.
Supporting Evidence
- The positional error in geocoding of schools was found to be substantial, with a 95% root mean square error of 196 m for street centerlines.
- Bias and error in proximity analysis to major roads were unacceptably large at distances of less than 500 m.
- Geocoding results consistently overestimated the number of schools at risk for exposure to traffic-related air pollution.
Takeaway
When scientists try to figure out how much air pollution kids are exposed to near their schools, they often make mistakes because the addresses they use aren't always accurate.
Methodology
The study compared geocoded school locations with actual locations using various geocoding techniques and analyzed proximity to major roads to assess exposure to air pollution.
Potential Biases
Positional errors in geocoding can lead to overestimation of schools at risk for exposure to air pollution.
Limitations
The study's findings may not be generalizable beyond suburban areas with large school parcels, and it only examined one county in the U.S.
Participant Demographics
Children attending public schools in Orange County, Florida.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website