Error and Bias in Geocoding School and Students’ Home Addresses
2008

Error and Bias in Geocoding School and Students’ Home Addresses

Sample size: 126 publication Evidence: moderate

Author Information

Author(s): Eric A. Whitsel

Primary Institution: University of North Carolina, Chapel Hill

Hypothesis

How does positional error in geocoding affect the assessment of traffic-related exposures for school children?

Conclusion

Positional errors in geocoding can lead to significant overestimations of the number of children at risk for traffic-related exposures.

Supporting Evidence

  • The study found a 200–500 m range of mean positional errors in geocoded addresses.
  • Lower thresholds for defining traffic-related exposure lead to overestimations of risk.
  • The sensitivity and specificity of the 100-m threshold were around 90% at positional errors of 250–300 m.

Takeaway

When we try to figure out how close kids' homes are to busy roads, mistakes in the addresses can make it look like more kids are in danger than really are.

Methodology

The study analyzed positional errors in geocoding school and home addresses and their impact on traffic exposure assessments.

Potential Biases

There is a risk of overestimating exposure due to positional errors in geocoding.

Limitations

The findings may not be generalizable beyond Orange County without further validation.

Participant Demographics

Participants were school children from Orange County, Florida.

Digital Object Identifier (DOI)

10.1289/ehp.11542

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication