177174 A methodology for geocoding California collision data

Monday, October 27, 2008

John M. Bigham , Traffic Safety Center, University of California, Berkeley, Berkeley, CA
Thomas M. Rice, MPH, PhD , Traffic Safety Center, University of California, Berkeley, Berkeley, CA
Junhak Lee , Traffic Safety Center, University of California, Berkeley, Berkeley, CA
Swati Pande , Traffic Safety Center, University of California, Berkeley, Berkeley, CA
David R. Ragland, PhD, MPH , Traffic Safety Center, University of California, Berkeley, Berkeley, CA
Background/Purpose

Geocoding is the process of assigning geographical coordinates to a location, referenced by a street address or intersection. Once a coordinate is established, a variety of spatial analyses can be conducted. The purpose of this paper is to develop a methodology and document steps needed to accurately geocode all severe and fatal injury collisions recorded in the California Statewide Integrated Traffic Records System (SWITRS).

Methods

Pre-processing of SWITRS must be completed to clean known errors. This consists of removing invalid addresses or characters, fixing common misspellings, and other tasks to ensure the highest level of geocoding possible. Spatial analysis software can then be used to geocode an intersection point using the primary and secondary streets listed in SWITRS. However, special consideration must be given to collisions that occur at an offset distance from an intersection or along major state routes and custom code must be developed.

Results/Outcomes

The act of geocoding SWITRS data must be taken in a stepwise approach. Pre-processing of data is necessary to avoid making incorrect matches and improving the efficiency. After pre-processing, different geocoding methods must be used depending on the location of the collision to ensure accurate representation. Quality checks are needed to identify and remove any false positives.

Conclusions

Providing agencies with guidelines for better data preparation will increase the efficiency of geocoding collisions. Using multiple software packages and custom developed programs can increase the accuracy of mapping the severe and fatal injury collisions from SWITRS up to 85%.

Learning Objectives:
1. Describe a methodology for geocoding California collision data. 2. Identify sources of error and workarounds for the geocoding process. 3. Develop a plan for accurately geocoding other sources of collision data.

Keywords: Motor Vehicles, Data Collection

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a principle investigator and author of the research paper.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

See more of: Motor Vehicle-Related Injury Posters
See more of: Injury Control and Emergency Health Services