Where does the data come from?
- Each morning by 9:30 AM, the Tucson Police Department and the City's Information Technology Department make available for download a file containing records of police incidents for the most recent 45 days. The records are extracted from the Departmentís I/LEAD computer system, geocoded to obtain X and Y coordinates for the approximate location of the incident and packaged up for download. The data file typically includes incidents that were reported as recently as 3:00 AM of the same day.
How is the neighborhood label established for each incident?
- When it becomes available, NSN downloads the police incident file. The file is processed for addition to a PostgreSQL database table. A PostGIS-enabled query matches the police incident data points with a file of neighborhood polygons that is based on the City of Tucson's neighborhood boundary file (nha_city.shp). Additional neighborhoods, not registered with the City, have been added to the base file. Incidents along boundary streets, particularly at intersections, may or may not be assigned to a given neighborhood. Finalized web pages are uploaded to the NSN website, usually shortly after 10 AM.
How reliable is the data?
- Police incident data used here covers the City of Tucson. It does not include data from Pima County, South Tucson, University of Arizona, Pima Community College, Davis Monthan AFB, Tucson International Airport, state, federal or other law enforcement agencies.
Police incidents were screened according to definitions included in the local version of the Uniform Crime Reporting program of the U.S. Federal Bureau of Investigation. Police incident information is approximate and subject to change. This is particularly true of the most recent information.
Data on reported and discovered police incidents is based on factors that change over time; therefore this data should be used with caution. Some factors affecting police data are:
- Actual numbers of police incidents as interpreted through Uniform Crime Reporting rules, including record updates based on new information
- Rate of reporting by citizens or crime victims
- Time lags in completion of or automation of police reports
- Neighborhood boundaries as included in geoprocessing tables
- Special efforts of police to reduce crime (which may increase the amount of crime known to police)
- Special efforts of victims to reduce crime, such as shoplifting prevention programs (which may increase the amount of crime known to police)
- Data processing practices, including geocoding match rate
- Changes in the law or in criminal justice system policies and practices
Did the updating work this morning?
- Some listings will not show any incidents from yesterday, making the viewer wonder whether the current file was updated this morning. No incidents may come from the fact that there were no police incidents in the neighborhood during the last 24 hours. Check other neighborhoods.
Another reason may be that the update process did not work correctly. The program is set to send an administrative email message to the database manager when errors occur. It is also possible that a computer was down when the timed update process was to occur. The entire update process is repeated several hours later to allow for a downed computer.
How can I contact you?
- This system was created by Donald Ijams, who can be reached at dsijams -at- gmail.com . Feedback and error correction information is most welcome.
Tell me about the columns on the listing.