Valentine’s Day update

There has been progress on several fronts the last couple of weeks, and i’ll use this page to capture the important bits:

  • We met 31 Jan with members of Oakland PD and IT department to work towards a mutually agreeable API for new data.  My slides from that meeting are available as a PDF file: opd_openData_130131_core
  • I’ve put some of the basic utility code I’ve used to digest the OPD, Alameda County and Urban Strategies Council crime data into a git repository: https://github.com/rbelew/OPD_util.git

I’ve also continued to dig into apparent discrepancies between the data provided by OPD and USC’s analysis of this.  While in general there seems to be pretty good agreement, I’ve found a number of situations where:

  • OPD and USC incidentID numbers are identical and
  • The two records agree on the same date of occurrance and
  • there is a single USC record associated with the OPD incident,

but the statute (and often UCR) code assigned by USC varies?!  I’ve put a sample of those identified using these probe OPD_CType examples and the statute typically (but not uniformly!) applied:

probeTbl = {"BATTERY": "PC242",
"BURGLARY-AUTO": "PC459",
"BATTERY:SPOUSE/EX SPOUSE/DATE/ETC": "PC243 (E)(1)",
"BURGLARY-FORCIBLE ENTRY": "PC459"}
the resulting anomolies are available in this file: USC_dbug_sample.csv.gz.  There are also 310k OPD records (within the overlapping dates) that do not seem to be in the USC data set?

Leave a Reply

Your email address will not be published. Required fields are marked *