I’ve added quite a few features to Portland Crime over the past few months: trends, history, and much more. However, one feature request pops up more often than others—public transit crime.
I’ve worked on adding this feature, but I find myself rushing it along due to time constraints and I’m just not happy with the results. Until I find the time to sit down and hash out the design for this feature, I’m going to keep it under wraps.
It’s not all bad news. Just because my efforts haven’t been successful, doesn’t mean others couldn’t use the data in some creative way. With that in mind, I want to share the data I’ve collected with anyone who wants it. Details below.
What counts as a transit crime report?
According to Wikipedia, Portland blocks are 200ft x 200ft. Given teleportation isn’t a reality, it’s assumed you’ll need to cover a minimum distance of 200ft to reach a stop. Therefore, crimes within a radius of 200ft from a MAX stop’s lat,lon point will be counted towards that stop.
There are many factors that can result in one stop having more crime than another stop. Population, ridership and activity to name a few. These numbers are raw counts. They are not adjusted.
- Location: Some crimes don’t include location data and will not be included in totals.
- Reports, not arrests: These are “reports”. They do not necessarily result in a conviction or arrest.
- Duplicates: Some stops are on multiple routes and are included more than once. You should not add counts from two stops with the same stop_id together.