Every year in December, the Lincoln Police Department puts a special focus on drunk driving enforcement. Since the vast majority of drunk driving arrests occur on the late shift, Capt. Marty Fehringer, one of the overnight Duty Commanders, was this year's coordinator, and sent me the results: 213 DWI arrests by LPD during the month of December. This is the all-time single month record for drunk driving arrests in Lincoln. As I have often said, there is nothing that a police officer does that has a greater impact on public safety than arresting a drunk driver. The officers who contributed to this project should be complimented for their efforts.
We weren't alone in focusing on DWI during the month, and the local news has reported on the results from the Lancaster County Sheriff's Office, and the statewide arrests by the Nebraska State Patrol.
Here's an interesting way of looking at LPD's drunk driving arrests: a temporal grid of the past few years, cases. I've blogged before about using this method for visualizing data sets that have strong time-and-day patterns, which is clearly evident in the DWI arrests. You can click on the image to enlarge the view.
This grid, by the way, is a screen capture from our latest version of CrimeView Dashboard. It took me less than five minutes to prepare this for my blog this morning. Performing this analysis and producing a graphic like this a few years ago would have made a decent project for the week. It also would have required a few thousand dollars worth of software on a fat honking PC, and a GIS analyst with extensive training and experience. Now it's a few minutes on my wife's low-end Dell with an iffy WiFi connection in our little home office, while sipping my first cup of joe. Pretty incredible, when you think about it.
Thursday, January 5, 2012
Subscribe to:
Post Comments (Atom)
9 comments:
Maybe if I spent more time, I'd figure it out, but the colors and legend at the bottom have me confused. Is that BAC levels? You know, Tom, the first thing I was taught when it comes to making good charts and graphs is the importance of labeling. :)
Hey Chief
Visualisation tip: Dont be afraid to shift the temporal scale on these charts. When I visualise temporal patterns like this I try to keep my main point of focus central and unbroken. For crimes at night, such as this offence or alcohol related violence, I do midday to midday. For burglary reporting times, I do midnight to midnight as we usually have a spike in the morning and one in the evening (people call in when they wake up and find they have been burgled during the night, or come home from work and find they have been burgled during the day). It means that the pattern is more naturally accesible to the reader, as long as the x axis is clearly labelled (using Midday and Midnight rather that 00:00 and 12:00 helps a lot).
Really good post btw! I cant imagine being given a week to produce a report these days!
Thank you for all the hard work with the DWIs. We'll never know how many lives were saved by your efforts.
Hey Rory,
You make a great point. When I do animations by hour, I normally start at 0500--sort of the start of my day--because it seems like a more intuitive way to visualize (ise) the time-of-day phenomenon. I wonder if I could construct a grid where you could enter a parameter to select the starting day and hour....
Glad I found your blog (I'd normally punctuate with an exclamation point, here).
Steve,
The legend at the bottom is standard deviation breaks: as the colors go from cool-to-hot, that's indicating the relative increase in the number of arrests during each hour on each day.
I didn't really think it was BAC, as I've never heard of a negative number, and any over three were probably dead, so thanks for clearing that up.
Rory's blog needs a pirate.
Chief, from all of us who have lost loved ones to Drunk Drivers, thanks to LPD'S continued effort to stop drunk drivers. Keep fighting the good fight. Thanks to all officers.
Tom, whoops. What I meant to say was, Director. Old habits sometimes are hard to change. Thanks.
Post a Comment