Saturday, February 04, 2006

Play to Win! - To Catch a Thief

The day has finally arrived!

After too many hours of preparation (as usual...), the To Catch a Thief contest has begun. We have released the first of four clues (map and police report) yesterday, so anyone who can correctly guess the means of transportation used by the burglar(s) can email such guesses to Only UTA students/staff can actually win the ultimate grand prize (1 free plotter map or poster), but anyone who so wished can...Play to Win!.

As I mentioned before ( Jan. 20 | Oct. 14 ), this was inspired by the TV show Numb3rs.

There will be 3 additional police reports/maps released before the hands-on workshop on March 2. The workshop will focus on two spatial analysis tasks. First, we will attempt to locate other targets with a high probability of being struck by the serial burglar(s). Second, we will attempt to track down potential neighborhoods where the burglar(s) might live.

Is the scenario fake?? Of course, but the data sources are actual.

Am I well versed in spatial criminology?? No way...just having some fun. The analysis will mostly depend on multivariate regression and suitability analysis. Who knows...maybe we'll even throw in a touch of an actual algorithm used in the field...

Should a criminology professor or student attend the workshop and scoff at the methods I use, then I will consider the exercise a success. This is actually my hope. I am not aware of any of our criminology faculty or students using GIS or spatial analysis, and would love the opportunity to collaborate with them. Got my fingers crossed.

Why did I have to spend many hours developing this contest with the clues?? Well, I had to ensure that the results would be statistically significant. I had to whip out some of my old stats textbooks to calculate the best way to select targets (points) that will yield a significant correlation. Yes, the analysis will work better than it could ever work in real life. ;) That's the fun of it, eh?

No comments: