Keep it simple

A crime analyst from another state got in touch with me yesterday. I'm not sure she would want to be identified, so I'll keep it anonymous for the time being. She is in the envious position of starting up a new crime analysis unit from scratch, and has been engaged in getting a lot of good training—a real luxury for most new analysts. She and her manager jumped into a webconference with me earlier this year, and she was now seeking my advice on the next step:

“In your experience, what would you consider to be the 3 or 4 most important things I could/should be doing (or products I should be producing)? I really want to make a difference here and do this job to the best of my ability.”

After replying, It struck me that my response might be valuable to other analysts who struggle with the same fundamental question, so here it is:

“Of course I remember talking to you. Glad you are moving forward. Your number one job is to provide timely, actionable information to officers and supervisors about important things they would not have known about otherwise. Make sure it's:

TIMELY

ACTIONABLE

IMPORTANT

...and not already obvious. An email from the crime analysis unit that tells everyone what they already know isn't especially helpful. All these things are relative. The theft of iPods from lockers at schools may not be a critical issue of public safety, but every one of those larceny/thefts counts as ONE part 1 crime--the same as a murder--so they may be important (particularly taken as a whole). It may not be entirely within the ability of the police department to prevent, but on the other hand I can think of a half-dozen action steps that we might engage in to prevent or interrupt such offenses. I may not realize just how common this crime is, or that I'm not the only officer who's investigated these, or that they are almost all in the girls locker room, or that the schools that don't allow students to bring these are immune from these crimes, or that the dollar loss actually exceeds the net from armed robberies this year, or that several of the victims have been members of athletic teams and have occurred during practice ...and so forth.

The best way to establish your credibility is to focus as much of your time as possible on providing such information, and not letting your day be entirely consumed with producing meaningless bar charts or reports for the brass, that have no utility for the women and men who have to do something to impact crime and disorder in your city. You're getting the Cadillac training, but I'm going to urge you to go slow and keep it simple. If you want to really cause eyes to glaze over, create a map of a crime series with a convex hull, ordered labels, and a set of standard deviation ellipses with a prognostication about the most likely date range of the next offense in the series.

Ask officers, sergeants, and detectives: "Is this worthwhile? What else would you find useful, or what would be useful instead of this?" With management-types (chiefs included) don't just point out the crime trend or pattern, suggest ways that it might be changed: "Do you think there is any chance that we could talk the other principals into prohibiting iPods at school?" Become an expert at this by reading everything you can at popcenter.org about the crime trends and patterns that are problems in your city.

I am attaching a document with three pretty easy ideas. This is low-hanging fruit, if it hasn't already been picked. I hope this helps.”