Police Service Calls in Madison

February 10th 2016

This week my neighbours have been reporting an increase in the number of thefts from unlocked vehicles. Discussions abounded as to whether the perpetrators were thugs or local kids and solutions ran the gamut from locking your vehicle to installing outdoor lights.

The underlying theme that dominates these discussions, however, is whether the neighbourhood is getting worse or better. Often feelings on this are anecdotal; based on newspaper reports, social media posts, or the direct experiences of the person and their friends & family. Some people think the area is going to hell in a handbasket; others don't think twice about leaving their cars unlocked.

In the below report (best viewed full screen), I set out to provide some facts about police service calls in the neighbourhoods of Madison, as well as trends & patterns.

Full Screen Version

The first tab shows where police service calls are taking place. At the top, you can filter by year, month, or elementary school catchment area (the easiest way I could think of to define a neighborhood). The map is zoomable and you can click any dot to see the list of incidents at that location.

The second tab shows trends over time. You can filter by year, month, time of day, elementary school catchment area, and incident type to see if the number of incidents is going up or down.

The 3rd tab shows patterns. Again you can filter by elementary school catchment area. You can see right away that 6am is a quiet time for the police. At the bottom left, you can click on an incident type and see how the time of day varies. (Stray animal reports peak in the morning for example.) The scatter chart shows incident numbers plotted against the average home value. You can see, for example, that more expensive houses have more bicycle thefts. Areas with cheaper houses have more gas drive-offs.

All of the charts on the 3rd tab are clickable, and they will filter the other charts. So if you want to see what types of incidents are more common at night, or which streets in your neighbourhood have the most incidents, you can do so.

Some caveats abound: I have to refresh the data manually, so it may be out of date. At the time of writing, it included calls from 2012 through 6 Feb 2016 (4 days ago). If it looks more than a week out of date, contact me and I can refresh it.

A service call does not mean a crime took place. This is not a report of crimes. People call the police for all kinds of reasons.

Incidents are indicated by block number, not address. So a dot over a particular house does not mean the incident occurred at that house.

Not all incidents are mapped. When there are too many incidents to plot, only some will be shown. (Use filters to narrow down the selection.) Not all incidents are mappable. Not all houses could be associated with an elementary school.

This is just a first draft that I whipped up last night - let me know if you have any suggestions for improvements. One thought I had was correlating weather (temperature & precipitation) with incidents. What type of incidents spike on hot summer nights vs. cool ones?

If you liked this report, and want to learn how to build reports like this for yourself, you can register for my Mastering Power BI pre-con at SQL Saturday Madison in April 2016.

Finally, I would very much appreciate it if you would consider voting for another of my reports in Microsoft's Best Report Contest (prior to February 14, 2016). You can vote for it here: What's in a Name?. You may have to register for the Power BI community website first. Learn more about it (and see a better version of it here: http://www.leonardmurphy.com/blog/power_bi_top_10_best_report_contest)

Go back...

Visitor Comments

Post Your Own Comment

Note that comments may not appear immediately on this site.

Commenting is not available in this channel entry.