This blog will be dedicated to examining and promoting civic data in Chicago, Cook County and Illinois.
WBEZ is partnering with the Smart Chicago Collaborative to promote civic data. This blog will be part of that collaboration.
We'll post original data sets @OpenSocrata
Calculating things can be hard for journalists.
Actually, most journalists go into journalism because we’re bad at math — or so the joke goes.
In my story about pot arrests vs. tickets, I first looked at the annual numbers. The ever-popular Crimes 2001 to present data set. I was trying to discern a trend, seeing if the ticketing ordinance had an effect on arrests.
At first glance, it looked like a yes-or-no situation. The reason: pot arrests seemed on a decline from a peak in 2007, briefly going up in 2010, then falling again in 2011.
Here’s a chart:
Now, from the chart, the decline in 2012 seemed like a natural part of the overall decline. If that were the case, then the tickets had very little impact on arrests.
It wasn’t until I broke out the monthly year over year totals that the staggering drops stood out:
The drops are much more acute in August. Now, taking the monthly totals from the tickets issued that month (110), a person can reasonably assume that a 831 drop in year over year arrests would not be natural, especially when the previous months were showing steady declines.
I first came across the arrest drop because in sifting through the numbers I had asked the police news affairs if the the arrests flags were in error because the numbers seemed so low. They asked why I thought that? But were quick to point out that it was the mayor’s office that maintained the data portal site.
(The data portal site is controlled by the Mayor’s Office of Innovation and Technology.)
I had created a sub-data set of the 2011 < 30g pot arrests
Then created another one for 2012 <30g pot arrests
I had to refine the month-to-month totals further and gather them in a Google Doc to break them apart and visualize.
In doing this, I encountered a handful of errors with some records and the date fields. I got one total for 2012 when I refined the set by year, but got a different total, off by about 4-6 when I went month to month and added them up.
I wasn’t able to find the individual broken dates, but they seemed centered on February 2011. Even more frustrating: the records were retroactively being changed, day-to-day, or rather being updated with old crimes being reclassified.
Despite that, the records remained relatively consistent with only slight changes by 1 or 2 records being added retroactively.
It’s almost not possible to get a concise record of breakdown of 15 grams (the ordinance amount) or 30 grams (the amount listed on the data portal).
When the police news affairs spokesperson got back to me, they gave me similar figures for amounts under 10 grams. Those numbers nearly, but not perfectly matched the <30 gram figures.
This would seem to indicate that the 30 gram number is a guesstimate at best for street cops.
And that for simplicity sake, it’s broken down as:
Less than 10 grams = you probably got caught smoking a joint
Less than 30 grams = baggie’s worth
Greater than 30 grams = you probably have enough to sell going to sell it somewhere
The reason that police, dealers and others pay attention to how much pot is carried on them might have something to do with this: