During our series on looking at traffic stop data in Connecticut, many commenters have offered differing theories on why certain town police departments issue more tickets than others. Some readers speculated, for instance, that towns with a small business tax base or a large number of commuters might issue more tickets.
In order to communicate how strongly two things are related, statisticians have come up with something called a correlation co-efficient. It’s a number between -1 and 1 that measures the strength of the linear association between two things. It works like this:
- The closer it is to 1, the stronger the positive correlation. In other words, when one thing goes up, the other things goes up — for example, if you spend more time at an hourly job, you’ll earn more money.
- The closer it is to -1, the stronger the negative correlation In other words, when one thing goes up, the other thing goes down — for example, if you spend more time at work, you’ll probably spend less time at home.
- The closer it is to 0, the more likely it is that there’s no correlation.
But correlation does not imply causation. For example, larger cities have more car accidents — but it doesn’t mean that larger cities are necessarily more dangerous places to drive. And just because two things share a similar trend doesn’t mean they are related.
Still, we tried to pick a handful of data sets that were probably affiliated. Test your assumptions. Can you guess the strength of the correlations in the quiz below? Move the slider to where you think the relationship exists: Is there a weak positive correlation? A moderate negative correlation? Zero correlation? Note: This only measures towns with police departments and excludes state police data.