Testing Iowa

You probably noticed that the bar plot of first digits is alarming: it looks quite different from what Benford's Law prescribes! Before you get ahead of yourself, though, realize that those bars each only contained a handful of counties, so you don't actually have that much data.

This is a prime example of when a hypothesis test is handy. It can tell you if the structure that you see (in this case, the deviation from Benford's Law) could just be due to the random variability of your small sample, or if it's indicative of a systematic difference.