Comment by achy

It's always interesting when statistics correlate across multiple frames (time, location, etc.); however that still doesn't prove causality. I'm inclined to think that the causes behind an area being slow to remove leaded gasoline could also affect crime rates.