Correlation or causation? combines seemingly disparate positions on social issues to build odd profiles of people by looking for statistically anomalous overlaps. What that means is they ask random groups of people a bunch of questions to figure out odd relationships. Take, for example, people who try to raise sea-monkeys. They are. Would you believe. More likely than average to dye their hair.

“In general, 44 percent of people have dyed their hair at some time. But among people who have tried to raise Sea-Monkeys, 63 percent have dyed their hair.

Based on a survey of 326 people who have tried to raise Sea-Monkeys and 1174 people in general.”

Love it.