Statistical testing is hard. Some of the common dangers are well known; repeat a study twenty times and there's a good chance you'll see a significant result at least once. The Replication Crisis was the outcome of such abuses of statistical testing in scientific
