Hacker News new | ask | show | jobs
by gavanwoolery 4778 days ago
Maybe I missed it, but I did not see a sample size. Without this, the study is meaningless - the odds of a correlation between two variables each with two possibilities is not that improbable (i.e. flipping two coins and hitting heads both times in one group, and tails in the other group, with 20 percent error).