In once case they released a study of how algorithms used by police were supposedly biased against minorities. Their results were "almost statistically signific

In once case they released a study of how algorithms used by police were supposedly biased against minorities. Their results were "almost statistically significant". In other words, they weren't statistically significant. But they went ahead and published it anyway, under an alarming clickbait headline. https://www.chrisstucchio.com/blog/2016/propublica_is_lying.html https://www.chrisstucchio.com/blog/2016/propublica_is_lying.... It makes me pretty upset, because that is the study/article that everyone references when discussing that issue.