3 ms·
Maybe, they should start with their fellow social scientists and change their behavior towards publishing garbage (ie non-reproducible, p-hacked) papers. In tr
by RcouF1uZ4gsC 9y ago
Maybe, they should start with their fellow social scientists and change their behavior towards publishing garbage (ie non-reproducible, p-hacked) papers.
In truth, there is likely to be a massive selection effect, just like with charter schools. Basically, the people who want to change will enroll in the study. We basically know that a subset of people have enough motivation to make themselves better and that is whom the studies will select.
Finally, this improving human "frailty" has a horrible track record in terms of perpetrating horrible atrocities - from the Inquisition to the Soviet gulags. Often, the most evil people in their own eyes have only the purest motives for the good of humanity.
- matt4077 9y agoSocial sciences may produce a lot of "garbage", but they do have at some point failed to avoid learning the most banal ideas about statistical thinking, which do include selection bias.