This paper builds a model of hypothesis testing with p-hacking and gives critical values that correct the inflated type 1 error rate caused by p-hacking. As a rule of thumb, robust critical values are classical critical values with one fifth of the significance level.
This paper develops a model of science. It shows that due to homophily in tenure decisions, false paradigms may persist when a science has low power. Low power may come from a lack of evidence, or from a reluctance to base tenure decisions on available evidence.