Are most published research findings false? Trends in statistical power, publication selection bias, and the false discovery rate in psychology (1975-2017).

This means that a strongly nonsignificant p -value is equally likely to be treated by publication selection bias and dropped with an equal probability in the case of publication bias as it is in the case of p-hacking (e.g., borderline significant results, p = 0.051 or far off p = 0.9).

also in 11,409 other papers