Table 1 The most appropriate p value significance threshold for a one-tailed Student’s t test for independent samples in a series of common study designs looking for a “medium effect size” of 0.5
Type of study | Prior odds of H1 relative to H0 | Seriousness of β relative to α | Sample size in each group | p cut-off value* |
|---|---|---|---|---|
Adequately powered randomized clinical trial | 1:1 | 1/4 | 100 | 0.0158 |
300 | 5.3 × 10−4 | |||
Adequately powered exploratory epidemiological study | 1:10 | 1/4 | 150 | 0.0013 |
300 | 1.3 × 10−4 | |||
1000 | 1.8 × 10−9 | |||
Typical psychological experiment | 1:4 | 1/12 | 150 | 0.0012 |
Discovery-oriented exploratory research | 1:1000 | 1/16 | 1000 | 6.2 × 10−11 |
Underpowered, poorly performed phase I/II randomized clinical trial | 1:5 | 1/16 | 150 | 7.8 × 10−4 |
Confirmatory meta-analysis of high-quality randomized clinical trials | 2:1 | 1/1 | 150 | 0.0230 |