New blogpost π¨ : What do you do if you found a significant result, but your study was underpowered? How reliable is your finding? I discuss Type M and Type S error
https://mzstats.blogspot.com/2023/02/what-not-to-do-with-non-null-results.html
#statistics #frequentist #NHST #pvalue #sensitivityanalysis #falsepositiverisk #rstats
#FalsePositiveRisk in #Medicine
We provide an empirical test of Ioannidis's prediction and find the false discovery risk is well below 50%. Now as citation friendly preprint.
The influential claim that most published results are false raised concerns about the trustworthiness and integrity of science. Since then, there have been numerous attempts to examine the rate of false-positive results that have failed to settle this question empirically. Here we propose a new way to estimate the false positive risk and apply the method to the results of (randomized) clinical trials in top medical journals. Contrary to claims that most published results are false, we find that the traditional significance criterion of $Ξ±= .05$ produces a false positive risk of 13%. Adjusting $Ξ±$ to .01 lowers the false positive risk to less than 5%. However, our method does provide clear evidence of publication bias that leads to inflated effect size estimates. These results provide a solid empirical foundation for evaluations of the trustworthiness of medical research.
Dr Mircea Zloteanu πΊππ
Ulrich Schimmack