0
$\begingroup$

I need some urgent advice for a grant application. I am evaluating a new diagnostic test and at the last minute a Professor has offered me the dataset from a completed prospective study of the disease my test diagnoses. He wants me to check if his data is adequately powered to achieve and AUC of 0.75.

The cohort is n=318 and the prevalence of the disease is 50%.

Can this be done - is it scientifically sound? Does this qualify as post-hoc power calculation?

$\endgroup$
0
$\begingroup$

I am under the impression that yes it would qualify as a post-hoc power calculation given the analyst here would perform the power analysis after the sample size is determined. However, if it feasible and/or ethical to further recruit participants, then the situation becomes more gray as the analysis would could then determine the suitable cohort size. Apologies; yet I would not be able to answer your other questions.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.