Honest sample size calculations: we had funding to include 500 patients in our study, so we fiddled with our estimates until the numbers fit
-
-
This Tweet is unavailable.
-
These practical issues are not the kind of topics you'll find in the papers, but you could talk to people working for major CROs.
End of conversation
-
-
-
funding is not the only factor, you can also save or invest into the amount and quality of data you collect per person (cost may range from near zero to >100k per person). Is there an economic analysis on this? Would love to have this for my courses.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
knowing what effect size you’re plausibly powered to detect is still useful, imo. Miettinen has an unusually honest section on economic determiners of sample size, including this glorious line: “Validity considerations alone are often sufficient to imply zero is the optimal size”pic.twitter.com/fOZXjo7JBu
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I'd say funding and the pressure to produce as many papers as you possibly can. Run one study with 1000 people? Naaah, run 10 studies with 100 ppts each! Publications go brrr!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
There’s a fuzzy line at the border between research & outbreak investigation, but in the latter the sample size is often beyond your control. I’ve occasionally been tempted to ask an Outbreak Control Team not to take a suspected food off the market till we have a few more cases.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.