Johannes Huesing wrote: > > I have little problems with scientists publishing all their findings with > p-values (or > confidence intervals) but big problems with people who use them as a > filter. >
While I am aware that CI and p-values have a common root, I do not to throw both into one pot for the average medical end-user. When two treatments for blood pressure show a significant differences, this only tells you that the sponsor had a lot of money to show his expensive new product is worth the money. When you tell the medical doctor that the CI of the difference has a range 1.1 mmHg , she would say: I don't care, it's not relevant, both products are good enough for me. I tend to strongly stress the difference between relevance (medically weighted CI) and significance (pure numerics) in lectures. After all, it's difficult enough when doing sample size estimation to get something like a "medically relevant difference". So I have to pay back by giving them a medically-to-be-weighted difference after study end. Dieter -- View this message in context: http://n4.nabble.com/OT-Sorta-Odds-Are-It-s-Wrong-tp1593626p1595554.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
