[R] acceptable p-level for scientific studies

Peter Dalgaard BSA p.dalgaard at biostat.ku.dk
Wed Dec 18 19:43:03 CET 2002


Frank E Harrell Jr <fharrell at virginia.edu> writes:

> - Confidence limits can be good compromise solutions (some journals are
>   almost disallowing P-values in favor of CLs)

I gather that some journals *are* doing that, to the extent of
actually disallowing multi-df testing (I'll say that it is a case of
proselytes being holier than the prophets). And of course a 95% CI is
strongly linked to testing at a fixed significance level, so the
information contained in the P-value is lost that way.

> - P-values are dangerous, especially large, small, and in-between ones.

<grin> Good one...

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907




More information about the R-help mailing list