| Herman Rubin 2006-10-18, 4:34 pm |
| In article <1161137923.362033.219150@f16g2000cwb.googlegroups.com>,
<awthrawthr@yahoo.com> wrote:
>Mark Probert wrote:
.................
[vbcol=seagreen]
[vbcol=seagreen]
[vbcol=seagreen]
>Charles Simone, M.D. has done a meta-study of over 100 studies. It was
>found that cancer patients who use dietary supplelements have TWICE the
>survival rate as those who don't. Checkmate. (You sure did walk into
>that one...did it hurt before the 'lights' went out?)
It is quite probable that using dietary supplements increase
survival rates, but the standard use of meta-analysis happens
not to be valid. Studies with "non-significant" conclusions
are not published. This cam SOMETIMES be treated, but not
many statisticians know how to do it if it can.
For a non-medical example, after it was found that there was
not conservation of parity in the decay of mu-mesons in air,
it was searched for in published emulsion data. The theory
stated that this should be 9-10%, and the emulsion data gave
15%. Both the theory and the data are correct.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hrubin@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558
|