home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!noc.near.net!hri.com!spool.mu.edu!olivea!news.bbn.com!hsdndev!binoc.bih.harvard.edu!rind
- From: rind@binoc.bih.harvard.edu (David Rind)
- Newsgroups: sci.med
- Subject: Re: Bashing, truth, etc.
- Message-ID: <2314@hsdndev.UUCP>
- Date: 22 Dec 92 16:50:50 GMT
- Sender: usenet@hsdndev.UUCP
- Organization: Beth Israel Hospital, Harvard Medical School, Boston Mass., USA
- Lines: 38
-
- In article <Bzo510.23n@mentor.cc.purdue.edu> hrubin@pop.stat.purdue.edu
- (Herman Rubin) writes:
- >In article <2311@hsdndev.UUCP> rind@binoc.bih.harvard.edu (David Rind) writes:
- >>a large enough N will prove an effect. Further, given chaos in
- >>the universe, it is unlikely that we could ever control things in
- >>such a way that we could have any likelihood that such a small effect
- >>would show the same results in an experiment performed twice.
-
- >The problem is even worse. For any N, there is a positive probability
- >that a "statistically significant" result will occur, even if there is
- >absolutely no effect; this is what is usually called the significance
- >level.
-
- This latter is the standard reason for using statistics and talking
- about p-values but is *not* the effect I am discussing. The
- randomness/chaos I am referring to is things like changes in
- the number of solar flares per month, or the gravitational pull
- of Alpha Centauri: By the same reasoning Herman Rubin is using
- to argue that saccharin must have an effect on cancer rates,
- we can similarly conclude that the gravitational pull of Alpha
- Centauri must be an effect modifier of saccharin on cancer rates.
- As the relative positions of Earth and Alpha Centauri change,
- that effect modification changes as well.
-
- Some effect modifiers are not merely quantitative but are actually
- qualitative. Thus, leaving out the randomness inherent in
- sampling from a population, the *true* effect of something
- that has only a miniscule effect to begin with can be expected
- to change if the experiment is repeated in a changing universe.
-
- For this reason, I think it is absurd to insist on the notion
- that we can "know" that the null hypothesis will always prove
- false. If the true effect of A on B is overwhelmed by chaos
- in the universe, I think it is fair to consider the null
- hypothesis true within the realm of human knowledge.
- --
- David Rind
- rind@binoc.bih.harvard.edu
-