Medicine's Ten Greatest Discoveries

Herman Rubin hrubin at b.stat.purdue.edu
Mon Sep 7 14:03:17 EST 1998


In article <EQrI1.2130$c3.3697154 at tor-nn1.netcom.ca>,
David Lloyd-Jones <dlj at pobox.com> wrote:

>Mitchel L. Galishoff wrote in message
>+ADw-35f1da2d.589447+AEA-news.mindspring.com+AD4-...
>+AD4AIg-David Lloyd-Jones+ACI- +ADw-dlj+AEA-pobox.com+AD4- wrote:

			....................

>Galishov's stupid truism -- which can be more briefly put as +ACI-coincidence is
>not causality+ACI- -- is one of the most common and fashionable snide remarks
>current among dolts and poseurs. The fact is that +ACI-post hoc ergo propter
>hoc+ACI- is not a logical proof -- but it is one of the most generally powerful,
>and most likely to be successful, of all hypotheses. Numerical coincidence
>would probably be its only competitor in these regards.

Galishoff is properly using what is known about inference, and we have
many examples to prove it where the causes are known.  

>In somewhat the same sense, multiple regression analysis, which was
>destroyed epistemologically by Locke and Hume a couple of centuries before
>it actually came into general use, has no claim whatsoever to Platonic
>logical rigor -- but is one of the most useful forms of explanation that we
>have.  Usually right, on rare occasions wrong.

Do you know what you are talking about?  I know exactly what multiple
regression analysis is, and what it is not.  It does not have the power
you ascribe to it, especially if there are many factors, as is usually
the situation.

>If Galishov has a list of other reasons for the failure of American medical
>practice to serve the American people, I am willing to bet money that my
>list is as long as his.  However I am unrepentant about my suggestion that
>letting PhD's from mail-order psych schools prescribe powerful drugs in not
>good for the general level of health.

Why should anybody be able to prescribe most drugs, or to proscribe them 
without someone else prescribing them?  Provide the information, and let
people decide for themselves.  

-- 
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, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
hrubin at stat.purdue.edu         Phone: (765)494-6054   FAX: (765)494-0558



More information about the Bioforum mailing list