[OT] Re: [R] p-values

Rolf Turner rolf at math.unb.ca
Tue Apr 27 22:53:53 CEST 2004

Greg Tarpinian wrote:

> I apologize if this question is not completely appropriate for this
> list.
> I have been using SAS for a while and am now in the process of
> learning some C and R as a part of my graduate studies.  All of the
> statistical packages I have used generally yield p-values as a
> default output to standard procedures.
> This week I have been reading "Testing Precise Hypotheses" by J.O.
> Berger & Mohan Delampady, Statistical Science, Vol. 2, No. 3, 317-355
> and "Bayesian Analysis: A Look at Today and Thoughts of Tomorrow" by
> J.O. Berger, JASA, Vol. 95, No. 452, p.  1269 - 1276, both as
> supplements to my Math Stat.  course.
> It appears, based on these articles, that p-values are more or less
> useless.  If this is indeed the case, then why is a p-value typically
> given as a default output?  For example, I know that PROC MIXED and
> lme( ) both yield p-values for fixed effects terms.
> The theory I am learning does not seem to match what is commonly
> available in the software, and I am just wondering why.

You shouldn't pay too much attention to the religeous ranting of
blinkered Bayesians.

Of course p-values should be taken with a grain of salt.  But
then so should everything else.  Including Bayesian methods.

Remember that all models are just that --- models.  They are not
reality.  George Box said something like ``All models are wrong.
Some models are useful.''

Experience has shown that p-values, ***interpreted with proper
caution and good common sense***, are very useful indeed.
And that's why they're there.

Bayesian methods and models are good too; they're just not the
be-all-and-end-all that the Zealots proclaim them to be.  And you
have to avoid the utterly ridiculous concept of ``personal
probability'' that Bayesians seem to find central.

A propos of the notion of ``personal probability'', the physicist
Carl Sagan (in his wonderful book ``The Demon Haunted World'')
referred to the ``dangerous doctrine'' that strength of belief has
anything to do with the truth of a proposition.  He was not writing
in any statistical context.  Rather, he was discussing the daffy
notion of ``alien abductions'' which apparently has been taken
seriously by some Ivy League psychologist on the basis that many of
those who claim to have been abducted so fervently ***believe***
their own tales. (So they must be true!)  However, the point still


					Rolf Turner
					rolf at math.unb.ca

More information about the R-help mailing list