Date:         Mon, 26 Apr 1999 11:12:04 -0400
Reply-To:     HERMANS1 <HERMANS1@WESTAT.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         HERMANS1 <HERMANS1@WESTAT.COM>
Subject:      ***Re: (MVS) Re[2]: How to trick Proc Freq?
Comments: To: dpider@hotmail.com
Content-Type: text/plain; charset=US-ASCII

I say that we should give Mr. Pider the benefit of doubt.  Ian Whilock has
posted a test of Mr. Pider's hypothesis (SAS procedures always more efficient
than GURU magic).  As I read the test results, any reasonable person would not
reject the null hypothesis.  Now Mr. Pider has an opportunity to challenge the
results with something more than 15 years of SAS experience (too many
statisticians in SAS-L to accept "trust me"'s).  I hope that you will respond,
Mr. Pider, so we can resolve this question.  Do you have any methodological
issues or evidence that would invalidate Mr. Whitlock's results?
If you are bluffing (or blustering), your fellow technical managers and
experienced SAS programmers will lose face.  I, for one, hope that you will not
let us down.  That would undermine our authority and force us to justify our
methods.  Sig