Date: Wed, 1 Jun 2005 13:44:13 -0700
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject: Re: select certain number of records
Content-type: text/plain; charset=US-ASCII
Sigurd Hermansen <HERMANS1@WESTAT.COM> replied:
> I, for one, prefer to have an opportunity to read all of your caveats
> and, if I feel entitled or obstinate, to ignore them. Besides, no one
> has to read anything written by 'cassell.david', though I would always
> do so if on a statistical topic of interest.
> Some of us subscribe to SAS-L to learn about the less than obvious
> of statistical and database programming ... It's what you don't know
> ask that hurts you. Sure you tend to be overbearing, but as long as
> continue to maintain such a high ratio of new information to noise,
> continue to tune in ...
'tend to be overbearing'???? Hey, I'll have you know that I work hard
on being EXTREMELY overbearing! :-) :-)
Yeah, I've noticed. I'll try to control myself. But when my hot
buttons get pressed...
And your comment about "the less than obvious sides of statistical and
database programming ... It's what you don't know to ask that hurts
How true. That's why I read SAS-L. I've learned far more from a long
of SAS-L exemplaries than from any SAS course I've had. It's also why
I nitpick. It's so easy to throw a glut of variables into a hopper and
turn the crank until some pseudo-statistical paste oozes out the chute.
Whether the results mean anything depends a great deal on the skills and
attention to detail of the user. See a few thousand of my rants on
stepwise selection, sampling from data frames, etc., etc...
David Cassell, CSC
Senior computing specialist