Actually, I am glad you eventually did sound off - it has been a while on my
watch, and by now, SAS-L has been missing your contributions a lot.
Paul M. Dorfman
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On
> Behalf Of Dale McLerran
> Sent: Monday, November 08, 2004 6:42 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: length of numeric variable
> Ron raises a point which seems quite lost to most people in
> todays world. When the slide rule ruled, it was very
> important to know just how much accuracy one could assume for a value.
> Now, with everything done by computer, many assume that the
> statistic spit out is precise to all reported digits. That
> is often (usually?) not the case. Ed gives a good
> description of what is meant by significant digits. I hope
> that folks take heed of the issue of significant digits.
> At the same time that we are discussing significant digits,
> let us not forget the other critical element for evaluating a
> measure - units of measurement! Remember the mission to Mars
> which failed because the engineers on NASA's team were
> operating with different measurement units? One can have a
> very precise measurement, but if you lose track of the units
> of measurement, the precision of the measurement may be of
> absolutely no value.
> Thanks for the opportunity to sound off on an issue which has
> bothered me much over the past few years.
> --- Ed Heaton <EdHeaton@WESTAT.COM> wrote:
> > Ron;
> > Significant figures, to an engineer or chemist, are those figures
> > which are more than just placeholders. That is, if I aver
> that there
> > are 300,000,000 people in the US, none of the zeros are
> > they simply move the three over to the left. However, if I
> claim that
> > there are 100 centimeters in a meter, both of the zeros are
> > significant. If I claim that a meter is 39.370 inches, the zero is
> > significant. But, if I claim that an inch is
> > 0.0254 meters, none of the zeros are significant; they just move the
> > 254
> > over to the right.
> > The scientific community clarifies this problem of
> significance with
> > scientific notation as follows.
> > 3*10**8
> > 1.00*10*2
> > 3.9370*10**1
> > 2.54*10**-2
> > With scientific notation, all zeros are assumed significant.
> > The question of significance should only apply to zeros.
> If I claim
> > that I live 2.97 miles from work, then I either live
> between 2.965 and
> > 2.975 miles from work or I lie.
> > Ed
> > Edward Heaton, SAS Senior Systems Analyst, Westat (An
> > Research Corporation), 1600 Research Boulevard, RW-3541,
> Rockville, MD
> > 20850-3195
> > Voice: (301) 610-4818 Fax: (301) 610-5128
> > mailto:EdHeaton@Westat.com http://www.Westat.com
> > -----Original Message-----
> > From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On
> Behalf Of
> > Fehd, Ronald J.
> > Sent: Monday, November 08, 2004 3:56 PM
> > To: SAS-L@LISTSERV.UGA.EDU
> > Subject: Re: length of numeric variable
> > really, people!
> > are we talking millimeters, miles or megaparsecs?
> > who remembers
> > what the significance of 'significant' digits means ... to an
> > engineer?
> > let me see, iirc:
> > "close only counts in horseshoes"
> > and Iraq :-Q
> > Ron
> Dale McLerran
> Fred Hutchinson Cancer Research Center
> mailto: dmclerra@NO_SPAMfhcrc.org
> Ph: (206) 667-2926
> Fax: (206) 667-5977
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection
> around http://mail.yahoo.com