Please accept my intrusion with a follow on comment.
What bothers me about any of these is the different meanings attached to
%.
20.0 +/- 0.5% *OR* 20.0% +/- 0.5%
could mean 20% (the original measurement was in percent) +/- 0.5
(original % units), OR it could mean 20(original%) +/- 0.5% of that,
which would be 0.04% of the original %.
Suppose the original measurement is percent of people in a sample who
score 'agree' or 'strongly agree' on a question with a Likert scale
response. If I report the numbers as "we expect the response of the
population at large to be 20% +/- 0.5% 'agree' or 'strongly agree,' I
sure hope that a reviewer will force me to clarify exactly which % it is
that I mean. Regardless of how many %'s I use in the statement.
In certain technical archival publications, I have usually seen the units
written as
20.0 +/- 0.5 m
Possibly on the grounds that the units are 'obvious' and take up less
space. Using David Smith's criterion, I think they also sound better,
spoken out loud.
Cheers,
Jay
Steve Bousquin wrote:
> Settle (or join) a debate: should a confidence interval with units be
> written as, for example:
>
> 20.0 +/- 0.5% *OR* 20.0% +/- 0.5%
>
> 20.0 +/- 0.5 m *OR* 20.0 m +/- 0.5 m
>
> Thanks!
>
> Steve
--
Jay Warner
Principal Scientist
Warner Consulting, Inc.
4444 North Green Bay Road
Racine, WI 53404-1216
USA
Ph: (262) 634-9100
FAX: (262) 681-1133
email: [log in to unmask]
web: http://www.a2q.com
The A2Q Method (tm) -- What do you want to improve today?
|