This is not a standard form of expression.  Technical language should be
standardized.  Sometimes variations are introduced to make matters clear but
only end up making them worse.  This is one.  For example, a standard error
is a single. positive value, not +/-.  A confidence interval is, often but
not always, an estimate plus or minus a confidence coefficient times the
standard error.  These words and symbols do not express this clearly.

Unclear language not only fails to communicate, it confuses the issue.  This
is basic and has nothing to do with statistics or any other form of
technical communication. Shakespeare knew this.  Writers of package inserts
for prescription drugs know it.  The rest of us can do no less.

Why do writers use non-standard language?  Often to make things clearer to
non-technical readers and users.  Once in a great while it works.  More
usually it doesn't.  This is a good example.

Regards,
David Smith

David W. Smith, Ph.D., M.P.H.

(518) 439-6421

45 The Crosway
Delmar, NY 12054

[EMAIL PROTECTED]
----- Original Message -----
From: "Doug Hoy" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, March 05, 2002 1:03 PM
Subject: Confidence limits--standard error reporting?


> I am not a statistician, nor do I play one on TV, but I
>
> I am considering hiring a research analyst who has submitted some examples
> of work performed. It involves reporting on market survey data.
>
> The work seems mostly OK, but a few oddities make me wonder...
>
> For example, when reporting on the statistic accuracy of a sample of 500,
a
> 'confidence' of 95% is given with a 'Std. Error' +/- 2.0%.
>
> I'm not used to seeing a Standard Error given in this context. Usually
it's
> a margin of error, which a non-statistically-inclined client can
> intuitively apply to their data for confidence limits (which at 95%
> confidence level, would be larger than +/- 2%). The impression given is
> that one can trust the various results to be within 2% of the population
> value.
>
> Am I misunderstanding something here?
>
> Later, the sample is split into two sub-groups of n= 350 and 150
> respectively. Results for a particular question are 56% and 68%. It seems
> to me that the sample is not big enough to declare these subgroups
> different, at 95%. No?
>
> Doug H
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> ................... http://jse.stat.ncsu.edu/ ...................
> =================================================================

=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
................... http://jse.stat.ncsu.edu/ ...................
=================================================================

Reply via email to