In response to a question about an audit where the pop. N=20,,
On Thu, 20 Jan 2000 03:07:40 GMT, [EMAIL PROTECTED] wrote:
< snip >
> population. If the sample size is a significant fraction
> of the entire population (as it will be in this instance,
> with only 20 people in the population), then a correction
> is needed to the usual formula for determining the correct
> sample size.
< snip; ... write for details >
> ... When all is said and done, with a population
> of only 20, the sample will need to be a large fraction of
> the population. Perhaps as many as 10 or 12.
> Charlie H.
RANDOM SAMPLING -- Will the auditors believe it? If the audit
requires that 75% (which would be 15 of 20) are informed about XYZ,
will the auditors believe it if you say, "All 14 of 14 that we asked
were informed, so we are sure that your standard of 15/20 would be
met" ? Maybe. Or will you be infinitely more convincing if you
actually get the 15th person?
Technically speaking, I can imagine that an AUDIT fulfills the
arbitrary, administrative-type need that can make it conceivable to
use a Finite-Population-Correction Factor. But it is really a rarity,
I think, and a bureaucratic mistake, that you would have a requirement
that would be written out, and that would expect you to invent the FPC
yourself. I *could* imagine that a regulation would point you to a
chart that you could be required to use, where the chart incorporated
the FPC.
I guess that I am saying: I doubt the usefulness of the advice or of
the proferred paper.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
