Selecting a sample from a small population

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Selecting a sample from a small population
Message posted by Kelly on December 28, 1999 at 12:00 AM (ET)

Hi,

I'm an Internal Auditor trying
to determine how large my attribute
sample needs to be if I'm looking
at a population of about 200.
I need a tolerable
exception rate of 0%, an
acceptable risk of overreliance
of 10% and I expect a population
error rate of 0%. Isn't there
some type of formula to determine
sample size within these parameters?

Thanks!

Kelly

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Selecting a sample from a small population
Message posted by Phil Rosenkrantz on January 2, 2000 at 12:00 AM (ET)

Could you explain a little more about what those three error rates are? Can't say I am familiar with them.

If these are equivalent in any way to Type I and Type II error rates and "AQL", then you could have an "Acceptance Sampling" situation. I would suggest using Dodge-Romig tables. You can find the procedure in any thorough Statistical Quality Control book.

The pure formulas here are rather complicated. Use tables if you can.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.