Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
finding the n
Message posted by MaryLee on October 29, 1999 at 12:00 AM (ET)
I have a population of 300 for a study. I need to find out what is a good sample n to have sufficient power to have a good study. I thought a confidence interval of 95 would be good.
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: finding the n
Message posted by jg on October 30, 1999 at 12:00 AM (ET)
You need to give more information. Are you estimation an average or a proportion ? What is your best guess about the populatioon standard deviation ? You say that you wish to have an alpha of .o5 . What beta are you looking for and what is the alternative hypothesis for this beta ?
Re: finding the n
Message posted by Mary Ann Wiedl on November 2, 1999 at 12:00 AM (ET)
The simplest formula for margin of error is 1 divided by the square root of n, with "n" being the number of people surveyed. It does not matter how many people in the group being sampled, of n equals 400, you'd have 1 divided by 20 or 5% margin of error. You see the problem here? You're universal set is only 300. Why not just send a survey to all of them?
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.