Sample size calculation

#1
Hi everyone, I'm a bit of a newbie here.

I am trying to determine the appropriate sample size. I have a survey with several dozen questions, most of which are of the 4 item-response type. My total population is about 400, but I need to subdivide it further into sub-groups, ranging from n=40 to n=120.

My question is the following: if I want a confidence level of 95% and a margin of error of +-10% for my estimates, what formula do I use to calculate sample size? First of all, I CAN have a conf. level of .95 and a margin of error of 10%, right?
 
R

Renāta

Guest
#3
Hi Aferim,

To calculate your sample size you can use the following formula:

Ns = (Np) (p) (1-p) / (Np - 1) (B/C) (B/C) + (p) (1-p)

Where
Ns = required completed sample size
Np = population size
p = how varied the population is expected to be (50/50 split is tipically assumed, which correspondes to the value of 0.5)
B = acceptabale sampling error (in your case 10%, that is 0.1)
C = Z statistics in relation to confidence inteval, 1.96 corresponds to 95% level (see calculator here http://www.measuringusability.com/zcalcp.php)

Source: Dillman (2000) Mail and Internet Surveys: The Tailored Design Method, NY: John Whiley & Sonc, Inc. (pp. 206-207).

Hope this helps :)

Renata