0
$\begingroup$

I am currently using a formula I got from a Qualtrics blog post to calculate the sample I require for a survey for the mean to be meaningful/reflect the population mean, as follows:

http://imgur.com/a/YKzBH

My survey has a large Std.dev (1.8), which I use instead of assuming a 0.5 Std.dev. This formula does not seem to take into consideration the population size (>10 000), however.

Is there an alternate formula to calculate required sample size with a known population size (but an unknown population std.dev)?

$\endgroup$
2
$\begingroup$

Population size is irrelevant unless the sample is a large portion of the population, which is clearly not the case here. There is a finite population correction: $FPC = \sqrt{\frac{N-n}{N-1}}$

where N is population size and n is sample size, but this is so close to 1 in your case that it can be ignored.

$\endgroup$
  • $\begingroup$ Out of interest and possible future application, what would one then do with the FPC value (which in my case is 0,98)? $\endgroup$ – The_Dude Jul 24 '17 at 12:39
  • $\begingroup$ One would multiply the sample size by its inverse. $\endgroup$ – Peter Flom - Reinstate Monica Jul 24 '17 at 12:49
  • $\begingroup$ So n * (1/FPC)? $\endgroup$ – The_Dude Jul 24 '17 at 13:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.