Hi everyone,
I have been asked to analyse a university survey which has race information
in it. They want me to identify areas of significant variation in
satisfaction levels according to ethnicity.
I have the total numbers of students by race and from there am able to
calculate the minimum sample needed for a given error and confidence level
with the formulas shown below.
The problem is that those populations are small and the sample size needed
is always bigger than the samples from my survey with a 5% error and
confidence of 95%.
Can I conclude from there that is not possible to obtain any significant
result from the survey for any of the ethnicities?
Are there any theories or models for sampling small populations (<1000)?
Thanks for your help
Luis Martinez
Sample Size Formulas
Here are the formulas used in our Sample Size Calculator:
Sample Size
ss = [Z*Z * (p) * (1-p)]/c2
where:
Z = Z value (e.g. 1.96 for 95% confidence level)
p = percentage picking a choice, expressed as decimal
(.5 used for sample size needed)
c = confidence interval, expressed as decimal
(e.g., .04 = ±4)
Correction for Finite Population
new ss = ss/[1+ (ss-1/pop)]
where:
pop = population
|