I'm looking for a way to measure the error of a market research. Lets say we have an universe of N and we want to survey a sample n, ¿how can I determine the error?
I'm trying to, using the Central Limit Theorem, demostrate the "**Murray & Larry" solution**, i.e.
n = ( N*(Z^2)p(1-p) ) / ( (N-1)*(e^2)+(Z^2)p(1-p) )
where: e = error Z = normal distribution based on the confidence level p = homogenous of the variables that determines our sample?
Thank you in advance.