Key factors in sampling size calculation

Population size

This is the total number of individuals or groups about which information is required. There are two types of populations: finite or infinite. For practical purposes, populations larger than 20,000 we can considered such as “infinite”. If we are dealing with a population smaller than 20,000 we try our best to estimate its size. To be on the safe side, we try to overestimating a little bit the population size.

Confidence level

This can be viewed as the degree of certainty that we have that the interval that we will estimate will contain the true value of the population (mean value or proportion). A 95% confidence interval is often explained like this: suppose we take samples from our population over and over again, and construct a confidence interval using certain procedure for each possible sample, we expect 95% of the resulting intervals to include the true value of the population parameter. 95% and 99% are the level most commonly used. The choice of confidence coefficient is somewhat arbitrary but it must depend on how secure we need to be that our interval estimation contains the true population parameter.

Higher confidence level requires a larger sample size.

Margin of error

A percentage that describes how closely the answer our sample gave is to the “true value” in our population. The margin of error is the amount of error that we can tolerate. The smaller the margin of error is, the closer we are to having the exact answer at a given confidence level. If 90% of respondents answer yes, while 10% answer no, we may be able to tolerate a larger amount of error than if the respondents are split 50-50 or 45-55.

Lower margin of error requires a larger sample size.