1 Answer | Add Yours
This is a good question and the same principles would apply; it is just that there is is more room for error due to the smaller population. Bear in mind that no sampling is perfect.
First, you would want a random sampling. This reason for this is that you would not want to skew the data in any possible way. For example, if you only sample older people who live in a certain area, then the data would only reflect older people in that area. Random is the key. Keep in mind that it is possible to skew data unwittingly. This will be your sample group.
Second, you want to get enough people to make the test significant. This number is not easy to determine. All samples differ in size, but statistics state that the larger the number the more accurate. Here is one caveat with a smaller population. The smaller the population, the more people you need to survey. There is an inverse relationship. So, if you are dealing with a rare population, you might consider reaching out to them all.
We’ve answered 319,199 questions. We can answer yours, too.Ask a question