The target population and the required level of accuracy must be determined before you can compute the sample size.

1. Population size

How many people in total are you referring to in your survey? You must know who belongs in your group and who does not to figure this out. If you want to know who owns dogs, for instance, you'll include anyone who has ever owned at least one dog. (Depending on the objectives of your research, you may consist of or omit people who have owned dogs in the past.) If you can't figure out the exact quantity, don't worry. Unknown or approximated ranges of numbers are frequent.

2. Margin of error (confidence interval)

The question is how much inaccuracy you'll tolerate because mistakes are unavoidable. Mean values describe the margin of error, also known as the confidence interval. You can decide how much variation between the means of your sample and population should be permitted. You've probably seen a confidence interval and how it's presented in a political survey featured on the news. It will read like this: "With a margin of error of +/- 5%, 68 percent of voters approved Proposition Z."

3. Confidence level

The confidence interval in step 2 with the same name differs from this one. It concerns your confidence that the actual mean will fall inside your margin of error. Ninety percent, ninety-five percent, and ninety-nine percent confidence intervals are the most typical.

4. Standard deviation

This step asks you to estimate how much the responses you receive will vary from each other and the mean number. A low standard deviation means all the values will be clustered around the mean number. In contrast, a high standard deviation means they are spread across a much more comprehensive range with tiny and huge outlying figures. Since you haven’t yet run your survey, a safe choice is a standard deviation of which will help make sure your sample size is large enough.