Sample Size Calculator

This sample size calculator tells you how many people you need to survey to estimate a proportion to a chosen accuracy. Whenever you survey a sample rather than the whole population, your results carry a margin of error, and the sample size controls how tight that margin is. You set the confidence level you want, usually 95 percent, and the margin of error you can live with, often plus or minus 5 percent, and the calculator returns the number of completed responses required. It also lets you enter the expected proportion, the share you think will answer a certain way; using 50 percent is the cautious default because it produces the largest, safest sample, so when in doubt leave it at 50. If you are surveying a small, known group, such as the staff of one company or the members of a club, you can enter the total population size to apply a finite population correction, which lowers the number of responses you actually need. The base formula multiplies the square of the critical z-value by the proportion times one minus the proportion, then divides by the square of the margin of error. The required sample grows quickly as the margin shrinks, so halving the margin roughly quadruples the sample. Enter your confidence level and margin to see how many responses you need.

Leave population size blank for a very large or unknown population. Uses p of 0.5 for the safest size when unsure. Estimate only.