Sample Size Calculator

Once a population exceeds a certain size (typically around 20,000), the sample size required to achieve a specific margin of error changes very little. This is because the variance of a sample mean depends on the sample size itself, not the proportion of the population sampled. Whether you are sampling a city of 100,000 or a country of 300 million, the target sample size remains nearly identical.

A population proportion of 50% (p = 0.5) represents the maximum possible variance in a binomial distribution. Defaulting to 50% ensures that your calculated sample size is large enough to achieve your desired margin of error regardless of the actual distribution of opinions or traits in the population. It represents the 'worst-case' or most conservative sample size scenario.

Reducing the margin of error increases the required sample size exponentially. Because the margin of error is in the denominator of the sample size equation and squared, halving your margin of error (e.g., from 6% to 3%) will require four times as many survey responses.

Confidence level represents how sure you want to be that the sample reflects the population (e.g., 95% confident that the true answer lies within the range). The margin of error is the size of the range itself (e.g., +/- 5%). Together, they describe the accuracy and reliability of your poll.

Use the Finite Population Correction (FPC) when your sample is expected to be a significant portion of the total population (typically if you are sampling more than 5% of a small, finite group). The correction reduces the necessary sample size because sampling a large fraction of the population decreases estimation uncertainty.