Calculator Uni

Sample Size Calculator

Find the survey sample size for your margin of error.

Survey design
±%

Sample size needed

370

Infinite-population size

385

Share of population

3.7%

Assumes maximum variance (p = 0.5), the conservative standard for surveys.

Precision is expensive

Halving the margin of error quadruples the required sample — the formula divides by e². A ±5% national poll needs ~384 responses; ±2.5% needs ~1,537. Population size barely matters once it's large, which is why 1,000 people can represent a country.

Sampling beats volume

The math assumes random sampling; a million self-selected online votes carry more bias than 400 random ones. Budget survey effort for response quality and randomness first — the sample-size formula can't rescue a skewed frame.

Related calculators