Free sample size calculator

How many people do you actually need in your survey or study? Enter your confidence level, margin of error, and population — get the answer instantly. Free, no signup.

Open the free sample-size calculator →

What it calculates

Using the standard formula (z²·p·(1−p) / E²) with a finite-population correction, the calculator gives you the number of responses you need for statistically meaningful results based on:

Confidence level — how sure you want to be (90%, 95%, 99%).
Margin of error — how much wiggle room you'll accept (e.g., ±5%).
Population size — your total audience (applies the correction for small populations).
Expected proportion — defaults to 50% (the most conservative, largest sample) when unknown.

Why sample size matters

Recruit too few and your findings are noise you can't defend. Recruit too many and you burn budget and time. The right sample size is the cheapest way to make research credible — especially for surveys, tree tests, and any quantitative study where you'll report percentages.

How to use it

Pick your confidence level (95% is standard).
Set your margin of error (±5% is common; tighten to ±3% for high-stakes decisions).
Enter your population if it's finite (e.g., 1,200 customers).
Read the required sample size — that's your recruitment target.

Then run your study in ResearchRocket — surveys, card sorting, tree testing, and usability — and analyze the results with real statistics, including non-parametric tests for ordinal survey data.

FAQ

How do I calculate sample size? Use z²·p·(1−p)/E²: z from your confidence level, p the expected proportion (0.5 if unknown), E your margin of error — then apply a finite-population correction for small audiences. The free calculator does it for you.

What sample size do I need for a survey? For a large population at 95% confidence and ±5% margin, roughly 385 responses. Smaller populations need fewer (the calculator applies the correction).

Is the calculator really free? Yes — it's a free public tool, no signup or credit card.

What confidence level should I use? 95% is the standard default. Use 99% for high-stakes decisions (it needs a larger sample); 90% is acceptable for quick directional reads.

Open the free sample-size calculator →