.svg){kind=small}
Free Sample Size Calculator
Find out how many people you need to survey for a result you can trust. Set your confidence level, margin of error, population size, and response distribution, and get the required sample size the moment you change a value. It uses the standard statistical formula, and every calculation runs locally in your browser, so nothing you enter is ever sent to a server. Free, no sign-up.
Your parameters
How sure you want to be that the true answer falls inside your margin of error. 95% is the common default.
How much error you will accept, in percentage points. A smaller margin means a larger sample.
The total number of people you are studying. Leave it blank for a large or unknown population.
Your best guess at how a yes/no answer will split. Leave it at 50% when you are unsure: that gives the most conservative, largest sample.
To be 95% confident within ±5%, survey about 385 people from a large or unknown population.
What these mean
Confidence level
How often the true value would land inside your margin of error if you repeated the survey. At 95%, you would be right 19 times out of 20.
Margin of error
The band around your result, in percentage points. A 5% margin on a 60% finding means the real figure sits somewhere between 55% and 65%.
Response distribution
Your expected split on a yes/no question. 50% is the safest assumption because it requires the most responses, so you never undersample.
This is a statistical estimate for a simple random sample. Real accuracy also depends on response quality, sampling method, and how representative your respondents are.
A sample size calculator tells you the minimum number of completed responses you need for a survey result to be statistically reliable, based on your confidence level, margin of error, population size, and expected response distribution. This free tool applies the standard formula for a proportion and shows the number instantly, with no sign-up, and it runs entirely in your browser so nothing you enter leaves your device.
How the sample size calculator works
01
Set your parameters
Pick your confidence level and margin of error, add your population size if it is small, and set the response distribution (leave it at 50% if you are unsure).
02
Read the sample size
The tool applies the standard sample-size formula for a proportion and shows the number of responses you need, updating the moment you change any value.
03
Copy and plan
Copy the summary and use the number to plan how many people to invite. Everything is calculated locally in your browser and never sent to any server.
Free, private, and built for better research
Most sample-size calculators want your email first, and many run the numbers on someone else's server. This one does neither. There is no AI model and no API call behind it: the formula is fixed, documented, and the arithmetic happens on your machine.
01
100% in your browser
Nothing you enter is sent to a server. The calculation is plain arithmetic that runs on your device, so your research plans stay private.
02
No account required
No sign-up, no email wall, no record of what you calculated.
03
Arithmetic you can check
The same inputs always produce the same number. You can read the formula below and verify it yourself.
The formula this tool uses
The standard sample-size formula for estimating a proportion, so you can sanity-check any result:
- Z is the z-score for your confidence level: 1.65 for 90%, 1.96 for 95%, 2.58 for 99%.
- p is the response distribution as a proportion, and e is the margin of error as a proportion.
- The large-population sample size is n0 = (Z squared times p times (1 minus p)) divided by e squared.
- If you enter a population size N, the finite population correction gives n = n0 divided by (1 plus (n0 minus 1) divided by N).
- The result is rounded up to a whole number of people, because you cannot survey a fraction of a person.
Where this formula comes from
It is the textbook method for sampling a proportion, not something invented for this tool:
- Sample size determination: the standard formula for estimating a proportion and the finite population correction.
- Confidence intervals: the basis for the z-scores used at each confidence level.
Built for every team that runs surveys
Anyone who has argued about whether a result is "enough people" knows the answer is a number, not an opinion. This tool gives you that number before you field the survey. It pairs well with an adaptive AI survey platform when the answers you collect are sensitive.
01
Researchers
Size your sample before fieldwork and state it in the method section with confidence.
02
Product teams
Know how many users you need before a feature survey is worth trusting.
03
Marketing & CX
Back a brand or satisfaction study with a defensible sample rather than a round number.
04
HR & people teams
Work out how many employees make an engagement result representative.
05
Academics & students
Calculate the sample your ethics submission or dissertation asks you to justify.
06
Agencies
Show a client exactly what a tighter margin of error costs in respondents.
Know your sample size? Now collect it privately.
BlockSurvey is an AI survey platform, encrypted end to end, so the responses you gather are never sold, mined, or used to train models.
More free AI tools
Explore more →
These tools help you govern and document your AI use. When the data behind your AI comes from people, collect it with an AI survey platform built for privacy.
Frequently Asked Questions
- It tells you how many completed responses you need for a survey result to be statistically reliable. You choose how confident you want to be (the confidence level), how much error you are willing to accept (the margin of error), and, optionally, how large the group you are studying is (the population size). The tool returns the minimum number of people you need to survey. It is a deterministic statistical calculation, not an AI model, so the same inputs always produce the same number.