About the Margin of Error Calculator
This free browser tool computes the margin of error and the corresponding confidence interval for a survey proportion or a sample mean. You pick the result type, enter sample size and either the observed percentage or the sample standard deviation, choose a confidence level, and optionally add a population size to apply the finite population correction. The page shows the margin of error, the lower and upper bounds of the true value, and the z value used (1.96 for 95% by default). Results update instantly as you type; nothing is uploaded. You can copy the result or create a share link that reopens with your inputs. The calculator uses the normal approximation formulas; it covers random sampling error only and does not correct for bias, weighting, or complex survey design.
How to use it
- 1Choose result type: A percentage (survey, poll) or An average (mean).
- 2Enter the sample size n and either the percentage in your result (%) or the sample standard deviation (SD).
- 3Set the confidence level (%) and, if known, enter the population size to apply finite population correction.
- 4Copy the margin of error, the true-value range and the z value, or share a link that reopens with your numbers.
Worked example
Example: result type = Percentage; sample size n = 1000; percentage = 50%; confidence = 95%; population left blank. Output: Margin of error = 3.10 percentage points; Range = 46.90% to 53.10%; z = 1.96.
The formula
Percentage: MOE = z × sqrt( p(1 - p) / n )
Mean: MOE = z × SD / sqrt(n)
With known population N multiply the standard error by sqrt((N - n) / (N - 1)).
z = 1.96 for 95% confidenceResults shown in percentage points for proportionsSource: OpenStax, Confidence interval for a population proportionPew Research Center, survey methodology referencesIntro Stats (finite population correction explanation). Last checked 6 November 2026.
What it is good for
- Instant results update as you type with no calculate button.
- Supports proportions and sample means plus finite population correction.
- Copy or share a link that reopens with your inputs for easy reporting.
- Covers random sampling error only; it does not correct for bias, weighting, or complex survey designs.
Questions people ask
What does the margin of error mean?
The margin of error is the half width of a confidence interval. It expresses sampling uncertainty: how far the sample estimate may be from the true population value under simple random sampling.
When should I use the finite population correction?
Use the finite population correction when your sample is a non-negligible fraction of the population, for example n is more than a few percent of N. Enter N to apply the correction automatically.
Which z value do you use for 95%?
This calculator uses z = 1.96 for a 95% confidence level. If you pick a different confidence level the tool uses the matching z critical value for the normal approximation.
How do I adjust for survey weighting or complex design?
This tool does not apply design effects or weighting. To approximate the effect, multiply the simple random sample standard error by sqrt(DEFF) or use an effective sample size n_eff = n / DEFF and enter that n in the calculator.
Is the normal approximation always accurate?
The normal approximation works well for moderate or large n and p not extremely close to 0% or 100%. For small samples or very small or large proportions, consider alternative intervals (Wilson, Agresti‑Coull, or exact) which this tool does not compute.
Can I calculate required sample size for a desired MOE?
This tool shows the margin of error from your inputs. To get a required sample size, rearrange the proportion formula: n = z^2 × p(1−p) / MOE^2, using a planned p. You can then enter that n here to confirm the MOE.
Help keep every tool free
FreeToolr stays free because people help it grow. Share a tool you like, tell us what to build next, or support us on Ko-fi.