About the Confidence Interval Calculator
This tool computes confidence intervals for a single mean or a single proportion right in your browser. You enter whether the interval is for a mean or a proportion, the sample mean, standard deviation, whether the SD is a sample or the known population SD, successes, sample size n, and the confidence level. Results update instantly as you type and include the interval, margin of error, point estimate, standard error, critical value, and a Wilson interval for proportions. The page runs entirely in your browser; you do not upload data to a server. You can copy results or share a link that reopens with your numbers. The calculator automatically uses t when the SD is from the sample and z when the population SD is known, and it shows the Wilson score interval because that behaves better for small n or p near 0 or 1.
How to use it
- 1Choose Interval for: A mean or A proportion (percentage).
- 2Enter Sample mean, Standard deviation, Successes, Sample size n, and Confidence level.
- 3Select whether SD is from the sample (use t) or the population (known, use z).
- 4Watch results update instantly and copy or share the link with your inputs.
Worked example
Input: Interval for a proportion, Successes = 8, Sample size n = 20, Confidence level = 95%. Outputs: Point estimate p̂ = 0.4000; Standard error (Wald) = 0.1095; Critical value z = 1.96; Wald interval = [0.1854, 0.6146]; Wilson interval = [0.2188, 0.6135]; Margin of error (Wald) = 0.2146.
The formula
Mean: estimate +/- critical value x SD / sqrt(n)With the sample SD the critical value comes from the t distribution with n - 1 degrees of freedom; with a known population SD it comes from the normal (z) distribution, 1.96 for 95%Proportion: p = successes / n, interval p +/- z sqrt(p(1 - p) / n) (the Wald interval taught in most courses)The Wilson interval is also shown because it behaves better when n is small or p is near 0 or 1Source: NIST e-Handbook of Statistical Methods - Proportion CIsStatCompute - Confidence Interval Formulas and Advice. Last checked 6 November 2026.
What it is good for
- Instant results as you type with no calculate button.
- Auto selects t or z for mean intervals and shows Wilson for proportions.
- Works fully in your browser; you do not upload data.
- Assumes a random sample. For a mean with small n, data should be roughly normal.
- This tool handles single-sample mean or proportion intervals only, not two-sample or paired intervals.
Questions people ask
When does the calculator use t versus z for a mean?
The calculator uses t when you mark the standard deviation as from the sample. If you specify a known population SD it uses the z distribution. For 95% z = 1.96.
Which interval does the tool show for proportions?
It shows the standard Wald interval p ± z·sqrt(p(1-p)/n) and the Wilson score interval. Wilson usually performs better for small n or p near 0 or 1.
Can I save or share my inputs and results?
Yes. You can copy results and generate a shareable link that reopens the calculator with the same numbers. All work runs in your browser.
What assumptions underlie the results?
You should have a random sample. For mean intervals with small n the measured data should be roughly normal. Proportion intervals assume binomial sampling.
Are the results exact for very small samples or zero successes?
For proportions the tool reports Wilson and the Wald interval. Wilson behaves better with small samples or zero successes. For guaranteed exact coverage use Clopper, Pearson outside this tool.
How should I pick the confidence level?
Higher confidence levels give wider intervals. 95% is common. Pick the level that matches your reporting or regulatory needs and state it in your report.
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.