How to use the Confidence Interval Calculator
- Choose a mean or a proportion.
- Enter your sample results and pick a confidence level.
- Read the interval and margin of error.
What does this tool do?
A 95% confidence interval is built so that, if you repeated the study many times, about 95% of intervals would contain the true population value. Wider intervals mean more uncertainty.
For means, the calculator uses the t-distribution when the standard deviation comes from your sample — the usual case — and the normal distribution only when the population σ is known. For proportions it shows the recommended Wilson interval alongside the simple Wald interval.
Worked example: a mean
A sample of 36 has mean 50 and standard deviation 8, calculated from the sample. The standard error is 8 ÷ √36 = 1.333. With 35 degrees of freedom the critical t value at 95% is 2.0301, so the margin of error is 2.0301 × 1.333 = 2.707, and the 95% confidence interval is 47.29 to 52.71. At 99% the critical t rises to 2.7238, and the interval widens to 46.37 to 53.63: more confidence costs precision.
Tick “This is the population σ” only if the standard deviation comes from outside your sample, which is rare. The calculator then uses the normal distribution instead of t.
Worked example: a proportion, and why Wilson
120 of 400 people say yes: p̂ = 30%. The 95% Wilson interval is 25.72% to 34.66%, and the simple Wald interval (p̂ ± 1.96 × standard error) is 25.51% to 34.49%. With a large sample the two agree. With a small one they don't: 1 yes out of 20 gives a Wald interval of 0% to 14.55% (clipped at zero, because the formula would go negative), while Wilson gives 0.89% to 23.61%. That's why the calculator leads with Wilson and shows Wald for comparison.
Related: Sample Size Calculator · Standard Deviation Calculator
Why use it?
- Uses t automatically for sample standard deviations.
- Wilson interval for proportions.
- Shows the critical value and standard error.
Common mistakes
- Saying “there's a 95% chance the true value is in this interval”. Strictly, 95% of intervals built this way would contain it; this particular one either does or doesn't.
- Applying the interval to a convenience sample (whoever answered). The maths assumes a random sample.
- Using the population σ option with a standard deviation calculated from the same sample.
The formula
Accuracy and limits
- Results are shown to several decimal places so you can check them against textbook tables; round them sensibly when you report them.
- Confidence levels available: 80%, 90%, 95%, 98%, 99% and 99.9%.
Privacy
The calculation happens instantly in your browser. The numbers you enter are not sent to our servers or saved. There's no account to create and nothing to install.
Frequently asked questions
Should I use 95% or 99%?
95% is the common default. 99% gives a wider, more cautious interval; 90% a narrower one.
Content last updated . See how M2Toolkit works.