How to use Confidence Interval Calculator
- Choose Student t for a mean with sample SD, z for a mean with known population SD, or Wilson score for a binary proportion.
- For a mean, enter mean, standard deviation and sample size. Under Student t you may instead supply a raw list; it replaces those three summary fields. Clear the list before choosing another method.
- For Wilson, enter the integer success count and total sample size. Mean and standard deviation are not used.
- Set confidence level and display precision. Review the estimate, interval center, margin and bounds alongside the method’s assumptions.
Example: Confidence Interval Calculator
Estimate a mean interval from 25 observations with sample mean 10 and sample standard deviation 2.
Options
- Deviation source
- Use Student t when the mean calculation uses a sample standard deviation. Select z only for a known population deviation under that method’s assumptions.
- Wilson proportion
- Use the observed success count and total sample count. They describe the entered sample rather than a forecast of future outcomes.
Supported inputs and limits
- Confidence level is 50.0001% to 99.999%.
- Summary sample size is at most 1,000,000;
- Student t requires at least two observations.
- The optional raw list supports Student t only and accepts up to 10,000 numbers or 200,000 characters.
- Mean intervals assume independent observations and an appropriate population distribution or sample size.
More details
Wilson models independent binary trials.
There is no weighting, clustering or finite-population correction.
Confidence concerns the repeated-sampling method, not the probability that this fixed interval contains the parameter.
Calculations use finite floating-point arithmetic.
A constant raw sample gives a zero-width computed interval, which does not establish zero population uncertainty.
Display rounding uses 4, 6, 10 or 12 significant digits and does not change the calculation.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
Confidence belongs to the repeated-sampling method
A 95% procedure is intended to cover the fixed parameter in about 95% of repeated samples under its assumptions. It does not assign a 95% probability to the parameter being inside this already computed interval. Nonrepresentative sampling and measurement bias are not repaired by a narrow interval.
Questions about Confidence Interval Calculator
When should I use the t method?
Use the estimated-deviation method when the population deviation is unknown and your sample provides its estimate, subject to the stated assumptions.
Does 95% confidence assign a probability to these fixed bounds?
No. It describes how the interval method performs over repeated suitable samples.
Why use Wilson bounds for proportions?
They remain within zero and one and account for the sample size, including all-success or no-success samples.