A confidence interval gives a range of plausible values for a population parameter. Choose what you are estimating: a mean, a difference between two means, a paired difference, two proportions or a variance. Then enter raw data or summary statistics.
How to interpret a 95% confidence interval
If you repeated the study many times, about 95% of the intervals would contain the true value. It is not a 95% probability that this particular interval contains it. Narrower intervals mean more precise estimates; collect more data to narrow them.
Which interval?
- Mean: one group’s average (t interval; z if σ is known).
- Difference in means: two independent groups (Welch by default).
- Paired difference: before-and-after data.
- Two proportions: difference in rates (Newcombe method).
- Variance / SD: precision or consistency (chi-square interval).
For a single proportion, use the proportion CI calculator, which offers Wilson and exact intervals.