Confidence Interval Calculator

Mean ± margin of error from sample mean, s, n, and z. Click any i for detail.

Jump to result

Sample

Sample mean
i
Mean

x̄ — the sample average.

Sample std. deviation s
i
s

Sample standard deviation (use σ if your course says so).

Sample size n
i
n

Number of observations.

z critical value
i
z

Common: 1.645 (90%), 1.96 (95%), 2.576 (99%).

How to use this calculator

  1. Enter mean, s, and n.
  2. Pick z for your confidence level.
  3. Read the interval and margin of error.
  4. Use t critical values instead when your course requires a t-interval.

Results explained

For large samples or known σ, a z-interval for the mean is x̄ ± z·(s/√n). The term s/√n is the standard error. Smaller samples often replace z with a t critical value from n−1 degrees of freedom — this page leaves z in your hands so you can paste either.

Quick reference: z intervals

x̄ ± z · s / √n.

ItemDetail
90%z ≈ 1.645
95%z ≈ 1.96
99%z ≈ 2.576
SEs / √n

Not a substitute for checking normality / sampling design assumptions.

How the estimate is built

CI = mean ± z × (s / √n).

Example scenario

mean 100, s 15, n 36, z 1.96 → SE 2.5 → interval about 95.1 to 104.9.

FAQ

Should I use t instead of z?

Many intro stats courses switch to t when σ is unknown and n is small. Enter the t critical value in the z box if you want.

Is this for a proportion?

No. Proportions use √(p̂(1−p̂)/n). This page is for a mean.

What does 95% confidence mean?

In repeated sampling, about 95% of such intervals would cover the true mean — not a 95% probability for one fixed interval in Bayesian language.

Can s be zero?

Then the interval collapses to a point at the mean (unusual in real data).