Statistics & Probability

Paired Difference t Test & Confidence Interval Calculator

Compare matched observations by calculating a t test and confidence interval for their within-pair differences.

How the calculation works

Compute paired differences, their sample mean and sample standard deviation. Standard error = SD/√n; compare (mean difference − null difference)/SE with a t distribution on n − 1 degrees of freedom. Confidence interval = mean ± t critical × SE.

Worked example

Example inputs — First measurement in each pair: 1, 2, 3; Matching second measurement: 2, 2, 2; Null mean difference first − second: 0; Two-sided confidence level (%): 95. Expected results — Number of pairs: 3 pairs; Mean paired difference: 0; Sample standard deviation of differences: 1; Standard error of mean difference: 0.5773502692; Degrees of freedom: 2; t statistic for entered null: 0; Two-sided p-value: 1; Confidence interval lower endpoint: -2.484137712; Confidence interval upper endpoint: 2.484137712.

Assumptions and limits

Pair observations by participant or matched unit in corresponding order; each pair contributes first minus second. Pairs must be independent and the distribution of differences should be approximately normal or the sample sufficiently large for the approximation. This is not an independent-sample Welch test; identical differences have zero standard error and are rejected. No multiplicity correction or causal conclusion is supplied.

Reference and further reading

Read the underlying subject reference. Calculith is an independent implementation; the reference does not endorse or certify this tool.

Related calculators

Browse all 1000 calculators →

Example values are editable. Results are estimates where assumptions apply. Calculations run in your browser. Updated 2026-11-16.