T-Test Calculator

How to use

1. 1Pick a test type: One-sample for one mean vs a hypothesized value, Two-sample (independent) for two separate groups, or Paired (matched) for repeated measures on the same subjects.
2. 2Choose how to enter data: Summary stats accepts n, mean, and SD; Raw data accepts numbers separated by spaces, commas, semicolons, or new lines (decimals must use a period).
3. 3Set the alternative hypothesis (two-tailed, left-tailed, or right-tailed) and the confidence level (90%, 95%, or 99%). For two-sample mode, leave Welch as the default unless you have a strong reason to assume equal variances.
4. 4Read the t statistic, degrees of freedom, p-value, decision verdicts at alpha 0.10, 0.05, and 0.01, the confidence interval, Cohen's d effect size, and the plain-English interpretation.
5. 5Use the t-distribution plot to see where the observed t statistic lands and which tail area is being measured. Use the example chips to load classic scenarios, or Copy summary or Copy full report to paste the results into homework, a spreadsheet, or a launch doc.

About this tool

T-Test Calculator runs the three standard Student's t-tests directly on raw data or on summary statistics. Pick One-sample to compare a single sample mean to a hypothesized population mean (quality vs spec, intake vs target, sensor reading vs reference), Two-sample (independent) to compare two groups where each subject sits in exactly one group (A/B test of completion time, class A vs class B test scores, treatment vs control with continuous outcome), or Paired (matched) to compare two measurements taken on the same subjects (before vs after a training program, reaction time pre and post intervention, weight loss in a clinical trial). For every mode you can enter either Summary stats (n, mean, standard deviation) or Raw data (numbers separated by spaces, commas, semicolons, or new lines, parsed and summarized automatically). The two-sample mode defaults to Welch's t-test, which does not assume the two groups have equal variances and computes the degrees of freedom with the Welch-Satterthwaite equation, a fractional value that is usually a few units less than n1 + n2 - 2. Tick the Assume equal variances box to switch to the classical pooled-variance form (Student's t-test with df = n1 + n2 - 2). The Alternative hypothesis selector lets you switch between two-tailed (mean differs from the null), left-tailed (mean is less than the null), and right-tailed (mean is greater than the null) tests, and the Confidence level selector returns a 90%, 95%, or 99% confidence interval for the mean or mean difference. For every test the page returns the t statistic, the degrees of freedom, the p-value, decision verdicts at alpha = 0.10, 0.05, and 0.01, the standard error, the confidence interval, the Cohen's d effect size with a qualitative small/medium/large label, a plain-English interpretation, a t-distribution plot with the observed statistic and the tail area shaded, and a per-mode breakdown showing every intermediate value used in the calculation. The numerics use the regularized incomplete beta function via the Lentz continued fraction, with the Lanczos log-gamma approximation underneath. This is accurate to roughly 1e-12 in the useful range, and the displayed precision is floored at 1e-12 so the reported numbers are never overstated. Useful for AP Statistics and college statistics homework, A/B test analysis when the outcome is a continuous measurement (time, revenue per user, score), survey research, clinical research, manufacturing quality control, sports analytics, and any time you need a defensible significance decision from two means or a mean against a target. Pair this tool with the P-Value Calculator when you already have a t statistic in hand, with the Chi-Square Test Calculator for categorical outcomes, with the Confidence Interval Calculator for interval estimates that do not need a hypothesis test, or with the Cohen's d Effect Size Calculator to dig deeper into how big the difference is once you have rejected the null. Everything runs in your browser, so the numbers you type for class, work, or research never leave your device.

Free to use. Works in your browser. No signup, no login.