About the T-Test Calculator
A t-test asks whether the average of one group differs from another, or from a known value, by more than chance would explain. Paste your numbers, choose the type of test and the significance level, and this calculator returns the t statistic, the degrees of freedom, the two-sided p-value, the confidence interval for the difference and Cohen's d as a measure of effect size. It supports Welch's test for two independent groups, Student's test when the spread is equal, the paired test for before and after measurements and a one-sample test. Your data stay in your browser.
Updated October 2026
How to use the T-Test Calculator
- 1Choose the kind of test that fits your data.
- 2Paste the numbers for group A, separated by spaces, commas or lines.
- 3Paste group B, or the value to compare with for a one-sample test.
- 4Pick the significance level, usually 0.05.
- 5Read the p-value, the interval and the effect size.
Picking the right test
Two groups of different people, such as customers who saw a new page against those who saw the old one, call for an independent test. Welch's version does not assume the groups have the same variance, so it is the default here. Student's version is slightly more powerful when the variances really are equal.
When the same people are measured twice, such as weight before and after a diet, the paired test is stronger because each person acts as their own control. The one-sample test fits a question like whether a machine fills bottles to an average of 500 ml.
Reading the results
- t statistic: the difference in means divided by its standard error.
- p-value: below your level means the difference is unlikely to be chance.
- Interval: if it excludes zero, the test is significant at that level.
- Cohen's d: tells you whether a significant difference is also a meaningful one.
- Always look at the means as well as the p-value.
Related tools
The Standard Deviation Calculator shows the spread behind the test, the Confidence Interval Calculator covers a single mean and the P-Value Calculator converts scores into probabilities. If your data are counts in categories rather than measurements, switch to the Chi-Square Calculator. Statistics are a guide to evidence, not proof.
Features
- Welch, Student, paired and one-sample tests
- Exact p-value from the t distribution
- Confidence interval for the difference
- Cohen's d effect size
- Means and standard deviations shown
- Helpful errors for lists that do not match
- Private, runs in your browser
Frequently asked questions
Which t-test should I use?
Use Welch's test for two separate groups unless you are sure the spread is equal. Use the paired test when the same subjects are measured twice, and the one-sample test when you compare one group with a known value.
What does the p-value tell me?
It is the chance of seeing a difference at least this large if the true averages were equal. A p-value under your significance level, often 0.05, counts as a significant difference.
What are degrees of freedom?
They describe how much independent information the test has. For a simple test they are the number of values minus one, and Welch's test uses a more exact fractional number.
How many numbers do I need?
At least two per group, but more is far better. Small samples give wide intervals and low power.
What is Cohen's d?
It is the difference between the means in units of the standard deviation. About 0.2 is small, 0.5 is medium and 0.8 is large by common convention.
Why must paired lists be the same length?
Each value in A is matched with the value in B for the same subject. The test works on the differences between the pairs.
Does the data have to be normal?
The t-test is fairly robust for moderate samples, but with very small or very skewed data a non-parametric test may be wiser.