Toolyard

T-Test Calculator

How to run a t test

  1. Choose a two-sample, paired or one-sample test.
  2. Paste the numbers for each group, separated by spaces, commas or line breaks.
  3. The t statistic appears with the degrees of freedom, the P value and the difference between the means.

Frequently asked questions

Which test should I use?

Two samples for two separate groups, paired for two measurements of the same subjects (before and after), and one sample to compare one group with a fixed value.

What does the P value mean?

The probability of seeing a difference this large if the groups really had the same mean. Below 0.05 is the usual threshold for calling a result significant, but it does not say the difference is large or important.

Should I assume equal variances?

Welch's test, which does not assume them, is the safer default and is what most statistics software now uses. Choose the equal-variance version only when you have a reason to.

What does the test assume?

That the values are independent and roughly normally distributed. For small, clearly skewed samples, a non-parametric test such as Mann-Whitney is a better fit.

Are my numbers sent anywhere?

No. Everything is worked out in your browser and nothing is stored or uploaded.

More calculators

Guides