A farmer tries a new fertilizer on one field and keeps the old one on another; harvest comes in 8% higher on the new field — did the fertilizer work, or did the weather just cooperate? The two-sample t-test settles it. It compares the averages of two independent groups and asks whether their gap is bigger than random variation can explain. Drug trials (new pill vs. placebo), website tests (red button vs. blue button), and manufacturing comparisons (Supplier A vs. Supplier B) all run on this test.

The mechanics extend the one-sample idea: compute each group's mean and spread, combine the spreads into a standard error for the difference, and form a t-statistic from the gap between means. A large t (small p-value) says the groups genuinely differ; a small t says the gap could be noise. The test assumes roughly normal data and independent samples — paired measurements (before/after on the same subjects) need the paired version instead.

This calculator handles the core paired computation. Enter the two groups' summary values in the Variable A and Variable B boxes, and the Result box combines them — the first step toward the t-statistic comparing the groups.

With the pre-filled defaults — Variable A 10, Variable B 5 — the Two Sample T Test Calculator returns Result 50.0000. Change any input and those results update instantly, so you can see exactly how each number you enter moves the answer.