Welch vs. Student t-test calculator
Drop in two groups of numbers and you'll get both t-tests side by side, each with the difference of means and its 95% confidence interval, plus the variance ratio for reference.
Numbers separated by commas, spaces, or new lines. At least 2 per group.
Numbers separated by commas, spaces, or new lines. At least 2 per group.
Student's t-test assumes both groups share the same population variance, which is rarely true for real biological data. Welch's t-test (Welch, 1947, Biometrika) drops that assumption and adjusts the degrees of freedom to match, so it holds up whether the variances are equal or not. Delacre, Lakens and Leys (2017, International Review of Social Psychology) make the case for defaulting to Welch's almost every time: it costs you next to nothing when variances are equal, and it saves you from a real mistake when they aren't. It's the test Ampliova runs by default whenever two independent groups are compared.
The stats are the last step. Ampliova takes your raw runs all the way there, efficiency correction and all. Try it free