Enter each group's measurements on its own line and run a one-way analysis of variance: are the group means more different than the spread within groups can explain? Get SS-between, SS-within, the F statistic and its p-value.
Each t-test carries its own false-positive risk; three groups mean three tests and roughly a 14% family-wise error rate at α = 0.05 instead of 5%. ANOVA controls that with one overall test first.
It says at least one group mean differs from the others, but not which pairs differ. Follow up with post-hoc comparisons (like Tukey's HSD) for the pair-by-pair story.
Independent observations, roughly normal data within each group, and similar variances across groups (homoscedasticity). Very unequal spreads or heavy skew call for alternatives such as Welch's ANOVA or Kruskal–Wallis.
SS-within measures how much individuals scatter around their own group's mean (noise). SS-between measures how much group means scatter around the grand mean (signal, scaled by group size). F is the signal-to-noise ratio after adjusting for degrees of freedom.
Grand mean over all N values. SS-between = Σ nᵢ(meanᵢ − grand)² with df = groups − 1. SS-within = Σ Σ (x − meanᵢ)² with df = N − groups. F = (SS-between ÷ df-between) ÷ (SS-within ÷ df-within), and the p-value is the right tail of the F distribution, computed via the incomplete beta function.
Inputs: groups [1,2,3], [4,5,6], [7,8,9].
Steps: Group means 2, 5, 8; grand mean 5. SS-between = 3×(2−5)² + 3×(5−5)² + 3×(8−5)² = 27 + 0 + 27 = 54. SS-within: each group contributes (1² + 0² + 1²) = 2, so 6. MS-between = 54 ÷ 2 = 27; MS-within = 6 ÷ 6 = 1. F = 27 ÷ 1 = 27, df (2, 6), p ≈ 0.0010.
Result: F = 27.00, p ≈ 0.001 — the group means differ far beyond within-group noise.
Add this calculator to your own website or blog — it is completely free, needs no sign-up, and always shows the latest version. Copy the code below and paste it into your page where you want the calculator to appear. Keeping the small “Powered by Loona Automation” credit line under the embed helps keep this tool free — but you are free to style or remove it; the tool works either way.