One-Way ANOVA Calculator (Analysis of Variance & F-Test

Statistical Foundations of One-Way Analysis of Variance (ANOVA)

Developed by Sir Ronald Fisher, One-Way ANOVA tests whether the means of three or more independent groups are statistically equal without inflating the Family-Wise Type I Error rate that occurs with multiple pairwise t-tests:

• The Partitioning of Variance: Total variation in the data (SST) is partitioned into two orthogonal components: .

• Between-Group Variance (MSB): Variance caused by the treatment effect plus random sampling error.

• Within-Group Variance (MSW): Variance caused solely by inherent random subject-to-subject variation (error variance).

• The F-Ratio Logic: If the null hypothesis) is true, MSB and MSW both estimate the same population error variance, yielding F ≈ 1.

• If real treatment differences exist, MSB > MSW, driving F > 1.

0.

• Effect Size Benchmarks (Eta-Squared η^2):

• Small Effect: η^2 ≈ 0.01 (1% variance explained)

• Medium Effect: η^2 ≈ 0.06 (6% variance explained)

• Large Effect: η^2 0.14

• Post-Hoc Tests: ANOVA is an omnibus test; a significant F reveals that at least one group mean differs, but does not identify which specific pairs differ. Follow-up pairwise tests (Tukey's HSD, Scheffé, or Dunnett's test) are necessary.

How to Use This Calculator

Enter your Group 1 Data (comma or space separated) (e.g. 12, 15, 14, 11, 13), Group 2 Data (comma or space separated) (e.g. 18, 17, 20, 19, 16), Group 3 Data (comma or space separated) (e.g. 25, 23, 24, 28, 26), Group 4 Data [Optional] (comma or space separated) (e.g. 30, 32, 31, 29 (Optional)), and Significance Level (α) — choose from α = 0.05 (Standard 95% Confidence), α = 0.01 (Strict 99% Confidence), α = 0.10 (Exploratory 90% Confidence). Results update instantly as you type — adjust any value to compare scenarios.

The Formula

One-Way Analysis of Variance (ANOVA) Omnibus formulas:

Grand Mean: X_grand = Σ_j=1^k Σ_i=1^n_j X_ijN

Sum of Squares Between: SSB = Σ_j=1^k n_j (X_j - X_grand)^2, df_b = k - 1

Sum of Squares Within: SSW = Σ_j=1^k Σ_i=1^n_j (X_ij - X_j)^2, df_w = N - k

Mean Squares: MSB = (SSB) / (df_b), MSW = (SSW) / (df_w)

ANOVA F-Statistic: F = (MSB) / (MSW)

Eta-Squared (Effect Size): η^2 = (SSB) / (SST) = (SSB) / (SSB + SSW)

Omega-Squared: ω^2 = (SSB - (k - 1)MSW) / (SST + MSW)

Worked Example

For example, comparing 3 independent groups with \ each: Group 1 [12, 15, 14, 11, 13]), Group 2 [18, 17, 20, 19, 16]), and Group 3 [25, 23, 24, 28, 26]): With \ and Grand Mean and . The F-statistic is \ (p < 0.0001), confirming highly significant treatment differences with \ (92.9% of variance explained).

Understanding Your Result

Evaluate your result in the context of your specific use case and relevant industry standards or reference values. A single calculated result is a starting point — test two to three scenarios to understand how sensitive the outcome is to your key input assumptions. For high-stakes decisions, verify your result against a secondary source or subject-matter expert before acting.