One-Way ANOVA

College Depth 79 in the knowledge graph I know this Set as goal
Unlocks 90 downstream topics
ANOVA F-test between-group-variance within-group-variance multiple-groups

Core Idea

One-way Analysis of Variance (ANOVA) tests whether the means of three or more groups are all equal, using H₀: μ₁ = μ₂ = … = μₖ. The F-statistic is the ratio of between-group variance to within-group variance: F = MS_between / MS_within. A large F suggests means differ more than would be expected from chance alone. ANOVA does not identify which specific means differ — post-hoc tests (like Tukey's HSD) are required for pairwise comparisons after rejecting H₀.

How It's Best Learned

Run a simple experiment: measure plant heights under three different fertilizers. Partition total variability into between-group and within-group components in an ANOVA table. Emphasize why running multiple t-tests inflates the Type I error rate — this motivates ANOVA as the correct approach.

Common Misconceptions

Explainer

You already know the two-sample t-test: compare two group means by asking how many standard errors separate them. One-way ANOVA extends this logic to three or more groups. The natural instinct is to run all pairwise t-tests — with five groups you'd run ten — but this has a serious problem you've encountered in hypothesis testing: Type I error inflation. If each test has a 5% false-positive rate and tests are independent, the probability of at least one false positive across ten tests climbs toward 40%. ANOVA provides a single test that handles all groups simultaneously, preserving the overall error rate at α.

The central idea is a decomposition of variance. Take all the observations together and measure their total variation around the grand mean — the overall mean of all groups combined. This total sum of squares splits into two additive pieces: SS_between, which measures how far each group mean sits from the grand mean (weighted by group size), and SS_within, which measures how much individual observations scatter around their own group mean. SS_within is the baseline noise — variation that cannot be explained by group membership. SS_between is the signal — variation attributable to the groups themselves.

The F-statistic is the ratio F = MS_between / MS_within, where MS (mean square) divides each sum of squares by its degrees of freedom to make the quantities comparable. MS_between uses df = k − 1 (k groups), and MS_within uses df = N − k (N total observations). Under the null hypothesis that all group means are equal, both MS_between and MS_within estimate the same population variance σ², so F should be near 1. When group means genuinely differ, MS_between inflates while MS_within remains anchored to within-group noise, pushing F above 1. The F-distribution gives the probability of observing a ratio this large by chance alone.

A significant F-test tells you "at least one group mean differs," not which ones. Post-hoc tests like Tukey's Honestly Significant Difference (HSD) perform all pairwise comparisons with a correction that controls the familywise error rate at α — solving the multiple-comparison problem that motivated ANOVA in the first place. The procedure assumes roughly normal data within groups and approximately equal variances (homoscedasticity). The normality assumption is fairly robust for moderate sample sizes by the central limit theorem, but unequal variances can distort the F-test; in that case, Welch's ANOVA provides a correction analogous to Welch's t-test for the two-group setting.

Practice Questions 5 questions

Prerequisite Chain

Counting to 10Counting to 20Understanding ZeroThe Number ZeroCounting to FiveOne-to-One CorrespondenceCombining Small Groups Within 5Addition Within 10Addition Within 20Two-Digit Addition Without RegroupingTwo-Digit Addition with RegroupingAddition Within 100Repeated Addition as MultiplicationMultiplication Facts Within 100Division as Equal SharingDivision as Grouping (Measurement Division)Division: Grouping (Repeated Subtraction) ModelDivision: Fair Sharing ModelDivision as Equal SharingDivision as GroupingBasic Division FactsDivision Facts Within 100Two-Digit by One-Digit DivisionDivision with RemaindersRemainders and Quotients in DivisionDivision Word ProblemsIntroduction to Long DivisionFactors and MultiplesPrime and Composite NumbersEquivalent FractionsRelating Fractions and DecimalsDecimal Place ValueReading and Writing DecimalsComparing and Ordering DecimalsAdding and Subtracting DecimalsMultiplying DecimalsDividing DecimalsDividing FractionsMixed Number ArithmeticOrder of OperationsInteger Order of OperationsVariable ExpressionsCombining Like TermsOne-Step EquationsTwo-Step EquationsSolving Multi-Step EquationsEquations with Variables on Both SidesAngle Pairs: Complementary, Supplementary, and VerticalParallel Lines and TransversalsCorresponding AnglesAlternate Interior AnglesTriangle Angle Sum TheoremExterior Angle TheoremTriangle Inequality TheoremSimilar Triangles: AA SimilaritySimilar Triangles: SSS and SAS SimilarityProportions in Similar TrianglesRight Triangle Trigonometry IntroductionTrigonometric Ratios ReviewRadian MeasureConverting Between Degrees and RadiansThe Unit CircleGraphing Sine and CosineGraphing Tangent and Reciprocal Trigonometric FunctionsDerivatives of Trigonometric FunctionsAntiderivativesIndefinite IntegralsBasic Integration RulesRiemann SumsDefinite Integral DefinitionProbability Density Functions and Continuous DistributionsCumulative Distribution FunctionsContinuous Random VariablesNormal DistributionCentral Limit TheoremConfidence Intervals for MeansZ-Tests and T-Tests for MeansOne-Sample Z-Test for MeansOne-Sample and Two-Sample T-TestsOne-Way ANOVA

Longest path: 80 steps · 376 total prerequisite topics

Prerequisites (3)

Leads To (3)