Analysis of Variance Lecture Notes Review

Flashcards covering key vocabulary and concepts from lectures on Analysis of Variance (ANOVA), including assumptions, hypotheses, and statistical terms.

EFT Quiz

A 1% assessment in canvas where feedback is only given for wrong answers due to canvas limitations.

Analysis of Variance (ANOVA)

A statistical method that builds models with multiple predictors or factors to analyze variance.

One-Way Analysis of Variance

An ANOVA with one treatment, factor, or predictor regardless of the number of levels.

Treatments/Factors

Categories within the analysis, often called predictors, used to predict the response.

Levels

The various conditions or groups within a treatment.

Replicates

The number of independent repetitions that is performed for the treatments.

Observations

The values obtained from the samples, indicating what is being measured.

Normality Assumption

The assumption that the samples collected have a normal distribution, tested using methods like box plots.

Equal Variances Assumption

The assumption that the all treatments have similar variances, checked by calculating the ratio of the largest to smallest standard deviation.

Null Hypothesis in ANOVA

Suggests all groups are the same; the overall mean represents everything.

Alternative Hypothesis in ANOVA

Suggests at least two groups are different.

Post Hoc Test

Tests conducted after ANOVA to reveal specific group differences.

Treatment Effects

Differences caused by the experimental treatments.

Random Effects

Differences caused by random, uncontrollable, or unexplained effects.

F Statistic (F Stat)

A test statistic in ANOVA representing the ratio of treatment variance to residual variance.

Residual Sum of Squares

The variance within each treatment, indicating random differences due to noise.

Estimated Marginal Means

Used to compare groups visually using confidence intervals, revealing statistical differences.