Applied statistics - Chapter 12: Analysis of Variance (ANOVA)

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/26

flashcard set

Earn XP

Description and Tags

Vocabulary terms and statistical concepts covering one-way and factorial Analysis of Variance (ANOVA), assumptions, components of variance, Mean Squares, F-ratios, post-hoc testing, and effect size.

27 Terms

1
New cards

Analysis of Variance (ANOVA)

A statistical hypothesis-testing procedure that provides a single omnibus FF test designed to evaluate differences among two or more population means without inflating Type I error.

2
New cards

Experimentwise Error

The probability of committing at least one Type I error when conducting multiple statistical tests (such as multiple tt-tests) from a single experiment,

3
New cards

Experimentwise Error (con’t)

If three tests were conducted, each at α = .05, the overall probability of making at least one Type I error would be approximately .14, nearly 3 times the original (intended) alpha level.

4
New cards

Omnibus F Test

A single overall test provided by ANOVA that is designed to detect any differences among group means while holding the Type I error rate at the specified alpha (α\text{α}) level.

5
New cards

Factor

An independent variable (IV) in an ANOVA design consisting of two or more conditions or treatment groups.

e.g. Control and experimental groups

6
New cards

Levels

The individual treatment conditions or specific values that make up an independent variable (factor).

e.g. Determine if teaching method (IV) affects student performance (DV).

  • IV: Teaching method (3 levels): 1. Traditional 2. Collaborative 3. Coaching


7
New cards

One-Way ANOVA (Single-Factor ANOVA)

An analysis of variance that evaluates the effects of a single independent variable with two or more levels (levels×1 factor×levels×or×more\text{levels} \times 1 \text{ factor} \times \text{levels} \times \text{or} \times \text{more}).

8
New cards

Two-Way ANOVA (Two-Factor ANOVA)

A factorial ANOVA design that assesses the simultaneous effects of two independent variables, each containing two or more levels (×2\times 2 levels).

e.g. Determine if teaching method (IV) and time of day (IV) have effects on student performance (DV).

  • IVs:

    • Teaching method (3 levels): 1. Traditional 2. Collaborative 3. Coaching

    • Time of day (2 levels): 1. Morning 2. Afternoon

  • DV: Student performance (interval/ratio)


9
New cards

Three-Way ANOVA (Three-Factor ANOVA)

Two factors, each with more than 3 levels

10
New cards

Between-Subjects ANOVA

The type of ANOVA conducted when the study samples across treatment conditions are independent of each other.

11
New cards

Within-Subjects ANOVA (Repeated-Measures ANOVA)

The type of ANOVA conducted when the study samples are correlated, such as when the same participants are tested across all treatment conditions.

12
New cards

Assumptions of between-subjects ANOVA

Sampling error is distributed normally around the mean in each treatment population.

13
New cards

Assumptions of between-subjects ANOVA (con’t)

Homogeneity of variance: the population variances of the treatment groups are equal.

14
New cards

Assumptions of between-subjects ANOVA (III)

Observations in a sample are independent of each other.

15
New cards

Homogeneity of Variance

An assumption of between-subjects ANOVA stating that the population variances of the different treatment groups are equal.

16
New cards

Between-treatments variance

Measures differences caused by:

  • Systematic treatment effects

  • Random unsystematic factors


17
New cards

Within-treatments variance

Measures differences caused by random unsystematic factors

18
New cards

Total Variance (Total Sum of Squares, SSTSS_T)

The total variability among all scores in a data set, partitioned into between-groups variance and within-groups variance (SST=SSB+SSWSS_T = SS_B + SS_W).

19
New cards

Between-Groups Variance (SSBSS_B)

The component of variance reflecting differences among the group means

20
New cards

Between-Groups Variance (SSBSS_B) [con’t]

Caused by both systematic treatment effects and random, unsystematic factors.

21
New cards

Within-Groups Variance (SSWSS_W)

An estimate of the average variability within each separate group, representing error variance

22
New cards

Within-Groups Variance (SSWSS_W) [con’t]

It’s caused solely by random, unsystematic factors not controlled by the researcher./

23
New cards

Mean Squares (MSMS)

Always positive variance estimates computed by dividing a sum of squares (SSSS) by its associated degrees of freedom (dfdf); these values are always positive.

24
New cards

Mean Square Between (MSBMS_B)

The always positive between-groups variance estimate which reflects both treatment effects and error (T+ET + E).

25
New cards

F Ratio (FobservedF_{\text{observed}})

The test statistic for ANOVA, defined as the ratio of between-groups variance to within-groups variance

26
New cards

Tukey HSD Test

A post-hoc multiple-comparison procedure conducted after a significant omnibus FF test to identify which specific pairs of means differ significantly, calculated as HSD=qcritical×MSWn\text{HSD} = q_{\text{critical}} \times \frac{\sqrt{MS_W}}{\sqrt{n}}.

27
New cards

Eta-Squared (η2\eta^2)

A measure of effect size in ANOVA that quantifies the proportion of total variance in the dependent variable accounted for by the independent variable, calculated as η2=SSbetweenSStotal\eta^2 = \frac{SS_{\text{between}}}{SS_{\text{total}}}.