Chapter 22 Multi-Level Analysis (625-635)

Question 1

Analysis of Variance (ANOVA) (3)

Answer 1

1. Statistical technique that
2. Compares variances within and between samples
3. In order to estimate the significance of differences between a set of means

Question 2

Capitalising on Chance (4)

Answer 2

1. Making too many tests
2. With alpha set at .05
3. On the same data,
4. Hence increasing the likelihood of a Type I error

Question 3

Within Groups Sum of Squares (4)

Answer 3

1. Sum of squares
2. of deviations of scores
3. around their sample means.
4. Also: error SS.

Question 4

Within Groups Variance (3)

Answer 4

1. Total variance of scores
2. around sample means.
3. Also. error variance.

Question 5

Between Groups Variance (2)

Answer 5

1. Variance of sample means

2. around grand mean.

Question 6

Grand Mean (2)

Answer 6

1. Mean of all scores in a data set,

2. irrespective of conditions or groups

Question 7

Total Variance (2)

Answer 7

1. Variance of all scores in a set

2. around their grand mean

Question 8

Error Variance (3)

Answer 8

1. Total variance of all scores
2. from their group means.
3. Also: within groups variance.

Question 9

F test/ratio (1)

Answer 9

1. Statistic giving ratio of between groups to within groups variance

Question 10

Sum of Squares (1)

Answer 10

1. Addition of the squares of deviations around a mean

Question 11

Variance Ratio Test (1)

Answer 11

1. Full name for the test producing the F statistic

Question 12

Mean Sum of Squares (1)

Answer 12

1. Sum of squares divided by df

Question 13

Between Group Sum of Squares (1)

Answer 13

1. Sum of squares of deviations of sample means from the grand mean

Question 14

Error Sum of Squares (2)

Answer 14

1. Sum of squares of deviations of each score from its own group mean.
2. Also: within group SS.

Question 15

Pairwise Comparison (2)

Answer 15

1. Comparison of just two means

2. From a set of means

Question 16

Post Hoc Comparisons/Tests (2)

Answer 16

1. Tests between means, or groups of means

2. Conducted after inspection of data from initial analysis

Question 17

A Priori Comparisons/Planned Comparisons (2)

Answer 17

1. Tests of differences between selected means, or sets of means,
2. Which, from prior theory, were predicted to differ

Question 18

Family-Wise Error Rate (3)

Answer 18

1. The probability of making at least one Type I error
2. In all the tests made on a set of data,
3. Assuming H0 is true

Question 19

Error Rate per Comparison (3)

Answer 19

1. Given the significance level set,
2. The likelihood of a Type I error in each test made on the data
3. If H0 is true

Question 20

Bonferroni t Tests (3)

Answer 20

1. Procedure for testing means pairwise,
2. Which involves raising the critical values of t
3. To lower the family-wise error rate

Question 21

Linear Contrasts (2)

Answer 21

1. Procedure for testing between individual pairs of means or combinations of means,
2. a priori (i.e. predicted)

Question 22

Linear Coefficients (2)

Answer 22

1. Values to be entered into an equation

2. For calculating linear contrasts

Question 23

Newman-Keuls Post Hoc Test (2)

Answer 23

1. Post hoc test of means pairwise

2. Safe so long as number of means is relatively low

Question 24

Tukey’s (HSD) Post Hoc Test (3)

Answer 24

1. Post hoc test of all possible pairwise combinations
2. Appropriate analysis choice with a large number of means
3. Considered conservative.

Question 25

Tukey’s (WSD) Post Hoc Test (1)

Answer 25

1. Less conservative post hoc test than Tukey (HSD)

Question 26

Phi (1)

Answer 26

1. Phi statistic for estimating power in ANOVA analyses

Question 27

Kruskal-Wallis Test (4)

Answer 27

1. Non-parametric
2. Between-groups
3. Test of difference between several groups
4. (Mann-Whitney is the two-condition equivalent)

Question 28

Jonckheere Trend Test (3)

Answer 28

1. Non-parametric
2. Statistical test for the significance of a trend
3. In the dependent variable across unrelated conditions

Question 29

MANOVA (2)

Answer 29

1. Statistical procedure using ANOVA

2. On more than one dependent variable

Question 30

Analysis of Co-Variance (ANCOVA) (3)

Answer 30

1. Statistical procedure that performs an ANOVA
2. While partialling out the effect of a variable
3. That is correlated with the dependent variable (the ‘co-variate’)

Question 31

Co-Variate (3)

Answer 31

1. A variable that correlates with a dependent variable
2. On which two groups differ
3. And which can be partialled out using ANCOVA

Question 32

Scheffé Post Hoc Analysis (3)

Answer 32

1. Post hoc test
2. That takes into account all possible comparisons of combinations of means
3. (Most conservative post hoc test)