1 Way ANOVA Flashcards

1
Q

How does running an ANOVA reduce the risk of family-wise errors, compared to doing lots of t tests?

A

ANOVA simultaneously accounts for all comparisons within the same analysis, so type 1 error rate stays at 5%.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Define single factor ANOVA.

A

Analysis of one IV that may have multiple levels.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the total variance composed of?

A

Experimental variance and random variance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How do you calculate an F ratio?

A

Experimental variance/ Random variance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What does a large F value suggest?

A

Lots of variance is explained by the experimental manipulation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does total sum of squares refer to in independent 1 way ANOVA?

A

How far participants are from the overall mean.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What does model sum of squares refer to in independent 1 way ANOVA?

A

Between groups variability: How far each group mean is from the overall mean.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does error sum of squares refer to in independent 1 way ANOVA?

A

Within groups variability: How participants are from their group mean.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How do you calculate model mean squares in independent 1 way ANOVA?

A

SSmodel/ dfmodel.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How do you calculate error mean squares in 1 way ANOVA?

A

SSerror/ dferror.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What do we know when an ANOVA is significant?

A

There are differences, but we do not between which group/ condition yet.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the five parametric assumptions required to carry out an independent ANOVA?

A

IV is categorical, DV is continuous, independence of observations, normality and homogeneity of variance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How do you check for normality in your data?

A

Combination of tests of normality table (Shapiro-Wilk), Q-Q plots, Skewness and Kurtosis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

How do you check for homogeneity of variance in you data?

A

Levene’s test.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is homogeneity of variance?

A

Similar variance within groups across different groups.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

When do you conduct planned comparisons?

A

When specific hypotheses and comparisons were decided on prior to seeing the data.

17
Q

What are the 5 rules of orthogonal comparisons?

A

Only compare 2 chunks of variation, positive weighted groups compared to negative weighted, sum of weights must be zero, groups not involved must be coded zero, and weights assigned to one chunk of variation must be equal to weights assigned to the other.

18
Q

Which planned comparisons do you run when you are comparing more than 2 chunks of variation?

A

Non-orthogonal.

19
Q

What is required when conducting non-orthogonal planned comparisons?

A

Bonferroni corrections, to be conservative and account for correlated p-values.

20
Q

When do you conduct post-hoc comparisons?

A

To explore all possible differences after seeing the data.

21
Q

Which correction should you choose for post-hoc comparisons when Levene’s test was significant?

A

Games-Howell.

22
Q

Which correction should you choose for post-hoc comparisons when there is a SLIGHTLY different number of groups?

A

Gabriel

23
Q

Which correction should you choose for post-hoc comparisons when there is a VERY different number of groups?

A

Hochberg’s GT2.

24
Q

Which correction should you choose for post-hoc comparisons when you are doing MANY pairwise comparisons?

A

Tukey.

25
Q

Which correction should you choose for post-hoc comparisons when you are doing FEW pairwise comparisons?

A

Bonferroni.

26
Q

What does total sum of squares refer to in repeated measures 1 way ANOVA?

A

How far participants are from the overall mean.

27
Q

What does condition sum of squares refer to in repeated measures 1 way ANOVA?

A

Between groups variability: How far each condition mean is from the overall mean.

28
Q

What does within GROUPS sum of squares refer to in repeated measures 1 way ANOVA?

A

Within groups variability: How participants are from the condition mean.

29
Q

What does within SUBJECTS sum of squares refer to in repeated measures 1 way ANOVA?

A

How far each participant’s mean is from the overall mean.

30
Q

What does within error sum of squares refer to in repeated measures 1 way ANOVA?

A

SSwithin - SSsubjects of SStotal - SScondition - SSsubjects.

31
Q

How do you calculate condition mean squares in repeated measures 1 way ANOVA?

A

SScondition/ dfcondition.

32
Q

What are the four parametric assumptions required to carry out a repeated measures ANOVA?

A

IV is categorical, DV is continuous, normality and sphericity.

33
Q

How do you check for sphericity in your data?

A

Mauchly’s test.