12. Comparison of Several Means 2

Question 1

assumptions of an ANOVA

Answer 1

• normally distributed data
• homogeneity of variance
• independence
• DV should be measured on interval scale

Question 2

ANOVAs and independence

Answer 2

applies to independent ANOVAs, not repeated measures designs

Question 3

How is the ANOVA a robust test?

Answer 3

it is still accurate even when certain assumptions are violated

Question 4

violations of normality: groups not equal

Answer 4

accuracy of f-stat affected by skew

Question 5

violations of normality: group sizes equal, have at least 21 subjects per group, low drop out rate

Answer 5

f-stat not affected by skew

Question 6

violations of normality: group sizes equal, at least 40 subjects per group, higher drop out rate

Answer 6

f-stat not affected by skew

Question 7

violations of normality: homogeneity of variance

Answer 7

when sample sizes are equal, ANOVA is robust to violations of homogeneity of variance

Question 8

planned contrasts aka

Answer 8

planned comparisons

Question 9

When do you used planned contrasts

Answer 9

• when you have an apriori directional hypothesis that you want to test (1 tailed)
• only needed if the f-stat is significant

Question 10

What do planned contrasts allow you to do?

Answer 10

know where the significant difference lies

Question 11

When are post-hoc tests used?

Answer 11

• 2 tailed test

- only needed when f-stat is significant

Question 12

What do post-hoc tests consist of?

Answer 12

pairwise comparisons

Question 13

What are pairwise comparisons?

Answer 13

• performing multiple t-tests to determine where significance lies
• correcting for inflated type I error rate

Question 14

What happens as type 1 error rate decreases?

Answer 14

type II error increases

Question 15

What does an increase in type II error rate = ?

Answer 15

decreased power

Question 16

What are the post hoc tests?

Answer 16

• REGWQ test or Tukey test
• Bonferroni correction
• Gabriel’s test
• Hochberg’s GT2
• Games-Howell test

Question 17

What post hoc test would you use if you have equal sample sizes and group variances?

Answer 17

REGWQ test or Tukey test

Question 18

What post hoc test would you use if you want guaranteed control over type I error rate?

Answer 18

Bonferroni correction

Question 19

What post hoc test would you use if sample sizes are slightly different?

Answer 19

Gabriel’s test

Question 20

What post hoc test would you use if sample sizes are very different?

Answer 20

Hochberg’s GT2

Question 21

What post hoc test would you use if group variances are not homogeneous?

Answer 21

Games-Howell test