Extended Anova Topics

Question 1

When do you do simple effects?

Answer 1

• When you get a significant interaction, you do simple effects.
• also, when you have a significant main effect of a factor that has more then 2 levels.

Question 2

How do you do simple effects?

Answer 2

Simple effects are done using t-tests. You do simple effects when you have a significant interaction and when you have a significant main effect of a factor that has more than 2 levels.

Question 3

What do follow up test do?

Answer 3

Follow up test locate where the effects are as anova tell you if there is a significant effect it doesn’t tell you where the effect is. So you do follow up test to find out where the difference is.

Question 4

What is an example of a follow up test?

Answer 4

Bonferroni correction is a follow up test that corrects for multiple comparisons by adjusting alpha level by dividing it by number of comparisons.

Question 5

What are the two types of follow up test for correcting multiple comparisons?

Answer 5

The two types of follow up test for correcting multiple comparisons including a priori (planned comparison) and post hoc tests.

A priori = or planned comparisons. It is chosen before the data is collected and a priori only corrects for comparisons that you plan to do.

post hoc = unplanned comparison. chosen after Data has been collected

Question 6

What is effect size?

Answer 6

Effect size is the magnitude (size) of the difference

Effect size in anova is the proportion of variance explained.

Question 7

What is the effect size of anova?

Answer 7

The effect size of anova is partial eta squared, np2.

Question 8

What np2?

Answer 8

Np2 is partial eta squared. It is the effect size in anova which is the proportion of variation explained. Effect size is the magnitude of the difference. It separates how big the difference is from the size of the sample.

Question 9

What is the rule of thumb for partial eta squared?

Answer 9

Small = 0.01 medium = 0.06 large 0.14

Question 10

How to report effect size in anova?

Answer 10

There was a significant main effect of … f(1,34) = 8.16 p = 0.03 np2 = 0.18.

Question 11

What is ANCOVA

Answer 11

ANCOVA is a blend of anova and regression. ANCOVA has one or more independent variables like anova and one or more covariates (continuous predictors of dv) like regression. ANCOVA examines the effect of independent variable on the dependent variable whilst controlling for covariates.

Question 12

Definition of MANOVA.

Answer 12

MANOVA is multivariate analysis of variance. MANOVA is similar to anova but has multiple dependent variables.