factorial anova

What is a factorial independent anova?

• Multiple IVs

• Different participants in all conditions

• Several independent variables

• One dependent/outcome variable

What is a factorial design?

How variables interact, the effect of one IV may depend on the level of another IV

What are the assumptions of factorial anova?

Independence

Normality

Homogeneity of variance

What are interactions?

Only happen in designs with more than 1 IV

The effect of IV1 changes depending on IV2

What are planned contrasts/ comparisons?

• more systematic and used for testing specific hypothesis

• Make a smaller number number of comparisons but can include mutliple conditions

• Useful in situations such as comparing groups against a control or if you expect an effect to get larger as the level of treatment increases

What is deviation planned contrast

• compares mean of each condition to overall mean

What is simple planed contrast

Compares the mean of each condition to either the first or last condition

What is helmert planned contrast

Compares mean of one condition to average of all other conditions

What is difference planned contrasts

Reverse of helmerts, compares the mean of a condition to average of all previous conditions

What is repeated planned contrasts

Compares sequential pairs of conditions

What is polynomial planned contrast

Looks for trends in the data

What are the rules of custom contrasts

• must be independent, contrasts must test unique hypotheses

• Only 2 chunks compared at once

• K-1, must have 1 less contrasts than number of groups

• Each group has a weight, positive/negatives compared

• Sum of weights= 0

What are the 2 main types of effect sizes?

Effect sizes based on the difference in means-scaled by variance

Effects sizes that tell you what proportion of the variance has been explained by the test

When is eta squared used?

Used for a one way anova, n²= SS effect/ SS total

When is partial Eta squared used

Used for factorial anovas

SPSS gives in ANOVA table

Tells your proportion of variance that is uniquely explained by the IV