week 7

Question 1

Why n-1

Answer 1

to get an unbiased estimate of how spread out the data really is especially when working with a sample

n-1 for the effect
For an interaction the df is df for IV1 x df for IV2
Within subject ANOVA- error df is number of participants per condition -1
Between subjects error df is total participants -1 number of groups -1

Question 2

2 x 3x 5 within subjects ANOVA

Answer 2

3-way ANOVA
3 IVs
IV 1 with 2 level, IV2 with 3 levels and IV 3 with 5 levels

All participants were tested in all conditions

Question 3

Mixed ANOVA

Answer 3

One IV is manipulated between participant
One IV is manipulated within participants

Question 4

Entering Data into SPSS

Answer 4

Each row is a participant
each column provides a bit of information about that participants

Question 5

Reporting ANOVA - Basic structure

Answer 5

Basic structure:
1- descriptive statistics: table of descriptive statistics
Details of how data meet assumptions
2- Results of ANOVA: main effect of IV1, main effect of IV2. Interaction
3- interaction plot

Question 6

Test of homogeneity of variance

Answer 6

Levene’s test in spss
The rule of four

Question 7

Non-parametric test

Answer 7

There is no non-parametric equivalent to multi-way ANOVA
you will need to split the analysis into the main effect, and analyse each main effect separately
Interactions cannot be tested using a non-parametric test
the choice of non-parametric test depends on: how many conditions are within the IV