lecture 3 - ANOVA

Question 1

What is ANOVA?

Answer 1

-Compares within group error variance to between group variance

Question 2

F-ratio

Answer 2

• Ratio of explained and unexplained variances
• Test statistic for ANOVA
• F = between group variance / within-groups variance
• Tests the likelihood of 2 variances being from the same population
• If F ratio large (F>1) = sig difference between groups

Question 3

Follow up analysis

Answer 3

• Post-hoc comparisons = checks where difference in sig ANOVA are located
• Problem = issues with multiple comparisons –> increases risk of making type 1 error

Question 4

Family wise error

Answer 4

• Usually if p<0.05 its significant but with multiple comparisons where p<0.05 they multiply the different comparisons together = creating greater significant = creating an error
• To get around this set an acceptable overall error rate –> divide the 0.05 by the number of comparisons

Question 5

Formula for calculating number of interactions in ANOVA

Answer 5

2(to the power of k)-k-1

k = no. of variables

Question 6

Interaction definition

Answer 6

-Interaction consists of any variation in the scores which isn’t due to error or main effects

Question 7

Interactions can be represented graphically

Answer 7

• Parallel lines = no interaction

- Non parallel lines = interaction