Testing Multiple Groups

Question 1

What is the purpose of an ANOVA?

Answer 1

(called ‘Analysis of Variance’ but actually ‘of mean’)
Statistical test analysing whether 2 or more means are significantly different (of normally distributed data).

Question 2

What is the null hypothesis (H0) of an ANOVA?

Answer 2

• samples drawn from the same statistical population
• will have the same mean

Question 3

What is the alternative hypothesis (H1) of an ANOVA?

Answer 3

• samples drawn from different statistical populations
• have different means

Question 4

When do we reject the null hypothesis (H0)?

Answer 4

When p-value < 0.05

Question 5

When p-value < 0.05, what does this mean about the groups?

Answer 5

At least one of the groups is different.

Question 6

What is the F-ratio?

Answer 6

F = between group variance / within group variance
(variance = mean)
(called F value in ANOVA table)

Question 7

What does a high F-ratio mean?

Answer 7

The higher the ratio, the more likely that the groups were not drawn from the same statistical population.

Question 8

What are the two assumptions of an ANOVA?

Answer 8

• normality of the dependent variable
• equality of variance

Question 9

How many rows does a single factor ANOVA have?

Answer 9

2

Question 10

What is a post-hoc test?

Answer 10

Test to determine which groups are different from one another

Question 11

What is an example of a post-hoc test?

Answer 11

Tukey test

Question 12

What is the purpose of a Tukey test?

Answer 12

Finds the significant differences between pairs of means after a significant Anova.