Week Ten - MANOVA

Question 1

What is a MANOVA?

Answer 1

When we have more than one DV (can be on different scales)

Question 2

what does a MANOVA promote?

Answer 2

Lower type I error

Question 3

When should we use a MANOVA?

Answer 3

If the goal of the design is to discover whether behaviour, as reflected by the DVs, is changed by manipulation of the IV

Question 4

What does MANOVA test for and what does this result in?

Answer 4

Differences on a linear combination of the DVs.

Results in potentially being more powerful/sensitive in identifying differences between groups - because it is examining differences along a combination of variables rather than just examining differences on a single variable

Question 5

For MANOVA, the DVs must be? (2)

Answer 5

Conceptually related (theoretical reasons)

Moderately correlated (use ANOVA if not) - cannot be too highly correlated either (results in redundancy of power)

Question 6

Too highly correlated DVs may be due to what? (2)

Answer 6

Multicollinearity: Very high correlations between variables

Singularity: One variable is a combination of a number of others in the analysis

Question 7

A correlation score of .85 or above means what?

Answer 7

That the DVs are just measuring the same thing and we should just look at one or the other based on the theory

Question 8

What are the assumptions of MANOVA?

Answer 8

INDEPENDENCE OF OBSERVATIONS: random allocation

MULTIVARIATE NORMAL DISTRIBUTION

HOMO OF VARIANCE (OF THE VARIANCE-COVARIANCE MATRICES)

Question 9

What is HOV(VC) in MANOVA?

Answer 9

We need HOV for each group AND the correlation between DVs needs to be the same in each group (assumes that the data in each cell comes from a single population)

Question 10

How do we check for HOV(CV)?

Answer 10

Levenes’s test

Box’s test: sig means violated

Question 11

When should we pay attention to Box’s test?

Answer 11

When we have unequal sample sizes as it is very sensitive to violations of multivariate normality

Question 12

MANOVA is reasonably robust to violations of HOV(CV) if?

Answer 12

There are equal numbers of participants per cell

Question 13

What multivariate test of significance does MANOVA produce?

Answer 13

Wilks Lambda

Pillai’s Trace

Question 14

When do we use Wilks Lambda?

Answer 14

Always unless there are some possible assumption violations

Question 15

When do we use Pillai’s Trace?

Answer 15

When the design is flawed/violated - eg small sample size, unequal numbers

Question 16

What do you do if MANOVA is significant?

Answer 16

Do separate univariate tests of all DVs (if correlations between dvs are low) so like normal anova with post-hoc follow up

Question 17

What do you do if MANOVA is not significant?

Answer 17

Nothing - could use effect size to build argument