MANOVA Flashcards

1
Q

When is MANOVA used?

A

When there are two or more DVs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What 3 questions can MANOVA help answer?

A

(1) Do changes in the independent variables (IVs) have significant effects on the DVs?
(2) What relationships exist among DVs?
(3) What relationships exists between IVs?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How does MANOVA answer these three questions?

A

MANOVA uses the covariance between DVs in testing significance of mean differences

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Running a single MANOVA avoids inflated _____ __ error rates that would occur from running multiple ANOVAs

A

type 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

MANOVA tests the difference between underlying ____ _____ while ANOVA only tests differences among ____ ___ __ _______ ______

A

latent variables

groups on an observed variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

MANOVA is more useful to social scientists because most topics research latent constructs that are not directly observable

A

Warne, 2014

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

When running a MANOVA, the null hypothesis is that…

A

all independent groups are equal on all dependent variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

4 statistical tests that should be run in order to reject null hypothesis

A

Pilai’s Trace,
Wilk’s Lambda,
Hotelling’s Trace,
Roy’s Largest Root

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Despite differences in their formulae, all 4 test the same null hypothesis and can be converted into…

A

an F-statistic which can then be used to calculate a p-value (Olson, 1976)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

If there are a large amount of DVs, the 4 tests may be contradictory due to differences in their ____ ____

A

statistical power

Haase & Ellis, 1987

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Roys largest root has the most power when…

A

DVs are highly correlated and represent a single construct.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Pilai’s Trace, Wilk’s Lambda and Hotelling’s Trace are more powerful for….

A

disparate outcome measures (Huberty & Oljenik, 2006)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

If the data violates any assumptions of the MANOVA, ____ _____ should be used as it is highly robust

A

Pillai’s trace

Olson, 1976

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

If these tests are significant and null hypothesis is rejected, the researcher will still not know which particular DV is influenced most by a particular IV….

what two follow ups?

A

Separate ANOVAs and Discriminant Function analysis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Who says that following a sig. MANOVA, subsequent ANOVAs are protected against type 1 errors?

A

Bock (1975)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

However, a significant MANOVA protects only the DV for which the group differences genuinely exist

citation?

A

Bray & Maxwell, 1985

17
Q

What should be applied if multiple anovas are run?

A

bonferroni corrections should be applied (Harris, 1975)

18
Q

who demonstrated that the same data can produce different results when analysis using a MANOVA or a series of ANOVAs.

A

Fish et al (1988)

19
Q

What does DFA attempt to do?

A

attempts to best discriminate a set of groups using several predictors

20
Q

How does DFA work?

A

DFA reduces and explains DVs in terms of a set of underlying dimensions which often reflect theoretical dimensions.

21
Q

How can you check how each DV contributes to that particular dimension

A

One can consult the standardized discriminant coefficients of significant dimensions

22
Q

n contrast to ANOVA, DFA is more in keeping with the ethos of MANOVA because….

A

it embraces the relationships that exist between DVs and can illuminate the relationships between DVs and group membership (Warne, 2014)