ANOVA Flashcards

0
Q

You could do 3 t-tests but that would…

A

Increase the likelihood of type 1 errors: finding significance when there is none

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

ANOVA used for…

A

Comparing groups or conditions

>2 groups = ANOVA (t-test 2 groups)

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

Parametric…

A

ANOVA

Almost always used, very robust

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

Non-parametric…

A

Kruskal-Wallis Test

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

Letter representing ANOVA

A

F

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

F-ratio represents…

A

Variance between groups / variance within groups

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

A significant F values means that we can…

A

Reject the null hypothesis

But it does not tell us which groups differ…

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

A post-hoc test…

A

Similar to a t-test

Correction for family-wise alpha errors built into them

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

Two most commonly used post hoc tests…

A

Scheffe’s confidence interval

Tukey’s honestly significance difference (HSD)

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

One-way ANOVA, what do you need?

A

One independent variable with three or more levels

One continuous dependant variable

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

ANOVA output…

A

F values

Multiple comparisons box

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

Repeated measures ANOVA, when do we need this?

A

One group
One dependant variable
3 or more time periods

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

Sphericity is…

A

An assumption of repeated measures ANOVA, replacing homogeneity of variance

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

If sphericity is violated…

A

We use a correct statistic: epsilon

Sphericity perfect= 1

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

Sphericity tests….

A

Mauchly’s test of sphericity- if answer yes we use epsilon to adjust degrees of freedom

Greenhouse-Geisser
Huynh-Feldt

Wilks’ Lambda- when sample sizes equal and observations independant

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

Mixed designs ANOVA

A

2 or more factors

i.e male and female, heart rate and two different classes (yoga, plyo)