Non Parametric Tests Flashcards

1
Q

Mann Whitney U is non parametric equiv to

A

Independent groups t test

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

Wilcoxon signed ranks test basic logic is to find

A

Each persons difference (or change) score from time 1 to time 2

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

When reporting Wilcoxon signed-rank t test you don’t give the

A

Mean

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

Wilcoxon you report like this

A

(Mdn =….., range =……..), T = ……., p = …… , r(squared) = ………

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

Mann-Whitney U rest is referred to as a

A

Rank sum test

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

Mann-Whitney U produces what statistic?

A

U

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

Mann-Whitney U uses a different formula sometimes why?

A

For different sample sizes

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

Mann-Whitney U is reported like this

A

(Mdn =……., range =……..), U =……., p =………, r(squared) =…….

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

Kruskal-Wallis test is non parametric equiv to

A

ANOVA (more than 2 groups)

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

Kruskal-Wallis H is based on

A

Ranked data

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

Kruskal-Wallis uses the

A

Chi square (xsquared)

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

Post hoc tests help us to determine

A

Which groups are sig

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

Bonferoni adjustment means that you

A

Divide level of p = .05 by number of comparisons to be performed to give new significance level

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

Watch out new alpha for post hoc comparisons because

A

Bonferoni post hoc divides alpha by number of groups!!!

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

Friedmans ANOVA is the I parametric equiv to the

A

Repeated measures ANOVA

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

Friedmans ANOVA is based in ranked data with key diff being

A

Groups are related. Ie one participant measured multiple times

17
Q

Post hoc test for the Kruskal-Wallis will always be

A

Mann-Whitney U - because it is also repeated measures

18
Q

Post hoc for Friedmans will always be

A

Wilcoxon

19
Q

Bonferoni only adjusts the

A

Significance level

20
Q

Effect sizes should be calculated for all non parametric tests (True /false)

A

True

21
Q

Only calculate effect sizes in the

A

Post hoc comparisons

22
Q

Post hoc effect size formula

A

r = z divided sq.rt N

23
Q

Chi square is used when there is what type of data

A

Nominal (ie head counts)

24
Q

Two types of data for chi square

A
  1. One way designs

2. Two-way designs

25
Q

One way chi - chi compares the expected frequencies (null) with

A

The actual observed frequencies (goodness if fit)

26
Q

The larger the chi then

A

The more sig results are

27
Q

Chi square issue is that it doesn’t tell you anything about the

A

Accounted variance.

28
Q

Nominal data we use

A

Chi squared

29
Q

Wilcoxon signed ranks test is the non parametric equivalent to

A

The paired t test