Non Parametric Tests

Question 1

Mann Whitney U is non parametric equiv to

Answer 1

Independent groups t test

Question 2

Wilcoxon signed ranks test basic logic is to find

Answer 2

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

Question 3

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

Answer 3

Mean

Question 4

Wilcoxon you report like this

Answer 4

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

Question 5

Mann-Whitney U rest is referred to as a

Answer 5

Rank sum test

Question 6

Mann-Whitney U produces what statistic?

Answer 6

U

Question 7

Mann-Whitney U uses a different formula sometimes why?

Answer 7

For different sample sizes

Question 8

Mann-Whitney U is reported like this

Answer 8

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

Question 9

Kruskal-Wallis test is non parametric equiv to

Answer 9

ANOVA (more than 2 groups)

Question 10

Kruskal-Wallis H is based on

Answer 10

Ranked data

Question 11

Kruskal-Wallis uses the

Answer 11

Chi square (xsquared)

Question 12

Post hoc tests help us to determine

Answer 12

Which groups are sig

Question 13

Bonferoni adjustment means that you

Answer 13

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

Question 14

Watch out new alpha for post hoc comparisons because

Answer 14

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

Question 15

Friedmans ANOVA is the I parametric equiv to the

Answer 15

Repeated measures ANOVA

Question 16

Friedmans ANOVA is based in ranked data with key diff being

Answer 16

Groups are related. Ie one participant measured multiple times

Question 17

Post hoc test for the Kruskal-Wallis will always be

Answer 17

Mann-Whitney U - because it is also repeated measures

Question 18

Post hoc for Friedmans will always be

Answer 18

Wilcoxon

Question 19

Bonferoni only adjusts the

Answer 19

Significance level

Question 20

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

Answer 20

True

Question 21

Only calculate effect sizes in the

Answer 21

Post hoc comparisons

Question 22

Post hoc effect size formula

Answer 22

r = z divided sq.rt N

Question 23

Chi square is used when there is what type of data

Answer 23

Nominal (ie head counts)

Question 24

Two types of data for chi square

Answer 24

1. One way designs

2. Two-way designs

Question 25

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

Answer 25

The actual observed frequencies (goodness if fit)

Question 26

The larger the chi then

Answer 26

The more sig results are

Question 27

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

Answer 27

Accounted variance.

Question 28

Nominal data we use

Answer 28

Chi squared

Question 29

Wilcoxon signed ranks test is the non parametric equivalent to

Answer 29

The paired t test