EBP2 NonParametric Testing and Chi Square

Question 1

What type of data is used in nonparametric testing?

Answer 1

nominal or ordinal

Question 2

What are non-parametric statistics based on?

Answer 2

comparisons of ranks of scores
or
comparisons of counts (yes/no) or “signs” of scores

Question 3

True/False: Nonparametric tests are less powerful compared to parametric statistics.

Answer 3

true (~65-95% of parametric equivalent)

Question 4

Match the nonparametric test to the comparison:

Two independent groups
Unpaired T-test

Answer 4

Mann-Whitney U test

Question 5

Match the nonparametric test to the comparison:

Two related scores
paired t-test

Answer 5

Sign test

Wilcoxon signed-ranks test

Question 6

Match the nonparametric test to the comparison:

Three or more independent groups
one-way ANOVA

Answer 6

Kruskal-Wallis analysis of variance by ranks

Question 7

Match the nonparametric test to the comparison:

Three or more related scores
One-way repeated measures ANOVA

Answer 7

Friedman two-way analysis of variance by ranks

Question 8

Match the nonparametric test to the comparison:

2x3 mixed ANOVA

Answer 8

NONE

unable to perform on more complex designs

Question 9

What type of test is used based on ranks (equivalent to independent t-test)?

Answer 9

Mann-Whitney U

Question 10

What type of test is used based on signs (equivalent to paired t-test)?

Answer 10

Sign test/Wilcoxon signed-ranks test

Question 11

What is the question to ask regarding rank tests?

Answer 11

Is the difference in ranks larger than would be expected by chance alone?

Question 12

What is the question to ask regarding signs tests?

Answer 12

Is the difference in sign frequencies larger would be expected by chance alone?

Question 13

What type of test looks for association between nominal variables?

Answer 13

chi-square

Question 14

What are the two types of chi-square?

Answer 14

goodness of fit and tests of independence (association)

Question 15

Describe goodness of fit test.

Answer 15

compare observed frequencies of 1 variable to uniform of another

Question 16

Describe tests of independence (association).

Answer 16

compare observed frequencies from 1 variable to observed frequencies of another variable

Question 17

What is the question to ask during chi-square?

Answer 17

is the difference between expected and observed frequencies larger than would be expected by chance alone?

Question 18

You are looking at baseline demographics between two groups in a table published in an article. You note a p-value of .87 (alpha set at .05) for the characteristic of gender. Which test was most appropriately used to compare genders between groups at baseline?

Answer 18

chi-squared

Question 19

You are wanting to compare two groups in a study you are designing. One group is getting eccentric exercises targeting the gluteus medius muscle group and the other group is getting concentric exercises targeting the same muscle group. Subjects were randomly selected from 3 different clinics located at different geographical locations. The outcome of interest is muscle strength which will be measured as either “strong”, “diminished” or “poor” via a manual muscle testing procedure. Which is most appropriate in this situation?

Answer 19

Mann-Whitney U

Question 20

You are wanting to compare four different groups in a study you are designing for individuals with persistent low back pain. One group is getting mirror box training, another group is getting laterality training, the third group is getting a combination of both and the final group is the control group. The outcome of interest is tolerance to pressure at the L4 segment and measured as “hypersensitive”, “normal” or “diminished”. The data will be collected prior to the intervention and then 4 weeks post intervention. The group is randomly selected from several clinics. Which is most appropriate in this situation?

Answer 20

Kruskal-Wallis

Question 21

When is a Fisher’s exact test used instead of a Chi square?

Answer 21

small samples

Question 22

You are performing a research trial looking at strength training in those with ACL-deficient knees. You are only able to recruit 10 subjects from 1 clinic that sees collegiate soccer players on a school campus. You will be comparing two groups (experimental and control) and look at strength via a Biodex machine (Newtons) over 3 time intervals. Which test is most appropriate?

a. two way ANOVA
b. 2x3 mixed-model ANOVA
c. Wilcoxon signed-ranks test
d. none of the above

Answer 22

none of the above

because it would be a 2x3 mixed but due to the non-parametric qualities cannot do