Non-parametric tests

If we are using an alpha level of .05 we judge a test to be significant if the p value is?

Less than .05

If you carry out a Kruskall-Wallis test and your results are significant what post hoc tests would you run?

Mann-Whitney tests

If you carry out a Friedman’s test and your results are significant what post-hoc test would you run?

Wilcoxon tests

You have a study which investigates the effect of caffeine. You have four different groups and each group drinks one of four different concentrations of caffeine drink. You carry out a Kruskall-Wallis test and find you have a significant result. How many post-hoc pairwise comparisons will you need to make?

6

You have a study which investigates the effect of caffeine. You have four different groups and each group drinks one of four different concentrations of caffeine drink. You carry out a Kruskall-Wallis test and find you have a significant result. You carry out post-hoc pairwise comparisons. What will your new alpha level be, against which you judge the p value?

.05/4 = .0125

The effect size for a Wilcoxon test is?

z statistic divided by the square root of the number of observations 

You carry out a study to see the effect of three different types of energy drink on runners’ performance. You stage a race and give each runner one of three drinks. At the end of the race, for each runner, you record which drink the runner drank and the position they finished in, in the race. Which test would be suitable to use on your results?

Kruskall-Wallis test

Which of the following is the non-parametric alternative to the one-way between-subjects ANOVA?

Kruskal-Wallis test 

The Kruskal-Wallis Test is used to compare:

More than 2 independent groups

Which test would you use for repeated measures instead of one-way ANOVA?

Freidman test

In the Kruskal-Wallis Test, the test statistic based on:

Ranks of the groups

In Friedman's test, the test statistic is based on:

Condition means