Unit 5 Non-Parametric analysis of variance: Kruskall Wallis test

Question 1

Whats a Kruskall Wallis test for

Answer 1

For comparing medians of more than two groups if parametric conditions are not met. It is the non-parametric equivalent of the one-way ANOVA.
Used to test whether:
The median of one or more groups are different from a control, or
The median of two or more groups are different from each other.

Question 2

Assumptions of Kruskall Wallis ranked analysis

Answer 2

Data must be from independent, random samples, and;
Data must be measured at least at the ordinal level.

Question 3

Whats some criticism with Kruskall Wallis tests

Answer 3

While KW test does not assume that data is normal, it assumes all groups have the same distribution. If your variance is different, regardless of normal or non-normal distribution, groups distributions are different.

Question 4

When do they recommend using KW

Answer 4

when we have a nominal (categorical) variable (or factor) and a ranked (dependent) variable:
the data is already in ranks