Module Components

• Overview

Topics

• Analysis of Variance (Anova)
• KRUSKAL-WALLIS TEST

NRSG 795: BIOSTATISTICS FOR EVIDENCE-BASED PRACTICE

Module 6: Significance Testing/Hypothesis Testing for 3+ Groups

KRUSKAL-WALLIS TEST

When you have data that are not normally distributed then the assumption of normality is violated and different statistical tests should be used.

It is used to analyze the relationship between a dependent variable that is ordinal and a categorical independent variable that has three or more levels. It tests the null hypothesis that the population distributions for three or more groups are identical against the research hypothesis that there are differences in the distributions for the groups. There should be at least 5 participants in each cell for this test to be used. Just like ANOVA, a significant result does not tell you where there is a difference between the groups and a post hoc analysis must be done to determine which groups are significantly different. Like all non-parametric tests, the Kruskal-Wallis Test is not as powerful as the ANOVA.

Here is some information in the event you want to try to run one (not required for class)

• Click here for an example illustration (calculated by hand)
• Intellectus Statistic: see the Nonparametric choice under Analyses
• Directions of how to do this in EXCEL:
• http://www.real-statistics.com/one-way-analysis-of-variance-anova/kruskal-wallis-test/
• https://www.youtube.com/watch?v=KmVTkVnWFuc&list=PLRUbsCrf-aQ8R1qdRG6Aj1ZwH27USHrNl&index=8 (10:02)

• On line calculator: https://www.socscistatistics.com/tests/kruskal/default.aspx

Required Video

• Watch this video about the Kruskal-Wallace test: (5:20)
  https://www.youtube.com/watch?v=BkyGuNuaZYw

Learning Activity

• Self test 6