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

Overview

The concepts of significance testing and hypothesis testing with two groups were covered in the previous module (T-tests, Mann-Whitney U and Wilcoxon ranked-sign test). This module extends group comparisons to three or more groups using both parametric tests (when the variable of interest is normally distributed in the population) and non-parametric tests (when the variable of interest is not normally distributed in the population).

As with the parametric t-test where the dependent variable is interval/ratio level, test assumptions must be assessed.  Violations of assumptions are particularly problematic when group sizes are very different.

There are several variations of ANOVA that you should be aware of but will not be covered in this class.  Indeed, there are newer statistical techniques that improve upon the older ANOVA approaches.

- A 1-way ANOVA is used when comparing the means of three or more groups – the one-way means that the groups being compared represent one variable with multiple levels (e.g., low, medium, and high levels of fish consumption).

- A 2-way ANOVA is used when comparing means across groups that are defined by two variables with multiple levels (e.g., low, medium, and high levels of fish consumptions for males and females).

- An Analysis of Covariance (ANCOVA) compares means across groups after controlling for an interval level variable like age (i.e., the means are adjusted for age).

- A repeated measures ANOVA compares means of groups over time.

This module has two subtopics. In the first, 1-way ANOVA is reviewed. Several videos are included to explain the conceptual background of ANOVA, how to conduct an ANOVA in Excel, and how to do post hoc testing with Bonferroni correction. In the second subtopic, the nonparametric alternative is described.

Objectives

At the conclusion of this module, the learner will be able to:

• Identify and test the assumptions of tests of mean differences for three or more groups
• Preform a mean differences analyses for more than two groups and interpret output

• Prepare text and tabular summaries that represent tests of mean differences for three or more groups
• Correctly interpret tests of mean differences for three groups or more in manuscripts particularly with regard to test assumptions, sample size, and generalizability of findings.
• Understand the basis of the Kruskal-Wallis test, and interpret the output.

Directions

The required videos are assigned within the subtopics. Learning activities within the subtopics are designed to help you apply what you learned.