e-Articles (Chapter 18)
Here are some full-length research articles that illustrate the use of nonparametric tests. To view any article, simply click on its title. (NOTE: No claim is made that these articles are perfect in all respects. By carefully reviewing them, you will hone your skills at being able both to decipher and to critique statistically-based research reports.)
Illustrates the use of the median test to make pairwise comparisons among 3 groups. In this investigation, the groups were business firms. These firms were put into the 3 groups based on the focus of their strategic plans. The dependent variable was a measure of profitable growth. See, in the article's "Results" section, Table 1 and the 6 paragraphs that follow that table. (Note that in each comparison, the test statistic is converted into a Z-value).
Because their data were either ordinal in nature or not normally distributed, the researchers used three nonparametric procedures to analyze their data: Spearman's rank-order correlation and Wilcoxon's matched-pairs signed-ranks test.
Illustrates the use of the Mann-Whitney U-test. For certain comparisons, the 2 groups were made up of those with and without chronic obstructive pulmonary disease (COPD). For other comparisons, the 2 groups were subsets of those with COPD: those who were either in or not in the top 25% in terms of cost of health care received. See Tables 1, 2, 3, 5, and 6, as well as the researchers' discussion of the findings presented in these tables.
In this study, each of 4 dogs was measured on 5 occasions. For each dog at each point in time, the median of several assessments of skin thickness was recorded. These data were analyzed with a Friedman's two-way ANOVA to see if the dogs' skin thickness had changed, with plans for Wilcoxon's test to make pairwise post hoc comparisons in case the Friedman test was significant. See the portion of the "Materials and Methods" section called "Statistics," the last paragraph of the "Results" section, and Table 1.
Copyright © 2012
Schuyler W. Huck