The Effects of Nonnormality on Parametric, Nonparametric, and Model Comparison Approaches to Pairwise Comparisons
Bibliographic Data
| ID | 20284146 |
|---|---|
| Authors | Robert A Cribbie (0000-0002-9247-497X, York University), H J Keselman (University of Manitoba) |
| Year | 2003 |
| Volume | 63 |
| Issue | 4 |
| Pages | 615-635 |
| Publication date | 2003-08-01 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Educational and Psychological Measurement (JOURNAL) |
| Journal identifiers | ISSN: 0013-1644 • E-ISSN: 1552-3888 |
| Publisher | SAGE Publications (PUBLISHER • US) |
| DOI | 10.1177/0013164403251283 |
| OpenAlex | W2009687691 |
| Language | EN |
| Citations received | 2 |
| References cited | 30 |
Researchers in the behavioral sciences are often interested in comparing the means of several treatment conditions on a specific dependent measure. When scores on the dependent measure are not normally distributed, researchers must make important decisions regarding the multiple comparison strategy that is implemented. Although researchers commonly rely on the potential robustness of traditional parametric test statistics (e.g., t and F), these test statistics may not be robust under all nonnormal data conditions. This article compared strategies for performing multiple comparisons with nonnormal data under various data conditions, including simultaneous violations of the assumptions of normality and variance homogeneity. The results confirmed that when variances are unequal, use of the traditional two-sample t test can result in severely biased Type I and/or Type II error rates. However, the use of Welch’s two-sample test statistic with the REGWQ procedure, with either the usual means and variances or with trimmed means and Winsorized variances, resulted in good control of Type I error rates. The Kruskal-Wallis nonparametric statistic provided good Type I error control and power when variances were equal, although Type I error rates became severely inflated when variances were unequal. Furthermore, for researchers interested in eliminating intransitive decisions or comparing potential mean configuration models, a protected model-testing procedure suggested by Dayton provided good overall results
Confidence interval · Econometrics · F-test of equality of variances · Homogeneity (statistics) · Levene's test · Nominal level · Nonparametric statistics · Normality · Pairwise comparison · Parametric statistics · Sample size determination · Statistic · Statistical hypothesis testing · Statistics · Test statistic · Type I and type II errors · Advanced Statistical Modeling Techniques · Mathematics · Optimal Experimental Design Methods · Statistical Methods in Clinical Trials
Robustness?
On the Comparison of Several Mean Values
Statistical Practices of Educational Researchers
A new look at the statistical model identification
Significance tests for multiple comparison of proportions, variances, and other statistics
The effects of violations of assumptions underlying the t test
A more realistic look at the robustness and Type II error properties of the t test to departures from population normality
New developments in pairwise multiple comparisons
The unicorn, the normal curve, and other improbable creatures
| Citations per year | 0.11 |
|---|---|
| Citation span | 2008 - 2011 (4) |
| Highly cited | No |