Avoid lost discoveries, because of violations of standard assumptions, by using modern robust statistical methods
R Wilcox, M Carlson, S Azen, F Clark - Journal of clinical epidemiology, 2013 - Elsevier
R Wilcox, M Carlson, S Azen, F Clark
Journal of clinical epidemiology, 2013•ElsevierOBJECTIVES: Recently, there have been major advances in statistical techniques for
assessing central tendency and measures of association. The practical utility of modern
methods has been documented extensively in the statistics literature, but they remain
underused and relatively unknown in clinical trials. Our objective was to address this issue.
STUDY DESIGN AND PURPOSE: The first purpose was to review common problems
associated with standard methodologies (low power, lack of control over type I errors, and …
assessing central tendency and measures of association. The practical utility of modern
methods has been documented extensively in the statistics literature, but they remain
underused and relatively unknown in clinical trials. Our objective was to address this issue.
STUDY DESIGN AND PURPOSE: The first purpose was to review common problems
associated with standard methodologies (low power, lack of control over type I errors, and …
OBJECTIVES
Recently, there have been major advances in statistical techniques for assessing central tendency and measures of association. The practical utility of modern methods has been documented extensively in the statistics literature, but they remain underused and relatively unknown in clinical trials. Our objective was to address this issue.
STUDY DESIGN AND PURPOSE
The first purpose was to review common problems associated with standard methodologies (low power, lack of control over type I errors, and incorrect assessments of the strength of the association). The second purpose was to summarize some modern methods that can be used to circumvent such problems. The third purpose was to illustrate the practical utility of modern robust methods using data from the Well Elderly 2 randomized controlled trial.
RESULTS
In multiple instances, robust methods uncovered differences among groups and associations among variables that were not detected by classic techniques. In particular, the results demonstrated that details of the nature and strength of the association were sometimes overlooked when using ordinary least squares regression and Pearson correlation.
CONCLUSION
Modern robust methods can make a practical difference in detecting and describing differences between groups and associations between variables. Such procedures should be applied more frequently when analyzing trial-based data.
Elsevier