Multiple Comparison Procedures: The Practical Solution
D. J. Saville · The American Statistician · 1990
A practicing statistician looks at the multiple comparison controversy and related issues through the eyes of the users. The concept of consistency is introduced and discussed in relation to five of the more common multiple comparison procedures. All of the procedures are found to be inconsistent except the simplest procedure, the unrestricted least significant difference (LSD) procedure (or multiple t test). For this and other reasons the unrestricted LSD procedure is recommended for general use, with the proviso that it should be viewed as a hypothesis generator rather than as a method for simultaneous hypothesis generation and testing. The implications for Scheffé's test for general contrasts are also discussed, and a new recommendation is made.
Cited by 696 works. doi:10.1080/00031305.1990.10475712
Authors
References
- Multiple Range and Multiple F Tests
- A METHOD FOR JUDGING ALL CONTRASTS IN THE ANALYSIS OF VARIANCE *
- Journal of the Royal Statistical Society.
- Comparing Individual Means in the Analysis of Variance
- An Evaluation of Ten Pairwise Multiple Comparison Procedures by Monte Carlo Methods
- A Bayes Rule for the Symmetric Multiple Comparisons Problem
- A Bayesian Approach to Multiple Comparisons
- The Present State of Multiple Comparison Methods
- Use and Misuse of Multiple Comparison Procedures1
- Detection of Differences Between Means: A Monte Carlo Study of Five Pairwise Multiple Comparison Procedures1