A Monte Carlo study of multiple comparison techniques
Bell, Merlin Gene
Larson, Harold J.
MetadataShow full item record
A study was conducted on the multiple comparison methods presented by Scheffe, Tukey, Student-Newman-Keuls, and Duncan under the experimental situation in which all populations were normal with equal variances and all means but one were equal. The characteristics of all four test procedures were compared for the case of multiple comparisons of pairs of means. These tests were conducted both with and without the prior performance of an analysis of variance. The Tukey and Scheffe procedures were compared in tests of linear combination of three means. Estimates were made of the power of the tests and of Type I error rates under both the null and alternate hypotheses. Scheffe's method was found to be too conservative for pairwise comparisons of means, but it was to be preferred over Tukey's method for combinations of more than two means. Duncan's method was the most powerful test of pairwise comparisons, but it maintained little control over one kind of Type I error. The S-N-K procedure showed a good balance between power and control of Type I errors.
Approved for public release; distribution is unlimited
Showing items related by title, author, creator and subject.
Kim, Kyung Cheol (Monterey, California. Naval Postgraduate School, 2004-06);High Frequency (HF) radar backscatter instruments are being developed and tested in the marine science and defense science communities for their abilities to sense surface parameters remotely in the coastal ocean over large ...
Barto, Jon L. (Monterey, California. Naval Postgraduate School, 1980-08); NPS67-80-005This report examines gas dynamic stabilization of high pressure, highly non-uniform electrical discharges for two electrode geometries. The geometries studied are the multiple-point to plane and the multiple-wire to plane. ...
Dendis, Alexandros. (Monterey California. Naval Postgraduate School, 2007-06);The mathematical pendulum is one of the most widely studied problems in engineering physics. This is, however, primarily limited to the classical pendulum with a single bar and mass configuration. Extensions to this include ...