The Duncan multiple range test (DMRT) was developed in 1955 by David B. Duncan and is classified as a post hoc test. This test uses a protection level of alpha for the collection of tests, rather than an alpha level for the individual tests. It is used to make pairwise comparisons that utilize a stepwise order to the comparisons. This test establishes an error level for the entire collection of tests, instead of an error rate for each individual test. In part, the DMRT was created as an alternative to the Student–Newman–Keuls test in order to have greater power. Notable is that the DMRT can be run regardless of whether an initial analysis of variance (ANOVA) resulted in a significant F value. This ...
Looks like you do not have access to this content.