Search Dictionary
Version history
- Current: Jun 6, 2022.
- Old versions: Dec 28, 2021
Duncan’s multiple range test
Duncan’s multiple range test (MRT) is one of the post-hoc tests used in multiple comparison.
In general, the use of Duncan's MRT test is not recommended because it has a high risk of making type I error (false positive) as pointed out by many studies [e.g., 1–3]. The increase in the alpha levels is intentional and inherited from the Student-Newman-Keuls method.
Definitions in the literature
- As noted by others, avoid Fisher’s LSD, Duncan’s MRT, and the SNK tests. These tests are very liberal as they do not make acceptable error rate adjustments [1].
- The Duncan's multiple range test is performed using steps similar to SNK procedure. Changed α error levels are applied following the step-down procedure. It shows more liberal tendency than the SNK procedure. Generally Duncan's multiple range test is not recommended when sample sizes are unequal because of the liberal tendency [2].
- We do not recommend Fisher's protected least significant difference (LSD) test, Duncan's multiple range test, or the Student–Neuman–Keuls (SNK) test, none of which can be relied on to keep EER at or below the nominal level (Day and Quinn 1989; Quinn and Keough 2002) [3].