What is Dunn’s multiple comparison test?

Viewing 2 reply threads
  • Author
    Posts
    • September 6, 2017 at 2:57 am #18454
      koreanraichu
      Participant

      Guten morgen.(means ‘Good mornong’ in Deutsch)

      Recently I read a paper entitled "Circulating Tumor DNA Mutation Profiling by Targeted Next Generation Sequencing Provides Guidance for Personalized Treatments in Multiple Cancer Types", the Dunn’s multiple comparisons test appeared in Figure 3 of that paper. When I search on Google, it is one of statistics, but I do not know statistics.

      I know p <0.05 that it means "to reject the null hypothesis when p <0.05", but for what is it a test, what is the multiple comparison and the Dunn’s test I do not know.

      1. What is Dunn’s test and what is test for blackening?
      2. What is the multiple comparison?
      3. Is Dunn’s multiple comparison test proceeding with multiple comparisons using Dunn’s test?

      PS: Are Dunn’s test and dunnett test the different tests?

    • December 17, 2018 at 10:23 am #116402
      rahilsaxena
      Participant

      Once your initial ANOVA has found a significant difference in three or more means, Dunn’s Test can be used to pinpoint which specific means are significant from the others. Dunn’s Multiple Comparison Test is a post hoc (i.e. it’s run after an ANOVA) non parametric test (a “distribution free” test that doesn’t assume your data comes from a particular distribution). It is one of the least powerful of the multiple comparisons tests and can be a very conservative test–especially for larger numbers of comparisons. For example, if you have 10 comparisons at an alpha level of .05 then the formula results in a very per-comparison error rate of .005.
      The null hypothesis for the test is that there is no difference between groups (groups can be equal or unequal in size).
      The alternate hypothesis for the test is that there is a difference between groups.

    • June 17, 2020 at 5:08 am #117148
      Jitu Singh
      Participant

      No. p<0.05 doesn’t mean “to reject the null hypothesis when p <0.05”. If you don’t know anything about statistics I suggest you start taking a course or at least pick a good introductory book otherwise any explanation we could write will not make sense to you and here is not the place to teach you basics in statistics.
      Online Clinical Research Training

  • Author
    Posts
Viewing 2 reply threads
  • You must be logged in to reply to this topic.