SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 

 View Only
Back to discussions
Expand all | Collapse all

Select correction test for multiple comparison with Kruskal-Wallis analysis

  • 1.  Select correction test for multiple comparison with Kruskal-Wallis analysis

    Posted Tue December 06, 2022 09:23 AM

    Dear SPSS users,

    I am experiencing a problem which it seems very difficult to overcome.

    I'm using SPSS for a Kruskal Wallis 1 way anova (independent k samples). I would like to not use the Bonferroni correction (which apparently is the default choiche) for multiple tests (I would prefere to use a Tukey's or Dunn's). Is there a way to change this default choice? If yes, how?
    Thanks a lot  



    ------------------------------
    Guido Rocchigiani
    ------------------------------

    #SPSSStatistics


  • 2.  RE: Select correction test for multiple comparison with Kruskal-Wallis analysis

    Posted Wed December 07, 2022 12:43 PM
    I forwarded this to one of our statisticians, who replied:

    ***
    We're using Dunn's method, which is sometimes referred to as Dunn-Bonferroni, by default. If the user would like something specific and different, we'd need a reference.
    ***

    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------

    Original Message