Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

ANOVA tukey post hoc

2012-06-29
2012-12-09
  • I am using 1 WAy ANOVA and Tukey post hoc to compare some data which is all located in one table. In the 'significant?' column some of the combinations of groups is 'not tested'
    - why is this
    -can I change it?

    K8

     
  • Daniel Kraus
    Daniel Kraus
    2012-06-29

    Hi K8,

    this is due to the way the Tukey test works. Read more at http://en.wikipedia.org/wiki/Tukey's_range_test#Order_of_comparisons

    With Tukey's test, you sometimes simply don't need some of the comparisons that are possible, as they would be redundant.

    If you want to have all comparisons in a posthoc test, choose Bonferroni-Holm or Holm-Sidak.

    Kind regards

    Daniel

     


Anonymous


Cancel   Add attachments