0
$\begingroup$

I used the dunn.test() function to run post-hoc comparisons after significant outcomes in an ANCOVA with dunn.test(df$number,df$group,method="holm" and got the following results:

Kruskal-Wallis rank sum test

data: x and group
Kruskal-Wallis chi-squared = 12.2539, df = 3, p-value = 0.01

                          Comparison of x by group                           
                                    (Holm)                                     
Col Mean-|
Row Mean |         g1         g2         g3
---------+---------------------------------
      g2 |  -0.450407
         |     0.6524
         |
      g3 |  -2.985400  -2.416702
         |    0.0071*     0.0313
         |
      g4 |   0.330601   0.729442   3.062257
         |     0.3705     0.6986    0.0066*

alpha = 0.05
Reject Ho if p <= alpha/2

This nicely compares all groups against one another, but how can I limit the comparisons to 3 groups with 1 group, e.g. compare g2-g4 with g1?

$\endgroup$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.