1
$\begingroup$

I have this dataframe:

group1<- matrix(c(rep(1,5),rep(2,5),rep(3,5),rep(4,5)), ncol=4)

group2<- matrix(c(rep(4,5),rep(3,5),rep(2,5),rep(1,5)), ncol=4)

I would like to compare these two groups with data of the preference ranking questionnaires. To my knowledge, the comparison of these two groups could be conducting the friedman.test using two matrices.

friedman.test(group1,group2)

However, I don’t know any post-hoc test for multiple comparisons using the two matrices in the friedman.test. I know there are some like the posthoc.friedman.conover.test or posthoc.friedman.nemenyi.test, that seem to use only one matrix, that is, these tests analize the effect in one matrix but not between both matrices.

If there is no post-hoc test, could you tell me another test to compare the two groups using matrices (or dataframes) and their multiple comparisons.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

You can use posthoc.friedman.conover.test and here is the sample code

posthoc.friedman.conover.test(y=y, p.adjust="none")

or Nemenyi-Test

posthoc.friedman.nemenyi.test(y)

y = matrix

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thanks Heinz, I know about your solution. However, I don´t know any post-hoc after coducting two matrices in the Frienman test. For only one matrix in the test, there are several post-hoc tests as you mention. $\endgroup$
    – Javi
    Commented Oct 11, 2023 at 10:28
  • $\begingroup$ we can merge it to single dataframe, then do that post test data <- data.frame(Group = rep(c("Group1", "Group2"), each = xxx), Preference = c(group1, group2)) $\endgroup$
    – Heinz Siahaan
    Commented Oct 12, 2023 at 1:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.