I have 4 levels for a single factor, which is called "Compliance".

Compliance has an ordinal scale (0, 1, 2, 3) and it is computed as the number of rules a team complies with.

My hypothesis is the following: "the more you comply, the more effective you are". Effectiveness is a real positive number.

I have 4 independent samples, one for each compliance.

My questions are: 1) Can I apply Jonckheere-Terpstra?

2) If I apply J-T and it succeeds, I can only conclude the existence of a weak monotony. I would like to check whether there are significant differences among pairs. In particular, I would like to check the following: C=0 vs C=1, C=1 vs C=2 and C=2 vs C=3. What test should I run? Should I apply any correction?



1) I checked the literature and the answer seems to be yes, I can apply that test. 2) Again, yes, independent t-tests are appropriate and no correction (e.g. Bonferroni) is needed.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.