A very beginner level question here: I'm trying to see whether there is a difference in how students assess the usefulness of 5 possible courses to be added to their curriculum. Each student was asked to rate the perceived usefulness of each course.

Would the fact that I have the same sample (respondents) in each group mean I have matched groups and am therefore supposed to use paired samples measures such as one-way repeated measures ANOVA (or much rather the Friedman test because my variables are ordinal?). Or are the groups considered independent because I am not expecting a "treatment" effect, like when comparing before-treatment assessment and after-treatment assessment, and I therefore need to use regular one-way ANOVA/Kruskal-Wallis?

Thank you!


1 Answer 1


The assessments of various courses by the same student are quite certainly dependent, so no, independence is not a valid assumption here.


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.