2
$\begingroup$

I have 4 judges who listened to a set of 35 musical auditions. Each judge was then tasked with ranking the auditions from 1-35 in order of performance ability. What test can I use to compare agreement of rankings among the four judges? I was looking at Kruskal-Wallis, but I'm not sure this is right. (I am a music education doctoral student with a very limited background in statistical analysis, trying to figure out the best way to analyze my dissertation data). Thank you!

$\endgroup$
2
$\begingroup$

For analysis of rankings by different judges and similar problems, you can use Friedman test. See following links:

http://en.wikipedia.org/wiki/Friedman_test

Nonparametric alternative to ANOVA for testing the difference between several brand preferences

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thank you. I looked up Friedman, and I have the same question about it as about Kruskal-Wallis: both include in their formulas a value that is the sum of the rankings for each judge; however, if each judge is ranking 35 singers, each will include rankings 1-35, and the sum will be equivalent for each judge. How does this then account for agreement of actual rankings? I appreciate your help! $\endgroup$ – Kelley Mar 20 '15 at 2:52
  • $\begingroup$ I think when they say sum, they may mean sum of ranks given by all judges to one particular singer and not sum of ranks give by one judge to all singers which obviously will be 1+2..34+35. $\endgroup$ – rnso Mar 20 '15 at 4:01

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.