In Section 5.6, we describe the test for the null hypothesis that the medians in two groups are the same. How could you adapt the χ 2 test to the null hypothesis that the medians in k  2 groups are...

In Section 5.6, we describe the test for the null hypothesis that the medians in two groups are the same. How could you adapt the χ²
test to the null hypothesis that the medians in k
 2 groups are all the same?

Under what circumstances might the χ²
value be large but the contingency coefficient be low? How would you interpret that result?