5.6. Test for Ordered Differences
Sometimes you have a contingency table in which the columns represent an ordinal outcome and the rows are either nominal or ordinal. One test of interest is whether there are location shifts in the mean response; this is evaluated with the mean score test as discussed in Section 5.2.2. However, you may also be interested in testing against an ordered alternative; that is, are the mean scores strictly increasing (or decreasing) across the levels of the row variable? The Jonckheere-Terpstra test is designed to test the null hypothesis that the distribution of the ordered responses is the same across the various rows of the table. This test detects whether there are differences in
d1 ≤ d2 ≤ ... ≤ ds or ds ≥ d
Get Categorical Data Analysis Using The SAS® System, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
