An evaluation of some tests of trend in contingency tables

Arthur Cohen, Harold B. Sackrowitz

Research output: Contribution to journalArticlepeer-review

24 Scopus citations

Abstract

Consider an r × c contingency table under the full multinomial model in which each classification is ordered. The problem is to test the null hypothesis of independence against the alternative that all local log odds ratios are nonnegative with at least one local log odds ratio positive. A number of tests have been proposed for this problem, including the Goodman–Kruskal gamma test; a family of linear tests studied by Agresti, Mehta, and Patel; and a test based on “C–D,” the difference of concordant and discordant pairs in the table. In this article we show that all of these tests can be improved on in some sense for most cases. In fact the preceding tests sometimes are inadmissible in a strict sense. Furthermore, we show by example that in some cases improved tests can yield substantially improved power functions. We suggest a test based on a linear statistic similar to that presented by Agresti, Mehta, and Patel but that is followed up with a test that orders points by their probabilities on sample points where the linear test would randomize. This latter test compares favorably with competitors and has optimal theoretical properties. Exact tests, which entail auxiliary randomization, are discussed, as are the p values of the test procedures, which do not use auxiliary randomization.

Original languageEnglish (US)
Pages (from-to)470-475
Number of pages6
JournalJournal of the American Statistical Association
Volume87
Issue number418
DOIs
StatePublished - Jun 1992

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Keywords

  • Concordant–discordant pair
  • Contingency table
  • Exact test
  • Goodman-Kruskal gamma test
  • Inadmissible test
  • Ordered category
  • P value
  • Unbiased test

Fingerprint Dive into the research topics of 'An evaluation of some tests of trend in contingency tables'. Together they form a unique fingerprint.

Cite this