A Comparison of Exact Tests for Trend with Binary Endpoints Using Bartholomew’s Statistic

Consiglio J. D. (), Shan G. () and Wilding G. E. ()
Additional contact information
Consiglio J. D.: Department of Biostatistics, The State University of New York at Buffalo, 3435 Main Street, Buffalo, NY 14214, USA
Shan G.: Department of Environmental and Occupational Health, School of Community Health Sciences, University of Nevada Las Vegas, Las Vegas, NV 89154, USA
Wilding G. E.: Department of Biostatistics, The State University of New York at Buffalo, 3435 Main Street, Buffalo, NY 14214, USA

The International Journal of Biostatistics, 2014, vol. 10, issue 2, 221-230

Abstract: Tests for trend are important in a number of scientific fields when trends associated with binary variables are of interest. Implementing the standard Cochran–Armitage trend test requires an arbitrary choice of scores assigned to represent the grouping variable. Bartholomew proposed a test for qualitatively ordered samples using asymptotic critical values, but type I error control can be problematic in finite samples. To our knowledge, use of the exact probability distribution has not been explored, and we study its use in the present paper. Specifically we consider an approach based on conditioning on both sets of marginal totals and three unconditional approaches where only the marginal totals corresponding to the group sample sizes are treated as fixed. While slightly conservative, all four tests are guaranteed to have actual type I error rates below the nominal level. The unconditional tests are found to exhibit far less conservatism than the conditional test and thereby gain a power advantage.

Keywords: trend test; exact test; isotonic regression (search for similar items in EconPapers)
Date: 2014
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1515/ijb-2014-0013 (text/html)
For access to full text, subscription to the journal or payment for the individual article is required.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bpj:ijbist:v:10:y:2014:i:2:p:10:n:9

Ordering information: This journal article can be ordered from
https://www.degruyter.com/journal/key/ijb/html

DOI: 10.1515/ijb-2014-0013

Access Statistics for this article

The International Journal of Biostatistics is currently edited by Antoine Chambaz, Alan E. Hubbard and Mark J. van der Laan

More articles in The International Journal of Biostatistics from De Gruyter
Bibliographic data for series maintained by Peter Golla ().