UNSPECIFIED (2004) A bound for publication bias based on the fraction of unpublished studies. BIOMETRICS, 60 (1). pp. 146-153. ISSN 0006-341X

Full text not available from this repository.

Abstract

Publication bias in meta-analysis is usually modeled in terms of an accept/reject selection procedure in which the selected studies are the "published" studies and the rejected studies are the "unpublished" studies. One possible selection mechanism is to suppose that only studies that report an estimated treatment effect exceeding (or falling short of) some threshold are accepted. We show that, with appropriate choice of thresholds, this attains the maximum bias among all selection mechanisms in which the probability of selection increases with study size. It is impossible to estimate the selection mechanism from the observed studies alone: this result leads to a "worst-case" sensitivity analysis for publication bias, which is remarkably easy to implement in practice. The method is illustrated using data on the effectiveness of prophylactic corticosteroids.

Item Type:	Journal Article
Subjects:	Q Science > QH Natural history > QH301 Biology Q Science > QA Mathematics
Journal or Publication Title:	BIOMETRICS
Publisher:	BLACKWELL PUBLISHING LTD
ISSN:	0006-341X
Date:	March 2004
Volume:	60
Number:	1
Number of Pages:	8
Page Range:	pp. 146-153
Publication Status:	Published
URI:	http://wrap.warwick.ac.uk/id/eprint/8625

Data sourced from Thomson Reuters' Web of Knowledge

Request changes to a record

Actions (login required)

View Item