BMJ  2005;331 (20 August), doi:10.1136/bmj.331.7514.0-d

Studies that find publication bias are no more likely to be published than studies that don't

There is no evidence that studies that find publication bias are preferentially published compared with studies that find no publication bias. In a systematic review Dubben and Beck-Bornholdt (p 433) identified studies presenting original data on publication bias. A funnel plot of effect size versus sample size of 26 studies included in the analysis showed no significant asymmetry in the data, indicating no evidence of publication bias. However, because only 26 studies were included, the authors warn that the power to detect asymmetry in a funnel plot is low.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Article

Systematic review of publication bias in studies on publication bias

Hans-Hermann Dubben and Hans-Peter Beck-Bornholdt
BMJ 2005 331: 433-434. [Full Text] [PDF]