Publication bias can be observed in the actions of both authors and journal editors. Editors are more likely to publish studies with “positive” results and if those results are counterintuitive, even better. In turn, authors respond to this perceived tendency by submitting articles with positive results. This shared perception of the higher value of positive results may seem innocuous when submitting articles for publication, but the corresponding lack of attention paid to negative results has even greater consequences for academic publishing and scientific research.
The Negative Side of Positive Results
Studies with negative results do not get cited with the same frequency as positive results (since researchers typically cite supportive references), and since journals depend on citations to maintain their rankings, low-citation articles are not attractive enough to them. If the negative results contradict previously published work in the same journal, then the level of interest in that study would probably fall to zero. Given the reticence of editors to acknowledge retractions of prior articles, or even to aggressively pursue concerns over the methodology or accuracy of articles that have been published previously, any submitted material that is likely to involve re-visiting previous studies and, perhaps, bring peer review processes into question will earn a rejection. However, if negative results remain unpublished, researchers may continue to conduct studies without realizing that the basic methodology has some challenges. Moreover, by having access to both positive and negative results, it would promote further investigation of alternatives rather than wasting resources and funding dollars on irrelevant studies.
Building Scientific Reputation
Career advancement in academic research is built on the publication and frequent citation of studies with positive results. Academicians may prefer to portray themselves as victims of having to work in such a system, but it could be argued that they owe a greater obligation to the larger scientific community to contribute negative results that still broaden the body of knowledge on a specific topic rather than giving priority to the suggestions provided by journal editors.
The Potential for Open Access
Since authors are paying to get their work published in an Open Access (OA) model, there would seem to be a greater opportunity to get studies with negative results out into the community. The low-citation factor will still be there, but the clear commitment to “the greater good,” should offset that to some degree. As the OA model enters the second decade, there may be room for journals that would want to publish negative results. Unless altmetrics breaks the stranglehold of citation volumes as the key metric of academic quality, we may have to wait until the next generation of researchers comes along to break the mold.