Are academic journals impartial? While many would suggest that academic journals work for the advancement of knowledge and science, we show this is not always the case. In a recent study, we find that two international relations (IR) journals favor articles written by authors who share the journal’s institutional affiliation. We term this phenomenon “academic in-group bias.”
In-group bias is a well-known phenomenon that is widely documented in the psychological literature. People tend to favor their group, whether it is their close family, their hometown, their ethnic group, or any other group affiliation. Before our study, the evidence regarding academic in-group bias was scarce, with only one study finding academic in-group bias in law journals. Studies from economics found mixed results. Our paper provides evidence of academic in-group bias in IR journals, showing that this phenomenon is not specific to law. We also provide tentative evidence which could potentially resolve the conflict in economics, suggesting that these journals might also exhibit in-group bias. In short, we show that academic in-group bias is general in nature, even if not necessarily large in scope.
To test the possibility of academic in-group bias, we examined four of the leading academic journals in international relations: World Politics, International Security, International Organization, and International Studies Quarterly. World Politics is published by Cambridge University Press on behalf of the Princeton Institute for International and Regional Affairs at Princeton University. International Security is published by MIT Press and edited by the Belfer Center for Science and International Affairs at Harvard University. The other two journals are not affiliated with a specific university and are considered our control group.
The basic idea of our methodology was to compare the citation counts of articles published by in-group members with those published by out-group members. In academia, citations are considered a marker of quality — the more citations a paper receives, the higher quality it is assumed to be. If papers written by researchers from Blue University and published in the Blue Journal get fewer citations than papers written by researchers from Red University and published by the Blue Journal, this could signal that the Blue Journal was willing to lower its standards for its own researchers; this could then indicate in-group bias. We found, for example, that the average article published in World Politics by an author affiliated with Princeton, which publishes the journal, gets roughly 80 Google Scholar citations, whereas papers written for that journal by non-Princeton researchers receive roughly 105 Google Scholar citations.
We look at faculty and PhD graduates from Princeton, Harvard, and MIT as the in-group members. Examining two journals and two in-groups provides more evidence that this phenomenon is general in nature. We consider MIT as part of the Harvard in-group because of the close ties between the institutions; both are located in Cambridge, Massachusetts, and researchers from these institutions have collaborated together on many research projects.
Our results confirm the existence of academic in-group bias. When published in Harvard- or MIT-related journals, articles published by graduates of Harvard and MIT receive roughly 60% fewer citations than papers written by out-group scholars. This difference is statistically significant and very large in magnitude. It’s also in contrast to what we see when we look at the control group journals. In these journals, Princeton authors get roughly the same citations as the out-group authors, while Harvard and MIT get more citations.
The favoritism we found did not look the same in each context. Unlike the PhD graduates, Harvard and MIT faculty do not seem to enjoy such favoritism. The picture for Princeton is the opposite — Princeton graduates do not seem to get special treatment, but faculty members do enjoy editorial favoritism, though to a lesser extent than Harvard and MIT. In the analysis we control for other confounding factors the best we can, such as the number of authors and the length of the article, and make sure the results are not driven by our choice of specification or empirical strategy.
While our results survive many robustness checks, there are several possible limitations to our study. The first is in the interpretation of our findings. Journals might choose their articles based on other factors than likely citation counts, such as better suitability with the journals’ scope. It is also important to bear in mind that most academic journals are not affiliated with universities or research institutions. Therefore, our results, while troubling, do not carry over to the majority of academic journals.
What harm can academic in-group bias create? First, it can tilt tenure decisions and other promotions based on an academic’s publications. Some competent scholars might lose while others who are less competent might benefit. This adverse effect can be minimized if the field incorporates this bias into its decision-making process, putting less weight on publications of in-group members in the home journals and assigning more weight to publications of out-group members. Another possible approach would be to strengthen the double-blind refereeing process by not allowing the editor to see the affiliation of the author. This, however, seems unpractical when the author and the editor have the same affiliation.
Another, potentially more severe, implication of our findings is the possible effect of academic in-group bias on the academic endeavor to advance science. If articles are not published based on merit, the dissemination of knowledge might be at stake. Having non-meritocratic systems might push out talented individuals, to the detriment of the academic community. These implications are troubling enough to call for immediate attention, and subsequent action, to address it.
from HBR.org http://ift.tt/2BSidKw
