Silenced Issues 3: Silenced Issues in Academic Publishing
10 April, 2017
Having published books and journals across a variety of disciplines for many years, I have inevitably been exposed to numerous concerns relating to ignored and forgotten voices. Indeed, it has been well documented how diverse perspectives within the global publishing landscape are often neglected due to the persistence of exclusionary processes and traditional networks. Of crucial significance to STS scholars, consequent distortions in published outputs (which have the ability to shape entire scientific, social and cultural belief systems) not only lead to epistemic concerns, but also can adversely impact future potentials for innovation and discovery.
It has been persuasively argued that the prevailing obsession with qualitative metrics within academia is skewing the current literature, and citation distortions have been shown to result in unfounded claims of authority. Moreover, ‘metrics chasing’ increases the risks that valuable (though not highly cited) scholarly voices are simply not being heard. For example, the Matthew Effect of accumulated advantage, first coined by sociologist Robert K. Merton in 1968, shows how only the most highly cited research is ever cited. Tellingly, by some estimates 90% of all papers in the social sciences are never cited at all. On a more subtle level, considerations of career and funding opportunities commonly lead to cases of subjective citing. Coercive citation is another (relatively more recent) phenomenon whereby editors encourage specific references in order to boost a journal's impact factor; and the impact factor itself is calculated for journals indexed in the US-based and still mostly English-language Web of Science.
To ensure more balanced assessment (and in turn diffusion) of published outputs, some have argued we need to consider a broader basket of metrics, designed to measure both academic influence and wider social impact. This has led to the rise of alternative metrics (essentially social and mainstream media attention), and increasing numbers of institutions are now using proxies such as Altmetric scores to monitor their researchers’ wider engagement. Advocates of ‘alt metrics’ believe they help reveal the relevant impact of an individual's work. Yet whilst we are certainly past the point of relying solely upon the impact factor to assess quality, the digital divide (viewed in terms of access, skills and self-perceptions) surely favours those with better access to the prerequisite technologies.
Furthermore, as the importance of having an effective online presence grows, many authors are now moving beyond the ‘publish or perish’ mantra, towards a intent to be discovered approach. Indeed, it has never been a better time to be a social media ‘DIY-PR’ expert to help guarantee your voice is heard. But is there an issue that these forms of communication are gaining too much value? Are we losing sight of what actually matters? This question has ironically led to the creation of even more measures, such as the Kardashian Index, which aims to highlight discrepancies between a researcher’s social media profile and publication record.
Science is, of course, not purely objective and scholarly communication is unavoidably prone to homogeneity and bias. Scholars from many parts of the world – including developed countries – are encouraged (and in some cases effectively mandated) to publish in English-language publications because this is seen as a measure of excellence. In a similar vein, there is a large peer reviewer ‘gap’ in China (where researchers are writing substantially more papers than they are reviewing), which arguably neglects the contribution of local voices and perspectives.
Language bias is particularly problematic in the social sciences which are perhaps more regionally and locally engaged. As John Rennie Short and colleagues noted, “(the) growing use of English privileges the discourse of the Anglophone world even when its members are writing about other parts of the world”. The traditional ‘gatekeepers of impact’ can therefore (whether intentionally or not) suppress critical issues of societal relevance. This problem is amplified in less-developed countries, with many researchers still only publishing their work locally (often in their mother tongue) which is an issue across academia.
Regional and racial bias compound such issues for researchers located in the Global South, where there is often perceived (and real) prejudice during peer review. Of particular concern, this has led some authors to actively conceal their country of origin, or ‘gift’ lead authorship to co-authors based at more prestigious Western institutions. Again, such concerns are especially relevant to the social sciences where double-blind review is not as conventional (and even double-blinding does not always safeguard the detection of ‘regionality’).
Gender bias within academic publishing is another clear problem. It has been shown that in many fields women publish fewer papers than men and are less likely to be listed as first authors. The Matilda effect, coined in 1993 by Margaret W. Rossier, names the systematic denial of contributions by women in the sciences (their work often being attributed to male colleagues). It has been further argued that some editors inadvertently take gender into account when selecting reviewers, and during the peer review process itself research confirms the pervasive implicit association of science with men. As recently as 2015, two female authors were advised by a reviewer that, “it would probably also be beneficial to find one or two male biologists to work with (or at least obtain internal review from, but better yet as active co-authors)”.
Within the vast global publishing landscape, the persistence of cognitive bias, ‘old-boys’ networks and cultural partiality undoubtedly results in important voices being silenced. In some areas we are seeing improvements. Overall, the proportion of women researchers has increased in recent years, with Brazil and Portugal leading the way (49% of researchers in these countries are now women). Open Access publishing and advances in machine learning are creating new ways to deal with language problems. Public Library of Science (PLoS) journals, for example, encourage non‐English‐speaking authors to provide a version of their article in its original language as supplementary material, whilst international publishers (such as Springer Nature) have large foreign language publishing programs. France’s Institute for Humanities and Social Sciences provides online English versions of the nation’s leading research, and makes content freely available to French‐speaking developing countries.
Until recently, reliable metrics have been difficult to obtain for longer-form outputs, but holistic performance indicator tools like Bookmetrix.com, which combines all kinds of impact of scholarly communication, are now helping us to better understand the impact of academic books. Publishers are regularly hosting masterclasses, publishing and peer review workshops, undertaking researcher surveys, and disseminating white papers to help raise awareness of some of the problems identified here. Guaranteed double-blind peer review – which many authors have been shown to have a strong preference for – may also help avoid perceptions of bias for those who don’t reside in the ‘right’ country or institution (it’s not readily apparent how the current trend towards open review will counter embedded cultural tendencies). It has even been suggested that a system of review for editors and reviewers may deliver a check for demeaning comments (an add-on service for ORCID or Publons perhaps?) In any event, clearly more needs to (and should) be done. By improving exposure to underrepresented voices and issues, we are maximising the chances of publishing higher quality research and reducing the risks that important advances in innovation and discovery are not overlooked.
Joshua Pitt is Senior Commissioning Editor for Science and Technology Studies (Palgrave Macmillan).