Literature Review – Open Access and Collaboration
Traditional academic publishing
Open Access Publishing
Encouraging greater collaboration
This paper aims to review the scholarly literature offering criticism on the concept of open access (OA) publishing, in particular with regards to the effect of OA on researcher’s collaborative working. Relevant literature is critically reviewed with the aim of identifying barriers to collaboration in academic publishing and as to what degree OA can mediate these.
OA relates to research outputs that are free of any restrictions on access (such as subscription walls) and free of any restrictions on use and re-use of research outputs (such as copyright and license restrictions) (Suber, 2004). Some scholars regard OA as a helpful tool for fostering collaborative work amongst researchers. Observing that scientific problems have become more complex and require a joint effort, these scholars argue that knowledge creation and dissemination could be more efficient if scientists worked together, emphasising ‘wisdom of the crowds’ and ‘multi-expert consideration’ as necessities in solving today’s scientific problems (Tacke, 2010). The core requirement for collaboration among researchers is commonly identified as their ability to share information freely (Nielsen, 2012). This relates not only to communication and collaboration technologies such as social networking sites or laboratory notebooks, but also to the degree of openness of scientific knowledge creation and dissemination. A researcher’s individual decision to collaborate is commonly understood on the basis of a cost-benefit analysis of the free sharing of information, with potential costs of collaboration being related to the risk of not being able to claim credit and profit (both reputational and monetarily) from freely available findings; and potential benefits being greater efficiency in the creation and dissemination of knowledge, the access to expertise, the aggregation of different kinds of knowledge and the possibility of raising one’s reputation (Haeussler, 2011). OA, as it enables researchers to make their findings available to other researchers and the wider public, is therefore by some scholars believed to eliminate barriers to greater collaboration amongst researchers.
Indeed, there is a case to be made in favour of OA encouraging greater collaboration. However, as Hitchcock (2013) observes, proponents of open access fail to address the practical concerns that many individual researchers would need to overcome in order to share their findings and engage in collaborative working with other researchers. Similarly, Nosek et al. believe that systematic weaknesses of OA create a research and publishing environment where it is not in the interest of individual researchers to adopt an open, collaborative and transparent approach to sharing their findings, code and data.
This paper aims to explore what barriers to collaboration there are, attempting to draw conclusions as to how OA could be amended in order to animate collaboration amongst researchers. Due to the nature of this topic, this literature review is not solely built upon traditional publications, but also includes scientific blogs and newspaper articles. Rather than providing a complete list of peer-reviewed articles on the subject, this paper aims to present a concise picture of the current scholarly discussion.
Traditional academic publishing
OA is a facet of open science, both of which have emerged in response to the current model of academic publishing. The ‘traditional’ publishing model relies on researchers submitting articles to journals for free, or paying the publisher to accept their research. Volunteers are recruited from experts in the field to conduct a peer-review quality control. Traditional publishers’ income is derived through subscriptions acquired from universities for their employees or students to access research outcomes. Author processing charges (APCs) are also common; in order to be considered for publication in a prestigious journal and to make their research freely available, authors need to pay fees. The rights for the published article usually remain with the publisher. This business model is characterised by systematic weaknesses, which many scholars believe to not only result in high costs or a limited quality of UK academic research and science, but also to create barriers in the processes of knowledge creation and dissemination that hinder collaboration amongst researchers (Haeussler, 2011). These barriers are mainly related to the inability to access research findings because of embargos imposed by the publisher, to restrictions on usage of data applied by the publisher and to the difficulty in reusing published data. As a result, researchers are neither able to publish preliminary research findings in real-time, enabling other researchers to engage in the research process (and potentially help in solving complex scientific problems); nor are their final research findings freely available to other researchers after being published unless high fees are being paid. Collaboration is further restricted through the focus on positive results. Articles that show a significant result in their findings are more likely to get submitted and published than those that do not evidence a hypothesis. Some scholars assume that negative or small results contribute crucially to the scientific knowledge process. Since, however, negative results are often dismissed, the entire knowledge created in such projects is not available to other researchers, preventing them from collaboratively reviewing and providing a clearer understanding of the effect under investigation. A further barrier to greater collaboration lies within the nature of the current approach to performance monitoring. Publishing in a prestigious journal effects the Impact Factor (IF) of the article; the number of high IF papers is the key performance metric for most academics and institutions. The IF is a flawed metric that has been shown to correlate poorly with the scientific quality of individual articles (Brembs et al., 2013). More than that, this metric does not capture the breadth of work carried out by a researcher, the researcher’s contribution to her field or her engagement in collaborative work with other researchers.
However, surveys of researchers show that they rate a journal’s reputation, and the subsequent effect on their IF, as the most important factor when deciding where to submit their papers for publication (Nature Publishing Group, 2015). Such incentives and practices that are not conducive to encouraging high quality research or collaborative work amongst researchers are thought to be prevalent in universities and research institutions that are increasingly competing for funds (Binswanger, 2014).
Open Access Publishing
In contrast to traditional publishing, open access journals are (mostly) published online and are funded via comparatively low APCs. Most OA publishing journals still rely on peer-review based assessments; they claim, however, to focus on “certifying the accuracy and validity of articles, not on evaluating their significance” (Frontiers, 2016).
To advocates, an OA approach would see research become a real time practice between collaborating researchers. Under OA, researchers are believed to be able to publish preliminary research findings in real-time, enabling other researchers to engage in the research process and potentially help in solving complex scientific problems. More than that, OA is believed to allow researchers to make their ‘nil result’ findings, failed replications and data that do not conform with predominant scientific theories freely available to the wider scientific community. This is argued to give researchers a truer reflection of the state of research in a particular field and, hence, to further eradicate barriers to collaboration amongst researchers (Brembs et al., 2013). These developments are believed to enhance the overall efficiency of knowledge creation and dissemination, creating benefits for both individual researchers and their research fields. Accordingly, Hitchcock (2013, Online) cites studies that show that OA approaches are beneficial to researchers, leading, for example, to “more citations, media attention, potential collaborators, job opportunities and funding opportunities”. Similarly, other advocates argue OA to enhance the efficiency in the creation and dissemination of research outputs, to enable researchers to access expertise and different kinds of knowledge and to raise their reputation and visibility, which is of particular interest for researchers at the beginning of their career who have few (if any) publications to their name (Alexy et al., 2014).
Many advocates assume OA to induce a pivotal shift in the scientific practice in the near future – namely from closed to collaborative behaviour – which, in turn, is likely to bring fundamental changes in the culture of science. Referring to collaborative knowledge discoveries that vary from small expert collaborations to large-scale amateur collaborations, Nielsen (2012), for example, promotes the idea of OA creating comprehensive scientific commons. As he states that “we need to imagine a world where the construction of the scientific information commons has come into fruition” (ibid., p. 111), it becomes obvious that Nielsen’s vision is based on the collective willingness to share on the part of scientists, showing that his notion of collaboration is bound to profound changes in scientific practice. Similarly, Haeussler (2011) assumes that a scientist’s sharing behaviour is built on a non-monetary system of return, arguing that if scientists expect the inquirer to be able to return the favour, they are more likely to share information even at temporary social costs. This argumentation, again, implies a fundamental change in the culture of science being induced by OA.
Nielsen, Haeussler and other scholars in favour of OA can be said to reveal visionary perspectives of scientific practice and culture under the regime of OA. These scholars, however, base their assumptions on a small number of individual cases of collaborative knowledge discoveries that happened to benefit both the individual researchers involved and their field as a whole. As Fecher & Friesike (2014) point out accordingly, more research needs to be done on the structural requirements of OA as well as on the importance of potential incentives for scientists to share information.
In response to this criticism, detractors of OA further elaborate on the potential costs and benefits associated with free sharing of information and place greater emphasis on the structural requirements of OA. Observing that OA has not managed to eradicate the systematic weaknesses of traditional academic publishing, these scholars believe that OA routinely failed to “encourage transparent, open and reproducible science” (Nosek et al., 2015., p. 1422). They believe that under the regime of OA, a hostile research and publishing environment has remained in force in which it is not in the interest of the individual researcher to adopt an open, collaborative and transparent approach to sharing their findings; even though their fields might benefit if all researchers were to adopt such behaviours. To these scholars, OA herald “the emergence of a new class of research parasite” (Longo & Drazen, 2016, p. 276). Lancaster (2016) posits that the increasing availability of data advocated by an open science approach will exacerbate the already increasing trend towards the marketisation of research, and the benefits of greater openness will disproportionately advantage those large firms most able to leverage their resources to take advantage of freely available data on a large scale. He echoes Longo & Drazen; large companies will be able to parasitically benefit from ‘collaborating’ with the freely distributed products of public investment shared by researchers. The same principle may hold true for the relationship between individual researchers; more established and renowned researchers might disproportionately profit from appropriating the expertise of those researchers nearer the beginning of their career who have few (if any) publications to their name and, hence, a smaller visibility and confidence. Power relations, therefore, can be said to have a crucial impact on the costs and benefits associated with collaboration.
Longo & Drazen (2016) raise a related and valid concern regarding the need to credit researchers when their data is used. Some have suggested that the researchers who created the data should be credited as co-authors, but this may not always be suitable, as the original researchers would have no control over the quality of the paper or where it is published.
- ISBN (eBook)
- ISBN (Book)
- Catalog Number
- Institution / College
- University of Sussex
- Open Access