Accessibility, sustainability, excellence: how to expand access to research publications/Success Criteria
6. Success Criteria
6.1. Our consideration of how best to meet the goal of increased access to published research results and findings in the environment we have described earlier in this report is built around a number of possible mechanisms, and a series of criteria against which to judge their likely success. The success criteria start from a common set of assumptions: that increases in access to the quality-assured findings of research conducted in the UK and across the globe will bring benefits to the UK economy and society of the kind set out in Section 3 above. The criteria themselves therefore describe in outline developments in or features of a research communications system that meets those ends. We discuss each of them in this section, before moving on to a consideration of the possible mechanisms. We are also aware that the criteria differ in kind. Those relating to increases in accessibility (A, B and C) and to high-quality research and services (G and H) describe outcomes in line with our core objectives. Those relating to costs, affordability and financial health (D, E, and F) are matters for attention in the process of developing a sustainable system of expanded access.
A. More UK publications freely accessible across the world
6.2. We noted earlier that UK researchers published over 123,000 peer-reviewed articles in journals in 2010, along with large numbers of monographs, reports, conference proceedings and other publications. No systematic attempt has been made to estimate the number of those articles that were immediately made accessible free at the point of use across the world; or even the number that are now accessible in that way. But the analysis in Sections 3 and 4 makes clear that only a relatively small proportion are accessible in any format on open access terms, and even then in many cases after a delay; and that while subscription-based access to major publications is provided to members of well-endowed research institutions, licensed access for other organisations and individuals, especially those outside the HE sector, is relatively meagre.
6.3. In order to meet this criterion, a greater proportion—preferably all—of those publications (including those written in collaboration with researchers in other countries) must be made accessible free of charge to anyone, anywhere in the world, who has access to the internet. The key aim, therefore, is to ensure that the results of research conducted in the UK—particularly if that research is publicly funded—should be freely accessible to the individuals and organisations anywhere in the world who may have an interest in them.
6.4. This criterion could in principle be met by a number of different mechanisms, or variants or combinations of them: through peer-reviewed open access or hybrid journals; through institutional or subject-based repositories; or through extensions to licensing (though the UK has little influence on licence arrangements overseas). Different mechanisms would have varying implications as to the version of the published findings that would be freely available; how easy it is to find them, and to navigate from them to related material; and the extent to which they can not only be read, but also analysed, manipulated, combined with other material, and used for a variety of purposes. Full accessibility would imply access to the version of record as published, in XML/HTML formats as well as PDF files, with full functionality and any semantic mark-up where that is provided by the publisher; and the ability to use and re-use the information with as few restrictions as possible.
B. More publications from across the world accessible to the higher education and research sectors in the UK
6.5. We have noted earlier that over 1.9m peer-reviewed articles were published in 2010 in c 25,000 journal titles, along with large numbers of other publications. Although the largest and most research-intensive universities and related research institutes have access to large proportions of those publications, no UK university has licensed access to all of them; and among the smaller and less research-intensive institutions, the proportion falls sharply.
6.6. In order to meet this criterion, more—preferably all—of the global total of research publications produced each year would have to be accessible to more—again, preferably all—of the members of the HE and research sectors as a whole, including those in smaller and/or less-well-endowed institutions. The key aim here is to ensure that members of the HE and research communities in the UK—students as well as academics—have access free at the point of use to the latest research findings wherever they are published.
6.7. Again, the criterion could in principle be met by a number of mechanisms, with varying implications as noted in paragraph 6.4 above.
C. More publications from across the world accessible to other sectors in the UK
6.8. For most people and organisations outside the HE sector—the health service; central Government and its agencies; other parts of the public sector including local government; the commercial sector, especially SMEs; the voluntary sector; and the public at large—it is at present often hard to secure access to journals free at the point of use. In order to meet this criterion, steps would have to be taken to make more—preferably all—of the global total of research publications accessible either to members of specific sectors or, again preferably, to everyone in the UK.
6.9. Like the previous criteria, this criterion could in principle be met by a number of different mechanisms. But since levels of awareness and understanding of the nature and scope of scientific and research publications is significantly lower outside the HE community and researchers in R&D-intensive businesses and other organisations, measures to increase access will have to be accompanied by a campaign to raise awareness, along with guidance on how to discover and navigate around such publications.
D. Financial sustainability for publishing 6.10. The research community, in the UK and worldwide, is supported by systems which provide effective and high-quality channels through which they can publish and disseminate their findings, and which ensures that those findings are subject to rigorous peer review. Effective communication of quality-assured findings and results requires a series of activities that involve significant costs. In order to meet this criterion, arrangements must be in place to enable publishers (whether they are in the commercial or the not-for-profit sector) to meet the legitimate costs of peer review, production, and marketing, as well as high standards of presentation, discoverability and navigation, together with the kinds of linking and enrichment of texts (‘semantic publishing’) that researchers and other readers increasingly expect. Publishers also need to generate surpluses for investment in innovation and new services; for distribution as profits to shareholders; and—for learned societies in particular—to support scholarly (and a wide range of related) activities for the benefit of their members and the wider communities that they serve. Finally, publishers need to take account of the sustained rise—3% to 4% a year—in the number of articles submitted to and published by them.
6.11. A number of studies have attempted to assess the costs involved in publishing peer-reviewed articles in journals. A report in 2008[1] demonstrated that there are considerable variations in costs per article between different journals, depending on the submission numbers; delivery formats (digital-only, print-plus-digital, or print-only); indirect cost structures; the level of surpluses generated by different publishers; and, above all, the rejection rate (i.e., the relationship between the number of articles submitted for peer review and the number that are finally published). Costs per article published, therefore, tend to be much higher for major journals with high submission and rejection rates—that is, those where there is the fiercest competition among researchers to publish their articles—than for those with lower rates.[2]
6.12. Subsequent reports also suggest that the costs for open access journals average between £1.5k and £2k,[3] which is broadly in line with the average level of APCs paid by the Wellcome Trust in 2010 , at just under £1.5k. The key point here is that no form of publishing is cost-free; and the key requirement is therefore that publishers—whether commercial or not-for-profit—should be able to generate revenues to meet the costs of those services they provide that are valued by researchers and their readers.
6.13. Some subscription-based journals—particularly in medicine and the life sciences—generate significant proportions of their income in addition from membership fees, advertising, the sale of reprints, and other sources.[4] Similarly, open access journals may have sources of income other than APCs, in the form of fees from membership schemes and so on. The scale of the market means, however, that advertising and similar sources are unlikely to generate significant amounts of revenue for more than a small minority of journals. Hence business models are likely to be built around moneys provided either by authors or readers, or those who provide funds on their behalf.
E. Costs and affordability for research funders
6.14. We have noted earlier that the great majority of funding to support research comes from Government and its agencies—primarily the Higher Education Funding Councils and the Research Councils—along with significant levels of funding from the research charities such as the Wellcome Trust. We have also noted that the overall costs of publishing and providing access to research publications have tended to rise over recent decades, but that they constitute a relatively small proportion of the total costs of research. Nevertheless, this criterion focuses attention on the need to ensure that costs are kept in check, and that the funds to support research communications in general, and increasing access to research publications in particular, are employed to best effect, both during a period of transition and for the longer term.
6.15. Assessment of the costs of different mechanisms and scenarios is therefore of critical importance in considering the most effective ways to increase access. We are also conscious of the current constraints on public expenditure, and also the different types of funding mix available in different subject/disciplinary areas. It is unlikely that significant increases in access—particularly to the publications from researchers outside the UK and for the benefit of people outside the HE and research sectors—can be achieved without some additional funding, or diversions from existing funds, particularly during a transition period; but such increases should be subject to a test of cost-effectiveness. We consider these issues further in the following sections.
6.16. In that context, we have taken account of the unique position of the UK in the global research communications system. As we have noted, researchers in the UK comprise just over 4% of the global research community; but they are responsible—often in collaboration with others from overseas—for over 6% of the publications produced each year. Hence in comparison with other countries, the UK’s production of research publications is disproportionate to its consumption of the publications produced elsewhere. Countries with lower rates of productivity tend, by contrast, to be net consumers of publications. This clearly has implications for the costs borne by different countries in supporting a global system characterised by a complex web of interdependencies.
6.17. It has been estimated that under current circumstances, where the subscription model still predominates (that is, where publishing costs are met in the main by readers, or the institutions that employ them), and where access beyond the academic and research communities is limited, the UK meets between 4% and 5% of the global costs of publishing and dissemination.[5] A global shift towards open access publishing funded by APCs is likely to lead to an increase in that proportion. The cash contribution from the UK may not rise—under certain optimistic assumptions it might even fall. But that will be the case only if market pressures keep publishing costs, and the level of APCs, in check; and if the UK does not during the period of transition take up open access publishing at a rate significantly faster than the rest of the world. Any significant move to extend licensed access beyond the groups that are currently covered is also likely to require an increase in funding.
F. Costs and affordability for universities
6.18. In considering costs to universities, it is important also to consider the roles of different funders of research, and how they might stimulate, or be affected by, changes in the current balance of research publications and business models. Under the dual support system, Government funding for research[6] is divided into two main streams: the Higher Education Funding Councils provide block grant to universities for research infrastructure and to support their strategic research priorities; and Research Councils provide grants to meet most of the full economic costs of specific projects and programmes of research. Under these arrangements, Funding Councils’ block grant plays a major role in meeting the costs of university libraries. But as we have noted earlier, there are no systematic arrangements for the payment of APCs, and that constitutes a significant barrier against the more widespread adoption of open access publishing. Higher levels of adoption will require some modifications to the rules relating to the use of Research Council grants in particular. Moreover, any new arrangements will have to take account of the policies and interests of other funders of research in the public, commercial and voluntary sectors.
6.19. Expenditure on academic libraries in total amounted in 2010 to 2.7% of overall university expenditure. For universities as for research funders, there is an imperative to keep costs in check even as the number of publications to which their staff and students want access continues to rise. But the precise impact of any changes in research publications and how they are financed—for the sector as a whole and for individual universities—will depend on three key factors: first, on any associated changes in the arrangements relating to the funds they receive, particularly from the Funding Councils and Research Councils; second, on the profile of individual institutions, in terms of size, research intensity, and mix of disciplines, as well as their current levels of expenditure on the library and its collections; and third, the speed of take-up of open access publishing across the world.
6.20. The sums currently paid by individual universities to secure access to journals and books vary widely, and are not necessarily related to size and research intensity: historic as well as current levels of provision may be significant too, not least because the pricing models for the big deals of many publishers still take account of individual universities’ levels of subscription in the print era. Patterns of library service provision and staffing levels also vary considerably; and all aspects of library services and expenditure—on content as well as staff—are coming under increasing scrutiny from senior managers in universities. The scope for increases in expenditure on libraries and their contents is generally seen as meagre, if it exists at all; more often, managers are looking for cuts. Few universities, if any, are seeking to extend the range of publications accessible to staff and students by purchasing more licences.
6.21. Many universities are also increasingly conscious of the impact that the VAT regime has on their decisions as to the nature and scope of their collections. For while books and journals in print do not attract VAT, e-books and journals do; and universities have only limited scope to reclaim the VAT that they incur on their purchases. This represents a significant barrier against moving towards e-only provision, despite the increases in efficiency that would result, together with lowering of costs, for both publishers and libraries.[7]
6.22. For all these reasons, individual universities will examine closely the financial implications for them of moves to increase access, and in particular at how they can avoid increases in costs. They will look at the likely transitional as well as continuing costs of measures to ensure that the university’s research is accessible across the world, and that more of the world’s research is accessible across the university. An example of the potential impact on an individual university is presented at Annex F.
G. Sustaining high-quality research
6.23. One of four principles guiding our considerations is the need to sustain the high levels of performance and standing of the research community in the UK. Its members operate in an ecology which provides high-quality channels through which they can publish and disseminate their findings; and which helps to ensure that they perform to best standards by subjecting those findings to rigorous peer review. It is critically important that in introducing any changes to the ecology, we do not put those key features at risk.
6.24. Learned societies play a significant role in that ecology in the UK, more important than in most other countries. Their central aim is to foster and promote the specific disciplines or subjects they represent, in three key ways: first, by facilitating two-way communication and engagement between researchers, policy-makers, practitioners, and the public at large; second, by nurturing researchers with opportunities for professional development and guidance at key stages in their careers; and third, by fostering a sense of professional collegiality and promoting good practice. Publishing and communicating the results of research are core to the missions of most learned societies, and they publish journals to meet the goal of disseminating high-quality research as widely as possible. Many of the journals published by UK learned societies are among the leading journals in their fields worldwide.[8] They also play a key role in sustaining the level of societies’ core activities, and that is of vital importance.
6.25. Quality assurance through peer review is enshrined in our terms of reference; and we believe that it is critically important to the users of research—both in the research community and in society at large—that published findings from whatever source, in the UK or worldwide, should be subject to peer review. Otherwise there is the risk that faulty or mistaken results can achieve currency, with damaging consequences. The risks can be especially severe in areas of research where findings may affect health and safety in the population at large.
6.26. Peer review is sometimes characterised an imperfect mechanism: it can take a long time and delay the publication of important results; it provides scant rewards for the efforts that researchers—hard-pressed for other purposes—devote to good reviewing; and since it depends on fallible human beings it cannot provide an absolute guarantee against the publication of faulty results. But most researchers regard peer review as overwhelmingly more reliable than other forms of review; and the principle that research publications should be subject before publication to rigorous review by expert peers—whether simply to check the rigour of the research or to assess its significance and likely impact in the field—is of critical importance. It becomes even more important as wider access to research-based publications leads to wider use by non-experts, who must have confidence in the quality assurance of the publication process, if they are to rely on the findings.
6.27. Nevertheless, it is important also to distinguish between the principle of peer review and the various ways in which it operates, with different degrees of openness and transparency. A number of approaches have been proposed, and experiments undertaken, with the aim of making peer review more effective. These have included measures to make reviewers’ names and/or the content of their reports open to authors and to readers; and to seek and publish feedback from a broad user community once an article has been informally disseminated or formally published. Different approaches appear to work more or less effectively in different disciplines, and post-publication review is widely seen as at best a complement to pre-publication review: while it may be useful for controversial or high-profile papers, it works less well for papers of more limited interest, not least because readers are unwilling to devote time to reviewing and commenting when they lack any incentive to do so.[9]
6.28. It is important, therefore, that there should be continuing monitoring and evaluation of peer review practices, and experimentation to seek improvements; and we support the recommendations of the House of Commons Science and Technology Committee to that effect.[10]
H. High-quality services to readers
6.29. Readers need help to discover information that is relevant to their needs and to navigate their way around the ever-increasing variety and volume of research publications. It is impossible for anyone to read and absorb all the publications that might be relevant in other than the most narrowly specialist fields of study. Hence the growing interest in machine-to-machine services. Readers have also come to expect the development of new services that enable them to interact with the content to which they have access, with enhanced links to other sources of information, and services that enable them to interrogate, manipulate and organise the content presented to them on a variety of platforms.
6.30. Publishers, libraries, aggregators and other intermediaries invest considerable sums in developing and implementing such services, and new entrants have added significantly to the range that is now available to readers. It is critically important that the research communications eco-system should continue to provide opportunities and incentives for new entrants to develop new services in this way. For as technology moves forward, readers will continue to demand more, and it is therefore crucial to sustain an environment that promotes innovation, investment in the infrastructure, and continued improvement in services.
Summary
6.31. We have noted in our discussion of the success criteria outlined above that each of them could be met in a number of different ways: none of them points unambiguously in one direction. We present a summary assessment of how different mechanisms might help to meet each of our success criteria in Annex D.
- ↑ Activities, costs and funding flows in the scholarly communications process in the UK, RIN, 2008.The report estimated that costs per article ranged from c£10k to c£2k, with an average of £2.8k
- ↑ The report estimated that costs per article ranged from c£10k to c£2k, with an average of £2.8k excluding the non-cash cost of the time spent by researchers and others in undertaking peer review. Houghton, J et al, Economic Implications of Alternative Scholarly Publishing Models, JISC, 2009 produced averages of between £3.2k and £2.3k for subscription-based journals, depending on whether they were published e-only, print-only, or in both formats.
- ↑ Houghton J et al, op cit; Heading for the Open Road: costs and benefits of transitions in scholarly communications, RIN, PRC, Wellcome Trust, JISC, RLUK, 2011. See also Solomon, D, and Björk, B-Christer,. A study of Open Access Journals using article processing charges. Journal of the American Society for Information Science and Technology , which suggests an average level of APCs for open access journals (including those published at very low cost in developing countries) of just over $900. It is difficult to judge – opinions differ – whether costs for open access journals are on average likely to rise as higher status journals join the open access ranks; or to fall as new entrants come into the market.
- ↑ Advertising is a particularly important for large-circulation journals such as The Lancet and Nature, but much less so for smaller-circulation niche journals.
- ↑ Those estimates take account of the disproportionate contribution that UK researchers make to the task of peer review. Activities, costs and funding flows in the scholarly communications process in the UK, RIN, 2008.
- ↑ For an analysis of the funding of research in the UK, see The Scientific Century: securing our future prosperity, Royal Society, 2010, page 17; and Making sense of research funding in UK higher education, RIN, 2010
- ↑ See E-only scholarly journals: overcoming the barriers, RIN, PRC, JISC and RLUK, 2010.
- ↑ The Association of Learned and Professional Society Publishers (ALPSP) estimates that about a third of all peer reviewed journals are published by learned societies, and that in 2006 three-quarters of the top 200 journals were published by non-profit publishers.
- ↑ It is worth noting, however, that post-publication peer review is the norm for Internet specifications and Web standards documents — RFCs (Requests for Comments) published by the Internet Engineering Task Force (IETF) and Candidate Recommendations published by the World Wide Web Consortium (W3C) — where the whole purpose of the initial publication is to make these documents available for post-publication peer review for a specified period of time, during which comments and criticisms from any interested parties are received and acted upon, before the new standards are formally agreed and published. Shotton, D, ‘The Five Stars of Online Journal Articles – a Framework for Article Evaluation’, D-Lib Magazine, 18 (1/2), 2012.
- ↑ House of Commons Science and Technology Committee, Peer review in scientific publications, HC 856, 2011