Research strategy drivers for the institution
Managing submission to external research assessment
The main driver towards an overhaul of the research information systems in many institutions has been the process of reflecting on the often messy and time-consuming experience of submitting to RAE2008. This was coupled with the prospect of a substantial change in research assessment in the form of the REF (of course subsequently there have been changes which make the REF much more similar to its predecessors). Many university research managements said ‘Never again!’ and set about reorganising systems and looking at the choices available, whether in-house development or procurement of specialist publications management systems or full-blown Research Information Management systems.
A showcase for the university’s research
As well as submitting research for external assessment, universities also like to draw together and publicise a whole picture of the work of the institution. This can also drive income to the university by supporting its business and community engagement strategies. Though the web site usually performs the function of portal to the research, and external users may often enter through departmental or personal web pages, the repository enables these to operate effectively by providing a solid, reliable infrastructure to make supporting research outputs available consistently. It can therefore be used to manage proliferation of multiple versions of an output, and to ensure that outputs do not become inaccessible when web pages are rewritten or reorganised.
The repository can be an important part of delivering the ‘showcase’ and demonstrating the benefits brought by the university to its local community and region. For example, Northampton’s NECTAR repository tags publications and research reports with codes to indicate regional relevance to subjects such as health and education. A JISC-funded project for the Open Access Implementation Group has created a resource pack to encourage universities to embrace the Open Access agenda, and make research outputs openly available. The Open Access Repositories Resource Pack has been developed by the University of Glasgow and it has surveyed the guidance currently available to HEIs in the UK and further afield on why and how they can practically implement a more open approach to the release of their research outputs.
Internal research performance assessment
In many institutions, the research information system is also designed to provide information for assessing the research performance of individual academics, as part of regular appraisals and for promotion and tenure. As well as individual assessment, the systems can allow the university to look at the best performing Schools and other research units in terms of winning grants and overall income.
The place of the library and the repository in this process
In many cases, the process of submitting to RAE2008 was an opportunity for the library to play a more central role in the university’s research effort, as it became clear that even where there were established publications databases (whether central or departmental), the data was usually incomplete and often inaccurate. It was usually the task of the library to deal with duplication, ambiguous author names, multiple variations on journal titles and to add DOIs.
This did not necessarily imply that the repository itself was seen as important in this effort. In some cases, where the repository acted at least to some extent as a central publications database (in other words where it was a hybrid of metadata and full text), it was clearer that it would continue to have a central role in preparing for the REF. Even in cases where a decision was made to move to a Research Information Management system, the involvement of the library in the RAE submission process has sometimes led to representatives of the library being included in decisions about the development or acquisition of the systems and in some cases, to being given responsibility for implementation. This presents an opportunity to integrate the repository firmly into the new infrastructure. In any case, repository and library teams have often been brought closer together with their colleagues in research administration as a result of learning the lessons from the RAE.
The Library subject/liaison librarians can have an important role to play as well as the bibliographic experts; they can be the link bringing disciplinary particularities to bear to make the repository a better home for all types of outputs and ensuring that deposit routes are eased. The wider expertise that the Library has in search and discovery should also be deployed and where possible, the LMS should be better integrated with the repository, Search and Discovery section, Integrating LMS and repositories section.
Too often the library staff and the repository team have been too separated; it is now being appreciated that Library should be more centrally involved in supporting the research activities of the university and researchers themselves, and the repository is one part of that.
Implications for the repository of being embedded
The main positive implication for the repository of being embedded is that it will capture a greater proportion of research outputs produced by its institution.
However it is worth mentioning some other possible implications and concerns:
- An embedded repository, especially one which effectively becomes the research publications database, is going to contain more metadata-only records as well as full text
- If an embedded repository is to play a larger role in a university’s submission to the REF, it will have to be capable of managing a diverse range of outputs; it may already do so, of course, but if the emphasis up to now has been on PDFs of research papers, it may well have to adapt to other media types, and perhaps in the slightly longer term, research data
- Some feel that the emphasis on managing publications information in embedded repositories has led to the original impetus for repositories, namely the Open Access agenda, being put to one side. However, it can equally be argued that having a more complete record of what is being produced, in a timely way, as part of the flow of research information in the university, gives a better opportunity for the repository to secure the outputs themselves and to ‘switch on’ Open Access when any embargo expires. Keeping a separate repository in addition to a CRIS, as at St. Andrews, is one way to maintain the level of OA activity as an independent strand.
- There may be implications for the control of the quality of metadata records in situations where the repository is on the receiving end of metadata from other systems
- Researchers could react negatively to heavy-handed institutional initiatives to measure their research productivity and this potentially could affect the repository as well
Benefits to researchers
It is critical to successfully engaging researchers to ensure that they perceive the process as delivering benefits to them, not as yet another demand for information from some central office. This is much stronger if other academics themselves advocate for the benefits.
The most important benefits include:
Data is entered once, and can be re-used by many systems
- It is important to minimise the effort that needs to be made by the researcher (Maximising deposit through embedding workflows section) and also to then re-use that data again and again in different contexts. Researchers are often asked to enter data in many different systems, leading to inconsistencies, inaccuracy, resentment and often non-compliance.
Exposure and citation
- A primary benefit for researchers in depositing their own copies of papers or other outputs in the repository is making them accessible to their peers and other users. Though many researchers feel that they have done this through normal publication in journals, there is evidence that making the outputs openly available increases usage and possibly citation (studies so far point to opposite conclusions on citation rates but as time goes on, it should become clearer).
- Although the REF is much less based on bibliometrics than was originally expected (which would arguably have been a greater driver for open access), the requirement to show the impact of research can also be part of a strong case for depositing outputs. Usage of freely available outputs definitely increases and in cases where for example, a paper is referred to in a Wikipedia entry, usage can increase dramatically. When advocating self-archiving to researchers, one Vice-Chancellor said simply “Impact! Impact! Impact!”
- The repository can also house outputs which reflect impact e.g. videos of researchers presenting results to a wider audience, media interviews, reports for practitioners in professions or business partners etc.
- It can provide links to policy documents, official guidelines and other evidence of impact, adding real value to the items held in terms of demonstrating impact
- Besides the REF, the Research Councils are also requesting output and impact data from researchers in receipt of their awards. See this RCUK webpage for details of the Research Outcomes project.
Dissemination of non-standard outputs or unusual media types
- While authors of journal papers usually feel that by publishing their paper they have effectively disseminated their work, researchers in other areas may lack such outlets to reach their peers, for example, creative arts researchers. In several cases, where repository managers have taken the time to engage with researchers and understand their needs, they have made workflows smoother (through the use of plug-ins such as MePrints) and ensured the proper display of video, audio and photography file types. They have found enthusiasm from arts researchers, who have then asked for additional facilities, such as showing the context of a series of photographs of an exhibition. Researchers in the past have often had opportunities to show artefacts in various ways, but not the particular research outcomes of their work.
- Researchers in disciplines where work is often disseminated through working papers or conference papers, or who regularly write reports for government bodies or professional practitioners, for example in health or social work, can also be happy to find a new way to make their work more accessible.
- These examples illustrate the need for repository managers to get more closely acquainted with different disciplinary practices.
- It also argues for a flexible policy approach to the type of outputs a repository contains beyond the conventional research paper.
Populating web profiles
- One of the key benefits of an integrated research information system is to provide up to date and accurate publication lists to researchers’ personal web profiles. This relieves the researcher of having to update it manually and ensures the university that its website accurately reflects current research being undertaken, associated correctly with individuals who can then be contacted by potential research partners.
- It can then be possible to lever the profiles to suggest to the researcher that they can help users to close the loop if the profile reference can link to a full-text output in the repository.
- MePrints – an extension to EPrints, enables this facility in repositories using the software.
Appraisal, promotion, CVs
- Though the use of research information systems for appraisal purposes is certainly more of a benefit for the institution than for the individual researcher (who might actually see it as the opposite of a benefit), such systems do potentially give them more control over the information held about them, with the opportunity to review and amend publications, professional activities, research income and esteem information. It also reduces the burden of compiling such information when applying for promotions and creating CVs.
Fulfilling funder requirements
- There are increasing numbers of requirements from funders on researchers, from formal acknowledgment of funding support in a publication, which allows funders to track exactly which outputs arise from their grants and awards, up to the requirement to deposit final versions of outputs, and in some cases, datasets, in open access repositories.
- Institutional repositories that are embedded in overall research information systems can help to satisfy these requirements more efficiently. The adding of data about a new publication to the system can be identified as having originated from a grant made by a funder through the project code or other master data element. This can trigger reminders to fulfil the requirements of the grant to deposit the full text (subject to any agreed embargoes), either in the IR and/or if specifically required in another repository such as UKPubMed. Repository managers can also ensure that metadata fields are correctly populated by the standardised names of the funders supporting the research and with unique grant Ids (in future).
It may be possible to quantify a number of these benefits financially using a suitable method such as TRAC; this may be required as part of the business case.
A number of institutions in the UK and elsewhere have agreed to require their researchers to deposit in institutional repositories. A very few have made promotion and tenure decisions dependent on the publications deposited in repositories, though this is much more common in small countries which effectively have national Current Research Information Systems, such as Norway.
Research indicates that mandates do affect the level of self-archiving, but only in the case of very strong mandates – such as that at the University of Liege – are very high rates achieved. Most institutions in the UK, even those that have mandates, have preferred to opt for positive advocacy and embedding strategies which ease the process of deposit and yield tangible benefits, rather than trying to enforce mandates.
Most barriers to deposit in repositories are now fairly well-known on the whole:
“Not another administrative chore/I haven’t got time/it’s too complicated”
- Researchers often still see the process of depositing work in a repository as another demand on their precious time with little perceived benefit to themselves. This has been addressed both by mediated deposit (though this can fail to scale and be unsustainable in the long run) and by embedding repositories in systems that minimise duplication of effort and make workflows smoother. Better user interfaces are a feature both of CRIS-type systems and work to improve and personalise deposit routes to DSpace and EPrints.
The quality of other repository content
- Researchers may be worried about the ‘company they’re keeping’ in the repository, especially if the policy is to hold non-peer-reviewed content. This is an issue for the particular institution’s wider strategy and decisions must be taken on the balance of benefits. It may be that the researcher prefers to deposit in a subject repository alongside his or her peers instead.
- Researchers are very uncertain about the copyright status of their work, both in terms of whether they have retained copyright and assigned a licence to the publisher or not, and in relation to the particular publisher’s policy on self-archiving. The practical issues around this have been largely solved by the creation of resources such as Sherpa-Romeo and by passing responsibility for checking to the library (though lack of staff resource may cause backlogs between a researcher depositing and seeing his or her work ‘live’, which can present problems).
“Won’t this upset my publishers?”
- The wider problem of researchers fearing that their publishers will be upset by them depositing the paper is a more difficult one which should ease over time and through the advocacy of peers rather than librarians.
- It can be an especially difficult issue for researchers just starting out after their PhDs, but again the evidence from other researchers’ experience will be crucial
Loss of royalties
- The loss of royalties is only an issue in the case of monographs and other books, but apart from a minority of academic works that make it on to a lot of reading lists, royalties are tiny, and there is very little evidence that they are negatively affected by free copies. In fact a number of academic publishers are considering ‘freemium’ models for academic works. Monograph publishing in many disciplines is in any case under threat and in the end, exposure and citation are more valuable than royalties.
Author hasn’t kept a copy of their own final version
- This is a considerable problem, especially where repositories are trying to ‘back-fill’ with past publications
- The only solution is to educate researchers to deposit in future as soon as they have the final version
We’re getting more journal articles deposited but they tend to be the published version of the article. But that’s okay, that gives us more contacts, we can explain the version issue to the depositor… and why we can’t make that particular version live.” Rachel Proudfoot
Publishers have told them to use templates or online editing tools so no author copy exists
- A recent posting to a JISC mailing list highlighted situations where “the author claims to have no author prepared copy because they have been encouraged/told to use either publisher templates or publisher’s online tools to edit their articles from an early stage. This clearly diminishes the chance of an author prepared copy at the end of the publication process and so we can’t archive (even where the publisher ‘allows’ deposit of author copy).”
- One reply to this noted that a learned society publisher which used templates (RSC) did allow the deposit of such a version, whereas some commercial publishers opposed any self-archiving of versions incorporating their logos or styles.
Publishers’ terms and conditions can prevent inclusion
- In a small number of cases (for example, Elsevier and the Institute of Physics Publishing) publishers have amended their terms and conditions for publication to allow deposit of pre-publications versions only when the author is not mandated to deposit. The value and validity of this strategy has been questioned, but repository managers should investigate and determine the risk and a suitable strategy for themselves.