The Research Councils UK (RCUK) policy — which came into effect on April 1st 2013 — requires that all peer-reviewed papers and conference proceedings (and eventually monographs too it is assumed) arising from research funded by RCUK are made open access, either by researchers paying to publish in open access journals (gold OA), or continuing to publish in the traditional (subscription) manner and then depositing copies of their works in an open access repository (green OA), usually after an embargo period.
The policy of the Higher Education Funding Council for England (HEFCE) — which will come into effect in 2016 — will require researchers to deposit all their final peer-reviewed manuscripts in an institutional or subject repository as soon after the point of acceptance as possible, “and no later than three months after this date”.
It has taken the OA movement twelve years to get the UK to this point (the Budapest Open Access Initiative was authored in 2002), but advocates believe that these two mandates have now made open access a done deal in the country. As such, they say, they represent a huge win for the movement.
Above all, they argue, HEFCE’s insistence that only those works that have been deposited in an open repository will be eligible for assessment for REF2020 (which directly affects faculty tenure, promotion and funding) is a requirement that no researcher can afford to ignore.
But could this be too optimistic a view? Dagmara Weckowska, a lecturer in Business and Innovation at the University of Sussex, believes it may be. While she does not doubt that the RCUK/HEFCE policies will increase the number of research outputs made open access, she questions whether they will be as effective as OA advocates appear to assume.
Weckowska reached this conclusion after doing some research earlier this year into how researchers’ attitudes to open access have changed as a result of the RCUK policy. This, she says, suggests that open access mandates will only be fully successful if researchers can be convinced of the benefits of open access. As she puts it, “Researchers who currently provide OA only when they are required to do so by their funders will need a change of heart and mind to start providing open access to all their work.”
In addition, she says: “Under the new HEFCE policy, researchers have incentives to make their best 4 papers accessible through the gold or green OA route (assuming that the REF again requires 4 papers) but they do not have incentives to make ALL their papers openly accessible.”
Further complicating matters, Weckowska points out that UK HEIs do not currently know how many research outputs their faculty produce each year, which would suggest that universities will struggle to ensure that faculty comply with the policies.
The conclusion would seem to be, therefore, that UK funders still have some work to do if they want OA to become the default for published research, both in terms of educating researchers about the benefits of open access, and ensuring that adequate compliance mechanisms are put in place.
And judging by a survey undertaken earlier this year by the publisher Taylor & Francis it would appear that there is still an urgent need to educate researchers in the specifics of what the mandates actually require of them. Only 30% of respondents to the T&F survey said they understood the RCUK policy, and many “appeared to be unsure whether the policy applies to them, since over half [55%] were unable to say whether or not their future articles would need to be published in accordance with the policy or not.”
The interview begins
RP: Earlier this year you gave a presentation on some research you have done looking at scientists’ motivations for open access publishing. This was focused specifically on the new RCUK OA policy and your conclusions were as follows:
· That the policy changes in the UK have been effective in strengthening pro-open access subjective norms among those who had little or no experience of open access publishing.
· That they were ineffective in promoting beliefs in the positive impact of open access publishing and creating the perceived behavioural control among those who had little or no experience of open access publishing.
· That they have shaken up confidence in the ability to provide open access among the researchers who had been publishing in open access journals before the changes in policies.
Can you talk me through the specifics of your study, and say why you reached these conclusions?
DW: Ok, let me explain the research process and the theoretical lens that I used to look at the data and draw conclusions.
The presentation that you refer to is based on a pilot project looking at broadly defined openness in biosciences. The project involved interviews with 22 Principal Investigators (PIs) aged approximately 40 to 60, who are working in the fields of systems biology, synthetic biology and bioinformatics and hold senior research positions in 11 higher-education institutions in the UK.
We conducted semi-structured interviews, lasting an average of two hours, at each institution between September 2013 and January 2014. This was a joint project between the Innogen Institute at University of Edinburgh and Egenis at the University of Exeter — team members at each university focused on different aspects of ‘openness’.
Part of our analysis drew on the extended theory of planned behaviour to understand motivations for open access publishing.
This theory assumes that whether we behave in a certain way or not depends on (1) our attitudes towards the particular behaviour (formed on the basis of our beliefs [so called behavioural beliefs] in the positive and negative effects of such behaviour), (2) our perceived control over performing such behaviour (formed on the basis of our beliefs [so called control beliefs] about ease or difficulty of performing the particular behaviour), (3) our subjective norms (formed on the basis of our beliefs [so-called normative beliefs] about social normative pressures, that is, what we believe significant others want us to do) and (4) our moral norms (formed on the basis of our beliefs [so-called moral beliefs] about whether a particular behaviour is the right or wrong thing to do).
The purpose of the analysis was to identify (behavioural, control, normative and moral) beliefs expressed by scientists during the interviews and identify differences between the beliefs of early adopters of open access publishing (those who had been publishing in open access journals before the changes in RCUK policies in 2012) and late adopters.
Differences in beliefs of early and late adopters illustrate differences in motivations for open access publishing. Slides 3 and 4 of my presentation portray the answers given to the question “Why have you made your publications openly accessible?” Slides 5, 7, 8, 14 provide all the beliefs expressed by early and late adopters during the interviews.
Please note that I made conclusions about the effects of policy change although I did not speak to scientists before and after the policy change. I made an assumption that late adopters did not hold any strong beliefs about open access prior to 2012 and that the beliefs that they hold now are the result of the new policy. I think it is a reasonable assumption and it was reported by many late adopters that open access just wasn’t on their radar before 2012.
RP: So you found that scientists’ beliefs about — and attitudes towards — open access and the RCUK OA policy are dependent on whether they were already making their papers open access before the new policy was announced?
DW: Yes. Late adopters expressed strong normative beliefs (they said that they now feel they should do open access because it matters to significant others — their employers or RCUK). Such beliefs were not expressed by early adopters. Thus late adopters are likely to have formed pro-open access subjective norms. This motivating factor is absent among early adopters.
On the basis of these observations one can conclude that the policy changes in the UK have been effective in strengthening pro-open access subjective norms among those who had little or no experience of open access publishing.
Late adopters did not express many (behavioural) beliefs in the positive effects of open access publishing as opposed to early adopters who believe that all sorts of good things will happen to themselves and to the world thanks to open access publishing. Thus it is unlikely that late adopters formed pro-open access attitudes while early adopters are likely to hold pro-open access attitudes based on their behavioural beliefs.
On this basis one can conclude that the policy changes were ineffective in promoting beliefs in the positive impact of open access publishing among those who had little or no experience of open access publishing.
Both early and late adopters expressed (control) beliefs about difficulties associated with open access publishing. On this basis, one can conclude that (1) late adopters are unlikely to have formed perceived behavioural control over open access publishing which could motivate them to publish open access and (2) the perceived behavioural control over open access publishing of early adopters has been shaken up (early adopters said it used to be easier to publish open access before 2012 when they had control over open access funds).
The second point of my second conclusion and the third conclusion that you cite above are based on these observations.
So these observations suggest that the policy changes were ineffective in creating the perceived behavioural control among those who had little or no experience of open access publishing, and they have shaken up confidence in the ability to provide open access among the researchers who had been publishing in open access journals before the policy changes.
All the written outputs of this pilot project are available online. We are working on an academic paper that will provide more detailed data and more in-depth analysis.
Not very optimistic
RP: To what extent do you think your findings can be generalised to other disciplines?
DW: One should not generalise these findings to other disciplines. Each discipline has its own publishing culture and the differences are stark between the sciences and the humanities. Bioscience, which was the empirical context of our study, has been one of the disciplines pioneering open access journals and open data movement.
Given the importance of copyright in arts and humanities, I would expect that researchers in these disciplines may be less motivated to provide open access to their work than researchers in, for example, biosciences. This however would need to be verified in another study.
RP: You said: “the perceived behavioural control over open access publishing of early adopters has been shaken up (early adopters said it used to be easier to publish open access before 2012 when they had control over open access funds)”. Am I right in thinking that this is because previously RCUK allowed researchers to pay for gold OA out of their research funds, but under the new policy they are no longer allowed to use these funds to pay article-processing charges but must instead request the money from their university?
DW: Yes, that is the reason. In the past researchers were allowed to budget for OA publishing in their research grant applications to RCUK and so were in control of spending. The changed funding mechanisms for OA publishing has shaken up the perceived behavioural control of the scientists who provided OA to their work for years.
This was certainly an unintended consequence of the changed funding mechanism. However, one should not jump to the conclusion that this change is bad. It could be a good change if it is well implemented.
In my opinion, the new funding scheme has significant advantages:
(1) under the old system scientists had to guess how many publications their research project will result in and in what journals they will publish before they started the research project. If they underestimated the OA fees they were left without funds.
(2) research publications are often written after the research grants are finished — that is when unspent grant-related money is inaccessible. In my opinion, if the universities develop fair, transparent and easy processes for allocations of OA funds, the feeling of control will likely come back.
RP: Would you say your research suggests that the RCUK policy will be successful?
DW: It depends on how you define “success”. The above research results are not very optimistic as they basically mean that late adopters will publish open access because they think they have to (e.g. when they publish the results of research funded by RCUK), not because they strongly believe that it is worth doing or the right thing to do.
Given that the late adopters also think that open access publishing is a bit of hassle, there is a possibility that they will choose open access licences only if they have to and continue to publish in a traditional way otherwise. HEFCE’s policy that sets the rules for future REFs addresses this potential problem to some extent.
No choice but to embrace OA?
RP: I think Green OA advocates view the situation more positively. They argue that as a result of the HEFCE policy OA is now a done deal in the UK, since researchers will have no choice but to embrace OA, and thus there will be 100% compliance. Are you less optimistic because you assume many researchers will choose green OA rather than gold OA, and so their papers may not be available with the RCUK’s favoured Creative Commons licence (CC BY), and only after an embargo? If so, green OA advocates would respond that this is good enough, so long as the embargoes are not overlong. They would add that embargoes can in any case be overridden by using the so-called “Request eprint Button” — where repositories provide a button to allow users to send an automatic request to the author(s) of a paper asking for a copy to be emailed to them.
DW: The point is that there is a possibility that the late adopters will choose gold (or green) open access only to show compliance to their research funders’ requirements and continue to publish in a traditional way (papers behind paywalls, no green OA) for the results of research that is funded by different means (e.g. university funds, industry). I would not consider this to be “good enough”. A green OA after a short embargo, I do consider adequate.
RP: I see. Your point is that researchers may not opt for open access at all, despite the RCUK and HEFCE policies?
DW: Yes. We did our research before the HEFCE policy was announced so the motivations of late adopters may have been strengthened by now. However, if the need to comply (with RCUK and HECFE) continued to be the only strong motivation for late adopters, one would worry whether all their research outputs will be made OA.
RP: I suspect open access advocates would respond by saying that since only those papers made available in an open repository will be considered for REF evaluation purposes, researchers will feel bound to make all their work freely available.
DW: I think people who believe, as you say, that “researchers will have no choice but to embrace OA” because of the HEFCE policy forget that academics need to submit only a few papers for REF (4 for the last REF). Good researchers publish much more than 4 papers in any REF period (~5 years) and they could strategically choose to make 4-6-? papers open access (green or gold) and keep the other papers behind a paywall.
The RCUK policy requires OA for papers resulting from research funded by RCUK. HEFCE policy requires OA for all papers submitted to REF — e.g. 4 papers per independent full-time scientist in the last REF period.
As I see it, there will be papers written from research that was not funded by RCUK (or other funders who require OA) and is not planned to be submitted to REF. Will late adopters make these papers (gold or green) OA? I would say it is unlikely, until they develop beliefs in the personal or societal benefits of open access publishing.
So I believe it is feasible that there will be near 100% compliance with the HEFCE policy — nearly all the papers submitted to REF will be accessible from repositories after an embargo period or from publishers under a CC BY licence. However, in my opinion, this does not mean that nearly all papers published by UK scientists will be open access.
Universities are bureaucracies
RP: In other words, because not all research papers published in the UK need to be submitted for the REF, and because not all the research published in the UK is funded by RCUK (or the Wellcome Trust etc.), your suspicion is that many papers could still end up stuck behind a paywall?
DW: Exactly. Under the new HEFCE policy, researchers have incentives to make their best 4 papers accessible through the gold or green OA route (assuming that the REF again requires 4 papers) but they do not have incentives to make ALL their papers openly accessible.
To submit their very best papers to the REF, scientists would need to provide OA to their first 4 papers published in a given REF period and then every paper that they feel is stronger than the ones that they already made OA.
RP: But does that not suggest that all their papers will therefore end up OA? And should we not assume that they will in any case want to make them all OA “just in case”?
DW: Scientists have their ways of judging the quality of a paper (e.g. on the basis of a journal’s impact factor [unfortunately] or the quality of a journal’s editorial board) and having gone through the last REF exercise researchers are aware of departmental strategies and criteria for a ‘REF-able’ paper.
So I believe that many researchers may develop strategies for identifying REF-able papers and making them open access and leaving other papers behind a paywall, particularly if the process of depositing papers in repositories is time-consuming (e.g. due to complicated repository systems or copyright issues).
However, if universities set up easy processes for depositing papers, or units dedicated to making all the institution’s papers accessible at least through the green route then it is more likely that 100% of UK scientific outputs will be openly accessible.
RP: The management and monitoring of OA policies, and ensuring compliance, is clearly going to be a big challenge. You may know that a recent Jisc report stated the following: “One of the most significant issues which arose during the TCoO [Total Cost of Ownership] data capture exercise was the fact that not one of the 24 universities involved in the project were able to tell the project the total number of APCs paid by their institution. In the vast majority of cases, the Library handled the RCUK and Wellcome funds and captured data on APCs paid from those funds. But none were able to establish how many APCs were paid from departmental/ faculty research project funds. A number of librarians asked the University Finance department for this information, but none were able to provide a figure. The speed that the data was collected at meant that institutions had very little time to compile data which they were not already collecting, which also meant that in some cases the data could not be fully checked for accuracy. In addition it was interesting to note that, despite a specific request, only one of the libraries was able to provide IPL [Information Power Limited] with the total number of articles published by their institution in a given academic year.” I wonder how likely you think it is that the units you feel are necessary will be created, and if they are how successful you think they might be — not just in terms of managing and distributing gold OA funds, but in ensuring compliance.
DW: Universities are bureaucracies and so take time to adapt to changes in the environment. The capability to manage open access publishing will not be developed overnight.
The problem increases when a university library does not have the spare capacity to take on the extra tasks and when the library staff lack the necessary skills and knowledge, e.g. legal knowledge of different copyright licence agreements.
In the late 1990s universities were expected to start collecting and reporting data about patenting, licencing and other forms of knowledge exchange with the business and community. At that time the UK government created a HEROBAC (now HEIF) fund to help universities develop the capacity to manage knowledge exchange activities, which led to the creation of many university technology transfer offices. New training organisations were also set up to train technology transfer managers and to share good practice across universities.
I think it should be assessed whether there is a need for a fund that would help universities to develop the capacity to manage open access publishing. Also the new training needs for librarians should be assessed and addressed.
No one knows
RP: You said: “I believe it is feasible that there will be near 100% compliance with the HEFCE policy — nearly all the papers submitted to REF will be accessible from repositories after an embargo period or from publishers under a CC BY licence. However, in my opinion, this does not mean that nearly all papers published by UK scientists will be open access.” I wonder if you have a sense of what percentage of the papers produced in the UK might therefore not be made OA. If you don’t have data on this, what is your gut feeling? Are we talking about a sizeable percentage of papers that could still end up behind a paywall, or a sufficiently small number that it will make little difference?
DW: I can’t answer that question. To make a guesstimate I would need to know what percentage of research is done in the UK with funding that has no OA requirements and how many papers are produced on the basis of such research. No one knows that.
If you are interested in looking into it you could check the percentage of research income received from different sources. Data on this can be purchased from HESA. However, looking at funders’ OA policies and the percentage of income received by universities from those funders, you would still not get a full picture because these statistics do not cover research activities done without external funding.
UK universities allow academics to spend a % of their time on research and publishing, for example, 40% — so two full days per week for research. Some academics do not have external funding for their research activities. I expect that the percentage of academics who do research without external funds will differ by discipline and type of university. Again, I don’t think anyone has an overview of how many papers are published on the basis of research that is not funded externally.
HEFCE’s policy on OA is currently the only way to encourage open access to papers produced on the basis of research that is not externally funded. But as I explained earlier, it would be possible to comply with HEFCE’s policy by making only a few papers OA in a given REF period.
RP: In a nutshell, then, your research suggests that if scientists do not have a personal belief in the value of OA then not even the combination of the RCUK and HEFCE (plus Wellcome etc.) policies is sufficient to ensure that all the papers published by UK researchers are made OA. For that to happen it will be necessary to engineer a change of hearts and minds within the research community?
DW: That is absolutely right. Researchers who currently provide OA only when they are required to do so by their funders will need a change of heart and mind to start providing open access to all their work. The future will show if the RCUK and HEFCE polices are enough to trigger such a change.
RP: What is your personal view about open access and the current UK OA policies?
DW: I personally believe that open access academic publishing can potentially help disseminate scientific knowledge around the globe and to a number of societal groups. For instance, journalists like yourself and civil servants will benefit from open access publishing, in my opinion.
However, I am not yet convinced that we will see local economic benefits (you may want to look at my presentation about Open Access and Innovation where I examine a set of assumptions underpinning the claim that open access will lead to more innovation and suggest that many of these assumptions are not yet supported by evidence).
I am also worried about the cost of open access publishing in the UK during the transition period, and the criteria that will be used by RCUK and universities to distribute open access funds (e.g. whose publications will be restricted because of limited funds).
Because of these worries I am yet to be convinced that the gold open access required by RCUK is a significantly better model than green open access model adopted in the US and promoted by HEFCE’s new policy.
RP: Do you think that researchers are more likely to opt for green or for gold OA in the UK?
DW: Our study did not examine in detail the differences in motivations for green and gold open access, so I cannot say whether researchers are more likely to opt for green OA on the basis of our findings.
RP: What do you see as the likely end game so far as OA is concerned, both in the UK and globally?
DW: I believe that a large proportion of scientific literature will be openly accessible in the foreseeable future, if the hearts and minds of scientists can be changed.
Open access policies have been introduced by public funders in many developed countries, including the US, the European Union, Australia, and in an increasing number of developing countries.
Right now it is important to monitor whether the policies are effective in stimulating a change in publishing behaviour and design other interventions if necessary. The publishing industry will be keen to maintain their business model while public funders will strive to maximize the benefits from research that they fund. Perhaps, some form of green open access will be the final compromise.
Dagmara Weckowska is a lecturer in Business and Innovation at the University of Sussex. She holds a PhD in Technology and Innovation Management from SPRU-Science Policy Research Unit at the Universityof Sussex. In the last 7 years she has researched the individual, organisational, legislative and regulatory aspects that affect the emergence of science-based innovation. More recently, while working as a research fellow at Innogen — the Institute for Innovation Generation in the Life Sciences at the University of Edinburgh, Dagmara started researching the relationship between open science and open innovation, including the issues of open access.