Contrary to what one might expect, not all the items in open access repositories are publicly available. Estimates of the percentage of the content in repositories that is not in fact open access tend to range from around 40% to 60%. This will include bibliographic records containing only metadata, plus full-text documents that have been placed on “dark deposit” — i.e. documents that are present in the repository but not freely available, either because they are subject to a publisher’s embargo or because the author(s) asked for the full-text to be deposited on a closed access basis. To enable researchers to nevertheless obtain copies of items that have been placed on dark deposit OA advocates developed the request eprint button. But how does the button work, and how effective is it? Below Eloy Rodrigues, Director of Documentation Services at the University of Minho, discusses the issues, and outlines the situation at UMinho.
RP: How many scholarly items are currently deposited in the University of Minho’s institutional repository RepositóriUM, and what are the growth rates?
ER: Currently we have more than 32,600 items in RepositóriUM, with around 5,000 being deposited yearly since the upgrade of our policy (effective since January 2011). Since 2011 more than 20,000 items have been deposited.
RP: Of these, how many are full text and freely available to the public (i.e. they are not metadata alone, not currently subject to publisher embargo, and not restricted to members of the university — as in requiring login)?
ER: Almost 26,000 (25,932) are freely available, which is more than 79% of the total.
RP: As I understand it, repository users can ask that a private copy of any document on dark deposit is made available to them by using the request eprint button built into the repository. In 2010 you co-authored a paper about this button, which was then more frequently called the “Fair Dealing” button. Your paper included data on “approval success rates” (i.e. the frequency with which authors sanctioned a copy of their work being made available to those requesting it). These data came from three universities: Southampton, Stirling and UMinho (your institution). The approval success rates were, respectively, 47%, 60% and 27%, with many requests simply ignored or lost. How has the situation at the University of Minho changed since then? What are the current figures?
ER: The overall response rate has remained basically the same, or even a little lower. In 2014 we had a global response rate of around 23%, with 21% sending the requested documents and 2% denying the request.
However the global response rate is highly “biased” by the effect of theses and dissertations. Theses and dissertations (T&Ds) account for around 21% of the total number of documents in RepositóriUM, and around 30% of the total number of restricted or embargoed access documents (currently around 6,700), but I estimate (based on some small “samples”) they represent far more than 50% (probably around 60% to 70%) of the requests received.
Because most authors of T&Ds don’t maintain any connection with the university after completing their thesis and dissertation, and they often change the email that was registered at the time the document was deposited in the repository (which is the email used to send the requests to authors), the T&Ds response rate is very low (probably lower than 10%), and that obviously affects the global response rate.
But we really don’t have data on this (we would need to “manually” look into the request logs we have, as we are not registering the document type from the requests) but based on some anecdotal evidence I estimate the response rate from UMinho members (professors and researchers) will be at least two times higher than the global average. So, excluding T&Ds, I “guess” the current response rate will be around 50%, or even a little bit higher (from 50% to 60%).
RP: In 2010 you made the following comment on a blog: “Our experience is that authors get ‘tired’ of replying to copy requests, especially when requests are very frequent. The consequence is that some start not replying at all, and others ask to change to open access articles/papers/theses there were in closed/embargoed access. We had more than 20 of those requests just on the last year…” Is that still your experience, or have author’s attitudes and behaviour changed since then?
ER: In the last couple of years I haven’t had regular conversations or feedback from Minho researchers about the copy requests, in the way I did in the first few years after the introduction of the button. But I know we still receive frequent (approximately on a weekly basis) requests to change the access status of closed/embargoed documents to open access.
RP: Presumably if a paper is on closed access as a result of a publisher embargo it is not possible to change the status to open access?
ER: Presumably yes. But there this a wide variety of behaviour from UMinho authors. While some are confident and fearless, others are fearful at the time of deposit, especially with papers published in journals or conference proceedings which do not have well formalised self-archiving/OA policies. Afterwards they tend to become less timid about their publications.
We inform authors about possible access permissions or restrictions to their deposited publications, but we respect their wishes about the access status.
RP: I assume most institutional repositories now have a request eprint button. But I think not all IRs implement the button in the same way. Can you talk me through the process at RepositóriUM once a user hits the eprint button? Is it fully automated, or is there some manual intervention? What happens behind the scenes when a user requests a copy of an item in the repository?
ER: The way we implement the process in RepositóriUM (and I assume it will be similar in other DSpace based repositories, as the request-copy addon to DSpace was developed here at UMinho) is the following: When users hit the button (actually it is a closed access logo) and fill in a form with their name and email (and an optional message), an automatic email is immediately sent to the author.
That message contains a token URL, directing the author to a RepositóriUM page, where there are two buttons – Send copy / Don’t send copy. After choosing one of the options another page is displayed with a template message, which can be edited by the replier. The final step is hitting the send button.
So, in summary, the text is always provided by the author (and not automatically or by the repository staff), and the process requires just 3 clicks, plus editing the reply message if the author chooses to do so.
RP: Advocates for use of the button believe that it is a much more effective way for researchers to get access to papers on dark deposit than, say, by directly emailing the authors. I note a paper published in PLOS ONE in 2011 tested the email approach. A group of researchers sent out a number of email requests for papers in the area of HIV vaccine research. The success rates they reported were between 54% and 60%, which is perhaps a little higher than the rates described in your 2010 paper. What do we make of that?
ER: I can only speculate about it. The button simplifies the process, both for the requester (who only needs to make two clicks and, if they want, customise a model message to the author) and for the author (who receives an email from the repository and just needs to make three clicks, and if they want customise a reply message). But maybe, at least for some people, this may appear completely impersonal and they prefer the more personal and human direct email contact.
That said, I’m not convinced that email contact will get a higher response rate than the button, and you cannot infer that from the PLOS paper. To test that hypothesis you would need to test both approaches for the same universe of publications and authors.
RP: The PLOS ONE study reported that two thirds of the papers (where the author responded positively) were received “on the same day or the next. However, the other third of respondents took on average 11 days to reply (median 3 days, maximum 54 days).” Do you have any information on turnaround time for those who use the button at UMinho?
ER: We just have data on the mean response time. In 2014 the mean response time was near six days for accepted requests, and 3.5 days for rejected requests. Again I think this result may be slightly biased by a higher response time from T&Ds authors, but that would need to be investigated.
RP: On March 2nd I tried to access a paper in RepositóriUM called “Academic job satisfaction and motivation: findings from a nationwide study in Portuguese higher education”. On trying to open the paper I was told that it was on restricted access and invited to request a copy of it, which I did. As the image below shows, I was informed that my request had been successful. However, I never heard anything further, and was left in the dark as to what had happened to my request. It is not a very user-friendly system is it? Might not most readers be inclined to give up after even a couple of such failed attempts to get a paper?
ER: Yes, I recognise that. It is not very user friendly, and people may be inclined to give up after a couple of “non-answers”. We’ve focused the development of the addon on making it very easy and simple to use by external readers and especially by UMinho authors.
At the time of development we really didn’t consider the issues around monitoring, reporting, collecting statistics on the use of the button, or providing feedback to requesters. And after the initial development we have really just made some minor improvements/adjustments (like spam control through a captcha feature) and upgraded it to the newest DSpace releases.
RP: My experience with the ORBi repository at the University of Liège was somewhat different. I tried the button there twice. On both occasions I received the full text (or a link to it) within 24 hours. Paul Thirion, Head librarian at the University of Liège, reports that the approval success rates for requests made using the button built into the ORBi repository are higher than average, ranging from 67% in 2009 to 81% in 2014. Do you have any sense of why Liège is more successful at getting researchers to approve eprint requests than other universities?
ER: I really don’t know. I imagine that, apart from some subjective aspects (like cultural and organisational differences and/or a different relationship to and perception of open access and the institutional repository between researchers at Liège and Minho etc.), there are some objective factors to explain it: probably the T&Ds effect is not present at ORBi, and I can speculate that there is a difference in the percentage of closed/embargoed access documents in ORBi (which I think is higher than in RepositóriUM), and maybe there is also a lower percentage of documents for which the access status is changed to open after deposition. [RP: Paul Thirion reports that around 62% of the documents in ORBi are full-text].
To what end?
RP: The paper you co-authored in 2010 goes on to say, “Given a significant number of button requests which are ignored or lost, one might be tempted to assume that it has not worked. However, this is not true. The principal impact of the Button has been to enable the adoption of institutional IDOA mandates.” This left me wondering as to the point of the button. I had assumed the sole purpose was to ensure that those who want access to papers under publisher embargo can nevertheless obtain a copy of them. For instance, in commenting on the open access policy being introduced by the Higher Education Funding Council for England Stevan Harnad described the purpose of the button as being to “tide over the usage needs of UK and worldwide researchers for the deposited research during the allowable embargo.” Your paper, however, suggests that the objective is rather to encourage funders and institutions to introduce OA mandates. What are your views today on the purpose of the button?
ER: I think the introduction of the button had both the immediate and practical objective of providing access to papers which were deposited with temporary (embargo period) or definitive access restriction, and the more strategic objective of helping in the introduction of mandates (by creating a mechanism that allows mandating universal deposit, regardless of eventual access restrictions, while offering a “second class” access procedure).
In my opinion both purposes remain important today.
RP: How would you describe the success of the button today, and what do you predict for its future success?
ER: I don’t know what the global response rate to the button requests is. But even if it is closer to the UMinho 50% estimate, than the Liege 80% result, it means that tens or hundreds of thousands of papers were made available to many readers that otherwise would not have access to them.
So, I think the button is relatively successful, both in actually providing access to closed/embargoed access publications and in helping institutions and funders to define self-archiving mandates, without pushing themselves into spending yet more money by paying APCs, on top of their subscription costs.
For the immediate future, I predict the button will remain useful and hopefully more successful, as the number of mandatory polices, as well as embargoes, grows.
RP: One thing I find striking is that advocates for the button seem to have done very little research into its efficacy. Why do you think that is?
ER: I can only reply for myself and for UMinho’s RepositóriUM. I think the first reason is that our main focus is on managing and running the repository as a critical service of the university, with limited capacity to do research and development. So we use that limited capacity for very practical and applied developments and not on “non-applied research”.
The second reason is that, despite being important and useful, the button is not on our top three priorities for work on the repository. We’ve devoted much of our efforts on improving the repository interoperability and integration with other services/systems, on facilitating and simplifying the deposit/self-archiving of publications into the repository, on collecting and providing usage statistics to authors of publications in the repository, on guaranteeing/improving repository visibility in the global search engines (especially Google), etc. All those issues have higher strategic relevance for us given the current state of policy implementation and repository development at UMinho.
RP: Do you think there is a danger that if the button were to prove too successful publishers might seek to curtail or prevent its use in some way?
ER: I don’t think so. It is at least very questionable that publishers would have any solid legal ground to act against the button use, and, on the other hand, it would give them very bad publicity. So, from a cost-benefit point of view, I think the button is not a high priority for publishers either.
RP: Thank you for taking the time to answer my questions.
I am currently working on a longer document about dark deposit and the request eprint button. As such, I would welcome people’s thoughts about and experiences of these two things. I can be contacted here.