Citation Advantage?

Citation Advantage?

1

You might think that after 20 years of research and more than 130 studies on the subject, we’d have a clear picture of the effect that open access publishing has on an article’s citation performance. Unfortunately, decades of poor studies and a mystifying unwillingness to perform experimental controls for confounding factors continues to muddy the waters around the alleged open access citation advantage (OACA).
 
In a new paper published in PLOS ONE, authors from the University of Minnesota Libraries attempted to perform a meta-analysis of the 134 studies they could locate on the subject. But to be valid, a meta-analysis must look at comparable experiments, and because the OACA studies were so heterogenous, this proved impossible. Definitions of “open access,” fields of study, time periods studied, etc. were all over the place, negating any possible conclusions that could be drawn.
 
While the authors note that the majority of these studies do indeed find an OACA, science is not a “majority rules” situation where the quantity of studies drawing one conclusion outweighs the quality of the studies performed. As Phil Davis wrote on the subject back in 2010, “If science were a matter of tallying consenting and dissenting views, the world would still be flat, the sun would still be revolving around the Earth, and Thalidomide would still be prescribed as a sedative for pregnant women.”
 
The authors make clear that, “Because of the limitations of the quality of the studies in our review, it is not possible to draw definitive conclusions and recommendations for authors deciding whether to make their work OA.” But there are several striking findings that should play an important role in how we think about the OACA.
 
First, only 2 of the 134 studies analyzed were randomized controlled trials (RCTs). Often referred to as the “gold standard” for experimental evidence, they offer significant advantage over observational studies in that they reduce bias and the effects of confounding factors beyond what is specifically being studied. Both of the RCTs found no citation advantage associated with OA publication.
 
Further, of the 134 studies examined, only 40 acknowledged the possibility of confounding factors, which is fairly astounding. As early as 2007, it was clear that things like selection bias (e.g., OA articles may overrepresent well-funded researchers; authors may save their precious OA funds and only spend them on their “most important” articles) were likely to invalidate any correlations that could be drawn. Yet even among the 40 studies that acknowledged potential confounders, only a subset made any effort to control for those factors. This is just bad science.
 
Regardless of the high-quality evidence arguing against the OACA, it remains a matter of faith for many. It has also led to decades of ethically questionable behavior, with promises being offered to researchers that they can essentially buy citations by paying for OA publication. Though this promise is likely to be snake oil, if true it would serve as yet another unintentional inequity that author-pays APC OA models create – the rich get richer because they can afford to pay for advantages that less well-funded researchers can’t. 

—
Source: Nature, PLOS ONE, The Scholarly Kitchen, Journal of Infometrics

Professional and Academic Publishing

2

As readers of The Brief are surely aware, the Plan S “Transformative Journal” designation allows funds from cOAlition S funders to be used for OA fees (article processing charges, or APCs) in hybrid journals with the designation. (Plan S otherwise frowns on hybrid journals—leaving to the side “transformative agreements,” of course.) In order to be designated as a Plan S Transformative Journal (TJ), publishers must agree to (try to) show an “annual increase in the proportion of OA research content of at least 5% points in absolute terms, and at least 15% in relative terms, year-on-year.” The journal must also publicly agree to flip to OA when 75% of its research content is published OA. 
 
A recent analysis of Plan S Transformative Journals by Delta Think shows that of the 2,250+ approved TJs, only a dozen have met cOAlition S’s target OA content growth rates over the last three years. Very few journals are on track to come anywhere close to the proposed 75% OA content point at which the journal is supposed to flip to fully OA by December 31, 2024, when funds for TJs are scheduled to be phased out. The report suggests that TJs’ OA content proportion is growing at about half the rate needed to be in compliance and is not that different from OA uptake seen in non-TJ hybrid journals.
 
To be fair to this cohort of TJs, the “Transformative Journal” designation did not exist until last year and even then most papers accepted under the TJ status would not be published until the current year (2021). A three-year retrospective analysis therefore is not necessarily representative of the future trends for these journals. But then again, in practical terms, journal publishers are not really in control of author decisions. In a hybrid journal model, authors can choose whether to publish their papers under an OA model or under a subscription model. While some funders, such as those of cOAlition S, mandate the OA route for research they fund, others (such as the federal agencies in the US) do not. If a journal does not publish an increasing number of papers funded by funders with mandates (according to Delta Think, cOAlition S funders fund about 5.2% of global research), it is not clear what publishers can do to meet Plan S’s TJ targets. 
 
That said, it is not clear that actually meeting Plan S TJ targets is really anyone’s goal. From a publisher’s perspective, becoming a TJ falls into the “yeah, why not?” category. As there is no penalty for failing to meet growth targets or not flipping to fully OA at term’s end, the TJ route makes publication in one’s journal available for authors with funding from cOAlition S, offering some much-needed breathing space to test new OA strategies. From the Plan S perspective, the TJ designation provides some quantifiable goals, benchmark data, and statements of intent from publishers. If publishers don’t meet the targets but do make some progress and test some new OA models, the cOAlition S funders can declare victory and come up with modified targets to keep pressure on publishers.

—
Source: cOAlition S, Delta Think

3

Transformative agreements (Publish and Read) pose a challenge for independent society publishers. Indeed, the rise of transformative agreements is the reason most cited by independent publishers (that C&E works with) for seeking a publishing services agreement with a commercial publisher. And so it was with keen interest that we read a new report, How to Enable Smaller Independent Publishers to Participate in OA Agreements, by Information Power. 

In reading the report, we were surprised (though given the report’s sponsors, perhaps we shouldn’t be) by its regional view, a limitation the authors acknowledge. This is very much a report about the problem of transformative agreements and independent society publishers in the UK and Ireland, with some applicability to the EU. The challenges of navigating transformative agreements in the North American market, or other regions of the world, are not addressed.  

One of the ways in which North America in particular differs is concentration – concentration of authors and their institutions and concentration of funders that are interested in transformative agreements. The report, for example, notes that: 

A single [transformative] agreement with an institution is much easier for a smaller independent publisher to administer than many article transactions, unless of course each library or consortium wants a different sort of agreement. 

The problem is that many independent society publishers publish journals with relatively smaller numbers of papers. A mid-sized society journal might publish 500 papers in a given year (many publish less). These papers will originate from authors at many hundreds of different institutions. There are over 250 research universities in the US. There are many more colleges and medical schools beyond that. And many more universities outside the US. The notion that an independent society journal will have “many article transactions” to administer with a given university is not a given. Perhaps this is true in some fields where there is a high concentration of researchers at a small number of universities. But for many journals, it is far easier to manage a couple of article processing charges (APCs) than to negotiate a complex transformative agreement with any given institution. More importantly, it is far easier for the institution. Why would an institution spend the time to negotiate a transformative agreement with a journal with which their affiliated researchers publish a handful of articles? It just is not worth the time for either the institution or the society. 

The other problem is that by signing a transformative agreement, the university is effectively shifting the costs of publication from the funder to the university (this is less true in fields where there is not a lot of research funding). The report’s authors acknowledge this point and suggest that funders either allocate block grants to libraries or “communicate to universities that 1%–2% of grant funding received should be allocated to the library to ensure OA compliance, including via OA agreements.” Leaving to the side the particularly contentious matter of institutional overheads, the larger issue is that few research funders provide sufficient funds to research universities to dictate broader open access and library policies. This might work in the UK where a smaller number of funders (such as JISC) account for a higher concentration of funding. In North America, funding (excepting that of the federal government) is too diffuse. The University of California’s “multipayer” model (see Issue 32 of The Brief) is a more practical approach to this vexing problem – but whether it will bring sufficient funder monies into such deals remains to be seen. 

Unfortunately, the regional orientation of this report identifies little in the way of strategies to push against the drift of independent societies (at least those outside the UK and Ireland) toward agreements with commercial publishers. While transformative agreements have much to commend them (they eliminate fees for authors and provide OA funding for researchers without grants that cover publication fees), they further privilege the largest publishers. It is not clear how the strategies employed to date by JISC and proposed by Information Power to help independent societies might be adapted in other regions that have very different dynamics. That is a report we would like to see.

—
Source: Information Power

4

A new report from ITHAKA S+R looks at the consequences of canceling journal access (especially “Big Deals”) for both researchers and libraries. The report is largely what you might expect, but it’s nice to see that ratified. When a Big Deal is canceled, researchers find other ways  to access papers (or find alternative papers), and this comes with a cost in the form of inconvenience and time lost. The report’s authors (Danielle Cooper and Oya Rieger) draw an interesting conclusion: As is the case with many other advocacy efforts intended to (ultimately) lead to more equity and access to scholarly literature, inequities are unintentionally created. In this case, more-established researchers (those with better networks of colleagues and those more in demand for positions as peer reviewers and editorial board members) are much better able to maintain access to journals after a cancellation than junior colleagues who are not as well connected. Further, by driving researchers into alternative access routes (asking colleagues, quasi-legal social networks, and piracy websites), libraries lose a significant amount of data about the information needs of those on campus.

There are big picture concerns for other stakeholders here as well. For funders and policymakers, the report serves as a good reminder of how unpopular the author-pays APC model for open access is with researchers. The interviews here show how APCs are perceived as a zero-sum game: for every article published via an APC model, it becomes that much more difficult to fund a graduate student’s project or pay a postdoc’s salary. For publishers there’s a lesson in figuring out how to balance non-subscription access to articles (“leakage”) in order to stay relevant and on mission with the risk of alienating the research community by driving inconvenience as a negotiating strategy.

The most cautionary message is for the libraries, and the continuing diminishment of their role on campus. Researchers believe the library’s primary role is in provisioning  content. Content discovery support seems to have long ago been ceded to the likes of Google Scholar:

The diminished role of the library as a locus of discovery underscores the library’s weakening ability to intervene in the scholars’ workflow, and by extension, its lessening influence on how sources are delivered. 

As libraries reduce their traditional focus on provisioning content (due to both budgetary constraints and the continuing drive toward open access), their importance on campus will continue to decline. As the report suggests, a move away from collecting content and toward the purchasing of author services makes it less logical for the library to be the key campus entity negotiating procurement.

—
Source: ITHAKA S+R

5

The State of Maryland passed a new law that requires publishers to license ebooks to public libraries on “reasonable terms.” Similar legislation looks likely to be signed into law in New York. “Reasonable” is not defined in either case, which means it is highly likely there will be lawsuits over the parameters of this term. The Association of American Publishers appears to be preparing to challenge the new laws on grounds that (per the organization’s March 24 testimony opposing the Maryland bill) federal copyright law supersedes state laws (among other finer legal points). But we wonder if there is another term that will also give some publishers heartburn. These laws will apply to “public” libraries. Is the University of Maryland’s library a public library given that the university is public? What about the SUNY system? We think of “public libraries” as distinct from “academic libraries” but is that a legal distinction? We will be watching the rollout of these two laws (and potential legal challenges) closely.

—
Source: Library Journal, Publishers Weekly

Dealmaking

6

“Interest rates remain low and there is a lot of capital looking for somewhere to park itself” remains perhaps the best explanation of the ongoing wave of dealmaking in the industry. Said wave is likely helped along by the expectation that US capital gains rates will be higher in the near future: if you are thinking about selling your company or cashing out of your investment, now is a good time both because of frothy valuations and historically low capital gains rates. In other words, cyclical macroeconomic factors, as opposed to industry dynamics, appear to be the main drivers of mergers and acquisitions announced this month (as well as the ProQuest acquisition, which we discussed last month in The Brief).
 
McGraw-Hill announced that Platinum Equity will be acquiring the company from Apollo Global Management. The deal announcement follows the dismissal of a lawsuit brought against McGraw-Hill – as well as Cengage, Pearson, and others – arguing that the inclusive access model of textbook provision does not adequately allow students to opt out. 
 
Kanopy announced that it will be acquired by OverDrive. Kanopy’s “Netflix for institutions” streaming video service has proven a hit with many library patrons (if not the library, given the potential for “cost-ballooning” the Kanopy business model can create). Students, faculty, and other patrons with a library card can download the Kanopy app and watch classic and independent films, documentaries, and other videos through the service (just like Netflix but the library pays for it). While Kanopy and OverDrive are similar in that they both provision digital media to libraries (video in the case of Kanopy and text in the case of OverDrive), the technologies and business models are different. We also do not know what the customer overlap is between the two services. That said, the cross-selling opportunity is likely significant as is the potential to consolidate sales, marketing, and corporate support (finance, HR, IT). 
 
EdEx will be acquired by 2U for $800 million in an all-cash deal. The online learning platform started by Harvard and MIT is being sold a few months after rival Coursera went public at a stratospheric valuation of $7 billion. Which raises the question as to whether $800 million for EdEx is a steal? The deal gives 2U a brand that is better known than its own, an archive of course offerings from Harvard and MIT, and access to the audience for some of the world’s most renowned courses (an audience 2U can then pitch other offerings to). 
 
BiblioLabs has been acquired by Lyrasis. The amount the not-for-profit library services organization paid for BiblioLabs was not disclosed, but the announcement did say this: “The acquisition is complete effective June 15, 2021 and was funded through cash purchase and charitable contribution by the owners of BiblioLabs.”. 

People

7

Annette Thomas has left her role as Chief Executive at The Guardian after just 15 months on the job. The cause of the departure is reported to be a clash with the paper’s editor. Thomas was previously CEO of Macmillan Science and Education, which includes the Nature portfolio of journals. If The Guardian is looking for someone else to not work at The Guardian under similar terms, please let us know! 

Briefly Noted

8

There was much chatter about Oxford University Press shutting its printing operation, ending a tradition going back over 400 years. It should be emphasized, however, that OUP will continue to offer its books and (some of) its journals in print. This is less a story about the end of printing (at OUP or elsewhere) and more a story about consolidation in the printing business. 

The New Yorker has published a thoughtful profile of “image detective” Elisabeth Bik, who has developed an uncanny ability to spot image manipulation without the aid of software. The profile also – unusually for a nontechnical magazine – does a good job of describing the problems Bik (and others in the industry) are trying to solve. In related news, thousands of scientists have signed an open letter in support of Bik, who is facing legal action from two researchers in France related to Bik’s criticism of the researchers’ published study on hydroxychloroquine as an effective therapy for COVID-19.

One-word article titles: a power move. 

Dan Rather taking Jon Stewart to task for a bad take on science was not on our 2021 bingo card, but there you go. (We find it charming that Dan Rather has a Substack.) 

A study published in JAMA Internal Medicine finds lack of diversity among leading biomedical journal editors. From an interview with lead author James Salazar: “It’s important to emphasize that our study found a lack of diversity despite the presence of widespread stated commitments to promote diversity among these journals.”

Cambridge University Press launches its Subscribe to Open model for books, which they have dubbed “Flip It Open.” 

Tim O’Reilly writes about how we live in a world with two economies that operate with two sets of rules (read the whole thing):

Most ordinary people live in a world where a dollar is a dollar. Most rich people live in a world of what financial pundit Jerry Goodman … called “supermoney,” where assets have been “financialized” … and are valued today as if they were already delivering the decades worth of future earnings that are reflected in their stock price.

Steven Sinofsky offers a deep dive into the impacts that the pandemic will have on the structure of corporations and how we work. Sinofsky, the former head of Office, and then Windows, at Microsoft, is always worth listening to. He suggests the pandemic as a generation-defining event, much along the lines of World War II in terms of its impact on how corporate culture evolved. Expecting things to return back to “normal” or hoping that a few small changes can be made around the edges is a recipe for failure “in the face of step-function change.” Along similar lines, Sinofsky’s Twitter thread on the performative nature of being physically present in an office creates a clear contrast between innovation and our new favorite phrase: “collaboration theater.” 

OurResearch announced it has been awarded a 3-year, $4.5 million grant from Arcadia. The grant will support ongoing development of Unpaywall and Unsub as well as two new services: a free and open database of journals and a bibliographic database designed to replace Microsoft’s Academic Graph. 

The University of North Carolina Board of Governors has decided not to reappoint Law Professor Eric Muller to the UNC Press Board of Governors. Muller suggests this decision is retaliation for his criticism of UNC’s handling of issues related to race and history. There are echoes here of the much higher profile controversy surrounding UNC’s decision to (initially) offer Nikole Hannah-Jones a prestigious endowed professorship without tenure. (Hannah-Jones has now decided to walk away from UNC and will instead join the faculty of Howard University).

Tim Berners-Lee sold the original World Wide Web code as an NFT for $5.4 million at auction. Is this a ridiculous amount of money to pay or not nearly enough?

Mike Shatzkin argues that “enterprise self-publishing” (meaning books developed by organizations) is on the cusp of ushering in a disruptive wave of book publishing. We are not convinced but it is a thoughtful and provocative read. 

eLife announces it will peer review preprints in clinical medicine on the medRxiv platform. The reviewer comments will be made public on medRxiv. We don’t really understand this (why is a life science journal reviewing clinical medical preprints it is not necessarily even publishing?) but OK.

Hear, hear for rapid retractions: “Scientific publishing needs to stop treating error-checking as a slightly inconvenient side note and make it a core part of academic research.”

Princeton University Press launches a speakers agency for its authors. Great idea, terrible name. 

Paper retracted not because of a fault in the research but because of disgruntlement over author order.

***
You can have data without information but you cannot have information without data. ―Daniel Keys Moran