Stony Brook University Author Perspectives on Article Processing Charges

INTRODUCTION The purpose of this study is to gain an understanding of Stony Brook University (SBU) author perspectives on article processing charges (APCs). Publishing an article without restrictions, also known as open access publishing, can be a costly endeavor. Many publishers charge APCs ranging from hundredsto thousands of dollars to publish an article without access restrictions. Authors who cannot obtain funding from grant agencies or their institution must pay APCs on their own. Do APCs fundamentally impact how authors choose their preferred publication venues? METHODS A cross-sectional survey was designed to learn SBU author perspectives on, and concerns about, APCs. RESULTS Responses mainly came from the sciences. Many SBU authors preferred to publish in a prestigious journal or journal of their choice rather than in an open access journal. Most authors published their articles in open access journals even if they were required to pay APCs. Many authors found that it was difficult finding funding for APCs and some expressed their concerns about the double charging practice. DISCUSSION SBU authors might believe that publishing in established and prestigious journals could secure their career’s advancement. Authors who chose to pay open access journals with APCs might be following publishing criteria. Libraries can encourage authors to negotiate with publishers to obtain a discount or waiver of APCs, when possible. Institutions should negotiate shifting journal subscription costs toward hybrid open access publishing. CONCLUSION Data will be used to inform how the SBU Libraries can help authors locate funding opportunities for APCs.


INTRODUCTION
The Open Access (OA) movement changed the scholarly publishing landscape in many revolutionary ways. Governments, research institutions, and funding agencies have started requiring scientists who receive public funding for their research to publish their findings as open access (Pavan & Barbosa, 2018). As publishers require "review and publication costs" to provide free public access to scientific publications, various journal business models have emerged (Sotudeh & Estakhr, 2018, p. 564).
The most significant models include "Green OA model," "Subsidized Gold OA," "Mandatory Gold OA or pure Gold OA," and "Hybrid Gold OA model or author-pays" (Sotudeh & Estakhr, 2018, p. 565). Among these models, "Green OA model" and "Subsidized Gold OA" do not require an author fee while "Mandatory Gold OA or pure Gold OA" and "Hybrid Gold OA model or author-pays" require authors to pay a publication fee (Sotudeh & Estakhr, 2018, p. 565). This fee is known as an article processing charge (APC) which publishers impose on authors for journal article publications at no cost to readers (Beasley, 2016;Reinsfelder & Pike, 2018). There are objections to APCs especially by young researchers who cannot afford to pay a fee for publication (Suber, 2003). However, Suber (2003) argued that authors can pay for APCs through funding agencies, institutional memberships with publishers, or can obtain financial hardship waivers from publishers such as BioMedCentral and Public Library of Science.
Stony Brook University (SBU), which is located on Long Island approximately 60 miles east of New York City, is a major research university within the State University of New York (SUNY) system. According to The Carnegie Classification of Institutions of Higher Education ® (n.d.), SBU is categorized as "Doctoral Universities: Very High Research Activity," or R1 research university. In 2017, the SBU Senate adopted an "opt-out" open access policy proposed by the SBU Libraries (Stony Brook University Libraries, 2020). As SBU faculty and researchers are heavily involved in publishing, the authors of this article sought to understand their perspectives on APCs and to gain insight into how they are funding these fees in pursuit of open access publishing.

LITERATURE REVIEW
The rapid growth of open access journals has challenged the fundamental question of making articles freely accessible in the face of high publication costs. Author publication fees are not new, however. To lower subscription costs in traditional publishing, society publishers would collect page charges from authors (Solomon & Bjork, 2012;Haynes, 2012). However, this practice declined by the late 20th century (Haynes, 2012, p. 153). To the contrary, commercial publishers have rarely used page charges as a funding source (Tenopir & King, 2000). This was, in part, because commercial publishers in the 1960s and 1970s depended on journal subscription fees which enabled authors to publish for free (Haynes, 2012).
Charging authors for publication resurfaced in 2000 from the publishers BioMed Central (BMC), Public Library of Science (PLOS), and Hindawi (Solomon & Bjork, 2012;Bjork & Solomon, 2014). Twenty years later, charging APCs for open access publishing is now a common practice. For instance, PLOS explains on their website the need to collect APCs, "PLOS offsets publication expenses-including those of peer review management, journal production and online hosting and archiving-by charging an Article Processing Charge (APC), to authors, institutions or funders for each article published" (PLOS, n.d.). Similarly, Springer (2005) stated in a press release that "the 3000 US dollar fee covers the costs of Springer's publishing service -including a parallel printed version of the article in an established journal" (para. 6).
A Hybrid Gold OA model or hybrid journal is a subscription-based journal that offers authors the option to publish their articles open access by paying an APC (Solomon & Bjork, 2012;Pavan & Barbosa, 2018). The publishers then allow free access to the journal section paid for by the authors (Sotudeh & Estakhr, 2018). Yet this hybrid journal practice raises a concern of "double dipping" (Bjork & Solomon, 2014, p. 93;Cook & Smith, 2017, p. 2) that warrants additional study.
Most gold open access journals charge APCs ranging from a few hundred to a few thousand dollars, with an average of $1500 to $2500 per article (Haynes, 2012). The practice of APC-funded open access journals originated in Europe (Cook & Smith, 2017), most notably in biomedical research which has substantial grant funding support (Willinsky & Rusk, 2019). However, open access publishing does not work as well for research in the social sciences and humanities disciplines (Eve & Willinsky, 2015;Eve, 2017). There is a resistance to APCs from authors in these fields, especially when researchers feel "economics are not in their favor" (Eve, 2015, p. 74).
In September 2018, The European Commission's Plan S was launched requiring results from public, grant-funded research "be published in compliant Open Access journals or platforms" (cOAlition S, 2019). Plan S proposed to end subscription journals and make scientific publications publicly available in 2020; it also aimed to eliminate hybrid journals that charged both subscription fees and author APCs to make the articles open access (Enserink, 2018).
In response to the Plan S proposals, universities and publishers voiced concerns about the APC fee structure and the elimination of the hybrid journal model, among others. Kowaltowski and Oliveira (2019) pointed out that the "capped fee structure" was not feasible in Brazil as APCs were deducted from "ongoing grant totals, meaning authors must choose between open access or lab materials" (p. 461). The American Chemical Society argued that price caps on APCs could result in increased publishing costs and thus, could impact an author's options for their preferred journals (Milne, 2019). Hybrid journals played an essential role in the "scientific ecosystem" allowing authors to publish in "top journals" (Shrier & Schmid, 2019, p. 461;Milne, 2019). Eliminating the hybrid journal model could restrict a researcher's choice to publish in journals that were appropriate for their scientific discoveries (Shrier & Schmid, 2019;Milne, 2019). Nevertheless, Plan S pushed back, aiming to achieve their goals in 2021 instead of 2020 (cOAlition S, 2019).
On the academic library standpoint, using the library collection budgets to support APCfunded open access research has diverted the library's intention to "provide access to community-supported resources versus supporting the self-interested economic decisions of individual article producers" (Scott, 2018, p. 264). A number of academic libraries secure separate funds for APCs, but this practice may not be sustainable as costs continue to rise (Reinsfelder & Pike, 2018). Crowdfunding, sponsorship models, and other alternative projects could be a way for libraries to support open access publishing (Reinsfelder & Pike, 2018).
Going forward, it is intriguing to envision how APCs will evolve in the realm of open access and how libraries and open access publishers will respond to the APC model in coming years.

METHOD
The authors designed a 13-question survey, including two demographic questions and 11 content questions with five options for further comments. The survey instrument for this study was Qualtrics and the full survey took approximately 4 minutes to complete. All survey responses were anonymous, except for demographic data such as academic department and employment status. The authors defined "SBU authors" as any author who published at SBU. Respondents were asked to select their SBU occupation (Faculty, PhD, Post Doc, Professional staff, Researcher, …) to ensure only university researchers responded.
The authors submitted the survey proposal to the Institutional Review Board (IRB) for approval. After receiving approval from the IRB, the survey was opened and distributed from April 22, 2019 to May 10, 2019. The survey was disseminated in a variety of ways including the Library's news blog (April 22), weekly campus announcements (April 22, April 29, and May 6), and a campus-wide email (April 25). Along with the survey link, an advertising script starting with "Have you published in academic journals? If so, we need your help" was included to prompt SBU authors to take the survey. Two hundred and thirty-four responses were received during the 3-week-survey period. Data was exported to Excel for further analysis and visualizations.

RESULTS
At the time of the survey, SBU had roughly 2700 faculty and 8734 graduate students (Stony Brook University, 2019). The survey was opened 234 times. The first question asked respondents if they agreed to participate in the study, to which 220 responded "Yes". Those who responded "No" were brought to the end of the survey. Data was analyzed from surveys with at least two responses beyond the initial agreement to participate. These stipulations reduced the measurable data to 183 surveys out of 220. These survey results served as a starting point for the SBU Libraries to examine the complexity of APCs.

Is open access publishing still preferable when the author must pay the APC?
Most participants disagreed when asked if they would still publish their articles in open access journals if required to pay the APCs on their own. The total number of responses went down slightly for this question to 179. Participants mostly selected "Disagree" (n=57, 31.84%), followed by "Strongly disagree" (n=51, 28.49%), "Neither agree nor disagree" (n=35, 19.55%), "Agree" (n=30, 16.76%), and "Strongly agree" (n=6, 3.35%). See Table 1.
7.1. Please share your experience if you answered "Yes" to 7 For participants who were confronted with APCs, the following survey question asked if they had the opportunity to negotiate the price. Out of 177 responses, most responded "No" (n=136, 76.84%). The 41 respondents that selected "Yes" to negotiating APCs were invited to leave a text answer to explain their process. A total of 28 respondents provided more information about their APC negotiation. These responses were brief. Most came from participants with no funding and thus experienced waived or reduced APCs. Successful negotiations also happened if the author's institution subscribed to the journal they were publishing in. Negotiations were also considered if an article was under a certain page count, the author was on a limited budget, the article was to be published in a special issue, or the author was personally invited to publish in the journal.

Who should be responsible for APCs if grant funding is not available?
Next, participants were asked who they thought should be responsible for paying APCs if grant funding was not available. Of 169 responses over half answered the "Author's institution" (n=104, 61.54%) followed by "Other (Please specify)" (n=24, 14.20%), the "Author's academic department" (n=20, 11.83%), "The author" (n=15, 8.88%), and the "Author's library" (n=6, 3.55%). A text field was available for those who answered "Other (Please specify)." See Figure 5.  Everyone who chose "Other (Please specify)" shared similar experiences in their text responses. Most responses stated the author should have minimal monetary responsibility.
One stated the APC model should not exist and Green OA is preferred. Some viewed their institution as responsible, stating if institutions expected their researchers to publish in open access journals, then they should "step up" to support fair and equitable access to scientific literature by funding APCs for open access journals, and make "every effort" to pay APCs specifically for researchers who contribute greatly to the research output of the institution. One suggested that APCs be shared among the author, department, and institution since all of these stakeholders are listed in the publication. A few participants suggested libraries and institutions should pay APCs instead of subscription fees to "greedy academic publishers," as one faculty member put it. In general, most responses declared that the author's institution or the dean of the school/department should be responsible for APCs. Surprisingly, most agreed the library should be the least responsible party for paying APCs.

Author interest in learning more about APCs from the library
Participants were asked if they would like the library to be a resource for APC information. Of 169 responses, 81 authors believed the library probably or definitely could be a place to learn more about APCs. The largest response to this question was "Probably yes" (n=43, 25.44%) followed by "Might or might not" (n=41, 24.26%), "Definitely yes" (n=38, 22.49%), "Probably not" (n=34, 20.12%), and "Definitely not" (n=13, 7.69%). See Table 1.

If yes, please provide comments
The final survey question asked participants if they had any other comments on, or concerns about, APCs. If they selected "Yes" they had the opportunity to leave a text response. Out of 169 responses, 45 (26.63%) responded "Yes." Text responses were received by 42 of the 45 who responded affirmatively. Many of these responses touched upon the participants' personal experience with publishing. One participant pointed out that open access allowed "resource-poor researchers" to gain better access to published materials but that APCs could limit publishing opportunities. Some felt APCs prohibited researchers without funding from sharing their research globally. Only authors with "deep-pockets" can publish, leaving out young researchers, students, adjuncts, retired researchers, and anyone producing good research without the financial means to cover publishing charges. A few participants stated that APCs were "unique" while some pointed out that the grant funding period does not often coincide with the timing for APCs. One participant was deeply concerned about publishing open access in general because they feared without a hard print copy there was no evidence to support the preservation of their work. One faculty member considered a journal that required payment to publish as a predatory journal.
Mathematicians and physicists at SBU used the arXiv.org OA repository and did not publish in journals where APCs are involved. SBU researchers in the engineering field did not view open access publishing as rigorous or valid for tenure and promotion. Another concern was the quality of peer review. For example, a faculty member from the biological sciences believed the open access business model discouraged rigorous peer review.
All responses expressed concerns about APCs. One notable concern was the double charging practice that occurs in the APC model, specifically when research is published by someone who is paid with public dollars. Some resented these charges and one even believed APCs cause "corruption."

DISCUSSION
Based on the survey results, the authors addressed two research questions: 1. How will the results inform future practices at SBU Libraries?
2. What are the implications for SBU and other libraries?

Primary area of specialization
SBU is a research-intensive institution with various academic departments and a teaching hospital. To track the discipline and role of each respondent, participants were asked two demographic questions. The survey results indicated that the majority of participants were faculty. This, in turn, signified that article publications were mostly coming from SBU faculty. Based on responses, APCs more frequently impact the science, health sciences, and social science disciplines than the art and humanities disciplines. Knowing participants' demographics helps SBU Libraries strategically focus more on specific disciplines to address the issue of APCs.

Importance of journal status
Most survey participants believed it is important to publish in a prestigious journal of their choice, regardless of if it is open access. Traditionally, publishing in prestigious journals safeguarded promotion and tenure in academics. Nowadays, "promotion and tenure committees may still favor older more established peer-review journals that are not open access to those that are open access" (Wical & Kocken, 2017, p. 113). This points to the idea that SBU authors believe they must publish in established, prestigious journals in order to secure their career's advancement.

Does an open access journal's quality override the APC?
Survey results revealed that APCs are not the main concern for publishing in open access journals. Instead the journal's quality is most important to authors. Cook and Smith (2017) conducted a survey and found that authors were not against open access but preferred to publish in reputable and high-quality journals where their publications met the scope of the readership, regardless of open access status. What is noteworthy is that if authors weren't familiar with the quality and prestige of open access journals in their field, they chose not to publish their work in them (Mullen, 2010).
As open access models become more prevalent, it is vital for libraries to highlight highquality open access journals to inspire change in author, and especially in promotion and tenure committee, perceptions.

Is open access publishing still preferable when the author must pay the APC?
The survey revealed that most participants preferred not to publish open access if they were solely responsible for paying APCs. It is understandable, for instance, that graduate students cannot afford APCs on their own. This burden may directly influence an author's decision to publish in a subscription-based journal over an open access journal. Libraries may want to consider establishing APC funds for graduate students. Libraries can expand education for these constituents on the different types of open access publishing models.

How often have authors been confronted with APCs when publishing?
More than half of participants had encountered APCs when publishing. It is not clear why SBU authors chose to pay open access journals with APCs over publishing in open access journals without APCs. This is an area for further exploration as these authors may be following publishing criteria that is unknown to the SBU Libraries. This also presents an opportunity for libraries to help constituents find high-quality open access journals without APCs.

Have journals ever negotiated, offered discounts, or waived APCs?
7.1. Please share your experience if you answered "Yes" to 7

How APCs were funded
Results from question 7 showed a majority of respondents did not negotiate with publishers on APCs. This indicated there was a significant number of authors who may not be aware of discount or waiver options from open access publishers with APCs. SBU authors who negotiated with publishers were invited to share their experience in question 7.1. Question 8 revealed that more than half of participants received funding either through grant funding agencies, their academic departments, or other channels. Approximately 8% paid APCs by themselves. Libraries as catalysts can provide open access publishers' practices and opportunities more widely to potential authors who plan to publish in journals with APCs. Libraries can and should encourage authors to negotiate with publishers to obtain a discount or waiver, when possible.

Amount authors should pay if there is no other funding source
The survey indicated that a majority of participants preferred not to pay any APC if funding was not available. Only 4 believed that authors should pay more than $500. It is concerning that authors who support open access initiatives often face publishing fees in order to do so.

Who should be responsible for APCs if grant funding is not available?
Most participants felt the author's institution should be responsible for paying APCs and that libraries and authors are the least responsible entities. Regardless of responsibility, libraries can play a role in assisting authors with innovative ways to fund APCs. Libraries and institutions should explore institutional memberships with hybrid journal publishers to reduce APCs for authors. BioMed Central provides membership that applies to BMC  Gruyter, n.d.). Both membership types promise discounted APCs for the authors seeking publication. But the responsibility of membership fees needs to be agreed upon. Libraries should consult their institutions to determine the best options for low APCs.

Author interest in learning more about APCs from the library
Unsurprisingly, the survey revealed that most participants encountered difficulty in finding funding for APCs. Respondents were interested in learning more about APCs from the library. Therefore, it is vital for libraries to set up an author support system for effectively locating APC funding. It is clear the SBU Libraries can offer more literacy sessions on open access publishing.

If yes, please provide comments
Notably, concerns about "double-dipping" were reflected in the survey results. The comments reflected that it does not seem fair for a researcher to pay APCs while their institution pays a high cost for that journal's subscription. It is equally unfair for research institutions to pay for journal subscriptions and at the same time pay for APCs for researchers who cannot obtain funding or can only pay a portion of the charge. In the UK, the government raised concerns about "double-dipping" stating that "institutions paying APCs for gold OA publication in particular journals should see some related and proportional discount in the total subscription fee" although they did not think this was appropriate for publishers to do on a global level (Parliament UK, 2013

LIMITATIONS OF THE SURVEY
After the survey closed, the authors quickly realized that a limitation to the survey was the absence of required answers (excluding the initial question where respondents indicated their willingness to participate). Therefore, some questions received fewer total responses. The survey was started 234 times however 54 were deemed incomplete. A survey was considered incomplete when only one answer was given past the initial participation question.
Another limitation was the list of options for the question "My primary area of specialization is…." Options were segmented by discipline for the sciences but not segmented for the arts and humanities. The results ultimately showed that APCs more commonly affect researchers in the science fields, however the authors of this article felt it was important to state this as a limitation of the represented specializations.
Another limitation was the lack of clarity for questions that ask "agree," "disagree," or "neither agree nor disagree." When participants answered that they "neither agree nor disagree" the authors wondered if "Not Applicable (N/A)" would have been a more appropriate response. Unfortunately, N/A was not an option. The survey may have been more accurate if it was.

CONCLUSION
The survey results indicated that the salience of APCs has a considerable impact on SBU author perceptions. After completing the survey, a few SBU authors reached out to the authors of this article to further express their concerns about APCs.
Overall, SBU authors were most anxious about acquiring funding for APCs. Without proper funding, researchers who could not afford APCs were faced with limited opportunities for open access publishing. Institutions should pay for APCs instead of journal subscriptions to avoid "double-dipping." Additionally, the authors found that SBU authors were interested in learning more about APCs from the library.
Open access scholarly publishing is a complex system as there are many business models and practices. The authors plan to further explore APCs with faculty and graduate students Although this study provided only a preliminary understanding of SBU's author perspectives on APCs, it highlighted the disparate views on APCs among disciplines. Future research may endeavor to explore APCs in hybrid journals where the institution also pays a subscription fee. With more transparency of APCs, libraries may be able to strategically reduce subscription costs during contract negotiations or consider joining a publisher's membership program to pay part or all of the APCs for their researchers. There is no perfect business model for addressing APCs and the issues will not be resolved in a short period of time. One thing that remains clear, however, is the challenge it poses to the fundamental tenets of open scholarship.