Publishing Practices of NIH-Funded Faculty at MIT Previous Contents Next Issues in Science and Technology Librarianship Summer 2010 DOI: 10.5062/F4959FH4 URLs in this document have been updated. Links enclosed in {curly brackets} have been changed. If a replacement link was located, the new URL was added and the link is active; if a new site could not be identified, the broken link was removed. Publishing Practices of NIH-Funded Faculty at MIT Courtney Crummett Bioinformatics and Biosciences Librarian crummett@mit.edu Ellen Finnie Duranceau Program Manager, Scholarly Publishing and Licensing efinnie@mit.edu Tracy A. Gabridge Co-Head and Electrical Engineering Librarian tag@mit.edu Remlee S. Green Librarian for Web Technologies and Neuroscience remlee@mit.edu Erja Kajosalo Librarian for Chemistry and Chemical Engineering kajosalo@mit.edu Michael M. Noga Librarian for Earth & Planetary Sciences & Mathematics Collection Manager mnoga@mit.edu Howard J. Silver Co-Head & Biological Engineering Librarian hsilver@mit.edu Amy Stout Computer Science Librarian astout@mit.edu Engineering and Science Libraries MIT Libraries Cambridge, Massachusetts Copyright 2010, Courtney Crummett, Ellen Finnie Duranceau, Tracy A. Gabridge, Remlee S. Green, Erja Kajosalo, Michael M. Noga, Howard J. Silver, and Amy Stout. Used with permission. Abstract Faculty and researchers who receive substantial funding from NIH were interviewed about their publication practices. Qualitative data was collected from interviews of eleven faculty members and one researcher representing six academic departments who received NIH funding. Interview responses were analyzed to identify a representative publication workflow and common themes related to the publication process. The goals of this study were to inform librarians about faculty publication practices; to learn how faculty are affected by and responding to NIH publication policy changes; and to inform planning and discussion about new services to support NIH compliance in addition to general faculty publishing. Major themes from the interviews included consistency in publishing workflows, but variety in authorship patterns and in data management practices. Significant points of pain for authors included difficulty finding quality reviewers, frustrating submission processes, and discomfort about the implications of publication agreements. Some authors found the NIH submission requirement to be burdensome, but most assumed their publishers were taking care of this process for them. Implications for library services are considered. Introduction The National Institutes of Health (NIH) Public Access Policy requires that peer-reviewed journal manuscripts reporting results of NIH-funded research be made available in PubMed Central, the National Library of Medicine's open access repository. The policy was announced on 11 January 2008, and was made permanent on 19 March 2009 as part of the 2009 Consolidated Appropriations Act. Principal Investigators and their institutions are responsible for ensuring that articles arising directly from NIH awards are submitted to PubMed Central in accordance with the policy. Compliance with the policy is a term and a condition of all grants and cooperative agreements active in Fiscal Year 2008 or beyond and for all contracts awarded after 7 April 2008. NIH funding accounts for approximately 34% of all Massachusetts Institute of Technology (MIT) sponsored research funding (Massachusetts Institute of Technology 2007). In Fiscal Year 2008, MIT researchers received approximately $800 million in NIH research monies per the MIT Office of Sponsored Programs. This new policy and compliance requirements underscore the need for a better, in-depth understanding of the decision-making and workflow processes that NIH researchers use when publishing their work, handling copyright transfer, and performing post-publication activities. A project team was formed within the MIT Libraries to learn about current publishing practices, how recent developments in the open access movement and the NIH Public Access Policy are perceived by MIT authors, and how these changes impact authors' workflow and practices. The project focused on a select group of NIH-funded MIT authors. This sub-population of MIT researchers were selected because NIH is such a large source of MIT funded research. As a qualitative study, our sample was not representative or random, but crossed many disciplines and focused on key NIH-funded researchers. The project team investigated the scholarly journal publication process, from the point when the author determines the need to publish, through selection of a venue, submission of the article, peer-review, revision, and acceptance, to publishing and post-publishing activities. Qualitative data collected from interviews were analyzed to develop a representative publication workflow process and common themes related to the publication process. The interviews were conducted in the Fall of 2008; analysis of interviews took place during the Spring of 2009. Review of the Literature The journal article publishing process for academic authors has received little attention among librarians and social scientists; a literature review revealed few articles relevant to our study. Using methods similar to those described in this study, Nicholas, Jamali, and Rowlands (2006) extracted themes from open-ended comments to examine senior authors' attitudes towards scholarly publishing. They stated that concerns of authors include the peer-review process, faster publishing, better access to journals, journal prices, and issues related to electronic publishing. Frey (2003) discussed behaviors of referees and editors of economics journals and the scholarly publishing system's effects on authors. He stated that authors are at the mercy of the referees, and that the journal's and the referees' interests are not in alignment. Recent reports about scholarly publishing touch on author practices. Ware & Mabe (2009) provided a wide-ranging overview of the Science-Technology-Medicine (STM) publishing market which included several aspects of authorship relevant to this study, such as articles published per author, trends in co-authorship, and attitudes about peer review. The University of California (2007) released a study that provides detailed data on faculty attitudes about publishing, with a focus on attitudes toward open access models, and also includes detailed information on motivations for publishing and specifics such as handling of copyright agreements. Methods The study used open-ended interviews, a qualitative, ethnographic survey method, to gather impressions of the publication process. We wanted to avoid predetermining outcomes by narrowly defined questions or directed questions that might simply reflect librarian or faculty preconceptions. The project team chose to study publication practices by interviewing faculty or researchers about one of their NIH funded articles each had recently published. Focusing on a recent article allowed each author to share more detail about the process and people involved. The intent was to avoid generalizations about publishing practices in favor of greater detail and gain deeper insights about authors' work and needs. Study participants were identified by two methods. The primary method was to approach NIH funded faculty; these faculty were identified via data available from MIT's Office of Sponsored Programs. The project team also identified study participants by asking for suggestions from initial interviewees. Interviews were conducted in teams, with one project team member, usually the subject liaison, acting as interviewer and a second acting as note taker. Interviewers prepared for each session by finding recently published articles by the interviewee that reported on NIH-funded research (Table 1). The interviewers investigated the policies of each journal to see how publishers did or did not help authors comply with the policy. Interviews were captured via voice recorder and recording pen. The interviews began by gathering general information, including a request to sign a consent form for recording the interview, an assessment of the author's knowledge of the NIH Public Access Policy, an estimate of how much of their funding is from the NIH, and whether the policy has affected their work in any way. The bulk of the interview involved asking authors to reflect on the process they used to publish a specific paper. Generally the intent was to let the story unfold, but questions were prepared to prompt authors if they did not supply sufficient detail (Appendix 1). Because not all questions were asked of all authors, we do not offer a quantitative assessment of responses, but use qualitative language. After the details of work practices were shared, general questions were asked about roadblocks to the publishing process, how often they publish, and how they handle copyright agreements. If any questions arose about the NIH mandate, authors were offered a handout on the topic. The interviews lasted thirty to sixty minutes. Interview teams met after each interview to draw out themes and user needs. After all the interviews were completed, the project team met to analyze the common elements of publishing practices across all interviews and to summarize the themes as they emerged in discussion. Table 1. Journal Articles Discussed in Interviews. Journal Number of Authors MIT Non-MIT MIT Faculty Publisher NIH Policy - Who submits? Angewandte Chemie International Edition 2 2 0 1 Wiley-Blackwell Author, unless select paid option, then Publisher deposits final published version for immediate access Arteriosclerosis, Thrombosis, and Vascular Biology 8 4 4 1 American Heart Association Author Cancer Research 6 6 0 1 American Association for Cancer Research Author Genes & Development 3 3 0 1 Cold Spring Harbor Press NIH Partner; Author does not need to take any action Journal of Biological Chemistry 7 3 4 1 American Society for Biochemistry and Molecular Biology NIH Partner; Author does not need to take any action Journal of Chemical Information and Modeling 2 2 0 1 American Chemical Society Author; unless select paid option, then publisher will submit final published version Lab on a Chip 3 3 0 1 Royal Society of Chemistry Publisher deposits author version Molecular BioSystems 6 6 0 3 Royal Society of Chemistry Publisher deposits author version Molecular and Cellular Biology 6 3 3 1 American Society for Microbiology NIH Partner; Author does not need to take any action Nucleic Acids Research 5 3 2 1 Oxford University Press Publisher submitted biomedical articles Nucleic Acids Research 4 4 0 1 Oxford University Press Publisher submitted biomedical articles PLoS Computational Biology 5 5 0 3 PLoS NIH Partner; Author does not need to take any action Proceedings of the National Academy of Sciences 4 2 2 1 National Academy of Sciences NIH Partner; Author does not need to take any action Proceedings of the National Academy of Sciences 6 3 3 1 National Academy of Sciences NIH Partner; Author does not need to take any action Proceedings of the National Academy of Sciences 4 3 1 2 National Academy of Sciences NIH Partner; Author does not need to take any action Proceedings of the National Academy of Sciences 5 4 1 2 National Academy of Sciences NIH Partner; Author does not need to take any action Tissue Engineering Part A 4 3 1 0 Mary Ann Liebert Publisher (post peer-review and copy -edited version) Average 4.7 3.5 1.2 1.3 Total 80 59 21 22 Summary of author information for journal articles discussed by authors. Some authors, when presented with a choice of papers, chose to discuss both. NIH Partners are journal publishers that have agreed to deposit final published articles in PubMed Central without author involvement. CC Results Eleven faculty members and one researcher, representing 6 academic departments, were interviewed. Collectively, these 12 authors received approximately $60 million in NIH funding, with an average award of nearly $5 million. Interviewed authors estimated that NIH provided approximately 70% of their research funding. Between 2004 and 2008, the authors collectively published approximately 500 articles in 151 different journals, averaging 8 journal publications per year per author (Table 2). Table 2. Authors' Journal Publication Practices and Research Funding (Massachusetts Institute of Technology 2007) Departments Authors NIH Awards In Force Federal FY2008 anticipated total amount ($) reported as of 2/27/08 Average Estimated NIH Award (%) Estimated total research funding ($) Number of papers published by authors Number of journal titles published in by authors Biology and Health Sciences & Technology Division 4 24,091,374 55 34,875,937 87 63 Chemistry 4 17,668,562 76 22,853,012 290 79 Biological Engineering, Chemical Engineering, and Electrical Engineering and Computer Science 4 17,354,647 78 20,038,239 110 68 Total 12 59,114,583 70 77,767,188 487 183 Author and publication year (2004-2008) were harvested from searches in Web of Science and PubMed. Citation information was imported to an Excel file, where duplicate citations and conference proceedings were deleted. Publications per interviewed author and number of different journals published in were reported. NIH funding data was provided by MIT's Office of Sponsored Programs. The total research funding estimate was calculated using the interviewed authors' estimate for the percentage of their overall funding provided by NIH. Publishing Workflow Analysis Steps taken by the authors to publish articles were largely as we anticipated. They included a fairly consistent set of tasks as illustrated in the following diagram (Figure 1). The following summary describes each of these steps in more detail. Research & Lab Work All publications discussed in this study began with research done at MIT, with more than half involving research partners at other institutions. The research was conducted with a graduate student or a postdoc and the author. Publishable Unit of Work Completed Once the research generated compelling results, the authors started the publishing process with their researcher teams. In many cases authors described it as having a story to tell: "At the start of project, the goal is to establish an overall research plan that will lead to a significant piece of data or significant result that will then be published ... [You] set up the overall experimental design so that you have a publishable unit at the end. At that point you figure out what experiments need to get done. When those experiments start happening and the data starts coming in, then we figure out where the story leads us." Select Journal Choosing the venue for publication requires careful match between the story the researchers want to tell and the characteristics of the journal. Authors want to match the subject matter to the audience, reach the greatest number of peers, and assure the greatest impact for their research: "Once we decide we have a story, the very next step is to decide where to publish it. In science there is a hierarchy of journals. Fundamentally scientists want to publish in the very best journal they can ... on the basis of the novelty of the paper, the originality, the quality of the paper, and the impressiveness of the results, one makes this determination." Other authors chose where to publish based on society memberships, on where they might find the most thoughtful reviewers, or where they expected the fastest turnaround time: "Why are [these journals] considered the best? Because I get thoughtful reviews from smart people. Good feedback from my peers is the most valuable input I can get." Journal selection usually happened early in the process because most journals have requirements for how articles are constructed and formatted, which affects how the manuscript is prepared. If their paper was rejected by one journal, authors chose a new journal and retooled the manuscript based on the new journal's requirements. Draft Manuscript The specific steps authors took to draft the manuscript varied in terms of how much control they ceded to the students and also in the approach they used to create the content. Most authors used manuscript writing as a teaching exercise, firmly believing that this is the process by which graduate students become independent writers and scholars. They spent a great deal of time carefully editing student work. Some mentioned that they write the comments on a paper copy instead of using the "Track changes" feature in Microsoft Word, because they want the students to fully engage in, absorb, and learn from the comments. Other authors let the student write certain sections, but wrote sections such as the introduction, discussion, and conclusions themselves: "It obviously would be much faster for me to sit down and write the paper ... I believe it is invaluable for the graduate student to do it themselves even if it is a pain for me to edit." To get the manuscript started, authors worked with students in a number of ways. Several mentioned they determined which figures to present and then crafted the text around the figures. "We agree on a set of figures ... the figures are crude in the beginning ... the figure becomes self-explanatory ... I spend a lot of time teaching them how to do that." "A huge portion of the time goes into thinking about how we can present the data and emphasizing what is important about the article that distinguishes it from previous publications." Most of the authors mentioned that EndNote is the reference-gathering tool they use to gather citations. Often the EndNote library was curated by the student, though there were a few exceptions to this. Cited references were carefully reviewed: "I require every paper that comes to me from students to be accompanied by a complete paper copy of every reference that is cited, and I check them." There were many variations in how research contributors were included as authors of the manuscript. Authors reported that a few individuals did the lion's share of the writing and work, but others were included as authors if they had been involved in the work in some way. The person awarded the grant that supports the work was always listed as an author. Acknowledgments were carefully constructed to ensure all players and grants in the conduct of the research were fully recognized as contributors. Revise Manuscript Getting the manuscript into publishable shape was an iterative process between the authors and their students. "It can take 10-20 iterations for each paper back-and-forth with my students and me to make it publishable." Sometimes more lab time is required to fill in gaps identified during writing. Authors knew they were almost finished drafting when comments shifted from substance to style or when the suggestions became "nice to have" versus necessary. Submit to Journal Authors noted that the online submission process is not difficult; others lamented that every system requires a password and every journal has different publication formatting requirements and standards. This part of the process is seen as time consuming, and many authors delegate this responsibility to staff assistants or students. Receive Reviewer Comments The reviewing process can be fairly smooth if the manuscript receives positive reviews. In some unusual cases principle investigators (PI) may select their own reviewers (e.g., as a fellow of the National Academy of Sciences publishing in Proceedings of the National Academy of Sciences), which makes the reviewing process easier and faster. Substantive reviews will improve the final article, but authors also commented that really thoughtful reviews are uncommon. "In the old days ... you used to get thoughtful reviews on your paper. Somebody would spend a lot of time thinking about it; they might make some really important suggestion that makes you think about things in a different way ... nowadays, that doesn't hardly happen anymore." The reviewing process also has many pitfalls. Negative reviews may result in a rejection of the submission, reviewers may have to be reminded to complete the reviews, or the PIs may have to write a response to a reviewer's critical comments. Many authors felt the reviewer comments must be distributed to all authors, including those who had moved on to other institutions and to other research. All these steps are time-consuming. Make Changes to Manuscript, Resubmit, & Review Final Proofs The next level of redrafting happened after the editors and reviewers had sent their comments on the manuscript. If all the reviews were positive, the article was accepted for publication. The redrafting can be as easy as making suggested changes and resubmitting. If the reviewer's comments are negative or the editors suggest that the article does not fit within the journal's scope, authors have to find another journal to resubmit the manuscript and aim the revisions towards that new journal. Some authors commented that they used to get fairly detailed comments from editors, but now that all the work is done online via publishers' web sites, it is hard to determine what changes the editors have made. Another challenge is that co-authors do not always send their feedback and the primary author has to decide what changes to make. Some authors stored only the final galley proof, not their own final drafts. Final Submission Administration Based on studies at other universities (University of California 2007; Covey 2007) and anecdotal data from MIT, we expected that MIT authors were not investing much time in reviewing publication agreements. MIT has an {addendum} to such agreements for authors of NIH-supported research. We asked how the publication agreement was handled for the particular paper that was the focus of the interview, and also how authors worked with such agreements generally (Table 1). Our interviews confirmed prior impressions that most authors don't spend much time reading their publisher agreements or storing them for retrieval. Some don't read them at all; others read them quickly but don't attempt to alter the terms or retain copies. Typically, authors read over agreements, but did not make any changes or did not keep careful track of past agreements: "I just sign it, especially for big journals ... I throw them away, [although] I scan copies of them and save them." I "scan the agreement quickly," but don't ask for any changes to it. "I pass all the paperwork to someone in the lab," returning for a quick signature on the agreement. While some indicated "I don't even read it. I just sign it ..." most authors indicated that they at least read the agreements: "One thing I've learned over the years is never sign anything you haven't read. I read it ... to be sure there's nothing totally off the wall." Naturally, the authors' focus is on research and publication as opposed to contract language, yet, some authors expressed some concern about not tracking the details of agreements: "I just sign it ... I probably could go to jail ... I don't really know what's in them. I don't have time to read [them] ... the print is so small you can't see it and lots of times what they say you can't understand anyway." "What are they going to do? Sue me? ... I'm sure I violate copyright ... but I don't do anything extremely wrong." "If I have to go to jail, I have to go to jail ... these are normal collegial interactions and if this violates the copyright form I signed, so be it. I feel if you do reasonable things ... [it should be OK]." None of the authors indicated that they currently attempt to alter their publication agreements, and few mentioned having tried in the past. One author felt there was little point in suggesting changes to the agreement: "You are negotiating with some secretary there who doesn't know anything." Another indicated that asking for changes is pointless, since: "You are entirely beholden to them in the ... publication process." "With the MIT form, they say 'our lawyers haven't looked at that; either sign our form or we won't publish your paper.'" One author had recently explored issues related to his rights through conversations with a member of MIT's Ad-hoc Faculty Committee on Open Access Publishing: "I didn't realize how much the publishers had consolidated; how much things are out of balance...They send you the form and ask you to sign it ... one wonders whether it should be a little bit more of a discussion, but basically I've always just signed it ... I can't say I've thought a lot about it when I do that." The author's thinking has recently evolved, so that now the author wonders: "Did I really know whether I could go on and put this on my own web site or not? It's not something I really thought about. ... I would like a little more [use rights] to be in the hands of the people who did all the work." The author would like MIT and other universities to solve this problem: "It's a problem; it's something I wish universities could work on together, more than they have in the past." Post Publication Activities Most authors reported that working with the NIH policy is an extra burden that they would rather avoid. One author's comment was characteristic: "I don't want to do any extra work ... there are a lot of things to do as is." Authors reported being unclear about their responsibilities and what actions publishers or NIH will make on their behalf. Most would like help from the MIT Libraries with submission or other parts of the process, though one author indicated that while help would be welcome, library staff time could be better used in other ways. Those who do not feel burdened by the policy believe, whether accurately or not, that their publishers are submitting for them. For example, one author who said there was "no problem" complying indicated that the publisher "promised me that they would submit for me. That is nice because more people can read [the paper.]" (Note that out of the 17 journal articles that were discussed in this study, 13 articles were published by publishers that submit to PubMed Central on behalf of the author while four articles were published by publishers that do not submit to PubMed Central.) Another author who had not submitted directly to PubMed Central, assuming the publisher was doing so, suggested a simple process to support submission: "If there was some way you could just dump [your PDFs] once a year; just click on something and dump them ... that would be good." Although some authors were not concerned about submission, they frequently expressed some concern about whether the publisher submission to PubMed Central was happening as it should. One suggested a library service verifying submission to PubMed Central: "One thing your library could do ... track the papers and make sure they end up open access [in PubMed Central]." Most authors indicated a desire to share their work, but the degree to which authors do share varied considerably, perhaps because about a third of the authors mentioned being confused about the legality of such sharing. One author indicated linking to papers but doesn't post them because: "It feels like a complete copyright violation; there doesn't seem to be any way around that." In other cases, generational differences seemed to come into play, with more senior authors less interested in sharing their work by posting it on the web. One such author commented that after the older generation is gone, younger faculty will maintain web sites with all their publications. Typical of this view was the comment from a non-poster that web sites are blatant self-promotion: "I'm just in a different generation ... web sites don't do it for me." One typical younger author consistently posts PDFs to a web page. When asked about preservation of these works, this author suggested that post retirement, when an author was no longer maintaining a web site with a list of publications, a repository such as DSpace@MIT could be used to store the final version of the author's web site. Most of the authors agreed that sharing data was a good idea, but few were actually doing it. "Access to other people's data and the ability to query these datasets is increasingly important." Some authors shared data because of outside requirements, explaining, "The rules for most journals now require that if there's microarray data that there's access to it." Most authors store their data in diverse places, including local PCs, external hard drives, and CD/DVDs. Most authors manage their own data. One author said that supervising a data manager in the laboratory would prove too time-consuming. One author noted that the unique methods of data collection for each research project might be too individual for systematic sharing of datasets. Some authors reported keeping hard copies of their data, especially in lab notebooks that are the property of the lab. One author stated that this was simply "because I'm a hard copy person." While a few authors said they rarely, if ever, went back to consult a laboratory notebook, one said that consistently checking data even 10 years back is typical. Another announced, "I have notebooks from 35 years ago." However, no authors kept all their data, but rather culled their data for the important information and disposed of the rest: "I get rid of a good half of... [the data] at least. The files are very big... Actually there's no good repository for all that electronic data." One author stated a concern over raw data versus processed data: "It's getting harder and harder to have the primary data versus some processed version of the data ... [the data managers] keep the very rawest data until they've done some basic analysis. They just can't store everyone's ... [data]." As to whether or not managing data is a problem, few authors expressed concern. However, one author admitted: "I don't annotate the data. I don't think I could find it next year at this time." One author, referring not just to data, offered, "I think somebody needs to find a solution to the information problem ... they have all this stuff in the computer but it's not organized in any way." Long-term data storage is an issue for faculty. They ask students to put their data and papers on a CD before they leave the lab after graduation. But some worry about what happens to those CDs after their retirement, and suggested a solution could lie in the MIT Libraries. Discussion Our study was qualitative and was therefore intended to highlight general themes about the publication process of scholarly journal articles of NIH-funded MIT authors. Major themes that emerged from the interviews were: Publishing workflows were similar in sequence. All authors described a process involving multiple drafts, iteratively edited and reviewed by a group. Many authors reported using figures as the starting point for a paper; others indicated that they started with an outline. Most chose to delegate portions of the writing to graduate students as part of their role as instructors and mentors. Many authors indicated that they rely on printouts of successive drafts to ensure that comments and changes can be seen and tracked to provide clear feedback to graduate students. Descriptions of data management practices were idiosyncratic, without the commonalities seen in other areas of the publishing workflow. In addition to these major themes, many authors expressed noteworthy issues and concerns. Authors frequently reported difficulty receiving thoughtful and timely reviews of their manuscript submissions. In some cases, authors stated that they chose to publish in certain journals based on the quality of reviewers. There may be implications for institutional repositories for authors that only retain their galley proof, as most publishers do not allow posting of that version. Authors revealed that they typically signed publication agreements without reading the content and only sometimes kept copies. Authors did not report negotiating agreements or referring back to them. Some expressed feeling powerless when signing agreements and felt frustrated by their loss of rights to their work. Others expressed a lack of concern with the details and implications of the agreements. Sharing work after publication was sometimes curtailed by concerns about rights; authors expressed discomfort and uncertainty about whether sharing their work was in violation of the law. The implications of the study's findings are limited because the small study size and its qualitative approach mean the findings cannot be confidently generalized. Nevertheless, we believe that the study supports potential library services that all research libraries may consider. Conclusion This study was intended, in part, to help inform decisions about how and whether the MIT Libraries could support NIH-funded authors in their publication process. Possible opportunities for support include: 1. Support for Publication/Copyright Transfer Agreements and Author Rights Authors clearly felt powerless to manage their rights when signing publisher agreements and expressed some confusion about their options for reusing and sharing their own work. The interviews reflect and support growing discussions on many campuses about the need to change the scholarly publishing system so that individual faculty are not solely responsible for negotiating author rights. Such discussions have recently led to open access policies at Harvard, MIT, Stanford, University of Kansas, and many European universities; the policies at these universities reduce the burden on individual authors. Another approach that is supported by this study is the expansion of efforts to negotiate license agreements for library content so that licenses include author rights for all authors on campus. Recent examples of success with this approach are the agreements between Springer and the University of California and MIT (Duranceau & Anderson 2009). There is also a suggestion that more support for individual authors -- interpreting whether and how they can share their work -- might also be welcome. If there were a centralized service for posting papers, for example, this service might include rights review so that authors inclined to share could do so, and those doing so could continue, but without the current level of concern. 2. Publication-based Approach to Services The interviews highlight the crucial importance of the publishing process to authors, suggesting that organizing library services around the themes and stages of publishing would make sense in communities supported by research libraries. Following up on this idea, a MIT Libraries team has just launched a web site that describes and links to library services in terms of publishing stages. Yale Medical School has also taken {this approach.} 3. Instructional Opportunities in Relation to Writing Research Papers A key theme of the interviews was the authors' deep commitment to teaching the next generation of academic authors. Authors indicated that not all graduate students are ready to write a publishable research paper, and they expend a great deal of effort training them to do so. There may be a role for libraries in supplementing faculty efforts by offering instruction to graduate students focused specifically on writing research articles. 4. Data Management Services Those interviewed did not express a need for support managing their data, and indeed suggested some barriers to taking advantage of such support. For example, authors explained that it is difficult for anyone other than the researcher to make full sense of the data; that it would be counterproductive to oversee other parties' attempts to help manage it. Some noted that key data is already stored in external, centrally managed repositories or linked from the journal article as supplementary material. The study suggests that existing data management methods will prove vulnerable over time, in ways not yet apparent to authors because the shift from paper-based data sources is recent and still in process. Libraries may want to consider positioning themselves to offer support for data storage as these vulnerabilities manifest over time. 5. Citation Management Software Many authors reported that they use EndNote to manage their citations, and most of their departments or labs have purchased the software for them. Based on this information, it seems worthwhile for any academic library system to evaluate whether purchasing institutional access for a bibliographic management package would provide overall savings and enhanced service. 6. Third Party Submission Service for PubMedCentral Some authors expressed interest in having the Libraries submit journal articles to PubMed Central on their behalf. Our interviews and information gathering suggested that there are many issues to consider before implementing such a third-party submission service. For example, it would be important to obtain more information about how these services are supported, how heavily they are used, and what kind of staff time is involved, including time to obtain accurate metadata and vetting for the proper publication version. To further leverage any staff efforts made by offering a submission process to PubMed Central, libraries could consider possible dual submission to PubMed Central and their institutional repository. Our interviews also suggest that it might be useful to set up a service to monitor PubMed Central to ensure that all NIH funded papers do end up there, though NIH's own compliance mechanisms may make this superfluous. Overall, it seems important for libraries to carefully weigh how they can best use their limited staff resources to support NIH-funded authors and their institutions. When exploring opportunities for new services, working closely with not only faculty, department heads, and research grant administrators, but also with administrative assistants and others in support roles in academic departments, would help ensure that services are targeted appropriately, thus encouraging the best use of limited library staff time. Future Challenges The publishing environment for scientific researchers is evolving. Along with new options for disseminating research come new mandates for sharing publicly funded research. This new environment brings challenges for authors along with opportunities for libraries to become engaged in supporting faculty publication. Understanding faculty publishing practices used by faculty at your institution is a key part of becoming an effective partner. References Covey, Denise Troll. 2007. Faculty Rights and Other Scholarly Communication Practices [Internet]. [Cited 7 August 2010]. Available: http://www.slidefinder.net/F/Faculty_Rights_Other_Scholarly_Communication/7487510 Duranceau, E.F., Anderson, I. 2009. Author-rights language in library content licenses. Research Library Issues. 263: 33-37. Frey, B.S. 2003. Publishing as prostitution?-choosing between one's own ideas and academic success. Public Choice. 116(1-2): 205-223. Massachusetts Institute of Technology. 2007. MIT Briefing Book. [Internet]. [Cited 7 August 2010]. Available from: http://web.mit.edu/vpr/www/reports/MIT_Briefing_Book_2007.pdf Nicholas, D., Jamali, H.R., & Rowlands, I. 2006. On the tips of their tongues: authors and their views on scholarly publishing. Learned Publishing. 19(3): 193-209. University of California Office of Scholarly Communication and the California Digital Library eScholarship Program. 2007. Faculty Attitudes and Behaviors Regarding Scholarly Communication: Survey Findings from the University of California. [Internet]. [Cited 7 August 2010]. Available from: {http://osc.universityofcalifornia.edu/wp-content/uploads/2013/09/OSC-survey-full-20070828.pdf} Ware, M. & Mabe, M. 2009. The STM report: an overview of scientific and scholarly journals publishing. [Online]. United Kingdom: International Association of Scientific, Technical and Medical Publishers. Available: {http://www.stm-assoc.org/2009_10_13_MWC_STM_Report.pdf [Accessed 6 Nov 2009]. Appendix 1. List of Interview Questions. The interviews were open-ended; questions below were used as basis for interviews. Introduction Are you aware of the new NIH Public Access Policy? What is the role of NIH funding in relation to your research? Roughly what percentage of your funding comes from this source? Has the new NIH mandated policy affected any aspects of your publication practice so far? Workflow/diagram questions We would like you to reflect on the process you used in publishing your last article. We have a recent publication of yours if you'd like to use it, or feel free to choose some other recent paper. What is the title of the paper? Where was it published? Were you the primary author? Were there co-authors? How did you choose a publication venue? Using this chart and markers, we'd like you to draw, or diagram, in as much detail as possible, the work flow (steps and tasks) you (or others) used in getting the paper published. We may ask questions along the way to clarify points or gain additional detail. Publication information What is the title of the paper? Where was it published? Were you the primary author? Were there co-authors? How did you choose a publication venue? Collaboration Who is typically involved in a publication process? How did the collaborative aspects of this process work for you? Citation management How did you manage the list of references related to this work? Did you use any content (tables, images, diagrams) from other authors; if so, how did you manage that? How did you manage permissions when publishing this content? Publication agreement How was the publication agreement handled? Did you experience any problems with it? Do you save/store/keep the publication agreements? Sharing the work post-publication & self bibliography Aside from publishing this in the journal, do you use any other strategies to make it accessible to a wider audience? Do you post it on a web site? How/where do you store it? What barriers or issues arise related to these post-publishing activities? Data Management Was there data associated with this paper? If yes: What is the data? What kind of data do you use or generate? How and where did you store the data? How is this data associated with the published article? What tools/software/formats did you use? Do you envision this data being used by others? If yes, What are the barriers to sharing the data? NIH Submission What are the challenges you foresee to meeting the NIH requirements? Did the paper get submitted to PubMed Central, and if so, through what means? Can you describe the exact steps and people involved in the submission? If yes, have you had to find, store, or use the PubMed Central ID numbers, and if so, how did that process go, and who was involved? If the paper is not affected by the NIH mandate, how do you think you will incorporate the NIH process in your work? How did you accommodate the NIH Mandate in this paper's author agreement? What is your normal practice in dealing with publication agreements -- copyright transfer agreements? Do you read them? Do you ever ask for changes to them? Do you keep them on file? What would be helpful to you to comply with the NIH mandate? What is the most difficult part of complying with the mandate? General, non-paper specific questions What are the major roadblocks to making this publishing process efficient for you? How do you manage your publications/publications list? How do you find/get copies of your own papers? How do you find/get copies of the papers of your colleagues? Are there aspects of this process you'd like to see streamlined or changed? Is your bibliography public (if so, where?), and how often do you update it or have someone update it? Who is involved in the process of keeping your personal bibliography/publications list up to date? What determines how often you publish? What is your normal practice in dealing with publication agreements -- copyright transfer agreements? Do you read them? Do you ever ask for changes to them? Do you keep them on file? What concerns, if any, do you have about managing data in relation to your research? If money were not an object, what would you hire someone to do in relation to managing data? Is there anything we can do for you to assist your compliance with the NIH Mandate? Can you recommend one or two other people with NIH funding that might provide some interesting insights on their own publishing practices within your discipline? Previous Contents Next