A word about supplemental materials

Abstract

As with other issues, it often takes a question to the Editor for which I do not have an answer to draw my attention to a matter. This is the case for the topic of this Editor's Message on supplemental materials. The Journal first started publishing supplemental materials in 2009, and there has been a steady increase in the number of published manuscripts in the Journal with supplemental materials over the past 6 years from 4% in 2009 to 31% in 2014. This increase is consistent with other journals in environmental sciences. For example, Kenyon and Sprague (2014) reported that in 2011 between 22–65% of the published articles in a subset of journals focused on biology, forestry, or fisheries were accompanied by supplemental materials. Their review of supplemental materials showed that journals publishing new research had more supplementary materials than those that published mainly reviews or meta-analyses, and that studies using coarse methods like climate change modeling or land cover mapping had fewer supplemental materials than work based on fine-scale methods such as genetic analyses despite each being data intensive. Journals that focused on short communications or were established since 2000 also had relatively high numbers of supplementary materials. The Journal of Wildlife Management guidelines (Block et al. 2011) currently indicate that “Supplemental material is manuscript information that adds depth to the manuscript but is not essential to readers' understanding of the manuscript (e.g., spreadsheets, detailed equations, video or audio files, in-depth tables and figures).” They currently differ from appendices by not being directly part of the results and in being published separately and only on-line. Supplemental materials are available to the reviewers, reviewed by the editors, and content edited by the Journal staff, but they are not copyedited or typeset so they should be considered ready for publication when the final manuscript is submitted. Wiley Blackwell, our publisher, hosts supplemental materials associated with Journal articles in an open access format, requires a formal decision by the Editor-in-Chief for their uploading, and will not add them at a later stage after acceptance (exchanges.wiley.com/blog/2004/09/01/supplementary-material/, accessed 11 Jun 2015). In the supplementary documents we require authors to entitle them as “Supplemental Materials” and to provide a date, citation for the associated manuscript, and literature cited. Supplemental materials fall under the same copyright agreement as the main article and errors found in them should be corrected via an erratum in the Journal. Our guidelines are not inconsistent with other journals, but there is high variation. In a review of journals in environmental sciences, Kenyon and Sprague (2014) reported that journal guidelines ranged from a disclaimer that the publisher of a journal (e.g., Wiley Blackwell) was not responsible for the content or functionality of supplemental materials (pointing the readers to the author for questions) to indicating that supplementary materials are expected to be of the level of quality as the articles with reviewers formally being asked whether all the materials are necessary for the article. Some journals had strict limits (5–150 MB) and placed tight restrictions on formats, whereas others accepted a wide variety of formats and file types. Some journals required a readme file for metadata, descriptions by type of file extensions, and additional text as descriptive captions, whereas others required minimal information on content. Approaches to hosting materials across journals ranged from posting on author-maintained websites to being hosted by the publisher with either open or paid access. Digital publishing indeed has extended options for submitting materials to support published articles. On-line supplemental materials allow long tables and data sets, and other graphical material to be made available that would be expensive to publish otherwise. However, as with other issues in scholarly publishing, the growing use of supplemental materials poses challenges. In 2010, the Journal of Neuroscience announced they would no longer allow authors to include or host supplemental materials (Maunsell 2010). The Editor-in-Chief, John Maunsell, initially considered the information in supplemental materials cost-effective but viewed the fact that their growth far exceeded the growth of information in articles (measured in megabytes per article) as problematic; he concluded “the sheer volume of material is adversely affecting peer review.” He argued that most reviewers are overburdened with requests to review manuscripts (see Merrill and Cox 2014) and asking them to also review supplemental materials, which often can be as extensive as the article itself, detracted from the quality of the review of the article. It also led to an “arms race” where authors submit supplemental information that may “immunize them again the reviewer's concerns.” In the past when publishing was restricted by the printed page, all materials integral to the article were printed. With the advent of electronic media that could handle multi-media, data sets, details on methods, and computer code among others items, the risk of integral materials getting detached from the article arose. Maunsell (2010) argued this undermined the principle of a “self-contained research report by providing a place for critical material to get lost.” It remains to be seen what effect the increase in availability of data repositories and more emphasis on data journals (see Merrill 2014) will have on these trends. Concerns about the long-term maintenance and regular access to supplementary materials also have been expressed. Anderson et al. (2006) in assessing biomedical journals over the period 1998–2005, found links to supplemental data maintained on publishers' websites were active in only 74% of instances. When publications sponsored by societies change publishers, such as The Wildlife Society (TWS) did in moving from Allen Press to Wiley Blackwell in 2011, access can be become problematic. Thus far, TWS has preserved the integrity to supplemental materials across publishers because articles published before 2011 directed authors to a TWS server (i.e., Table S1 at www.wildlifejournals.org) that is linked to the current publisher's website. Williams (2013) in her interviews with soil scientists also reported difficulties with accessing data due to changing file formats. Supplementary materials easily can become unreadable when new software versions no longer support obsolete file formats. Further, García-Pérez (2015) argued that on-line supplementary materials were “black holes” for citations. Based on a randomly selected sample of articles published in the Proceedings of the National Academy of Sciences in 2013, he found that references only available in the on-line supplemental materials added an additional 10% to the references cited and, with rare exceptions, they were not incorporated into citation databases. Requesting authors to produce a single list of references from the main paper and supplemental materials and posting this in the on-line version of the article as is done for articles in Science may address this issue (Hanson et al. 2011). The disarray caused by the growth in supplemental materials also poses technical challenges for librarians, abstract and indexing services, and repository administrators. Their concerns resulted in a joint initiative by the National Federation of Advanced Information Services (NFAIS) and National Information Standards Organization (NISO) in the United States to make general and technical recommendations for best practices for on-line supplemental materials (NISO and NFAIS 2013). They emphasized the conceptual division of materials into those that are integral and those that are additional to the article. As long as journals remain in print or when page charges apply, the distinction between categories may remain blurred. In any case supplemental materials need to be justified by the authors. The Journal promotes peer review of supplemental materials but recognizes reviewer overload exists. Because reviewers are likely to put less effort into supplementary materials, Associate Editors, the content editor, and Editor-in-Chief need to play pivotal roles in safeguarding their relevance and redundancy. Future standardization and emphasis on metadata in supplemental materials may be required not only to facilitate review but to simplify linkages to other digital repositories as funding agencies require more detailed data management plans. Assignment of digital object identifiers (DOI) also may assist in long-term preservation. Societies working with their publishers can promote access and discoverability in citation indexing services, and find new approaches to facilitate linkages for readers to move easily back and forth between the article and supplemental materials. As the forms of digital publishing continues to morph, effective guidance to its authors, reviewers, and editors in addressing these challenges will be key for The Society to effectively meet its mission of providing science-based research to our professional members. —Evelyn Merrill Editor-in-Chief