European Science Editing 1 February 2016; 42(1)
Editorial Board From the editors’ desks Chief editor Ksenija Baždarić EASE newsletter welcomes this decision, as a forum Managing editor Dado Čakalo EASE has a monthly newsletter from should be a member-only discussion December 2015, the editor of the group as it represents being a part of Production editor Lynne Rowland newsletter is our vice president Pippa the EASE community. EASE editors Smart. It covers EASE news and are posting not to the public but to Original articles and essays Ksenija Baždarić events and other news and events of their community. So the EASE forum [email protected] interest to our members. If you have digest will be avaliable only in the Book reviews any problems with the newsletter printed issue of the journal. Laurence Mabile please contact our secretary Dalibora Current issue Viewpoints and My life as an editor Behmen ([email protected]). Rhianna Goozée This issue brings Moira Hudson’s [email protected] FaceBook page editorial on her experiences as Correspondence With thanks to Duncan Nicholas an editor (page 2), Cagney et al Hrvoje Jakovac and Cristina Vasiliu, EASE has a original article about retraction [email protected] new FaceBook page: https://www. and republication from the editor’s Meeting reports facebook.com/EurSciEd, that is more perspective (page 3), Denys Hannah Cagney interactive and can ensure effective Wheatley’s essay about independence [email protected] communication with members. of editors (p7), Michele Nujiten’s EASE-Forum digest essay about statistics (page 9), Elise Elise Langdon-Neuner Membership renewal Langdon Neuner’s viewpoint about [email protected] Don’t forget to renew your the proper use of the word multiple This Site I Like membership for 2016 to keep your (page 15), an interview with the Jong Silvia Maina member benefits: any queries, please K. Ha, President of the Council of [email protected] contact our secretary Dalibora Asian Science Editors (page 21), News Notes Behmen ([email protected]). meeting reports from Ishan Dave John Hilton [email protected] (page 17) Anna Maria Rossi (18) and EASE forum digest Nuria Salade (20) plus our regular The Editor’s bookshelf Anna Maria Rossi At its last meeting, the EASE Council sections. [email protected] decided that from this issue on, We hope you will enjoy the read. EASE Council the EASE forum digest will be for Joan Marsh (ex officio) members only. The editorial board Ksenija Baždarić
International Advisory Board EASE Council 2015–2018 Denys Wheatley (UK) President: Ana Marušić, Croatia; [email protected] Ana-Maria Simundic (Croatia) Vice-Presidents: Chris Sterken, Belgium; Pippa Smart, UK Karen Shashok (Spain) Members: Ksenija Baždarić, Croatia; Paola De Castro, Italy; Shirin Heidari, Switzerland; Moira Hudson, UK; Rachel Lammey, UK; Duncan Nicholas, UK; Elaine Seery, France; Ines Steffens, Sweden; Sylwia Ufnalska, Poland Past-President: Joan Marsh, UK Contributions should be sent to the Treasurer and Company Secretary: Roderick Hunt, UK Chief Editor or the appropriate section editor. See the Instructions to Authors Secretary: Dalibora Behmen, Croatia; [email protected] on the EASE website (www.ease.org.uk). The journal is published in February, European Science Editing is indexed/listed in SCOPUS, SCImago Journal and Country May, August and November, free to paid-up members of EASE and available Rank (SJR), SCIRUS, CAB International, Ulrich’s Periodicals Directory, NewJour, on annual subscription of £77.50 to Genamics JournalSeek, Chemical Abstracts Service (CAS), Electronic Journals Library libraries and other non-members. (EZB), J-Gate, ZETOC, The British Library, The Library of Congress, Cornell University Library, The John Rylands Library of the University of Manchester, Geneva Foundation Disclaimer: The views expressed for Medical Education and Research (GFMER), Academic Index,VINITI of RAS, GeoRef, by contributors are their own. The Association does not necessarily Google, Index Copernicus Journals Master List, NLM Catalog, NLM LocatorPlus, endorse the claims of advertisers. EBSCO, ProQuest: LISA, HINARI, WorldCat, ResearchGate, and RIN, Cabell’s Directory of Publishing Opportunities. ISSN 0258-3127 Printed by Qwerty Ltd, The Markham To advertise in this journal, or on the EASE website, please contact EASE Centre, Theale RG7 4PE ©EASE 2016 Secretary, Dalibora Behmen, [email protected] European Science Editing 2 February 2016; 42(1)
Editorial Lament of the publications professional: 1. Managing author expectations and the obstacle of peer review
The aim of all authors, whether they be academics, clinicians their target journals with extensive reviewer comments. or pharmaceutical companies is to publish their research in With the first two manuscripts to come back, the response the best possible journal – in terms of suitability, audience, of my client was to flatly refuse to consider addressing the and impact factor. The carefully crafted manuscript, once reviewer comments as “they are fundamentally flawed”. submitted, hopefully makes it to the desks of two or more In other words, the client felt that the reviewers had peer reviewers, who may have been suggested by the author not understood the studies, or taken into account their themselves as having sufficient expertise to make a fair and inherent limitations. I was instructed to simply reformat informed evaluation. the manuscripts for alternative (lesser IF) journals, and get Dating back to the earlier part of the 18th century, peer them resubmitted as soon as possible. A cheap job, but a review has come a long way – evolving from a means of false economy. Despite my advice that this was possibly assisting editors to select manuscripts to, according to the not the wisest course of action, the client persisted, and chief editor of Science and Engineering Ethics, Raymond E the reformatted manuscripts were duly submitted to the Spier1, alternate journals and – although hard to believe – the very “a turf battle with the ultimate prize of knowledge, same reviewers suggested during the submission process. science or doctrine being published. On one side we By the time these two manuscripts came back from have the writers, on the other the editors and critics.” their second-choice journals, the other two manuscripts were back from their first-choice journals. Perhaps now the But what do authors really think of peer review? Do they client would listen to my advice, select a journal with the welcome critical feedback and use it to the full to improve correct specialist audience, not the highest ranking cancer their manuscripts, or do they regard it as an obstacle in the journal in print, be realistic about the impact factor of a way of publication? journal that is likely to be sufficiently niche to consider the One would hope the former, that authors would appreciate manuscript, and address those reviewer comments that the comments and suggestions of their peers – constructive could be deemed valid while pre-empting those considered criticism given freely and in good faith. But my experience misguided, perhaps because the nature of the disease, or the during several of years as a publications professional in a design of the study made them impossible to address. (smallish) number of medical communications agencies I lost count of the number of times I invoked GPP-32, tells me that this is not always the case. and appealed to common sense, but in the end we got The example I will use to illustrate this is that of a there, by the end of my stay in that particular agency, all pharmaceutical company working on an orphan disease four manuscripts were accepted for publication, following – a rare disease affecting a very small percentage of the various degrees of modification, and in journals reflective population, defined in the European Union as fewer than of the “real-world” interest in the data. Of course, the case 5 individuals in 10,000 of the general population. Clinical of the deceased author is a matter for another time. trials of drugs in such diseases are already on the “back foot” in terms of robustness due to the small numbers of subjects References available for study – achieving a sufficiently powered study 1 Spier R. The History of the Peer Review Process.Trends in is often beyond the reach of such companies, as the patients Biotechnology. 2002 20(8):357-8. simply aren’t available. This is not to say though, that the 2 Battisti WP et al. Good Publication Practice for Communicating data aren’t valid – they most certainly are. Further, orphan Company-Sponsored Medical Research: GPP3 Annals of Internal disease areas are usually of interest to a restricted specialist Medicine. 2015; 163(6): 461-4. doi:10.7326/M15-0288 audience – such that the pool of peers available to act as reviewers is small. Author’s note This year I found myself with responsibility for four All experiences and opinions mentioned in this article are manuscripts describing studies carried out with one drug those of the author. No clients, publication professionals, or in two orphan diseases. The aim, obviously, was to get them authors were harmed during the events leading up to this all accepted for publication in the “best” journals possible. editorial. Despite trying to manage the client’s expectations, all four manuscripts were submitted to fairly high ranking journals. Moira Hudson One after the other these manuscripts were returned from [email protected] European Science Editing 3 February 2016; 42(1)
Original articles Retraction and republication—a new tool for correcting the scientific record?
Hannah Cagney, Richard Horton, Astrid James, Sabine Kleinert, Zena Nyakoojo, Laura Pryce The Lancet, London, UK; [email protected] Emma Grainger, Diana Stanley The Lancet Respiratory Medicine, London, UK Helena Wang The Lancet, Beijing, China
Abstract: Mistakes in the scientific literature are reasonably the erroneous findings, but a retraction alone might also common; however, the available methods to communicate remove valuable and important data. Further, retraction corrections to published articles or retract unreliable findings stigma could unfairly penalize authors who have been do not always meet the needs of readers, authors, editors, proactive in disclosing problems with their research, and and publishers. In this article, we propose an innovative potentially discourage other researchers from being honest process of retraction and republication: the purpose of this about mistakes in the future. instrument is to correct the scientific literature for situations Therefore, the aim of this article is to describe an in which a published article is realized to contain pervasive innovative process of retraction and republication to error, but still contains important scientific findings. We transparently correct the scientific record, and present two present two case studies of its use in The Lancet and The case studies of its use. Lancet Respiratory Medicine. Case 1: The Lancet—editorial retraction Keywords: retraction, publication, ethics, correction In January, 2014, The Lancet received an epidemiological investigation of the burden and treatment of heart attacks in Introduction China across a 10 year period. The study involved analysis Retraction is the formal withdrawal of a scientific article of data from 162 hospitals, which were then weighted to or other publication. The Committee on Publication Ethics derive nationally representative estimates. The findings were recommends that editors should consider retraction of articles important but disturbing: although China had increased in cases of duplicate publication, plagiarism, or reporting of the coverage of critical interventions for heart attacks, unethical research, or in cases with clear evidence that the in-hospital mortality had not changed. In other words, study findings are unreliable.1 The purpose is to ensure that China faced a crisis in the quality of its rapidly expanding the scientific record remains accurate, and to inform readers health services. After peer review and several rounds of that the research findings and conclusions cannot be relied revision, the article was accepted in May, 2014, copy edited on; COPE explicitly states that the purpose of retractions and published online on June 24, 2014,5 together with a is “to correct the literature and ensure its integrity rather commissioned accompanying comment that quoted some than to punish authors who misbehave”. However, although of the study’s findings.6 Print publication was planned for a unreliable findings can result from misconduct or genuine special issue in late August. error, retraction is often associated with high-profile scandals On August 14, the authors alerted the journal to an error involving scientific fraud, with reputational implications— in the published article. A miscalculation in the weight of so-called ‘retraction stigma’—for the authors of retracted one of the urban areas in the study had affected data in the articles irrespective of the cause for retraction.2 results section, figures, tables, and appendix. The effect of Authors of retracted articles suffer a ‘citation penalty’ the error was fairly small, with the magnitude, direction, of 10%, and authors of papers that are retracted because and significance of the results virtually unchanged (for of “honest errors” similarly have lower citations.3 Other example, the use of aspirin within 24 hours of hospital researchers in the field can also be affected by retractions, admission in 2001 changed from 79.3% [95% CI 77.3–81.3] with a 5% to 10% reduction in citations and a decline in to 79.7% [95% CI 77.9–81.5]). However, the miscalculation funding activity for future research.4 affected almost every numerical result in the paper. Further, This leaves journal editors with a dilemma when there were just 8 days between when the error came to light presented with evidence that research they have published and the press deadline for the scheduled print issue. contains extensive or pervasive errors. A correction notice The planned print publication of the article was put on could be used, but this does not represent a transparent hold while the corrected data were peer reviewed by two mechanism for papers with large numbers of errors, does clinical reviewers and two statistical reviewers involved. not necessarily remove the erroneous results from the All agreed that, although the errors were extensive, the scientific record, and is unsatisfactory when errors in the corrections were appropriate and the conclusions of the data affect the interpretations and conclusions of a study. A study were unchanged. retraction would correct the scientific record by removing European Science Editing 4 February 2016; 42(1)
Retraction and republication notice, and the journal’s editors drafted the retraction The journal’s editors had several options. We could kill the notice. After approval of the corrected proofs by the paper, by retracting it. The errors were sufficiently extensive respective authors of the research article and linked to potentially damage our confidence in the work. But comment, and extensive proofreading of all versions, the we knew that the findings could make a vitally important retraction and republication was scheduled for print and contribution to advancing the quality of care for one of online publication in the issue of January 31, 2015. the commonest causes of death in China. We wanted to Our attention now turned to the mechanics of publication. preserve the paper in some way. But how? Our proposed Our original intention was to instruct the publisher’s web solution was to retract the original article, then immediately vendor to retract the original article under its original republish it as the corrected version7, along with a web Publisher Item Identifier (PII), in a process the publisher appendix containing the original article and outlining the terms “tombstoning” (used for whenever published content changes made between the two versions. Additionally, as must be withdrawn, for example due to retraction or legal is our usual procedure with retractions, a comment would issues). This would result in the original PDF and XML be written by the journal’s editors outlining the reasons for versions on the publisher’s website being replaced by a retraction, and communicating the procedure to readers.8 ‘RETRACTED’ watermarked version. The corrected version The quoted data in the accompanying comment6 would would then be republished under a new PII. be simultaneously corrected through normal erratum However, after the files for the January 31 issue were procedures. compiled and supplied to the vendor, it emerged that due We communicated this proposal to the paper’s authors to a miscommunication between the production team, the and the peer reviewers of the correction. After some web production team, and the vendor, a new PII had not deliberation, the authors, reviewers, and authors of the been obtained for the republished version of the article. linked comment article agreed that this would be the fairest To avoid delay to delivery of the entire issue, the corrected and most transparent way to proceed. version was therefore republished under the original PII. The in-house copy editor set about preparing the files for Consequently, it became impossible to formally retract the retraction and republication. These comprised: the original version through Elsevier’s tombstoning 1. The original version of the paper, with incorrect procedure. Despite this difficulty, the ‘editorial retraction’ data highlighted was successful, with publication of the retraction notice in 2. The corrected version of the paper, with corrected the issue8 and clear identification of the retracted version of data highlighted the article in the online web appendix of the article. 3. A ‘clean copy’ of the corrected version of the paper. As of January 18, 2016, the study had 13 citations Figure 1 shows the retraction and republication process (including one from before the retraction and for the China PEACE study in The Lancet. republication11b), with an Altmetric score of 139 and a The copyeditor prepared the new files and an erratum Scopus Field-Weighted Citation Impact of 21.2811c.
Figure 1: Editorial retraction and republication of The Lancet China PEACE study
Originally published article
Corrected Corrections data made highlighted
Corrected version Republished, errors highlighted Republished under original PII Included in republished appendix
Errors highlighted
Retracted, errors highlighted Included in republished appendix European Science Editing 5 February 2016; 42(1)
Case 2: The Lancet Respiratory Medicine—full Respiratory Medicine issued an Expression of Concern retraction and republication to alert readers that data presented in the paper on ICU In April, 2014, a study on tracheostomy mortality in mortality were unreliable, with the next steps to be decided intensive care units (ICU) was submitted to The Lancet by a panel of experts.13 Respiratory Medicine. The article was a meta-analytical This panel comprised of peer reviewers of the original investigation, aiming to assess the effect of tracheostomy article and those who had not previously seen the paper. timing (early vs late or no tracheostomy) on mortality in The members of the panel did not agree as to the most critically ill patients who were receiving ventilation. The appropriate course of action: some reviewers felt that article was peer reviewed and revised in several rounds, a correction notice would be sufficient, whereas others accepted, copyedited, and then published online on June 27, disagreed, feeling that the error undermined the results of 2014 ahead of planned print publication in September11,12. the entire paper and that there was no choice but to retract On July 4, 2014, the journal received a letter from two it. The Lancet Respiratory Medicine editors agreed that an researchers who raised concerns about the use of mortality erratum notice would be insufficient, particularly as the data for one of the trials that was included in the meta- interpretation and conclusions of the study had changed as a analysis. The editors of The Lancet Respiratory Medicine result of the data error. Discussions on a possible retraction contacted the authors of the article to inform them of these and republication of the China PEACE study were then in concerns and ask for a response. After some discussion process at The Lancet, and given that the error had been and clarifications between the letter authors, editors, and identified, the corrected findings were indeed reliable, and authors of the article, the authors noted that an incorrect the analysis still represented an important contribution assumption had possibly been made when extracting data to the literature, the Lancet Respiratory Medicine editors for two of the trials in their meta-analysis: patients who proposed to retract and republish the tracheostomy study had not been discharged from the ICU were assumed to to ensure a transparent correction of the record, to which have died in the ICU; however, this assumption was not the study authors agreed. necessarily correct (for example, patients might have been In this retraction and republication, the incorrect transferred to other wards or hospitals, or might still have article, published under the original PII, was successfully been in the ICU when follow-up ended). The authors were retracted via the publisher’s tombstoning process.11 Figure unable to obtain the correct mortality data for the two trials 2 shows the retraction and republication process for the from the original study authors, so re-analyzed their results tracheostomy meta-analysis. The corrected article was using an approximate measure. This reanalysis affected then immediately published, under a new PII14, with a web both the numerical findings and the overall outcome of the appendix containing marked copies of both the retracted study’s co-primary endpoint; the finding of a significant and republished versions. The republication was similarly difference in overall survival between the two tracheostomy accompanied by a retraction comment, outlining the groups in the original paper was not shown by a corrected reasons for retraction and republication15, along with a analysis. On October 14, 2014, the editors of The Lancet correction notice for the linked editorial. The article was
Figure 2: Full retraction and republication of The Lancet Respiratory Medicine tracheostomy meta-analysis
Originally published article
Corrected Corrections data made highlighted
Republished, errors highlighted Retraction Corrected version Republished under new PII Included in republished appendix
Errors highlighted
Retracted version Retracted, errors highlighted Watermarked, remains on publisher’s website Included in republished appendix European Science Editing 6 February 2016; 42(1) retracted and republished in the February, 2015, issue of or commentaries explaining the source of the error and The Lancet Respiratory Medicine. the reliability of the new findings17. The approach reported As of January 18, 2015, the retracted version had here—using appendices to show changes made between no citations and an Altmetric score of 39, whereas the versions—although transparent, does however raise several republished version had six citations, an Altmetric score of difficulties. 15, and a Scopus Field-Weighted Citation Impact of 9.73. An important limitation of retraction and republication is the resources required to implement it. Even though we Discussion benefited from full-time, professional peer review editors, In these case studies, we have presented two uses of a novel production editors, copy editors, and web team editors, publishing mechanism of retraction and republication. most of whom were physically located in the same offices, Although the cases differed slightly, they each followed the practicalities of this retraction and republication were a core set of principles: retraction of the original article; very complicated, with manual editing and proofreading immediate republication of the corrected article; publication of up to four different versions of each paper proving of an appendix, clearly marked with the changes made particularly challenging. It is unclear whether a more between both versions; and publication of a retraction traditional publishing operation—with part-time or notice, outlining these procedures and the reasons for voluntary academic editors, contracted or outsourced copy retraction. editing and proofreading services, and a single publisher The issue of data error presents journal editors with a responsible for a portfolio of journals—would be capable of significant dilemma. Obviously, erroneous data need to be handling such a departure from usual publishing processes. removed from the scientific record as quickly as possible. However, retraction and republication does benefit from However, the best way to do this is often unclear. utilization of several familiar publishing workflows (ie The first option is often to issue a correction notice; retraction, online web appendices). With clear guidelines some journals, including The Lancet and other journals and authorial assistance in preparing the web appendix in the Lancet family, also provide a corrected PDF and (showing the changes made between version), the burden XML version of the article online and, if possible, in print. on the editorial offices of smaller journals could be This option has previously been used by The Lancet for considerably reduced and retraction and republication research articles requiring extensive correction. However, might prove practical in a wide range of circumstances. this strategy is arguably unsatisfactory because it lacks Additionally, online tools such as Diffchecker (https://www. transparency; readers might understandably be concerned diffchecker.com/) could be used to facilitate transparent about the study’s reliability, yet a brief correction notice comparison between the two article versions. would give no space to explain the background of the error, It is unclear how retraction and republication might the material impact it had on the presented data, or the affect a journal’s impact factor, altmetrics, or indexing. steps taken by the journal to ascertain the reliability of the This is likely to depend on whether the corrected version corrected version. is republished under the same PII and title (as for the Further, in these two cases we were faced with a logistical Lancet article) or under a new PII and title (as for the issue in how to communicate the correction purely through Lancet Respiratory Medicine article), and rely on the precise an erratum notice. A summary statement that particular categorization of references to retracted items. Thomson sections of the article “have been amended” would be Reuters Journal Citation Reports uses titles to create unique inappropriately vague, particularly as the HTML/XML and 20-character abbreviations for each individual bibliographic PDF files of the original version of the article on our online entity, with non-title elements (such as city or country of publishing platforms (ScienceDirect and TheLancet.com) publication) to distinguish between entities with similar would be automatically replaced by the updated version. titles.18 Whether retracted and republished items would be However, a full description of the changes made would be classed as separate or linked bibliographic entities under both unwieldy to publish and impenetrable for the reader to this system—and how they affect a journal’s impact factor— interpret. Additionally, for journals or publishers that only remains unclear. Elsevier uses PII numbers to register issue erratum notices and do not correct the published PDF each article’s Digitial Object Identifier (DOI); therefore, or XML files, a correction does not remove the incorrect republished articles with a new PII have a new DOI, and are data from the scientific record. registered as separate articles under the CrossRef system and other indexing systems using DOIs as identifiers (eg Strengths and challenges of retraction and Scopus). The US National Library of Medicine, which republication runs the MEDLINE database, independently indexes The obvious strength of the retraction and republication journal content. Of note, it uses a publication type called is that it combines the best aspects of both retraction and “correction and republication”, which creates a new citation correction, and ensures that the scientific literature is for a republished article and indexes it, enabling the new corrected while not depriving the scientific community and original PubMed citations to be linked; both The Lancet of valuable and reliable research findings. Versions and The Lancet Respiratory Medicine republished articles of retraction and republication or “retraction and are labelled with this publication type. Similar approaches replacement” has previously been reported, but without have recently been adapted by JAMA Psychiatry20,21. These provision of web appendices showing the changes made16 case studies from The Lancet and The Lancet Respiratory European Science Editing 7 February 2016; 42(1)
Medicine, and the JAMA Psychiatry example, were discussed 9 Meng Q, Fang H, Liu X, Yuan B, Xu J. Consolidating the social health at the International Committee of Medical Journal Editors insurance schemes in China: towards an equitable and efficient meeting on November 9–10, in Delhi, India, and changes to health system. The Lancet 2015;386(10002):1484–1492. http://dx.doi. the ICMJE Recommendations will be agreed after further org/10.1016/S0140-6736(15)00342-6. discussion with the National Library of Medicine. 10 Horton R. Offline: Health—the Chinese dream.The Lancet 2016;387 Correction of the scientific literature represents a (10015);212. http://dx.doi.org/10.1016/S0140-6736(16)00041-6. challenging and dynamic aspect of journal publishing. 11 Siempos II, Ntaidou TK, Filippidis FT, Choi AMK. RETRACTED: As research becomes more collaborative and statistical Effect of early versus late or no tracheostomy on mortality of critically methods grow more complex, pervasive correction of ill patients receiving mechanical ventilation: a systematic review and published research could play an increasingly important meta-analysis. The Lancet Respiratory Medicine 2014; published online part of the peer review and publication process, but June 26. http://dx.doi.org/10.1016/S2213-2600(14)70125-0. stakeholder engagement remains crucial19. 12 McCredie VA, Adhikariemail NJK. Early tracheostomy in critically ill patients: still too fast. The Lancet Respiratory Medicine 2014; published Conflicts of interest online June 26. http://dx.doi.org/10.1016/S2213-2600(14)70141-9.113. All authors are employees of The Lancetgroup of journals, 13 The Lancet Respiratory Medicine Editors. Expression of concern.The an Elsevier publication. HC is a board member of European Lancet Respiratory Medicine 2014;2(11):871. http://thelancet.com/ Science Editing. AJ is The Lancet representative to ICMJE. journals/lanres/article/PIIS2213-2600(14)70235-8/abstract. 14 Siempos II, Ntaidou TK, Filippidis FT, Choi AMK. Effect of early Contributions versus late or no tracheostomy on mortality and pneumonia of EG and DS are editors of The Lancet Respiratory Medicine. critically ill patients receiving mechanical ventilation: a systematic RH, AJ, SK, and HW are editors of The Lancet. LP is Senior review and meta-analysis. The Lancet Respiratory Medicine Deputy Managing Editor of The Lancetjournals. HC and 2015;3(2):150–158. http://dx.doi.org/10.1016/S2213-2600(15)00007-7. ZN copyedited the retracted and republished articles. HC 15 The Editors of The Lancet Respiratory Medicine. Retraction and drafted the first version of this manuscript. All authors republication—Effect of early versus late or no tracheostomy on critically revised this manuscript. mortality of critically ill patients receiving mechanical ventilation: a systematic review and meta-analysis. The Lancet Respiratory Medicine References 2015;3(2):102. http://dx.doi.org/10.1016/S2213-2600(15)00005-3. 1 Committee on Publication Ethics. Retraction guidelines. Available 16 Fontanarosa PB, DeAngelis CD. Correcting the literature—retraction at: http://publicationethics.org/files/retraction%20guidelines.pdf and republication (accessed November 13, 2015). 17 Hanna VN, Ahmadb A. Corrected and republished: Suicide in the 2 Van Noorden R. Science publishing: The trouble with retractions. Kurdistan Region of Iraq, state of the art. Nordic Journal of Psychiatry Nature 2011;478:26–28. http://dx.doi.org/10.1038/478026a. 2013;67(2):140–144. http://dx.doi.org/10.3109/08039488.2012.761401. 3 Azoulay P, Bonatti A, Krieger JL. The career effects of scandal: evidence 18 Hubbard SC, McVeigh ME. Casting a wide net: the Journal Impact from scientific retractions. NBER Working Paper No. 21146; May 2015. Factor numerator. Learned Publishing 2011;24:133–137. http://dx.doi. 4 Azoulay P, Furman JL, Krieger JL, Murray FE. Retractions. NBER org/10.1087/20110208. Working Paper No. 18499; October 2012. 19 Retraction Watch. Retraction Watch “mischaracterized the reason for 5 Li J, Li X, Wang Q, Hu S, Wang Y, Masoudi FA, et al, for the China a retraction:” Harlan Krumholz responds to a post. Available at: http:// PEACE Collaborative Group. ST-segment elevation myocardial retractionwatch.com/2015/02/01/retraction-watch-mischaracterized- infarction in China from 2001 to 2011 (the China PEACE- reason-retraction-harlan-krumholz-responds-post (accessed Retrospective Acute Myocardial Infarction Study): a retrospective December 1, 2015). analysis of hospital data. The Lancet 2014; published online June 23. 20 Lopes AC, Greenberg BD, Pereira CAB, Norén G, Miguel EC. http://dx.doi.org/10.1016/S0140-6736(14)60921-1. Notice of Retraction and Replacement. Lopes et al. Gamma ventral 6 Kirtane AJ, Stone GW. STEMI care in China: a world opportunity. capsulotomy for obsessive-compulsive disorder: a randomized clinical The Lancet2015;385(9966):400–401. http://dx.doi.org/10.1016/ trial. JAMA Psychiatry. 2014;71(9):1066-1076. JAMA Psychiatry S0140-6736(14)61033-3. 2015; published online October 28. http://dx.doi.org/10.1001/ 7 Li J, Li X, Wang Q, Hu S, Wang Y, Masoudi FA, et al, for the China jamapsychiatry.2015.0673. PEACE Collaborative Group. ST-segment elevation myocardial 21 Heckers S, Bauchner H, Flanagin A. Retracting, Replacing, and infarction in China from 2001 to 2011 (the China PEACE- Correcting the Literature for Pervasive Error in Which the Retrospective Acute Myocardial Infarction Study): a retrospective Results Change but the Underlying Science Is Still Reliable. JAMA analysis of hospital data. The Lancet 2015;385(9966):441–451. http:// Psychiatry 2015;72(12):1170–1171. http://dx.doi.org/10.1001/ dx.doi.org/10.1016/S0140-6736(14)60921-1. jamapsychiatry.2015.2278 8 The Editors of The Lancet. Retraction and republication—ST-segment elevation myocardial infarction in China from 2001 to 2011 (the China PEACE-Retrospective Acute Myocardial Infarction Study): a retrospective analysis of hospital data. The Lancet2015;385(9966):402. http://dx.doi.org/10.1016/S0140-6736(14)62216-9. European Science Editing 8 February 2016; 42(1)
Essays
Preventing statistical errors in scientific journals
Michèle B. Nuijten Tilburg University, The Netherlands; [email protected]
Abstract since a single psychology paper on average already contains There is evidence of a high prevalence of statistical reporting about ten statistical tests1. In the tangle of statistical output, errors in psychology and other scientific fields. These errors it is imaginable that a p-value (or test statistic or degree of display a systematic preference for statistically significant freedom) is copied incorrectly. Matters probably become results, distorting the scientific literature. There are several worse because many researchers are not in the habit of possible causes for this systematic error prevalence, with double checking their own or their co-authors’ analyses publication bias being the most prominent one. Journal who sometimes do not even have access to the raw data in editors could play an important role in preventing statistical the first place8. However, sloppiness alone does not explain errors in published literature. Concrete solutions entail the apparent systematic preference for significant findings. encouraging sharing data and preregistration, and using A possible explanation for the excess of p-values wrongly the automated procedure “statcheck” to check manuscripts reported as significant is publication bias: significant for errors. results have a higher probability to be published than non-significant results9-11. It is imaginable that researchers Keywords just as often wrongly report a significantp -value as a Statistical errors, publication bias, statcheck, data sharing non-significant p-value. However, because of publication bias, only the gross inconsistencies that wrongly present In a recent study1, we documented the prevalence of a p-value as significant are published, resulting in a statistical reporting inconsistencies in more than 250,000 systematic bias in favour of significant findings. Conversely, p-values from eight major psychology journals, using the it is also possible that researchers suspect that their findings new R package “statcheck” 2. The programstatcheck : converts will not be published if they do not find a significant effect, PDF and HTML articles to plain text files; extracts results of and because of this, they more often wrongly round down null hypothesis significance tests that are reported exactly a non-significantp -value to obtain a significant finding, according to APA style3; recomputes the p-value based on than vice versa. This would be in line with the finding of its accompanying test statistic and degrees of freedom, and John et al12, who found that 22% of a sample of over 2000 checks if the reported p-value matches the recomputed psychologists admitted to knowingly having rounded down p-value, taking rounding of the reported test statistic into a p-value to obtain significance, which would lead to an account. We found that in half of the papers at least one excess of false positive findings. Of course it could also just p-value was inconsistent with the test statistic and degrees be the case that researchers unknowingly maintain double of freedom. In most of these cases, the reported p-value was standards concerning the checking of their results: they only marginally different from the recomputedp -value. would inspect their results with more scrutiny when the However, we also found that one in eight papers (12.5%) result is unexpectedly non-significant, but not when it is contained gross inconsistencies that may have affected the significant. statistical conclusions: in those cases the reported p-value I believe journal editors can play an important role in was significant, but the recomputed p-value was not, or vice preventing, detecting, and/or correcting statistical errors versa. We found a higher prevalence of gross inconsistencies in scientific literature. There are several concrete steps in p-values reported as significant, than p-values reported that could be taken to improve the state of the published as non-significant, implying a systematic bias towards literature. statistically significant findings. A possible solution to the problem of statistical reporting This high prevalence of statistical errors in psychology errors is to promote data sharing. In previous research it papers is alarming, and there is evidence that this problem is has been found that if researchers were unwilling to share not unique to psychology. Similar inconsistency rates have data of a certain paper, there was a higher probability that been found in, for instance, the medical sciences in general4 the paper contained reporting errors, often concerning and psychiatry in particular5. Even though small reporting statistical significance13. This finding could illustrate that errors might be inconsequential, wrongly reporting a authors are aware of the inconsistencies in their paper and p-value of 0.37 as 0.36 will probably not have serious effects, refuse to share their data out of fear of being exposed. An the apparent focus on significant results is worrying and alternative explanation for this finding is that researchers can have far-reaching consequences. It may have added to who manage their data with more rigour both make fewer the excess of (false) positive findings in science6 7. There are mistakes and archive their data better, which makes data several explanations for this high error prevalence. First, sharing easier. In both cases the prevalence of reporting most of the inconsistencies could have been caused by mere errors might decrease if journal editors were to encourage sloppiness. Especially in psychology this is easy to imagine, data sharing. European Science Editing 9 February 2016; 42(1)
Besides the possibility that authors themselves may This research plan is then “registered” somewhere online become more precise in reporting their results if they (eg in a repository for clinical trials such as https://www. have to share their data, encouraging data sharing has clinicaltrialsregister.eu), or even submitted to a journal. In more benefits. If authors would submit their data and the latter case, the research plan is peer reviewed, and if analysis scripts alongside their manuscript, it would allow the plan meets the standards of the journal, the researchers for so-called analytic review14. In analytic review, peer can receive an “in principle acceptance”, no matter what reviewers or statistical experts verify if the reported analyses the results will be – given that they will adhere to the and results are in line with the provided data and syntax. research plan (see eg the guidelines for registered reports Not only will this encourage authors to manage their data in the journals Cortex, Comprehensive Results in Social more carefully in order for a third party to understand it, Psychology, and Perspectives on Psychological Science). This statistical errors that were overlooked at first have a higher way, the decision to publish a paper cannot be influenced probability of being detected before publication. by whether the results were significant or not, avoiding the Editors could decide to make data sharing mandatory, selective publishing of p-values wrongly rounded down as taking into account certain exceptions concerning privacy compared to the ones wrongly rounded up. On top of that, etc (see eg the policy of PLoS One). Another option is to it takes away an incentive for researchers to deliberately simply reward authors who share data. For instance, the report a non-significantp -value as significant. journal Psychological Science awards badges to papers that Besides side-stepping publication bias and avoiding are accompanied by open data and also awards badges for systematic reporting errors, preregistration also solves the open materials and preregistered studies. Although at first problem of HARKing: Hypothesizing After the Results sight these badges might seem trivial, they can be considered are Known16. When researchers are HARKing, they first a quality seal and have inspired many researchers to share explore the data to find interesting patterns, and then their data. present these findings as having been predicted from the Of course, researchers could still conceal deliberate start. If a researcher performs a lot of exploratory tests, he rounding errors towards significance by manipulating or she is bound to find at least one significant result purely the raw data before submitting them. However, falsifying by chance. Reporting only the tests that were significant research data like this is explicit scientific fraud. Data leads to an excess of false positive findings. However, if the from self-reports show that scientific fraud is much more research plan and hypotheses are registered beforehand, uncommon than questionable research practices such as there is a clear distinction between confirmatory and wrongly rounding a p-value12, so it seems implausible that exploratory tests in the paper, which allows for a more encouraging data sharing will result in researchers hiding reliable interpretation of the results17. rounding errors by manipulating the raw data. In any case, To conclude, there is evidence of a high prevalence of there will always remain ways to commit fraud in science, statistical reporting inconsistencies in scientific literature. but encouraging data sharing will definitely make it harder. Even though many of these inconsistencies are minor Another way to avoid reporting errors and to facilitate errors that are probably due to mere sloppiness, there is analytic review, is for editors of journals that adhere to also a high prevalence of gross inconsistencies that may APA reporting style to make use of statcheck2. As described have affected the statistical conclusion, mainly in favour of above, statcheck is a package for the statistical software statistical significance. Even though we can only speculate R15 that can automatically scan articles, extract statistical why there are more results wrongly presented as significant results reported in APA style, and recompute p-values. (deliberately rounding down, publication bias, less rigorous Editors could make it standard practice to use statcheck checks of findings in line with expectations, etc) it remains to scan papers upon submission to check for statistical a worrying finding, reflecting a systematic preference for reporting inconsistencies. This takes almost no time; on “success” and leading to an excess of false positive findings average, statcheck can scan approximately 250 papers per in the literature. minute. Since many journals already have an automatic There are several concrete steps that journal editors can plagiarism check, it is a small step add a check for reporting take in order to avoid or reduce the number of reporting inconsistencies. Results that are flagged as problematic can errors. For instance, editors could encourage data sharing then be corrected before publication. R and statcheck are and preregistration, or use the program statcheck to both open source and freely available. For more information automatically check for inconsistencies during the review about statcheck and an extensive analysis of its validity, see process. Besides decreasing the prevalence of reporting our paper1. For instructions on how to install statcheck, see errors, these measures also reduce publication bias, http://mbnuijten.com/statcheck. HARKing, and other questionable research practices. The excess of results wrongly presented as significant Statistical reporting errors are not the only problem we is probably caused by publication bias. A promising way are currently facing in science but at least it seems like one for editors to try to avoid publication bias is to encourage that is relatively easy to solve. I believe journal editors can preregistration. Preregistration can take many forms, play an important role in achieving change in the system, in but in general the idea is that researchers write a detailed order to decrease statistical errors slowly but steadily and research (and analysis) plan before collecting the data. improve scientific practice. European Science Editing 10 February 2016; 42(1)
References Independent editors – how secure is 1 Nuijten MB, Hartgerink CHJ, Van Assen MALM, Epskamp S, Wicherts JM. The prevalence of statistical reporting errors in psychology your position? (1985-2013). Behavior Research Methods. 2015. doi: 10.3758/ s13428-015-0664-2 Denys N Wheatley 2 Epskamp S, Nuijten MB. statcheck: Extract statistics from articles and BioMedES Ltd (www.biomedes.co.uk) recompute p-values. R package version 1.0.1. http://CRAN.R-project. Member of the International Advisory Board, European org/package=statcheck2015. Science Editing 3 American Psychological Association. Publication Manual of the American Psychological Association. Sixth Edition. Washington, DC: Prior to the start of the millennium, scientific papers were American Psychological Association; 2010. published largely as hard copy, usually by editors working 4 Garcia-Berthou E, Alcaraz C. Incongruence between test statistics and for particular organisation and societies, which either paid p-values in medical papers. BMC Medical Research Methodology. to have them published by firms such as Academic Press 2004;4:13. doi: 10.1186/1471-2288-4-13 and Wiley, or more usually the journals were sent out 5 Berle D, Starcevic V. Inconsistencies between reported test statistics and to subscribers (usually institutional libraries rather than p-values in two psychiatry journals. International Journal of Methods individuals) who paid for each one or for a “basket” of those in Psychiatric Research. 2007;16(4):202-7. doi: 10.1002/mpr.225 in a similar field. Most publishers did not see any particular 6 Francis G. The frequency of excess success for articles in Psychological profit in providing this service for the scientific community. Science. Psychonomic Bulletin & Review. 2014;21:1180-7. doi: 10.3758/ But all this was soon to be quickly superseded by electronic s13423-014-0601-x publishing, with almost no journals continuing to produce 7 Fanelli D. “Positive” results increase down the hierarchy of the sciences. only hardcopy, and increasingly fewer offering both PLoS One. 2010;5(3):e10068. doi: 10.1371/journal.pone.0010068 hardcopy and online versions, as had been the case with Cell 8 Veldkamp CLS, Nuijten MB, Dominguez-Alvarez L, van Assen Biology International in the last few years, for which I was MALM, Wicherts JM. Statistical reporting errors and collaboration Editor-in-Chief for 14 years before recently demitting office. on statistical analyses in psychological science. Plos One. Around 2000-2001 when electronic online publishing 2014;9(12):e114876. doi: 10.1371/journal.pone.0114876 began in earnest, there was a rapidly growing demand for 9 Greenwald AG. Consequences of prejudice against the null hypothesis. publication, and the “author-pays” model took off. Authors Psychological Bulletin. 1975;82:1-20. were charged for papers that were accepted for publication, 10 Sterling TD. Publication decisions and their possible effects on and the papers were immediately made accessible to readers inferences drawn from tests of significance–or vice versa.Journal of free of charge once online. Although the converse method the American Statistical Association. 1959;54(285):30-4. still remains available, ie the reader or the institution pays to 11 Sterling TD, Rosenbaum WL, Weinkam JJ. Publication decisions download the full paper (only the abstract being free), this revisited - The effect of the outcome of statistical tests on the decision has become less and less common. Crucially, the author- to publish and vice versa. American Statistician. 1995;49(1):108-12. pays system means that publishers get paid up-front. 12 John LK, Loewenstein G, Prelec D. Measuring the prevalence of Taking an overview of the present situation, publication questionable research practices with incentives for truth-telling. costs are miniscule compared with the cost of hardcopy Psychological Science. 2012;23:524-32. doi: 10.1177/0956797611430953 journals published before the 21st century. Unlike authors 13 Wicherts JM, Bakker M, Molenaar D. Willingness to share research of novels and other publications, who sell their work or data is related to the strength of the evidence and the quality of receive royalties, scientists and other academics do not get reporting of statistical results. PLoS One. 2011;6(11):e26828. doi: paid for their contribution to the literature; on the contrary, 10.1371/journal.pone.0026828 they pay for the privilege. And the costs can be hefty, with 14 Sakaluk J, Williams A, Biernat M. Analytic review as a solution publishers charging thousands of pounds or euros per to the misreporting of statistical results in Psychological Science. accepted article. The scientific paper has become a cash- Perspectives on Psychological Science. 2014;9(6):652-60. doi: cow, which is why there has been a burgeoning of new 10.1177/1745691614549257 journals. Many less reputable outfits have jumped on the 15 R Core Team. R: A Language and environment for statistical band-wagon – the predatory journals. Beall’s list shows how computing. http://www.R-project.org/2014. prolific these have become as the entries are coming close to 16 Kerr NL. HARKing: Hypothesizing after the results are known. a thousand; see http://scholarlyoa.com/publishers/. Personality and Social Psychology Review. 1998;2:196-217. Now I come to the crux of this article, which concerns 17 Wagenmakers EJ, Wetzels R, Borsboom D, Maas HLJvd, Kievit RA. An my own experience as an independent editor. When these agenda for purely confirmatory research.Perspectives on Psychological changes in electronic publishing began, some publishers Science. 2012;7:632-8. doi: 10.1177/1745691612463078 created “core” journals edited by their own staff. However, they also encouraged the creation of new journals that Author’s note would be independently edited. This meant that someone The preparation of this article was supported by VIDI grant not employed by the publisher would work hard to establish number 016-125-385 from the Netherlands Organization a journal, raise an impact factor and get a healthy submission for Scientific Research (NWO). rate of good articles. Anyone who has attempted this exercise will know how much time and effort is required to succeed. In my own case, I created two new journals, European Science Editing 11 February 2016; 42(1) both of which are now in good shape, viz. Theoretical There were two further concerns. First, the publisher’s Biology and Medical Modelling (IF 0.95) and Cancer Cell sub-editor, who had my journal in her portfolio, waded International (IF 2.77). The former has since had two in without my permission almost immediately after the editors-in-chief, both highly competent experts who took termination notice was sent, effectively pushing aside the up their posts after invitation. Both journals are “owned” by editorial role that my assistant and I had been operating the same publisher, and the editor-in-chief receives a small smoothly for nearly 15 years. Second, the publisher had annual honorarium and a little of the article processing already decided who they wanted to take over, and this charge (APC) paid by the authors of accepted papers. This person was not identified until well through the 6-month recompense, however, amounts to only about 10% of the termination period. No information was provided about heavy charges levied by the publishers. During the last 10 the new editor and his standing in the field of the cell years or so, the remuneration of editors per accepted article biology of cancer (necessary for a journal called Cancer has remained essentially unchanged, whereas the APC has Cell International). Passing on an editorship usually at least doubled, hardly fair on those who have to do the involves extended consultation between the editor and the most work, ie the editors, some - but not all - of whom will publishers, since it is essential that the journal is passed be helped by paid assistants. on to trustworthy and competent persons with sufficient Independent editors need to be respected for the hard experience of editing. work put into creating and running successful journals Independent editors must be made aware that under while maintaining the high standards of research integrity. present day publishing practices they can find themselves The fewer manuscripts accepted by a journal, the smaller the out of office at short notice, as has been experienced salary becomes. If weaker manuscripts start being accepted, by others with little or no redress3. The case for legal an independent editor’s income will increase but standards regulations in such circumstances has been well argued by will fall. This raises another issue regarding the treatment Matko Marusic4. Therefore it is particularly important that of independent editors by publishers; the publishers are a contract or “agreement” between an independent editor there to make money, not primarily for the science or its and the publishing company (usually on an annual basis) is veracity. The more papers are published, the greater the carefully drawn up and renegotiated as necessary each year, publisher’s profit. Pressure on independent editors to relax along with negotiations on an increase in remuneration their standards is therefore inevitable. at least commensurate with any increase in APC by the Independent editors have an agreement with a publisher publisher. Second, the business considerations mentioned regarding the continuation of the journal annually. Since above (particularly in the case of predatory journals) can the publishers “own” the journals, agreements can come to seriously compromise and undermine the integrity of a close by either party provided it is mutually understood scientific publications. This problem needs to be recognised why this is being done; this should be through a process before it gets much worse, and this is why WAME and other which is fair and ethical. WAME (www.wame.org/), COPE editorial organisations have asked that instances such as (www.publicationethics.org/) and other similar bodies in this are made public. several countries have set out regulations regarding the Finally, it seems odd that many independent editors, like way in which independent editors should be treated.1 But myself, can be and have been exploited in helping to fill the business is business, and independent editors can find coffers of wealthy publishing houses. We seem to act as the themselves out of a job. These regulations are not readily “piggy-in-the-middle” that gets the raw deal. This raises enforceable, but when they are flouted, cases need to be the question of who is truly important in the publishing of brought to the attention of everyone2, especially to other scientific articles. Cannot independent editors provide the independent editors, who must be aware that their editorial scientific fraternity with a better service than having authors position and independence can be quite fragile. spend a great deal of valuable funding to get their papers In my case, not one of these regulations was taken into published? Thanks to ever advancing IT, there are other consideration, and therefore there was no redress after I ways of going about this business that can greatly improve was literally dumped by the publisher as the chief editor of the situation compared with the now “conventional” online my journal. According to the WAME regulations, the main publication process. issues that were ignored are as follows: 1. There was no negotiation before the termination References notice was sent. 1 European Science Editors’ Handbook, 2nd Edition, 2013. Several 2. All parties involved were not informed. chapters, notably 6.10. 3. No clear reasons were given for terminating the 2 Poynder R. Open and Shut. Follow his Blogs at http://poynder.blogspot. agreement. co.uk/. Many deal with open access issues. 4. Notice of termination should have come from the 3 http://www.oapublishinglondon.com/news/43# senior management of the overall owning company, 4 Marusic M. The importance of legal regulation for scientific journals. based on information from both sides following prior European Science Editing. 2015;41(2):36-38. discussions (some major publishers having a small committee dealing with such matters). European Science Editing 12 February 2016; 42(1)
Computer tools for preparing stylesheets
Paul Beverley Archive Publications, [email protected]
Abstract: When editing technical and scientific documents, spelling generally, (d) the use of accents and (e) a range of a stylesheet is an essential tool for maintaining consistency, other issues including punctuation and capitalisation. both internally within the document and with the publisher’s house style. Paul Beverley has provided some Hyphenation issues free computer tools, in the form of Word macros, that will With the ‘non(-)linear’ example in mind, Table 1 shows analyse a document and highlight potential inconsistencies. an extract from a Word file that my macroHyphenAlyse This allows the stylesheet to be populated early in the editing generated for one book that I was editing. process, thereby saving time and allowing the editor to spot inconsistencies that might otherwise have been missed. Table 1: Part of a hyphenation frequency list produced by The tools cover inconsistencies with hyphenation, spelling the HyphenAlyse macro (including proper nouns) and accents, and certain aspects non-circular…1 of punctuation and capitalisation. non-collinear…2 non collinear…2 Keywords: consistency, hyphenation, macros, spelling, non-commutative…3 stylesheets non-commuting…2 non-conforming…5 nonconforming…3 This essay describes the use of some free computer non-conservative…4 programs that I have created to help editors with the process non-equilibrium…4 non equilibrium…1 of creating a stylesheet, an essential tool for maintaining nonetheless…2 consistency, both internally within the document and with the publisher’s house style. It is important to make these non-euclidean…1 style decisions early in the editing process, and so using non-existent…1 the computer’s power to analyse your document is a very non-existing…1 time-efficient way to assess the document’s current (in) non-isothermal…1 consistency. non-issue…3 The creation of this software has its their origins in non-linear…5 nonlinear…207 a situation that you might recognise. When editing a book, I noticed that in chapter 1 the author had mainly non-linearity…1 nonlinearity…18 used ‘nonlinear’ but occasionally ‘non-linear’, so I applied non-negative…6 ‘nonlinear’ consistently. Unfortunately, the frequency non-negativity…2 of ‘non-linear’ increased in chapter 2, but I continued ‘correcting’ it. By chapter 3, I could see that the author was First, the macro has listed all the hyphenated words in now consistently using ‘non-linear’. I then had to decide the whole book (Table 1 only shows a small extract) and has whether to persevere with ‘nonlinear’, or go right back given me the frequency of each, but also if they appear a through the first two chapters. single words and/or as two separate words. In addition, it I realised then that making the effort to check this sort has also found the frequency of all the words that begin with of thing before I started to read would save time in the anti, hyper, inter, non, etc, whether they are hyphenated not. long run. So over the past few years, I have developed I can look through this list and see what the author a number of computer tools that check for several forms has done, but from the extract in Table 1, you can see that of inconsistency, and I make these tools available free of the macro tries to draw my attention to those words that charge to fellow editors, in the form of Word macros, from: appear both as a single word, and in hyphenated form (eg www.archivepub.co.uk/macros. nonlinear/non-linear), because those are the words most These are computer tools and so they can only follow likely to represent inconsistencies in the text. rigid rules and point up discrepancies that might possibly By looking through this list I am able to see the author’s be inconsistencies; it’s up to you as the editor with an (in)consistency as regards hyphenation and make decisions understanding of the vagaries of the English language to about whether some, all or none of the words using, say, the interpret the findings. ‘non’ prefix should be hyphenated. I can then list these as The analysis is done using five main macros, plus a part of a word list in my stylesheet – see Table 2. And having number of smaller ones, covering possible inconsistencies made these sorts of decisions before starting to edit chapter with: (a) hyphenation, (b) the spelling of proper nouns, (c) 1, time-wasting problems over hyphenation can be avoided. European Science Editing 13 February 2016; 42(1)
Table 2: An extract from a word list, forming part of a Table 4: Some possible proper noun inconsistencies stylesheet detected by ProperNounAlyse hyper
Varying use of accents serial (or not) commas, punctuation (e.g. p6/p.6/p 6/p. 6 My tools were written for my own use with the English or Eqn 2.1/eqn 2.1/Eqn. 2.1/eqn. 2.1/Equation 2.1), and language, but I gather (not being a linguist) that some many others. Of course, some of these variations (possible of them can be used with documents written in other inconsistencies) might not occur in a particular document, languages. What’s more, at the suggestion of a group of but the macro only lists those that do represent possible editors that I met in Spain, I wrote the AccentAlyse macro, inconsistencies. which looks for possible inconsistencies such as those shown in Table 5. Other useful tools One other important related macro is MultifileWord. Table 5: A set of word pairs showing different accent use This is used when your document arrives as a number of Ängquist ...... 1 separate chapter files. Clearly, to do the analyses above, you need one big file of all the text from the whole document. Angquist ...... 2 MultifileWord allows you to look into a folder and combine the text from all (or a selection) of those files. CIGRÉ ...... 1 With very big files, some of the analyses can take quite CIGRE ...... 47 a long time to run, so one way to speed things up is to use MultifileText instead of MultifileWord. This alternative macro creates a pure text version of the combined file Dennetiere ...... 1 which, since there is no formatting information, will be Dennetière ...... 8 smaller and easier for Word to handle.
Worth the effort? Dube ...... 4 Using these macros does take time and effort – both initially Dubé ...... 4 to learn how to install and use them, and also to apply them to any given job, so only you can decide if any or all are edition ...... 1 worthwhile. Certainly, the longer the job the bigger the pay- Édition ...... 1 back from the preparation time. But regardless of the time saved, these tools enable me to produce a better standard of work (a higher level of facade ...... 1 consistency), which seems to me a good investment. façade ...... 3 These macros are all available, including instructions, in the form of a book from my website (http://www. archivepub.co.uk/TheBook). If you are new to macros, you Gerin-Lajoie ...... 1 might like to try the ‘Proofreader’s Pack’ (equally applicable Gérin-Lajoie ...... 2 to editors), which gives you very simple instructions for loading and installing a set of macros (including most of Saint-Germain-des-Prés ...... 1 those mentioned here). These will get you off to a good start (www.archivepub.co.uk/documents/ProofingPack.zip). Saint-Germain-des-Près ...... 1 And if you have other ideas for programs that could make you more efficient and effective, do please let me Other inconsistencies know – I’m adding more macros to my list all the time. Finally, DocAlyse produces a frequency list of a whole range of different consistency issues, such as capitalisation (eg Paul Beverley has been an editor and proofreader of technical Chapter 6/chapter 6, Section 4.3/section 4.3), alternative documents for over ten years. He’s partly retired now, but spellings (eg focus(s)ed, co(-)operate, learnt/learned, etc), doesn’t want to stop altogether because he enjoys his work far too much. European Science Editing 15 February 2016; 42(1)
Viewpoint
Multiple: do we know what we are talking about?
Elise Langdon-Neuner Freelance editor and publications consultant, Vienna, Austria; [email protected]
Have you noticed a dramatic increase in the use of the the word for its precise meaning is especially important in word multiple in recent months? The BBC reported that science, where the word has been coupled to some nouns the group New World Hacking, which brought down its for a meaning of this nature: website, said at the beginning of this year, “It was only a Multiple choice is a form of assessment in which test, we didn’t exactly plan to take it down for multiple respondents are asked to select the best possible answer (or hours.” Another news report, this time from The Daily answers) out of the choices from a list (Merriam Webster) Beast in November last year, stated, “Late Friday night in Multiple regression is a statistical tool used to derive the Paris, multiple gunmen opened fire on diners and concert- value of a dependent variable (criterion) from several other goers.” It seems almost every news item must have the word independent (predictor) variables. It is the simultaneous multiple. This morning (14 January 2015) it was “Multiple combination of multiple factors to assess how and to bomb and gun attacks in Jakarta”. I suspect the word has what extent they affect a certain outcome (https://www. become popular with journalists because it sounds more techopedia.com/definition/27369/multiple-regression). sensational than many, several, numerous, various etc. But it Multiple exposure is the superimposition of two or more is replacing those words in everyday speech too. And, more exposures to create a single image (https://en.wikipedia.org/ to the point, its use in academic circles has also increased. wiki/Multiple_exposure) The trend has been creeping up on us: Google Ngram An element of complexity is explicitly added into Viewer shows that the frequency of multiple in its corpus multiple in medical writing. When qualifying a disease or of fiction books increased 2-fold from 1975 to 2008 and injury multiple means the disease is complex in its nature the increase in PubMed® articles was 2.3-fold from 1975 to or effects; affecting several parts of the body (Oxford 2010.1 (For an explanation of how the author used Google Dictionary). Multiple sclerosis, a chronic disease that Ngram Viewer and PubMed to obtain these statistics see2.) attacks different parts of the central nervous system, namely However, my impression is that the trend is becoming an the brain, spinal cord and optic nerves, is an example. explosion. Multiple’s precise meaning is betrayed by its general The word multiple can be a noun, with a meaning increase in use as a substitute for many, several, numerous, confined to mathematics, or an adjective. We are concerned various etc. We should be careful not to confuse readers by here with the adjective multiple. It means, “consisting of, writing multiple when all that is meant is one of these words. having, or involving several or many individuals, parts, Alistair Reeves gave an example of the title “Managing elements, relations” (dictionary.com but other dictionaries dyslipidaemia—multiple patients and multiple approaches”. give similar definitions).Multiple is therefore more complex While ‘multiple approaches’ indicates the approaches were than many/several . The Macmillan dictionary gives the different in his view, he asked if ‘multiple patients’ meant following example of the usage of multiple: more than one patient or patients with dyslipidaemia of “Words can have multiple meanings”. Fair enough, different origins.2 If more than one patient presumably words can have different types of meanings but how is the they were Siamese twins. Multiple is the indefinite version sense of this sentence different from “Words can have many of single, double, triple and so forth, which might translate meanings”? The many meanings must be different. Alistair’s title to something like “Triple approaches (3 I tried the web service at http://the-difference-between. treatment arms) to sextuple patients (N = 6)”. Talking of com. It told me multiple means “having more than one approaches, how has the term repeat-dose study, meaning element, part, component, or function” and many means participants receive more than one of the same dose, “an indefinite large number of”. The Merriam Webster become interchangeable with multiple-dose study, which and Oxford dictionaries make a similar distinction. But Alistair suggested should mean participants receive multiple is not finite either. So, is many quantitatively more different doses? Ifmultiple can include elements that are than multiple? Not according to the Cambridge dictionary, the same or different, we are left with multiple-dosea study which defines multiple as “very many of the same type, or of being one where very many more doses are given than in different types”. This definition makes multiple bigger than a repeat-dose study or the doses are of an infinite number. many—very many: “Words can have multiple meanings” = I have encountered manuscripts reporting on “multiple “Words can have very many meanings”? animals”, which were just a bunch of ordinary laboratory I suggest restricting the use of multiple to where we mice—not genetically engineered animals created from a want to emphasize the element of “several/many in one”. I variety of species. Elsewhere I found multiple meetings, also think that the several/many can be of the same or of all with the same purpose and following the same format: different types as the Cambridge dictionary indicates. Using The forms constituting the electronic health record were European Science Editing 16 February 2016; 42(1) developed based on individual professional experience during types, but in such cases I would suggest keeping our hands multiple dedicated working group meetings held by the off multiple and describing exactly what we mean. Overuse ophthalmologists working at the diabetic retinopathy consult. and misuse of multiple will only make it unusable and lead I changed this sentence to: The electronic health record to its disuse. forms were developed from input given by ophthalmologists from the diabetic retinopathy department during several References working-group meetings held for the purpose. 1 Goodman NW, Edwards MB, Langdon-Neuner E. Medical writing: A I chose to replace multiple with several but if there had been Prescription for Clarity 4th ed. Cambridge: Cambridge University a large number of meetings many would be appropriate and Press; 2014 if the number had been even larger numerous, after which 2 Goodman NW. The increasing pseudodignification of medical prose. we get into the realms of an enormous/vast/gargantuan ESE 2015;41(2):31-35. Available at: http://www.ease.org.uk/sites/ number of meetings—but not multiple meetings, unless default/files/article_goodman_0.pdf#page=3&zoom=auto,-158,481 maybe we are concerned with more than many meetings, 3 Reeves A. Multiple problem: Different or more of the same? TWS an infinite number of meetings or meetings of different 2007;16(3):122-123
New member of the editorial board ESE would like to welcome Laurence Mabile as a new member of the editorial board. Laurence will be responsible for the book review section of the journal. 2016 Asian Conference Laurence Mabile graduated as PhD in Medical Biochemistry “Best Practices in Scholarly in 1995 and has joined the team “Genomics, Publishing” biotherapies and public health: multidisciplinary approach” directed by April 4-5, 2016, Singapore Dr Anne Cambon- Thomsen, initially as The conference agenda has been confirmed the BRIF ‘Bioresource and registration is now open for ISMTE’s first Research Impact Factor’ Asian conference. project manager, in the department of More information is available at Epidemiology and Public Health at University of Toulouse III, France. She studied at the University “Pierre et Marie www.ismte.org/?page=2016AsianConference Curie” in Paris and at the University “Paul Sabatier” in Toulouse. She held fellowship positions in UMDS Guy’s Hospital in London, UK and in the Clinical Research Institute of Montreal, Canada, in the research context of Nutrition and Diseases. She also holds a master in “Museology: Sciences, Cultures and Societies” (National Museum of Natural History, Paris). She is interested in many societal aspects of research in life sciences, notably in the way science is disseminated. European Science Editing 17 February 2016; 42(1)
Reports of meetings
Looking at New Perspectives–2015 annual meeting of the Society for Scholarly Publication 27–29 May 2015, Arlington, Virginia, USA
The annual meeting of the Society of Scholarly Publishing (SSP) was held between 27th and 29th May at Arlington, VA, USA. The theme of this year’s meeting wasThe New Big Picture: Connecting Diverse Perspectives. The attendance for this year’s meeting was >900, which to date has been the highest attendance and indicates the diverse audience this meeting has been attracting year on year. There were over 150 speakers at the meeting with a mix of keynote speeches, concurrent meetings and exhibits by various SSP members. This year, I had the opportunity to both participate and exhibit at this year’s meeting, and it was a truly enriching experience to collaborate and network with editors, authors, companies and other professionals from multiple countries. Interestingly, this year is the 350th anniversary of Philosophical Transactions, the first scientific journal that (PLOS), Judy Leather (Informed Strategies), Alice Meadows was published. Thus, this meeting was a very good forum (OrCID), Kari Roane (University of Chicago Press) and to understand how the industry has evolved in the past Jamie Hutchins (IOP Publishing). During the session, the 350 years as well as the challenges publishers are facing speakers extensively discussed the newer aspects (technology in ensuring that they are able to sustain themselves in an and business models) of journal publishing, the introduction increasingly diverse market. Moreover, as this was my first of newer journals every year, as well as the various features time at the SSP conference, it was a great experience to know and formats in which content is being provided by journals. and understand the career path for an editor in this industry. Moreover, they stressed the impact of various service Considering the overall time constraint and multiple sessions providers on the publication industry, eg OrCID, which that are being conducted simultaneously at this meeting, helps authors ensure that their work is noticed, tracked and attending all of the sessions was not possible; therefore, we followed by a particular audience. had already planned the sessions that we intended to attend The session on Journal Publishing – Operational Focus had before the meeting. If you are considering attending the next Helen Arkins (PLOS), Linda Sussman (Cold Spring Harbor meeting at Vancouver, BC, Canada, I suggest you pre-select Laboratory Press) and Judy Hum-Delaney (Canadian the sessions you wish to attend, without fail. Science Publishing) as speakers. The discussions were on The highlights of this year’s meeting were the keynote the production-related processes that have been automated speeches and the inclusion of speed networking sessions by and/or optimized to ensure reduction in the time taken for the organizers. The speakers selected for this year’s keynote publishing papers. Helen Arkins discussed the improvements speeches were Charles Watkinson (a librarian), Ken Auletta PLOS has made to their production processes, which has (a writer), and Jennifer Lawton (ex-CEO of MakerBot); helped them in resolving manual workarounds in their this selection especially highlighted the theme of ‘diverse systems, which was done via the “Process Improvement perspectives’. Moreover, at this year’s meeting, there were Initiative” of PLOS. Because of this initiative, PLOS several engaging panel sessions that provided a live platform achieved improvement in the quality of publishing, started for the audience to share their views with the panel. The co-ordinating with service providers and reduced the overall novel inclusion at the meeting was the real-time interaction time taken for publication after acceptance of any paper. of the graphic reporter Greg Gersch who engaged with the This year’s concurrent sessions included excellent topics SSP community in person and online, to bring the ideas such as The Researcher’s New Big Picture, The Future of and themes of the meeting to life and colour in real time Scholarly Publishing, The Future of Publishing: A University’s with large graphic murals on public display while he was View, Where Are All the Users Going? and How Today’s Startups creating them. are Changing Tomorrow’s Scholarly Communication. Each On 27th May, the pre-meeting seminars included Publishing session was well attended and had participants from diverse 101: The Business of Publishing, Smart Linking & Supplemental backgrounds like societies, publishers, vendors and journals. Data: Turning Funder Data Mandates into Opportunity, During this year’s meeting, gender diversity in the Journal Publishing – Operational Focus, and Implementing scholarly publishing industry received due attention with Dr. Next Generation ID Standards for the New Machine Age: “The Amy Brand presenting on the Demographics and Education Ties That Find.” I attended the session on Publishing 101 – The of Scholarly Publishing and Communication Professionals. In Business of Publishing, which was conducted by Kerry Koffe fact, one of this year’s keynotes was from Jennifer Lawton who European Science Editing 18 February 2016; 42(1) talked about her experiences from a career in technology at Editors as promoters of good practices the executive level and her path to success. in bioresource research Among the concurrent sessions, The Researcher’s New Big Picture had authors themselves as speakers, which provided 9 October 2015, Toulouse, France them with an opportunity to share the inherent challenges they face when they have to publish a paper in any journal. Soon after the development and publication of the guideline For example, authors mentioned issues such as difficulty to standardise the Citation of BioResources in journal in template usage and how it was frustrating to ensure Articles (CoBRA)1, a workshop took place in Toulouse adherence to formats while publishing. Many of the authors on the 9th October 2015. Bioresources are collections of also highlighted the fact that they feel disengaged when biological samples with associated medical, epidemiological, communicating with editors, and thus, journals need to biological or social data (biobanks), as well as collections examine communication at the author–editor level to assure of data of biological origin (databases) or bioinformatics authors that their work is being examined in an effective and tools. Many important biomedical publications refer to data time-bound manner. The contributors for this session were obtained from collections of biosamples, but the recognition from various institutions such as University of Maryland of such resources is often neglected or highly heterogenous. College Park, AAAS/ASM, and American University. The CoBRA guideline introduces a standardised citation of One of the most delightful sessions for me was Where bioresources in scientific publications, and was developed Are All the Users Going? This session addressed the issue within the BRIF (Bioresource Research Impact Factor) with the current business model of publishing, which is initiative, aiming to improve transparent reporting of completely based on the accessibility and visibility of the bioresource-based research, proper sharing and optimal use research published in journals to the end users of any and of bioresources2. The guideline recommends the citation of every journal. To be aware of their current target audience, each bioresource according to a specific reference structure, publishers have to keep evaluating how they are receiving providing specific examples for its adoption. It aims to trace the traffic on their websites and what possible changes occur in use of bioresources, encourage their sharing and create tools user behaviour over a period of time. This helps them retain to recognise the importance of their contribution to research. advertisers and prolong the use of the subscription-based The CoBRA is included in the list of reporting guidelines model. The users also help assess and validate the efforts of the EQUATOR Network (Enhancing the QUAlity and of any publisher and ensure that the products they have on Transparency Of health Research, www.equator-network. the market are used over a period of time. Importantly, the org/), that supports editors in promoting and practising session addressed the issue of steady fragmentation in the responsible research reporting in their journals. manner in which a journal gets user traffic. By evaluation The workshop was organised by the European Association of data, it has been observed that there is a steady increase of Science Editors (EASE), the Istituto Superiore di Sanità- in traffic through NIH, PubMed and Twitter. However, Italian National Institute of Health (ISS), the Institut National e-mails are still the most important generators of web de la Santé et de la Recherche Médicale (INSERM), the traffic for publishers; thus, it is very important that e-mail Université Toulouse III-Paul Sabatier, the Centre National de etiquette and author communication keeps improving with la Recherche Scientifique (CNRS) - Department of Scientific time from the publisher end. and Technological Information (DIST), and the Biobanking To conclude, this year’s meeting truly highlighted the and BioMolecular resources Research Infrastructure- diversity of the scholarly publishing industry. The variety of European Research Infrastructure Consortium (BBMRI- companies, vendors, publishers, societies and authors that ERIC). It was also partially supported by the European attended the meeting demonstrates how popular this meeting Commission Seventh Framework Programme. has become and how it encourages professionals to network Based on a positive long-lasting collaboration between and grow in the future. The keynotes and the concurrent the BRIF initiative and EASE, the workshop gathered meetings covered the challenges and opportunities we need some of the main experts in the field, who discussed how to address in the coming years, from the number of new journal editors, research institutions, researchers, and other products to basic issues like author–editor communication. stakeholders could promote the use of the CoBRA guideline I am definitely looking forward to next year’s meeting and in general the best practices in scientific research. and hope to see you all there too! The morning session was dedicated to presentations by researchers involved in the BRIF initiative, in the EASE, in Acknowledgements the EQUATOR Network, and in the BBMRI-ERIC. They I thank Anupama Kapadia (Senior Manager, Crimson all remarked upon the importance of a strong collaboration Interactive) for her guidance in developing this report. between editors and scientists, that worked successfully and led to the development of the CoBRA guideline. The Ishan Dave guideline was analysed from different points of view: that Senior Business Associate, Enago Crimson Interactive, of the editors as gatekeepers of science and the power of a Mumbai, India multidisciplinary approach (Paola De Castro, EASE and [email protected] ISS); its role as a key element for the recognition and the assessment of the use and research impact of bioresources (Elena Bravo, ISS); how CoBRA is integrated in the vision European Science Editing 19 February 2016; 42(1) of the BRIF initiative (Anne Cambon-Thomsen, INSERM- officials responsible for research integrity (for example, Université Toulouse III); the BBMRI-ERIC value as a new the network of French universities contacts for scientific governance tool for biobanking in Europe (Jan-Eric Litton, integrity, and the US National Organization of Research BBMRI-ERIC); and the CoBRA in the context of other Development Professionals). The groups also recognised research guidance (Iveta Simera, EQUATOR Network). the important role that funders could have in promoting In the second part of the workshop, space was given to CoBRA. They suggested that grant agencies should include all participants, who were asked to give their contribution to CoBRA reporting as part of the grant application and the discussion and were then split into three working groups. yearly research reporting. Finally, it was considered part They focused and discussed the following three issues: of the CoBRA dissemination strategy that other reporting 1. How can editors enhance the implementation of the guidelines mention and refer to CoBRA. For this reason, CoBRA guideline? the developers of other guidelines - through the EQUATOR 2. How can researchers enhance the implementation of Network - should be notified of the CoBRA guideline, and the CoBRA guideline? include it in the next revision of their own. 3. How can universities, research institutes or research The working groups explored other main issues related infrastructures incentivise researchers to use CoBRA? to the CoBRA promotion strategies. One of them referred Referring to point 1, the working groups suggested that to the role of biobank and bioresource managers, who have the guideline could be added in the instructions to authors a strong interest in implementing CoBRA. They could for and to reviewers, and/or in the reviewers’ checklists so that instance include the guideline in MTAs, and also reprimand the editors can make sure authors properly cite the use researchers who do not cite them according to the guideline. of samples in their submitted manuscripts. The CoBRA In addition, referencing software and citation platforms guideline could be added in the list of reporting guidelines, should be updated to allow the citation of bioresources, such as the one in the EMBO journal. Furthermore, in the and CoBRA should then be promoted at companies and submission process a compulsory section could be added websites such as Reference Manager, Mendely, Zotero, on the ‘use and description of biological samples’. The idea ReadCube, ResearchGate, CrossRef, and Orcid. Finally, the of a ‘CoBRA compliant publication’ stamp or logo was also fundamental role played by the patients’ associations that are suggested. The BRIF community, and more specifically often involved in the creation of biobanks was recognised. the people present, should also disseminate the existence These associations should be informed about the CoBRA of the guideline as much as possible using their networks initiative, so that they can follow, through the citations, the and personal contacts to reach journal editors, suggesting research based on the biobanks they participate in. they write editorials about CoBRA. Some participants also The workshop was full of ideas. The audience became pointed out that authors would be more likely to notice, and deeply involved in discussions and many proposals and adhere to, CoBRA (or any reporting guideline) if it were suggestions came from it. So, with a wide consensus and a mentioned in the Methods sections of papers. great help from everyone, the roots were put down for further In the next discussion about point 2, participants action to promote and disseminate the CoBRA guidelines in suggested that a new ‘document type’ could be added to the most effective ways. reference managers such as Mendeley, Zotero or Endnote. This would enable researchers to cite the bioresource in References an automatic manner when they write their manuscript 1 Bravo E, Calzolari A, De Castro P, Mabile L, Napolitani F, Rossi AM, directly on the text editor. This point was also analysed Cambon-Thomsen A. Developing a guideline to standardize the under a different perspective, that is how researchers can be citation of bioresources in journal articles (CoBRA). BMC Medicine encouraged to use CoBRA guideline, and the important role 2015;13:33 of professional associations of writers and editors was then 2 Mabile L, Dalgleish R, Thorisson GA, Deschênes M, Hewitt R, Carpenter strongly remarked upon. It was suggested that associations J, Bravo E, Filocamo M, Gourraud PA, Harris JR, Hofman P, Kauffmann and scientific and relevant professional societies present F, Muñoz-Fernàndez MA, Pasterk M, Cambon-Thomsen A; BRIF CoBRA at their national/international meetings, in their working group: Quantifying the use of bioresources for promoting their newsletters and websites, all considered useful means sharing in scientific research. Gigascience 2013, 2(1):7 for spreading news about CoBRA. In addition, video presentation or webinars about CoBRA could be produced, Anna Maria Rossi, Paola De Castro, Elena Bravo, Alessia which these associations could link to or include as part of Calzolari, Federica Napolitani their educational material. Istituto Superiore di Sanità, Rome, Italy Finally, some interesting suggestions came from [email protected] discussing point 3. Research institutions could require researchers to include a description of their use of CoBRA Anne Cambon-Thomsen directly in the projects data management plan. This would UMR 1027, Inserm, Université Toulouse III – Paul Sabatier, incentivise them to plan for a correct citation of a bioresource Toulouse, France in the early stages of the design of their research project. Furthermore, institutions could be motivated to use CoBRA Laurence Mabile if it is envisaged as an element to ensure research integrity, BBMRI-ERIC, Graz, Austria also considering that in some countries there are university European Science Editing 20 February 2016; 42(1)
HEIRRI: Integrating responsible research and innovation into universities and higher education institutions
16 September 2015, Barcelona Biomedical Research Park, Spain
Research and innovation have to be responsible. That is it in the long term, so that future scientists and engineers the main basis for the European project HEIRRI (Higher (as well as others involved in the process of R+D+i) have Education Institutions and Responsible Research and the knowledge and skills to integrate RRI in their activity. Innovation). The project aims to integrate the concept of Thus, research and innovation could be in line with the Responsible Research and Innovation (RRI) at all stages of expectations, needs and values of society. Integrating RRI education of scientists and engineers, so that research and into university education may increase social benefits of innovation meets societal needs, values and expectations. research and innovation in Europe. The project started in September 2015 and will last three years. It’s one of the first initiatives that Horizon2020 sets How can RRI improve the university of the future? in motion to integrate RRI into universities and higher To incorporate RRI into daily scientific practice has two clear education institutions. The HEIRRI project is coordinated benefits, according to Van den Hoven et al. (2013): firstly, by Universitat Pompeu Fabra (Spain), accompanied it has the potential to make a more efficient investment by University of Bergen (Norway), Aarhus University in research and innovation; and secondly, to encourage (Denmark), Institute for Advanced Studies (Austria), it to focus more on global social challenges. Awareness, University of Split (Croatia), ECSITE, the European knowledge and acquisition of tools to incorporate RRI into network of science centres and museums, with more than professional practice should be taught as early as possible. In 400 institutions from 50 countries, “La Caixa” Foundation this sense, higher education institutions play an essential role. (Spain), the Catalan Association of Public Universities The first HEIRRI conference, “Teaching Responsible (ACUP), which chairs, the Global University Network for Research and Innovation at University” will be hosted in Innovation (GUNi), with 208 universities in 78 countries, Cosmocaixa (Barcelona) on March 18th 2016. It will focus and Innovatec (Spain). All in all, nine European institutions on raising the awareness of RRI and its diverse aspects integrate the HEIRRI Consortium. among the HEI community, as well as aiming to spark the The HEIRRI kick-off meeting connected the members of debate on how institutions can include the concept of RRI in the nine European institutions participating in the project their curricula to train future scientists and engineers. The for the first time and provided an extensive overview of the conference will bring together professionals from European three years of the initiative. During the meeting, hosted in organisations working in Research and Innovation, including the Barcelona Biomedical Research Park (PRBB) on the 16th higher education institutions, research centres, public of September 2015, the institutions discussed the phases of entities, industry, and civil society organisations. It will also the project to promote the application of RRI through six key give visibility to the project and will provide input for future areas: public engagement, ethics, science education, gender actions. A preliminary programme of the event is available at equality, open access and scientific governance. They also http://www.rri-tools.eu/documents/10182/37230/1stHEIRR emphasised the potential of RRI as a transformative, critical IConf_DraftProgramme.pdf and radical concept. All participants presented their involvement in the Marta Cayetano project, which included the elaboration of training Catalan Association of Public Universities (ACUP) materials, pilot experiences, the implementation of the [email protected] ongoing project RRI Tools and a communication and internationalisation strategic plan, among others. Gema Revuelta Pompeu Fabra University (UPF) Getting to know HEIRRI [email protected] Interview with Gema Revuelta, Director of the Studies Centre Núria Saladié on Science, Communication Pompeu Fabra University (UPF) and Society from Universitat [email protected] Pompeu Fabra. She is the HEIRRI project Coordinator.
What do you expect of the HEIRRI project when completed in 3 years’ time? We expect the project will contribute to the integration of Responsible Research and Innovation (RRI) within the education of sciences and engineering. The aim is to promote European Science Editing 21 February 2016; 42(1)
This site I like
Board of Editors in the Life Sciences (www.bels.org)
BELS for Editors Alternatively, editors living outside the US can contact the Board of Editors in the Life Sciences (BELS) came into Registrar for instructions via email. The website also lists existence in 1991 to award credentials to professional dates and places of upcoming exams. editors in the field of life sciences. After years of research A manuscript editor who has been certified for two years and planning, the first certification examination was held and able to document six years of editorial experience, in May 1991 for a total of 23 candidates. Today BELS is can then apply to become a Diplomate Editor in the Life an entirely volunteer-driven organisation with a 25-year Sciences. Applicants are required to submit their portfolio history of tremendous accomplishments. of work and two short essays on topics provided by BELS, Until 1991 manuscript editors had no idea about how which are then evaluated and reviewed independently by to showcase their proficiency in an objective way. The three board-certified editors. The application form can be authors and clients also had absolutely no way to assess the found on the website. So far, only about 30 editors have expertise of an editor. Personal references and ad hoc tests been awarded the diplomate status. were not proving sufficient when it came to gauging the Finally, Honored Editor in the Life Sciences is the standard of an editor’s capabilities. It was being realised that designation conferred on editors who have done distinguished the professional requirement of an editor in the subject was work in the field of editing. This is in recognition of an editor’s increasing with every passing year. A lot of new people were hard work and excellence and a great honour. To date, only stepping in to cater for this increasing need but the quality four people have achieved this status. of the editorial services offered by these new editors was a serious concern. There was a strong need to devise a system Looking for a Board-Certified Editor? that would establish and maintain a standard of quality for BELS website is also useful for authors or journal managers this fast growing profession. who are looking for a qualified editor. The organisation The BELS website provides useful information and maintains “BELS Roster” which lists all the certified editors documents for anyone who is either interested in becoming and their details. A direct link available on the website a board-certified editor or who is looking for a board- allows authors or clients access to more information about certified editor. the editors they might want to employ. In addition, you can search the BELS database to find information about members Interested in becoming a Board-Certified Editor? of the Board of Editors in the Life Sciences who have chosen BELS offers three credentials aimed at recognising to advertise their freelance editorial services. A search can be professional excellence. done by area of expertise, by the type of service required (eg The first level credential, Editor in the Life Sciences, or copy editing or substantive editing), or by the media format ELS, is referred to as “certification”, and is awarded based on (eg journals, books, or web pages). a multiple-choice test used by other professions. This three hour test, about 100 questions, evaluates an editor’s editorial Sangeeta Gaur proficiency, language skills, and copy editing skills. The copy Project Manager at Thieme Medical and Scientific of the study guide is available for download on the website. Publishers India Ltd European Science Editing 22 February 2016; 42(1)
EASE-Forum Digest: September to December 2015
You can join the forum by sending the one-line Pippa thought another explanation was that authors could message “subscribe ease-forum” (without the quotation not be bothered to make the corrections requested upon a marks) to [email protected]. Send in plain text, provisional acceptance. The journal wondered if asking the authors to pay a not HTML. Details at www.ease.org.uk/node/589. reviewing fee or banning them from future submissions if they failed to provide a valid reason for their withdrawal could Is peer review in the social sciences easier than in the resolve the problem. Aleksandra Golebiowska recognized applied sciences? these withdrawal problems. Staff at her journal suspected Pippa Smart asked, “Is peer review easier in applied sciences that when this happened the authors had submitted to a (where the methodology, results and analysis are checked) number of journals simultaneously and one of the other than in social sciences (where the argument and logic are journals had been quicker to accept the manuscript. Sylwia checked)?” In Andrew Davis’ experience of reviewing and Ufnalska pointed to the EASE Form for Authors (http:// writing in the applied sciences, logic and argument were www.ease.org.uk/publications/ease-form). Authors who indeed checked as a matter of routine. It was interesting to signed the form agree to publication of their paper in hear what Will Hughes had to say about this as a variety of the journal and certify that the paper is not considered types of articles are submitted to his journal Construction for publication elsewhere. Aleksandra considered such a Management and Economics. The journal’s board found statement, which she thought to be standard, would not that social science articles were held to a higher standard of deter authors intending to act unethically. methodological arguments and theoretical positioning and argument than applied science articles. The board had to Regional excellence in research, Spanish translation work hard to ensure that all papers met the same standards of open access guide and why an author withdrew and carefully select referees to make sure quantitative Karen Shashok kindly posted the following interesting scientists did not let through material that would not hold its URLs on the forum. own against that of a qualitative researcher. 1. “Defending Regional Excellence in Research Eric Lichtfouse noted that peer reviewers in or Why Beall is Wrong About SciELO”, environmental studies were faced with the problem that published by Phill Jones on Scholarly Kitchen: reproducing exact conditions of research, eg relating to http://scholarlykitchen.sspnet.org/2015/08/10/ agricultural plots in another place, was often impossible defending-regional-excellence-in-research-or-why- because of the number of variables. I asked my son, a political beall-is-wrong-about-scielo/. She quoted his last scientist, what he thought of the question. He considered it sentence, “There is a real danger that the current tone outdated or in need of amendment as many manuscripts in the discussion of predatory publishing could lead in social sciences use a lot of statistics, when reviewers also to a guilt by association of all publishers based in the have to check the results and methodology. In his view, non-English speaking world and that would not only be political science fell into both categories and ideally peer entirely unfair, but damaging to the public good.” review should seek to speak to both dimensions. But it 2. The freely available Spanish version of Peter Suber’s book was not about an easy dichotomy, rather about reviewers on Open Access (MIT Press, 2012): http://ri.uaemex.mx/ being asked to focus on different aspects given their fields’ handle/123456789/21710 (click on ACCESO ABIERTO priorities. Pippa also told me later that she felt her question para repositorio (1).pdf). The book was translated by had perhaps been badly phrased. What she had meant to EASE member Reme Melero and includes an excellent ask is whether other editors found it harder to get reviewers Introduction on OA in Latin America written by Indrajit to objectively critique discursive articles that challenged Banerjee, Dominique Babini and Eduardo Aguado. their (or established) opinions than to get them to evaluate 3. An explanation by Jonathan Eisen (at Tree of Life, articles that reported factual research. brother of Michael at It is NOT Junk) of why he withdrew his name as coauthor of an article published Problems with authors withdrawing articles in Science even though he planned and wrote the paper: A second question raised by Pippa was how journals deal http://phylogenomics.blogspot.com.es/2015/10/ with authors withdrawing papers after review and provisional removing-my-name-from-author-list-of.html acceptance or when the articles were in proof. She also asked 4. URLs discussing the resignation of Lingua’s editorial how journals dealt with authors failing to respond to emails board (See the News section on page 23). after review and acceptance, forcing the journal to drop the article. The question was posed on behalf of an open access Elise Langdon-Neuner (compiler) journal with a low impact factor that does not charge authors. [email protected] The authors who withdrew were suspected to be using the journal as a “first review” service and then taking their Discussion initiators articles to higher impact journals, which wasted reviewers’ Pippa Smart: [email protected] and editors’ time and, if the articles had been typeset, costs. Karen Shashok: kshashok@kshashok European Science Editing 23 February 2016; 42(1)
My life as an editor – Jong K. Ha
Between 1984 and 2013, I taught and researched ruminant to have maintained the reputation of AJAS, which is now nutrition at Seoul National University in Korea. My an open access journal indexed by most global databases, involvement in journal publishing began when I helped including Web of Science, Scopus, PubMed Central, and create an international journal, Asian-Australasian Journal others. We are fortunate to have maintained over 1000 yearly of Animal Sciences (AJAS), the official journal of the Asian- submissions despite several competing international animal Australasian Association of Animal Production Societies science journals that have recently emerged from Asia. (AAAP). Professor In K. Han, the first editor-in-chief, One of my most valuable experiences as editor thus together with several leading animal scientists in AAAP far has been my involvement in creating KCSE. Founded regions, published the first issue ofAJAS in March 1988. in September 2011, KCSE aims to improve the quality Since then, I have served the journal as an editorial board and international status of scientific journals published in member, business manager, assistant editor and in 2001, Korea. Since KCSE’s founding, this vision has been pursued took over as its editor-in-chief. through the concerted efforts of several academic journal AJAS, the first English journal of international stature editors. I have been thrilled to see KCSE grow, thanks in its field from Asia, was established with the mission to to support from major scientific journals in Korea, into serve the animal industry and academia in Australiasia an organisation with over 280 member journals in just a through efficient publication and distribution of scientific little over four years. This rapid progress was possible information on animal science. Initially, the journal was primarily thanks to the dedication of our executive board published quarterly but since 2000, has been a monthly. members and to the strong support of member societies This growth was possible thanks in part to the increase in and individuals, many of whom are also members of EASE. research activity in many Asian countries over the years. Developing KCSE has allowed me to learn about new The development ofAJAS owes much to Professor In K. publishing technologies through personal communication Han, Professor Emeritus of Seoul National University (and with fellow editors and through seminars, conferences, and my adviser during my MS course). His vision and skills were training sessions, where I made the acquaintance of many essential in the creation of AJAS. He nourished AJAS from experts from other organisations. the outset until 2001, when he stepped down as editor-in- A major outcome of collaborative effort by KCSE was the chief to take a still more prestigious position as President of creation of the Council of Asian Science Editors (CASE) the Korean Academy of Science and Technology. in 2014. For several years, KCSE prioritised scholarly When I took over the operation of AJAS, I was still exchange among editors in Asia. In 2013, 18 editors from employed by the university, making it difficult for me to several Asian countries gathered at the Korean Federation find the time required to manage the journal properly. of Science and Technology Societies (KOFST) and agreed One major task I faced at the beginning was reducing the that an organisation dedicated to the advancement of workload of handling and responding to letters and hard- scientific research publication across Asia was needed. copy manuscripts. Commercial online platforms were not Thus, CASE was conceived. After a year of preparation, then widely used in Korea. We were therefore required the council was officially established during the Asian to commission a local IT company to develop a system to Science Editors’ Conference and Workshop 2014 at handle manuscript submission and review, which saved KOFST. As a non-governmental, non-partisan, non-profit me a lot of time until the new, current submission system organisation, CASE aims to improve the quality of science was adopted. Another challenge was keeping pace with journals published in Asia through consulting and sharing publication technology. Although I did not initially have information on editing and publishing, and to become much experience in journal publishing and editing, I was a counterpart to existing international science editors’ aware of the tremendous developments in journal publishing associations. worldwide. I was able to access new technologies through the Many editors of scholarly journals affiliated with small efforts of the Korean Council of Science Editors (KCSE) and societies work as volunteers while holding another primary the European Association of Science Editors (EASE). job. Few such volunteers can devote themselves fully to Publishing a small academic society journal like AJAS editorial work and take editorial positions without adequate is not easy, and editors encounter many challenges in this professional preparation. From my own experience, I global era. Competing with giant publishers and surviving suggest that editors obtain as much new information as requires constant hard work by core members of the possible from various sources, such as EASE, Council of editorial team, especially the editor-in-chief. The journal Science Editors, CASE, and other publishing organisations. is unique in that the society grants the editor-in-chief Additionally, editors must fully grasp the performance absolute power over publication. However, the editor-in- metrics of their journals. Any major change in metrics chief must take full responsibility, looking after all aspects should be detected and proper measures taken at an early of the journal including manuscript handling, editing, and stage. Finally, I recommend that editors make both short- publication; marketing and other business development; and long-term development plans with achievable targets. global outreach; and maintaing the journal’s reputation. Periodic checks will show you where you are and where you Although much remains to be done, I am very proud are heading. European Science Editing 24 February 2016; 42(1)
News notes
News Notes are compiled by John and researchers across the research research”. The initiative, launched on Hilton ([email protected]) lifecycle” and follows on from the 1 October 2015, was prompted by ODIN project (odin-project.eu). concerns about ‘predatory’ journals or Some of these items are taken Ten partner organisations across publishers, and has been endorsed by from the EASE Journal Blog the research infrastructure will the Committee on Publication Ethics (http://esebookshelf.blogspot. work to build relationships between (COPE). com) where full URLs may be contributors, research artefacts (including data), and integrate RIO launch editorial found organisations into the ORCID (orcid. An editorial by the editors of new org) and DataCite (datacite.org) journal Research Ideas and Outcomes Editors of Lingua resign systems. (RIO; riojournal.com) sets out in In October 2015 the editors and detail the aims and processes of entire editorial board of the journal PRO Initiative the journal. The editorial RIO( Lingua (journals.elsevier.com/lingua) A group of academic scientists have 2015;1:e7547) emphasises that science resigned after the journal publisher established the Peer Reviewers’ is a cyclical process, rather than just did not agree to the editors’ proposals Openness (PRO) Initiative a set of findings. The journal enables to make the journal fully open access. (opennessinitiative.org), which researchers to create a “permanent The departing editors intend to start a proposes that peer reviewers should public record for every step within new open access journal, to be called only fully peer review manuscripts their research cycles – including those Glossa, published by Ubiquity Press. that meet the minimum requirements not traditionally published.” Elsevier is recruiting new editors for openness. These requirements and editorial board for Lingua. The focus on the public availability and ELIXIR EXCELERATE dispute was widely reported, and documentation, where possible, of ELIXIR (www.elixir-europe.org) there was support for the departing all data, materials, code or any other is a pan-European infrastructure editorial team from a number of fundamental components of the for biological information, library and academic organisations. research. The PRO Initiative website and the project moved into the includes detailed guidelines for implementation phase in September Making ORCID mandatory authors seeking to ensure openness 2015, with €19 million of EU In November 2015 a number of and transparency. Horizon 2020 funding. The ELIXIR publishers agreed to require authors EXCELERATE programme will aim to supply their ORCID IDs when Annotating all knowledge to integrate Europe’s bioinformatics submitting manuscripts. The Royal More than 50 publishers, technology resources. Society, PLOS, and eLife Sciences organisations and scholarly joined with ORCID to publish an websites have formed a coalition Most-cited retractions open letter (available on orcid.org) with Hypothes.is, with the aim of Many retracted papers continue explaining the decision and offering applying the principles of annotation to be cited after they have been guidance on implementation to technology to scientific publications. retracted. The Retraction Watch other publishers, with EMBO Press, The aim is to implement a “native team has compiled a list of the Science journals, ScienceOpen and and universal collaborative top 10 most cited retracted papers others publishers signing up. Authors capability” rather than relying on (retractionwatch.com; 28 December submitting to the The Royal Society an array of proprietary commenting 2015). Top of the list was a paper by journals or to eLife from January systems. As well as being used for Fukuhara et al published in Science 2016 onwards will be required to note-taking and post-publication in 2005, and retracted in 2007, with supply ORCID IDs, while the other discussion, such systems could at least 750 citations since retraction. signatories will implement the be used for peer review, copy Many of the papers were cited more requirement in the near future. editing, and linking/classification. after retraction than before, and You can read more at hypothes.is/ retraction seems to have no impact on Project THOR annotating-all-knowledge. citation rate, although post-retraction THOR (Technical and Human citations may have acknowledged the Infrastructure for Open Research; Think Check Submit retraction, with the retraction notice project-thor.eu) is a project funded Think Check Submit itself gaining over 100 citations. by the European Commission as part (thinkchecksubmit.org) is a of the Horizon 2020 programme. The checklist developed by a coalition of Matters project aims to “establish seamless organisations “to help researchers A new journal, Matters integration between articles, data, identify trusted journals for their (sciencematters.io), aims to publish European Science Editing 25 February 2016; 42(1) single scientific observations, on the China tackling fraud six months after publication, for basis that these are the “true unit of The problem of fraud in peer review clinical trials that start enrolling science”. The journal uses an online is being tackled by Chinese research patients a year after the requirements guided writing tool for submissions organisations. An investigation by are adopted. Authors would also and will operate a triple blind peer the Chinese Association for Science be required to include a data- review system, whereby neither the and Technology led to major research sharing plan as part of clinical trial peer reviewers nor the handling agencies demanding that offenders registration. The proposals were editors know the identity of the return funding. The story was published simultaneously in member authors. The journal will guarantee reported in Science (20 November journals (20 January 2016), and publication of all “scientifically solid” 2015). Six editing companies formed feedback is sought on the ICMJE observations, while higher rated the Alliance for Scientific Editing in website (icmje.org). observations will be published in a China to try to raise standards. sister journal, Matters Select. Once The One Repo the first ‘seeding’ observation is Hijacking journals There are more than 4000 published, subsequent observations A report in Science (sciencemag.org; institutional repositories listed in the can be published, and a proprietary 19 November 2015) explores a recent Registry of Open Access Repositories algorithm will create links between development: hijacking of journal (ROAR; roar.eprints.org), and observations. Once authors have websites. while there have been regional published sufficient observations, they approaches to aggregating content, are encouraged to submit a “narrative PRISMA-P checklist there is no way of searching across all integration of their observations” to a The PRISMA statement (prisma- repositories. Index Data, with support third journal, Matters Narratives. The statement.org) provides reporting from SPARC Europe (sparceurope. journals are open access and authors guidelines for systematic reviews. org), is building The One Repo, which must pay a fee to submit. Some of Its extension for systematic review aims to capture content from all that fee is passed on to editors and protocols (PRISMA-P) was published repositories. You can try out the demo reviewers. in 2015. To help authors follow (onerepo.net) or read more about the PRISMA-P when submitting project on The One Repo Blog (blog. OpenCon Community Collaborate protocols, a new checklist has been onerepo.net). OpenCon (www.opencon2015. published (Systematic Reviews org) is a conference for early career 2016:5:15) and will be a requirement COPE guide updated researchers interested in open access, for authors submitting protocols to The Committee for Publication open education and open data, the Systematic Reviews journal. Ethics’ Short Guide to Ethical supported by the Right to Research Editing for New Editors has been Coalition (www.righttoresearch. ICMJE recommendations updated revised and updated. The new guide org) and SPARC (sparc.arl.org). In The International Committee of is available from the COPE website November 2015 the team launched Medical Journal Editors (ICMJE) (publicationethics.org). OpenCon Community Collaborate has updated its Recommendations (www.opencon2015.org/collaborate) for the Conduct, Reporting, Editing Hijacking journals as a forum for connecting and and Publication of Scholarly Work A report in Science (sciencemag. supporting individuals and in Medical Journals (previously org; 19 November 2015) explores organisations undertaking initiatives called the Uniform Requirements a recent development: hijacking of in open access. for Manuscripts Submitted to journal websites. This goes beyond Medical Journals). The updated the practice of stealing another EQUATOR Oncology Project recommendations, published in journal’s name or branding, or even The EQUATOR Network has December 2015, are available from using a very similar looking website started compiling a set of resources the ICMJE website (icmje.org), and an address, and involves the actual relating to oncology. The aim is to annotated PDF indicating the changes takeover of another journal’s website help clinicians and researchers to is also available. domain. This can happen when an implement reporting guidelines in organisation neglects to keep up cancer research, and the project is ICMJE data sharing proposal payments on hosting (as happened to supported by Cancer Research UK. The International Committee of CrossRef in 2015 when DOIs stopped You can read the collected resources Medical Journal Editors ICMJE) working briefly after the doi.org and plans for the future on the is seeking feedback on a series of domain registration expired). EQUATOR website (www.equator- proposed data-sharing requirements network.org). for those submitting clinical trial John Hilton reports to its member journals. Editor, Cochrane Editorial Unit, These will require authors to share Cochrane Collaboration, London, UK individual patient data no later than [email protected] European Science Editing 26 February 2016; 42(1)
The editor’s bookshelf
Please write to annamaria.rossi@ ETHICAL ISSUES suggest that the Internet may not be iss.it if you wish to send new items significantly impacting the prevalence or become a member of the EASE Bohannon J. How to hijack a journal. of plagiarism in advanced levels of journal blog (http://ese-bookshelf. Science 2015;350(6263):903-905 higher education. In the past few years fraudsters have doi:10.1007/s10805-015-9233-7 blogspot.com) and see your been snatching entire web addresses, postings published in the journal right out from under the noses of Menezes RG, Kharoshah MA, academic publishers, erecting fake Madadin M, et al. Authorship: few EDITORIAL PROCESS versions of their sites, and hijacking myths and misconceptions. Science their journals, along with their web and Engineering Ethics e-pub 15 Breuning M, Backstrom J, Brannon traffic. The usual method is to build December 2015;1-5 J, et al. Reviewer fatigue? Why a convincing version of a website at This article addresses and dispels scholars decline to review their a similar address and then drive web some of the popular myths and peer’s work. PS: Political Science & traffic to the fake site. Unsuspecting misconceptions surrounding the Politics 2015;48(4):595-600 visitors who log into the hijacked authorship of a scientific publication, The double-blind peer review process journal sites might give away their as this is often misconstrued by is central to publishing in academic passwords or money as they try to pay beginners in academia, especially journals, but it also relies heavily on subscription or article processing fees. those in the developing world. While the voluntarily efforts of anonymous doi:10.1126/science.350.6263.903 ethical issues in publishing related reviewers. To evaluate the degree to authorship have been increasingly to which scholars suffer from the Hvistendahl M. China pursues discussed, not much has been written resulting “reviewer fatigue”, the authors fraudsters in science publishing. about the myths and misconceptions empirically evaluated the reasons Science 2015;350(6264):1015 about who might be an author. scholars offered when declining to China’s main research agency is doi:10.1007/s11948-015-9742-1 review for the American Political cracking down on scientists who use Science Review. Just over one-quarter fake peer reviews to publish papers, LANGUAGE AND WRITING of them declined requests to review. demanding that serious offenders For those who decline, reviewer fatigue return research funding. Since 2012 Gilliver S. Online plain English is only one of several reasons: scholars scores of authors, many of them and readability resources. Medical are willing to review but they often Chinese, have been snagged in a Writing 2015;24(1):20-22 face many demands on their time and peer-review scandal involving papers To encourage individuals and substantial workloads overall. published in international journals. businesses to write in simpler English, doi:10.1017/S1049096515000827 Journals discovered that authors private and government-backed provided email addresses to accounts enterprises have created a number Menéndez J. More on double-blind controlled by the perpetrators, and of freely available online resources. review. APS News 2015;24(10):4 then reviewed their own work. While most relate to general English This letter gives reasons why double- doi:10.1126/science.350.6264.1015 use, some are devoted to writing about blind review should not be optional medical matters. In this summary and also why it would be valuable in Ison DC. The influence of the article, the author takes a brief look at avoiding institutional or country bias. Internet on plagiarism among what is available and how good it is. doctoral dissertations: an empirical doi:10.1179/20474806 Slavov N. Point of view: making the study. Journal of Academic Ethics 4Z.000000000272 most of peer review. eLIFE 2015 2015;13(2):151-166 November 11;4:e12708 This study collected empirical data PUBLISHING Many of the legitimate concerns to investigate the potential influence about papers raised on blogs and the Internet has on significant higher Bandrowski A, Brush M, Grethe JS, other platforms are being ignored education artefacts by comparing et al. The Resource Identification by journals. Journals should publish dissertations written prior to the Initiative: a cultural shift in referee reports, and referees should widespread use of the Internet publishing. Journal of Comparative be encouraged to sign their reports. with those written in a period of Neurology 2016;524(1):8-22 Journals should also consider ubiquitous Internet use. It studied The Resource Identification Initiative non-anonymous post-publication Doctor of Philosophy (PhD) was launched as a pilot project to comments submitted to certain dissertations written in English and improve the reporting standards for platforms within a certain time after published by accredited universities in research resources in the Methods the paper has been published. the USA and Canada. A sample of 384 sections of articles and thereby doi:10.7554/eLIFE.12708 dissertations were analysed. Results improve identifiability and scientific European Science Editing 27 February 2016; 42(1) reproducibility. The pilot engaged scientific discipline, or status as The purpose of this narrative over 25 biomedical journal editors a review journal. Results showed literature review is to synthesize from most major publishers, as well that the majority of journals had evidence that describes processes as scientists and funding officials. a retraction policy, and almost all and conceptual models for assessing Authors were asked to include of them would retract an article policy and practice impacts of public Research Resource Identifiers without the authors’ permission. health research. The literature is (RRIDs) in their articles prior to COPE’s guidelines appear to have characterised by an over reliance publication for three resource types: had a significant influence on journal on bibliometric methods to assess antibodies, model organisms, and retraction policies. research impact. Future impact tools (ie software and databases). doi:10.3163/1536-5050.103.3.006 assessment processes could be doi:10.1002/cne.23913 strengthened by routinely engaging Krumholz HM. The end of journals. the end-users of research in Jorm AF. Can a medical researcher Circulation e-pub 2015 November 10 interviews and assessment processes. have too many publications? According to the author, there are doi:10.1186/s12961-015-0003-1 The Medical Journal of Australia at least 9 deficiencies in the current 2015;203(5):230-1 publication model that fuel the sense SCIENCE COMMUNICATION Most prolific researchers may not be that journals as we have known adhering to authorship guidelines: them are approaching their final act. Helbing D, Pournaras E. Society: the author argues that very high Among them: the publication process build digital democracy. Nature 2015 publication rates should be seen as is too long; the expense of publishing November 5;527(7576):33-34. indicating poor authorship practices is growing rapidly; the configuration Open sharing of data that are and should be discounted when of articles prohibits a comprehensive collected with smart devices would evaluating track record. and in-depth approach to a scientific empower citizens and create jobs, doi:10.5694/mja15.00194 question; peer review and the journal say the authors of this article. A decision-making process occur research team has started to create a Jubb M, Goldstein S, Amin M, et al. without much external scrutiny and distributed, privacy-preserving ‘digital Monitoring the transition to open transparency. nervous system’ called Nervousnet. It access. A report for the Universities doi:10.1161/ uses the sensor networks that make UK Open Access Co-ordination CIRCOUTCOMES.115.002415 up the Internet of Things, including Group. Research Information those in smartphones, to measure Network. August 2015;105 p. RESEARCH EVALUATION the world around us and to build a Reliable indicators should be gathered collective ‘data commons’. on key features of the transition to Bourne PE, Lorsch JR. Green ED. doi:10.1038/527033a open access (OA) in the UK. The Sustaining the big-data ecosystem. findings presented in the report Nature 2015 November 5;527 Whelan J. Medical journalism: from this study are a first attempt at Biomedical big data offer tremendous another way to write about science. generating such indicators covering potential for making discoveries, but Medical Writing 2015; 24(4):219-221 five sets of issues: OA options the cost of sustaining these digital True journalism differs from public available to authors; accessibility; assets and the resources needed to relations and uncritically reproducing usage; financial sustainability make them useful have received press releases. It involves doing for universities; and financial relatively little attention. Funders background research into the sustainability for learned societies. should encourage the development context surrounding the findings of new metrics to ascertain the being reported, seeking comments Resnik DB, Wager E, Kissling GE. usage and value of data and when from independent experts, and Retraction policies of top scientific we have a better understanding of highlighting the negative as well as journals ranked by impact factor. data usage, we can develop business positive aspects. In this article, the Journal of the Medical Library models for storing, organising and author pulls together information Association 2015;103(3):136-139 accessing them. Tools and rewards for medical writers interested in The purpose of this study was to that incentivise researchers to submit journalism or science writing. provide updated information on the their data to data resources in ways doi:10.1179/20474806 retraction policies of major science that maximise both quality and ease 15Z.000000000327 journals. The specific aims were to: of access, are also needed. determine the percentage of the doi:10.1038/527S16a Thanks to John Glen top 200 science journals ranked by impact factor that have a retraction Milat AJ, Bauman AE, Redman S. A Anna Maria Rossi policy; analyse the content of journal narrative review of research impact Publishing Unit retraction policies; and ascertain assessment models and methods. Istituto Superiore di Sanità, Rome whether having a retraction policy Health Research Policy and Systems [email protected] is associated with impact factor, 2015;13:18 European Science Editing 28 February 2016; 42(1)
Correspondence Promoting global health: the role of medical journal editors
In May 2015, WAME issued a policy statement on the Lancet journals commission a comment to accompany social responsibilities of medical journal editors to each research article and where relevant, we do invite publish “whenever possible, research that furthers health authors from low and middle-income countries. The Lancet worldwide”.1 It called for editors in high-income countries Psychiatry also publishes stand-alone comments which to invite researchers from low and middle-income countries have proven to be a good outlet for topics on global mental to participate on editorial boards and as peer reviewers and health. Some of these reports need substantive editing but to submit editorials and commentaries on local context. we feel that this investment is worthwhile. We support this call and would like to add some points A new initiative relevant to this is being trialled by from our experiences at The Lancet group journals. The The Lancet Psychiatry in the form of an Editorial Board Lancet has long been an advocate of global health and Development Programme.3 Six psychiatrists from in 2013 launched a journal dedicated to this topic. The developing countries will receive training in peer review, preference of The Lancet Global Health is that papers should manuscript assessment and other aspects of the editorial be authored or at least co-authored by researchers in the process. They will share their acquired skills and knowledge country where the research is conducted. with peers and other researchers and students in their The Lancet policy is that research pertaining to a given countries, thereby building capacity and hopefully raising country should be peer reviewed by a researcher based the profile and standards of global mental health. We hope in that country and familiar with the population, setting, that if this initiative is successful, it might be adopted by databases or questionnaires used, as appropriate. Finding other journals for the benefit of researchers, healthcare such reviewers can be difficult: in addition to the usual workers and patients around the world. database searches, it often requires asking multiple board members or respected researchers in global health for Joan Marsh suggestions. In a recent post on the Scholarly Kitchen,2 Deputy Editor, The Lancet Psychiatry Angela Cochran suggested searching databases of dissertations and conference programmes as a way of Zoe Mullan expanding a journal’s pool of reviewers, not necessarily just Editor, The Lancet Global Health from low and middle income countries. It is important to give feedback to inexperienced reviewers which will both References encourage them to review for the journal in the future and 1 WAME website (www.wame.org/about/policy-statements). improve their performance. One way, if a journal doesn’t 2 Cochran A http://scholarlykitchen.sspnet.org/2015/11/04/is-reviewer- do this as standard practice, is to send all the reviewers’ fatigue-a-real-thing/ Accessed 4 November 2015 comments to junior reviewers after the final decision has 3 Marsh J, Boyce N The Lancet Psychiatry Editorial Board Development been made on the paper. Programme. The Lancet Psychiatry
Letter to the editor
I much enjoyed the article in the last issue by Omar Sabaj there are more than three volunteers, and twisting arms (or et al1 on the periods of time taken between submitting an re-circulating the paper titles) if there are no immediate article, finding referees, receiving their reports, making respondents. This process, which we call “peer choice” editorial decisions, and eventual publication in three appears to work well. A recent account has been provided Chilean journals. What was of particular interest was the by Hartley, Cowan and Rushby2. amount of time taken here for the editors of one of the journals to find appropriate referees. References As the authors themselves suggest, one way of reducing 1 Sabaj O, Valderrama J O, Gonzalez-Vergara C, Pinar-Stranger A. this is to submit the titles and the abstracts of submissions to Relationship between the duration of peer review, publication a pre-established database of potential referees. TheBritish decision, and agreement among reviewers in three Chilean journals. Journal of Educational Technology uses such a system. Here, European Science Editing. 2015;41(4):87-90. groups of anonymised abstracts are forwarded regularly to 2 Hartley J, Cowan J, Rushby, N. Peer choice - does reviewer self-selection a panel of volunteer referees, and members of the panel are work? Learned Publishing. 2015. DOI: 10.1111/bjet.12360 asked to choose (say up to three) papers that they would be willing to referee. The editor then sends these papers out appropriately – choosing which respondents to ask if