Ethical Considerations in Digital Scientific Communication and Publishing

Poor editing, sloppy bookkeeping, fudgy analysis, falsification? Misunderstanding, variable technical practice, outride fraud? Sightings in the published scientific literature of apparent data manipulation raise eyebrows and many such questions. A community blog discussion on a controversy around inconsistently-reported elemental analyses last summer suggests that several layers of action by multiple parties might be involved in such issues, some intentional, some perhaps not, and very few publically disclosed (http://blog.chembark.com/2013/08/09/the-om-paper-vs-drinkels-phd-thesis/#comment-52402). There is very little understanding through the copies of record of this research to indicate to the reading public what the process might have been regarding the review that the articles underwent and how both the editors and authors approached any confusion over data representation and adjustment.

One way or another, it appears that the community is becoming more aware of potential concerns with the responsible reporting of data and other ethical issues with the scientific publication process. How much of this awareness might be arising from more transparent community discussion via blog and twitter-spheres, less transparency of handling data from measurement through analysis and eventually as publication quality figures, and/or greater pressures on the research and publication systems globally is complicated to sort out. Engaging the chemistry research community in conversations is certainly an important part of the process. To members of the American Chemical Society (ACS) Ethics Committee (ETHX) and Division of Chemical Information (CINF) it seemed timely to organize a symposium on the ethical challenges that arise in the course of preparing for presentation. 

ETHX and CINF teamed up with the ACS Divisions of Chemistry and the Law (CHAL), Professional Affairs (PROF), and Publications to bring together a diverse group of editorial professionals at the recent ACS National Meeting in Dallas, TX to discuss a range of challenges, old and new, and strategies to re-enforce responsible conduct in the publication process. The international speaker set included senior staff and science editors from several core chemistry publishers including the ACS, the Royal Society of Chemistry (RSC) based in the UK, and Wiley-VCH based in Germany and publishing the journals of the Gesellschaft Deutscher Chemiker (the German Chemical Society, GDCh). Also represented were several other publishing and supporting organizations concerned with ethical issues, some with particular focus on data representation, including the Cambridge Crystallographic Data Centre (CCDC), and the American Physiological Society (APS), the Committee on Publication Ethics (COPE), and CrossRef. Abstracts and some slides are available on the CINF site at: /node/557. Following are my own reflections from my notes; any omissions or misrepresentations are my error.

Considering the above case, curation of data is an important concern in the publication process, including best practices for representation, processes for checking and validation, and communicating with authors on subsequent corrective action. Some data types such as crystallographic data are regularly deposited with articles as part of the review process and subsequently curated in databases. The CCDC receives and curates standardized files of coordinates from a variety of publishers, validates the structures in-house and sends back identified errors to authors for correction. The accumulated database is searchable with software that enables further analysis, visualization, curation, and system development that are sustained through distinct subscription streams. Another example is the representation of visual types of data in figures, such as protein separation gels used in physiological research. The APS has found that figure manipulation accounts for a large majority of problems that raise ethical flags in the course of article submission, including general presentation issues: splicing images, adjusting contrast or dropping backgrounds, and poor resolution. An in-house discipline-trained scientist has been assigned to analyze the types of problems and detection scenarios, develop a communication process with authors, and publication guidelines including a policy of transparency concerning image rearrangement as well as a list of “don’ts” to curb misplaced efforts up front. These organizations are focused on engaging authors and the research community through community curation scenarios and graduate education outreach.

Data manipulation and similar ethical concerns have accompanied the exchange of scientific information for centuries. An article on fraud in science in a Wiley journal quotes 19th century British scientist Charles Babbage of the Royal Statistical Society classifying misconduct: “hoaxing, forging, trimming, and cooking,” (DOI: 10.1111/j.1740-9713.2007.00215.x pg 24). The most common types of ethical concerns brought forward by the journal editors presenting in this session included questions of authorship, prior publication and self-plagiarism.  Apparently, not all authors are always aware that their names have been included on submitted manuscripts, and the ACS Publications Division now issues letters to all authors listed on manuscripts to verify submission and that all are aware. Prior publication can be particularly confusing as it is acceptable in some situations such as theses, earlier communication articles and in some fields, preprint repositories. Specific policies concerning prior publication often lie with the editorial offices of specific journals to contend with the shifting research priorities appropriate to the subject coverage and it is important for both the editors and authors to be in communication and up front about handling prior publication of research results. Other concerns include duplicate submission, almost impossible to detect before publication; self-plagiarism, particularly problematic in review articles intended to build across the art of a research area; and “dry-labbing,” reporting of procedures not actually performed that becomes evident with reporting of unrealistic reaction conditions or inappropriate results. 

Many organizations are interested in the questions of ethics arising in the course of scholarly publication.  A memo from the Office of Science and Technology Policy (OSTP) in 2010 directed US funding agencies to “establish principles for conveying scientific and technological information to the public” (http://www.whitehouse.gov/administration/eop/ostp/library/scientificintegrity). In Europe, promotion of responsible conduct is discussed by the European Association for Chemical and Molecular Sciences (EuCheMS, http://www.euchems.eu/divisions/ethics-in-chemistry.html). The ACS Committee on Ethics serves as an educational resource and clearinghouse on ethical considerations and coordinates communication and programming activities (see the committee overview in this issue of CIB). Publishers take their editorial role in these concerns very seriously. They associate through various professional organizations to streamline and network best practices and collectively support the process of addressing challenges. Various tools are available to support the work of editors, reviewers and the overall publication process, including the CrossCheck screening tool developed by CrossRef to flag overlapping text against a growing corpus of full text scholarly literature for further review and implemented by publishers as part of the editorial workflow (http://www.crossref.org/crosscheck.html). It is not a plagiarism detector or a comprehensive sweep of all published literature, missing most supplemental information and interpretation of image data, including equations. Collaboratively-sourced tools for developing clear and consistent guidelines and process flowcharts are also available from the publisher member-based COPE (http://publicationethics.org/about). Community-based discussions of specific example cases are collated and anonymized into a knowledge bank of lessons learned available for the broader public. A recent analysis of the cases suggests a shift in focus towards more discussion of conflict of interest, correction of the literature, data, misconduct or questionable behavior, and peer review.

It is increasingly important to support clear and consistent processes for all parties involved in review, handling identified concerns, reaching resolution and developing understanding of the broader issues and responsibilities. RSC has assigned a dedicated staff position to follow the overview of cases and address consistency, with a goal of striving for agreement among all parties, including in cases of retraction, where authors sign retraction notices before they are posted. The ACS also supports authors through a suite of educational materials related to issues of ethics, including episodes of the Publishing Your Research 101 video series on ethical considerations, copyright and the review process (http://pubs.acs.org/page/publish-research/index.html).  

Representation issues are all around us in the irrepressibly malleable digital environment. How much manipulation is inadvertent? How much of this activity is concerned with attempting to look professional, how much is involved in striving for the right story, how much wrapped up in building reputation, how much is simply confusion over proper procedure? Creativity in research that builds usefully on the scientific corpus is inherently a juggling act between consistency and aberration.  Researchers are entrusted with due diligence in their experimental design, analysis and documentation. Feeding back into the corpus involves additional juggling of representation and expression of data and rationale. The importance of the moment of publication in defining a line of inquiry and the critical role of trust in upholding the integrity of the scientific record speaks to the long-standing oversight of the editorial process in scientific publication. As the acceleration of research output outpaces traditional publication outlets and the digital environment opens new opportunities for data sharing and communication, both curated and wild, the supportive processes of ensuring responsible conduct in research and publication are put to the test. The questions of what is involved in the editorial process in the era of digital scientific publishing and what are the ethical considerations that arise continue to evolve with the practices of both science and publishing.

The author would like to thank the presenters, the co-sponsors of the symposium, with particular acknowledgement of the organizing efforts of Heather Tierney, for bringing this endeavor to successful fruition. Supporting sources are referred in-line. 

Leah McEwen, Symposium Co-Organizer

Image

Oral presentations captured at the ACS Spring National Meeting in Dallas will be available to ACS Members after April 28, 2014 at www.presentations.acs.org

The following four presentations were recorded at the “Ethical Considerations in Digital Scientific Communication and Publishing” symposium:

 

CINF52 Tools for identifying potential misconduct: The CrossCheck service from CrossRef. Rachael Lammey

CINF53 Mapping the terrain of publication ethics. Charon Pierson

CINF55 Ethics in scientific publication: Observations of an editor and recommended best practices for authors. Kirk Schanze

CINF57 Role of the journal editor in maintaining ethical standards in the changing publishing environment. Jamie Humphrey.