Scientific misconduct has been a matter of debate ever since scientific communication began.1a–c However, chemists have largely considered themselves above the matter, trained to have the highest respect for scientific honesty. Professor Sir Derek Barton commented on scientific misconduct, dismissing it by saying “Chemists don′t do that sort of thing. We are better.” But eventually “The Black Cloud”2 hit us too, and obscured our world. In the late 1960s,3 the chemistry community became aware of the growing issue of scientific misconduct,4 and although it had been relatively immune to the problem, it was not without some major failures. To safeguard scientific integrity, the current situation should be viewed with disenchanted eyes.
While there are several types scientific misconduct, such as fabrication of results, data augmentation, and plagiarism in all its forms,5 these are well understood by the scientific community and are covered by numerous ethical guidelines adhered to by researchers and publishers alike. For example, ChemMedChem and the other ChemPubSoc Europe journals published by Wiley-VCH apply the guidelines set forth by the European Association for Chemical and Molecular Sciences (EuCheMS).6 However, over-fragmentation and premature publishing are often overlooked because they represent a grey area in which the definitions are subjective and open to interpretation. These practices flood the literature with partial results, lowering the overall quality of the scientific literature, and most importantly, making it increasingly difficult for researchers to keep up with the papers being published in their field (let alone any adjacent areas of interest).
EuCheMS Ethical Guidelines for Publication in Journals and Reviews does, in fact, deal with the issue of fragmentation.6 In section 3.0, the document lays out the author′s responsibility to ensure the ethical publication of their work. Subsection 3.4 clearly states that undue fragmentation is to be avoided and even goes so far as to outline what might be considered a redundant publication: “[Authors have a responsibility] a) To avoid undue fragmentation of their work into multiple manuscripts. Editors have the right to reject submitted articles on the grounds of undue fragmentation. In particular, a piece of work should not be split into a number of manuscripts for publication as Communications. b) Not to engage in redundant publication, which occurs when two or more papers, without full cross reference, share the same hypothesis, data, discussion points, or conclusions. Previous publication of an abstract or preprint of the proceedings of meetings does not preclude subsequent submission for publication, but full disclosure should be made at the time of submission. Re-publication of a paper in another language may be acceptable, provided that there is full and prominent disclosure of its original source at the time of submission”.6
In contrast to over-fragmentation of research results, premature publication is not dealt with in the current version of the document, but this may well be because it is a relatively new phenomenon.
…over-fragmentation and premature publishing are often overlooked because they represent a grey area in which the definitions are subjective and open to interpretation…
So what has caused this new trend of prolific publication of marginal scientific data? The likely answer is a mixture of necessity, lack of instruction, and possibly even scientific culture misalignment. The present system of funding and career progression in many countries is based not only on quality but on quantity as well,7a,b where research grants and promotions are given based on the number of articles published, sometimes with little regard for, or even understanding of, the differences between scientific disciplines or other factors. Cultural factors play a role too; we are witnessing a tremendous increase in scientific publications from authors from a wide range of cultural backgrounds and with diverse ethical standards who may unknowingly and unwillingly breach the rules.8 Poor appreciation of established research ethics is not limited to the emerging research regions; we should all ask ourselves: are we sufficiently preparing our students for a career in science,9 or do the pressures to publish start even before obtaining a PhD? This lack of understanding could well be at the root of the problem.3 These cultural misalignments might prevent some people from making clear distinctions between proper and improper, fair and unfair, and between having ambitions and being ambitious.9
The continued issue of unethical behavior and misconduct has sometimes called into question the validity of the peer-review system. However, although the current system is not without its flaws, a viable alternative that offers significant advantages has yet to be proposed.10a–i From a positive standpoint, the peer-review process often identifies ethical issues prior to publication. Cases of plagiarism, fragmentation, duplicate submission, and many others are caught by editors and reviewers alike, going some way to ensure the integrity of the published record. However, with the rise in the number of publications and the additional time constraints on both editors and reviewers, it is clear that over-fragmentation and premature publishing wastes resources and clogs the system; ultimately such manuscripts mask important research behind their noise.
So how do we address this issue? Authors, reviewers and editors all play clear and significant roles in ensuring the integrity of the scientific literature. But so too do governments and funding agencies, as their policies have an impact on the actions of the scientific community, and as a consequence, the publishing habits of scientists.
Authors, reviewers and editors read a given manuscript through different eyes: authors see their hard work laid out, the fruits of their labor, and the potential means to attract more funding or career advancement. Reviewers see the manuscript with a critical eye, noting the flaws and holes in the rationale; reviewers are obliged to provide explanations in support of their comments, to provide a balanced and unbiased report. These requirements are actually laid out in ethical guidelines (e.g., Section 4.0 of the EuCheMS Ethical Guidelines for Publication in Journals and Reviews).6 Finally, editors look at a manuscript as it pertains to the journal. In considering scope and impact, editors cast a critical eye over the manuscript upon submission, and should they feel it is suitable for the journal, they send it to reviewers for evaluation. A decision based on two reports in accordance is a typical scenario, but of course an editor should and does have discretion over the decision. Interpretation and a “reading between the lines” ensure that due consideration is given to all manuscripts. For the system to stay true, reliance on the quality of the reviewers′ comments is inevitable.
Authors, reviewers and editors all play clear and significant roles in ensuring the integrity of the scientific literature. But so too do governments and funding agencies…
Attempts have been made to quantify and rationalize the system, to understand the behavior of peer reviewers, and to predict the ultimate outcome in terms of the quality of published articles.11 Of course, the evaluation of reviewers, at least qualitatively, is not a new concept, and this is commonly done—certainly at ChemMedChem—by the editor, evaluating the quality of the reports and keeping a record of reviewers′ behavior and recommendation patterns. By evaluating the referees, an editor is able to build a reliable pool of reviewers; with experience, both on the side of the reviewer and editor, comes useful comments and fair decisions.
The solution to curbing the trend toward premature publication lies with all those involved: with authors in avoiding the temptation, with reviewers in remaining alert to it, and with editors in establishing a firm policy against it. However, the most important action is the continued education of others at the global level. Policymakers and research funders should be well informed on the distinctions between research conducted in various disciplines. The rate and nature by which new results are acquired in medicinal chemistry, for example, differ greatly from those of astronomy, sociology, and botany. This is crystal clear to most active researchers regardless of background, but is an important fact that risks being overlooked when it comes to gauging the performance of a given scientist by using simple bottom-line figures, such as the number of publications per year, or the impact factors of journals in which he or she has published.
Important decisions such as hiring, tenure, promotions, and research funding should always be based on a close qualitative look at a person′s track record. This requires greater effort, of course, but the temptation to make such assessments primarily on the basis of easily quantified metrics should be avoided. Doing so will not solve the problem of over-fragmentation and premature publication overnight, but it will encourage and foster a healthier climate in which greater value is placed on the quality of results reported than on the number of papers in which such results appear.
Professor Giorgio Tarzia Co-chairman of the Editorial Board ChemMedChem