Notices and Policies for Retractions, Expressions of Concern, Errata and Corrigenda: Their Importance, Content, and Context

Abstract

A retraction notice is an essential scientific historical document because it should outline the reason(s) why a scientific manuscript was retracted, culpability (if any) and any other factors that have given reason for the authors, editors, or publisher, to remove a piece of the literature from science’s history books. Unlike an expression of concern (EoC), erratum or corrigendum, a retraction will usually result in a rudimentary vestige of the work. Thus, any retraction notice that does not fully indicate a set of elements related to the reason and background for the retraction serves as a poor historical document. Moreover, poorly or incompletely worded retraction notices in fact do not serve their intended purpose, i.e., to hold all parties accountable, and to inform the scientific and wider public of the problem and reason for the paper’s demise. This paper takes a look at the definitions and the policies of clauses for retractions, EoCs, errata and corrigenda in place by 15 leading science, technology and medicine (STM) publishers and four publishing-related bodies that we believe have the greatest influence on the current fields of science, technology and medicine. The primary purpose was to assess whether there is a consistency among these entities and publishers. Using an arbitrary 5-scale classification system, and evaluating the different categories of policies separately, we discovered that in almost all cases (88.9 %), the wording used to define these four categories of polices differs from that of the Committee on Publication Ethics (COPE), which is generally considered to be the guiding set of definitions in science publishing. In addition, as much as 61 % deviation in policies (wording and meaning), relative to COPE guidelines, was discovered. When considering the average pooled deviation across all categories of policies, we discovered that there was either no deviation or a small deviation, only in the wording, in the definition of policies when compared to the COPE guidelines in 1 out of 3 ethical bodies, and in 40 % (6 out of 15) STM publishers. Moderate deviation from the COPE guidelines was detected in 26.7 % of STM publishers and one ethical body but a large deviation in one ethical body and 20 % of STM publishers was observed. Two STM publishers (13.3 %) did not report any information about these policies. Even though in practice, editors and publishers may deviate from these written definitions when dealing with case-by-case issues, we believe that it is essential, to serve as a consistent guide for authors and editors, that the wording be standardized across these entities. COPE and these entities also have the responsibility of making it clear that these definitions are merely suggestions and that their application may be subjected to subjective interpretation and application.