European Science Editing 42 May 2019; 45(2) fair is to be expeditious, which is in tension with the need to get a suitable set of reviews. It is also very important for the editors to recognise that we are the stewards of the journal for our time in that role. We wanted to make sure that what was published under our watch was solid, sound research, well-written, and relevant to the stated scope of the journal. While we hope our perspectives are useful for others, what ultimately worked reasonably well for us—mistakes and all—will not necessarily work for others. 1. Should journals invite reviewer recommendations, either built into manuscript handling systems or within the reviewer reports? Te most important part of what a reviewer does is provide advice to the editor about the scientifc value and quality of the work. Obviously, the action editor ideally has a decent sense of the topic, but it is typical for additional expertise to be necessary. EM:IP used a fairly simple holistic rating: Accept, Minor Revision, Major Revision, and Reject. Our feeling, as specialists in measurement, is that this four- point scale is enough to provide such guidance and to help orient the editor without being burdensome to the reviewer. While raters may be uncertain, for instance between Minor Revision and Major Revision or Major Revision and Reject, the confdential comments to the editor can be (and in our experience, ofen were) used to indicate this uncertainty. We endeavoured not to use the holistic rating in a mechanical fashion that reduced to “vote counting” among the reviewers. We discuss this further below. 2. Should such recommendations be mandatory or optional? For EM:IP, recommendation was mandatory. Our belief is that a measurement system—which a reviewer recommendation clearly is—operates best when applied in a uniform manner that is understood by its users, which, in our view, implies that the recommendation should be required. However, we also recognise that the decision made by the editor is, ultimately, a holistic one based on all the evidence presented in the reviewer’s comments as well as other evidence. 3. Do recommendations form part of reviewer best practices? As we said previously, we question the notion there is a clear set of “best practices” just as there is no clear set of “worst practices.” Tere are multiple ways to be bad: We are happy to have the opportunity to react to the editorial by Jonathan P Tennant, et al (2019), “Boon, bias or bane? Te potential infuence of reviewer recommendations on editorial decision-making”. 1 Te editorial poses seven questions about the role of reviewers and editors in scientifc publishing. We were Editor in Chief (Everson) and one of the two Associate Editors (Verkuilen) of the journal Educational Measurement: Issues and Practice (EM:IP) from January 1, 2016, to December 31, 2018, which is the typical three-year term. Te journal is sponsored by the National Council for Measurement in Education (NCME). Tis area is explicitly multi-disciplinary, involving statisticians, psychologists, educational psychologists, educational policy analysts, and other scholars interested in assessment, testing, and measurement. It is US-based but has substantial international authorship and readership from both academia and industry, refecting the increasingly international nature of assessment. We ofer our refection on our time in these roles and how it bears on the role of the reviewer and the potentially blurry border between editor and reviewer. Our core view is that the reviewer’s job is to provide advice to the action editor and, ultimately, to the EIC. Te way we operated, the EIC made all fnal decisions, with action editors taking the paper through the review process and making recommendations. Other journals may, of course, operate diferently. We do not suppose there is one way for a journal to function well, just as there are many ways for one to function poorly. It is the editor who is responsible to the chartering organisation and editorial board, not the reviewers. It is not at all unusual for reviewers to be from other disciplines and they may not even be members of the chartering organization, depending on the nature of the article. Tis was something NCME took very seriously. For us, having clear and consistent lines of authority minimized friction and helped maintain, insofar as is possible in a challenging environment like editing, uniformity and sense. Tis was both for our own sanity and our prior preferred working arrangement but, more importantly, to ensure that the community who had entrusted us with the role as editors were able to understand our choices. It is ofen an editor’s role to deliver possibly career-afecting decisions to authors, so making sure that the process is both as fair and seen as being fair as possible is essential. Tis is not easy, however: A big part of being Viewpoints Seven questions about the border between the role of reviewers and editors and how we tentatively answered them Jay Verkuilen City University of New York, USA Howard T Everson Stanford Research Institute, City University of New York, USA DOI:10.20316/ESE.2019.45.19003