Quality in science publishing

Abstract

The purpose of this Theme Section (TS) is to discuss various aspects of quality, and quality assurance, in science publishing.While we will each have our own views about exactly what this means, the editorial policy statement of the Council of Science Editors uses the following adjectives in reference to the quality of a scientific journal's content: accurate, valid, reliable, credible, authoritative, relevant to the journal's scope and mission, readable, and comprehensible.A standard dictionary definition of quality control is: '…a method of establishing and maintaining a high level of quality in a product or process through careful organizing, constant checking and painstaking corrections.'In this context, most of us would probably agree that the publisher, the editorial board, the reviewers and (not least) the authors are all involved in creating, defining, measuring and maintaining quality.Thus, we asked a cross-section of people with experience of one or more aspects of science publishing to present their views on the theme.The main points of consensus that emerge from the essays follow.Most contributors identified the importance of good scientific writing.When a manuscript has not been well written by the author(s), thorough and professionally astute editing is a necessity.This requires great care to preserve the exactness of the text despite possibly significant revisions, restructuring, and/or deletion of what is unsuitable, superfluous and/or badly written.This task is shared by the reviewers of the manuscript's scientific content, and by copy editors.The former focus on what is, or is not, essential to the scientific story being told, and the latter on the clarity of the prose and the correctness of the grammar and syntax.This process takes time, but it is worth waiting for if the result is something that will be more easily read.This is the essence of production quality.The fundamental importance of good scientific judgement in deciding a manuscript's fate was mentioned by all contributors.This crucial decision must be taken by a hands-on Editor who is an acknowledged authority in the subject area of any particular article, and who is a respected scientist capable and courageous enough to arbitrate openly and fairly when confronted with divergent views among authors and reviewers.The sheer number of manuscripts received by our discipline's most successful scholarly journals (over 700 per year at MEPS alone), and the wondrous breadth of their subject matter, makes this an extremely challenging task.The views expressed here suggest that the 'Subject Editors plus Editor-in-Chief' model is probably the best way of assuring scientific quality, as well as being more transparent and accountable than alternative models of editorial decision making.Authors themselves are best positioned to decide upon the most appropriate Subject Editor to handle the evaluation process.There should be a sufficient number of Subject Editors to cover the journal's scope: no more, lest they become idle and their title meaningless, and no less, to ensure that the workload is reasonably distributed and that they are only responsible for papers that fall within their own field of expertise.Subject Editors must be actively engaged in the review process, able to dissect the arguments of authors and reviewers alike and to provide guidance and leadership, and thus arrive at decisions that are transparent and well founded.Their decisions should still be open to appeal to an Editor-in-Chief, a person of recognised eminence and broad experience, and that decision would be final.The Editor-in-Chief, presumably in consultation with the rest of the Editorial Board, and the Publisher, would set a common Editorial policy and ensure that it is consistently applied.This system is not yet widely applied in marine science publishing, but it is the most common decision-making framework

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.