评审的十条简单法则

David

Ten Simple Rules for Reviewers0


Bourne PE, Korngreen A  doi:10.1371/journal.pcbi.0020110

The rules for getting articles published included advice on becoming a reviewer early in your career. If you followed that advice, by working through your mentors who will ask you to review, you will then hopefully find these Ten Rules for Reviewers helpful. There is no magic formula for what constitutes a good or a bad paper—the majority of papers fall in between—so what do you look for as a reviewer? We would suggest, above all else, you are looking for what the journal you are reviewing for prides itself on. Scientific novelty—there is just too much “me-too” in scientific papers—is often the prerequisite, but not always. There is certainly a place for papers that, for example, support existing hypotheses, or provide a new or modified interpretation of an existing finding. After journal scope, it comes down to a well-presented argument and everything else described in “Ten Simple Rules for Getting Published”. Once you know what to look for in a paper, the following simple reviewer guidelines we hope will be useful.

Rule 1: Do Not Accept a Review Assignment unless You Can Accomplish the Task in the Requested Timeframe—Learn to Say No

Late reviews are not fair to the authors, nor are they fair to journal staff. Think about this next time you have a paper under review and the reviewers are unresponsive. You do not like delays when it is your paper, neither do the authors of the paper you are reviewing. Moreover, a significant part of the cost of publishing is associated with chasing reviewers for overdue reviews. No one benefits from this process.

Rule 2: Avoid Conflict of Interest

Reviews come in various forms—anonymous, open, and double-blind, where reviewers are not revealed to the authors and authors are not revealed to reviewers. Whatever the process, act accordingly and with the highest moral principles. The cloak of anonymity is not intended to cover scientific misconduct. Do not take on the review if there is the slightest possibility of conflict of interest. Conflicts arise when, for example, the paper is poor and will likely be rejected, yet there might be good ideas that you could apply in your own research, or, someone is working dangerously close to your own next paper. Most review requests first provide the abstract and then the paper only after you accept the review assignment. In clear cases of conflict, do not request the paper. With conflict, there is often a gray area; if you are in any doubt whatsoever, consult with the Editors who have asked you to review.

Rule 3: Write Reviews You Would Be Satisfied with as an Author

Terse, ill-informed reviews reflect badly on you. Support your criticisms or praise with concrete reasons that are well laid out and logical. While you may not be known to the authors, the Editor knows who you are, and your reviews are maintained and possibly analyzed by the publisher's manuscript tracking system. Your profile as a reviewer is known by the journal—that profile of review quality as assessed by the Editor and of timeliness of review should be something you are proud of. Many journals, including this one, provide you with the reviews of your fellow reviewers after a paper is accepted or rejected. Read those reviews carefully and learn from them in writing your next review.

Rule 4: As a Reviewer You Are Part of the Authoring Process

Your comments, when revisions are requested, should lead to a better paper. In extreme cases, a novel finding in a paper on the verge of rejection can be saved by (often) multiple rounds of revision based on detailed reviewers' comments and become highly cited. You are an unacknowledged partner in the success of the paper. It is always beneficial to remember that you are there to help the authors in their work, even if this means rejecting their manuscript.

Rule 5: Be Sure to Enjoy and to Learn from the Reviewing Process

Peer review is an important community service and you should participate. Unfortunately, the more you review, in all likelihood the more you will be asked to review. Often you will be asked to review boring papers that are of no interest to you. While it is important to serve as a reviewer, only accept papers in which you are keenly interested, because either they are close to your area of research or you feel you can learn something. You might say, should I not know the work very well to be a reviewer? Often a perspective from someone in a slightly different area can be very effective in improving a paper. Do not hesitate to indicate to the Editor the perspective that you can bring to a paper (see Rule 10); s/he can then decide how to weigh your review. Editors would of course like to see you review papers even if you are not particularly interested in them, but the reality is that good reviewers must use their reviewing time wisely.

Rule 6: Develop a Method of Reviewing That Works for You

This may be different for different people. A sound approach may be to read the manuscript carefully from beginning to end before considering the review. This way you get a complete sense of the scope and novelty of the work. Then read the journal's Guide to Authors, particularly if you have not published in the journal yourself, or if the paper is a particular class of article with which you are not overly familiar, a review for example. With this broad background, you can move to analyzing the paper in detail, providing a summary statement of your findings as well as detailed comments. Use clear reasoning to justify each criticism, and highlight the good points about the work as well as the weaker points. Including citations missed by the author (not your own) is often a short but effective way to help improve a paper. A good review touches on both major issues and minor details in the manuscript.

Rule 7: Spend Your Precious Time on Papers Worthy of a Good Review

The publish-or-perish syndrome leads to many poor papers that may not be filtered out by the Editors prior to sending it out for review. Do not spend a lot of time on poor papers (this may not be obvious when you take on the paper by reading only the abstract), but be very clear as to why you have spent limited time on the review. If there are positive aspects of a poor paper, try to find some way of encouraging the author while still being clear on the reasons for rejection.

Rule 8: Maintain the Anonymity of the Review Process if the Journal Requires It

Many of us have received reviews where it is fairly obvious who reviewed the work, sometimes because they suggest you cite their work. It is hard to maintain anonymity in small scientific communities, and you should reread your review to be sure it does not endanger the anonymity if anonymous reviews are the policy of the journal. If anonymity is the rule of the journal, do not share the manuscript with colleagues unless the Editor has given the green light. Anonymity as a journal policy is rather a religious rule—people are strongly for and against. Conform strictly to the policy defined by the journal asking you to review.

Rule 9: Write Clearly, Succinctly, and in a Neutral Tone, but Be Decisive

A poorly written review is as bad as a poorly written paper (see Rule 3). Try to be sure the Editors and the authors can understand the points you are making. A point-by-point critique is valuable since it is easy to read and to respond to. For each point, indicate how critical it is to your accepting the paper. If English is not your strong point, have someone else read the paper and the review, but without violating other rules, particularly Rule 2. Further, as passionate as you might be about the subject of the paper, do not push your own opinion or hypotheses. Finally, give the Editors a clear answer as to your recommendation for publication. Reviewers frequently do not give a rating even when requested. Provide a rating—fence-sitting prolongs the process unnecessarily.

Rule 10: Make Use of the “Comments to Editors”

Most journals provide the opportunity to send comments to the Editors, which are not seen by the authors. Use this opportunity to provide your opinion or personal perspective of the paper in a few clear sentences. However, be sure those comments are clearly supported by your review—do not leave the Editor guessing with comments like “this really should not be published” if your review does not strongly support that statement. It is also a place where anonymity can be relaxed and reasons for decisions made clearer. For example, your decision may be based on other papers you have reviewed for the journal, which can be indicated in the Editor-only section. It is also a good place to indicate your own shortcomings, biases, etc., with regard to the content of the paper (see Rule 5). This option is used too infrequently and yet can make a great deal of difference to an Editor trying to deal with a split decision.

David

评审的十条简单法则
    让你文章发表的规则中有让你在事业中早成为一个评审者的建议。如果你顺从这些建议，通过完成你导师要求的评审工作，你就会发现这十条规则非常有用。没有决定一篇论文好坏的万能公式，大多数论文都是平庸之作，作为一个评论者你期望的是什么？我们建议，最重要的一点，你希望你所评审的期刊会为自己骄傲。立题新颖，科学论文中有太多的“我也”经常是它的一个必要条件，但也不总是这样。当然也有这样的论文，举例说，支持存在假设，或者为存在的发现提出一个新的或者修正的解释。在期刊范围之外，它归结为一个著名的争论并且其他的问题在“让你文章发表的十条简单规则”中提及过。一旦你知道什么样的文章是你所期望的，那么下面这十条简单的评审指导原则我们希望会非常有用。
    Rule 1: Do Not Accept a Review Assignment unless You Can Accomplish the Task in the Requested Timeframe—Learn to Say No
学会说不——除非你能在规定的时间内完成，否则不要接受评审的工作
    评审延误无论对作者还是期刊社都是不公平的。假设下次你有一篇文章被评审而评审者不负责的情况。你不会喜欢延迟，别人同样如此。此外，非常重要的一点出版成本与评审者的延误时间是息息相关的，没人会因此受益。
Rule 2: Avoid Conflict of Interest
避免利益冲突
    评审有多种形式，匿名，开放，双盲（作者与评审者互不相知）。无论过程怎样，这种行为有很高的道德原则。匿名的外衣不是为了掩盖科学造假。不要接受评审哪怕有极小的可能和你有利益冲突。举例来说，论文很差有可能被拒，但是其中有一个很好的点子可以用之于你自己的研究；或者，有人的工作与你的实验成果非常接近，这时候，冲突就有可能升级。大多数评审先提供摘要，只有你同意接受评审工作后才给你论文。如果存在明显冲突，就不要接受论文。存在冲突时，经常有一些不甚明朗的地方，无论是什么都要和作者商量。
Rule 3: Write Reviews You Would Be Satisfied with as an Author
像一位作者一样对自己的评论感到满意
  简洁，没有内涵的评论不可取。让你的批评或者表扬有理有据，逻辑条理。可能作者不知道你是谁，但是编辑知道，并且你的评论会为出版社的手稿追踪系统所分析。如同你作为评审者的形象为期刊社所熟知一样，你的评论质量也会为编辑所知，并在合适的时间成为令你骄傲的理由。很多期刊，包括这本，在论文接受或被拒之后提供给你其他评论者的评论，仔细阅读学习，让你下次的评论做的更好。
Rule 4: As a Reviewer You Are Part of the Authoring Process
作为评审，你是写作过程中的一环
    你的注解，当校订需要的时候，将成就一篇更好的文章。在极端的情况下，一篇处于被拒边缘关于新发现的文章会因为评论者详细的注解多次修改而最终通过，并被多次引用。你就是论文成功一名不为人知的合作者。牢记你是帮助作者的工作是有益的，即使这意味着拒绝他们的来稿。
Rule 5: Be Sure to Enjoy and to Learn from the Reviewing Process
确定喜欢并且从评审工作中不断学习
    同行评审是一项重要的基本服务工作，你应该参与。不幸的是，你评审的越多，你就有越多的可能被邀请参与评审。经常被邀请评审你不感兴趣的无聊的论文。作为一名评论者服务是很重要的，但是从接近自己的研究领域或者自己可以从中学习到东西的角度来说，只接受自己非常感兴趣的论文。你可能会说，我能在不很了解的情况下作一名评审者吗？经常从一个轻微不同的角度看问题会对提高论文的很有效。向编辑指出你看法的时候不要犹豫。编辑会乐于知道你的意见即使你对这些论文不是特别感兴趣，但是一个好的评审者应该会精明的利用自己的评审时间。
Rule 6: Develop a Method of Reviewing That Works for You
制定自己的评审方法
对不同的人来说应该是不同的。在评审之前将文章从头到尾读一遍。这样你会对论文的范围和新颖性有一个完整的认识。然后读期刊的作者须知，尤其你自己没在此类期刊上发表过文章，或者文章不是你熟悉的类型，比如一篇综述。在这个背景之下，你就可以详细的分析这篇论文了，提供一个概要或者详细的评论。评论有理有据，标出文章出色和不足的地方。包括作者的引用缺失经常是非常有效的提高论文质量的方法。一篇好的评论应该兼顾到主要问题和细枝末节。
Rule 7: Spend Your Precious Time on Papers Worthy of a Good Review
将时间花费在值得你认真评审的论文上
  出版或枪毙综合症使很多拙劣的论文在外审之前没被过滤掉。不要花费大量的时间在垃圾论文上（在你看摘要的时候可能不明显），但当你试图用有限的时间为它写评审的时候就显而易见了。如果这篇文章还有可取之处，在清楚表明被拒理由时适当的鼓励一下作者。
Rule 8: Maintain the Anonymity of the Review Process if the Journal Requires It
如果期刊社要求，维持匿名评审
  谁在评审这些文章，有时因为他们提议你引用他们的工作。在小的学术圈里很难维持匿名，所以如果匿名政策是期刊规定的话，你应当反复阅读你的评审以确认不会使你暴露。如果匿名性是期刊的规定，除非编辑同意，不要让同事看见文稿。匿名性只是期刊的规则而不是人们激烈争辩的宗教规则。严格遵守这些规则决定于期刊社的要求。
Rule 9: Write Clearly, Succinctly, and in a Neutral Tone, but Be Decisive
评审要清楚、简洁，不带感情色彩，但是要坚定。
一篇差的评审就像一篇拙劣的论文一样。明确作者和编辑明白你所指出的问题。点对点的批评非常有价值，因为它容易阅读改正。指出每一点对你接受他的论文来说有多关键。如果英文不是你的强项，让别人来读论文和评审，但是不要侵犯规则，尤其是规则2。另外，你可能被论文的主题所影响，不要妄加揣测。最后，关于这篇文章应否发表给编辑一个非常明确的答案。评审者即使被要求，也经常不给予等级评价。提供一个骑墙的态度毫无意义。
Rule 10: Make Use of the “Comments to Editors”
利用编辑评论
很多期刊让编辑有机会阅读作者看不到的评论，利用这种机会可以用简洁的几句话提出对论文的个人看法。无论如何，你的评审应该非常清晰的支持你的评论，在给编辑的评论中说“此文不宜发表”， 但在评审中却没有指出明显的不足，只会令编辑困惑。编辑评论中匿名性没那么重要，但要非常清晰的表达你对文章的看法。比如说，你的决定可能依赖于本期刊中你阅读的其他论文，可以在编辑板块中引用。也可以在编辑评论中指出着急的不足、偏见，关于文章内容（参看规则5）。这个主意经常用到，可以有效的处理与编辑之间的分歧。