Surviving peer review part III: how to write a response to reviewers

​Receiving the decision letter from a journal is one of those ‘heart in mouth’ moments. The outcome of months, or years, of work might be decided in this one email. Being given the opportunity to resubmit a revised version of the manuscript can be a huge relief. However, this rarely means that the job is finished—and doing revisions to address the reviewers’ comments can still represent a substantial amount of work (in some fields, it can even mean conducting new experiments). One of these jobs is having to write up a set of responses to the reviewers’ comments. Here I provide some advice on where to get started, and the do’s and don’ts of responding.

Writted by Damien
@DamienFarine

What is involved in revising a manuscript?
 
When receiving the invitation to resubmit a revised version of the manuscript, the journal should provide guidelines on what they expect this revision to contain. In general, this almost always includes at least two documents: the manuscript with changes highlighted (i.e. all of the edits in track changes) and a document containing a detailed list of responses to the reviewers’ comments. The latter is particularly important, as it is where you outline how you took the reviewers’ comments on-board and used them to improve the paper (at least in their eyes). I note here that invitations to resubmit does not, by any means, guarantee eventual acceptance—I have had a number of papers rejected after resubmission—and doing a good job of the revision is always important.
 
Where to get started
 
The first step is to drop the editors and reviewers’ comments (which we usually receive by email) into a word document. Here you will alternatively provide the comments and your responses (making the response clear, e.g. by using a different font colour). Once this is done, the task can seem extremely daunting: a set of reviewers’ comments can easily run into many pages (I’ve had some with literally hundreds of comments). My strategy here is twofold. I first read the reviewers comments thoroughly, to really understand the main points of concern (see previous post). However, I then do not work on these yet, but rather I start with fixing all of the really small comments and minor line changes, such as small typos or wording changes. This achieves two aims: first, it rapidly and substantially reduces the size of the task at hand (e.g. from hundreds to maybe a dozen comments), and second, it gives the opportunity to familiarise myself even more with how the reviewers’ major comments map out onto the manuscript. It’s usually been several months since I’ve read the manuscript, so working through the small changes helps to reload my previous thinking into my memory.
 
What is contained in a response to reviewers’ comments?
 
Different authors have different ways of constructing the response to reviewer comments, and editors also have different preferences. In general, it is always necessary to address ALL the comments. ‘Addressing’ means writing a sentence or more on what has been changed in the manuscript to address the comment by the reviewer (and, ideally, why this particular change was made), or a justification for why nothing has been changed. Responses to reviewers can be time consuming because it is commonly-expected to include details on where the change was made (e.g. line numbers in the track changed document) and/or the changed text in quotation marks. For example:
 
Reviewer: The authors do not provide a definition for term X.
Response: We now provide the following definition on line 55: “Term X is defined as Y”.
 
In addition, I like to provide a short summary of the major changes at the very start of the response document. This usually involves thanking the editors and reviewers for their time, summarising the key points raised by the reviewers (i.e. those things that are preventing the paper from being published with minor changes), and summarising how these have been addressed. For example:
 
We thank the editor and reviewers for taking the time to provide constructive feedback on our work. We have addressed all of the reviewers’ comments, and these have substantially improved the clarity and presentation of our work. In our revisions, we paid particular attention to three key concerns raised by the reviewers and editor:
 

1. The reviewers both questioned the logic of A, and we now clarify this by X.
2. Reviewer 1 suggested that we needed to explore a larger range of parameters in our model, and we address this by conducting a sensitivity analysis (details are found in Y).
3. Reviewer 2 proposed an alternative mechanism, which we now acknowledge in our discussion (L502: “The patterns we observed could also be driven by an alternative mechanism, specifically …”).

 
Strategies for writing responses
 
Below I list some strategies that can help smooth the pathway to successful acceptance of the final manuscript.
 
Be positive and avoid confrontation: ALWAYS write the responses using a positive tone. Even if you disagree with the reviewer, or if they totally misunderstood a segment of the paper, do not criticise them. For example, a statement such as ‘The reviewer clearly does not understand the literature on this topic’ would be better written as ‘Our understanding of the current literature, after rechecking it in light of the reviewer’s concerns, is that our original framing of the hypothesis is correct’.
 
Be succinct: Reviewers are busy, but they (almost always) care about trying to help you improve the work. Don’t burden them with pages and pages of details. In general, the response to reviewer comments should not contain information not available in the main text. Provide the necessary information for the reviewer to understand where you are coming from, or what changes you have made, but keep the response text as short as possible (e.g. the longest should be 2 to 3 short paragraphs).
 
Avoid addressing reviewers directly: The tone of the responses always sounds much less per personal when reviewers are addressed indirectly. For example, use ‘The reviewer makes a useful observation’ instead of ‘He/she makes a useful observation’ or (even worse) ‘You make a useful observation’.
 
Leave the quotes and line numbers to the end: As co-authors read and proof-read the final text, they might make many small edits to the revised text. These edits can change the final line numbers and the text that is being quoted in the revisions. It doesn’t look good to the reviewers if there are mismatches between the manuscript and what’s in the revised text, so don’t spend hours entering the line numbers or finalising the quotes until you are sure that they won’t change again before resubmitting. On that note, also make it clear (in the start of the response) if the line numbers correspond to the highlighted manuscript or the clean manuscript (some journals also ask for a version with no track changes).
 
Keep the original submitted PDF handy: As you go through the edits, have the original PDF that was produced by the journal handy. It is what the reviewers will have had access to, so it will contain the original line numbers, formatting, etc. This document will help you identify where in the document the comments are addressing, because as soon as you update the manuscript the mapping of text to line numbers will change.
 
Conclusion
 
Peer review can be excruciating. For better or worse, it’s designed in such a way that triggers all of our natural insecurities—it directly critiques a product we’ve spent months (at best) or years (usually) working on, it often comes with terms such as ‘reject’, it usually involves a great deal of anonymity, and the outcome of peer review can make or break careers. The good news is that there is no one journal that is the gatekeeper of publishable research, giving us the chance to address and overcome the perceived limitations of our work and, eventually, get it published. Revising does take time, and sometimes involves several rounds of rejection. Remember, this is normal—most top journals have an acceptance rate of <20%, meaning that the vast majority of papers are rejected at least once.
 
If you are early in your publishing career and your paper is rejected, do try and seek some mentorship. The journey can be particularly arduous for those who have had few opportunities to see first-hand the benefits of peer reviewer feedback on their own research, or to go through the process of peer reviewing themselves. It might be helpful to (i) be walked through some more senior researcher’s own experience with a manuscript repeatedly failing but eventually finding its feet, and (ii) to try get some experience with reviewing a manuscript you’ve never seen before (e.g. through a journal club, or by offering to read a colleague’s work). These might help give insights into how simple it is to fail to clearly convey crucial information to the reader.
 
In the end, whatever happens, try to avoid taking any of this process personally. Your own Science can seem like a very personal journey, but for reviewers it is usually just another part of a very busy job. It is unlikely that they will have any intentions to cast personal judgement on your quality as a researcher.