Featured Posts
Posts are coming soon
Stay tuned...
Recent Posts
Archive
Search By Tags
No tags yet.
Follow Us
  • Facebook Basic Square
  • Twitter Basic Square
  • Google+ Basic Square

How to revise a manuscript for peer reviews

So, it's been more than a year since the last blog. I would like to think that this means I was productive over the past year. Now I am an Assistant Professor at the University of Alberta (Faculty of Kinesiology, Sport, and Recreation).

Today, I want to discuss the process to revise and resubmit your manuscripts in peer reviews. This is a clearly important issue, as many manuscripts get either "Major Revision" or "Revise and Resubmit" decision after the initial reviews. Based on my review/editing experiences as well as my own papers, I feel that 90% or more papers that would be eventually published go through this (so, if you get one of them, congratulations on the good start!). However, I found that surprisingly little is written about this process. Most of us get some copies of response/revision material from our mentors, and perhaps see them going through a couple times as a co-author, before we do it ourselves. Although these are good practices, I believe that more can be done. Specifically, developing an efficient approach to this crucial component of publication could systematically improve your productivity (e.g., less rounds of reviews, more time for other works). In this blog, I share my process so far. First I discuss the general principle, and then describe my process from the beginning (so you can follow it if you want). If you have other ideas, let me know.

The general principle or something you have to keep in mind throughout revision is that revision is about both logic (science, rigor, etc.) and emotion. Many people get the former, and forget the latter. The fact that you are invited to revise means that at least most of issues identified by reviewers are not fatal, and thus fixable. You have to CAREFULLY understand what each comment means, and logically address it (more on why I "yelled" carefully below in #1). This is not so difficult when you are trained well as a (future) Ph.D. But, don't think that this process is purely logical and you can say whatever as long as (you think) you are right. Reviewers (and editors) are humans and have emotion. Upsetting them never works for you.

One way to manage emotions in reviews is to watch out your tone. This is especially difficult for non-native English speakers. For example, I was writing sort of a "commentary" piece that directly builds onto particular papers by others. When I stated Dr. X "failed" to do Y, Reviewer 1 (presumably Dr. X) reacted strongly and told me to take out all "fail" from my paper. So it's important to avoid the language that could cause negative emotions, especially in a small field. The above example also shows another bad practice: sounding like critiquing the person not their work/comments. Make sure that in revision, you focus on comments--not Reviewers, even if you have a very good hunch of who they might be--and science, research, or whatever you do.

Another common mistake is to try to "explain away" most comments. As an editor/reviewer, I sometimes come across with a response sheet where authors look reluctant to make any changes and list all excuses not to change the manuscript. Often these excuses are not good reasons (at least not good enough), and you could get slammed (i.e., rejected) after that. This type of unwillingness does not communicate respect for reviewers' expertise and appreciation for their time, which makes reviewers mad. A mistake in the opposite direction is the overuse of flattering. Does every response of yours begin with "thank you very much for this comment...." or "this was a brilliant point!"? That may be a cultural issue. But, at least for English journals, I say this rather does harm. If there are 10 comments, I would use this type of "special thanks" for one or two really good comments. That sounds genuine. 10 out of 10 "brilliant point!" is annoying or at least distraction. Remember, as a reviewer, nothing feels better than authors' taking your comments seriously, making diligent changes, and improving their paper. Let your work speak the "thank-you."

Now, I go through my revision procedure from the top. They are:

1. Read through the comments and take time away.

2. Use a response sheet as a compass of revision.

3. Revise the main document (and others like Tables).

4. Read the whole manuscript to ensure the flow.

5. Revise the response sheet and insert the page/line numbers in the response sheet.

Note that at the end, I will briefly discuss where I would involve collaborators in this process to maximize the efficiency. However, this would greatly depend on which discipline you are in and how many co-authors you have.

1. Read through the comments and take time away.

The very first step of revision is to understand the comments. My advice here is: take your time. First, quickly read the comments right after you get the decision email. Although the assumption here is that your paper did not get outright rejected, it's likely that there are some tough comments, some of which make you feel upset a bit. I say that's normal. I heard some people tell me "detach yourself and don't get upset." You will be over time, but that's a useless advice, as we can be like that only with experiences. You get upset because you care about your work and you have pride in it -- that's a good sign. Inform your co-authors about the decision, and close the email. Ideally, go somewhere to have fun.

This part would be very much a personal preference and situational (e.g., how much time you have for the revision). But, I tend to spend a couple weeks, without actually revising or doing anything. Yet, those comments are somewhere behind my mind. Especially those tough and upsetting ones -- how can you forget? The interesting part is that letting the comments float in your mind like this, at least for me, helps exploring the meanings of the comments, logically dissecting them, and finding good responses. I often find some of earlier thoughts are more emotional and not necessary effective or appropriate. Also, even better is that while letting the comments sink in, I often hear, read, or watch something that's quite useful for the revision. Dumb luck -- maybe. But, I would like to think that we are often surrounded by those useful pieces of information, but the situations make us attend to them. Give it a try.

2. Use a response sheet as a compass of revision.

Now that you are back after a while (days or weeks, perhaps), go over the comments again with hopefully less emotional turmoils. You open up a Word file. Put your paper title, a cover letter material (Dear Editor... Thank you for ....), and make some space for an overview of responses. You don't write this now. Then, you create a long table with 2 columns. The left side is for the comments, and right side is for your responses. And I copy and paste all the comments (except for positive ones) into the left side. One cell per comment. Some people say that you should paraphrase comments, especially long ones. But, I find that that's not necessary and waste of your time. Just copy them.

Now the fun begins. Now that the comments are separated into the cells, take a close look at each one. What is the Reviewer saying? This is critical. As editor/reviewer, I often see miscommunication in revision. Reviewer said A, which Authors took as B and responded accordingly. Waste of time. Sometimes, one comment can be multifaceted. Once you think a good idea of what the Reviewer meant, write up a response. A major distinction here is whether you make a change or not; the latter comes with strong yet respectful rebuttal. When you are making a change -- and you should make as many changes as possible within a reasonable time and space, describe how you change the commented part (or elsewhere). When you write this response sheet at first though, you just say something like "ADD THE INFO ON A RESPONSE RATE," where the uppercase indicates that you would come back later and add actual revised parts. Also put the pager and line numbers so that the Reviewer can identify it in the revised paper. At this point, I just put "......... (p. X, L X)." Easy, right?

One thing you should NEVER do: write a response like "the commented part was expanded." Reviewer 1 is shouting, "HOW!?" Explain how the response was made, and how it addresses the concern raised in the comment. The latter part is important as sometimes, there are responses but it's not clear how they address the comments. You may want to drop some details like re-run statistical results (e.g., "based on the requested re-analysis, the coefficient b* was .XX). Or, you can even quote the revised parts, if that's not too long or the exact words are important. Note that some reviewers may only read this response sheet (hopefully not, but certainly a possibility). At least, many reviewers read this response sheet before going into a longer revised paper, which means whatever emotional responses your response sheet is giving would set up the tone for their reading of the revision. You want to keep them as happy as possible.

If you choose not to make a change, have a darn good reason. Perhaps, the Reviewer misunderstood what you wrote (then you want to clarify a bit), the Reviewer suggested an alternative approach that has been discredited in the literature, the Reviewer made a problematic assumption, etc. Either way, double-check you understood the Reviewer's comment right. Sometimes, we see what we want to see, find a problem(s) in the comment that do not exist, and critique it. This does not do well for the emotion issue I mentioned above. Also be respectful. Acknowledge where the Reviewer's comment is coming from, and then present a counter-argument(s). Be succinct and direct at the issue. Cite the references to support your point(s) in your response sheet (but not necessarily in your revised paper).

Sometimes, you come across with comments you don't really understand. That's not uncommon. What I would do is to have a fresh set of eyes. If you have co-authors, great. Have they take a look of those comments and ask how they would interpret them. Better yet, if you have co-authors who are specialized in the commented parts, you might want to ask them to address them (before letting them address the parts in the paper, you should have them write the response first and see where they are going and if you are okay with it from the overall perspective). I understand that some of reviewers' comments are not so clear, which could cause frustration. However, the old expression fits here: the onus is on authors. YOU as the author have to make the full effort to understand what the comments mean and address them. Nothing is more annoying than a counter-argument based on a wrong interpretation of comments, from a reviewer's point of view. Plus, they get to decide if your paper will be accepted or not; you don't.

"Well, Shin, but sometimes, we don't know what reviewers mean!" Yes. When that happens, be honest. You can start your response by stating that "We were not so clear about what Reviewer 2 meant by X." Then, add your interpretation: "We interpreted as "Y." And continue with your response by assuming that Reviewer 2 meant Y by X. And towards the end of that comment, you "could" invite them to clarify what they meant by X if you want, by saying: "If our interpretation of X was incorrect, we would appreciate a clarification on it." At least this response communicates that you sincerely tried to understand and address the comment, while acknowledging the potential miscommunication.

Last, but very importantly, do not spend too much time on this process of making a DRAFT response sheet as a compass of your revision. It's a draft. You may revise your interpretation of certain comments later, or change your response strategies. If a few comments were very long, unclear, and/or complex, skip them and go through other comments first. My field tends to be comment-heavy: I often get anywhere between 2 and 5 pages single-spaced, if not more, for Major Revision or sometimes even Minor Revision. But, I spend perhaps 1-2 hours on crafting this draft response sheet. Note that you don't have to have actual revised sentences or re-run statistical results. The point of this draft is to give you a sense of direction in revision.

3. Revise the main document (and others like Tables).

Finally you actually change the submitted documents and other files like Tables, Figures, and Appendices. Use your compass: the draft response sheet. A good thing about having the response sheet is that it helps you focus on your responses, rather than, for example, how unclear the comments are. It saves me a lot of time and emotional upheavals. I can be very mechanical. When you do change the documents, make sure that you use the Word's track-change mode. Plus, "save as" the file with different names (e.g., "rev1, rev2, rev3") occasionally so that you have a track of revision at different stages. Also what I recently started to do is to color-code revised parts: for example, red for Reviewer 1 and blue for Reviewer 2. This is because that I often also edit the entire manuscript mostly to make it shorter and generate space for revision. After editing, there are so many "bubbles" due to the track-change mode, which is quite frankly ugly. More importantly, these editorial changes are now difficult to be distinguished from more substantive revisions based on the comments. This does not make reviewers happy. So, unless the journal specifies otherwise, I use a clean file with these color codes (Word's highlights) so that Reviewers can jump to the important parts.

4. Read the whole manuscript to ensure the flow.

I know you are busy, but do read the revised manuscript from the top to the end, after revision is largely complete, at least once. Often, we change some terminology, which needs to be consistent throughout the paper. If you added something in the literature review, perhaps you want to reflect that in your discussion section (or vice versa). Slightly different (hopefully) statistical results due to re-runs should be reflected in both Tables and text (and sometimes Figures and Appendices). This read-through is especially important when you had your co-authors directly change the manuscript file. From a reviewer's perspective, reading a "patch-work" paper with different writing styles is not a pleasant experience (I have received this comment myself).

5. Revise the response sheet and insert the page/line numbers in the response sheet.

Now, you have a fully revised text and additional files. Revise the response sheet so that your responses and revisions are consistent with each other. Insert quotations of revised parts, if you want. Insert the page and line numbers of the revised parts. Usually, I add line number to the revised manuscript file (Word does this). That way, you can avoid the issue of confusing Reviewers about the line number you counted and proof-read PDF automatically generated. Make sure Reviewers can find the revised parts very easily, which again makes them happy. Remember: "Happy reviewers, happy academic life."

As you can see, the most of my revision work is done in the Steps 1 and 2. The important things are that you should give yourself time for the Step 1 (letting the comments float in your mind) and doing the Step 2 mechanically. When you get the direction right, the subsequent revision steps become surprisingly smooth.

Finally, about when to invite your co-authors into the revision process. Again, this depends on disciplinary norms, group dynamics, if you are student or not, etc. That said, you could invite them into Step 2 in terms of understanding (and ensuring that you understood right) the comments. I often have my co-authors to read the draft response sheet so that we can agree on the direction of revision. I ask them to take up a few responses and associated revision. When my co-authors are too busy, I sometimes only ask them to take a look of the draft response sheet, just to make sure that my interpretations of the comments are not too off. As such, what this draft response sheet does is to make co-authors' roles in revision more proactive. They are not reading your responses and revised parts, after they are done. What if you are wrong about the interpretations of the comments? Hopefully this process would increase your efficiency and productivity.

Best,

Shin