[wild_reilly]

Hi there,

I don't have my phd (considering starting one soon), but the area of research I was involved with when I did my masters is quite specialized and there aren't a lot of us... so I got asked to review an article for an academic journal that is doing a themed issue in that subject area.

Having never reviewed an article, I hesitated... but the response I've received is that I know as much as anyone (well, not anyone - but a lot of people) about this, so why not?

What I'm looking for is some advice on how those of you who review articles go about doing it! What do you look for? How do you provide your feedback (in what format, etc.)?

The person running the show said he doesn't use a particular template, and said to "just review it as I would any other academic article". I do have the support of my masters committee, but I am not close to any of them anymore and would love some help... I'm feeling quite out of my element here!

Thanks so much!

 

[claras_mom]

I haven't done peer review, so take this as you will...

First, it's a terrific opportunity, so be brave!

Second, I'd read the article as if I were doing a critical reading for a good friend. In other words, is it factually correct? whether or not you agree with conclusions, are the arguments well-presented? Are there any wacky typos? Read with the idea that you want the best for the writer. I suspect that as you read and make notes, the review will pretty much write itself, with the positives as well as the areas that might need more work.

Are you being asked to make a recommendation as to acceptance or rejection or is this more a case of feedback for the purposes of polishing an article?

 

[emilysmama]

Did the editor provide you with a form? Not that it really matters, but that would be a good start in boosting your confidence. Do you know anyone that has reviewed any articles, even if it is outside of your discipline? If so, then ask them to let you glance at what they wrote and it will help you a lot, even if you don't understand the subject matter.

If not, then that is fine too.

First, read the article really carefully. I'm in physics, so I always go through and check the math in the equations too, but that may not apply in your discipline.

A typical review will be about a paragraph or two long, more likely one paragraph.

I would write the last sentence first. It will be one of the three. "I recommend this manuscript for publication without any changes." or "I recommend this manuscript for publication with the following (minor) changes." or "I recommend this manuscript for publication contingent on the following (major) changes". or "I do not recommend this manuscript for publication." or "I suggest that the author submit this manuscript to Different Journal, instead".

Then explain why, in a few concise yet complete sentences.

If you like/don't like the article, explain why you like/don't like the paper.

In science, I might say that the author has developed and demonstrated a novel measurement technique and why it is better than what is already known in the current literature.

If there are minor modifications, like typos, etc, you can point them out.

If there are minor suggestions that you think would help make the article easier to understand, but you want to allow the author to have the option to decide whether or not to implement your suggestions, then write a couple sentences describing them.

If there are substantial changes that you think absolutely must be made before the paper would be acceptable for publication, then say so, and describe exactly what they are. (For example, maybe some point that needs to be explained in more detail, or a diagram/figure that you think the author should put in the article, or for a scientific article - a graph illustrating the data in a different way.)

Don't recommend the article if the there is something fundamentally flawed about the premise of the paper (like the data violates the laws of physics), if the paper is not demonstrating what it says it does (like the data is not convincing), etc. Above all, if you don't recommend the article, be very diplomatic. You don't have to sugar coat things, but you don't want to sound like you have an axe to grind. I have had direct competitors review my papers, and even though the reviewers were anonymous, it was clear from their bashing comments exactly who they were. I've never been an editor before, but I would think that not being diplomatic does not do anything to raise the editor's opinion of the reviewer.

I have had reviewers point out that I need to cite a certain article. (Usually an article that was written by the reviewer.

) That is fine up to a point.

 

[emilysmama]

Oh yes, one other thing. Every single reviewer's comment I have ever gotten, started out with a quick one or two sentence that summarized what my article was about. I've never understood why. You would think that if I wrote the article, I of all people would not need to know what the article was about. I guess it is so that the author will believe that the reviewer actually read the author's article before passing judgment on it?

As for what kind of format, I would say that a simple paragraph would be fine. Is that what you mean by format?

 

[kate3]

Generally, I start out the review by stating "Thank-you for the opportunity to review ...... I found it to be an interesting article/topic/study, etc. I always try to say something positive first, even if I'm not going to recommend it for publication.

Then I go section by section: Intro, background, lit review, methods, analysis, findings, implications, etc. If it all seems fine I may write nothing or just leave a positive comment. If it is something specific, I refer to the line # and ask for clarification.

I judge whether or not I recommend it for publication if it contributes something new to the dicipline. It could be extremely well-written, but if it doesn't move the current knowledge forward than I don't recommend it.

 

[tarahsolazy]

I review for medical journals, and mostly patient oriented or epidemiology research, and I do what Kate3 does, go section by section. I pay close attention to study design, subject selection, outcome definitions, etc.

I typically review for journals that do it as an on-line form, with sections to write stuff.

Maybe someone from your committee can email you reviews of papers they've written, or maybe you have some from your master's days. That was helpful to me when writing the first few reviews.

 

[~pi]

First, it's a great opportunity, so definitely do it.

Second, I've done a little peer review, not tons, so take this FWIW, but for journals with no particular rubric, I do the following:

1. Quick summary of what the article is about (helpful for editors.) I try to mention how it fits (or doesn't fit) with the journal and audience.

2. List any major issues with the article (e.g. faulty conclusions, missing info, other major methodological flaws, missing reference to other key issue in topic, etc.) I word things like, "The authors did not mention ...," "The authors draw the conclusion that ...; however, they do not note that this could also be due to ..."

3. List any minor issues with the article (e.g. minor stats issues, grammar, etc.) Note: If the manuscript does not have line numbers, make reference to pages, paragraphs and lines. E.g. "Page 2, paragraph 3, line 4: {incorrect spelling} should be {correct spelling}."

4. Make a recommendation: Accept as is (rare), Accept with minor revisions, Accept with major revisions, or Reject.

I second the pp who suggested asking your committee to send you a couple of copies of their reviews to give you an idea of how it's done.

 

[wild_reilly]

This is some great feedback! Thank you everyone for your suggestions.

I think I'm ready to go now - one of my old committee members said he would also read the article for me and let me know if he thought my review is on track. So that combined with everyone's advice here should get me through!

You know, I have to say that when I was pregnant and started snooping around MDC I never knew I would have such a great resource for this kind of stuff in here!

Thanks!

 

[Geofizz]

Quote:

Originally Posted by emilysmama Oh yes, one other thing. Every single reviewer's comment I have ever gotten, started out with a quick one or two sentence that summarized what my article was about. I've never understood why. You would think that if I wrote the article, I of all people would not need to know what the article was about. I guess it is so that the author will believe that the reviewer actually read the author's article before passing judgment on it?

I review way too many articles each year.

The review serves two purposes. It informs the editor of the suitability of the paper and it provides feedback to the author. That first statement is for the editor -- it puts the paper in context to aid in deciding whether the manuscript is suitable for that journal.

There really is no set formula for writing a review. I focus on the science and the suitability of the inferences and conclusions based on the data. I don't comment on every writing/spelling issue I find, but if something stands out that a copy editor won't find, I call it out. Otherwise, I might make a statement like "the paper needs to be revised with the assistance of an editor."

OP, ask your committee to see reviews they've written for that journal or similar ones. I'm sure review style varies between fields (and I get a huge variation in review style that I get on my papers.)

 

[simcon]

As an editor who deals with peer reviews every day, I would second much of this advice... most useful to me as an editor (and our authors) are reviews that engage substantively with the ideas presented and the ways in which they're developed and supported, offer detailed and constructive critiques, and comprehensively assess the manuscript.

Less helpful are reviews that (1) ask that a different project be written, (2) offer heaps of general high praise and no detailed or critical feedback (3) instruct the author to primarily reference the reviewer's own work (4) offer only a sentence or two, whether positive or negative (5) demonstrate a lack of knowledge about other major works in the field (6) come in very very late.

I'd add that it's really not worth your time to give specific copyediting feedback about typos/misspellings, etc.--the journal will be paying someone else to do that, and you offer a different kind of expertise. The most I'd suggest is saying something like "there were a number of typos/second-language issues/grammatical inconsistencies that a copyeditor will need to correct" That way the editor knows what instructions to give the copyeditor, and you can save your time and energy for the feedback that's more important.

I'm sure it'll go very well--you've got excellent feedback here and it's great that you have the support of your committee as well!

 

[LynnS6]

I teach students to peer review, and I use the guidelines developed in this article:
http://wac.colostate.edu/journal/vol3/chisholm1.pdf

While the article is aimed at writing courses, it can be easily modified for journal peer reviews. The major points of a peer review that I use are:

1. Identify Values in the paper: Give positive feedback for your colleague's paper. What did you like about it? What are the best parts? What are the strongest points?

2. Describe the paper: Explain the main ideas of the paper and how it is organized. What did you hear as the main points of your colleague's paper?

3. Ask Questions about the Paper: Ask questions about your colleague's meaning and wording. For a journal review, I would then turn those questions into comments about where more clarity is needed. Be explicit about what you see to be problems.

4. Suggest Points to Revise: Give suggestions for improving the paper. Suggest places that need more information, more clarity or re-thinking. Tell what you wish the paper had said or what it might have said.