Science has a thousand problems, but the time it takes for our manuscripts to be peer reviewed ain’t one. At least, that’s how I feel. How about you?
I hear folks griping about the slow editorial process all the time. Then I ask, “how long has it been?” And I get an answer, like, oh almost two whole months. Can you believe it? Two months?!”
That boggles me. I don’t know if my attitude is the norm or the outlier. I guess this poll will tell us.
I bet most of us have stories about extraordinarily outliers. When I was in grad school in the mid-late ’90s, manuscripts were still submitted by post. You had to send a few photocopies, and the editor would pop them in the post to prospective reviewers. Once, an editor physically lost and then forgot about my manuscript while moving to a new job. About nine months later, when I got around to asking, he was like, “Whaaa….? oh crap I’m sorry,” and expedited the reviews. Which took maybe six weeks. Just last year, I had a colleague who had a manuscript linger for about a whole year with no notice. The Editor-in-Chief failed to respond to emails and phone calls. That’s beyond hideous.
So I’m totally cool with 2-3 months to get reviews and a decision. It’d be nice for a desk reject to come within a few days, but if it takes the editor another week to get to it, that’s fine with me.
Could the review process be faster? I guess, but I think getting it much lower than what we have right now is really hard to do.
I realize that the progress of my career isn’t contingent on how quickly new papers are added to my CV. For grad students, postdocs, and junior faculty, the speed of the publishing pipeline matters more.
Most of what we publish doesn’t need to get out there right away. On one occasion I’ve dealt with an author who had a urgent reason to expedite the path to publication, for conservation/policy purposes. But typically, speed is a matter of personal preference or career utility. (That’s also why I say ‘meh’ to posting preprints in repositories. It’ll come out when it comes out. Whatevs.)
What might it take to speed up the time to decision beyond what we have now? There’s a great post about how to minimize time-to-decision by Emilio Bruna, on the Biotropica blog. He would know because he runs a tight ship at Biotropica where decisions happen mighty quickly.
If you haven’t been on the editorial side of a journal, please know that there are a lot of steps in the process.
The routing of manuscripts varies among journals, so the process varies, but there are a lot of moments where things pause and each paper goes through the digital hands of several people:
- The editor-in-chief gets it, then decides to desk reject or assign to a handling editor. (hours to a few days)
- The handling editor looks over the paper and decides whether to desk reject. (hours to a few days)
- The handling editor needs to find reviewers who are willing to review the manuscript. (days to multiple weeks. There is little written about this online as far as I can tell, but it is a bugbear for handling editors. I have a post about this in the queue of posts-to-be-written.)
- The reviewers return their reviews, probably within a few days before or after the deadline. (2-4 weeks, but perhaps much much longer)
- The editor reads the reviews, scrutinizes the paper to a certain extent, and makes a decision. (hours to a week, especially if the reviews are conflicting or tricky in some way)
- The decision then may go through the editor-in-chief. (hours to days)
I should go to a great length to point out that even though I’m not picky about how long it takes for me to get my manuscripts back, I work hard as an editor to keep review times low. Even though it’s not a big deal to me, it’s something that sizable fraction of the community thinks is really important. I’m pretty proud of my record as an editor in keeping review times low. I want the society journals that I support to have good relations with authors who value the society’s journals. So I do my best to handle these things promptly. Sometimes I’ll have a complete set of reviews for more than a couple days while I work on the decision, because I’d rather do things right than do things quickly. But it’s always going to be a high priority for me. Authors deserve that kind of treatment. I imagine most other editors would agree with this sentiment.
When a review is taking a long time, I suggest there are three major possibilities:
- The editor is having a really hard time finding people to review the paper.
- A reviewer is being irresponsible.
- Someone dropped the ball due to negligence or circumstances beyond their control.
When I was dealing with some personal stuff a few years ago that really interfered with work, there were a few weeks when I just didn’t even log on to the manuscript management system. And that was fine. Maybe there was an author venting on twitter that I was taking too long. But so what, I had other priorities at that moment in time.
It takes 3-4 people to get a manuscript to decision in a short time frame and if any one of them has something come up and they don’t do their job promptly, then the whole process gets gummed up. As an author, when it takes a while for me to get a manuscript back from the editor, I’m more concerned that everybody is okay than I am about that line on my CV. The community is small enough that there’s a good chance that I know who the editor is, and perhaps we’re even conference buddies or have collaborated together. When I don’t get a review back quickly, I can be confident it’s not because they have it in for me, and not because the process is broken. It’s just because peer review involves human beings.
[Update, 2 hours after posting: Meg Duffy brought to my attention this great and highly relevant post by Stephen Heard, about how long peer review should take. By his numbers, under the best of circumstances, getting a paper back within seven weeks is pretty good, and any little slow down can get it well past the 2 month range. Stephen also pointed to a peer-reviewed paper that showed how a sizable fraction of scientists have entirely absurd expectations about how long peer review should take: “Peer-review speed was generally perceived as slow, with authors experiencing a typical turnaround time of 14 weeks while their perceived optimal review time was six weeks.” One caveat is that article is about conservation biology journals, which is one corner of our discipline in which the prompt dissemination of findings might be of import.]
For me, it depends on the journal, and my previous experiences with it. If in the past, things tended to move apace, I’d expect the same. If I know going in that it will take 6 months, that’s fine too. I’m not necessarily under the same pressures, though, as others (e.g., students that need n accepted/published papers to be awarded a degree, which is very silly https://labandfield.wordpress.com/2015/04/12/publication-requirements-for-graduation-are-a-terrible-idea/).
I’m also a subject editor at a small(ish) journal, and handle maybe 2-3 submissions a year. Our goal is submission to first decision in 6 weeks, but there are obviously lots of factors (as pointed out in Biotropica).
Someone I know is an AE for a journal that requires 3 reviewers for papers. He recently told me he had asked 23 people so far and still had not had 3 who said they could do the review. Presumably it takes a while to go through that whole process that many times!
I generally try to get reviews done within days of getting the request, figuring I can do my little part to speed the process along. But, as you said, sometimes things come up. I had been planning on not doing any reviews until my baby starts daycare, but agreed to one for several reasons that aren’t worth getting into. I thought I was going to be able to do that review last Thursday. But then my 5 year old was home sick from daycare that day and so the review didn’t get done and now it’s seeming much trickier to fit it in.
Finally, as a potential reviewer, I can say no if the next couple of weeks look like they’ll be too busy. But, as an AE, I can’t control when the reviews come back in, and sometimes that ends up being at a bad time.
I think it depends a lot on the paper and journal (I edit for two different journals, and review for many others). In my field of paleontology, lengthy manuscripts (50 to 100 pages) are relatively common, and in those cases it really is unfair for anyone to expect a turn-around time from submit to first decision of less than 8 weeks. As an author, I wouldn’t think about bugging the editors/journal staff until the paper passes the 12 week mark (unless it’s something like a one page note).
That said, in my experience most reviewers will take as much time as you give them — if they have a two month deadline, they’ll get it back in two months, and if they have a two week deadline, they’ll get it back in two weeks. It’s a delicate game to balance being fair to the reviewers with being fair to the authors. In general, and based on my own experiences as both reviewer and editor, around 3-4 weeks as a reviewer deadline is a good (and humane) rule of thumb for most papers in my field.
As editors, I think it is also good to keep track of reviewer turn-around times to some extent. If there is a reviewer who is consistently super late (e.g., 8 weeks to submit a review promised in 3), it’s best to find other folks to do the reviews. We’re all late from time to time, but consistent and egregious delays aren’t fair to anyone.
Terry, I think it’s actually a good thing that you wrote this without seeing my post – as I think we came out in pretty close agreement. That makes me think we’re actually right, and maybe will lead some other folks to think so too. Cheers!
Reviewers are us. Wonder how many people who complain that it takes 6 whole weeks (or whatever) for a journal to handle their paper say yes to every review request they receive and complete every review the same day, without fail (and if they do, will continue to do so as they number of review requests they receive, and their other duties, ramp up over time).
Everybody wants to go to heaven, but nobody wants to die.
Seems like 6-8 weeks is reasonable given your estimates of each stage. (6 weeks is the sum of maxes for all stages except finding reviewers.) So why was the typical turn-around time 14 weeks (in the linked paper)? I still don’t get it.
I don’t really like reviews to come back too quickly (i.e. 2-3 weeks). Once I submit an ms, I move on to something else and try to forget about the submitted ms for at least a month. I think it helps to regain perspective when the reviews do come back. Obviously, a year is way too long, but faster than ~4 weeks is not necessarily better in my opinion.
Having just become an editor at new journal-like operation [1], I am rapidly gaining sympathy for other editors and an understanding of both the effort and time to find reviewers as well wrangle them to submit reviews on time. That said, I think having pre-prints makes this a lot less of a problem. If you want your work to be sharable (say, prior to a conference), the preprint can be made available on your own schedule.
[1] https://ropensci.org/blog/2016/03/28/software-review
Oh, and I recommend that whoever voted for “2 weeks” be made EiC of a major journal immediately. :-)
Great post. Agree with Jeremy & jeff above – accepting review requests and finding the time to do a thoughtful review within all your other personal & professional commitments is the best way to keep some perspective while waiting for your own papers. I think longer (but not too long) waits make better papers – both because the reviewers & editor have had time to read it thoughtfully, and because the author has had time to refresh their perspective on the research.
Thanks for this post! Becoming an AE has drastically changed how I think about this process. I get turned down a lot, and a lot of reviewers never even respond to a request. And then there’s the question of what to do when someone goes AWOL, which seems to happen every other paper. If we’re collectively unhappy with the length of time this process takes, collectively stepping up to be more responsible as reviewers would help a lot.