How long until my paper is published?

Some N = 1 observations on the publication process.

How long until my paper is published?

Some N = 1 observations on the publication process.

For me, one of the many (so many) sources of stress in graduate school was the fear that I wouldn’t have something published by the time the job market came around. At least in my program, there was a strong sense that no publications effectively meant no chance at a job. Now that I’m on the other side of things I know that the process is much more “idiosyncratic” than that.1

And yet the worry stands, especially since (overwhelmed, understaffed) academic journals can be slow to respond to submissions. How long should you expect to wait before finding out if a paper has made it to the Revise & Resubmit stage?

I’ve been keeping track of my own paper submissions over time2 and thought it would be interesting to visualize the lifecycle of these projects.

The Timeline

I’m using the ggalt::geom_dumbbell() geometry to visualize the trajectory of each paper.

p1 = ggplot(df, aes(x = submit, xend = decision, 
             y = paper, label = duration, 
             color = type)) + 
  geom_dumbbell(size = 1, size_x = 1) + 
  labs(x=NULL, color = 'Submission round:',
       y=NULL, 
       title="Timeline of Journal Submissions", 
       subtitle="Start date, decision date, and wait time (in days) for my papers.") +
  theme_ipsum_tw() + 
  ggrepel::geom_label_repel(nudge_y = -.5, show.legend = FALSE) + 
  theme(legend.position = 'top') + 
  NULL
p1

The first, obvious barrier to publication is getting past reviewers in the initial submission stage and to an invitation to revise and resubmit (R&R). My papers have varied a lot in clearing that initial submission hurdle. Some can take a long time to find a home; dyads has been submitted to four different journals (actually, I think I’m missing a submission here) before landing an R&R. Other papers, like hierarchy-order, landed an R&R on the first try.

Getting through that first barrier is crucial and the time to decision can vary a lot. In my experience, the median time to decision on a first submission is about 84 days, but can vary a lot: I’ve waited as long as 270 days (!!) for a response. R&R’s are a good deal faster (median = 52.5) but also vary.

df %>% 
  filter(type != "3rd revision") %>% # too few of these to bother
  ggplot(aes(y = type, x = duration)) + 
  geom_density_ridges(quantile_lines = TRUE, fill = "black",
                      quantiles = 2, color = "white", size = 1) + 
  theme_ipsum_tw(grid = "X") + 
  theme(legend.position = "none") + 
  labs(x = NULL, y = NULL, 
       title = "Waiting to Hear Back from Journals", 
       subtitle = "Number of days between submitting a paper and hearing back at each stage of publication process.")

Getting to an R&R is only part of the battle: actually revising the paper to meet reviewer comments takes time, as does waiting for the journal and reviewers to evaluate your revisions. With my first publication, worlds-apart, I received a rejection from a journal after about 45 days. I spent about a month and a half making revisions based on reviewer comments, then sent it out again to a new journal where I got an R&R after roughly another 45 days. It took me about a month to do the revisions (this was August, before start of school, plenty of free time), but then I had to wait about 4 months to get a decision back on that R&R. I spent another 20 days or so turning that around, and then the final decision took about a month.

The Tradeoffs

These timelines put pressure on graduate students that shapes their writing in different ways. Working backwards from a job market that starts in September, I would have needed to submit worlds-apart in December of the prior year to have something accepted for publication by application season. But note that this is assuming only two tries to an R&R. If I had sent it to an additional journal first and failed there, I would have to add another 2 months or so to that timeline. And getting something R&R’d in two tries is not easy! At some point, this means graduate students might need to be pushing papers out for submission earlier and earlier in graduate school, when most programs assume students are still just learning the ropes.

The other big consequence is that students face a trade-off in terms of ambition of publication venue and having pre-tenure publications. Do you send the paper to multiple top journals with miniscule acceptance rates and risk not having anything for the market, or do you send it to a lower-tier journal but miss out on a potential “big hit” that could have big implications for your career down the line?

These pressures and trade-offs are obviously all starker for students who don’t come from wealthy families or who have families of their own they need to support. Without a TT job, workshopping a paper for years to boost its chances at a top hit is a tremendous privilege.

Conclusions

My main advice to graduate students on this front would be to pursue co-author opportunities early, with other graduate students or with faculty. The best strategy probably involves developing a mix of: a) your own, high-investment, shoot-for-the-stars projects related to your dissertation, and b) co-authored, lower-investment projects that are faster out the door. I know this is much easier said than done, especially for students with lots of teaching/family/work responsibilities. The last thing I’ll say is that there’s no perfect paper, and I think there’s a risk in waiting too long to send things out.

But these are just my two cents.


  1. At this point I’ve seen folks with multiple publications, “top-3” publications, records that would put them at 2/3 or more of the way to tenure at many universities – not get jobs. Little about this process feels fair or systematic.↩︎

  2. This data is imperfect. Some of these papers have been submitted more than seen below, but I forgot to log the submission. Oh well!↩︎

Avatar
Juan F. Tellez
Assistant Professor of
Political Science

I research and teach on armed conflict, security and development, and quantitative methods.