Ideal Primary Data:Review Article Ratio

Jan 26 2015 Published by under Careerism, Science Publication

Odyssey is pondering review articles today. That led to a question from Dr. Becca about the ideal ratio of reviews and primary research articles.

I am not a fan of authors publishing essentially the same review in multiple journals. Nor am I a fan of the incrementally updated review published every year or two. And I am really not fond of burgeoning subfields where everyone spits out a me-too review which then outnumber the primary research articles!

So, my views on this question are likely more negative than average.

19 responses so far

  • How field specific is this? I'm not sure I could get away with primary:review of less than 5:1 or 4:1. IMO, there is no point in writing a review unless you have a reasonable expectation that it will be well-cited - that is the carrot, not an extra line on my CV. But, do some departments consider this as acceptable output?

  • (1) You should never agree to write a review unless it's for a high-impact-factor journal.

    (2) The proper ratio should be *at least* 5:1, with 10:1 more appropriate.

  • drugmonkey says:

    Given that no matter your place in the JIF ecosphere, one's review articles are guaranteed to get more cites than one's primary research reports, how do you argue the first point, PP?

  • AcademicLurker says:

    In the sub-field I worked in during my first postdoc, there was (is) a group that published about 4 review articles for every 1 research paper.

  • odyssey says:

    AL: I wonder if that's my sub-field...

  • Philapodia says:

    Isn't the point in writing reviews trying to step back and see the big picture? Seems that reviews from different authors can be useful in synthesizing data and presenting alternative points of view, which can stimulate new experiments in the field that most in a field may never think about because they're focusing on their little areas.

  • Susan says:

    How do you feel about infinity (ie no review articles)?

  • eeke says:

    I think anyone can write a review about anything. Publishing x number of papers in a given field is irrelevant - anyone with enough background can read, compile information, and write out a comprehensive review.

    That said, it boils down to what you should spend time on. If all you write is reviews and never produce any original research articles, or your reviews outnumber research articles, I would guess that you are in retirement mode, have no funding, or some combination of both.

  • Dr Becca says:

    I've written a couple of mini-reviews since starting this job, and though I kvetched while writing them, in retrospect they are all different and each served a purpose.

    The first was a much needed pub filler in a lab-startup productivity gap. Yes, most people are able to publish some remaining stuff from their postdoc to achieve this, but that wasn't my situation. Even though it's not in a fancy journal, it's been surprisingly highly cited so far. My h-index went up a point, so I'll take it.

    The second, I felt I couldn't say no to my BSD postdoc mentor. It allowed me to put some ideas out there that I think are really new and laid the foundation to what we were about to publish experimentally. It only took a few weeks, and it is honestly one of my favorite things I've ever written.

    I chaired a symposium last summer and was told that every speaker for the meeting had to submit a review to their special journal. I made my grad student and postdoc write it.

    All that said, unless Nature calls, I'm done with reviews at least until post-tenure.

  • ecologist says:

    Write papers because you have something to say. Say it. Say it clearly, and in a way that others will find interesting and useful. All this ratio calculating and strategizing is a bunch of crap.

  • drugmonkey says:

    How do you feel about infinity (ie no review articles)?

    I'm definitely okay with that.

  • jmz4 says:

    Does having at least a couple of reviews (between the 5:1 and 10:1 ratio) count positively on grant review committees (versus if they were primary research)? I've heard anecdotally that for K99 applicants, having at least one review on your CV is seen as a major bonus.

    I've no problem personally with excess reviews for a given topic.
    That being said, some reviews are terrible. If you've got Google Scholar Alerts for any of your high impact papers, you will be linked to reviews in the Swedish Journal of Immunology where they horribly misreference your paper. And there's nothing you can do about it.

    AL: we had one of those, the PI just thought they were good exercises for his students, so he would churn out a couple every year. Not the worst mentoring activity.

    eeke: "I think anyone can write a review about anything."
    -But writing a good review is very challenging.

    ecologist: "All this ratio calculating and strategizing is a bunch of crap."
    -A big part of this blog is careerism in science, hence the discussion of how various ratios of reviews are received among the peer groups. Also, it's my impression that PIs often get requests to write reviews at the behest of journals. So, its not always about writing a review when you *want* to write a review.

  • drugmonkey says:

    All this ratio calculating and strategizing is a bunch of crap.

    Yes, we all have infinite time to do all the things that would be good in our lives and careers so therefore strategy is unnecessary.

    /eyeroll

  • Dennis says:

    What else are you going to do with the introduction chapter of your thesis if you can't publish out a junky review?

  • becca says:

    Dennis- interpretive dance? science comics? blog? illustrate it to turn it into a children's book?

  • Professa says:

    I have exactly the opposite experience as this:

    "drugmonkey January 26, 2015 at 12:01 pm
    Given that no matter your place in the JIF ecosphere, one's review articles are guaranteed to get more cites than one's primary research reports, how do you argue the first point, PP?"

    I don't think my reviews even rank above the avg and are no where near the top. Do my reviews suck? I prefer to think there are just a lot to choose from in my field.

  • drugmonkey says:

    This seems very unusual to me Professa. But maybe your subfield has great respect for citing the primary lit and not so much for the reviews? I'd say Bravo, if this is the reason.

  • qaz says:

    An observation made writing a paper today - Reviews (*) serve a very useful purpose. Sometimes there is a lot of original literature to cite, and given readability limitations, sometimes citing everything makes the text unreadable, particularly given author-year citations in APA and the like formats. (Do I really need a half of a page of citations to give credit?) This is especially a problem where it's not my place or my interest to take a side in a priority fight - I just need to prove that X is true.

    Citations were originally supposed to be used just to defend your point. We know our data mean Z because all these other people have shown that X is true. Go look there if you don't believe me that X is true.

    This problem occurs because we use citation count as a proxy for impact/importance/influence/success/rewards. It would be better to have a different method of measuring scientific impact, especially given all the problems with citation counts in the first place (GlamourPreference, FieldDifferences, FamousPersonA**Kissing, etc.)

    * As we've discussed before, there is a difference between a "review" that is merely a container for lots of other literature, and a "theoretical advance", which is a new idea and a direct contribution in its own right.

Leave a Reply