It's all just political

Jun 01 2015 Published by under #FWDAOTI, NIH, NIH Careerism

First of all, if you don't understand that anything featuring groups of humans is in the broader sense "political" than you are a fool.

The typical charge that NIH grant review is "all political" made by disappointed applicants, however, always sounds a little more...specific. Take this guy:

Nice and truthy. But what does it mean?

As you might expect I set about trying to get home slice here to define terms and be more specific about what "politics" there are that are making the decision among grant applications which survive triage. Naturally he started dodging and weaving and refused to define what he meant by "politics" save for

which is ridiculous. Yes, big scale stuff like this involves a lot of real actual political behavior. But this has very little to do with the round-by-round review of grants in study sections. In fact, the Brain Initiative folks launched their political effort precisely because they were not enjoying the success they thought they deserved in the usual NIH grant review process!

The closest our friend came to honesty was

which is nice and wishy washy as a definition. Obviously it means that he has decided that the people who he thinks should not get funded do win NIH grants. Since he has determined, in his wisdom, that it is unjustified that they are funded then clearly it is because of "undue personal influence".

It cannot possibly be that the many players in the system come with their own unique constellation of beliefs about what constitutes the most-meritorious proposals, see? It has to be politics and undue personal influence.

And this is such an important factor in deciding what gets funded out of the 40-50% of proposals that do not get triaged, that he is suggesting wholesale revision of the process to award below the triage line via lottery.

I find this laughable. Yes, there is a great deal of randomness as far as which grants get selected for funding in a given round. I continue to believe, however, that non-random factors are important and that over the entirety of NIH grant selection, the 5%ile grant is likely to be selected over the 45%ile grant for nonpolitical reasons. We may not agree individually with all of these reasons, but I think dismissal of it all being "undue personal influence" is wrong. YHN is a prickly and unfriendly customer in real life and yet is funded. I know of many really friendly and awesome scientists who struggle to get NIH funding. Time after time on study section I hear the crappy application from the highly successful PI being lauded on the basis of past accomplishments and never once on the personal influence. The vast majority of the time, people are reviewing grants from people they don't really even know personally.

I remain confused as to what this charge of "politics" really means, if it is anything other than personal disgruntlement. But I am eager to learn.

So by all means, Dear Reader, have at it.

What does it mean to you to say grant review is "political"? Be specific in your terms. How could we reduce undue influence? What changes to regular old unsolicited grant review should be made to combat this truthy sounding boogeyman?

59 responses so far

  • SidVic says:

    I have heard a couple of blatant quid pro quo- you take care of my grant and i will look after your application type of stories lately. SROs are implicated in the shenanigans. Given the level of desperation out there; i wouldn't be surprised if significant corruption is creeping into the system.

    This is not the good ol boys type of corruption "i've known Bob for 20 twenty years and boy does he do good work.. " More overt.

  • drugmonkey says:

    SidVic- I have also heard one verified (the dumasse was stupid enough to email it) attempt at this. He picked the wrong person to try this on but yes, it gives one pause that perhaps this type of slippery fish does this habitually and because he thinks that it works. maybe it does. maybe it does.

  • drugmonkey says:

    the good ol boys type of corruption "i've known Bob for 20 twenty years and boy does he do good work.. "

    I don't see how this is corruption. We have a very vigorous set of arguments at present that NIH should shift to more explicitly fund people over projects. Your "corruption" is just the subrosa version of this that has been going on for a long time. Many, many people in the system view this as a feature and not a bug.

    I tend to disagree but I would not attempt to dismiss the people-over-projects fans as illegitimate.

  • physioprof says:

    When people like your interlocutor say "undue personal influence", what they almost always mean is "more prestigious in their field than me". It's an intentional conflation, because they know that genuinely inappropriate personal influence--"I will destroy your career if you don't give my grant an awesome score"--is exceedingly vanishingly rare. And what they probably are in denial about is that the system of at least three assigned reviewers and final vote of an entire panel completely eliminates any possible effect of real "undue personal influence" if it were to occur. This is because--as anyone who has served on study section knows--reviewers take it *very* seriously if they think another reviewer is trying to pull the wool over their eyes, as a matter of intellectual integrity and scientific rigor. So for real "undue personal influence" to actually move the needle in the positive direction and pull up a grant that is unworthy on some metric of "scientific quality" would require a real and exceedingly unlikely conspiracy.

    I know we flog this fucken horse to death, but it would do wonders for the credibility of NIH peer review with junior PIs if they were given once again the opportunity to start serving on panels as soon as they take the reins of their new labs (as was the case back in the day).

  • drugmonkey says:

    And not so junior PIs as well.

    Nothing like being asked to make the sausage for dismantling paranoid conspiracy thinking about relative impacts of various real and suspected factors.

  • Dr Becca says:

    I served on study section for the first time this year, and it was quite the eye-opener. I found everyone on the panel to take the whole process very seriously. They were 100% respectful of each other's expertise and I witnessed no egos or anything otherwise suspect in terms of the integrity of the reviews.

    That said, there was a HUGE amount of randomness in the focus of discussion from grant to grant that I found both surprising and demoralizing. There were simply too many grants to discuss and not enough time to discuss them. The issues that were brought up completely depended on who the reviewers happened to be, and so two grants that may have had the same "flaw" could have been penalized or not. I got the sense that most reviewers had not had time to look at grants that weren't part of their assignment load, and so the panel's impact score votes mostly just averaged the primary reviewers' range.

    Finally, one of the biggest things that surprised me was how independent each discussion was. I think there's a sense that in study section everyone is able to compare each grant against the others and that there are little things that will explicitly put one grant ahead of another. It's possible that happens on a different admin level, but no one in study section was saying "these two grants are equally good but this PI just had a Nature Neuro paper and this one's best journal is J Neuroscience so lets' rank the 1st one higher." For the most part, we discussed a grant, scored it, and moved on.

    YMMV, but those were some things that stood out to me.

  • Ola says:

    This DM. All of this.

    I too am a rather "prickly" person who has rubbed a LOT of people up the wrong way during the course of my relatively short career to date (hence using a pseud here). Hell, a few years ago I was on the verge of losing my job as a result of such behavior. Deespite everyone in the field knowing that I'm a royal pain in the ass, I've still managed to remain funded. Naturally, as soon as I stop getting funding then I have a ready-made excuse - it must be all those people I pissed off!

    Unless you've actually done something to annoy the powers that be, it's unlikely they even know you exist, let alone how to use "politics" to hurt you. So go ahead, whine about it in public - maybe that's the kick the bee's nest needs? Also be sure to ask Steve McKnight how that strategy is working out for him.

  • SidVic says:

    DM- one case, one verified (the dumasse was stupid enough to email it)

    Yeah, well it was verified because the idiot emailed it, a clever guy would never commit such to writing. As with the scientific fraud, the dummies don't worry me, it is the clever amoralist that concern me. I have long maintained that you would have to be an idiot to get caught reusing a blot or doing what the guys that get caught do.

    As far as the good ol boy network- i think i agree with you; i consider it a not damning byproduct of human interaction. My point is i worry that it has begun to change into something that one could characterize as a overt corruption of the system.

  • physioprof says:

    That said, there was a HUGE amount of randomness in the focus of discussion from grant to grant that I found both surprising and demoralizing. There were simply too many grants to discuss and not enough time to discuss them. The issues that were brought up completely depended on who the reviewers happened to be, and so two grants that may have had the same "flaw" could have been penalized or not. I got the sense that most reviewers had not had time to look at grants that weren't part of their assignment load, and so the panel's impact score votes mostly just averaged the primary reviewers' range.

    A good chair (and other conscientious members of the panel) should be trying to nudge the panel towards consistency from grant to grant.

    Finally, one of the biggest things that surprised me was how independent each discussion was. I think there's a sense that in study section everyone is able to compare each grant against the others and that there are little things that will explicitly put one grant ahead of another. It's possible that happens on a different admin level, but no one in study section was saying "these two grants are equally good but this PI just had a Nature Neuro paper and this one's best journal is J Neuroscience so lets' rank the 1st one higher." For the most part, we discussed a grant, scored it, and moved on.

    You supposedly aren't allowed to explicitly compare one specific grant to another. But you definitely can get around this by discussing "consistency". It sounds like your study section is lacking somewhat in the leadership department, both by the chair and by other experienced members of the panel. On my study section, I would say we spend between 1/4 and 1/3 of our discussion time on such "meta" issues like consistency, fairness, spreading of scores, etc.

  • drugmonkey says:

    It is often the case that the "consistency" and relative comparison only focuses on a subset of issues for any one grant. Perhaps that is due to the specific area of disagreement of the reviewers. Perhaps that is due to the particulars of the application. I would argue that you can't necessarily complain about inconsistency when the discussion about Grants A, B and C is over the hypothesis testing and discussion of Grants X, Y and Z is over the impact of the PI's previous publication output. it may be that Grant Z is awesomely tight on the factors being debated in A, B and C discussions but for whatever reasons it doesn't come up and does not give Grant Z any sort of boost. Conversely it could be the case that Grant B rocks the JIF but Grants A and C have pedestrian publication outputs compared with Grants X, Y and Z but again, this particular facet doesn't happen to come up in the A-B-C discussions.

    Perhaps because of sub-topic domain. Perhaps because the reviewers assigned don't give a crap about JIF on the one hand or tight, hypothesis-testing-bodacious, alternate-considering research plans on the other.

    This can look like maddening inconsistency.

    I say it is part of the diversity-strength that keeps all the grants of the NIH from going to curing mouse cancer in cell-line faked up bullshit assays.

  • Dr Becca says:

    My point is that there is so much handwringing here and elsewhere about absolute minutiae of granstmanship and CVs and whatnot, much of which is based on the hypothetical "all things being equal except this one extra piece of preliminary data/co-first author/Ivy League institution/BSD mentor etc," X gets the grant and Y doesn't. My impression was that the variance in priority scores is bigger than any one of those things. There is no single tie-breaker.

    PP, you may be right about the panel management, but in this instance there were almost 40 of us, with over 50 grants on the agenda for discussion (another 50 triaged), and I think the SRO and chair did what they had to do to keep things on schedule. I have no idea how we could have built in another 4-5 hrs for "meta" discussion.

  • physioprof says:

    One thing that has really opened up time for important "meta" discussion in my panel is using the "read phase" system, where there is a few minute silent period for each grant where everyone can read the specific aims page and look at the biosketch. This *really* helps prevent the ten-plus minute sub-aim-by-sub-aim pointless descriptions of the grants by the reviewers. We are really strict on "limit your descriptions to key strengths and weaknesses that drive your scores". If any reviewer starts out with, "There are three aims. In aim one...", we shut them down *immediately* and tell them to focus solely on strengths and weaknesses. This leaves time for the really important "meta" discussions. And also, to be fair, our SRO has let us know that we do spend more time than average at CSR per grant, but we triage a relatively higher percentage as well.

  • physioprof says:

    Oh, and as far as handwringing about minutiae, I think you will find that DM and I are focused not on minutiae, but on the single most important thing you can do to make your grants fare well: do everything in your power to make the reviewer's job easy, and for the path of least resistance to be to give you a kick-ass score.

  • drugmonkey says:

    there is so much handwringing here and elsewhere about absolute minutiae of granstmanship and CVs and whatnot

    It is very understandable that when you read a summary statement and there seems to be a single issue that killed your grant score to assume that everything else was awesome and but for this minutia you would have been funded.

    A lot of times that is not really the case. There can be other flaws that were also concerning and added up to a distinct lack of reviewer enthusiasm *even if* the summary statement comments are not quite as pointed about those other flaws.

    It can be reviewer short-cut behavior to focus overly on the one major flaw (as that reviewer sees it) and to not really go through point by point about all of the other problems. Especially when the one issue justifies triage all by itself (again, in the eyes of that reviewer). This may make it appear that the reviewer is focusing on minutia when in point of fact your entire package just isn't doing the job.

    I will also take a little bit of blame for often focusing my blog entries on one particular aspect of the grant application at a time. This can make it seem more important than it deserves, I suppose.

  • Philapodia says:

    Grants are in essence sales documents. You have to get the customer (i.e. reviewer) emotionally invested and excited about your product (project) so they want to see more of it. Spending the time to make it a pleasure to read will help smooth over some rough spots.

  • Busy says:

    In my field a subgroup took control of the purse strings and funded proposals mostly within their subfield, for about a dozen years. Eventually that subfield went out of favor due to lack of major results and granting bodies have regained more balance. The subfield clique still controls some major journals though, again with preference given to their own pet subject.

    I would call that political.

  • drugmonkey says:

    And people inside that supposed clique call it merit. How do you distinguish for the disinterested audience?

  • I know this is a bit off topic, but I'm often fascinated by the differences in review at NIH and NSF. Every NSF Bio panel reserves time at the end to revisit the ranking and discuss whether panelists are still in agreement with the meta-ranking. This often causes a handful of proposals to be adjusted from one category to another and is a last ditch opportunity for people to advocate for or against specific proposals in light of the full discussion. I've often found that time incredibly valuable.

  • Grumble says:

    "This can look like maddening inconsistency.

    I say it is part of the diversity-strength that keeps all the grants of the NIH from going to curing mouse cancer in cell-line faked up bullshit assays."

    I agree with this sentiment. But consider this: you could get the same inconsistency, and the same diversity, by putting all the grants on a dartboard and throwing darts. And you would do it without wasting the time of applicants, reviewers, SROs, and POs.

  • Not a member says:

    the good ol boys type of corruption "i've known Bob for 20 twenty years and boy does he do good work.. "

    I don't see how this is corruption. We have a very vigorous set of arguments at present that NIH should shift to more explicitly fund people over projects.

    There is a profound difference between Bob has an excellent project/ cv (if you want to be personal) and Bob is an old friend/ fun at meetings so I suppose he'll do a good job. The latter (=Old Boys Club) excludes diversity (women, minorities) that are not members of the The Club. I strongly oppose using membership in The Club as a criterion for funding. (Obviously I'm not a member).

  • another young FSP says:

    A lot of times that is not really the case. There can be other flaws that were also concerning and added up to a distinct lack of reviewer enthusiasm *even if* the summary statement comments are not quite as pointed about those other flaws.

    This. Especially if the reviewers do not go back to edit their comments after discussion, particularly if that discussion changed their score.

  • drugmonkey says:

    Nobody explicitly deploys " Bob is an old friend". If people think this is what happens on study section....phew.

  • BWJones says:

    I don't think you've been entirely fair in your portrayal of our interactions here. I invoked "political" as one example of influence in the context of grants that can happen when the decisions on which grant gets funded is increasingly in the statistical noise. In fact, the study your quoted and sent the link to me on supported this by concluding that small influences have large effects in this environment.
    http://blog.research.chop.edu/simulation-of-peer-review-sheds-light-on-bias-in-grant-funding/

    You clearly object to the term "political" and that is fine, but come on now.

  • potnia theron says:

    It is far, far more subtle than that. "Oldfriends" get sniffed at.

  • drugmonkey says:

    BWJones-
    What I object to is deploying "political" for truthy attack purposes and then refusing to clearly describe what you mean by this charge. And yeah, you did try to imply it was a huge factor even if you backed down to a "one of many" position later in.

  • BWJones says:

    Nobody was refusing to clearly describe anything. Its Twitter. You have 140 characters and asynchronous conversations.

  • drugmonkey says:

    Huh. Gee, I wonder why I took it to the blog form?

  • BWJones says:

    I offered to take the discussion to email or a direct conversation, so letting me know you were taking it to your blog might have been nice.

    In the interest of transparency and all...

  • BWJones says:

    Also... Nobody was "attacking" anybody. We are all in the funding crisis together, right?

    My point was simply that when funding decisions become this tight, small influences matter.

    I was talking with a friend of mine who recently got off study section and was complaining that he wanted to go home and drink afterwards. I asked why and he noted that every grant they triaged was better than anything he had written. Its a super tough climate and I understood this was discussion.

  • drugmonkey says:

    Really dude?

    At any rate, care to expand on what "political" means now that you have rom to stretch your legs out?

  • BWJones says:

    Really. Honest and for true.

    What is the deal with that chip on your shoulder here? You act as if someone is attacking you.

    My statement on Twitter was that it would be interesting to have a lottery after the triage step as that would reduce the impact of small influential effects like "politics". As I've stated in the tweets and here, my invocation of the word "political", was made as one example of influence that could sway opinion on who gets funded. Given the paper you cited, http://blog.research.chop.edu/simulation-of-peer-review-sheds-light-on-bias-in-grant-funding/ small effects have big impact in situations such as tight funding where decisions on what gets funded are in the noise which I thought was supporting what we both were saying.

    So, I am confused as to what the beef is here.

  • Dave says:

    This is an interesting and important topic I would gladly participate in and learn from. However, I cannot as you chose to once again use a potentially serious discussion to shame and ridicule a person with under-the-belt jabs. Calling someone a fool and dishonest and his arguments ridiculous and laughable are NOT a legitimate starting point for a serious discussion about his or your opinions. While this type of behavior is your bread and butter, I am very disappointed to see seemingly respectable scientists engaging the discussion as if one of their colleagues was not just publicly humiliated. And this is putting aside the evident hypocrisy, in that suddenly EVERYBODY agrees there are no politics involved in the scientific process, politics as in this paper was accepted because of the name at the top and not the science. Yeah right. Bunch of chickenshits. That's scientists and power structures for you.

  • Dr Becca says:

    the single most important thing you can do to make your grants fare well: do everything in your power to make the reviewer's job easy, and for the path of least resistance to be to give you a kick-ass score.

    LOL if only this were a "single" thing.

  • LOL if only this were a "single" thing.

    It's a single thing with a lot of sub-parts!

  • Grumble says:

    "politics as in this paper was accepted because of the name at the top and not the science"

    Why is that necessarily politics? When I get papers from BSDs to review, I tend to be hypercritical because they always make these big claims and I want to make sure that they are well supported, so that readers don't sit there scratching their heads saying "how did this piece of crap get into this journal?" ...

    Wait a minute. Why is it that I never get those papers to review?

  • profduder says:

    Maybe I'm old and nieve, but the real politics occurred when the study sections got reorganized a while back. I remember some scientific societies working hard at getting an additional study section devoted to their area.

  • Dave says:

    This is escalating nicely. Subscribed.

  • Juan Lopez says:

    "I am very disappointed to see seemingly respectable scientists engaging the discussion as if one of their colleagues was not just publicly humiliated. ". Come on, have you seen CPP's comments? Go back and read what he wrote about one of my comments a few days ago in the discussion about invited talks in the CV. Nobody was humiliated. Nobody was insulted.

    About the politics: I have heard complaints that my field is an Old boys club, and that your papers and grants will be rejected if you don't have the right pedigree. That's one option. The other is that anyone thinks they can come and teach us how to REALLY do science. So they write a paper ignoring the last 20 years of discussion, and submit their "hypothesis". They get slammed and conclude that it's because of pedigree and politics. When true new ideas come, they have done very well.

    Is this politics?

  • Busy says:

    >> And people inside that supposed clique call it merit. How do you distinguish for the disinterested audience?

    You ask the disinterested audience for their neutral opinion.

  • Lurker says:

    "I know we flog this fucken horse to death, but it would do wonders for the credibility of NIH peer review with junior PIs if they were given once again the opportunity to start serving on panels as soon as they take the reins of their new labs (as was the case back in the day)."

    PhysioProf, what about:
    http://public.csr.nih.gov/ReviewerResources/BecomeAReviewer/ECR/Pages/default.aspx

    isn't that what the program is designed for?

  • drugmonkey says:

    It is too limited.

    Assistant Profs never topped more than 10% of reviewers (not reviews, reviewers). Even that wasn't enough but Scarpa purged them.

  • physioprof says:

    ECR is *way* too limited in scope.

  • physioprof says:

    I just looked at my long-form CV, and within four years of starting my lab, I had served on four different review panels: two SEPs and two regular study sections.

  • drugmonkey says:

    My first service on study section was at 4.5 yrs into my appointment (6 assignments) and I was empaneled by the end of year 5.

    I do think Assistant Professors need to think carefully about taking on a permanent role on a study section. It is a lot of work, particularly with the old style where you were expected to do all three meetings per year for four years.

    I would like to see occasional ad hoc service earlier than I experienced it, however. Ideally, biased for traditional, in-person meetings too. The various online forums and conference call variety do not do the trick.

    From the perspective of the review system itself, I think it would be great* to get postdocs involved. Maybe the K99 awardee population would be the place to experiment with this although even this compromises* the main goal of this proposal.

    *What I mean is that the system is inherently self-reinforcing and conservative because reviews are conducted by those who have been successful in the system as it is. Therefore, either by selection of like-minds or by beating people into the mold, the system reinforces itself.

  • NewbiePO says:

    Aside from ECR which is a CSR program, some ICs also recruit early investigators for their in-house panels, which is another way to put in ad hoc service on study sections. These panels often don't require the same breadth of expertise as would be expected on a standing study section, and may be more likely to recruit early investigators. I suggest contacting SROs at your ICs with your CV so they have you on file.

  • physioprof says:

    Aside from ECR which is a CSR program, some ICs also recruit early investigators for their in-house panels, which is another way to put in ad hoc service on study sections. These panels often don't require the same breadth of expertise as would be expected on a standing study section, and may be more likely to recruit early investigators. I suggest contacting SROs at your ICs with your CV so they have you on file.

    One of my earliest study section experiences was on one of these. This one was actually broader in subject matter than most CSR study sections of relevance to the IC.

  • Dave says:

    I frequently take part in local reviews of post-doc/grad student applications as part of the NIH CTSA program. These are run exactly like NIH SS and have very broad topic areas. It's been very important, and is as close to the real NIH experience as I have gotten so far.

    I genuinely have no idea why it is not policy to have a mix of rank in every SS. I mean, I know all the arguments for and against this, but still.......

  • drugmonkey says:

    It's because senior investigators who are mad about grant review pitch a fit about how all those Asst Profs on study section killed their grant.

  • Pinko Punko says:

    Homo sapiens riffus raffus

  • jmz4gtu says:

    I really think they should start letting some postdocs do review of grants and papers for technical feasibility to start, but also let them come down and sit in on a study section once a year. The PD reviewers could even be invited/screened by the study section, and could help lighten the workload of the section itself.
    After all, there are a *ton* of us, and we're always looking for an excuse to step away from the bench and get some of this vaunted "training" we're always hearing about.

  • Juan Lopez says:

    Jmz4gtu, this is training only in the sense of "do it". There really is no mentoring. They give you a bunch of pdf's with instructions, and speeches, but everybody acts like it's all old news and you should know this already. It was clear to me that if I had questions, then I was not ready and I belonged on the triaged pile myself.
    Completely worth doing, but calling it training is generous.

  • Juan Lopez says:

    Actually, I want to ask: anyone has seen it different? Has anyone actually been part of the reviews and get good mentoring while doing it?

  • physioprof says:

    Dude, you've never heard of "learning by doing"? When you said stupid shit, didn't other people on the study section call you out?

  • physioprof says:

    You didn't have dinner with the study section and discuss without reference to specific grant applications what you were doing and how it all works??

  • Juan Lopez says:

    Mentoring is more than calling someone out when they do something stupid. It is necessary, but sure not enough. There was also the "do as I do", which was awesome.

    Dude, ha.

  • AcademicLurker says:

    @Juan Lopez,

    My experience also was "just do it". Going to dinner with members the evening before the meeting gives some opportunity to pick things up by listening to the old hands conversing.

    I sort of eased into reviewing by serving on an AHA study section starting in year 2 of my tt appointment, so I had a few years experience when I finally starting reviewing for NIH.

  • postdoc says:

    The lottery notion raises an interesting possible fix to the problem of having more fund-able applications than funds to distribute. What if there were a lottery for who gets to write grant applications and apply on any given funding cycle? The total number of funded PIs would be the same, but the hours spent writing any one grant application would be way more likely to result in research funding, and PIs would spend a larger portion of their time writing papers and analyzing data, and a smaller portion writing funding applications. Society would benefit because way more science would get done per funding dollar. Of course, there still has to be some sort of evaluation for who gets to enter the lottery - perhaps a combination of how long that PI has been waiting for a chance to apply, how their previous application ranked, and some assessment of their publication productivity, adjusted for the norms of their sub-sub-specialty.

  • drugmonkey says:

    once you open the door to "some sort of evaluation for who gets to enter" you have undercut the entire value of your premise. The fact you feel it necessary to add this, and are unwilling to commit to a truly random process, should be enough to show you that the proposal is on the face of it absurd.

  • […] aspect of human endeavor that involves teaching newcomers how to do something involves both didactic and practical […]

Leave a Reply