FSP, we need to kick this arrant twaddle to the curb, not give it a fair hearing!

December 19, 2008

Female Science Professor has been posting on a specific area of career mentoring frequently expressed to new faculty as “don’t get to big for your britches, junior!”. It is a familiar theme of internal departmental advice and a not-infrequent StockCritique of NIH grant review as well. In her posts, FSP has been promulgating this sort of harmful, fake helpful career “advice” that makes my blood absolutely boil. In not just one but two consecutive posts!
Even the doyenne of prof-blogging needs to be taken to the woodshed now and again…


Interestingly, FSP muses on how she was ticked off by getting this type of advice as a junior faculty member but then turns right around and peddles the same garbaggio.

I can see how it would be bad advice to say to an early career person: Sure, go ahead, write 3 or 5 grant proposals this year and take on 8 grad students, sink or swim, good luck. And it is perhaps rare for a mentor (a program officer or a faculty colleague) to know enough about their mentee to gauge how much of a research program they can reasonably manage and sustain in the first few years of their career.
Perhaps it is therefore better to err on the side of caution and advise a gradual building of a research program.

Wrong, wrong, WRONG! It is the worst sort of paternalistic nonsense to look out for the junior investigator in this way. The fact is you cannot know how much s/he can handle. The junior investigators themselves don’t know how much they can handle until they get overloaded, now can they? And the worst part of this is a universal prescription when the capacity of junior investigators (not to mention competing responsibilities) vary tremendously.
This nonsense comes up in NIH grant review, most often covered over with the StockCritique of “too ambitious” but also snuck in under “This extremely well funded junior investigator”, “high level of funding” or some such. There are also more-or-less explicit comments made at study section to the effect that the junior PI who has been storming the section with proposals and finally gotten her R01 funded needs to “settle down” and “work on that project for awhile”. This is maddening and wrong. I had a post up awhile ago entitled “New Investigator, don’t cut yourself off at the knees” but clearly I need to keep emphasizing the point.
Junior investigators need to grow their labs as big as possible as fast as possible to succeed.
This is a fundamental truth I have understood since about the middle of my last postdoc stop. Look around at the mid to senior people in your field with whom you compete or plan to compete. What size are their groups? How many techs, grad students and postdocs have they sustained over the past decade or two? How many grant $$ did that take? That is your target. That is the demonstrated basic requirement to pull off the kind of science you want to do so you had better get there as quickly as you possibly can, no? At the root of everything is money so keep those proposals rolling out.
The flip side of looking at the successful people is to look at what happens to those investigators who have marginal funding from the beginning. I have seen them operate more under obligation to BigCheez and the crumbs that drop from the BigMech. This is bad for independence and wastes time (relatively speaking) working on stuff for which someone else is going to get the majority of the credit, no matter who’s work it is.
I have seen underfunded junior PIs struggling to get anything done on their own core interests because they can’t afford a full time tech, full time postdoc and their own salary percentage while still having enough money for supplies. It is VERY hard to get all the pieces that are necessary from one R01. I developed my appreciation in the era of FIRST (R29) mechanisms in which it was essentially expected that new PIs would get this crippled ($100K direct per year, no more that $350K over 5 yrs) starter award. Maybe after about year three the study sections would permit you to be considered for an R01….gak. You can’t do anything for this kind of money. You are begging for training grant slots to support a postdoc. Postdocs are already hard to recruit because of your newbie status but when you have to gate on US citizenship? Cannot say “yes the money is sitting in my grant award and I can pay you tomorrow” but have to screw around with the annual start date of the training grant? It makes it hard. I’ve seen this happen to several investigators who later went into an exponential growth phase later down the road, thereby proving they were worthy and capable and all that.
What’s the risk?
Tied up in the advice to slow down, not be too ambitious and to stay humble is the implicit or explicit concept that there is some risk. That somehow the junior investigator is going to be worse off with two R01 grants than she would have been with one. From the grant review perspective, there is some implication that the second grant will be wasted in some poorly specified manner. Show me the evidence. I’ve yet to see a junior investigator fail out of science because she had too many R01s of which she was the PI. I’m not saying it has never happened, just I’ve never seen it. I have seen people give it up and go to industry or other jobs because it was too hard to get the lab up to a sustainable size whilst wasting effort on collaborative projects early on.
Too “distracted” to focus on a defined area of expertise that is necessary come time for tenure? That one’s a bit more arguable. But you are damned if you do, damned if you don’t. One single underfunded project may result in papers only on one closely related topic- but they are going to be few and/or of lower profile. You put all your eggs in one basket and what if the handle breaks on you? With two R01s you at least have increased your chances of payoff in any given interval. (Ideally you want projects that interact and combine key infrastructure demands but differ in scientific application.)
From the grant reviewer perspective the idea seems to be that the junior investigator will be less scientifically productive per grant if you give them two awards. I believe this rests on a common bias when subjectively reviewing the “productivity” of mid to senior investigators. The issue comes up now and again when someone is trying to promote a clearly crap proposal on the basis of the fantastic productivity track record of Dr. Greybeard. It frequently turns out that when you put together all the awards to Dr. Greybeard and look carefully at attribution in the acknowledgments of the papers that his per grant productivity isn’t all that. I look at the record of the Dr. Greybeards and Professor Bluehairs and think to myself that what this shows is what is necessary to generate that kind of productivity. Not that they are so fabulous that they “deserve” that sort of money but they were able to express this fabulosity once they got the necessary resources in place. So by extension, if we think JuniorMint and Yun Gun have the scientific chops, we should be happy that they are trying to put in place the kind of resources that are needed for sustained excellence. Right?

No Responses Yet to “FSP, we need to kick this arrant twaddle to the curb, not give it a fair hearing!”

  1. Dave Says:

    Yes! Well said!
    The other mistake I see is people slowing down when they hit a rough financial spot. All slowing down will do is sink you into a hole from which you’re even less likely to recover. My best advisors always said: ‘Get the work done. Worry about how you’ll pay for it later.’
    It’s WAY easier to get money for accomplishments than promises.

    Like

  2. Dave Says:

    The other thing I might add is: “No dean or department head gives start up money and lab space to a junior faculty member in the hope that person will get off to a slow start. FSP’s advice will just frustrate the people in charge of your tenure. Better at tenure time to be the person burning brightly but in a temporary slow spot, than the lame junior faculty member who never really seemed to try very hard.

    Like


  3. If FSP is to be the doynenne of science blogging, I shall like to be its princess.

    Like

  4. acmegirl Says:

    DM, what you are saying makes a lot of sense. But it can be really painful to either watch or be directly involved in the building of a lab when the strategy you suggest is employed. My PI is a YunGun type, and we have ten grad students and three post docs crammed in a lab that was designed for six people. We have at least six major projects going on, and our equipment is in unconventional spaces like the undesirable windowless office space that was created by a poorly thought out partition of a larger space that was done for some group that has been re-located. At times we fear that, as a lab, we are spread too thin, and at the same time always bumping into each other. And that’s not even to mention the people from other labs that come in to make use of our equipment and expertise to do collaborative projects. For a while, now, every time PI takes on another new student, there has been plenty of grumbling. And some people have said that they think PI should be more “selective” when it comes to taking on collaborative projects.
    However, each new project required some time to get up and running and producing data, and all had some risk that they would not produce anything of interest (or anything at all). Some of the projects are beginning to bear fruit. And, as a result of increased grant income, we will likely be given a larger space, soon. So, I think it has been worth it. For all the inconveniences, I think it has also been an invaluable experience to see how a lab grows.

    Like

  5. Jim thomerson Says:

    I came on board as a new assistant professor in 1965. There were no mentors, so I had no problem with bad advice.

    Like

  6. DrugMonkey Says:

    But it can be really painful to either watch or be directly involved in the building of a lab when the strategy you suggest is employed.
    Trouble is that it is even MORE painful if you are in a lab that has no resources to hire techs, to buy necessary equipment and has random bouts of “we gotta get rid of the mouse colony” or staff layoffs…

    Like


  7. I wonder if part of the disagreement is due to the fact that FSP is in a different field. In my field, which I believe to be closer to FSP’s than yours but still not the same, most people do not have techs.

    Like

  8. jake haberson Says:

    A lot of what you are saying is field specific. A lot of prestigious senior labs in my field have have 1-2 grad students, 1 post doc (usually self-funded), some undergrads on research credit, a tech and that’s that. They get science done at a steady pace. One nice thing about the smaller lab model is that all the people in the lab including the PI can actually sit down at one table and have real conversations/collaborations. I think you’re more likely to produce grad students who will go on to great things in this setting, something which is as important to the “advancement of science” as is producing papers. In big labs it’s sometimes tempting to treat grad students/postdocs like pipette monkeys – I think we’ve all seen it.

    Like

  9. DrugMonkey Says:

    Do not get too caught up in detail. The essential point is that junior people should not be advised or reviewed with an expectation to take it slower than mid career peers.

    Like

  10. neurolover Says:

    “The essential point is that junior people should not be advised or reviewed with an expectation to take it slower than mid career peers.”
    True, but they need to look at what the mid-career peers are doing in their own field, and they need to perform at the same level as their mid-career peers.
    I can guarantee that a early investigator who loads on 2 (or 3!) RO1 grants and then falls anywhere short of the productivity expected is *not* going to get any breaks (say, in contrast to the established investigator who will be excused for a slow period. The established guy relies on the fact that he’s already published a 100 papers. The new investigator won’t have that fall back.
    The expectations of one RO1 will be less. And, those who aren’t fully on soft money have it a lot easier — funding 100% of the PI salary off of one’s grant sucks significant resources out of a project.
    But, then I think big labs, are in principle bad, and that everyone should be prevented from having them (junior and mid-level and senior labs).

    Like

  11. DrugMonkey Says:

    neurolover, this is nonsense. or, you are trying to bring about your last point about big = bad by rendering bad advice.
    look, someone who has it in for you is always going to find something to criticize. if you do get lucky with 3 R01s is someone going to say you “could have done more”? Sure. Whether this is fair or accurate in comparison with your mid career peers or not. But it is the sucker’s bet to think that you can have one grant, produce a little bit and rely on the “hey, I only had one grant” excuse down the road. Much better is having the chance to produce a lot because you had the necessary resources in hand. Hard numbers on the CV are always better than excuses such as “I only had one R21”.
    You are also neglecting to consider that appearances mean a lot in this biz and having a lot of grants at an early stage is something that impresses people. At the very least they will assume you are better than average at grant writing, even if they think your science sucks. Deans don’t always know anything about impact factors or numerical expectations in your field but they sure as heck know from indirect costs.
    there is very little downside to having that extra grant in hand. you scenario is not one of the credible downsides IMO.

    Like

  12. yolio Says:

    Little of this applies in the kind of science that I do. My science is cheap, but highly skilled. Students and post-docs usually take more in training than they give back in productivity. There is a trade-off between time spent chasing dollars and time spent getting it done. Sometimes the time is more useful than the dollars.

    Like


  13. Who needs to woodshed whom? FSP concluded:
    Even so, if a young colleague wanted to try to do more, I would not discourage them unless I had specific reason — based on their record, not on a one-size-fits-all philosophy — to counsel caution.
    She comes right out and says that too many grants for one person may be just fine for another, and that I think the advising philosophy of “You should be just like me” is probably not a good one in general.
    Furthermore, having watched a junior faculty take on three simultaneous grad students and flail pathetically at mentoring even one, I would have to say that “take it slow” is not the worst advice one could give. Sure, this faculty might have flailed with even one, but at least only one grad student’s career would have been toast, instead of three. Bigger is not always better.

    Like

  14. S. Rivlin Says:

    Although both FSP’s and DM’s points make sense, each on their own merits, focusing on the scientific career as the scientist’s most important goal, while the science takes a back sit, is, I think, the beginning of the scientific venture’s demise. There’s no bailout plan for the NIH and I fear that many federally funded PIs, especially the big ones, will collapse similarly to the collapse of the big SUVs of the American auto industry and, possibly, the industry as a whole.

    Like

  15. Max Says:

    Doing excellent science should be the scientist’s main goal. While bigger is better up to a certain point, no one is immune to the law of diminishing returns. Biting off more than you can chew and consequently producing mediocre output or failing to deliver on your promises are very real possibilities for early career scientists who, unlike their mid-career role models, do not yet have the management experience to juggle lots of different projects successfully (no one starts off with this skill right off the bat, rather it is gradually acquired, note the word ‘gradually’). Nor do early career scientists have the big name to recruit the top notch students and postdocs to do your research for you with minimal effort on your part (and if you have a lot of projects going on you will be spending minimal effort on any one of them), nor the cushion of a long list of past accomplishments to back you up to offset any lull in productivity. Taking on too much and consequently failing to deliver early in one’s career, can lead to inability to get future funding in the long run.

    Like

  16. SurgPA Says:

    “focusing on the scientific career as the scientist’s most important goal, while the science takes a back sit, is, I think, the beginning of the scientific venture’s demise…There’s no bailout plan for the NIH and I fear that many federally funded PIs, especially the big ones, will collapse similarly to the collapse of the big SUVs”
    I don’t think DM is arguing in favor of career-building for it’s own sake (ie for personal-professional prestige/ego-massage), but rather that the ability to Do Science for the long term is dependent on having a financially stable platform on which to do the work. If the prejudices of federal funding favor prior proven success as DM argues(ie bigger labs with more grants) then the ability to survive, especially during a financial downturn, depends on becoming one whom the systems prejudices favor. One might argue that the system shouldn’t have those biases, but that’s little solice to the new PI trying to establish him/herself.
    Yes, federal funding is only going to get tighter, it doesn’t sound like that will preferentially hurt the big labs. The size analogy to Big-3-Auto doesn’t work for me Their collapse has many contributor that have nothing to do with their size (years of inferior quality harming their reputations, reliance on short-term profits from gas-guzzlers rather than long-term forecasting and planning, labor/union/healthcare/pension costs that exceed their competitors to name a few.) Their business plan would put them in financial difficulty no matter what their scale; their size only factor in publicizing their collapse due to the potential financial impact of their failure on the entire economy.

    Like

  17. plato Says:

    I’m a little baffled by this whole debate. Imagine a young asst. prof comes into your office and says “Hi, I’m trying to figure out how many grants/projects I should apply for as I’m growing my lab.”
    Shouldn’t the answer always be “whatever you are comfortable with, but getting at least one grant is vital to your future success?” i.e. one size doesn’t fit all? Or are we really saying that bigger always equals better? If not, then I don’t know why there’s such a strong reaction to what FSP wrote.

    Like

  18. qaz Says:

    DM –
    I have actually seen junior PIs with labs too big. The problem with junior PIs with big labs is not that the junior PIs fail. Rather, the problem is that their students get lost. That being said, I have also seen junior PIs with labs too small. Isn’t the real issue that we need to encourage junior PIs to push themselves to find the best lab set-up that suits them? To be willing to risk?
    I think the biggest problem isn’t so much the junior PI trying to write too many grants or too few. I think the problem is that junior PIs are encouraged to concentrate on getting grants instead of doing science. Junior PIs need to understand that grants are a means to an end, but they are not the end. The best mid-range labs I’ve seen have always come from people who found ways to do the science they wanted to do. Sometimes that entailed big labs with lots of money, sometimes that entailed little labs with little money. Sometimes that entailed little labs with lots of money, and sometimes (yes, it can be done) that entailed big labs with little money.
    PS. Don’t trash training grants! A well-run training grant is an opportunity for a senior PI (who is actually interested in mentoring junior PIs) to help a junior PI create an appropriate-sized lab before the junior PI has its own funding. [I am forever indebted to my senior faculty for putting my first students on training grants, which allowed me to have enough students to do some major breakthrough work before I had my first R01. Believe me, that made getting that R01 MUCH easier. I worked it out once and the cost of students on training grants was in the same ballpark as my startup. And I had a decent sized startup-package.] I wish NIH put more of its money into training grants over R01s! [Especially because training grants simply say find the best students, do the best work – rather than this “tell us what you plan to do and we’ll decide if it’s a good idea or not” crap. But that’s another rant.]

    Like

  19. leigh Says:

    i guess what i take away from the two sides of this equation is that one should without exception do as much as one is possibly capable of doing.
    you don’t want to take on so much that you neglect a graduate student or postdoc who relies upon you, but you don’t want to take on any less than your full capacity.
    the problem, how i see it, is that finding out exactly what i’m capable of usually involves taking on work until i pass the “too much” line. but then you’re usually kind of stuck, and you flail. imo flailing is a very important growth experience too, because you have to learn to put your feet back on the ground somehow, but at whose expense?

    Like


  20. Furthermore, having watched a junior faculty take on three simultaneous grad students and flail pathetically at mentoring even one, I would have to say that “take it slow” is not the worst advice one could give. Sure, this faculty might have flailed with even one, but at least only one grad student’s career would have been toast, instead of three. Bigger is not always better.

    Alright, DJMH, I’m tossing the bullshit back to you. Do you think this faculty would have been any more successful with 2 or 1 student as opposed to three?
    And, for my dear friend Sol, doing good science and being concerned about the path of one’s career are not mutually exclusive and does not signal the death of the sciencez. This is ridiculous. I think it is crucial that one considers how to structure their research program to be most productive and efficient with their time and resources

    Like

  21. Definitely anonymous Says:

    I am one of those rare PIs with too many grants too soon. I can see both sides. Of course I’m glad I’ve got the grants — only a fool wouldn’t be — and I know that it will count heavily in my favor come tenure, and obviously it is better than not having enough. But the fuck if I know how I’m going to get the work done. The awards in my field are not big enough to hire an army of techs/postdocs and I have had poor luck in attracting decent grad students for a number of reasons. I can’t help but feel my early success is going to come back and bite me hard…though hopefully post-tenure.

    Like

  22. S. Rivlin Says:

    My dear Isis, if you somehow have managed to understand that I believe that science funding and good science are mutually exclusive, I apologize. Of course, they are not. However, statements such as the following are strong indicators that science takes a back sit to career:
    You are also neglecting to consider that appearances mean a lot in this biz and having a lot of grants at an early stage is something that impresses people. At the very least they will assume you are better than average at grant writing, even if they think your science sucks. Deans don’t always know anything about impact factors or numerical expectations in your field but they sure as heck know from indirect costs.
    I do agree with most of DM’s statement above, which I believe support my contention that the focus is on the money and the career, not on the science. After all, the dean will be impressed by the bottomline ‘indirect costs’ rather than the science that generates the money. This is a sure way to prefer the rich science over the good science, and thus, the rich scientist over the good one. Under such preferences, in lean times the good scientist with the less money will collapse first along with her good science; the rich scientist may survive or will collapse along with her not so good science.

    Like

  23. whimple Says:

    But the fuck if I know how I’m going to get the work done.
    If you really do have more money than you need and are desperate to get some serious experimental talent to get some actual work done, why not consider taking a 6-month sabbatical at the bench in your own lab?

    Like

  24. drdrA Says:

    ‘If you really do have more money than you need and are desperate to get some serious experimental talent to get some actual work done, why not consider taking a 6-month sabbatical at the bench in your own lab?’
    If I had the $$, and papers continuing to go out… I’d do what you suggest in a New York minute!!

    Like


  25. I don’t think DM is arguing in favor of career-building for it’s own sake (ie for personal-professional prestige/ego-massage), but rather that the ability to Do Science for the long term is dependent on having a financially stable platform on which to do the work.

    Rivlin understands this very clearly. He is not here to engage in honest discourse. He is a petty vandal, and seeks only to disrupt and derail discussion. I urge the readers of this blog to ignore his sad attention-seeking antics, and hopefully he will get bored and find some other community to interfere with.

    Like


  26. What’s with the reading comp problems, DM? I said Sure, this faculty might have flailed with even one, but at least only one grad student’s career would have been toast, instead of three.
    Of course, that’s more advice from the grad student side of things–wait and see how the first “guinea-pig” student does with a brand-new PI before you jump in…
    Anyhow, this particular faculty would probably have sucked anyhow, BUT as one of the major student complaints was that the faculty member never spent any time in lab talking with the students about their projects, it is at least possible that a lone student would have received 3 x (not very much) attention, which is better than 1 x.
    The faculty member’s view, as I understand it, was that money was so scarce that grant-writing took priority over lab mentoring. This is understandable but one does wonder if biting off a smaller chunk (fewer students, fewer grants) would have allowed for more mentoring time.

    Like

  27. Perceval Says:

    I find it fascinating that so far no one has suggested transferable skills courses in phd supervision or project management to help people like definitely anonymous or the challenged supervisor of three cope . I have been working three days a week on three projects in postdoc and pi positions – very useful since one tends to bear fruit while the other one languishes. But impossible if you don’t actively work on those soft skills…

    Like

  28. S. Rivlin Says:

    It is comforting to notice my guarding angle, CPP, following me around on multiple blogs pasting the very same message on all of them. Yet, I am worried about his obsessiveness and its effect on his blood pressure.

    Like

  29. Definitely anonymous Says:

    A six month sabbatical at the bench? That sounds lovely. Now can you tell me how to do it? Because, my appointment is 50% teaching, and I also have service responsibilities (impossible to blow off completely). And since I’m untenured, a real sabbatical is out of the question. I am already working at the bench, in fact I am the primary worker in the lab these days. The more established commenters may correct me, but I’ve always thought that it is a very bad sign when the PI starts autoclaving pipet tips…. And project management? There are not enough hours in the day.
    But before I start shouting “the end is near!” let me say I am now fully behind DM’s original post. After making my earlier comment, I started thinking about my situation around the time when I applied for my second major grant, the one that led to my current frantic state. I was facing a fairly tough departmental review that semester, and when the second grant came through, I remember thinking “Ha! They can’t fire me now!” I think it may have saved my skin.
    So, yes, new PIs have to push themselves to find their limits. I suspect I went past mine, given what I have to work with in my department, but I would not have done anything differently. And hopefully I will land on my feet.
    Now back to the autoclave!

    Like

  30. whimple Says:

    D.A., you’re crazy. Hire an undergrad to autoclave the pipet tips. My guess is that your real problem is that you’re a micromanaging control-freak who doesn’t know how to say ‘no’. 🙂

    Like

  31. Eli Rabett Says:

    It’s called a leave without pay.
    Probably the best strategy is to write as many grants as you can before you get there and in your first year which things are being set up, using your set up funds as cost sharings and leverage. You then have years 2-4 to accumulate students and papers. You might write one new grant a year in year 3 and 4, which brings you to year 5 and renewal city. At that point you can consider going for a BIG project.

    Like


Leave a comment