<p>This is so sad. The graph is frightening. They ask an interesting question:
[quote]
Although most scientists in Ph.D. and postdoctoral training aspire to become tenure track independent PIs, the fraction of people who succeed in doing so are actually the vast minority. Despite ever decreasing job prospects in the academia, PhD and postdoctoral programs are not decreasing student enrollment.</p>
<p>Should the emphasis of F32 postdoctoral awards be changed to focus on jobs for which over 80-90% of trainees have some realistic expectation of obtaining?
<p>There is no doubt that obtaining grant funding is difficult and highly competitive. I think that the graph is slightly misleading about the per cent of NIH post-doctoral researchers who will eventually obtain their own grant funding, though. As the writer notes, the average age of receiving a first NIH R01 grant is now over 40–in fact, the last that I looked, it was 42. So if a person had completed a Ph.D. in 2004, then assuming that undergrad work is completed at age 22 and allowing for 5 years to complete the Ph.D. (which would be typical for the kind of person who receives an NIH post-doctoral fellowship), the person would have been 27 in 2004. The person is still 5 years shy of the average age at award of the first NIH grant, so they won’t show up in the award data.</p>
<p>There are many companies that hire people with education in the NIH-supported areas–all of the biotech firms are interested in them. Perhaps there is a change in the fraction of people going into biotech industries who have post-doctoral experience, as opposed to coming directly from a Ph.D.</p>
<p>(Dr.)molliebatmit recently posted that she had submitted an NIH grant, and on the first go-around, it was rated in the top 14% of proposals–not high enough to be funded. She and a co-PI revised it, and wound up with a rating in the top 7% of proposals–which is enough to make the funding cut-line. Considering the population of people who submit NIH proposals, this is a really difficult screen to pass.</p>
<p>I have some perspective on this, as I frequently review F32 and other fellowship applications for NIH. The applicants who get fundable scores are extremely strong. I imagine that the depressing statistics illustrated on the article just reflect the increased difficulty of obtaining any NIH funding, for anyone, in recent years. And it is much easier for established investigators. The payline for R21s and R01s is under 10% in most institutes. And, as QuantMech wrote, it’s a very competitive group.</p>
<p>I am incredibly impressed that molliebatmit was able to get her grant funded! Congratulations, mollie!</p>
<p>Happier to hear about Mollie’s good news. Dh’s last postdoc took a job in industry because of the grant situation. I can’t begin to tell you how stressful the current situation is, top 10% used to be a good enough score. Not any more. It’s horrible at institutions like med schools where they expect you to get your entire salary from grants.</p>
<p>Sigh. I read recently how more and more American scientists are considering working overseas because money and opportunities are beginning to dry up here. More and more, America is beginning to lose its status as a world power - economically, educationally, culturally. In one generation, we’ve gone from a science and research magnet for the planet, to seeing brain drain. Very depressing. </p>
<p>Raising my hand here, mathmom. I have had uninterrupted R01 funding for 20 years. My last few grant applications have not been funded. They’ve been in the top 20%, top 25%, etc., but nowhere near the payline. </p>
<p>I received a F32 14 years ago and then I left academia and worked for a small biotech company. I have seen many of my former colleagues suffered in the last 5-10 years when R01 is getting more and more difficult to be funded and the research budget has been cut several times. Not only the new faculties are hard to get a grant, the established labs and research centers are all having the same problem too. I was getting sick and tired of grant cycles and decided to go to industry. Ironically, I landed on a company that is running out of money and I needed to apply for SBIR grants. For SBIR, it was less affected by the NIH budget until last few years. We have been receiving SBIR non-stop for 10 years (with 1 month gap) under several projects.</p>
<p>As F32 is for people received their Ph.D. less than 5 years ago and it will fund up to 3 years, one may be on the 8th year of postdoc when the fellowship ends. Most typically it would be around 5-6 years after PhD that one would need to find other funding source or work in affiliation with an established PI. When there was more opportunities in academia, one may find a junior faculty position and receive some money to set up a lab to run for a couple years while trying to get the first R01. With all these, it pretty much add up to the age range of the first R01.</p>
<p>It is a sad day when resources for the development on hard sciences is dwindling. On the other hand, it is hard to measure the real ROI for most research. With less desire to spend liberally, it is inevitable that more discernment in prioritizing research dollars has to come. Those good ol’ days ain’t coming back! </p>
<p>The graph in the link shows the year of the F32 award on the x axis, and on the y axis it shows the % of those who received F32 awards in that year, who subsequently received R-category awards of any type. (The R-category awards typically go to independent principal investigators.)</p>
<p>The people who received F32 awards in a given year are relatively recent Ph.D. grads. They might have obtained a Ph.D. in that year, the previous year, or two years back.</p>
<p>My point is that if you figure out how old the 2004 cohort of F32 award recipients is, they are almost certainly too young to have reached the average age of first-time R-category grant recipients. So age (or really, years since Ph.D. and likelihood of having already received an R-category award at any given number of years post-Ph.D.) does factor into the decline shown in the plot–in a substantial way, in fact.</p>
<p>If you developed a similar plot that showed % receiving tenure at a university vs. year of F32 award, there would still be a precipitous drop off. This happens due to lag time, because essentially no one is offered a tenured position coming straight off the typical post-doc. It’s not a tenure crisis.</p>
<p>There is a problem, yes, but it’s not nearly as calamitous as the graph makes it seem.</p>
<p>Just wanted to restate my opinion that the current grant funding situation is pretty bad. The fact that the average age of a first-time R01 awardee is 42 should be sufficient to indicate that all by itself.</p>
<p>But the graph doesn’t do anyone a favor, by representing the situation as worse than it really is.</p>
<p>I picked 2004 because it looks as though that is a point of drop-off in the probability of having obtained an R-category grant. The analysis applies more strongly to any time more recent than 2004, for receipt of the F32 award. </p>
<p>Or, going back further in time:
Suppose that we consider a scientist who received R01 funding in 2014 for the first time, at the average age of first-time R01 grantees (42) and suppose that the person completed a Ph.D. at 27 and then received an F32 grant two years later, at 29. Since the person is 42 in 2014, he/she was 29 in 2001 (give or take, depending on the birthday month) when the F32 was awarded. So the person shows up as receiving an R* award, at 2001 on the x axis. Further, not all of those who received F32 support in 2001 are likely to have received their first R01 award yet–some people must receive first-time R01 support when they are older than the average age.</p>
<p>If you go all the way back to an F32 recipient in 1985, and suppose that the person was 29 in 1985, then in 2014, that person would be 58 (oops! bad arithmetic! I had 48 here before). By now, he/she is pretty likely to have received R* support, if he/she is ever going to. There could be a few outliers who receive their first R* awards in their 60’s, but the number of those is probably pretty small.</p>
<p>So, roughly 100% of those who received F32 support in 1985 and are ever going to receive R* support have already done so. Essentially 0% of those who received F32 support in 2014 will have received R* support by now. Those who received F32 awards in between 1985 and 2014 have a decreasing probability of having <em>already</em> received R* support, as the year of the F32 award increases. That doesn’t mean that they will never receive R* support–I’d guess that 25-30% of them probably will.</p>
<p>In the comments on the article, Jill makes an interesting point that it is not possible (within the bounds of the study) to track women who received the F32 while using their “birth” names, and who may have changed to a spouse’s name before receiving the R01. This is really too bad, because it would be interesting to see how the results play out by gender. The National Science Foundation managed to track me despite a name change–I’m not sure how they did it.</p>
<p>I genuinely deplore his rhetoric in the discussion of the “mining” analogy for scientific education. It’s too bad that he wrote it that way. However, since he is criticizing the “mining” model, perhaps he used the incendiary terms satirically. His overall point about improving science education for everyone is good. </p>
<p>Goodstein also makes a valid point point that unsustainable growth of the scientific enterprise is inherent in a model in which each university research scientist educates 15 Ph.D. students, each of whom wants to become a university research scientist, and educate another 15 Ph.D. students of his/her own. I suspect that Goodstein was really writing about physicists, and that the average number of Ph.D. graduates for a professor in the biomedical sciences is higher.</p>
<p>Goodstein claims that the “crunch” point for physics came in the 1970’s, but was not recognized at the time. Perhaps the crunch point for the biomedical sciences is just being recognized now.</p>
<p>It is for recent PhD grads, but it goes up to 5 year back (not sure that is by the time of funding started or application). I grad in 96 and apply F32 in 2000 during my second postdoc. I was funded through the beginning of my 7th year postdoc.</p>
<p>That would shift the time-frame a bit, billcsho, though the underlying analysis still holds (with appropriate modifications). The pattern of lengthy post-docs and/or multiple post-docs has become more common. I think it is a real problem.</p>
<p>The German government recognized that they had a problem with the age at which their scientists could begin truly independent research (it was too old), and created funding mechanisms to permit independent work earlier.</p>
<p>I hope that something similar will happen in the US. The length of the post-doctoral period is a problem for anyone, but I think it is particularly problematic for women scientists who wish to have children.</p>
<p>Sorry to hear that, bookworm. This is really a problem!</p>
<p>The scientific frontier is very broad, and highly valuable research can be done by a lot of people. There is overcrowding in terms of funding, but no overcrowding in terms of the scientific problems to be solved.</p>
<p>I think that the issue raised by Goodstein, the difficulty of communicating the significance of the research to people in general, is greatly reduced for biomedical science. Even though the population of biomedical scientists cannot continue to grow indefinitely, I think that we would reap great returns from permitting people who are already educated to pursue their fields, and from encouraging a reasonable number of younger scientists.</p>