Is science underfunded or overstaffed?

The bane of most academic research scientists is the decreasing chance of getting funding from agencies like the National Science Foundation. Certainly one way to look at this is to argue that basic research funding is inadequate (which is nearly entirely from the government; very little private R&D money goes to basic research). The current situation encourages grantsmanship over scholarship, probably not the best means of improving science.

We can approach this from a different direction: maybe there are too many people at the trough. Derek John de Solla Price suggested in 1965 that good scientific literature doubles in about 15 years, but the literature as a whole doubles in 5 years. If correct, we risk drowning in mediocrity. Price warned of a problem like this, attributing this to the creation of big science (science supported by substantial financial support, in a sense).  If you could make a career doing science, but not really be all that dedicated to the field, you could clutter up the literature with your brain droppings. All we need to do, it would seem, is drop the mediocre people out of the field.

Well, where are all these scientists?  There are scientists in industry, scientists in academia, scientists in government institutes, and scientists in research labs. Pure basic research in industry is fading and not growing (and, in any event, is usually not looking for handouts from government), so let’s not worry about that. Of the remaining three, arguably the one that had grown most rapidly is academia. Over the past half century universities and colleges have perceived that there is money and prestige in scientific research. Meantime, Congress has encouraged the government to spread the wealth around, most obviously through programs like EPSCoR. Schools that used to focus entirely on teaching came to first allow and then encourage funded research. So maybe we have our culprit; all we need to do is ban university academics from research money and all will be well, right?

Let’s look then and see who is doing good science.  There is no trivial metric, but arguably Nobel Prize recipients have done good science. So where are these folks? If you said universities, you are right: most are. But let’s parse things down; maybe we just want to encourage a few top notch schools to do research and ban the others. So let’s look at Nobel prizes in chemistry and physics and see if we are seeing new affiliations crop up over time. So let’s count the number of first-time U.S. institutions over time and see what we get. We’ll start in 1950 as “big science” wasn’t really present before then.

’50-4 ’55-9 ’60-4 ’65-9 ’70-4 ’75-9 ’80-4 85-9 ’90-4 ’95-9 ’00-4 ’05-9 ’10-6
New 1 1 3 1 3 2  0 3 1 4 3 2 3
Old 3 5 5 6 2 6 12 5 6 11 11 9 15
Non-U  0 3  0  0 3 1 1 5  0 1 5 5 5

It is no surprise that so many awards go to schools that previously housed prize winners (e.g., Harvard, MIT, Caltech, Stanford, U. Chicago, UC Berkeley)-many have resources to supplement or replace any government support-but the growth of institutions hosting such high-level work is what we want to notice here. 27 institutions had their first Nobel winner in physics or chemistry since 1950, 22 of them since Price suggested that science was being diluted. This is probably an underestimate because the nominations for Nobels can only come from very specific individuals, notably including past prize winners, so there is likely a bias towards past prize winners’ home institutions. (If you wonder, the numbers increase to the right as more and more prizes are split and U.S. capability and prestige increase. Some non-university (non-U) prizes are to organizations closely affiliated with a university).

This suggests that the growth of research in academia does produce good work in some of the new places.  We aren’t just cutting off mediocre science by limiting research funds to specific schools, we also are cutting off some really good stuff. So maybe this isn’t a good approach. [If you want to argue that cutting off those 27 schools would have simply pushed those researchers to the existing “good” schools, reflect on the likelihood that these people applied for open positions at one or more of those “good” schools and didn’t get the job.]

The discussion above presumes that all mediocre literature came from mediocre researchers. But maybe the problem in publication is not the scientists but encouragement to publish nearly anything. For instance, a grant that produces no papers will greatly hurt that scientist the next time she or he submits a new proposal; the pressure to put something out there is high. The lazy use of counting in evaluating faculty also encourages submission of marginal papers. We’ve discussed this before, so GG will stop ranting here and move on to, what about the money? If good researchers are in fact putting out not only good science but some dreck, we can’t simply forget folks who have ever published a less-than-stellar paper.

Arguably the way you would do this is on a case-by-case basis, looking at the proposed work and the background and history of the proposers.  And this is exactly what NSF’s peer review system is supposed to do. If it worked perfectly–meaning that the quality of a proposal was fully objective–then the outcome would be fine. Yes, only 10% or 15% of the submissions would get money, but they would be the best of the best. The mediocre people would get nothing, all the time.

But of course the outcomes are not perfect.  Program managers will fund some projects not highly ranked because of other factors; Congress will mandate some money go in certain directions or add goals independent of scientific quality (e.g., training a STEM workforce), reviewers and panelists are rarely perfectly objective, and of course the connection between quality of a proposal and significance of the resulting science is tenuous. Bold proposals that challenge ruling paradigms rarely are greeted with open arms, yet these might be the most significant proposals of all.

Maybe science is overstaffed.  Maybe science is underfunded. It is easy to argue for more money, both logistically and logically: having served on a panel, GG can say some really promising work regularly fails to get money. Lobbying Congress is straightforward if tedious.  It is hard to figure out a good strategy for pruning the scientific workforce. It is likely, as Price argued, that we are increasing the numbers of bodies at a far faster rate than increasing the cleverness or originality of the practitioners in a field. Price’s solution of small science basically means starving out all but the most dedicated; in a sense we are starting to do that. But it isn’t clear that starving roots out the worst; funding is capricious enough that you probably have to be pretty bad to never get funding, while good science (perhaps evaluated by mediocre scientists) might not get rewarded. And so we see why the usual solution to the problem is to increase the size of the fiscal pot. Maybe somebody out there will come up with a better plan…



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: