Great article in Ars Technica by Ben McNeil about how government funding agencies tend to promote well-established mediocrity. Although it is mainly about NIH, it also applies to much of NSF despite NSF trying to adopt special categories of “high risk, high reward” and “transformative science.” Although GG agrees with the basic premise of the piece that incremental science is more easily funded (and the quote about proposing what you have already done is one GG heard more than 30 years ago), it isn’t so clear that NSF is starving young scientists quite to the degree of NIH (NSF is starving nearly everyone). Part of the deal is the nature of biologic research these days, which echoes the European professorial system, with a senior scientist overseeing a cadre of postdocs or junior scientists who are actually generating ideas and results who in turn oversee graduate students doing the bulk of the labor; such a system will always skew to older investigators. Earth science doesn’t see that kind of pyramid to the same degree (at least not since the demise of COCORP) and EAR at NSF funds postdoctoral and graduate fellowships as well as making Young Investigator awards. (It is worth noting too that many universities will not permit non-tenure track scientists to act as a Principal Investigator, so the NIH numbers quoted in the Ars article may be skewed for that reason, too; GG has seen a number of proposals where the PI was in fact just window dressing needed to get the proposal out of his or her university). But it would be kind of interesting to see the age breakdown of NSF grants by division….
Another statistic we wish existed: citations/NSF-funded paper vs. citations to work NSF rejected.