Benajmin Jones, of the Kellogg School of Management at Northwestern University writes about the science funding cuts:
The real challenge is that we do not know what to cut. Unless we acquire a deeper understanding of the ‘science of science’, it is hard to deploy limited resources for their highest return. We need data — rigorous empirical evidence born in experimentation. We need to turn the scientific method on science institutions themselves.
Funding institutions should identify operational features that they are unsure about and then experiment with change. For instance, some programmes can be put into ‘treatment’ groups, while keeping others in a status quo ‘control’ group. There are numerous ‘operational experiments’ from which we could learn and improve science programmes. As just one example, take winners of grants from the US National Institutes of Health. A subset of these beneficiaries could be randomly selected to receive 10% less funding (treatment group 1) and then grants could be awarded to extra projects that scored just below the funding line (treatment group 2). By tracking project outcomes over time, we could determine the causative effects of both dollars and grant numbers on the progress of science, thus informing a better balance between grant size and grant number for future programming.
Crisis can breed opportunity. The opportunity here is to learn how to improve the use of science funding. If we take this moment to experiment with the science of science, a 5% cut could ultimately produce substantial gains.
If I remember correctly (paging DrugMonkey), the data suggest that the tail end of the funding curve doesn’t really differ in output (as measure by papers). Also, I think some program officers are unofficially conducting Jones’ experiment anyway.
It’s an interesting idea, but I still would rather have the money. I like that form of opportunity much better.