Monday, July 18, 2022

How do you make scientists more productive?

Tyler Cowen of Marginal Revolution asks an interesting question: 
For instance, the management of science is a radically underappreciated issue. How many managers of scientific labs receive any management training at all, even in the basics? On a scale of 1 to 10, how well are most labs or non-profit scientific ventures run? I’ve asked a number of people in the hard and biological sciences this question, and more often a laugh is the response, rather than a citation of a very specific number. I’ve never heard anyone say they are run just great.

On the other side of the market, the rest of us are failing too. In our social discourse, we have not elevated better scientific management as a social priority. This could be done in our universities, non-profits, research labs, government agencies, and of course in the private sector too. It’s not a sexy policy issue, but science is one of the most significant means for improving society. In the language of finance, you could say that science is a major source of social alpha.

You should probably read the whole post, which is here. 

I broadly think that Professor Cowen is onto something, in the sense that "since we give out all this public funding for science, shouldn't we try to improve the amount that we get over time?" That's a reasonable question, even if I think the answers are completely unknowable. 

Here's the thing - I don't think anyone knows how to get more "productivity" out of scientists at all. I know plenty of ways to get scientists to get less innovative and less productivity. Stress them out, make them subject to randomly changing short-term goals, make them write lots of reports, and you'll soon have a poorly performng scientific organization. 

More productivity? I have no idea. Give them lots of freedom (i.e. don't tell them how to do things, just what the goal is (but not too narrow!) and just enough resources (i.e. I'm not completely convinced that lavishly funded teams work better.) That's all I got. (Maybe "make them interact with people from other fields routinely.") 

Readers, your thoughts? 


  1. i'm willing to say 99% PIs didn't receive proper management training - and too often you'd hear great science minds manage their labs horribly. I guess ideally postdocs/senior grad students should manage junior grad students/undergrads as they move towards a faculty position, which Ive heard of some Chinese institutions practice this (albeit it's often lecturers/associate professors joining another big academia group to manage a subgroup there before establishing their own, and is also a drag to ones academia career?). I didn't see any of my peers actually managed a portion of their lab in grad school.

  2. The fundamental missing premise in Cowen's piece is how one defines "performance" in science (the term he uses). How do you measure it? How do you know you're improving it, on a timescale meaningful for decision-making?


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20