Some Consequences of Poverty

I don’t know if the idea of poverty causing disproportionate concentration on short term thinking is new. But the exploration here is quite telling. It’s a great pleasure when papers are so clearly written that I can read the whole thing, understand something and still want more by the end, even though this subject is not one that I’m an expert in at all.

Science 2 November 2012:
Vol. 338 no. 6107 pp. 682-685

Some Consequences of Having Too Little

  1. Anuj K. Shah1,*,Sendhil Mullainathan2,Eldar Shafir3

The authors of this paper studied the attention and performance of “poor” and “rich” players in a set of tailored computer games. To be clear, the real world status of the players isn’t discussed. These people played games where the rules made them “rich” or “poor” in terms of turns, or resources. The interpretation of the results – that poor players concentrate too much on the short term, to their own long term detriment – is intuitively satisfying. The performance of “poor” contestants was much worse when they were allowed to borrow on their playing allowances, which is a great argument for banning payday loans. The overall findings are a great argument for a more generous social system, for example, less like the USA and more like Germany.  On the other hand, there are unintended consequences of more generous social provision, and it’s clear that many people choose to join the American system even when they are unwelcome – proximity within the Western hemisphere being a factor, no doubt.

My mind then turned to academics trying to get grants for research.  Less profound but still worth worrying about. It’s obvious that when success rates are small, devoting a lot of time (numerous attempts) to the process is necessary in order to get success. This applies even if all applications for a given round are equally good. The scale of the effect is quite stark, which may surprise some people. But this paper suggests that scientists who regularly compete to secure funding (because they lack it) might concentrate excessively on obtaining grants, to the exclusion of other pressing needs, beyond the net gain of actually obtaining the grant. I don’t worry about research funding too much, at least that’s my self-perception. Writing grants isn’t hell and it can be a creative exercise that brings focus. But it isn’t, of itself, a very useful activity. Another way to look at it is, what if tenured professors did original research, by their own hand, or invested months of time closely supervising a junior colleague, rather than only writing, thinking and worrying themselves to death about grants. One would expect such a strategy would lead to smaller research groups with reduced research output, but possibly would provide much better value for money (because the average quality and training of the scientist doing the work would be higher) and perhaps everyone would derive greater job satisfaction. Perhaps they would have a great deal less to worry about too? Sometimes senior scientists respond enviously when I mention that I still work in the lab (only one or two days a week though) – most of them would like to, but don’t have the time.

In the academic environment, what solutions does this study suggest? It suggests that funding agencies should make it very easy for almost anyone to get a small amount of money. In the UK, the trend has been rather the opposite, with increasing disparity between the haves and have-nots following elimination of a large number of small grants in favour of a small number of large grants (I’m looking at you, Wellcome Trust). Small funding instruments would reduce the current obsession with super-sized grants, which people are always talking about but few people ever receive (full disclosure: I’ve benefitted from the odd super-grant myself).  There is a place for big grants, but I doubt anyone can ever be certain that a single act of funding to the tune of 2 million will pay off. That is, will such a grant have the better value for money, as compared to say 8 grants of 250 thousand. You can put pounds, euros or dollars there. Those are the ballpark figures. For a start, it’s probably impossible to measure the output fairly, and therefore, hard to judge it properly. To be fair, Wellcome intends to catalyse earth-shattering research by funding “Investigators”, and it’s their money to do with as they will.

It’s hard to measure the success of big grants, but again, this paper has an insight. Rich players spent less time lining up their slingshots in game two, an Angry Birds clone, than poor players did. This is akin to one strategy that is widely presumed to be practiced by big labs, along the lines of sink-or-swim. If 1/4 of the postdocs from a big lab get a Cell-Nature-Science paper at some point, then the lab is world-famous and very succesful, and who cares about the rest of the ones that publish next to nothing. The boss can concentrate on the stars. People are generally the most expensive thing in a lab, except for the biggest, shiniest toys, and so this strategy is very expensive for funders. One can always argue that high-risk, high-payoff research must entail failure (I agree) but imagine if you (the lab head) were judged on the performance of the median postdoc. Full disclosure: 4 postdocs, 3 students at the moment (not big by German standards, but…). Darwinian approaches may improve output episodically, but the consequences for the “human consumable” aren’t palatable.

One can take the view that players will always “game” any system. The German system of a large number of smaller, shorter grants isn’t a panacea. One consequence of a flatter funding landscape is that Germany (like France) is working to introduce more “excellence” into its University research sphere, using big team grants. But this simple study strongly suggests that “treat ‘em mean, keep ‘em keen” won’t ever work, either for poverty, or for the advancement of science. It leads to self-destructive behaviour.

About these ads

4 thoughts on “Some Consequences of Poverty

  1. Pingback: The funding ‘lottery’: how many grants should we be applying for? | Infectious Thoughts

  2. I couldn’t agree more. I’ve been pursuing the idea for a while (thanks for the link). I’d prefer not to rely on psycholgy though (specially given the recent publicity about how little of it can be replicated). What I’d really like is decent information on value as a function of group size (or funding). The problem, as you say, is the great difficulty in measuring value. I may be unduly biased by the observation of three Nobel prizewinners (Huxley, Katz and Neher & Sakmann who all had remarkably small labs. at least at the time they did the work that got them the prize. They did the work themselves. All the pressure now is to do exactly the opposite. I can’t think that is a good idea.

  3. Contra recent high-profile cases, most psychological research can be replicated. Of course there are failures to replicate, but I’d challenge David to provide any evidence that psychology replicates less often than any other discipline.

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