Peters Against Aggregation

23 April 2009 at 4:27 pm Peter G. Klein Leave a comment

| Peter Klein |

When I saw the title of Brayden’s post, “Don’t Give Up on Aggregation Yet, Peter,” I thought he’d been reading my macroeconomics posts. Alas, Brayden, prefers meatier fare, such as this post by Barnard College sociologist Peter Levin. Levin is worried about the aggregation of knowledge represented by the open-source, wikified, crowdsourcing movement about which people are all, well, atwitter. (We’ve expressed more than a few reservations about this stuff ourselves.) His main concern, if I understand correctly, is the possibility of information cascades. However, much of the cascades literature deals not with the wisdom of crowds, but the wisdom of experts (tulip-bulb traders, mortgage-backed securities underwriters, etc.). The more expertise decision-makers grant to their peers, the more likely they — in the face of uncertainty — will interpret their peers’ (ostensibly expert) opinions as reliable indicators of underlying reality, and hence the greater the likelihood of cascades.

Brayden takes a different tack, arguing that aggregation mechanisms can be designed to mitigate the chance of outliers biasing the results. I think Brayden is right but am not sure his comments address the underlying mechanism — the microfoundations, to use a certain co-blogger’s favorite term — that Levin is worried about.