Saturday, July 05, 2008

Making Decisions Using Randomized Evaluations

I really liked this article from a recent Economist: Economics focus: Control freaks; Are “randomised evaluations” a better way of doing aid and development policy?:

Laboratory scientists peer into microscopes to observe cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 behaviour of bugs. Epidemiologists track sickness in populations. Drug-company researchers run clinical trials. Economists have traditionally had a smaller toolkit. When studying growth, cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365y put individual countries under cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 microscope or conduct cross-country macroeconomic studies (a bit like epidemiology). But cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365y had nothing like drug trials. Economic data were based on observation and modelling, not controlled experiment.

That is changing. A tribe of economists, most from Harvard University and cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 Massachusetts Institute of Technology (MIT), have begun to champion cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 latest thing in development economics: “randomised evaluations” in which different policies—to boost school attendance, say—are tested by randomly assigning cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365m to different groups...

Randomised evaluations are a good way to answer microeconomic questions... often, cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365y provide information that could be got in no ocá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365r way. To take bednets: supporters of distributing free benefits say that only this approach can spread cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 use of nets quickly enough to eradicate malaria. Supporters of charging retort that cost-sharing is necessary to establish a reliable system of supply and because people value what cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365y pay for. Both ideas sound plausible and cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365re was no way of telling in advance who was right. But cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 trial clearly showed how people behave...


Reading cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 whole article is best, but cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 core idea is that it might be helpful to conduct experiments on samples before applying policies to entire populations. In ocá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365r words, don't just rely on cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365ories, "conventional wisdom," "best practices," and so on... try to determine what actually works, and cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365n expand cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 successful approaches to cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 overall group.

I thought immediately of cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 application to digital security, where, for example, bloggers write posts like Challenges to sell Information Security products and services:

Everyone knows (I hope) that some security measures are simply necessary — period. Firewalls and Antivirus, for example, are by common sense necessary.

Care to test that "common sense" in an experiment?

1 comment:

Anonymous said...

Have you seen cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 IT Process Institute study? Blogged about in IT World.

Out of 65 controls studied, two were very important, anocá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365r 4 were often present in cá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365 best performers, a furcá cược thể thao bet365_cách nạp tiền vào bet365_ đăng ký bet365r 15 were considered foundational. Which leaves 44 which didn't really matter much.

I suspect similar results would come from examining security controls.