An absurdly and I mean QUITE ABSURDLY overdue somewhat-review with caveat of Freakonomics
Freakonomics is a book exploring questions about society with the tools of microeconomics and statistics. It's a book full of interesting questions--why did crime fall in the 1990's? can we detect cheating by teachers in standardized testing? do real estate agents try harder with their own houses? why do drug dealers live with their mothers if they're making riches?--which are asked with some economics framing and some very clever experiment set-ups. The thing with research into these fields is that they're impossible, either because of ethics or sheer impossibility, to control variables the same way you might an inorganic chemistry experiment. But Levitt uses opportunities in different data ingeniously to extract data. For instance, Chicago's public school system created a lottery assignation of spots in different schools in the city--this is the sort of thing a researcher can only dream of, to randomly be able to place subjects in different "treatments" i.e. schools (without regard for socio-economic status and similar factors that might be the reason for attendance at a particular school) and thereby tease out differences in education. This is what they call randomized controlled trial, by the way.
The questions are treated in an economics way, by which I mean there's discussion of common concepts in economics (incentives, etc). But the questions themselves are rather a departure from most economics texts--for one, they're not macro! (I feel macroeconomics--you know, GDP, unemployment, monetary policy, exchange rates--has practically dominated the field, and the perception of the field, for a long time. Then again, what do I know?) The book's written in a very non-technical manner. If you look very closely Levitt and Dubner actually talk about multiple regressions and dummy variables when discussing methodology, but for the most part the technical part about statistics and economic theory are completely elided.
For me, the most interesting part of the book wasn't necessarily the conclusions that Levitt reached or the anecdotes, though both were excellent, but instead the way Levitt got them. Designing experiments to test hypotheses is sometimes really hard, and Levitt extracts meaningful data from larger pools (e.g. government records on whatever) to make sense of it. I really love experimental economics and the elegance of these experiments are amazing and fascinating.
However: the reason why this review has languished since June 2013--Levitt's conclusion regarding abortion and crime rates came under criticism. Not from random pro-choice people but rather pretty respected economists like ones who work at the Boston Fed. Here is a link to the paper that Foote and Goetz wrote: http://www.bostonfed.org/economic/wp/wp2005/wp0515.pdf It's been ages and I've still not read that stupid thing so I have been holding off posting this review but really, enough is enough and it's clogging up my desktop (purposefully, I mean, I put documents on my desktop when I want them to be done ASAP because it drives me mad to have a cluttered virtual desktop. But less about me.) So grain of salt, as with every economics paper you ever read.
(Reinhart and Rogoff, anyone?)
Finally, I would like to quote a part of the Q&A that is appended to my version of the book, which made me laugh:
Tell us about the criticisms you have received from traditional/academic colleagues over Freakonomics--J Plain.
Levitt's academic colleagues tend to react in one of two ways. The majority of economists thought about it like economists: the success of Freakonomics probably increased the number of students wanting to take economics courses, and since the supply of economics teachers is fixed in the short run, the wages of academic economists should rise. That makes economists happy. A second group of economists decided that if Levitt could write a book that people would read, surely they could too. So there has been a flurry of "popular" books by economists--some good, some not so good. And then, inevitably, there are a handful of economists who feel that he violated the secret handshake of economics by showing the outside world that what economists do really isn't that hard or complex. They will never forgive him.
supply of economics teachers fixed in the short run, it's like the most basic essence of economics distilled into one sentence. I'm laugh-crying into my fingers at this paragraph.
Crosspost: http://silverflight8.dreamwidth.org/150706.html.
The questions are treated in an economics way, by which I mean there's discussion of common concepts in economics (incentives, etc). But the questions themselves are rather a departure from most economics texts--for one, they're not macro! (I feel macroeconomics--you know, GDP, unemployment, monetary policy, exchange rates--has practically dominated the field, and the perception of the field, for a long time. Then again, what do I know?) The book's written in a very non-technical manner. If you look very closely Levitt and Dubner actually talk about multiple regressions and dummy variables when discussing methodology, but for the most part the technical part about statistics and economic theory are completely elided.
For me, the most interesting part of the book wasn't necessarily the conclusions that Levitt reached or the anecdotes, though both were excellent, but instead the way Levitt got them. Designing experiments to test hypotheses is sometimes really hard, and Levitt extracts meaningful data from larger pools (e.g. government records on whatever) to make sense of it. I really love experimental economics and the elegance of these experiments are amazing and fascinating.
However: the reason why this review has languished since June 2013--Levitt's conclusion regarding abortion and crime rates came under criticism. Not from random pro-choice people but rather pretty respected economists like ones who work at the Boston Fed. Here is a link to the paper that Foote and Goetz wrote: http://www.bostonfed.org/economic/wp/wp2005/wp0515.pdf It's been ages and I've still not read that stupid thing so I have been holding off posting this review but really, enough is enough and it's clogging up my desktop (purposefully, I mean, I put documents on my desktop when I want them to be done ASAP because it drives me mad to have a cluttered virtual desktop. But less about me.) So grain of salt, as with every economics paper you ever read.
(Reinhart and Rogoff, anyone?)
Finally, I would like to quote a part of the Q&A that is appended to my version of the book, which made me laugh:
Tell us about the criticisms you have received from traditional/academic colleagues over Freakonomics--J Plain.
Levitt's academic colleagues tend to react in one of two ways. The majority of economists thought about it like economists: the success of Freakonomics probably increased the number of students wanting to take economics courses, and since the supply of economics teachers is fixed in the short run, the wages of academic economists should rise. That makes economists happy. A second group of economists decided that if Levitt could write a book that people would read, surely they could too. So there has been a flurry of "popular" books by economists--some good, some not so good. And then, inevitably, there are a handful of economists who feel that he violated the secret handshake of economics by showing the outside world that what economists do really isn't that hard or complex. They will never forgive him.
supply of economics teachers fixed in the short run, it's like the most basic essence of economics distilled into one sentence. I'm laugh-crying into my fingers at this paragraph.
Crosspost: http://silverflight8.dreamwidth.org/150706.html.