Freakonomics tries to turn the scalpel of the analytical and statistical methods intrinsic to Economics onto questions that the authors feel do not have definitive answers. Mostly because no one thought to ask the questions that would allow us (the world at large, not Economics students) to solve the problems that would lead to the answers, the authors feel.
Because of this, Freakonomics is attended by all the problems of the so-called soft sciences. The data, and the conclusions often seem susceptible to multiple interpretations. There is a relatively large “eye of the beholder” problem with books like Freakonomics, compounded by the authors’ failure to provide (or intentional decision to omit) the data from which they draw these conclusions. Because the data is hidden, it’s difficult for the reader to look behind the curtain and determine if he would come to similar conclusions. We can’t say if they’re right. All we can do is say whether we believe them or not. Not if we agree. Faith.
Freakonomics is of those books where if you agree with the conclusions then you say: Exactly. But if you don’t, you have to take it on faith that the authors are correct and you aren’t. There doesn’t seem to be any method to debunking this, other than to call out inconsistencies. I am troubled by this, it’s almost like turning science into a religion, rather than the explanation of phenomenon that we need Science to be so we can be better, or build the better world.
That being said, Freakonomics sometimes seems to come to inconsistent conclusions. Or at least ones that definitely seem susceptible to dispute.
Freakonomics concludes that more police officers leads to less crime. They come to this conclusion they say by analyzing the crime rate in a city that recently had an election with cities that have not had an election. According to Freakonomics, cities that recently had an election usually have politicians who hire more police officers to seem tough on crime.
Data that shows that a politician who is savvy enough to try to tip the election in his favor by hiring more police while suggesting that these additional officers will contribute to lower crime, also tends to suggest that this politician will lean on heads of police departments to underreport crime so as to be able to show a dip in crime and pander to the law and order constituency. AKA, the people most likely to vote.
There are other questions served poorly by the quick gloss we’re asked to accept. How is he measuring a drop in crime? Drop in crime reporting statistics? Why would hiring more officers cause a drop in crime? Why aren’t more officers arresting more people? And why would hiring more officers prior to election cause a drop in crime? Are these officers being trained at super speed? How long after the election is he measuring? How can we know that the hiring of the officers had an effect in a particular time frame?
It seems like more than one conclusion can be...