« The Surge | HomePage | A Good Day »

Monday, January 28, 20081201567009

The Quantitative Revolution

Revolutions break eggs to make omelets. Omelets are tasty. Broken eggs are messy. Hence, the essential problem of revolutions.

Of all the controversies I have learned about since entering the University of Nebraska, none has fascinated me so much as the Quantitative Revolution. The Quantitative Revolution, or QR, has radically transformed social research in academia. It is as much of a revolt against all that has gone before as Marxism. QR is a rejection of all that would interest a bright adolescent in social research. It is also, I think, all that can save social research from Marxism.

To understand this war, think about politics, or psychology, or geography, or any of those subjects that interested you when you read an Encyclopedia as a kid. Think of the Plato and Machiavelli pondering Politics, Freud and Adler plumbing the subconscious, and explorers and theorists deciding what is a Sea and what is a Bay. This is social research as it existed from antiquity to sometime in the 20th century.

Now throw that out. Instead measure things, and note what varies with what.

That's the Quantitative Revolution. It's very powerful, because it's actually science: It provides a way of showing you when you are wrong, and a methodical way for supporting your intuition when it is right. Is man, for instance, truly a political animal? Well, measure where his nature comes from (neatly dividing it into biological influences, non-biological influences shared with one's siblings, and non-biological influences not shared with one's siblings) among a diverse enough population, regress it, and suddenly you get answers. More than that, you get repeatable answers which allow you move on to something else without throwing your old work away.

Yet QR is a profoundly dull revolution. All the great questions become matters for vertical thinkers and technicians. An academic career in the era of the QR essentially is the process of limiting your imagination to one or two good tools, and measuring variation with those tools. The sort of people who enjoy being accountants, I think, love life under the Quantitative Revolutionaries.

Yet the enemy of my enemy is my friend, and QR targets its wrath most consistently against the Marxists, dead-ender followers of a 19th century Revolution that have burrowed themselves deep into academia. Marxists have spent a century developing a self-consistent toolbox of rhetoric that has dispatched non-Marxists in nearly every academic field. Every place the Quantitative Revolution has not taken and held, it seems, is territory in which Marxists rapidly make their home.

I despise, I think, the Quantitative Revolution for depriving academia of the qualitative give-and-take that is so common in the better parts of the blogosphere. But I delight in the ease at which the Quantitative Revolution unseats the Marxists every time it gains a foothold, overwhelming the Marxist immune system through dull questions of covariation and how-do-you-know-if-you-are-wrong?

Comments

You talk like qualitative and quantitative can't coexist! Yet if qualitative analysis and thought isn't what decides where to use the quantitative tools and how to interpret the results, then what is?

Posted by: Michael | Monday, January 28, 2008

I find there is can be a balance. I am not sure if the battle between the two can ever end, epic as it is, but some people I have come across do a brilliant job balancing the two without any Marxist undertones.

Have you read KI's take on this? The money quote:

Scientisim in intelligence analysis is a particularly seductive heresy. It offers the false promise of greater insight, should only additional efforts be applied more systematically, more rigorously, or with more and better data.

Posted by: Younghusband | Monday, January 28, 2008

Michael,

By qualitative I meant qualitative analysis of data, as opposed to theoretical research, metareviews, etc.

Younghusband,

Who's KI?

Posted by: Dan tdaxp | Tuesday, January 29, 2008

Dan,

KI = Kent's Imperative blog (I think)

Are you familiar with Ian Ayers? His book looks pretty cool:
http://www.amazon.com/Super-Crunchers-Thinking-Numbers-Smart/dp/0553805401

Posted by: vimothy | Tuesday, January 29, 2008

Younghusband & vimothy,

Gotcha!

Strongly agree on the value of qualitative data anlysis within applied analysis.

My purpose in the post was to weigh the benefits of Quantitative Analysis (practically, in terms of usable results; ideologically, in terms of marginalizing Marxists) against its costs (the death of the romance of social research), within academia.

Clearly policy analysts must exploit all sources of information, and scientific research is just one of these.

Posted by: Dan tdaxp | Tuesday, January 29, 2008

Sounds a bit like "Revenge of the Nerds" to me... :-) What's ironic is that I was schooled in that most quantitative of sciences (physics) and came to abhor the "qualitative sell-out" of quantum theorists -- Heisenberg chief among them. So while the "soft" sciences try to get more specific, physicists have been getting more lax.

My first encounter with Dr. Tom Barnett (shortly after the turn of the Millennium) was a dialog on "Perturbation Theory". Tom has used the term to great effect in articulating his grand strategy, while to physicists the term means "get as close as you can quantitatively then just fudge the rest".

The rub always comes when you have to select the parameter(s) to "quantize" -- Michael's opening comment above is a great way to describe the coexistence (codependency? :-) of qualitative and quantitative methods. To rely on one too much will skew the results and lead you farther from the truth.

Posted by: deichmans | Tuesday, January 29, 2008

Dan,

Can you demonstrate how you are not just blanket stereotyping the whole spectrum of the American Left as Marxist.

There really is no denying that most academia are on the American Left (but whether they really go out of their way to suppress opposition is another argument for another time), but I have yet to ever know a professor who considers themselves a Marxist.

Posted by: Jeffrey James | Tuesday, January 29, 2008

BTW, I have nothing against your assertion of the Quantitative Revolution that you say is taking place in academia, but I don't see Marxism as being the establishment that rivals it.

Posted by: Jeffrey James | Tuesday, January 29, 2008

Also, I'm very much in love with Bruce Bueno de Mesquita's work at present. He performs ultra rigorous game theoretical analyses of international relations scenarios for US government agencies, as well as leading the charge for uber-mathematical political science.

I have a (totally non-quantitative) post on one aspect of his approach here:
http://vimothy.wordpress.com/2008/01/10/the-political-economy-of-democracy/

Or check out his (excellent) book here:
http://www.amazon.co.uk/Logic-Political-Survival-Bruce-Mesquita/dp/0262524406/ref=sr_1_10?ie=UTF8&s=books&qid=1199962598&sr=8-10

Posted by: vimothy | Tuesday, January 29, 2008

Umm, maybe this would be a good point to rest the Marxism part of the argument and concentrate on the romance part? What is this romance (aside from cute coeds, that is) that draws students into psychological research? What part of the dream is made impossible by quantitative analysis?

Posted by: Michael | Tuesday, January 29, 2008

Deichmans,

Well said. Issues of operationalization, measurement, statistical analysis, etc, are very powerful and very easy to misuse.

Jeffrey,

"There really is no denying that most academia are on the American Left (but whether they really go out of their way to suppress opposition is another argument for another time), but I have yet to ever know a professor who considers themselves a Marxist."

Doubtless few businessmen consider themselves (Adam) Smithist, either.

In my experience at Nebraska, the bulk of the professariat is liberal, and thus non-Marxist.

I'm pretty sure we've both studied under at least one professor (I'm thinking of specific names here) who has not been able to execute his program of research because of the influence of Marxists.

vimothy,

Minimum winning coalition is a well-established idea in political science. Re: your post, how is it cynical?

Michael,

An easier example to get across is geography. Articles on geography before the Quantitative Revolution read like more thorough versions of National Geographic articles from that time. Any kid who spent hours examining maps or globes would instantly recognize the pursuit, and be appreciate of how much more is now known because of the article.

Post QR, Geography is a social science that uses space as an independent variable.

Posted by: Dan tdaxp | Wednesday, January 30, 2008

"There really is no denying that most academia are on the American Left (but whether they really go out of their way to suppress opposition is another argument for another time), but I have yet to ever know a professor who considers themselves a Marxist."

Come to the UK -- in my experience, pretty much everyone in social sciences is a Marxist!

Dan,

The book is cynical in that it views governments as being fundamentally constrained by their institutional context, and that it is this context that variously produces development and prosperity or stagnation and poverty. The Logic of Political Survival sees all leaders as being fundamentally self-interested, and considers that policy will be chosen on the basis of maintaining the incumbent in office: if that coincides with good policy, fine; if not, tough. It has nothing to do with the nobility or altruism of the actors -- be they democrats or dictators -- quite the opposite, in fact...

(Though of course -- this shouldn't be news to anyone).

Posted by: vimothy | Wednesday, January 30, 2008

KI is Kent's Imperative. Blogspirit stripped my post of HTML so the link was lost. Here is the linktext:

http://kentsimperative.blogspot.com/2008/01/perils-of-arbitrary-and-false-precision.html

Posted by: Younghusband | Wednesday, January 30, 2008

Dan, is that the fault of QA or of article writers who don't know how to (or are too lazy or rushed to) write? Heck knows I'd done my share of "Here's the graph/spreadsheet. What do you think?".

Put this another way. Think through all the articles you've read AND UNDERSTOOD WELL that were rendered dull and romance-less this way; try not to commit suicide in the process (I don't want Fei tracking me down). How many could you rewrite in an interesting manner, with the dull mathematical portions tucked off in an over-sized appendix somewhere?

Posted by: Michael | Wednesday, January 30, 2008

Michael,

For a flavor of geography before the quantitative Revolution, check out Williams (1927) "Geographical determinism in Nicaragua" [1]. The question is not one of writing style, but rather the qualities or quantities of an object of interest is the primary focus of an article.

Younghusband,

KI's point is well made. Ultimately, it comes down to this: analysis is not a science.

vimothy,

Well said.

The book looks like a combination of institutionalism and rational man theory... Rational choice is very powerful, as its very mathematical. However, it's also unfalsifiable: as rational choice describes all actions are resulting from an internally consistent preference schedule, any action is thus confirmation of its assumption.

[1] http://links.jstor.org/sici?sici=0002-7162%28192707%29132%3C142%3AGDIN%3E2.0.CO%3B2-P

Posted by: Dan tdaxp | Wednesday, January 30, 2008

Dan,

'To verify the accuracy of his model, the CIA set up a kind of forecasting face-off that pit predictions from his model against those of Langley’s more traditional in-house intelligence analysts and area specialists. “We tested Bueno de Mesquita’s model on scores of issues that were conducted in real time—that is, the forecasts were made before the events actually happened,” says Stanley Feder, a former high-level CIA analyst. “We found the model to be accurate 90 percent of the time,” he wrote. Another study evaluating Bueno de Mesquita’s real-time forecasts of 21 policy decisions in the European community concluded that “the probability that the predicted outcome was what indeed occurred was an astounding 97 percent.” What’s more, Bueno de Mesquita’s forecasts were much more detailed than those of the more traditional analysts. “The real issue is the specificity of the accuracy,” says Feder. “We found that DI (Directorate of National Intelligence) analyses, even when they were right, were vague compared to the model’s forecasts. To use an archery metaphor, if you hit the target, that’s great. But if you hit the bull’s eye—that’s amazing.”'

http://www.goodmagazine.com/section/Features/the_new_nostradamus

Posted by: vimothy | Thursday, January 31, 2008

HAS THE APPLICATIONS OF QUANTITATIVE TECHNIQUES HELP GEOGRAPGY AS A SCIENCE?

Posted by: SELELO | Monday, February 18, 2008

Selelo,

Yes, because without quantitative techniques there can be no science. Science is built on the proposition and refutation of quantitative hypotheses.

vimothy,

Long Term Capital Management's model was tested and verified the same way, if I believe.

Having developed a model-simulation for my master's thesis, I'm pretty skeptical of claims they work: There's a lot of arbitrariness in any model.

Posted by: Dan tdaxp | Monday, February 18, 2008

Post a comment