So: exploratory factor analysis exploits correlations to summarize data, and confirmatory factor analysis stuff like testing that the right partial correlations vanish is a prudent way of checking whether a model with latent variables could possibly be right. What the modern g-mongers do, however, is try to use exploratory factor analysis to uncover hidden causal structures. I am very, very interested in the latter pursuit, and if factor analysis was a solution I would embrace it gladly. But if factor analysis was a solution, when my students asked me (as they inevitably do) "so, how do we know how many factors we need?", I would be able to do more than point them to rules of thumb based on squinting at "scree plots" like this and guessing where the slope begins. (There are ways of estimating the intrinsic dimension of noisily-sampled manifolds, but that's not at all the same.) More broadly, factor analysis is part of a larger circle of ideas which all more or less boil down to some combination of least squares, linear regression and singular value decomposition, which are used in the overwhelming majority of work in quantitative social science, including, very much, work which tries to draw causal inferences without the benefit of experiments. A natural question but one almost never asked by users of these tools is whether they are reliable instruments of causal inference. The answer, unequivocally, is "no".
(And further down in the summary):
Building factors from correlations is fine as data reduction, but deeply unsuited to finding causal structures. The mythical aspect of g isn't that it can be defined, or, having been defined, that it describes a lot of the correlations on intelligence tests; the myth is that this tells us anything more than that those tests are positively correlated. It has been known for almost as long as factor analysis has been around that positive correlations can arise in many ways which involve nothing remotely like a general factor of intelligence. Thomson's ability-sampling model, with its myriad independent causes rather than a single general cause, is the oldest and most extreme counter-example, but it is far from the only one. It is still conceivable that those positive correlations are all caused by a general factor of intelligence, but we ought to be long since past the point where supporters of that view were advancing arguments on the basis of evidence other than those correlations. So far as I can tell, however, nobody has presented a case for g apart from thoroughly invalid arguments from factor analysis; that is, the myth.
---------
Max Sawicky posted the link, but here it is again:
http://cscs.umich.edu/~crshalizi/weblog/523.html
Read the whole article. The bottom line is that data from IQ tests establish a correlation to a factor called the g-factor. However, there is no causal relationship between g-factor and a putative phenomenon called general intelligence, period.
The above is really worth the few minutes to go through the essay. I went through it too quickly this morning on the way to work. I'll go back later and go over it some more.
The concluding paragraph:
``In primitive societies, or so Malinowski taught, myths serve as the legitimating charters of practices and institutions. Just so here: the myth of g legitimates a vast enterprise of intelligence testing and theorizing. There should be no dispute that, when we lack specialized and valid instruments, general IQ tests can be better than nothing. Claims that they are anything more than such stop-gaps \u2014 that they are triumphs of psychological science, illuminating the workings of the mind; keys to the fates of individuals and peoples; sources of harsh truths which only a courageous few have the strength to bear; etc., etc., such claims are at present entirely unjustified, though not, perhaps, unmotivated. They are supported only by the myth, and acceptance of the myth itself rests on what I can only call an astonishing methodological backwardness.
The bottom line is: The sooner we stop paying attention to g, the sooner we can devote our energies to understanding the mind.''
Well, we could say that if IQ tests were better than nothing, we could also say, that judging IQ by looking at stool samples was better than nothing. Like the IQ tests, the stool sample must occasionally be a causal indicater at least once in a while.
Most of the day I was thinking about the broad implications of various policy institutes and the think-tank industry that supplies the US political establishment and it adminstrators with the various directions for domestic and foreign policy.
The total politicalization of private and public university institutes, not to mention the think-tank apparatacheks, combined with such devastating methodological critiques make a utter mockery of the ideal of an enlightened, empirically based policy apparatus for modern states. Sure we all knew this, but really, did we quite understand that there is no there, there at all?
With the White House censorship of the report from public health agency studies attempting to find correlations between various ER admissions and health problems and possible climate changes, it seems that we have completely lost any creditibility in even the most basic protective service capacity of elected government.
We talk about the general decline in popular understanding of the world, which could be debated forever. But, consider that the endless politicalization of both health and social science studies has managed to corrupt these policy tools to such an extent that they can not contribute anything to our understanding of society and the world.
Now I expect the cynics to say, well that's always been the case.
But that is not true. There was a day, when at least some public policy was based on accurate and enlightened empirical studies that did indicate various directions in which we could hopefully change society for the better. I am thinking of many of the early to mid-60 work done on poverty, education, discrimination, and the findings that identified economic deprevations, loss of jobs, break up of families, under funded and segregated schools, poor housing, and other factors that lead to the well known profiles in poverty that could be changed at the policy level.
Many of these sorts of studies, layed the foundation for my own intellectual radicalization. Guys like Jonathan Kozol, Death at an Early Age (1967) and William Domhoff, Who Rules America (1969) provided devastating empirical accounts of US realities.
I think that process is still available and it is now up to the left intelligencia to provide that kind of material. Most of the liberal establishment can no longer be counted on, since they have uniformly fallen into the neoliberal camp of capital apologists.
Doug's first book Wall Street peeled my eyes pretty much the same way, although I was already so jaded about the US economic system, it was difficult to get through the orderly and detailed hatchet job on bastards I already hated---and would have shot, come the revolution. I ordered Michael Yates's Cheap Motels and expect a dark narrative on the state of the nation---gritty stuff that I probably already suspect, but haven't seen in print. Mike Perelman's several books a few years ago the patents and the internet did similar things. There are Susan Faludi, Stiffed, Naomi Klein's last few, Barbara Ehrenreich, Nicked and Dimed, and others. Most of the latter I haven't read.
Yes, I knew it was bad, but I just didn't have the detail to argue the case. That's the stuff that counts. It counts in the old fashioned liberal way, that any empirically accurate assesment counts to shape the policy apparatus of the state, for the benefit of the people, as oppossed to the benefit of the economy, i.e the power elite.
Just about anything that uses IQ is probably not only mythological, but utterly reactionary at this point in time.
(BTW, the photo of Ehrenreich is a perfect for a Diebenkorn figure study, for anybody interested...):
http://en.wikipedia.org/wiki/Image:Barbara_Ehrenreich_by_David_Shankbone.jpg
