It is one of those statements that seems to be trotted out whenever political rivals debate, inevitable someone on the left will counter some argument or another with the claims that "intelligent people" are overwhelmingly liberal. They may word it differently, there are a number of variations. The most educated people are liberal (much more likely, as we shall discuss, but also much less meaningful). People who are interested in ideas are liberal. Creative people are liberal. And so on. Some are more nebulous than others, some much harder to prove, but the idea is the same, by some measurement or another, liberalism is the philosophy of choice for the more clever, inventive, creative and so on. Nor is it simply trotted out as an unsubstantiated taunt, the way some on the right accuse leftism of being the philosophy of the immoral, hypocritical, what have you. No, those on the left who put forth this claim actually believe it is a fact, one which can be proven*.
There are a host of technical issues with such a theory, so let us address those first, before going into the other problems.
Most noteworthy, the studies are invariably based on self-identification, and, more importantly, are almost always done in some academic setting. As one who was quite a pariah among his art school friends due to being an admitted conservative, I know that in most modern academic settings it is simply habit to call oneself liberal, as any other answer seems to inspire troubling levels of hostility from certain individuals, not to mention the assumption by many that anyone self-identifying as conservative is unintelligent, perhaps racist, sexist and other, less savory, things. Even asking about specific political stands versus one's political stance would not be trustworthy, as, again, most learn to say they favor a host of liberal positions, lest they be ostracized in the college community.
A second issue is self-selection. Any study that relies upon volunteers, rather than a random selection from a non-volunteer pool, is going to automatically self-select for those inclined to volunteer for such studies. Of course, non-voluntary studies have a different issue, potential resentment or apathy on the part of those dragooned into a study. I know during my high school years my standardized test scores dropped as I aged, mostly because I started viewing them as a waste of time, and did not put in as much effort as I could have**. So, similarly, those forced into tests may not test well if they are the kind disinclined to volunteer, making the results similarly skewed either way.
Finally, or at least the final one we shall discuss (there are definitely more issues than these three), the studies done on young participants make the error of confusing youthful political inclinations with one's lifelong politics. The famous Churchill quote is not far from reality, most young people are much more liberal than later in life, even lifelong liberals tend to have been more liberal in early life. There are many, many reasons for this (see "Ritual Abuse, Backwards Logic and Conspiracy Theories", "The Path of Least Resistance", "Prelude to a Future Essay on Heroic Ethics and Romanticism"); here are a few I have pointed out before. For example, young and immature people tend to be attracted to grand struggles to "save the world", and thus find the revolutionary, missionary attitude of the left, especially the far left, much more appealing than the staid and sensible conservative attachment to gradual change and piecemeal reform. (See "Conservatism, Incremental Change and Federalism", "In Praise of Slow Changes") Likewise, with many in education, both secondary and collegiate, tending toward liberal beliefs, it is not surprising that many young individuals would adopt the liberal beliefs of their teachers. Finally, with much less to lose, and a general dissatisfaction with their status quo -- as well as many embracing the youthful tendency toward rebellion -- it is understandable that the young would be attracted to revolutionary politics. (See "Juvenile Intellectuals", "Deadly Cynicism", "Self-Serving Cynicism and Our Cultural Immaturity", "O Tempora! O Mores!, or, The High Cost of Supposed Freedom", "An Immature Society", "Trophy Spouses", "Cranky Old Man?", "Pushing the Envelope")
But there are many issues beyond the technical questions of the studies themselves. (Though we shall look at one more of those later.) For example, there is the underlying assumption implicit in this theory that liberalism is somehow "right". After all, if all the most intelligent people are liberal, does that not imply that liberalism is the correct answer? Of course, this also implies there is a single correct answer to many complicated questions, and that only those in the know, this intellectual elite, is privy to this knowledge. Such arrogant elitism, and dismissal of the ignorant masses is an inherent part of liberalism (and other interventionist theories), so it is not exactly shocking to find it repeated here. (See "Liberalism, Its Origins and Consequences", "Man's Nature and Government", "Appealing to Arrogance", "The Intellectual Elite", "The Citizen Dichotomy", "The Essence of Liberalism","Liberalism, "Idealists" and Internal Contradictions", "Big Government, Arrogance and Part-Time Psychopathy", "For Your Own Good -- The Problem with Subjective Rights", "The Road to Violence", "The War of All Against All") The surprising thing is how many supposed scientists will lend authority to this prejudice by giving it their support.
Of course, it is a bit hypocritical as well. If liberals were to be told intelligent people were predominantly Episcopalian, or spoke English, or, worse still, were white or male***, they would be quick to dismiss these claims as mistaken, or, at most, the result of environmental, cultural or other confounding factors. Yet, in this case, they are completely blind to the many confounding factors that exist.
First, even using IQ there is the problem of confusing intellect and education. There are a number of IQ tests which try to avoid relying upon education, but regardless of how well designed, to some degree the expression of intellect relies upon the information one has. If you are not aware of the basic principles of geometry, such as counting sides or vertices, even simple pattern problems, which are supposedly free of educational requirements, become more difficult. In the end, education will always be something of a confounding factor.
And that is not a trivial issue, as education -- as pointed out above -- tends to correlate with liberal viewpoints. This does not mean more intelligent people are liberal, but rather that greater education tends to incline one towards liberalism (just as in the past, or in other nations, greater education tended to instill a more conservative viewpoint****), and there is always going to be some confusion between intellectual ability and education.
Allow me to explain with a simple, if imprecise, analogy. Intellectual ability is something like a motor, while education, or to be more precise, information, is like fuel. Without any information upon which to work, the motor cannot function, and similarly, intellectual ability divorced from information is useless. One could have the greatest capacity in the world, but if he never studied physics, he will not develop a Grand Unified Theory as he will not be aware of what questions remain to be answered, or even that there are questions at all. And so, even though we like to think IQ tests measure pure intellectual potential independent of education, the fact remains any test embodies some elements of education. Word comparisons, shape or numeric pattern recognition, and so on, all require some external information, and thus require certain education. Beyond that, even if the test does not require specific education, having certain information can make things easier. For example, as mentioned above, knowledge of the importance of sides, vertices and so on may make it easier to identify patterns in shapes than it would be for those alien to the study of geometry. Thus, intellectual ability, while it may exist as a potential independent of knowledge or education, cannot be measured completely in isolation. And thus, in the end, educational level will skew tests of intellect, and the liberal bias of modern schooling will cause it to appear the correlation between IQ and liberalism is greater than it truly is, if it exists at all.
Then there is the fact that, even if there might be a correlation, it tells us nothing about either the truth of liberal beliefs or the intellect of any given liberal. After all, one may be intelligent, one may even be a genius in a specific field, yet be incompetent in another. For example, Linus Pauling's famous, and completely mistaken, belief in the benefits of massive doses of vitamin C is not proved correct by his genius in other fields, and, likewise, Einstein's gift for physics does not mean that his personal politico-economic views are superior to anyone else's. Just because some brilliant men were liberals, or socialists, it proves nothing. Many brilliant men held conflicting views.
The similar error, that of assuming liberalism proves one to be intelligent, is akin to the mistaken reasoning of racist or sexist theories. Even if it could be shown, for example, that whites were more intelligent, or men smarter than women, it would mean nothing in terms of individuals. Just because one groups has an elevated average, any specific member can still fall anywhere on the scale. To draw an example from a less controversial subject, the mean height in the US (163.2cm for all Americans, 163.2 cm for black Americans) is higher than the mean height in China (160.1 cm on the mainland), that does not mean Yao Ming is shorter than Gary Coleman. Similarly, even if it could be shown more intelligent people were liberal, it would say nothing about any given liberal's intellect. While it may please him to bask in the reflected glory of sharing views with someone clever, any given liberal could be just as foolish, or even more so, than the most foolish conservative.
It would not be hard to go on, as not only do the various studies have their own flaws, but the meanings assigned to them by various liberals go even farther in drawing erroneous conclusions. Whether or not there is a correlation between intelligence and liberalism, even if it could be shown to be something more than coincidence, or the result of various confounding cultural factors, in the end it says nothing about either the truth of liberal beliefs or the abilities of your average liberal. Taking it in the best possible light, all it tells us is that at a given time there are more people with high IQs who choose, at the time of the study, who choose to identify themselves as liberal. Beyond that, any other conclusions are just speculation.
* Recently I saw someone who put forth this argument citing a Time article, which is an odd choice, as the reporter actually questions the assumptions, even going so far as to question self-identification, though not going into all the possible issues I describe. At least these studies use IQ, which is a fair representation of actual intellectual capacity (as good as these things get at any rate), rather than other, less trustworthy proxies for intellectual ability.
** And I also know it was not a sign of declining aptitude because (1) aptitude is not supposed to change, being an innate trait and (2) scores on tests where I actually cared, SAT, GRE and LSAT, grew over time, SAT going from 1040 in 6th grade to 1440 in 10th, GRE at age 20 being 800/800/800 and LSAT at the same age being 48 of 48 (old scoring). So I definitely tested well when I cared.
*** It is funny to see them embrace IQ scores in this context, when they were so eager to dismiss them in face of the conclusions of The Bell Curve.
**** For example, German universities in the 19th and early 20th centuries tended to have strong royalist and nationalist tendencies. Most of the Ivy League schools had very conservative, protestant, upper middle class biases throughout the same era. Thus, it would be foolish to look at the modern era and assume liberalism always goes hand in hand with education. Education can be a bastion of free thinking, or of the status quo, and has been both at various times and places.