"While, as I would undertand, no-one has ever determined precisely what it is that IQ tests are measuring, if you turn things around and use them as predictors then they are quite reliable in predicting behaviours and performance. There is an interesting (American) chart called Economic and social correlates of IQ in the USA here."
Bearing in mind your caveat that no statistical analysis is going to predict the outcome for any individual, I gather this seems to be so (see below for exceptions/possible refutations), and wonder why exactly this should be the case. Could it at least partially be a matter of reliably being able to produce the goods on the day and that this consistency in problem solving translates well into the skills required for working environments?
Having been subjected to quite a number of IQ tests during my time (starting from the age of 4), I know full well that IQ results can be completely unreliable. At various times, I have produced scores with a difference of over 60 points, which would not seem particularly useful in terms of prediction. In fact I have just done two on-line tests to see how similiarly I score, and even within a space of 20 mins have managed a difference of some 14 points, which according to this list
140 Top Civil Servants; Professors and Research Scientists.
130 Physicians and Surgeons; Lawyers; Engineers (Civil and Mechanical)
120 School Teachers; Pharmacists; Accountants; Nurses; Stenographers; Managers.
110 Foremen; Clerks; Telephone Operators; Salesmen; Policemen; Electricians.
100+ Machine Operators; Shopkeepers; Butchers; Welders; Sheet Metal Workers.
100- Warehousemen; Carpenters; Cooks and Bakers; Small Farmers; Truck and Van Drivers.
90 Laborers; Gardeners; Upholsterers; Farmhands; Miners; Factory Packers and Sorters.
would mean that my life course would be different. Given my own experience of this variance, I find it hard to take IQ scores too seriously as a reliable predictive mechanism across populations, unless the consistency argument holds.
But on the subject of exceptions/possible refutations to the theory that IQ will, across populations, predict success, the New Scientist this week, in piece entitled "How to be a Genius" states:
"No accepted measure of innate or basic intelligence, whether IQ or other metrics, reliably predicts that a person will develop extraordinary ability. In other words, the IQs of the great would not predict their level of accomplishments, nor would their accomplishments predict their IQs. Studies of chess masters and highly successful artists, scientists and musicians usually find their IQs to above average, typically in the 115 to 130 range, where some 14 per cent of the population reside - impressive enough, but hardly rarefied as their achievements and abilities.
"The converse - that high IQ does not ensure greatness - holds was well". (Study from a selective elementary school quoted).
Instead of IQ, the circumstances that did seem to contribute to the fostering of exceptional talent almost invariably involved the "10 year rule"; in other words, a decade of hard and focused work in order to master something.
"Pete Sampras didn't possess more talent than Andre Agassi, but he won 14 grand slams to Agassi's eight because he worked harder and more steadily. And as cellist Yo-Y0 Ma once said, the most proficient and renowned musicians are not necessarily those who outshone everyone as youths , but rather those who had "fire in the belly".
The 10 year rule of course requires resourses: time and space to work, a mentor, support and fire in the belly - a love of what you are doing."Bloom came to see great talent as less an individual trait than a creation of environment and encouragement. "We were looking for exceptional kids, " he said "and what we found were exceptional conditions". He was intrigued to find that few of the study's subjects had shown speical promise when they first took up the fields they later excelled in, and most harboured no early ambition for stellar achievement. Rather, they awere encourgaged as children in a general way to explore and learn, then supported in more focused ways as they began to develop an area they particularly liked. Another retrospective study, of leading scinetists, similarly found that most came from homes where learning was revered for its own sake."
From the editorial:
The notion that people love doing things because they're good at them is back to front- they're good at them becaue they love doing them and will spend hours practising."
All in all, although the reasons for doing it are entirely different, you can't help feeling that autonomous education is coming out of this pretty well, what with there being an explicit aim to help children foster their interests. Autonomous educators set out to facilitate the interests of their children because they believe it is the morally right thing to do to, not because they want to create a genuius; but the fact that this appears to be the way in which talent is fostered seems to suggest that the epistemology underlying the faciliation of freedom is sound, and that being able to enact the theory that is active in the mind does look to be the most effective way to learn.
There's more in the article about how the talented manipulate information, but I guess I shouldn't really blow the NS's last chance of a subscription.