On page 292 of the book The g Factor, scholar Arthur Jensen writes:

There is a high degree of agreement among people when they are asked to rank occupation titles according to their impression of (1) the occupation’s socially perceived prestige, (2) the desirability (for whatever reason) of being employed in the occupation, and (3) the estimated level of “intelligence” needed to succeed in the occupation. When a number of people are asked to rank a large number of different occupations titles, from highest to lowest, on each of these standards, the mean rank for each occupation remains fairly constant.  The over-all rank order-correlations in various studies fall between .95 and .98.   The high consistency of rank order holds up across rankings by people from different occupations, social class backgrounds, industrialized countries, and generations.

From a Darwinian perspective, intelligence can be defined as the cognitive ability to adapt: to take whatever situation you’re in, and turn it around to your advantage.  People who spend their lives working in a good job, a job that others admire and wish they had, have generally turned their situation to their advantage.  By contrast, one who spends his life working in a job that others look down on and would never want to do, have generally failed to adapt, assuming one has the same goals as most people (and by definition most people do). Thus, one should expect a positive correlation between IQ and occupational status.

On page 293 Jensen writes:

The correlation of individuals’ IQs with occupational rank increases with age, ranging from 0.50 for young persons to about 0.70 for middle aged and older persons, whose career lines by then are well established.  

The relation of IQ to occupational level is not at all caused by differences in individuals’ IQs being determined by the differences in amount of education or the particular intellectual demands associated with different occupations.  This is proved by the fact that IQ, even when measured in childhood, is correlated about 0.70 with occupational level in later adulthood.

A 0.70 correlation is absolutely massive, especially by the standards of the social sciences. It’s roughly the correlation between IQ as measured in childhood and IQ as measured in adulthood.  If what Jensen is saying is true, one could make a very strong case that IQ is destiny.  The idea that a test given to a child could predict with that much accuracy, their place in society 40 years later is absolutely fascinating.  It’s almost as if life is a valid IQ test, and occupation is your score.

As a kid, an Indian woman in a sari drove from far, far away, to get to my school, just to test me, on the most expensive, in-depth IQ test the school board had.  A test so closely guarded and so carefully scored, no teacher was allowed to give it.  At the time I thought of Indian women in saris as fortune tellers because I would see one reading palms on downtown streets, so I remembered thinking of the school board Indian woman who asked me to play with blocks, jig-saw puzzles, and cartoon pictures that resembled tarot cards, as a fortune-teller too.  I was more right than I knew.  Childhood IQ predicts your future with alarming accuracy.

On the other hand, the correlation between IQ (both in childhood and adulthood) and older adult occupation sounds too high to be true, especially if the correlation could be forced to fit a bivariate normal distribution and extended to the extremes.

For example, there are 76.4 million baby boomers.  That means the baby-boomer with the best job is +5.53 Standard Deviations above the mean of a normalized occupation distribution.  This implies an IQ that is 0.7(5.53) = 3.87 SD above the mean: IQ 158, a truly jaw-dropping figure.

So who has the best job in America?  Apparently it’s not the president since White House occupants seem to average IQs around 130.  Maybe this is a case where the linear models I love so much, simply don’t work, no matter how much I want them to.

It’s also a bit strange that childhood IQ would correlate as much with older adult occupation as adult IQ itself does.  I hope this isn’t because socioeconomic background is influencing both childhood IQ and later adult occupation.

Advertisements