Psychologists have long divided IQ variance into three major parts:
- Genes (measured by subtracting the IQ correlation of DZ wins raised together from the correlation of MZ twins raised together and then doubling the result or simply the correlation of MZ twins raised apart)
- Shared environment (measured by subtracting the IQ correlation of siblings raised apart from the correlation between siblings raised together or simply the IQ correlation of unrelated people raised together)
- Non-shared environment (measured from subtracting the IQ correlation of MZ twins raised together from the correlation between the same person tested twice).
One thing both Arthur Jensen and James Flynn agreed on was that the role of genes roughly doubled from about 40% of the variance in early childhood to 80% in later adulthood. To make room for this doubling, the role of shared environment shrank from roughly 40% in early childhood to 0% in later adulthood.
Meanwhile the effect of non-shared environment remains constant at 20% throughout the life span.
Flynn believed non-shared environment was just luck, and since luck by definition is random, it can’t favor adults or children on average which explains why it stays 20% while the other sources of variance rise and fall. Flynn also implied (though I don’t want to take him too literally) that non-shared environment was “free will”. By this I assume he meant, we don’t choose our genes or the homes we are raised in, but that 20% IQ variance that remains is up to us.
Jensen also believed non-shared environment was luck but he viewed it as specifically biological luck, writing:
a large part of the specific environmental variance appears to be due to the additive effects of a large
number of more or less random and largely physical events— developmental “ noise” — with small, but variable positive and negative influences on the neurophysiological substrate of mental growth
Now what’s interesting is Jensen not only believed that 80% heritability applied to differences within U.S. races, but also to differences between them. He called this the default hypothesis because if we don’t know what’s causing the IQ difference between races, the default should be that it’s the same cause as within them: Occam’s razor.
In 2006, Rushton & Jensen wrote:
Shuey’s (1966) compendium to document that the average Black-White difference was 0.70 standard deviations in early childhood, 1.00 standard deviations in middle childhood, and 1.20 standard deviations in early adulthood….Until the results of several such studies allow reassessment of the situation, the best estimate of Black-White convergence over the past 100 years is between 0 and 3.44 IQ points – a maximum effect size of 0.23 – well within the predictions of our estimated heritability of .80 for the Black-White g difference in the United States
In would have been nice if Rushton and Jensen had shown their math. Maybe their logic was, if 80% of the IQ variance is genetic, and 20% is non-shared environment, and the square root of 20% is 0.45, then we can at most expect the black-white IQ gap to shrink by 45%. So if the gap was originally 1.1 standard deviations (17 IQ points) in WWI, the most it could shrink by would be 1.1SD(0.45) = 0.5 SD or 9 points.
On the other hand, if genes explain 80% and the square root of that is 0.89, then the black-white gap should never shrink to less than 0.89(1.1 SD) = 0.98 SD = 15 points.
Something’s not adding up.
But there’s another problem with Jensen’s elegant default hypothesis. If non-shared environment is luck as Flynn and Jensen implied, how can it differ at all between races? Luck by definition is random so an entire race can not be lucky or unlucky. Rather the lucky and unlucky members of both races should cancel each other out leaving the non-shared environment equal and genes as the only cause of the racial IQ gap.
This seems to fit the data better.