By contrast, here’s an article that explores how certain racial groups appear to be more violent, until you control for class and demographic, when they suddenly aren’t.
That is to say, violence and class have common causes. Intelligence, self-control, personality, etc., all immediately come to mind as relevant, but let’s call this cluster of common causes of violence “harmness”. Even though the presence of harmness may screen off the ability of race to contribute to predictions of violence and class, one still observes more harmness in the population of blacks than the population of whites.
For example, regarding heredity of IQ:
I’m sorry, but this test in no way differentiates between the hypotheses of biological heredity of IQ and verbal transference of IQ. As is, this just shows that there’s an attribute which is positively correlated with being talked to by your parents and academic success, and negatively correlated with welfare. To quote the newspaper article:
“One thing is to say we can change adult language behavior,” Suskind said. “Another thing is to show that it is sustainable, and that it impacts child outcomes.”
Even among links which are strongly supported- like breastfeeding raising IQ- there are selection effects that are hard to estimate, because almost all of our evidence is observational. (Breastfeeding is still a good idea for most parents, even if we’re not quite sure how good it is.) If cleverer parents are more likely to breastfeed because of the purported IQ-boosting link, then in order to start separating out the effects of breastfeeding and the effects of heritability we would need to measure the IQs of the parents of both groups, moving from a simple observational study to a massive project (that would still have possible unmeasured selection effects, like the amount they care about health and cleanliness).
do worse than professional-class children entirely because their parents talk to them less.
Emphasis mine. The probability of this conclusion being made by correct reasoning from the evidence supplied is epsilon, and if you didn’t realize that when reading that quote, I beseech you to give that lapse of rationality solemn contemplation.
I can’t be confident it’s more than 2%.
The probability of an unbiased survey of the literature returning this conclusion is epsilon. I suggest updating.
meaning even if you can keep these thoughts separate in your head, they can be dangerous memes to spread, because self-fulfilling prophecies, etc.
I think it is worth considering the mirror of this concern. “Anyone who says X must mean Y” is its own self-fulfilling prophecy, which can only be counteracted by saying X and not meaning Y.
That is to say, violence and class have common causes. Intelligence, self-control, personality, etc., all immediately come to mind as relevant, but let’s call this cluster of common causes of violence “harmness”. Even though the presence of harmness may screen off the ability of race to contribute to predictions of violence and class, one still observes more harmness in the population of blacks than the population of whites.
I’m sorry, but this test in no way differentiates between the hypotheses of biological heredity of IQ and verbal transference of IQ. As is, this just shows that there’s an attribute which is positively correlated with being talked to by your parents and academic success, and negatively correlated with welfare. To quote the newspaper article:
Even among links which are strongly supported- like breastfeeding raising IQ- there are selection effects that are hard to estimate, because almost all of our evidence is observational. (Breastfeeding is still a good idea for most parents, even if we’re not quite sure how good it is.) If cleverer parents are more likely to breastfeed because of the purported IQ-boosting link, then in order to start separating out the effects of breastfeeding and the effects of heritability we would need to measure the IQs of the parents of both groups, moving from a simple observational study to a massive project (that would still have possible unmeasured selection effects, like the amount they care about health and cleanliness).
Emphasis mine. The probability of this conclusion being made by correct reasoning from the evidence supplied is epsilon, and if you didn’t realize that when reading that quote, I beseech you to give that lapse of rationality solemn contemplation.
The probability of an unbiased survey of the literature returning this conclusion is epsilon. I suggest updating.
I think it is worth considering the mirror of this concern. “Anyone who says X must mean Y” is its own self-fulfilling prophecy, which can only be counteracted by saying X and not meaning Y.