Jason Malloy on P.Z. Myers

This is strange; you’re doing the same thing here that people at this forum have done both of the other times I’ve debated about this topic here. The thread was dead for a week, so I stopped checking it, and then a week later you revived it with a new response to one of my posts. I wish I understood what keeps causing this kind of delay when people are replying to me about this topic.

I’ve read the entirety of what he said in that thread, as well as a few other threads where he’s made similarly unsupported claims, but I don’t read his blog consistently. So if you’re right that these various claims he makes do not accurately describe his position, then there’s a different problem: he’s basically creating a strawman of himself by making claims that present a caricature of his own position.

I’ve never heard of anyone doing that before, but I don’t see how you can consider it not a problem. How would it be possible to have any kind of debate with him if his posts themselves are strawmen of his position, and it will be attacking a strawman if you respond directly to any of them?

About the adoption study: since I haven’t seen the study itself, I’m not able to respond to this in detail other than to say what I’ve said before. Which is that for any one of these studies, it’s usually possible to come up with a completely plausible environmental explanation, so the real question is why so many unrelated environmental factors would all produce the same result.

I’m currently reading The g Factor by Arthur Jensen, which seems to be generally regarded as the best book that exists about this topic, so that’s something I definitely recommend reading if you’re going to be looking into race differences in IQ.

One thing I’ve learned about from this book is a line of data I hadn’t been aware of before, that I think it’s important for you to cover if you’re going to discussing this topic: Regression toward the mean.

In genetics, regression toward the mean refers to the fact that when the genes of one individual place it far from the mean of its population—in terms of size, color, or any quantifiable trait—then statistically, this trait in any individual which shares 50% of the first individual’s genes will tend to be halfway between that of the first individual and the population mean. So for example, I’m 6 feet 4 inches tall, while the average height for men is 5 feet 10 inches. So if I had a brother, the statistical most likely height for him to have would be 6 feet 1 inch, since he has 50% of my genes, and this is halfway between my own height and the population mean. For any genetic trait, the larger a sample you take, the more exactly their siblings will regress exactly 50% towards the mean.

On the other hand, with purely environmental traits, if you take a sampling of a large enough group, regression towards the mean will be zero. The environment for siblings reared together certainly isn’t identical, but when you’re taking an average of a group where the individuals furthest from the mean includes both the oldest and youngest sibling, as well as both possibilities in every other area where environmental influences for siblings reared together can differ, all the individual differences tend to cancel each other out. So among all populations that reproduce sexually, the extent to which siblings regress towards the mean is generally considered a reliable measurement of the extent to which a trait is genetic—50% for a purely genetic trait, 0% for a purely environmental trait, and for anything in between the two, the extent to which siblings regress towards the mean varies linearly with the extent that the trait is genetic.

Jensen’s hypothesis, that the difference in IQ between races is a little over 50% genetic, predicts that the extent to which IQ regresses towards the mean in both populations would be between 25% and 50%. Page 470 of The g Factor describes a study that was done of IQ test results in a California school district, covering around 900 white sibling pairs and 500 black sibling pairs, which found the regression towards their population mean to be 40% for the white siblings and 38% for the black siblings. So when a white student had an IQ of 120, their sibling would, on average, tend to have an IQ of 113. Whereas when a black student had an IQ of 120, their sibling would on average have an IQ of 99.

This is a specific prediction made by Jensen’s hypothesis which has been tested and shown to be correct. The 100% environmental hypothesis would not have predicted this, and I’m not sure if it’s even possible to come up with an explanation for it that doesn’t involve genetic factors. Using the standard method from genetics by which regression towards the mean can be used to measure the extent to which genes are responsible for any trait, the extent to which genes are responsible for the average IQ of African-Americans is 2 x 38%; that is, 76% genetic.

Since this line of data is based on a commonly-accepted way of measuring the extent to which any trait is genetic, which is considered reliable for all organisms which reproduce sexually, I think this is the most compelling evidence for the IQ difference being genetic that I’ve discovered thus far. So if you’re going to be covering this topic in particular, it’s important that you include this topic also.

The g Factor also cites the evidence of transracial adoption studies. Three studies have been carried out on Korean and Vietnamese children adopted into White American and White Belgian homes. Though many had been hospitalized for malnutrition, prior to adoption, they went on to develop IQs ten or more points higher than their adoptive national norms. By contrast, Black and Mixed-Race (Black-White) children adopted into White middle-class families typically perform at a lower level than similarly adopted White children. In the well known Minnesota Transracial Adoption Study, by age 17, adopted children with two White biological parents had an average IQ of 106, adopted children with one Black and one White biological parent averaged an IQ of 99, and adopted children with two Black biological parents had an average IQ of 89.

Not even in the threads you gave in the beginning was his position that the only determining factor, or even the most important factor. You jump on a single statement he made about SES when even in the very thread you jump on he never makes a statement about it's relative importance or that it is the only factor.

Because the factors are related to each other and because different study designs will give a different relative importance. If you look at the results of the interracial adoption studies and other studies, the different factors are related to African American culture. Worse adoption experiences, identification with African American culture, lower quality of education, conscious and unconscious effects of discrimination, caste-effect. All these tie into the complex that is formed by the living circumstances for different races both in the USA and the rest of the world.

I also see that you ignore the fact that the IQ-figures of Asian/Indian adoptees do not support the genetic hypothesis in the slightest. Basically, you are completely ignoring the actual results of the adoption studies and jump onto another topic. You wanted to talk science, do it.

I'll get back to you on this later. What were the other factors taken into account in the study with the California school district?

Also, is this indeed a part of the argument in Jensens book?

[…]

If so, does the term "thin ice" ring a bell with you?

Evolutionary scenarios must be evaluated much in the same way that jury members must judge a prosecutor’s narrative. Ultimately they must make their judgment not on the basis of any single fact or observation, but on the totality of the totality of the available evidence. Rarely will any single item of evidence prove pivotal in determining whether a prosecutor’s scenario or the defense’s alternative is most likely to be correct. Many single details may actually fail to favor one scenario over another. The most probably account, instead, is the one which is the most internally consistent—the one in which all the facts mesh together most neatly with one another and with the motives in the case. Of paramount importance is the economy of explanation. There are always alternative explanations of any single isolated fact. The greater the number of special explanations required in a narrative, however, the less probably its accuracy. An effective scenario almost always has a compelling facility to explain a chain of facts with a minimum of such special explanations. Instead the pieces of the puzzle should fall into place.

As I stated before, the real issue here is that he stated that SES is a more important cause of IQ than IQ is of SES, when the evidence shows that the reverse is true. It’s not a matter of it being the “only” cause, but the relative strength of this particular causality compared to its opposite.

Except that these do not explain the results for the rest of the world, or even all of the results in the United States. In countries such as Kenya where Africans are a majority, there is no discrimination against them, caste effect, or most of the other things you mentioned. Yet the IQ difference there between Africans and Caucasians is even greater than in the United States. And in the USA, differences in education or culture would not affect something as basic as reaction time, which is primarily a function of synapse speed. It could be affected by something like malnutrition, but as with most of these tests, the difference still exists even when SES and education are equal.

So to say that all of these results can be explained by the network of factors you described is an oversimplification. And since the APA has stated that there is no known environmental explanation for the IQ differences, they evidently agree with me on this point.

You’re asking me to comment on this study when I haven’t actually seen it, which is kind of difficult.

The fact that Asian and Indian children had lower scores than the white children is obviously because of an environmental influence, but it might be an environmental influence that didn’t exist for the black families. According to what you quoted, the Asian adoptees suffered from malnutrition prior to their adoption, which would have depressed their IQ; was this true of the African children also? If not, then this is a confounding factor for the Asian/Indian scores, but it isn’t evidence against a genetic basis for the black/white difference.

There are also other things you’re describing about this that don’t seem to make sense. The main thing I’m wondering is, was the adoptive experience of the half-black children who were considered to have two black parents basically the same as that of the half-black children where it was known that they were of mixed ancestry? Their average IQ was the same, so that’s the only way this can be accounted for with a 100% environmental explanation. But it seems very implausible that people would have treated these kids in accordance with their actual ancestry, and not in accordance with what their ancestry was assumed to be. How can you treat a person differently based on something you don’t even know?

The study also looked at SES, although since it was comparing siblings within the same families, differences in SES wouldn’t have depressed the scores of the black families. The main thing it noted related to this was that the higher the SES of the family was, the stronger the regression towards the mean tended to be. Since one would expect environmental influences on IQ to be greater at lower SES, this trend is the opposite of what one would expect to see if the difference were purely environmental.

There are a few other details of this data that Jensen points out, but it’s probably easier if you can just read the book yourself.

The book talks about the transracial adoption studies also, but they aren’t really the emphasis; the explanation of them occupies around 5 pages of the 650-page book. His main point about race and intelligence is fairly similar to mine: that when considering all of the data that exists about this topic, an explanation involving genetic factors is more parsimonious than any of the alternatives. The basis by which he considers a genetic explanation most likely is best summarized by a quote he includes from the anthropologist Owen Lovejoy:

Evolutionary scenarios must be evaluated much in the same way that jury members must judge a prosecutor’s narrative. Ultimately they must make their judgment not on the basis of any single fact or observation, but on the totality of the totality of the available evidence. Rarely will any single item of evidence prove pivotal in determining whether a prosecutor’s scenario or the defense’s alternative is most likely to be correct. Many single details may actually fail to favor one scenario over another. The most probably account, instead, is the one which is the most internally consistent—the one in which all the facts mesh together most neatly with one another and with the motives in the case. Of paramount importance is the economy of explanation. There are always alternative explanations of any single isolated fact. The greater the number of special explanations required in a narrative, however, the less probably its accuracy. An effective scenario almost always has a compelling facility to explain a chain of facts with a minimum of such special explanations. Instead the pieces of the puzzle should fall into place.