Nereid
Staff Emeritus
Science Advisor
Gold Member
- 3,392
- 3
Sorry to say that, as reported in the link you posted Nachtwolf, the Minnesota and German studies only very weakly support Jensen's (and hitssquad's, and yours, and Apollo's, and Adam's?) assertion.
By far the biggest problem is the conflating of 'unknown' with 'random'; a second (but still large) problem is identifying the adopted children as 'black' or 'white'
{bold blue text is from http://psycprints.ecs.soton.ac.uk/archive/00000692/ [/color]}
A study with flaws is not unusual; a study which is accepted without a serious attempt to estimate the potential size of systematic errors is worthless.Nachtwolf: To give a more specific analysis:
{the adoption study has several flaws[/color]}
Nachtwolf: Certainly. Scarr & Weinberg admit that no study is perfect.
Ignoring a leading alternative hypothesis is intellectual dishonesty (at least). Translation of Nachtwolf's comment?: I don't care what the alternative hypotheses are, I know I'm right, and won't countenance any dissenting interpretations.{Maternal effects were not considered as a serious possibility for the lower IQ of blacks relative to whites.[/color]}
Nachtwolf: Maybe not.
Er, did you realize that this is, in fact, fatal to your case?{The IQ of parents of adopted children was unknown.[/color]}
Nachtwolf: Indeed.
IIRC, you, hitssquad or Jensen actually stated that this was quite important. Unless these differences are well accounted for, it's likely the study's results say little or nothing wrt your key assertions.{The mean age of adopted infants also differed between treatment groups.[/color]}
Nachtwolf: Just so.
{And most fundamentally, the parents of children were not selected randomly from the population. Infants of parents of varying socioeconomic status and race might give children up for adoption, or have them taken into care, for very different reasons.[/color]}
Nachtwolf: Quite true.
Hmm, and the Minnesota study is the landmark study which most clearly makes the hereditarian IQ case??Nachtwolf: But:
We don't know whether maternal effects were involved - they are a "possibility."
We don't know whether the parents were unusually high or low for IQ - their IQs are "unknown."
We don't know what effect, if any, the (unspecified) mean age difference would have had.
We don't know whether it's true that parents give up kids for different reasons, only that they "might." And we don't know whether this could have any effect - we're just expected to assume that it could.
If Nachtwolf's comments are indicative of how the scientific method is applied in IQ studies, I can only say that there would clearly be prima facie grounds for ignoring the entire field as serious science. Fortunately, reading through the papers in the link I originally posted, I conclude that few of the professionals are so cavalier as Nachtwolf appears to be.
Those are confounding factors which need to be taken into account; they are not reasons to dismiss the German study out of hand (or, if they are, there are equally good reasons for dismissing the Minnesota study).Nachtwolf: But Jensen brings up real, solid points which mitigate the results and interpretation of the German study:
The black fathers of the German kids were smarter than average. This isn't a possibility - it's what we know.
I couldn't believe this when I first read it, and I'm still find it hard to believe Nachtwolf was serious when he wrote it. Jensen himself cites studies which show that groups of 'blacks', geographically, have a 'white' ancestry which ranges from (IIRC) 4% to 40%; he also makes it very clear that within groups of people (such as 'Californian blacks') there is considerable variation in the degree of 'white' ancestry. Further, though he didn't dwell on this (I wonder why not?), the same thing can undoubtedly be said of 'whites' (even for the German mothers; for example, what is their 'Roma' ancestry?).Nachtwolf: Heterosis probably boosted the IQ scores - this isn't just a guess; you can see heterosis showing up in Scarr & Weinberg's study, too.
And there were no fully black kids for comparison, just half-black children - there's no speculation about this, as they all had white mothers.
Yet, without at least some reliable data on the 'blackness' and 'whiteness' of the parents, any conclusions about the 'racial' hereditability of IQ are meaningless. And it's even worse when so much about the parents is 'unknown' (esp in the Minnesota study).
Translation: Nachtwolf doesn't really understand his own case, how hereditability works, why analysis of systematic errors and confounding factors is important, and so on.Nachtwolf: So while there's room for doubt, and probably good reason to do a better follow-up study, the mitigating points brought up to explain the Scarr & Weinberg study are generally... well... vaporous, while the mitigating points brought up against the German Study have some actual substance to them.
Accuracy and attention to detail: 1
Natchwolf: 0
Last edited by a moderator: