Ethnic origin vs. intelligence

The data from the 1997 cohort of the National Longitudinal Survey of Youth include ASVAB scores for a large sample of American youth born between 1980 and 1985, as well as the youths' answers to the following question:
What is your origin or descent?
You may choose as many as three answers in the order with which you identify with them most. For example an Irish-Italian might identify most with the Irish and second most with the Italian.

Here are median (mean) ASVAB Math/Verbal score percentiles for various (first listed) origin groups:
Russian 83.625 (72.724)
British 77.916 (71.252)
Danish 75.19 (61.958)
Polish 72.194 (67.316)
Swedish 71.638 (63.886)
Scottish 71.083 (63.875)
Welsh 71.068 (65.105)
Dutch 68.623 (64.306)
Irish 67.978 (62.7)
Norwegian 67.864 (64.519)
German 65.08 (61.711)
French 62.016 (58.34)
Italian 59.912 (57.836)
Greek 56.368 (52.965)
Portuguese 51.088 (57.063)
Cuban 35.621 (39.861)
Mexican-American 30.107 (36.019)
Black 26.696 (32.532)
Mexican 24.822 (30.5)
[Note: I've limited the sample to respondents identified as non-Hispanic and "White", apart from the last four groups listed.]

I assume most of the "Russians" and a substantial fraction of the "Polish" are Jews. I plan to do more analysis of this data in the next week or two. NLSY97 contains much more background information, including questions about religion, as well as parent interviews which contain questions about the parents' ancestry and religion. Other test scores are included, as well.

Data on which of four census regions (Northeast, North Central, South, and West) the respondent lives in are also available. The single most intelligent region-ethnicity group seems to be British-identified Northeasterners, whose median ASVAB percentile is 91.543 (n=23). Next highest is "Russians" in the North Central region, at 89.556 (n of only 7).

The overall trend appears consistent with patterns of achievement in America identified by Nationiel Weyl in the mid-1980s.
The most basic trend is that the people of primarily Germanic stock (Germans, Scandinavians, Dutch, British and Irish) tend to outperform the Latins, the Slavs, and, to a lesser extent, the Hungarians. Considering the seven core indices as a whole, these Germanic peoples are clustered together with performance coefficients ranging from 90 to 117. The French, Italians, and Slavs range in the 60s, or at one-half to two-thirds the Germanic level.

[. . .] When we consider the differences within the Germanic majority stocks, several interesting profiles emerge. Among the Scandinavians, the Swedes are generally, though not always, in the leadership and Norwegians rank below the other two groups in achievement.

[Nathaniel Weyl. The Geography of American Achievement. pp. 23-24]

17 comments:

Anonymous said...

The Armed Services Vocational Aptitude Battery, or ASVAB, was created by the Department of Defense to help in the recruitment and placement process with the U.S. armed forces. Anyone who is serious about joining the military needs to understand what this test consists of, how it is measured and used and how you can prepare for it to ensure the best score possible.

Your conclusions are BO-GUS! Way too many reasons to list, as if you do not already know. And if you don't, then you have no fucking business putting shit like this up.

Anonymous said...

BO GUS: Elaborate or shut the fuck up. I think it will become clear very quickly who has no business posting on the topic.

dudhduddhd said...

I was going to post about the same thing but thought better of it since it is your blog and you seem perfectly capable.

I did want to drop a note of appreciation to you though as well and here it is: fantastic blog, keep up the good work

Lastly, I enjoy your comments at Molbug's site. It is nice to see someone call him out.

Anonymous said...

Elaborate or shut the fuck up.

Pffft. What a surprise that your range of knowledge varies between fuck all and jack shit.

Listen dipshit, the cohort involved less than 9000 kids. In your "expert" opinion, is this an adequate sample to make generalizations about "Ethnic origin and intelligence"? Do you know the representative ethnic breakdown? Do you know the regions where these different ethnics came from? City or country and who came from where? What types of schools they came from? Inner city versus suburb? Were they in single or double parent homes? Poor? Rich? Disabled? Did they take this study seriously? Who did and who didn't? How were the surveyors able to select out any pitfalls and biases? Is a longitudinal survey accurate? Is it more accurate than a blinded study?

Just a SMALL sampling of how little you fucking thought this through. Now why don't you STFU.

Anonymous said...

Anonymous;
HMMM....
You run out a good set of questions. But the data from the 1997 cohort of the National Longitudinal Survey of Youth include ASVAB scores for a large sample of American youth born between 1980 and 1985.
One can imagine that a survey with this name did adequate sampling in the manner you suggest. Also such surveys usually need to be peer reviewed for funding.
Your knowledge of survey techniques is in contrast to your knowledge of statistics. 9000 is quite a large survey. Again funding and peer review would require that their sample size be demonstrated to be adequate to get meaningful results.

Anonymous said...

9000 is quite a large survey.

Yes but it's questionable whether it's large enough to draw accurate conclusions about the smaller ethnic groups which might only constitute one or two percent of the sample.

Still, the results are consistent with what one would expect, so no cause for alarm there.

Longitudinal studies aren't ideal for assessing the intelligence of various groups but if the data is there obviously you'd look at it.

Anonymous said...

As expected, bo-gus -- whose hysterical (non-)argumentation bears striking resemblance to koob-lie-con's -- has nothing.

If it wasn't clear in the post, I regard these numbers as preliminary. But sample size isn't the issue.

The number of respondents listing major European origins first:

British 154
Dutch 111
French 163
German 685
Polish 108
Italian 224
Irish 454

As for representativeness: The survey sample is designed to represent U.S. residents in 1997 who were born during the years 1980 through 1984.

This is as large and representative a sample as is ever likely to be assembled of U.S. subjects in this age range, with ethnicity and test score data.

(Note: The one legitimate problem I see with these data is that origins are self-reported by the youth, and about 1/4 of the ASVAB test-takers don't list any ethnicity; those listing "Other" and "None of the above" score lower on average than those listing an origin. My guess is those not listing an origin are disproportionately Northern European. My guess is that absent this effect, the magnitude of intra-European differences would be reduced, though the directions would remain essentially the same. Spot checks using information reported by parents tend to confirm this.)

danielj,

Thanks for the comments.

Anonymous said...

"The most basic trend is that the people of primarily Germanic stock (Germans, Scandinavians, Dutch, British and Irish)..."

How come the Irish are considered to be of primarily Germanic stock? Rather than Celtic?

Any ideas?

Thanks.

Anonymous said...

As expected, bo-gus ...has nothing.

I'd do a "LOL" about now, though this deserves more of a "Guffaw".

The numbers are not a problem, huh?

The number of respondents listing major European origins first:

British 154
Dutch 111
French 163
German 685
Polish 108
Italian 224
Irish 454


So, nothing about those numbers raises any questions in your "big brain"? Let's ignore the heavy German Irish slant for now and focus on the total numbers where you felt enough confidence to title this worthless post, "Ethnic origin vs intelligence"? LOL You felt the numbers were large enough to make such broad generalizations, right? What if the 154 British first respondents above listed South Asian/Indian as second, or worse, Negro? Would that have any effect on your stupid thoughts? Do you know that breakdown?

Give it up, douchebag. You've been punk'd!

Longitudinal studies aren't ideal for assessing the intelligence of various groups but if the data is there obviously you'd look at it.

Looking at it and making broad, unfounded generalizations are two separate issues. Something our dear host has not yet grasped.

Polak said...

"I assume most of the "Russians" and a substantial fraction of the "Polish" are Jews."

There's no reason to assume that, especially for the Poles.

They're all Catholic and ethnically Polish.

Shalmaneser said...

Interesting site. I've a few questions though:

A.) Echoing Polak: why do you assume that the Russians and Polish in the sample are Jews?

B.) Haven't the claims of Jewish intellect been exaggerated?

C.) How can Irish be considered Germanic (I know there was Irish-Scandinavian intermarriage but that goes back 1000 or more years)?

D.) For the sake of argument, we'll accept your conclusion that Germanic people are elite of Europe. But if so, of what practical significance is that (i.e., to the problems of the diverse Euro peoples today)?

n/a said...

Polak,

They're all Catholic and ethnically Polish.

No. I just checked. Looking at only one parent's (the mother's in most cases) answer to which religion they were raised in, at least 5/92 of the "Poles" have definite Jewish connections. 63/92 of the parents list Catholic.

The corresponding numbers for "Russians" are 10/36 Jewish and 0/36 Eastern Orthodox.

Shalmaneser,

why do you assume that the Russians and Polish in the sample are Jews?

It's common knowledge that U.S. "Russians" are substantially Jewish. As for Poles, at least at the time Weyl published The Geography of American Achievement, Slavs appeared to be underperforming NW European Americans, and I'm aware of no evidence that has changed in the past two decades.

Haven't the claims of Jewish intellect been exaggerated?

Yes: "115" is an utterly baseless exaggeration. Average Jewish IQ in English-speaking countries is probably around 107-8. That still likely makes them the single most intelligent ethnic group (though not the most intelligent religious group).

How can Irish be considered Germanic (I know there was Irish-Scandinavian intermarriage but that goes back 1000 or more years)?

I wouldn't describe Irish Catholics as "Germanic" (I'm not sure Weyl meant to either), but they are certainly Northwestern European, and more closely related to Germanic than to other Europeans (see this and any number of autosomal studies). Irish Protestants tend to be Germanic to one degree or another.

For the sake of argument, we'll accept your conclusion that Germanic people are elite of Europe. But if so, of what practical significance is that

For one thing, it (as well as any number of other reasons) should give pause to NW Euros who imagine it's a good idea to dissolve all distinctions among Europeans and merge the various "diverse Euro peoples".

Anonymous said...

I'm curious about something. Maybe you can answer it for me. Why did the youths who identified themselves as German do so poorly compared to so many other Euro ethnics? I seem to recall reading an article by Prof. Richard Lynn that stated ethnic Germans had the highest IQs among Europeans. Shouldn't they be at or close to the top of the list? Any thoughts?

Polak said...

So how many of the 5 Poles with Jewish "connections" are actually Jews?

Seems like this Polish sample is quite Polish.

n/a said...

Why did the youths who identified themselves as German do so poorly compared to so many other Euro ethnics?

I don't know. I could imagine it being a result of differential migration and/or an artifact of the youths being asked to pick which group they identify with (if, e.g., the less intelligent are more frequently only aware of their recent immigrant ancestry and ancestry from large countries).

I'd hoped to minimize the second possibility by looking at parental responses. That analysis looks like it will be a bit more involved than I initially anticipated, but I should get around to it eventually.

Polak,

That's a tough one. What percent of Americans with mothers who were raised Jewish would you say are Jewish?

Anonymous said...

No one identifying as American? I know in Australia there are lots of people with an English parent or grandparent or some other nationality like Irish yet they would never identify as English or Irish. Maybe Americans are queer in that regard.

Why 1997? So long ago.

Anonymous said...

@Ponto
As an Australian I've always found interesting the need Americans that have to publicly advertise the symbols of their ancestry, even if authentic practices are no longer a part of their everyday life. Perhaps its because there's much more division in the US, and those without discernible roots need to dig deeper to feel a part of an ethnic club.