« I thank Allah I was born a man | Gene Expression Front Page | New Mx script/data library made available »
September 30, 2003


In comments on a previous post, I mentioned that I had seen a study of the IQ of immigrants to the Netherlands.

Here is the full reference:

Jan te Nijenhuis and Henk van der Flier: ‘Group differences in mean intelligence for the Dutch and Third World immigrants’, Journal of Biosocial Science, 33 (2001), 469-475.

The authors’ Summary is as follows:

“Evidence from eleven samples indicates that the mean IQ of third world immigrants in the Netherlands is lower than the Dutch mean by approximately one standard deviation for Surinamese and Antillians, and by approximately one and a half standard deviations for Turks and Moroccans. Since IQ tests provide the best prediction of success in school and organizations, it could be that the immigrants’ lower mean IQ is an important factor in their low status on the Dutch labour market. The IQs of second-generation immigrants are rising.”

For more details......

The article is a review of Dutch studies of immigrant IQs. It is stated that ‘only studies of acceptable methodological quality were included in the review’. Unfortunately not much is said about the tests used, which are identified by their (Dutch?) acronyms, such as ‘RAKIT’, ‘GATB’, or ‘DAT’. If any Dutch readers know what these are, please tell!

The four immigrant groups studied are Turks, Moroccans, Surinamese, and Antillians. You all know who Turks and Moroccans are, but Surinamese are from the South American former Dutch colony of Surinam (Dutch Guiana), and Antillians are from the Dutch Antilles islands in the Caribbean. The Surinamese and Antillians are predominantly black, with maybe a touch of white and Amerindian ancestry.

It is not explictly stated whether mixed-race individuals are excluded from the studies. Clearly it would be misleading to describe them simply as ‘Turkish’, etc., if they are Turkish-Dutch. If anyone has access to the original Dutch studies maybe they could check?

The data from the various studies are given in a table. There are a few studies of adults, and more of children. The children’s studies are classified as ‘1st generation’, ‘2nd generation’, or ‘Mixed generations’.

I give the key results below. A = Antillians, S = Surinamese, T = Turkish, M = Moroccan. In some of the studies Surinamese and Antillians are grouped together, and in some Turks and Moroccans are grouped together. I have omitted two studies which cover ‘various’ immigrant groups. [For those who consult the original table, I have assumed that the study by van der Vijver covers Surinamese and Antillians, like the one immediately above it in the list.]

You can all group and average the data however you wish. I have calculated averages for Surinamese and Antillians together, and for Turks and Moroccans together, if only because many of the raw data are grouped in this way. For children, I have averaged 2nd generation and mixed generations combined, as there are so few studies which distinguish 2nd generation as such.

So here are the data. The mean IQ of the samples is stated in standard deviations below the Dutch mean:

Surinamese and Antillians
S: 1.05, 1.08. A: 1.17. Average: 1.1
Turks and Moroccans
T: 1.43 M: 1.86 Average: 1.5

Children (1st generation)
Surinamese and Antillians
S: 0.93, S & A: 1.09 Average: 1.01
Turks and Moroccans
T: 1.45. M: 1.70. T & M: 1.13. Average: 1.43

Children (2nd generation and mixed generations)
Surinamese and Antillians
S & A: 0.77 (2nd gen.), 0.81 (2nd gen.), 0.22 (mix), 0.67 (mix), S: 0.70 (mix),
A: 0.21 (mix) Average: 0.56
Turks and Moroccans
T: 1.20 (2nd gen.), 1.16 (mix), 1.47 (mix), 0.60 (mix) M: 1.43 (2nd gen.), 0.82 (mix), 1.38 (mix), 0.79 (mix). T & M: 0.89 (mix) Average: 1.08.

It will be seen that Turks and Moroccans perform considerably worse than Surinamese and Antillians. Presumably no-one will argue from this that Turks and Moroccans are innately less intelligent than Surinamese and Antillians. The obvious fact is that Surinam and the Antilles are former Dutch colonies, and their people presumably speak reasonable Dutch, while the Turks and Moroccans wouldn’t know Dutch from douche. Both groups perform better after the first generation, as the authors point out.

It is odd that the studies do not cover people of Indonesian origin. As anyone familiar with the Netherlands will know, there is a large Dutch Indonesian community, especially from the Moluccas. Te Nijenhuis and van der Flier say that data on the IQs of various generations of Moluccans are not available. This is surprising.

Posted by David B at 04:05 AM

Below is a rough translation of information on the RAKIT test, found on www.plato.caiw.nl
The parts marked with [?] are ambiguous in the original Dutch.

"RAKIT refers to Revisie Amsterdamse Kinder Intelligentie Test (Revised Amsterdam Intelligence Test for Children). The tests are intended for children aged 4-11 yrs, are conducted individually, and take about 2.5 hours. RAKIT consists of 12 subtests. For the youngest age bracket (4-5), the subtests measure: verbal learning ability and verbal fluency, spatial-perceptual reasoning, sequential memory and quantity [?]. For the 5-11 yr bracket: perceptual reasoning, verbal learning ability, spatial orientation, tempo [?] and verbal fluency. […] The raw sub-test scores may be converted to standardized scores."

Posted by: Syarikat at September 30, 2003 05:41 AM

The lack of data on "Indonesian" Dutch is partly due to the policy definitions used for ethnic minorities in the Netherlands. Because the notion of "racial" classifications is tainted because of the sad legacy of World War II, when the Nazis had no trouble at all pinpointing Dutch Jews because of the detailed civil records, the government had to come up with a way of singling out (ethnic) groups which demanded special attention (such as affirmative action programs, which have taken a rather different approach than in the US, for instance). So "ethnic minorities" called "allochtonen" in the PC speak of the last few decades, are defined on the basis of the country of birth of the parents (needless to say, this sometimes produces odd results as some white persons may have been born in Surinam, for instance, and thus officially counted as "allochtoon" for some purposes.
The arrival of "Indonesian" Dutch (actually, most of them were Eurasians, also known as "Indische Nederlanders" or "Indo's" who often felt more akin to the Dutch than to the Indonesians while they were still in the colony) and Moluccans took place in the early 1950s. The "Indische Nederlanders" were usually fluent in Dutch and relatively well-educated. Although some second and third generation Indische Nederlanders are culturally drawn to Indonesia, they are on average very well integrated into the Dutch mainstream. The Moluccans were a more difficult (but much smaller) group. First generation Moluccans were often not fluent in Dutch, and resentment of Dutch decolonization policy made them a volatile group. However, time has removed most of the sharp edges by now, nowadays young Moluccans speak Dutch as well as Malay, most of them have jobs and have become political realists. Not unimportantly, most of them are Christians.
Turks and Moroccans are overwhelmingly Muslim, have arrived more recently and in much greater numbers, and have more language difficulties. However, the young Moroccans who are perceived as difficult in Dutch society are usually fluent in Dutch, so that is no longer the main problem. (What IS, is a topic which would take too long to discuss here).
Anyway, "Indonesian" Dutch are absent from the statistics because of the way the criteria for being considered "allochtoon" were formulated. These criteria were formulated intentionally so as to include "difficult" ethnic groups without using overtly "racial" or "ethnic" criteria. "Indische Nederlanders" are not considered especially "difficult"; and most of them, or their parents/grandparents, have arrived to long ago for the "country of birth" criterion to apply.

Posted by: Syarikat at September 30, 2003 06:09 AM

GATB = General Aptitude Test Battery
Quote from

"The General Aptitude Test Battery (GATB) was designed to measure the work-related skills of applicants for white-collar civil service jobs in the U.S. government. It consists of nine tests of aptitude, including general ability, verbal aptitude, numerical aptitude, and other occupationally related aptitudes such as clerical perception and manual dexterity. Of the nine, general ability and verbal aptitude are the most closely related to the GED."

DAT = Differential Aptitude Test

provides more details on the composition of this test.

Posted by: Syarikat at September 30, 2003 06:59 AM

1 immigrant minorities are not necessarily representative of their countries of origin. for example Black Africans in the UK tend to be okay in terms of education and income. in terms of Fluid IQ they are not nearly as bad as African Negros or as bas as Black Americans (90 v 70,85).

2 along the same lines immigrants from a country are not necessarily representative of that countries ethnic break down. for example over 70% of Arab Americans are Christian. Dutch Antillies is about 85% Creole (Mixed Black) with the remainder mostly Amerindians, East Asians and Whites. Dutch Guyana is over 1/3 South Asian, about 1/3 Mixed Black, 1/6 Indonesian and the 1/10 Negro with some Chinese and Whites. i would guess that from these countries non-Negros would be more likely to leave than the Negros.

3 David B's Dutch language theory is not consistent with the fact that Antillians and Guyanese improve more across the generations than Turks and Moroccans.

Posted by: Acheson at September 30, 2003 01:14 PM

Speaking of Dutch-Indonesian blends, the best-known in the U.S. are the Van Halen brothers. I've never met them, but I know their sons Aric (Alex's kid) and Atilla (Eddie's kid). Aric is small and blond and Atilla is small, dark, and sensitive looking, with his mother Valerie Bertinelli's big dark eyes. I wish him luck in life -- he's going to need it being named Atilla Van Halen!

Posted by: Steve Sailer at September 30, 2003 03:18 PM

that dude who played zack morris in "saved by the bell" is also indonesian-dutch...and know I feel he does look a tad ethnic...of course when i was a kid i thought he was a WASP's wasp :)

Posted by: akshay at September 30, 2003 03:22 PM

Syarikat: thanks for the in-depth briefing!

Acheson: interesting info on the racial makeup of Surinames and Antillians. The authors of the article themselves describe them as 'predominantly Black'.

- also a fair point about the relative change (or lack of change) over time. I don't know whether the Turkish and Moroccan 2nd-gen children are entirely fluent in Dutch - in Britain some Muslim communities (especially Bangladeshis) tend to speak their own language to the children and don't let them mix much with the 'natives', so the kids still start school with a serious language handicap. (Though this is changing.)

Of course, I agree that immigrants are not a random sample of their population of origin. In general I would expect them to be somewhat superior in IQ, if only because the very-low-IQ people are more likely to be left behind.

Posted by: David B at October 1, 2003 01:37 PM

"Antillies is about 85% Creole (Mixed Black)"

Most ordinary Dutch, who are only vaguely aware of the ethnic intricacies of their former colonies and remaining overseas dependencies, would consider the “mixed black” group simpy “black”. The presence of the ore “purely” black group found in Surinam (the Maroons or “Bosnegers”, descendants of escaped slaves) is negligible in the Netherlands.

"I would guess that from these countries non-Negros would be more likely to leave than the Negros."

Not necessarily, because the East Asians, Chinese (and also the numerically insignificant whites) are economically dominant, so there is somewhat less incentive for them to leave. While is a significant number East Asians (“Hindostanen”) in the Netherlands, they have the strongest presence in The Hague, while most of Surinam’s Creoles are concentrated in Amsterdam and Rotterdam. The former group is socio-economically relatively well-off and well-educated, so it doesn’t often appear in publications designed to deal with "problematic groups" (the same goes for groups like the Chinese, who mostly keep to themselves socially, but are successful economically and educationally and are virtually absent from Dutch crime statistics). When speaking of “Surinamers” most Dutch people primarily think of the Creoles.

"David B's Dutch language theory is not consistent with the fact that Antillians and Guyanese improve more across the generations than Turks and Moroccans."

Surinamese usually speak much better Dutch than lower-class Antillians. Surinam has been independent since 1975, and the last big immigration wave from that country occurred around that time. Even then, Dutch is still dominant as the language of administration and education, while in the Antilles, which is still part of the Kingdom of the Netherlands, Papiamento (a Portuguese/Spanish based creole language with some Dutch vocabulary) is dominant. Only the higher social classes are really fluent in Dutch. The small well-educated Antillian elite in the Netherlands generally arrived in the 1960s and 1970s. Recent Antillian “immigrants” (they are citizens) are almost invariably from the lowest social strata, speak an impoverished form of Dutch, are barely educated and often unemployed. While it is hard to tell a Surinamese from an Antillian from their appearance, their behaviour and related problems are quite different. Lower-class Antillians’ command of Dutch is not much better than that of Turks and Moroccans, so they have comparable room for improvement. But Turkish and Moroccan improvement is seriously impeded by (voluntary) social segregation and the role of (some manifestations of) Islam.

"I don't know whether the Turkish and Moroccan 2nd-gen children are entirely fluent in Dutch"

“Fluent” yes (like, African-Americans are very fluent in English), but (academic) proficiency levels differ greatly. Boys drop out of school much more frequently than girls.

"in Britain some Muslim communities (especially Bangladeshis) tend to speak their own language to the children and don't let them mix much with the 'natives', so the kids still start school with a serious language handicap.

Same in the Netherlands, although there are some peculiar differences between Turks and Morrocans. Turks keep speaking their own language among themselves, are proud of it and of their nationality and heritage, are relatively well organized, and are rising in socio-economic and educational achievement considerably faster than Moroccans. Relatively speaking, criminal behaviour among Turks this group is slowly declining.

Moroccans, on the other hand, often have a relatively poor command of their own language. This is complicated by the relationship of Arabic dialects to the standard language, the fact that many Moroccans have Tamazight (a “Berber” language) as their first language, in combination with limited literacy to begin with. Many Moroccans consider their dialect not really much of a language (in contrast with the official, high-prestige Arabic, which they often have a limited command of).

Turkish, by contrast, has high status in the Turkish community, which is quite nationalistic. Neither Turks nor Moroccans have much social interaction with Dutch people, but Turks usually have better Dutch language skills than Moroccans despite the fact that the latter use Dutch among themselves far more often and also have somewhat more contact with Dutch people than Turks.

By the way: intermarriage between Morrocans and Dutch is rare, but considerably less so than between Turks and Dutch.because of Islamic and tradition attitudes, Muslim male – Dutch female couplings are probably about 10 times as common as the other way around. As you can see in other posts on this site or elsewhere, for a Muslim girl to marry, date, or even socialize with non-Muslim males is quite dangerous. In most of the cases I’ve seen of Moroccan or Turkish girls with a Dutch boyfriend or husband, the girls were university students who had arrived on their own. By contrast, most Turks and Moroccans living in the Netherlands arrived by “family reunification” chain migration, following the labour migration of the 1960s and 1970s).

Posted by: Syarikat at October 2, 2003 02:20 AM

Syarikat: this is fascinating info. I hope you will keep commenting on other posts.

Posted by: David B at October 2, 2003 03:02 AM