Are the children of intermarried couples smarter?

Ever since my research for my 2002 book Bilingual Couples Talk I’ve regularly been told by people – or been asked to confirm their belief – that a cross-cultural relationship is beneficial once the couple have children. The children are expected to not only be bilingual but also to enjoy cognitive advantages from growing up with more than one culture and to be more open minded and better communicators. I’ve always struggled how to respond because, of course, nothing is ever this simple. A 2011 study of the cognitive and linguistic abilities of various groups of preschoolers in Germany confirms the assumption – children of intermarried couples outperform all other groups on a cognitive ability test – and, simultaneously, explain why it is a fallacy that confounds ethnicity and class.

The study by Birgit Becker examines the cognitive and linguistic abilities of three- and four-year-olds with different types of parents:

• Children whose parents and grandparents were all born in Germany (the ‘native’ group)
• Children whose parents were both born in Turkey (the ‘second generation’)
• Children whose parents were both born in Germany but each parent had at least one parent born in Turkey (the ‘third generation’)
• Children with one first-generation and one second-generation Turkish parent (the ‘2A generation’)
• Children with one ‘native’ parent and one first- or second generation Turkish parent (the ‘intermarried’ group)

The cognitive abilities of a total of 1,008 children were tested with the Kaufman Assessment Battery for Children. The German version of the test was used but it was administered by bilingual researchers and the children could choose to do the test in German, in Turkish or they could mix the two languages as they pleased. So, language proficiency is unlikely to confound test results here, as it so often does in cognitive testing of bilingual and minority children.

The diagram shows that the children from the intermarried group outperformed all the other groups, including the natives. It also shows that, with the exception of the intermarried group, all the other ‘Turkish’ groups performed significantly lower than the ‘native’ group. Children in the ‘2A group’ – with one first-generation and one second-generation Turkish parent – performed particularly poorly. In fact, ‘2A’ parents might be considered ‘intermarried,’ too; but, obviously, their intermarried status is not beneficial for the child.

Once the full diagram is revealed, part of the conundrum is solved.

Once parents’ socio-economic status (as measured by their level of education and their occupational status) and educational resources (as measured by the number of books in the home; the frequency of bedtime stories; or the number of visits to the zoo) are controlled, the ethnic differences disappear and the influence of all the above ethnic groups/generations is reduced to non-significance.

All group differences regarding children’s cognitive skills can be fully explained by families’ socioeconomic status and educational resources. (Becker 2011, p. 447)

What seems like an ethnic effect (‘children of intermarried couples are smarter’ or ‘German children are smarter than Turkish children’) is, in fact, an effect of socioeconomic status and educational resources; in other words, a well-known class effect. However, class maps onto ethnicity, in this case, as elsewhere. The vast majority of Turkish families in the sample, which can be assumed to be representative of Turks in Germany (or, at least, southwest Germany, where the study was conducted), are poorly educated, work in low-status occupations, and have few educational resources at their disposal.

As far as the two ‘mixed’ groups – ‘2A’ and ‘intermarried’ – are concerned a process of negative and positive selection can be assumed to apply respectively.

Having a first-generation mother and a second-generation father constitutes some sort of ‘double jeopardy’ for the child: the mother is much less likely to speak German than even first-generation women married to first-generation men; and the father is even less likely to have completed secondary education than other Turkish second-generation men. As the researcher explains, second-generation men who ‘import’ brides from the country of origin are likely to be negatively selected on various dimensions and their ‘imported’ brides will lack knowledge and resources that are useful to raising a child in the destination country.

By contrast, a process of positive selection works in favor of a child with a native and a migrant parent. Not only will the native parent ‘automatically’ have country-specific knowledge and resources but the migrant parent is likely to be positively selected with regard to level of education, proficiency in German, and general ‘openness’ and ‘integration.’ This is particularly true in the case German-Turkish intermarriages, which are comparatively rare and only account for five percent of all marriages of first- and second-generation Turks in Germany.

In sum, if intermarriage is an expression of parental cosmopolitanism, it is beneficial for children. Not because there is any intrinsic value in intermarriage but because that is how educational reproduction works: well-educated parents with stable jobs, parents who read to their children and who engage in a wide range of family activities confer an advantage on their children. It is just that the advantages – as well as the injuries – of class are increasingly mapped onto ethnicity, race or ‘culture.’

Becker, B. (2011). Cognitive and Language Skills of Turkish Children in Germany: A Comparison of the Second and Third Generation and Mixed Generational Groups International Migration Review, 45 (2), 426-459 DOI: 10.1111/j.1747-7379.2011.00853.x