Matt Yglesias has a short post on immigrant achievement across Anglophone countries, drawing on a new study:
Elizabeth Washbrook, Jane Waldfogel, Ali Ghanghro, Miles Corak, and Bruce Bradbury looked at this through an education lens. If you look at immigrant children in Anglophone countries when they start school, they’re just as advanced in terms of cognitive skills as their peers from nonimmigrant families with one important exception—weak vocabulary and language skills. That’s what will come from growing up in a household where the local native language isn’t spoken or isn’t spoken proficiently. One major task of the school system is to help those kids whose home environment is less conducive to learning. And it happens. Whether in the United States, the United Kingdom, Australia, or Canada, the gap closes somewhat. But to quote Corak: “The resulting disadvantages in reading skills are overcome to a much greater degree as they progress through school in Australia and Canada than they are in the United Kingdom and the United States.”
As my friend and colleague Arpit Gupta notes, however, the Bradbury et al. also addresses selection effects. Table 3 compares the socioeconomic characteristics of native-born parents, foreign-born parents who speak the official language at home, and foreign-born parents who do not. The results are fascinating. In Canada, 53.9% of foreign-born parents who speak the official language at home and 55.5% of foreign-born parents who speak a foreign language at home have achieved the highest education level as opposed to 52.8% of native-born parents. In Australia, 46.6% of foreign-born parents who speak the official language at home and 44.3% of foreign-born parents who speak a foreign language at home have achieved the highest education level while only 34.3% of native-born parents have done the same. That is, foreign-born parents, whether they speak the official language at home or not, tend to have higher academic credentials than native-born parents — a slight difference in the Canadian case.
The numbers are, suffice it to say, striking different in the United States and Canada. In the U.S., 45.7% foreign-born parents who speak the official language at home have achieved the highest education level while only 21.1% of foreign-born parents who speak a foreign language at home have done the same. This is a dramatic difference. As for native-born parents, 35.4% have achieved the highest education level. The British numbers are roughly comparable.
As Arpit suggests, one possibility is that immigrant children do better in school in Canada and Australia due to the cumulative impact of immigration selection policies than rather than instructional quality.
The authors offer a different take:
In summary, we find some support for our hypothesis that immigrant parents are more positively selected in Australia and Canada than in the United Kingdom and the United States, but also evidence of a more nuanced picture. First, the distinction between language groups seems crucial, with official-language-speaking immigrants showing high levels of parental resources that differ little across countries. It seems likely then that any cross-national differences in relative child outcomes will show up more among the foreign language group, where socioeconomic resources differ systematicallybetween the two sets of countries. Second, however, there is variation across countries in the degree of which children of immigrants experience other advantages (such as lower single parenthood) and disadvantages (such as low birth weight) that does not necessarily map closely to these socioeconomic patterns.
They apply a variety of statistical controls (remember Jim Manzi’s Uncontrolled?), and conclude their discussion with the following:
We began by noting the substantial variation across countries in the outcomes of adolescent children of immigrants, and in particular the poor relative performance of this group in the United States compared to their equivalents in Australia and Canada. We find little evidence that this difference exists prior to school entry (particularly with respect to Canada), but instead that children of foreign-language-speaking immigrants have similar difficulties with early language skills in all four countries.
That is, if I understand correctly, the children of foreign-language-speaking immigrants in Canada and Australia, where 55.5% and 44.3% of foreign-born children who speak a foreign language at home have achieved the highest education level have similar difficulties with early language skills as the children of foreign-language-speaking immigrants in the United States and Britain, where 21.1% and 25.4% have achieved the highest education level respectively.
For some reason difficulties in this one domain appear associated with a widening of disadvantage in literacy, math, and science in the United States, but not in Canada or Australia. One possibility is that the different trajectories ofchildren of immigrants have a good deal to do with the extent to which public policy, the education system, and other integration policies fail or succeed in addressing this shortfall. In Australia, for example, there has been an increasing emphasis on the development of appropriate language skills before arrival, coupled with a policy of fast-tracking foreign university students through the immigration process. Canada has long held a multicultural policy that supports a positive sense of identity, along with language and work support policies. In the United States there is a dominance of Spanish-speaking migrants, many lacking documentation, who may have less of a tendency to learn English, and more difficulties integrating into schools (OECD, 2006).
Note the unique circumstances the authors identify in the United States: the critical mass of Hispanophone students might delay linguistic assimilation, as might the volume of undocumented immigration. The strategy of reducing the size of the undocumented population through attrition is extremely controversial in the United States, but somewhat less so in Canada and Australia. Moreover, Canada and Australia’s policies should be understood in the context of its strong preference for skilled migrants.
Another possibility is that unobservable differences between immigrant groups in cultures and countries of origin affect children more strongly as they age. Although our sample sizes are relatively large, we lack the power (and in the U.K. and U.S. cases the required data) to distinguish immigrant parents by their countries of origin. The contribution of our research is to provide evidence against substantially different patterns of school readiness of children of immigrants across countries, despite differences in countries of origin and selection policies. Although not our focus here, the results clearly raise the question for future research of what happens to children of immigrants during the school years that results in such different outcomes in these countries.
This is an extremely difficult question to answer. One wonders if the differences between highly-educated foreign-born parents and less-educated foreign-born parents who speak a foreign language at home become more significant over time, e.g., highly-educated parents might acquire stronger language skills over time, which might facilitate linguistic assimilation on the part of children.