The demand for French immersion education in Vancouver so far outstrips the supply that the school board allocates places by lottery.

But why? Is it because French is a useful employment skill? Because learning to speak French makes you a better person? Or is it because parents know intuitively what economists can show econometrically: peer effects matter. Being with high achieving peers raises a student's own achievement level.

A study by Douglas Willms found that children in French immersion tend to have better educated and higher income parents, while children with special needs tended to be concentrated in the English ("core") programs:

If students with special needs were equally distributed among all classes, each teacher would on average have 3.4 students with special needs. However, in schools that have early immersion programs, the average in core English programs is about 5.7 students.

French immersion streams children.

You might ask: so what? Every public school system in the world has some form of explicit or implicit streaming. Sometimes the streaming is explicit, as with "gifted" programs in parts of Canada. Sometimes students are streamed on the basis of where the parents live. High income/high parental education children go to school together in West Vancouver; low income and/or low parental education children go to school together on Vancouver's East Side. Sometimes streaming is based on religion. In England, where church attendance rates are well below those in Canada or the US and ads on London buses cheerfully proclaim "there's probably no God, now stop worrying and enjoy life", fully one quarter of primary schools are affiliated with the Church of England. 

Streaming happens. The questions are (1) who gets in the stream that has the high achieving pupils and (2) are the side effects of streaming positive or negative.

French immersion streams children on the basis of linguistic ability. Perhaps for this reason, girls are more likely to enroll in French immersion. I could not find any statistics on the proportion of recent immigrants enrolling in French immersion, but I suspect it is not high.

If we had another kind of streaming – say math immersion, science immersion, hockey immersion, music immersion – different students would be in the high achievement stream. Would that be better? I don't know. But it would be different.

The other way of evaluating streaming is to consider the consequences – what are the side products? French immersion streaming produces students who know more French, math immersion would produce students who know more mathematics. In Canada, how valuable is knowledge of French?

Several studies have found that Anglophones who can speak French enjoy an earning premium. The question is: do bilingual Anglophones earn more because speaking French is a valuable skill in the workplace? Or do they earn more because they're on average smarter and more capable people (after all, they've mastered two languages)?

A recent study by Louis Christofides and Robert Swidinsky published in Canadian Public Policy compared the earnings of Anglophones (outside Quebec) unable to speak French, bilingual Anglophones who did not use French at work, and bilingual Anglophones who frequently used French at work. 

Bilingual Anglophones earn more. But they earn more whether or not they speak French at work – speaking French at work seems to increase earnings slightly, but the effect is statistically insignificant. Christofides and Swidinsky's results are compatible with the idea that being bilingual signals that you are a high ability individual, and so you are more valuable to an employer – but actually speaking French is irrelevant.

This is how they interpret their results:

In the ROC [Rest of Canada, that is, outside Quebec], during 2002–2003, 7 percent of all children in elementary, intermediate, and secondary schools were enrolled in French immersion programs and, in addition, many children were enrolled in core French programs (the proportion of bilinguals in our ROC sample in Table 2 is, roughly, 8 percent). Our findings suggest that enrolment in these important programs may be motivated by social/cultural/political reasons and may reflect superior ability and a commitment on the part of parents to education and knowledge. While knowledge of both official languages carries a modest earnings advantage, no significant additional earnings advantage can be discerned for those who actually use French in the ROC.

And they conclude:

Efforts to promote French in the ROC should be continued not so much because of the earnings advantage that bilingualism confers but because it results in many social/cultural/political benefits, strengthening the fabric of Canadian society and serving as an example to countries torn by ethnic, religious, and linguistic divisions.

These are fine words. But while we're strengthening the fabric of Canadian society, there's a stack of boys, recent immigrants and children with special needs thrown together in English classes. Are we serving them well?