In an earlier post, I noted that
- Ontario Catholics are able to choose between sending their children to the (Catholic) separate schools system or the public system, while non-Catholics can only go to public schools.
- Separate schools systematically outperform their public school counterparts, even though they receive the same funding and draw from almost-identical populations.
This was consistent with a story in which separate schools face incentives that public schools do not and in which they respond to those incentives in predictable ways. Since separate schools do not have a captive market, they have a greater incentive to provide higher-quality education.
There are some interesting echoes of that story here in Quebec. As in Ontario, there are two public systems: English and French. And as in Ontario, only one system has a captive market: unless a child's parent was educated in English in Canada, she must go to a French school.
If we wanted to apply the Ontario story to Quebec, we'd predict that the English system generated better outcomes than the French system. And we'd be right.
Clearly, the English/French distinction in Quebec is not so clean as the separate/public distinction in Ontario: many (most?) students in the English system are there because they are not sufficiently proficient in French. But if it were widely known that English schools were not doing a good job, they would have difficulty retaining their clientèle. Kids pick up second languages pretty quickly.
Competition. It's a good thing.
Worthwhile Canadian Initiative 
Good find! My first thought was that immigrants’ kids (some of whom have poor French) are required to go to French schools. But I see they have checked for that.
That’s another reason why it’s not a clean experiment. But apparently the results are similar outside Montreal, so it’s not a complete explanation.
I don’t know of any study in this, but I have this notion in my head that kids enrolled in French Immersion programs in Ontario tend to do quite a bit better than kids English-only streams. This might take away a bit from your point about Quebec schools.
Don’t disagree with the basic premise, though.
I think the story we’d tell there is that Ontario’s French immersion schools are cherry-picking. It’s hard to tell if they’re delivering good outcomes, or if they’re picking kids who would have done well no matter where they went.
1 above only refers to the elementary system, and some catholic boards can opt to open enrol (PDF). Secondary Catholic schools have open enrolment, presumably as a result of the funding extension beyond the Constitution imperative.
The last election was an interesting argument – the Premier argued that opening the public funding trough to other religions would “take $400m out of the system” which when translated to the system his children attend and in which his wife works would work out to more like $4bn. Only the Greens had the honesty to forgo chasing the catholic vote and argue for a non-denominational system where parallel bureaucracies – and expense loving trustees – weren’t funded by the public purse.
I’m not sure if this is the case today but when I went to catholic elementary school in ontario back in the late 70’s and early 80’s non-catholics attended my school so I had a couple of Sikh and Muslim kids in my grade.
Can’t see why a catholic school wouldn’t want more enrollment as long as it had the space to cater to their core market. I don’t know how schools are funded but I assume its by enrollment.
All I know is after we left high school my parents (catholic board supporters up to that point) had their support automatically revert to the public board.
Mark Dowling: in the previous post that I linked to, the results are all based on elementary school outcomes, where enrollment is (with a few exceptions) limited to Catholics.
My recollection of catholic school in Ontario was that all the schools across the province were way overcrowded with schoolyards full of portables and the last thing anybody wanted was more students.
A short list of other factors which might explain the results you are seeing:
Smaller school sizes perform better
Smaller school systems perform better
Minority groups with their own schools perform better
Schools with a more homogeneous student base perform better
Schools with a less homogeneous student base perform better
Schools with fewer immigrants perform better
Schools that aren’t the last resort for poor students perform better
Random variation
I’m not saying you’re wrong, just that we have nowhere near enough information to really draw any conclusions with any confidence. I’d actually find anecdotal evidence more persuasive in a situation like this (i.e. stories of actual families that switched systems to get a better education, stories of different choices made by the catholic board because they were afraid of losing students, etc.)
In Quebec, although French-language schools have a higher drop out rate that English-language schools, the French-language schools score higher on provincial exams.
Quebec francophone high school students also score well for math achievement in international comparisons such as PISA.
Quebecers score high marks on national test
I think the main difference is cultural, in that francophones are more elitists than anglophones. In anglophone culture, we try to have as many people pass as possible, at the cost of mediocrity, while francophones are more likeley to accept that laggards fall by the wayside.
There’s a similar phenomenon when you compare the Quebec junior hockey league to junior hockey leagues outside Quebec. I recall reading an article about Sydney Crosby, explaining that the reason he played for the Rimouski Oceanic before playing in the NHL was because of the much tougher standards in the Quebec junior hockey league.
David Card (yes, the one of minimum wage fame), Abigail Payne and Martin Dooley are pretty smart applied econometricians, and I didn’t see any obvious holes. An ungated, less-technical version (no equations) of their paper is here (pdf). They’re pretty careful about controlling for everything but the the school competition measure.