Brazenly illogical?

Is David Cameron’s proposal to attract the “brightest people” into teaching based upon a logical error? I ask because of this:

Finland, Singapore and South Korea have the most highly qualified teachers, and also some of the best education systems in the world, because they have deliberately made teaching a high prestige profession.

Let us grant that he’s right on the facts: Finland and South Korea do score highly on the Pisa. But what’s the causality?
One possibility is that highly qualified teachers cause pupils to become good. If this is the case, then Cameron’s policy might well be justified.
But there’s another possibility. Maybe these countries have a culture in which education is especially highly valued. If so, bright people will want to become teachers, and pupils will do well at school, but there won’t be causality from the former to the latter. Instead, both will arise from that culture.
If this is the case, then attracting brighter people into teaching might not raise standards.
There’s some empirical evidence here. Research in Sweden has found that teachers with high measured cognitive skills are actually bad for lower-ability pupils. And researchers (pdf) in North Carolina have found that the effect of teachers’ qualifications upon pupil’s achievements are often statistically insignificant, or even negative, once they control for the fact that better-qualified teachers tend to teach better pupils anyway*.
This evidence is roughly consistent with Claude’s claim that there’s a “total lack of correlation” between someone’s ability to pass exams at age 21 and their ability to inspire pupils years later.
This raises the question. Why, then, does Cameron seem so confident that improving teachers’ qualifications is a good idea?
One possibility is that he’s just ignorant about the evidence, and about the difficulties of distinguishing between correlation and causality in social research. Tories do, remember, have a lamentable tendency to display sheer pig-stupidity in interpreting social statistics.
The other possibility is that Cameron does know these problems, but just doesn’t care. His policy is designed to win the approval of numpties, not to actually improve school standards. After all, only ghastly oiks go to state schools anyway.
* This evidence is quite consistent with Cameron's claim that "good" teachers are very important. It's just that "good" is a matter of experience, inspirational ability and suchlike, rather than mere formal exam passes alone. 

Comments

[Mike Woodhouse]

Unless improving school standards (defined and measured how, I wonder?) can be shown to have a positive effect on electoral prospects within five years, one is disposed to assume that all such posturings are intended to win voter approval. This is likely to be equally true for the present incumbents, of course, who appear also not to give the proverbial tinker's cuss for the long term.

[Luis Enrique]

Some relevant research that uses a dataset matching individual teachers to pupils, and tries to control for all manner of possible confounding factors, does find a relationship between teachers and attainment:

http://www.bristol.ac.uk/cmpo/publications/papers/2009/abstract212.html

but the effects are not explained by observed teacher characteristics (like whether the teacher is highly qualified)

[ortega]

Probably GB is the country that has applied more plans to come back from the cul de sac where the 70's reforms brought it on. We must remember that those, like GB and Sweden, who were the leaders in the constructivist reforms, are now the less succesful .
Everything has been tried: more autonomy for the schools, less autonomy for the schools, unified CV, more money, specialized head masters, less people for room, more strict inspections,...It simply seems imposible to do an eficient counter revolution and statistics get worse every year.
The most succesful european systems (i.e. Bayern and Finland) are those who do not need to start any counter revolution because they haven't had any revolution.
The last unions proposal: to eliminate qualifications. It seems that with no proof of failure, at least there will be no bad conscience.

[Pat]

I'm puzzled- because this proposal seems to run counter to the voucher idea.
With parents deciding which are the best schools (and on what grounds) we only need to get out of the way and let the schools provide what is wanted. If in fact, a good degree makes a good teacher then competition will make this happen- if not then those with a good degree will re-deploy elsewhere where their abilities are more needed, thus benefiting the country outside the education system.
And I can't help pointing out that for many there is no point in high educational achievement- they lack the ability to benefit from it and the interest to take notice. I'm sure there is some shop assistant somewhere who had the potential to become (say) a doctor- but by and large they chose their careers wisely, and would not have been helped by more schooling.

[Steve]

As a teacher in South Korea, I feel I might be able to shed a little light on how exactly the wonderful performance of Korean students is attained - they spend vastly more time in school. For all but the very poorest, there are two schools in a school day; a state school in the morning and early afternoon, and a private school in late afternoon and evening and into the night (and at weekends during exam time) for older students.

You might have immediately noticed a couple of problems with this; first of all, social mobility is utterly ignored, with the poorest students spending much less time in school, therefore guaranteed to end up with low-paying manual jobs. The second is it might be hard to persuade teachers and students in Britain to spend 3 or 4 more hours a day in school or take much less vacation - if that's the plan, I'll see you on the other side of the crippling strike to say 'I told you so'.

The other two problems here are that, as noted, higher intelligence doesn't necessarily make better teachers, and as I think was argued at Left Foot Forward, making new recruits come from 'good' universities is going to considerably reduce the number of candidates available.

Once again, a Tory policy that won't work, that we're all desperately hoping will be quietly dropped after the election.

[Corey Dixon]

Luis - I'm not convinced on the quality of that study. Its regressions seem to omit several variables that are outside 'school' such as socio-economic class, poverty, single parent families, regional controls and such. I would take from this that the stats are significantly suspect.

I believe Hanushek brought together all the studies from reputable journals and found that teachers over the most studies can make an impact - but that it only is roughly 1 grade. They also can help overcome income gaps - these tend to be related to experience, education and formal test scores.

If the proposal of Cameron is to expand schemes such as TeachFirst which do not only look for bright pupils - but are very selective over who they take - accounting for inspirational impact etc. then this is a policy that is likely to work. The testimony of schools and pupils involved in the Teachfirst scheme seems enough to say they are having an impact.

Equally, schemes such as the above generate more bright individuals into schools who rise rapidly and quickly through the school system. This is often highly beneficial to have strong leaders in the key administrative and strategy roles for the school - rather than having many of these people attaining prominence in business.

[chrisg]

Just to make sure I've got this right, is this post arguing that it would be better for state school pupils if people who leave university with 3rd class degrees can become state school teachers than the alternative that they cannot?

Are there other professions where we could improve outcomes by lowering entry standards, or is it just teaching?

[ad]

"I'm puzzled- because this proposal seems to run counter to the voucher idea."

Pat: I agree. I suspect that the voucher idea is something the Tories think will work, but not help them get elected.

But burbling on about higher standards for teachers will help them get elected in the first place.

Policies that sound good may get you elected. Policies that work may get you re-elected.

[oldandrew]

I can't believe you are making me defend the Tories, but you are being incredibly rude about David Cameron's grasp of the evidence when your own doesn't seem up to much.

Your first piece of research does not suggest that bright teachers have no benefits, nor does it state categorically that they are bad for low performing peers. It states that "while high performing students benefit from high cognitive teachers, being matched to such a teacher can even be detrimental to their lower performing peers." While it does show a mixed bag of effects the benefits of bright teachers on the most able are stated far more strongly than their disadvantages for the least able.

As for your second piece of research, I just don't get your interpretation at all. It seems to show a number of positive effects based on teacher qualification.

[Alena]

I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.

Alena

http://grantsforeducation.info