(1.) A teacher who is in the top 10 percent of the current distribution of value-added raises student achievement by several times what a teacher in the bottom 10 percent does.
(2.) If all US teachers had value-added equal to what the current top 10 percent has, the average American student would achieve at the level of students whose parents have incomes in the top 10 percent of the family income distribution. This is approximately equivalent to the level at which the average student in Singapore achieves.
Clearly, this contradicts my null hypothesis, which is that there is no intervention that can achieve a significant, durable, replicable impact on student outcomes.
She also favors choice and competition, as well as attempts to introduce technology in a cost-effective way.
Perhaps most interestingly, she begins the essay by saying that she is describing what might be feasible ideally, not with what the political system is likely to produce. Implicit in this view is that the political process is likely to be distorted by interest groups. This is something that progressives tend to ignore when they advocate for government playing a larger role in education, health care, and so forth.
“Clearly, this contradicts my null hypothesis, which is that there is no intervention that can achieve a significant, durable, replicable impact on student outcomes.”
Not to be disagreeable, but to me your null remains intact. First, what is “the intervention?” “Assume a magic wand”? Didn’t all these teachers get some kind of teacher training? Why didn’t it work for the mid-to-lower tier teachers? In other words, are they utterly failing in teacher training, are they utterly failing to teach the right things, or does teaching simply not work? When they get to the classroom, 90% of the 100% who went through training designed to make them great teachers apparently aren’t good enough to even be allowed to stay teachers. We can safely assume 10% of people who end up in a career of their choosing are just naturals. Is all teaching that useless?
Second, for example, I had a “good” teacher for Thermodynamics (you can use literature or high school calculus if you prefer) who made me prioritize thermodynamics over other things. Then I entered life and aside from burning calories trying to artificially apply what I learned in that class I have not needed one time and I worked in a job involving heat transfer. This may sound impossible, but let’s just say that entropy and partial differentials never came up on the job. Even if I could start throwing around thermo at a plant they would have (literally) not allowed me back in. Maybe academia is the only job where knowing things your colleagues can’t possibly understand is reliably beneficial.
So, good teachers maybe affect learning of something useful in one world followed by the rest of your life where you are on your own in a completely different world.
How are the measuring “Value Added”? Test scores?
Is there any solid reason, other than our intuition, that higher test scores and the measures used to get them, improve our lives or even the economy?
I agree with Andrew- the null hypothesis is safely intact. Hoxby could apply the same logic to the 100M dash by noting that Usain Bolt is 4 standard deviations faster than most humans, therefore if everyone were Usain Bolt the average human would be much, much faster. This is true, but does not imply that there is some way of actually making human’s faster, anymore than saying “some teachers are better than others” implies that there is some way of actually changing the distribution of teacher’s effectiveness.
It raises the stakes. And it seems there is a lot to try since it seems we are doing nothing. My kid has a new teacher this year and as far as I can tell the school did nothing for her.
So how does Singapore train, select, and retain teachers?
And how does Jamaica train, select, and retain sprint coaches?
Is there any association between well liked and performance? Probably weak at best.
Liberals don’t ignore it; they just realize it occurs in all institutions.
One problem here is that you need a lot of years to determine how much a “top 10%” or a “bottom 10%” teacher raises student achievement. No one with any sense believes that if mutual fund x has outperformed the market by 9% last year and 7% the year before that it will consistently outperform the market by about 8%.
An interesting article on Singapore:
http://theconversation.com/why-is-singapores-school-system-so-successful-and-is-it-a-model-for-the-west-22917?curator=MediaREDEF
While I agree with most of Hoxby’s recommendations, I am disappointed by the argument. Why must we worry whether “we” are above average, lament changes in “our” comparative advantage, or raise the specter of (gasp!) educated foreigners competing against “us”? Perhaps this resonates with the general public, but in fact we should be celebrating a potentially richer world that will benefit everyone. Moreover, these sentiments run counter to the decentralization advocated: Choice with competition.
I have seen enough to be quite confident that gifted teachers can and do raise performance in the children they teach, in a way that is sustainable – possibly not over a lifetime, but certainly over several subsequent years – across multiple dimensions including academic performance, self esteem (achievement-based emotional resilience), social-emotional skills, and assertive enthusiasm for learning.
What I have not seen, is much evidence that gifted teachers are made. They appear, instead, to be born (and then they gain experience). Some of the very best teachers have unimpressive credentials; some of the most impressively-credentialed are unimpressive teachers (this is to be expected by looking at the 2-D distribution of performance – being a standout on two dimensions is less likely than being a standout on either).
Either better teachers must be made – which I cannot rule out – or teaching may be made more selective, to identify and retain only the more gifted teachers. Since people who have gifted-teacher potential are fewer, pay would likely have to be higher to attract a higher percentage of them into teaching. But, also, the system would have to be much more ruthlessly selective – and a lot less focused on academic qualifications. I see little chance of either – and I feel bad for gifted teachers, who remain under-appreciated (and often resented by their less-gifted peers).