David Leonhardt at NYT:
How much do your kindergarten teacher and classmates affect the rest of your life?
Economists have generally thought that the answer was not much. Great teachers and early childhood programs can have a big short-term effect. But the impact tends to fade. By junior high and high school, children who had excellent early schooling do little better on tests than similar children who did not — which raises the demoralizing question of how much of a difference schools and teachers can make.
There has always been one major caveat, however, to the research on the fade-out effect. It was based mainly on test scores, not on a broader set of measures, like a child’s health or eventual earnings. As Raj Chetty, a Harvard economist, says: “We don’t really care about test scores. We care about adult outcomes.”
Early this year, Mr. Chetty and five other researchers set out to fill this void. They examined the life paths of almost 12,000 children who had been part of a well-known education experiment in Tennessee in the 1980s. The children are now about 30, well started on their adult lives.
On Tuesday, Mr. Chetty presented the findings — not yet peer-reviewed — at an academic conference in Cambridge, Mass. They’re fairly explosive.
Just as in other studies, the Tennessee experiment found that some teachers were able to help students learn vastly more than other teachers. And just as in other studies, the effect largely disappeared by junior high, based on test scores. Yet when Mr. Chetty and his colleagues took another look at the students in adulthood, they discovered that the legacy of kindergarten had re-emerged.
Students who had learned much more in kindergarten were more likely to go to college than students with otherwise similar backgrounds. Students who learned more were also less likely to become single parents. As adults, they were more likely to be saving for retirement. Perhaps most striking, they were earning more.
All else equal, they were making about an extra $100 a year at age 27 for every percentile they had moved up the test-score distribution over the course of kindergarten. A student who went from average to the 60th percentile — a typical jump for a 5-year-old with a good teacher — could expect to make about $1,000 more a year at age 27 than a student who remained at the average. Over time, the effect seems to grow, too.
The economists don’t pretend to know the exact causes. But it’s not hard to come up with plausible guesses. Good early education can impart skills that last a lifetime — patience, discipline, manners, perseverance. The tests that 5-year-olds take may pick up these skills, even if later multiple-choice tests do not.
David Shenk at The Atlantic:
For those who still need proof that great teachers matter, the NYTimes has an excellent piece today by the economics columnist David Leonhardt about an “explosive” new study showing great kindergarten teachers have huge effects on future success and well-being. Leonhardt quotes one economist as estimating that a top-notch kindergarten teacher is worth $320,000 per year.Early development matters. Parenting matters. Schools and teachers matter.
Ryan Avent at Free Exchange at The Economist:
Economics Nobelist James Heckman has found that the earlier one pursues efforts at remediation with underperforming students the more effective the interventions are. And studies have indicated that while the academic knowledge gained from remediation programmes tends to fade, social knowledge is more durable (and it may well be more important over the long-term). In general, it seems like the importance of educational reforms at the secondary and undergraduate level is wildly overstated, while the importance of improvements in education at the primary level (and earlier) is given far too little attention.
The findings haven’t been published yet, and I imagine it will give us far more to chew over. For now, I have some tentative thoughts:
* Standout kindergarten teacher isn’t terribly specific — of the teachers in the survey, how many would be considered standouts? The study tracked 12,000students . Roughly speaking, let’s say this amounts to 600 classrooms and 600 teachers. Were there 100 standouts? Or 10? I’d love to see the curve. Highlighting the standout teachers might give us a misleading picture of the scale of the challenge.
* How much is a standout parent worth? Some children wil generate a great deal of tax revenue over a lifetime. Should we, in light of this fact, give more generous tax subsidies to parents with the characteristics the best predict high future earnings for their offspring? I imagine the answer is no — this would strike most observers as unacceptably illiberal. But improving incentives for parents, as Robert Stein and Ramesh Ponnuru have recommended, seems like a pretty good idea.
Leonhardt offers a few immediate responses:
They can pay their best teachers more, as Pittsburgh soon will, and give them the support they deserve. Administrators can fire more of their worst teachers, as Michelle Rhee, the Washington schools chancellor, did last week. Schools can also make sure standardized tests are measuring real student skills and teacher quality, as teachers’ unions have urged.
Others are less nuanced than Leonhardt. Erik Brynjolfsson, an MIT economist I greatly admire, writes the following:
We would be better off if we paid our teachers a lot more, and attracted more of the best and the brightest to this profession.
Note that while Leonhardt suggests we pay our best teachers more, Brynjolfsson suggests that we pay all teachers a lot more. But again, how many teachers were standouts? And were the standouts drawn to the profession for the cash compensation? One can imagine compensating teachers by giving them more autonomy and flexibility, or by limiting the number of disruptive students.
Rather than offer blanket pronouncements on how much teachers ought to be paid, we need to give more thought to allowing more business model innovation in the education sector. As new compensation schemes emerge, we’ll eventually learn more about what works and what does not.
Where does the amount $320,000 come from? It’s “the present value of the additional money that a full class of students can expect to earn over their careers.” And that’s not counting the social and psychological values that may flow from excellent early schooling. But salaries aren’t calculated by the actual benefit the employee bestows on the clients. If it did, there would be some negative numbers. But what if kindergarten teachers were very highly paid? Different individuals would pick kindergarten teaching for a career….
In fact, it seems a bit hard to believe. If kindergarten teachers matter as much as this new research suggests, then you would think that parents would have a large influence on their kids’ adult outcomes. After all, you spend a lot more time with your parents than in your kindergarten class. But much research in behavioral genetics finds very little evidence for significant parental effects. (See Judith Harris’s The Nurture Assumption.) So I am puzzled.
However, as I read the story and the findings it is based upon, some questions crept into my mind. I relate them not out of any desire to discredit the study, which is enterprising and newsworthy, but simply as a warning to parents and policymakers not to go overboard. Yes, good teaching is important. But the key policy issue is raising average teaching standards throughout the education system, not simply seeking out and rewarding exceptional kindergartens. In affluent areas of New York City, parents’ efforts to capture every possible educational advantage for their young children have already reached absurd levels. If this new research is accepted as gospel, I can already see admissions officers at snooty private schools waving it in the air and saying to parents: ‘Look, I told you our $30,000 fees were justified. Where else can you find a kindergarten teacher with a Princeton Ph.D. in Victorian literature?” (In the public sector, meanwhile, the teachers’ unions will cite the new research in support of higher pay claims, ignoring the fact that it doesn’t even address the issue of whether higher pay leads to better-quality teaching.)
All this emanating from a dense academic article that hasn’t been published or peer reviewed. At this stage, there isn’t even a working paper detailing how the results were arrived at: just a set of slides. Why is this important? Because economics is a disputatious subject, and surprising empirical findings invariably get challenged by rival groups of researchers. The authors of the paper include two rising stars of the economics profession—Berkeley’s Emmanuel Saez and Harvard’s Raj Chetty—both of whom have reputations for careful and rigorous work. However, many other smart researchers have had their findings overturned. That is how science proceeds. Somebody says something surprising, and others in the field try to knock it down. Sometimes they succeed; sometimes they don’t. Until that Darwinian process is completed, which won’t be for another couple of years, at least, the new findings should be regarded as provisional.
A second point, which is related to the first, concerns methodology. In coming up with the $320,000 a year figure for the effects that kindergarten teachers have on adult earnings, the authors make use of complicated statistical techniques, including something called a “jack knife regression.” Such methods are perfectly legitimate and are now used widely in economics, but their application often adds an additional layer of ambiguity to the findings they generate. Is this particular statistical method appropriate for the task at hand? Do other methods generate different results? These are the sorts of question that other researchers will be pursuing.
From the perspective of someone whose statistical education didn’t extend much beyond rudimentary probability and econometrics, a few things struck me about the evidence that is presented in the slides. The study confirms that smaller classes have a significant impact on test scores, which was the primary conclusion of the original Tennessee study. And the correlation between kindergarten test scores and later life outcomes—not just earnings, but also home-ownership values, and even death rates—seems to be pretty robust. (See Figures 3a to 3f.) However, the relationship between test scores and earnings doesn’t necessarily have much to do with the quality of teaching. It could simply reflect students’ innate ability, the impact of other students in the class, or other unknown factors.
The researchers try to control for some of these factors, but they can’t rule them out. In arguing that individual teachers play an important role, they rely on evidence that is somewhat indirect, which is inevitable because the Tennessee experiment wasn’t set up to study the impact of teacher quality: quite the contrary. It was designed to gauge the impact of smaller class sizes regardless of who the teacher was. Unlike their students, the kindergarten teachers didn’t get graded; consequently, the new team of researchers didn’t have any data on whether specific teachers were good or bad. They were forced to rely on student test outcomes as an indicator of teaching quality. Effectively, they assumed that if a given a kindergarten class achieved outstanding results its teacher must have been responsible. In many cases, this may well have been true, but we have to take it largely on trust.