John Thompson examined the studies comparing the relative cost and benefits of older and younger teachers, and he reads the findings differently from the Education Week reporter.
Here are my thoughts on your question.
These studies had different purposes so, if used properly, they would have different effects on policy discussions. For instance, the North Carolina study investigates, “different responses to pension incentives.” It develops “a conceptual model of teacher retirement behavior and employ(s) a unique data set to estimate the causal effect of pensions on teachers’ exit decisions.” It explains, “Teachers in my sample are in their fifth or higher year of teaching … .”
In other words, it offers no support for reformers seeking to replace veteran teachers with TFAers or other inexperienced teachers in the hope that student performance will increase.
Also, in North Carolina “the most- and least-effective teachers in North Carolina are the first to leave, a new study finds. By six years out, however, more-effective teachers are much more likely to retire than less-effective ones.” So, if we conclude that inexperienced teachers are as effective and cheaper as experienced ones, and keep the buy-outs in perpetuity, what would happen after the least-effective veterans are gone? That question should give pause to “reformers,” who in my experience are committed to driving Baby Boomers out in order to keep young teachers away from our professional judgments, as well as save money.
Secondly, the Los Angeles study found an increase in student performance after retirements and it focused on peer effects and the decision to retire. So, it could be an anomaly (due to that unique retirement law and its effects on one district) or, it could have been the most important study for policy purposes. After all, West Ed had discovered that for every $1,000 cut from per-student spending, teachers in the state were 4 percent more likely to retire. That suggests that conditions inside schools can have a big effect on who takes early retirement, and that has a big effect on whether those early retirees are a valid sample for discussing the effectiveness of teachers.
The LA study found “that the retirement of an additional teacher in the previous year at the same school increases a teacher’s own likelihood of retirement by 1.5-2 percentage points.” It conducted “robustness checks indicate that teachers’ responses to colleagues’ retirements in the previous year are not driven by coordinated retirements of spouses, a subsequent increase in workload or a distaste for working with less experienced teachers.”
But, it did not check for the factors that teachers would cite as likely explanations of variance in who retires and why. After all, we are more likely to throw in the towel after being worn down by the challenges of high-poverty schools and/or mismanagement. So, the chances are that the sample of early retirees was not representative but that the economists did not ask teachers to help design a better methodology for comparing teaching effectiveness.
Thirdly, the Illinois study found that the poorest and lowest-performing schools saw the biggest test-score gains after early retirement. Those results may say a lot about the nature of those schools, but thus say very little about the teaching profession as a whole. The sad truth is that the top talent in the toughest schools tend to be worn down and move to schools that are less maddening. Moreover, low performers tend to be channeled towards low-performing schools.
The question is how these serious problems should be addressed. Some “reformers” want to move teachers around like chess pieces, and they will claim that these articles give support to their top down policies.
I suspect that many relevant findings reflect early retirement packages (especially when they use data back to 1992) being used as a substitute for a lot of missing policies. Yes, low performing teachers were more likely to take the offer, suggesting that they were used in lieu of the dismissing ineffective teachers. The solution to that issue is fair and efficient methods of removing ineffective teachers, as opposed to today’s “teacher quality” gimmicks.
High-performing teachers were also more likely to retire early and that reflects a lack of a career ladder. So, the studies document the need to better capitalize on the strengths of the best teachers. To take a military metaphor, if the best lieutenants kept getting pay raises, but they could not be promoted, they would get better at leading their own platoon, but their wisdom would not affect more than those few soldiers. A better system would be for systems to institutionalize ways of drawing on the experience of top teachers – experience that they are paying for – for setting effective policies.
We should not be like the “reformers” and deny truths such as the reality that “many teachers may feel ‘pulled to stick it out a few more years’ in order to receive their full pension benefits, even if they are no longer interested in teaching.” As one local union leader explained to me, the best tool for removing older, ineffective teachers would be the passage of universal health care. His efforts to counsel out such teachers were undermined by the reality that older persons with health problems are locked into their jobs by the lack of health care options. Similarly, Toledo’s Dal Lawrence describes his decision to fire a friend. His fellow teacher later said that the job’s stress had gotten to him and the union’s dismissal of him through peer review saved his life.
The following may sound like special pleading, and I have less confidence in it, so I would not showcase the following speculation. But, in regard to the Illinois study, in the early 1990s the crack and the murder epidemic were peaking. Their replacements in the mid-1990s entered a profession where NAEP scores were increasing. The same could also be true of the L.A. study which covers the peak of the Clinton economic boom 1998 to 2001. So, the veteran teachers might have seen additional increases in their test score growth if they’d remained during the up years.
My district did early retirements in the “jobless recovery” of the mid-90s. It thus got the budget problems behind it in the least disruptive way. Soon afterwards, test scores increased as Oklahoma City finally got out of our two-decade Great Recession. And, that influences my views on how the studies should be read.
During the 2007 Great Recession, my district rejected the buy-out option. Oklahoma embraced the Colorado teacher evaluation law and Oklahoma City used the SIG and other “reforms” to “exit” veteran teachers who it thus labeled as “culture killers.” In the most notorious example, a Transformation school “exited” 80% of its teachers. Now, 5% of that school’s juniors are on track to graduate. The elementary school that feeds my old school brought in so many young teachers that it made the newspaper because of the rampant fights and chaos that resulted, so that they even had to close the school to get reorganized.
The first step in analyzing the economic studies should be to consider “Rational Expectations.” Why would a talented young person commit to a profession, start a family, and buy a house when he or she would become expendable after their effectiveness peaked? We should also ask what would be more cost effective – periodic buy-outs that we all acknowledge aren’t an optimal approach or the churn of today.
Reformers condemn buy-outs and other practical but unlovely policies as “the status quo.” But, they should honestly face all of the facts and ask whether their policies have been worse than the imperfect ones of the “status quo.” They should not cherry pick economists’ findings. They should do a cost benefit analysis of their theories.
As I argued this week, neither we should not be afraid of admitting hard truths.
http://www.huffingtonpost.com/john-thompson/the-challenge-of-overcomi_b_2521436.html?utm_source=Alert-blogger&utm_medium=email&utm_campaign=Email%2BNotifications
We should be transparent when discussing the difficulty in creating learning environments where equally good teachers in rich and poor schools can get equally good results. Especially in the inner city during an age of “accountability,” teachers get burned out. After all, in the inner city the biggest beneficiaries of such policies would not be teachers, as much as the students who are also burned out by our deplorable conditions.
If the evidence shows that teacher effectiveness increases steeply in the first few years and then levels off, why should we feel threatened by that? Isn’t it likely that the same is true of most jobs? Would we get better doctors or better UPS drivers if we started to harass them out of jobs after their first decade or so?
Even President Obama, last week, returned to the position that we can’t balance our budget by reneging on Social Security and Medicare. It is only the contemporary school “reform” movement that argues that teaching is the only profession that would attract more talent if contacts signed in good faith could be torn up at the whim of non-teachers.
And, finally, while pure research may yield information on high- and low-performing teachers, policy should focus on the vast majority in the middle. The ultimate pyrrhic victory is using abusive teacher evaluations the way we are doing now – undermining the entire profession to get rid of low performers.
Diane Ravitch's blog 