Much as I'd feared, preparations for round two of Race to the Top (RTT) seem to be impelling states to overshoot the mark when it comes to teacher evaluation and pay. Generally laudable proposals like Florida's Senate Bill 6 and Colorado's Senate Bill 10-191 suffer from the "fix the world in one pass" syndrome. Advocates who are waging an admirable fight to end or dramatically scale back hyper-rigid, industrial-era state policies governing teacher tenure and compensation display a worrisome tendency to mandate that, henceforth, teachers will be evaluated in large part on (thus far) largely nonexistent, hyper-rigid, value-added metrics. This is all playing out in Colorado right now, where SB 10-191 passed the state senate on Friday and goes to the state house this week.
Now, don't get me wrong. I supported Florida's Senate Bill 6 and heartily support Colorado's more measured 10-191, but the impatient rush to "fix" teacher quality in one furious burst of legislating is leading to troublesome overreaching and putting the cart before the horse. The result: hugely promising efforts to uproot outdated and stifling arrangements get enveloped in crudely drawn, sketchily considered, and potentially self-destructive efforts to mandate a heavy reliance upon value-added assessment.
We have trouble with the notion that unwinding a century's worth of accumulated policies needs to be a staged process. The first task is to uproot anachronistic policies and structures to create room for smart new solutions to take root. (Think of the first decade of welfare reform in the 1980s). Only after a couple years in which we've given districts a chance to feel their way, and after a handful of alpha locales have crafted some promising approaches, does it make sense for state legislatures to start offering more direction. I blame a lot of this current "one fell swoop" mindset on Race to the Top.
As I've said before, "It's rarely been noted that RTT actually embodies two schools of reform. The first type of reform cracks open systems hampered by anachronistic statutes and policies." Measures such as those tackling tenure, step-and-lane pay scales, or data "firewalls," which prohibit the use of performance data to evaluate teachers, are all to the good. These measures don't tell states or local officials what to do with the newfound freedom; they merely create the space in which to act. However, as I observed, RTT also pursues "reform by specific prescription." I've noted that these prescriptive measures account for the lion's share of RTT. This, to my mind, is precisely what we're seeing as states hustle to emulate Tennessee's round one RTT-winning promise to have as much as 50% of teacher evaluation based on value-added calculations.
What matters most is uprooting the old intrusive superstructure, not the frenzied efforts to impose a new one. In particular, there are three big problems hampering the "fix it now" approach, even if one stipulates on the front end that all of the technical challenges to reliability can be readily resolved.
First, and least important, there's the question of small n sizes for many teachers and the challenge of devising good value-added assessment metrics for a raft of subjects. These challenges aren't necessarily susceptible to technical fixes and will require a little time, adjustments, and workaday compromises to take shape.
Second, systems built around individual value-added calculations can stifle the kind of smart personnel use that reformers are trying to encourage. Principals who have their best 3rd grade reading teacher step in for grade-level colleagues, rotate faculty by strength, or augment classroom teachers with niche providers or online instruction are going to clash with a system predicated on evaluating how teachers are doing in their classroom with their students. If teachers are only teaching their kids 75% of the time, or if key instruction is being provided by others, then the results will be muddy.
Third, none of this matters yet. Florida's SB6 went down over resistance to value-added calculations based on tests that don't yet exist and that won't exist until the middle of the next decade. Colorado's system won't be functional for years, given the need to create assessments and build out the data systems. Rather than using statute to impose systems that are currently nothing more than concepts, it makes more sense to strip out troubling mandates, create room for new solutions, and then build out the tests and data systems. This way, by the time those tools start to come online, we will have experience that can inform the deliberations of state lawmakers.
What to do right now? The best move is to insist that districts base a substantial part of teacher evaluation and pay on measures of performance, that states flag indicators deemed legitimate, and then that districts and providers have leeway to employ metrics. Over the next few years, this will begin to shift norms, allow value-added systems and evaluation models to develop and ripen, and set the table for adoption when value-added metrics are more readily available. If advocates are nervous that this is their big window, then they can write a trigger into the statute that will force the legislature or state board to revisit the question in 2013 or 2014.
I understand the frustration with the status quo and union resistance that has fueled "fix it now" thinking. I understand fears that nothing much will change if states don't mandate it. But K-12 schooling is a big, complex exercise. Large, hurried solutions have a way of working less well than hoped. Impatience and lashing out in frustration can lead to bad policy--as with NCLB's 100% target for 2014 and its Kafkaesque remedy cascade. Both helped undercut support for the law's more sensible provisions and are now consigned to the trash heap. The patient path of welfare reform--create opportunities, nurture and monitor successful efforts, and then talk about new policy requirements--is the more promising one.