The Problem with One-Size-Fits-All Approaches to Teacher Quality
Today's debates over teacher evaluation mostly just leave me tired. On the one side, we've got "reformers" who've accurately identified real problems, suggested sensible principles (like we should work to identify teachers who are better and worse at their jobs)... and then rushed to champion crude, inflexible policies that turn good ideas into caricatures.
On the other side, we've got teachers and "public school defenders" who aren't content to challenge simple-minded solutions, but who argue that we can't really distinguish good educators from bad ones...and ought to instead spend lots of time worrying about whether teachers are happy.
I've no use for either camp. I don't want to split the difference, find a "middle course," or any of that. I think that both camps just get it flat wrong. It's good to identify problems and to respond. But it's a mistake to imagine that those responses can always be translated into policy solutions.
On Wednesday, in a terrific piece, the Washington Post's Bill Turque penned a page one story on Sarah Wysocki, a former DC Public Schools teacher who was terminated under DC's IMPACT teacher evaluation system because of low value-added scores. Wysocki may or may not be a good teacher. She sounds good in the account, and had promising evaluation scores. But there are certainly reasonable concerns about whether her value-added scores were compromised by cheating that may have inflated fourteen students' prior year scores--as well as all the usual questions about how much weight we want to put in these numbers, and how fully we think they reflect a teacher's performance.
Especially intriguing is that Wysocki was quickly snatched up by neighboring Fairfax County, one of the nation's highest-performing school systems. Fairfax superintendent Jack Dale told me on Wednesday that he questioned how much faith to put into Wysocki's value-added scores, and added that Fairfax is focused on more than just a teacher's reading and math scores. Dale said that Fairfax parents probably regard instruction reading and math as no more than "twenty to twenty-five percent" of what they expect from their schools and that, "More and more parents are saying we test too much, so they're not really looking at those test scores that much... I see less and less emphasis on test scores now than even a couple years ago."
I grew up in Fairfax and know the system fairly well. It's fair to say that Fairfax families generally chose the system's schools because they desire a broad emphasis on science, world languages, gifted programs, music, and an array of aptitudes that those assessments don't capture. DCPS's laser focus on reading and math gains, and the assumption that these scores are a good proxy for a teacher's general performance, doesn't make sense in the Fairfax context.
At the same time, that doesn't mean they're wrong-headed for DC. The DCPS leadership is trying to turn around a historically low-performing system that has long failed at even its most basic responsibilities, and where vast numbers of students lack rudimentary skills in reading and math. Thus, it's certainly reasonable for DCPS to build a teacher evaluation system that seeks to base 50 percent of a teacher's evaluation on reading and math value-added.
In other words, neither Fairfax nor DCPS are necessarily right (or wrong). Rather, they're confronting different challenges and needs, and trying to make reasonable choices about how to proceed. The world is a complex place and adopting mechanistic, one-size-fits-all solutions, like so many of the statewide teacher evaluation and pay systems being championed today, make it likely that thousands of schools and millions of teachers and students will be snared by systems that are a poor match for their needs.
Also on Wednesday, MetLife released the 28th annual Survey of the American Teacher, which reported that 44 percent of the nation's teachers are "very satisfied" with their jobs. This was the lowest reading since 1989. The decline in satisfaction occasioned the usual hand-wringing and angst. (At the same time, 77 percent of teachers said they're treated as professionals by the community, suggesting that those who claim teachers feel under assault may be exaggerating just a wee bit.)
I don't get the angst. Why? This is just another face of the one-size-fits-all problem. I don't care that teacher morale is down in the aggregate. I would care if we knew that morale is lousy among teachers who are doing a good job and working hard. I want those teachers to feel valued, energized, enthusiastic, and all that. On the other hand, if a teacher is lousy or doing lousy work, they should have lousy morale. Hopefully it'll encourage them to leave sooner. And we've got plenty of reason to worry about teacher quality in the aggregate. For instance, 29 percent of teachers say they are likely to leave the teaching profession within the next five years--up from 17 percent in 2009. Is this a good thing or a bad thing? I think the right answer is: neither. Getting it just right, Regis Shields, director of Education Resource Strategies, told Ed Week's Liana Heitin, "We need more information on who the 29 percent" are. Shields said, "If these aren't effective teachers and this increases the effectiveness of the teaching force, that's great. If they're high-quality teachers, then we have some concerns."
And that's about as smart a response to one-size-fits-all thinking as I can muster.