Yo Teach…! Or how to avoid teaching like Jason

Closing the Teach For America Blogging Gap
Aug 20 2013

Why nitpick about charter school data when there are so many low performing district schools?

I had the pleasure last night of meeting someone in the teachforus comments who was to my right on educational policy issues. The rarity of such an occasion speaks more to those who populate this website than to my own beliefs, which I would describe as disagreeably centrist. Matt, this commenter, challenged me in a way that I found worthy of reflection, and therefore, a blogpost. He wrote :

“I would like to see you dedicate the same analytics towards the hundreds of failing publics that go unnoticed for generations. We have a few hundred schools with passing rates less then 15% yet not a word bout them gets passed on these blogs, yet charters who almost always beat the co-located schools get massive amounts of criticism.“

Why do we–let’s say, the “blogosphere”–generally focus on nitpicking data of the most prominent charter schools, rather than the schools failing the most students? Here are the reasons I can think of:

1. It’s hard to know which public schools are failing students, and which just serve a higher-need group of students. Everything Mike Petrilli and Seth Andrew said about how schools that have higher-need students are biased against in the most prominent proficiency data applies doubly so for district elementary and middle schools. As a former first, second, and fourth grade teacher, I can attest to the facts that some academic and social gaps are not going to totally diminish by third grade even with generally above average teachers. Annual growth measures may help with this, as do attempts by states and cities to give schools grades that reflect this demographic data (as well as some observational data), but often these more complex indicators miss another important factor: trajectory. Are these schools improving? Are strong and capable leaders in place that are fighting the long, hard, slow battle of taking a historically underperforming school and improving it? Life would be easier if there were indicators that reliably and unambiguously told us which schools were lost causes or getting worse, and which were on their way up. But there are not.

2.  We don’t have access to enough information. Most of us can’t get inside failing public schools, can’t influence behavior of public school leaders, and, most importantly, don’t have enough information to know exactly what would work even if we could single out the lowest performing schools. Schools are bad for a variety of reasons, ranging from corruption, to under-resourced staff, to inept leadership/organizational culture, to recalcitrant teachers, to bad curricula, or any combination of the above (and hundreds of reasons I’m certainly leaving out). Each one of those problems has a set of twenty or more possible responses policymakers and administrators can’t agree upon. Why? Because it’s very difficult gathering research on school improvement given the heterogeneity of problems schools face, confounding variables, and outcomes we could measure. It’s also hard to differentiate between well implemented and badly implemented responses in large-scale studies. So the result is, the only people who may have the faintest idea about how to improve these schools have intimate relationships with the communities of staff and students surrounding the organization. In general, I think the people who criticize the publicized charter data are of this belief: that schools should not be micromanaged from afar. And that is also true for charter schools. We are not trying to tell them how to operate, we are simply trying to correct assumptions about their success. See point 4 for more.

3. Even if we knew what to do, the history of education reform teaches an important lesson: capriciousness is antithetical to school improvement. Read Charles Payne’s So Much Reform, So Little Change or Rick Hess’s Spinning Wheels, the books come from different perspectives but tell a similar story: At the policy, state, research, and district level, education is ruled by cycling fads. These fads can be organizational, pedagogical, behavioral, but they all involve a school administrator or external consultant telling teachers what they’ve been doing their whole life is wrong, teaching them a new way of instructing, and then coming back a year or two later, right when the teacher or school is perfecting this new method, to say that everything they know is wrong and the new best practice is something totally different. I had this conversation every day with teachers at my school, and when I asked about my grandma (a 40 year NY public school teacher) about her experience decades ago, that’s the first thing she mentioned (she continually told phonics was either horrible or the only thing worth teaching, and would just do everything she could to teach at her best behind closed doors). So how is this related? Because, if we were to demand changes from every struggling school based on annual test scores, they would be too busy adjusting to our changing demands to implement any kind of meaningful changes. Moreover, there’s Peter Pan phenomenon most teachers can attest to, instead every time we are micromanaged from afar, a little part of our professional motivation drops down dead. Compound this with the points from part one and it’s an even more perilous enterprise.

4. It is *vitally* important to discuss and analyze and nitpick about charter school data because the practices of successful charter schools spread across the country, and not enough people are looking critically at which charter schools are being copied and how. In an educational market, change still happens from the top-down. Rarely (but not never) do innovators have some new idea of about how to educate students and organize schools based on family needs or holes in the market and start up a school. This is for a few reasons. One is there are a lot of upfront costs in starting a school, which can be prohibitively expensive if you don’t have access to foundation money, and most foundations only support schools that align to an already established philosophy. Second, you need a lot of political connections to receive a charter to start a school (either you must be tied to existing charter schools or some other foundation of importance). This environment engenders institutional isomorphism: all practices at the organizational and classroom level tend to resemble each other. Because when the media or policymakers or politicians talk about successful schools they almost always are alluding to one of five charter networks, these practices indirectly spread across all school systems. TFA uses Uncommon school’s professional development. Almost all schools use “Do Now’s” and “exit slips” (classroom techniques of apocryphal origin I am sure began in some charter school). Low performing schools get turned around or taken over by charter networks or leaders of charter networks. This is currently happening in Detroit all over, and it’s happening with Democracy Prep, which is starting up/taking over a number of elementary schools. This isn’t necessarily bad, but it certainly means we better be sure that these models are not just successful in the best cases, with the best leaders, but also exportable. When I question the quality of certain charter schools, the stakes are high because these practices are likely spreading across low-performing public schools in a number of direct and indirect ways. When Mike Petrilli, one of the most respected voices in education, defends (in rhetoric, not data) Democracy Prep in spite of the evidence, he is contributing to an environment where continually exported best practices are assumed to be good, and therefore less likely to be improved or replaced by something better. Maybe it’s Uncommon or Success that should be exporting their models. Maybe it’s a high achieving public school. When we are not critical of the very top, we are essentially being uncritical of the practices spreading to our most at-need students. Maybe more of a market system would fix this problem. I doubt it, but that’s a debate for another day.


All that being said…

1. It’s important to apply our critical gaze to all schools being held as replicable examples, public, private, magnet, or charter. I do think many bloggers on this platform are too eager to disprove miracle charters and hail miracle district schools. I think that is analogous Mike Petrilli’s actions.

2. It’s important not to lose sight of the big picture. I may diverge from many others on this platform in this regard, but I think we should not forget that even mediocre charter schools are often performing above average in terms of many indicators, but going back to point one, that doesn’t take into account growth, student profiles, or trajectory. Any reader of my blog knows the loathing I save for the charter school that turned me off from teaching, but even I admit that the students it served were better off there than at a typical Detroit Public School. But that does not mean we can just import this charter school’s practice, shrugging “It’s better than the alternative!” People don’t improve through that method, organizations don’t improve through that method, and, most importantly, I believe it sets our education on a path dependency towards deprofessionalized organizations and mediocrity. Still, let’s put our criticisms of high performing charters in context.

This is a graph of elementary NYC public (red) and charter (blue) schools.On the x axis is peer index (0-100, lowest need to highest need), on the y axis is 2013 common core score (adjusted to fit with 2012 scores). The first formula  on the right applies to public schools, the second charter. Keep in mind, this is proficiency data, so all the caveats above apply. Ignoring that, if we look at the lowest performing schools, a handful are charter, and much larger handful are public. Charter schools also seem very capable of educating students between a 45-55 on the peer index scale (ironically, there charter schools that seem to do the most cream-skimming are also the least effective- I’ll dig into that another day). But the lowest performing public schools are on average serving a much higher-need population (60+ on peer index). A population zero charter schools in the sample show any effectiveness at serving. So we can conclude that yes, charter schools are on average doing well, but they have not shown an ability to run schools with the most at-need students. Most of us write because reformers like to ignore the second half of the previous sentence.

2 Responses

  1. Gary Rubinstein

    Nice graph! It shows that charters don’t, in general, serve the ‘same kids’ as they are grouped together in that small band, but that for that small band, some are doing pretty well.

  2. yoteach

    Thanks! It took hours of painstaking copy/pasting and vlookups to combine a few different NYC excels that of course had just enough inconsistencies to drive you crazy. But yeah, I would like to see more charters to discuss transparently how they plan on helping students in the bottom right, or at least compare themselves with the schools in the middle.

    Though, all this research was making me wonder if it’d be better to just get rid of that bottom right corner altogether by having all schools target a mid-range peer index, so that classes are a bit more diverse. I wonder your thoughts on what the top left third of public elementary schools can do, if anything, to play a role.

About this Blog

Pontifications of the Unplaced


Subscribe to this blog (feed)