Do Small High Schools Truly Work? Part MMM: Last week’s Dropout Nation commentary on Bill Gates’ wrongheaded opposition to publicly releasing teacher performance data elicited plenty of responses. One of them came from charter school operator and ed school founder Michael Goldstein, whose MATCH schools in Boston are among the exemplars of high-quality teaching and curricula for all students. But Goldstein spends less time on the arguments of the piece (which he enjoyed) than on my statement that Gates and his foundation had also gotten it wrong on its abandonment of the small schools effort it had pushed during the early part of the last decade.

From where Goldstein sits, I’ve wrongly “jumped in” with  George W. Bush Institute scholar Jay P. Greene in declaring that the report from research outfit MRDC, which looked at New York City’s approach to implementing small high schools, vindicated the concept. As far as Goldstein is concerned, my statement (along with Greene’s own conclusion), fails to consider failed implementations of the small schools approach in cities such as Boston; he also argues that Greene’s position that randomized studies of small high school implementation such as MRDC’s were more valid than non-randomized studies that yielded different conclusions. Writes Goldstein: “I take a back seat to no one on valuing randomized studies… but it’s crazy to leap from that to ONLY randomized studies mean ANYTHING.”

Your editor isn’t going to spend much time weighing in on the debate over the value of randomized and non-randomized studies. While I have my own thoughts on the issue, I’ll let Goldstein and Greene argue among themselves over that one. What I will say is that there is enough research, including Stephen F. Austin State University scholar Lee Stewart’s 2009 study, to show small schools can be effective in improving student achievement.  While the overall evidence doesn’t show that small schools in itself is an effective school reform strategy, there is evidence that it can work as part of a wide array of effective measures. This is something I should have said more clearly in last week’s piece.

Which hits upon one important issue that Goldstein rightly (if inadvertently) hits upon: That there is a tendency among some reformers to simply embrace one strategy as the proverbial silver bullet. It was clearly a flaw in the Gates Foundation’s approach to the small schools concept. Much of the dollars Gates spent on small schools went to districts which did little more than just break up large high schools into smaller operating units without revamping reading, math and science curricula; overhauling how teachers were hired, evaluated and rewarded; ditching failed approaches to selecting and training principals; and developing new ways to make parents lead decision-makers in schools. This was the true in the case of Indianapolis Public Schools, where then-Superintendent Duncan “Pat” Pritchett merely moved the proverbial deck chairs on that Titanic disaster of a district without doing much else; his successor, Eugene White, then abandoned the small schools effort as part of his own disastrous tenure.

Reducing school size certainly did contribute to New York City’s success with small schools. But so did other efforts undertaken by the city that flowed into that effort. The city shut down more than 91 failure mills between 2002 and 2010 and replaced them with new high-quality traditional and charter schools. The city also became more aggressive in improving teacher quality, and also improved the quality of school leadership across the district. In short, it wasn’t just about breaking apart small high schools. It was also about revamping a district that had been an abject failure for its students for at least four decades. This is an important lesson Gates and other reformers should embrace as part of their reform agenda.

Another issue which does come up in this discussion is the importance of patience in measuring the success or failure of a reform strategy. Education traditionalists rightly criticize many school reformers for being too quick to embrace a potential solution and being even quicker to ditch it when it doesn’t immediately yield fruit. This is already happening in the charter school arena, in which the idea of increasing diversity within the ranks of charter school operators (including bringing in community groups and recruiting black and Latino charter school operators) is being pushed aside for a stronger focus on scaling up successful existing operators such as KIPP and Green Dot. Why? Because reform-minded philanthropies, concerned about improving school quality (and dismayed that far too many charters are not succeeding in improving student achievement) have decided that it is better to put their money on successful incumbents on the ground. While such thinking is understandable, it fails to consider that perhaps the real problems of quality lie more with issues such as how authorizers select potential charters than with the efficacy of encouraging diversity itself. The fact that scale is not all that people think it should be (and that American public education has already that scale is more often a recipe for failure than success) also doesn’t foster into the thinking of reform-minded philanthropists.

Certainly we should be impatient in demanding and pushing for systemic reform. At the same time, reformers and the philanthropists who back them need to be patient and thoughtful in achieving results. This is tough because we are dealing with the lives of children, who only have one shot every day to get the high-quality education they need for success in life. But sometimes, in education, patience is a virtue.

Petrilli Shoots and Misses on Contrarian Thinking: One of the flaws of so many Beltway reformers is the constant pursuit of provocative contrarianism. Looking to distinguish themselves as being different from the crowd, otherwise thoughtful think-tankers end up writing polemics that may appear to the less-observant as being new takes on longstanding issues. But upon closer examination, the flaws in the thinking point back to more-conventional (and right-headed) positions.

Thomas B. Fordham Institute research czar Mike Petrilli indulges this constant pursuit of contrarianism yesterday on the think tank’s Flypaper blog, when he took a look at the New York Times‘ analysts of the recently-released data on performance for 18,000 New York City teachers. Playing off a weak argument he made two years ago, Petrilli declares that the Times analysis proves his view his fellow reformers are wrong in embracing “the notion that our achievement gap is caused in large part by a “teacher quality gap,” with the worst teachers clustered in the neediest schools”. From where Petrilli sits, reformers can now “we can take “closing the teacher effectiveness gap” off our to-do list” and move on to addressing other issues.

Petrilli’s conclusion is off the mark. Certainly we now know that the quality of teaching can easily vary from one classroom to another. Data also reveals that there can be greater variations in teacher quality within schools than across them. But the latter conclusion is based solely on looking at schools across a state level, as the National Center for Analysis of Longitudinal Data in Education Research showed in its 2010 report. Given that inter-district school choice remains a rarity for four out of every five children in this country, it means that there are open questions about variations of teacher quality within school districts, where teacher assignment policies governed by collective bargaining contracts and state laws are more likely to be a more significant factor than at the macro level of a states. We also must consider how kids are actually assigned to teachers within schools. Given the rationing of education through gifted-and-talented programs, and other gatekeeper-controlled regimes, it is just as likely that high-quality teachers within a school are only working with children that are considered top-performing; essentially, poor and minority kids, who are often denied opportunities for comprehensive curricula, are probably losing out. Essentially, Petrilli is offering a conclusion that isn’t supported by available evidence. (Editor’s Note: One even less supported by data when one learns that the city’s rating system is essentially designed to find high-quality and low quality teachers within schools, and not across the district as a whole.)

Meanwhile Petrilli fails to consider is that a decade of reform efforts under Mayor Michael Bloomberg and a series of chancellors (including the legendary Joel Klein and current schools boss Dennis Walcott) may have actually improved the distribution of teacher quality. Thanks to rules that allow principals to remove laggard teachers from their schools (and rhe infamous rubber rooms they helped foster), more has been done to improve the quality of teaching in Big Apple schools. Add in the aggressive effort to weed out new hires before they attain tenure, and the addition of more than 10,000 new teachers in the past decade (many of whom the city selected with the help of reform outfits such as TNTP) and it is quite likely that the city has managed to improve distribution of teacher quality across the district. If one just measures teacher quality just on qualifications — which, by the way, has almost no correlation to impact on student achievement — a 2008 report by researchers led by Donald Boyd of the State University of New York, University at Albany, shows that New York City has at least done a better job of choosing teachers to go into the classroom.

What the Times analysis shows is that perhaps New York City’s reform efforts have been far more successful given the constraints (including strong opposition from the American Federation of Teachers’ Big Apple affiliate) than either education traditionalists or reformers skeptical of Bloomberg’s efforts (notably Petrilli) will admit. Given that we have no similar data on teacher performance from 2002 (when the district was placed under mayoral control), it is almost impossible to know. But when one looks at the performance of Big Apple students on the National Assessment of Educational Progress (which I discussed in Dropout Nation‘s analysis last November on the effectiveness of mayoral control of districts), it can be surmised that the city has made good progress so far.

There is other data that Petrilli should have considered. This includes the Education Trust’s recent report on Los Angeles Unified Schools, which shows that black and Latino children are three times as likely to get a low-quality teacher in reading (and twice as likely to get a low-quality math teacher) than white and Asian counterparts. There’s also the Los Angeles Timeslandmark series on teacher Quality in L.A. Unified’s elementary schools, which revealed, among things, that schools serving poor and minority kids were more-likely to lose high-quality teachers because of reverse-seniority layoffs. There is also the data out of Washington, D.C., which shows that . And then Petrilli should consider the U.S. Department of Education’s own survey on teacher distribution based on Value-Added analysis of teacher performance, which shows that there is wide variation of teacher quality distribution within districts, with some equally providing high-quality teachers to all students and others not doing so at all; the study also suggests that the differences in distribution of teachers widens once kids enter middle school while children regardless of background at the elementary school level are more-likely to get equal access to good-to-great teaching. Considering that Zip Code Education policies still restrict most students to zoned schools, it means that far too many children from poor and minority households can’t access high-quality teaching — and that their families can’t simply buy their kids into good-to-great classrooms.

This isn’t to say that Petrilli may not have a kernel of a point about access to high-quality teaching. He is correct that simply moving good-to-great teachers around to struggling schools will not work; overhauling how we recruit, train, compensate, and evaluate teachers would be far more effective in expanding the pool of good-to-great teachers working in all schools. He is also right in noting that schools serving kids from middle-class and affluent backgrounds aren’t necessarily providing high-quality education; studies such as the George W. Bush Institute’s Global Report Card have clearly made that point. But Petrilli should be more careful to avoid the kind of contrarian-and-provocative-just-to-be -contrarian-and-provocative approach embraced by a few too many Beltway reformers who should be more-thoughtful in their polemicism.

The Dance that is the “Student Success Act”: Your editor has already said enough about House Education and the Workforce Committee Chairman John Kline’s latest efforts to gut the No Child Left Behind and the Adequate Yearly Progress provisions that have spurred much-needed systemic reforms. Considering that today’s markup of the bills are meaningless (because the reauthorization of No Child won’t likely happen until next year), why bother devoting more time to it? All I’ll say today is that the Student Success Act is most inaccurately-named piece of legislation coming out of Congress this year. And does anyone really expect the markup to be anything more than kabuki?