The Tony Bennett flap
In light of the news of Tony Bennett’s resignation, Gadfly asked several top education-policy analysts to tell us what it means for school accountability going forward. RiShawn Biddle, Kevin Carey, Anne Hyslop, Kathleen Porter-Magee, Marc Porter Magee, Mike Petrilli, and Andy Smarick responded.
RiShawn Biddle: Four lessons to be learned from the Tony Bennett revelations
There are four key lessons reformers should learn from revelations that former Indiana Superintendent Tony Bennett made changes to the Hoosier State's A-to-F grading system last year affecting thirteen schools, including Christel House Academy South (whose founder was a donor to Bennett's unsuccessful re-election campaign):
1) Transparency matters: Accountability systems can only be effective when people can trust it. Bennett and his staff should have publicly revealed the grade changes last year and explained them thoroughly. It’s understandable that Bennett didn't want any of the problems with the system to hinder his other reform efforts (a matter with which he expressed clear concern). But the lack of transparency has put Bennett’s actions in an even worse light than it may deserve.
2) A-to-F grading isn’t ready for prime time. Because it doesn't accurately break out for families, especially from poor and minority backgrounds, how well schools serve their children in all aspects of learning, it isn’t very useful for making smart decisions.
3) Reformers must make sure that state school leaders take good care in developing their systems. In Indiana, the approach used to calculate scores for the A-to-F grading system is so convoluted as to make one's head spin. And this was even before the changes Bennett had made.
4) The fiasco is a reminder of why the Obama Administration, which approved A-to-F grading as part of its No Child waiver gambit, must embrace the subgroup accountability approach contained within the law. Because it simply-yet-sharply focused on how well schools served all kids, families and others gained the information needed to spur a decade of reforms that have led to 217,432 fewer fourth-graders being functionally illiterate in 2011 than eight years earlier.
RiShawn Biddle is the editor and publisher of Dropout Nation.
Kevin Carey: Falling short of equal justice under law
I live three blocks from the United States Supreme Court, where the words "Equal Justice Under Law" are chiseled in stone. It's bedrock principle of democratic governance and the first obligation of public officials. This is the idea that Tony Bennett failed to uphold in Indiana. In his tenure as schools superintendent, Bennett was a strong advocate for certain education-reform ideas, including charter schools. That was his right as an elected official. But he also had an obligation to apply Indiana's laws fairly, regardless of his personal relationships and preferences. It's clear from emails obtained by the AP that he was working backward from a pre-determined outcome in applying the state's accountability rules to charter schools he favored—even while, as the Indianapolis Star reported today, refusing similar accommodations to the regular public schools that he frequently and publicly criticized. That's the opposite of equal justice under law. The whole point of a rules-based school-accountability system is to let the data speak, regardless of whatever personal impressions and opinions local officials may have. The fact that Bennett offered a less-than-accurate account of how and why grades were changed to Rick Hess just yesterday underscores the impression that he wasn't playing fair. Once that line is crossed, it's very hard to regain public trust.
By definition, accountability systems exchange complexity for clarity and human judgment for objective rules. If you believe those exchanges are worth it—as I do, and presumably Bennett does—then you react to situations like Christel House with smart communication and reasonable intervention, not by subverting the rules themselves.
Kevin Carey is director of the education-policy program at the New America Foundation.
Anne Hyslop: A betrayal of public trust
In changing Christel House’s grade and those of twelve other schools, Tony Bennett undermined the very principles of school accountability that he sought to defend. Bennett wrote, “They need to understand that anything less than an A for Christel House compromises all of our accountability work.” In fact, changing Christel House’s grade from a C to an A is what compromised the work. Bennett’s “statistical anomaly” is, in fact, just another “loophole” in a series of previous loopholes states have used to avoid accountability.
This is unacceptable. Inflating grades denies improvement resources to schools and students that may legitimately need the extra support. It betrays the public trust that is essential to the legitimacy of these kinds of ratings, or any measure of school quality. And it’s false advertising to families in a state where school choice is a linchpin of education policy and A–F grades are touted as the best measure to use in making these choices.
School-accountability systems—including rigorous standards, assessments, and performance data—are already under fire by critics on the left and right. Bennett’s actions and those of his staff just make it all the more difficult to defend them. This is a sad day for Indiana, for Florida, and for anyone who believes in better and meaningful school accountability.
Anne Hyslop is a policy analyst with the New America Foundation’s education-policy program.
Kathleen Porter-Magee: School grades need to be more science than art
The purpose of the state evaluation systems is to set “objective” benchmarks, to judge schools against the criteria, and to let the chips fall where they may. In the days immediately following the Indiana flap, accountability supporters suggested that the process of setting accountability metrics “was more art than science.” Perhaps that’s true, but if it is, we need to rethink the role and form of statewide school grading.
If accountability is indeed more art than science, we need to be very careful about the consequences we tie to them. The quality of art, after all, is judged by personal taste, not by some unassailable objective measure. If that’s the case, then the accountability tied to the results should be driven more by choice than by top-down state mandates.
On the other hand, if we are going to tie serious consequences to the results of a statewide accountability plan, we need to be far more certain that the metric we’re using is more science than art. Then, we need to make sure there is a clear and transparent process by which the metrics are developed or changed, and we need to erect checks and balances that help preserve public trust in a system that so many parents, children, and community leaders rely on.
Kathleen Porter-Magee is senior director of the high quality standards program at the Thomas B. Fordham Institute and, later this summer, will become the senior advisor for policy and instruction at the College Board.
Marc Porter Magee: Is a single letter grade the right approach?
My graduate school advisor once told me that when weighing the results from a regression analysis, the most important factor was not what you saw on the page but, rather, how much you trusted the person who ran the analysis. There are so many ways you can tweak complex equations to get the results you are looking for. In the end, it’s the ethical—not the scientific—aspect of the work that’s most essential. I think the same holds for state accountability systems.
Trust took a big hit this week, not just in Indiana or Florida but also across the country. If we are going to make these systems work, state chiefs and their staffs will need to regain the trust of principals, teachers, parents, and the public through the rigorous and transparent way they develop and evolve these systems.
The Bennett scandal has also brought to light the challenges of putting a single grade on a school. I think we’ll see a real debate about whether that is a sustainable approach. Letter grades can be an amazing way to summarize performance. But when you shift from providing multiple grades for each of the key components of a school’s performance to providing just one grade, you risk masking subjective judgment calls about which factors to prioritize that perhaps are better left in the hands of parents.
Marc Porter Magee is the president and founder of 50CAN.
Mike Petrilli: School grades, stakes, and signals
A few days ago I urged people not to rush to judgment in the Christel House affair, and that remains my advice today. Even now, it’s not entirely clear what Tony Bennett and his staff did to raise the school’s grade, and whether they were careful (or not) to apply the same rules to schools in similar situations. Selectively reported emails are unlikely to provide a full or fair account.
What matters most, though, is how reformers react to the bright spotlight now on school-grading systems. To be sure, these haven’t exactly been out of view before; how we measure school success (or failure) has been a raging conversation in the policy community since before No Child Left Behind. Arne Duncan’s waivers were an admission that the law’s Adequate Yearly Progress approach was hardly infallible as a metric—and that states should have more leeway in designing measures going forward.
Yet we’ve never been good at explaining to the public that a school rating (whether A–F or anything else) is the sum of a multitude of judgments about what makes for successful (or failing) schools. For instance: How much should the attainment of a particular performance level (like “proficient” or “college ready”) count, versus students’ progress over the course of the year? Which subjects are included in the measure? Are they weighted equally? Should all students’ scores count the same, or should we put greater stress on those of the lowest-performing kids? What about high-performers? What role should the size of a school’s achievement gap play, and how should we measure that? Should the performance of students with cognitive disabilities count? Children who are new to the country and can’t yet speak (or read) English? What about graduation rates, or attendance data, or other indicators? And, quite relevant in this case: If a school spans the elementary, middle, and high school grades, should it get one grade or three?
There are no “right” answers to these questions—which is one reason the federal government should leave these decisions to the states. But it also implies that school grades might be more appropriately seen as “signals” to parents and taxpayers than as unassailable evidence that, for instance, a school deserves to be closed.
Mike Petrilli is executive vice president at the Thomas B. Fordham Institute.
Andy Smarick: First, think about the kids
I first met Tony Bennett over the phone in 2009. He called me at the Fordham offices to talk about something I wrote regarding Indiana and Race to the Top. He was very sharp, extremely reform oriented, strong-backboned, and—in fine Midwestern tradition—decent and straightforward. Every interaction I’ve had with Tony since has only reinforced my early impressions.
So my initial reaction to his resignation was personal: I’m terribly disappointed for him and his family; these things are grueling to go through. I hope people have some level of sympathy. Of course, we should also be disappointed and frustrated if additional news, beyond this excellent analysis, makes it appear that this was more a case of playing favorites than fixing an error.
But—without going all cliché on you—we need to think about kids right now. First, Indiana’s: The reforms Tony advanced in the Hoosier State were invaluable. The student-performance progress that state has made is notable. The Christel House situation puts Indiana at a fork in the road. Do they use this as a reason to roll back the last era’s reforms? Or do they see this as an isolated incident (big or small, depending on what is ultimately revealed) that won’t be allowed to undermine the smart path they are on? I’m more than a little worried that retrenchment may be in the offing. For example, consider Governor Pence’s recent statement pulling back on the Common Core and common assessments.
This whole thing is also very unfortunate for Florida, a state that was setting the curve in recent years. It’s now gone through five chiefs in fewer than four years. That’s a mess. States that have made progress in recent years (e.g., Massachusetts and Maryland) have had great continuity in educational leadership and strategy over a long stretch of time. All this Tallahassee turmoil serves to distract. I’m sorry to say that Florida is likely to have a hard time landing a new chief. Potential candidates will look warily upon this environment.
And we need to keep an eye on what this means for Common Core and PARCC. Bennett’s been among the nation’s strongest CCSS backers and maybe the leading conservative supporting the standards. He’s now left without a public perch—at least for the time being.
The influence on PARCC, I fear, will be even greater. Florida has served as PARCC’s fiscal agent. Tony has been a driving force behind PARCC from the start. PARCC is losing Tony as a chief and maybe Florida as a member.
My last thought is about these new A–F accountability systems that are at the root of this entire situation. They try to simplify into a single letter grade an entire school. That makes it easy for consumers to get a quick take on a school. But it obviously obscures lots of nuances. Crafting these systems is enormously challenging. There is no “right” answer for which components ought to be used and how each should be weighted. States are trying their best to come up with fair, transparent systems, but no matter what, there will always be questions about how final grades shake out. That’s an unfortunate but inevitable consequence of these things.
Obviously, if any state official tampers with the formula to help a favored school, that is unethical and undermines public confidence in accountability frameworks. But sometimes the decisions made are actually about good policy, not sinister politics. I know that today’s hyper-cynicism will favor the latter interpretation in most cases. But we have lots of deeply committed state chiefs trying to do right by kids. Regardless of how the Indiana case ends, let’s try to give some of our leaders the benefit of the doubt.
Andy Smarick is a partner at Bellwether Education Partners and a Bernard Lee Schwartz senior policy fellow at the Thomas B. Fordham Institute.