Teacher evaluation was one of President Obama’s signature policies, and a controversial element of education reform during his tenure. With the passage of the Every Student Succeeds Act (ESSA), which does not require states and districts to implement performance-based teacher evaluations like No Child Left Behind waivers did, teacher evaluation policy has largely fallen out of the public narrative. But that does not mean states or districts know how they are going to proceed with teacher evaluation policy—in fact, its future remains unclear in this new era of lessened federal oversight.
In December 2016, Bellwether Education Partners and The Thomas B. Fordham Institute independently released two reports centered on teacher evaluation and its consequences. Bellwether’s report summarizes the teacher evaluation policy landscape and points out potential risks for teacher evaluation in the wake of the passage of ESSA. The Fordham Institute’s report studies twenty-five districts to determine if those districts can terminate veteran teachers once evaluation systems have deemed them ineffective.
Both reports offer a glimpse into ongoing challenges and opportunities with teacher evaluation reform, but they have very different analyses. To understand our different approaches and the places where we might overlap on teacher evaluation policy, Bellwether and Fordham hosted an email conversation between the report authors. Below is a transcript of the exchange between Bellwether’s Kaitlin Pennington and Sara Mead and Fordham’s Victoria McDougald and David Griffith.
Fordham’s report emphasized dismissing ineffective veteran teachers and Bellwether’s report highlighted how the field has switched the focus of teacher evaluation to professional development. Are these aims incompatible? Understanding the core purpose of teacher evaluation systems is especially important as states and districts consider making changes in the ESSA era. Each exchange below ends with a question for the other organization’s authors to respond to. The conversation has been lightly edited for clarity.
---
Bellwether’s Kaitlin Pennington and Sara Mead: When we wrote our report, we grappled with what policymakers, teachers, and the public see as teacher evaluation’s purpose or “theory of action.” From the onset of reforms, there was inconsistent messaging. Some said teacher evaluation would be used primarily to inform employment decisions. Others said the systems would inform teachers’ professional development. Over time, many said the systems would do both. The inconsistent messaging about the purpose of teacher evaluation enabled opponents to define the systems as primarily punitive measures targeting teachers.
This led advocates to lose hold of the teacher evaluation narrative. Although very few teachers are rated ineffective and—as you write about in your report—even fewer actually lose their jobs because of poor evaluation ratings, many teachers currently view performance-based teacher evaluation systems as mechanisms to harm them.
But focusing narrowly on ineffective teachers may be the wrong emphasis. We believe that teacher evaluation is important because these systems can be used to define common expectations for effective teaching practice, facilitate data-driven conversations about instruction, and reward effective teaching in order to retain the most-skilled teachers. In the past several years, many states have used the systems to do this—in fact the most differentiation of teacher practice is happening at the high end of the spectrum between “effective” and “highly effective” teachers.
So while it is troubling that the systems have not exited poor performing teachers from the profession like many early teacher evaluation advocates hoped they would, the reformed systems have made progress over the binary systems they replaced. Yet the focus on firing teachers continues to dominate the public narrative.
Now because the Every Student Succeeds Act allows for more changes to teacher evaluation, many states are trying to rid their systems from connection to the “firing bad teachers” narrative. They are doing so by making it even harder to objectively determine ineffective teaching (and exit teachers because of it) by reducing or eliminating the main quantitative data point in many teacher evaluation systems—student achievement and growth data. The loss of this objective data to balance subjective measures like classroom observations and student and parent surveys negatively impacts the reliability of the systems. Moreover it reinforces a troubling age-old message, that great teaching is determined by teacher input and is divorced from student output. This has implications for the future of the teaching profession and the human capital talent interested in joining a profession that is evaluated on these metrics.
When you wrote your report, did you consider teacher evaluation’s theory of action? Do you think performance-based teacher evaluations will survive in the ESSA era if the narrative around them continues to focus on firing ineffective teachers?
Fordham’s Victoria McDougald: While teacher evaluation and dismissal are obviously closely intertwined, our report—Undue Process: Why Bad Teachers in Twenty-Five Diverse Districts Rarely Get Fired—shines light on the latter, equally important yet far less-studied issue. In it, we ask: after nearly a decade of teacher evaluation reform, is it any easier to exit an ineffective veteran teacher from the classroom? Dismissal data are notoriously difficult to come by, but as recently as 2013, just 0.2 percent of tenured teachers in a typical state were dismissed for poor performance.
After combing through collective bargaining agreements, employee handbooks, and state laws, our study yielded bleak findings. Overall, we found that dismissing underperforming teachers remains far too hard (in all twenty-five diverse districts included in our study, significant barriers remain in place to do so).
Should the conversation around teacher evaluation focus exclusively on firing ineffective teachers? Certainly not. Well-designed evaluation systems can and should provide teachers with valuable information and lead to appropriate supports to drive improvement. But dismissal has to be part of the conversation. Once teachers are proven to be ineffective, they simply should not be entitled to lifelong employment, given the decades of research showing what that means for students’ achievement and long-term prospects (such as college outcomes and future earnings). Based on what we learned in our study, it’s imperative that we make the dismissal process simpler and easier for districts. We can’t shy away from what’s best for kids because we’re afraid of angering some educators and teacher unions.
As for whether teacher evaluations will survive ESSA unscathed, it’s possible that a handful of states may follow New York and Oklahoma’s lead and backpedal on teacher evaluation in the face of political pushback. But my hope is that the majority will stay the course, and that those that do are committed to using student achievement and growth data to inform evaluations, support and develop educators, and exit the chronically lowest performing teachers from the classroom (regardless of how long they’ve been on the job). One of the few bright lights in our study was that all of the districts and states we examined required some measure of student performance as one component of their teacher evaluations. It would be a shame to lose that hard-fought ground.
Fordham’s David Griffith: Let me piggy-back on what Victoria wrote and then offer a (slightly) different perspective on the topics she raises in her final paragraph. As far as our theory of action is concerned, I agree that dismissal has to be a part of the conversation. And in fact, I would go even further: To me, evaluation runs the risk of being meaningless (and potentially even counter-productive) if nobody can be fired, nobody can get a raise or promotion for doing an exceptional job, and the whole process is so formalized and bureaucratic that supervisors spend all their time worrying about compliance instead of actually talking to people. That characterization may not apply to D.C., but I think it describes a lot of places. Put differently, evaluation is important, but if you gave me a choice between reforming evaluation and reforming the dismissal process I would choose the latter every time. Meaningful consequences are more likely to lead to meaningful evaluations than vice versa.
As for teacher evaluation systems that incorporate test-based value-added, I agree with Victoria that we’ll probably lose a few but keep the rest. And despite what I wrote in the previous paragraph, so long as we’re stuck in a district-based system [as opposed to a system of charter schools that gives principals greater autonomy], I think that’s probably for the best. So yes, test-based value-added is a valid measure of teacher quality, and getting rid of it entirely would be a mistake. But having said that, in the long run, I think we can improve on current evaluation formulas by moving away from formulas altogether, because they often tell us things that we already know. As you acknowledge in your report, there’s a strong case for locating test-based accountability at the school level rather than the classroom level. So in my ideal world, states would send teachers’ value-added scores to principals, who could use them as they saw fit. (In other words, they would use them informally to augment their professional judgment.) After all, principals know their kids, so they know when Mrs. P had a particularly tough group of students or Mr. J was working with outdated textbooks that weren’t aligned with the test. To me, having this sort of context for value-added measures makes them more reliable rather than less. But of course, all of that assumes that principals are empowered in ways that are almost impossible under the current system.
OK, now here are our questions for you…
- It’s clear from the report that you think existing teacher evaluations have done at least some good by encouraging low-performing teachers to exit the profession voluntarily and promoting substantive dialogue between teachers and principals. Current politics aside, what would make these systems more effective or useful to teachers than they are now? What’s the next step?
- If the primary purpose of teacher evaluation systems is to inform professional development, then why are value-added measures so important in teacher evaluation? Do teachers and principals actually use them for this purpose? Or is their true purpose to ensure some amount of differentiation?
Pennington and Mead: Our answer to both of your questions requires a larger conversation about the intersection of professional development and teacher evaluation. In doing research for our report, we saw a trend in the post-ESSA era: many in the education reform community began shifting the teacher evaluation narrative to focus less on making employment decisions to influencing teacher professional development. This shift likely has to do with the fact that the “firing bad teachers” narrative gained more adversaries than it did advocates.
For those of us who believe that accountability is necessary for teacher evaluation systems to identify effective teaching, we need to grapple with the changing tides and figure out a way to marry both professional development and employment accountability in the same system. As we caution in our report, focusing solely on professional development has two major risks:
- Current evaluation systems built around a small number of observations and little feedback are not likely to contribute much to teacher professional growth. Instead, prioritizing professional development would require a completely revised teacher evaluation system created and implemented with a focus on professional development at the forefront.
- It’s not clear that we have the research or knowledge base on how to create evaluation systems that support professional development and lead to improved practice and student learning. In fact, research suggests that most current teacher professional development systems do little to improve teachers’ practice. And without teacher evaluations or other objective ways to assess teacher performance, it will be difficult to know if professional development affects the quality of instruction.
So to your first question about what would make teacher evaluation systems more effective and useful to teachers, we think the policy and practitioner communities need to come together to address the elephant in the room: how to make professional development useful and tied to teachers’ identified needs. This will require significant, intentional investments in the design and capacity of teacher evaluation systems instead of trying to use current systems for evolving needs. In our report, we spotlight a charter management organization, Achievement First, which has been successful in marrying a performance-based teacher evaluation system with professional development that helps teachers improve. But as TNTP’s Mirage report shows, this success is possible in part because leaders and teachers at the CMO invest many more resources into the system than most districts do in their evaluation and professional development systems.
To your second question about the role of value-added measures in a teacher evaluation system that focuses, in part, on professional development, we think value-added measures serve as one measure to differentiate teacher practice. This is important because they connect teachers’ practice with student outcomes. But value-added measures will not be helpful in directing teachers and the leaders who support them to decipher the types of development activities they’ll need to improve their practice, and by extension, their value-added measures. That’s where the other measures and resources, like observations, surveys, coaching, etc., come into play.
To be clear, however, as we write in our report, if states move more toward teacher evaluation systems that are focused on professional development, practitioners and policymakers must consider how the loss of accountability “teeth” will affect practitioners’ implementation of and response to these systems. It’s not clear that teacher evaluation systems will be taken seriously in the absence of pressure. As you write, the systems could quickly become meaningless.
This conversation serves to underscore another point we made in the paper: If we’re serious about ensuring that all children have access to quality teaching, we can’t afford to focus solely on one lever for improving teacher effectiveness, whether that lever is teacher evaluation, professional development, or dismissal. It’s simply not possible to develop or dismiss our way to effective teaching. Rather, efforts to improve teacher effectiveness must consider and address the entire human capital ecosystem—including recruitment, preparation, professional development and support, opportunities for professional growth, compensation, and dismissal. And that requires a more sophisticated conversation than our current politics typically support.
To close out this conversation, what is your reaction to the idea that we can’t develop or dismiss our way to effective teaching?
Griffith: I agree with that sentiment wholeheartedly. We can’t anything our way to effective teaching, so I think what we’re really arguing about (if we’re arguing at all) is which levers to pull first. If there were an obvious answer to that question, we wouldn’t be having this discussion. But personally, I think the sooner we actually reform the dismissal process the sooner we can all stop talking about it and the sooner the 95 percent of teachers with nothing to worry about will grasp that fact.
McDougald: I think most people would agree that adequately training, developing, and evaluating teachers (and dismissing the lowest performers) are all interconnected. For example, administrators would probably be more motivated to differentiate between effective and ineffective teachers if the process for dismissing persistently low performers wasn’t so time-consuming and fruitless. But on the other hand, it’s clear that many teachers aren’t getting the kind of feedback and training they deserve, which contributes to the perception that the system is rigged against them. Ultimately, we need to find a way to move beyond these sorts of “either-or” debates, and until we do, researchers and policymakers need to continue tackling this issue from all sides.
Editor’s Note: This post also appeared on the Bellwether Education Partners' blog Ahead of the Heard.