A common theme among teacher evaluations and effectiveness debates is the many unanswered questions that persist. Education leaders and policymakers seem to hold firm on one side of the fence or the other in whether or not they think student achievement data should play a significant role in teacher effectiveness scores and impact a teacher’s job status. The most important question on my mind is that if we still have this much uncertainty about whether we are even evaluating the right things and how closely teacher effectiveness correlates with student achievement data, then how can we, as Joanne Weiss put it, “walk forward” through the fog? Should we walk just for the sake of moving, despite whether those steps end up leading us along the right path, take us backwards or send us off a cliff?

I am concerned with this issue not only because it has such a great impact on where K-12 education is headed, but also because the same debates on evaluation and effectiveness are occurring around programs that prepare teachers for the classroom. For the record, I believe there should be greater accountability for all educator preparation programs in how well they prepare their graduates to teach effectively to all students. In fact, this is a primary goal of edTPA, a new performance assessment of pre-service teachers developed by Stanford University in collaboration with practicing K-12 teachers and teacher educators across the country. It requires teachers to center their practice around student achievement, but through a multiple-measure assessment system that is aligned with state and national standards, including the Common Core State Standards and the Interstate Teacher Assessment and Support Consortium.

Decisions about teacher evaluation systems should be evidence-based, and the current evidence we have is, at best, mixed. In considering the use of controversial value-added measures for teacher evaluations, we risk imposing a unilateral, one-size-fits-all learning calculation back on teachers that could increase the likelihood of unintended negative consequences. The chances of classifying a good teacher as ineffective, or vice versa, are too great. These systems should be based on professional teaching standards and used to establish a data system that allow teachers to reflect on multiple measures of student achievement. We need to ensure that teachers have the skills to analyze achievement data and draw inferences about their own practice. We should encourage recognition of effective teachers so that they become leaders in their schools’ efforts to advance student learning. We also need to ensure that evaluations are a piece of a larger system to improve student learning. This includes accompanying evaluations with detailed feedback, appropriate professional development strategies, and peer-to-peer collaborative work environments. Further, teacher leaders should be engaged in the process of reviewing evaluations. Just as with edTPA, it is essential to bring in the voices of experts who do this work every day and do it well.

Before we use teacher evaluations for high-stakes decisions, we must consider the logistical challenges and other intervening variables that you cannot just manage out of the formula. This will lead us on a path to building evaluation processes that are relevant and meaningful throughout the education system, and thus, truly beneficial to closing our nation’s student achievement gap. Otherwise, we will continue to wander down the teacher evaluation path that leads us to less meaningful, misinterpreted or inaccurate results.

At the Carnegie Foundation seminar on teacher evaluation, Joanne Weiss of the Department of Education commented, “When, in fog, walk forward.” One audience member responded, “What if you’re on a mountain ledge?” I think this is an important point. Yes, teacher evaluation systems must be revamped; right now, they don’t tell teachers, principals or policymakers very much at all. But state and district officials should be judicious in implementation and cautious in making the evaluations high-stakes. There is still much we don’t know about how to adequately observe teachers, provide useful feedback and development, and measure teacher effectiveness, especially in the early grades.

As states and school district officials redesign their evaluations systems, they should consider observation tools that assess teachers’ interactions with their students. Last fall, my colleague Lisa Guernsey, director of the Early Education Initiative at the New America Foundation, co-wrote the paper entitled Watching Teachers Work: Using Observation Tools to Promote Effective Teaching in the Early Years and Early Grades. According to the paper, high-quality observation tools allow for measurements that are far less subjective than many of the checklists and rubrics currently in use. In addition, good observations can capture the kinds of stimulating and content-rich conversations happening between teachers and students, which are so important to children’s cognitive and social-emotional development.

School districts also need to make sure that principals – if they are doing the observing – are trained and have ample time to observe teachers and discuss their strengths and weaknesses. Additionally, districts need to improve the overall quality of professional development, making sure offerings align with the areas in which teachers need additional support.

I think what states and school districts are struggling with most is how to measure student growth in currently untested grades and subjects, in order to potentially encompass such data into teacher evaluation systems. There is no clear path for how to do this. In some places, early grade teachers are being measured by third, fourth or fifth grade students’ scores on state achievement tests. This does not necessarily provide the best picture of how early grade teachers are influencing their students learning. Other districts are creating assessments for the early grades, allowing teachers to create their own assessments or using existing diagnostic assessments as a growth measure. These options are problematic as they raise questions about reliability of the assessments, comparability among teachers and about the appropriate use of assessments with young children.

So while states and school districts should continue to move forward in their work to redesign teacher evaluation systems, they should do so responsibly. New observation measures and assessment measures should be piloted and evaluated, teacher input should be sought, and research should be considered before the ratings from these systems are used in staffing and salary decisions. It does not benefit students or teachers for state and district leaders to hastily implement evaluation systems that don’t provide accurate information about the role teachers play in their students’ learning.

A child’s ability to master each grade level’s material and succeed is largely determined by the quality of his or her teacher at each grade level. That’s why we must work through the tricky and complicated task of devising a model for fairly and effectively evaluating educators.

Colorado has shown that we can arrive at an educator evaluation model that is thoughtful, rigorous and fair. The state developed a student academic longitudinal growth model for its standardized tests that measures how each student is progressing every year. It’s the student’s progress—not the assessment score itself—that comprises half of the educators’ evaluations. The other half of the evaluation will include multiple measures that are being collaboratively developed by state and local leaders, such as principal and administrator observations, parent and student comments, peer input, and school-wide student performance measures like graduation, dropout and truancy rates.

At the federal level, I have co-led a teacher and principal evaluation bill (STELLAR – Securing Teacher Effectiveness, Leadership, Learning and Results) with Congresswoman Susan Davis of California, a former local school board member, which includes fair measures of effective teaching, based on strong standards and assessments. As co-chairs of the New Democrat Education Task Force in the U.S. House of Representatives, Congresswoman Davis and I proposed this model for the task force principles and the New Democrats, composed of 43 House Democrats, approved it unanimously.

By including objective, standards-based student test growth data, educator evaluations are fairer to teachers and less subject to arbitrary reviews by school administrators. No student test is perfect, but that doesn’t mean it’s useless. A lack of transparency and results would be even less perfect. Rather, assessment growth data helps inform all of us about how well students are learning, practices that work, and teacher performance. The policy goal of improving teacher and principal performance through better evaluations is achievable. Student assessment experts are developing appropriate ways to measure student growth in all subject areas and that data can be included in educator evaluations. This is a first step in improving teaching and learning. Our students deserve nothing less.

For too long we have handed teachers the “key to the castle,” closed their classroom door and wished them good luck in their profession. To say we have done a poor job in the past supporting the growth and development of teachers is an understatement. But even today, most teachers in this country do not have access to high quality comprehensive support to improve their practice over the course of their career. When we examine other highly skilled professions, teaching pales in comparison regarding our efforts to systematically and continuously recruit, develop, support and retain talent. The intense focus on teacher evaluation at the federal, state and local levels currently may be our last best chance to right this ship and ensure that teachers are systematically supported in their profession.

Two major shifts in thinking (and then acting) need to occur if we hope to sustain forward momentum in our quest to create systems that comprehensively support teacher growth. First, we need to stop thinking of teacher evaluation by itself and rather think about how evaluating teachers is part of a larger approach to talent management in a state and in a district. Teacher evaluation is only one necessary component of a system to ensure students have access to high quality instruction. Preparation, recruitment and hiring, induction and mentoring, professional growth, compensation and incentives and educator environment are the other essential elements of a comprehensive approach to managing educator talent.

While designing comprehensive systems to educator talent management is critical, it is not enough. We also need to integrate (rather than layer on) initiatives aimed at improving teaching and learning—shift #2. For example, educator effectiveness initiatives should be integrated with college and career ready initiatives. In many cases, states and districts are approaching these areas as separate related reforms rather than planning for their seamless integration. What teachers need to know and be able to do (as articulated in professional teaching standards and actualized in an evaluation system) ought to be integrated with what students should know and be able to do to be college and career ready (as identified in the Common Core State Standards and enacted in changes to curricula and assessment). Integration of education initiatives through a systems approach to change and reform needs to become business as usual rather than innovative thinking.

At AIR, we are working with states (such as Ohio, Massachusetts and Nevada) and districts (such as Hazelwood, MO and Decatur, IL) to develop comprehensive approaches to managing educator talent to improve teaching and learning. In Nevada, for example, AIR staff worked intensely with the Governor appointed Teacher and Leader Council to identify the purpose and goals of educator evaluation and the categories of performance against which teacher and administrators will be evaluated. This process helped to guide and focus design decisions and, when integrated with other state and district initiatives, can provide a greater sense of coherence and credibility among stakeholders. Nevada is using this opportunity to recognize the strong connections within educator performance evaluations by aligning educator preparation and licensure, student standards, curriculum and instruction, and in-service professional learning.

My colleague, Kelly Burling, PhD, oversees our educator effectiveness work. I asked her to weigh in this week:

Most teachers want from their evaluations what any highly skilled, motivated person wants: substantive, objective, actionable feedback; the energizing feeling that comes from contributing and being valued for your contribution; a clear idea of what it means to be excellent and help in reaching that level of performance. As several of the teachers in the Carnegie Foundation for the Advancement of Teaching video stated, they want to be involved in developing the systems by which they’ll be evaluated and take responsibility for their own improvement.

There’s a lot that we don’t know about how to best evaluate teachers. What we do know is that in the past evaluations have been done poorly and didn’t provide much information about how teachers’ practices and their effects on student learning differed. Across the country, many people are involved in developing and implementing new evaluation systems; most include a focus on multiple sources of data, including the impact of educators on student learning, and better observations. At Pearson, we are focusing on creating fair, valid and reliable systems that give us both more and better information; . The more information we have, the better we will be able to understand educator performance, and the better we will be able to identify strengths and areas of improvement that can impact student outcomes.

Earlier this year, we published a guide on developing evaluation systems so that they are as defensible and useful as possible. Undoubtedly, evaluation systems will change and improve over time as more becomes known about their impact and as researchers work to discover what information is best and most important in understanding educator performance. Teaching is too important and the time students have in school too precious to wait until the perfect system is defined and developed. We need to do all we can now to strengthen the profession through better evaluations, useful feedback, and effective professional development and support. At Pearson, we will continue to implement better and more efficient evaluation systems using what is most valid today; while simultaneously focusing our research on questions that will not only strengthen evaluation, but the education profession in the longer term.

Here are some of the other questions we’re asking at Pearson:

· What can empirical research tell us about the rate at which typical teachers deepen their knowledge of content and improve in their ability to teach it?

· What can task analyses tell us about how the job changes as teachers grow beyond novices into experts?

· What are the practices of exceptional teachers and how do they differ from those who are just good?

· What are the methods and technologies that can improve teacher performance at scale in a cost effective way?

We also want to know which measures of educator effectiveness are the most valid and reliable. It is those measures that should be given the greatest weight in evaluations. But all data about teachers’ performance must be investigated by a knowledgeable evaluator who knows them and the context in which they work before making any decisions about what should come next—specific recommendations for how they can improve or, in the worst possible cases, removing them from a classroom.

Of course, teachers must be involved in designing these systems. It is not enough to just give them a voice. They know their students and their practice and they need to take ownership of what it means to be excellent and contribute to defining the best ways to get there.

There’s a lot that we don’t know about educator evaluations. But we do know that stronger, more effective evaluations are one critical element of any well-run organization that involves skilled professionals. Personnel evaluations in many fields, not just in education, are often poorly done and do not yield the improved performance that they should. It’s time for those of us in education to figure out how to capture meaningful data, diagnose and improve performance, and do all we can to change the trajectory of student learning.

Teacher resistance to evaluation is a red herring. The skill of evaluators, not the nature of evaluations, is the real issue.

Who among us hasn’t been evaluated at least once by an evaluator who was not competent do so? We all know what it feels like. Many of us also know how damaging it is to individuals and organizations when this practice persists and its consequences fester.

The evaluation issue in education has been festering far too long because we’ve wasted time on the wrong part of it.

Perhaps we did this because it is easier to believe the worst about people than it is to empathize with them, to realize that they are not so different from ourselves, and that were we in their position we might act in similar ways.

Or maybe we just did things backwards. Had we put capacity before consequence, the situation today might be very different. Good evaluation systems that provided teachers and evaluators with proper training up front could have eased tensions and and avoided declining morale.

There are many reasons why education reform moves at such a tedious pace. The simplest reason is because we spend a lot of time complaining about the wrong things when we could be collaborating on the right things.

To move the teacher evaluation issue forward, we need to do the following:

1. Stop arguing about whether or not teachers want to be evaluated and start thinking about ways to ensure they get thorough, constructive, and actionable feedback from competent evaluators along with the support and resources they need to implement suggested improvements.

2. Operationalize policies that increase the capacity of schools to improve the evaluation instruments they use and the skills of the people who use them.

3. Reframe evaluation as feedback, balance summative consequences with formative support, and prioritize coaching over contempt.

4. Re-evaluate the notion that the path to better academic outcomes is to pressure the people who play a direct role in determining them, recognize that the solution lies in ensuring that teachers do their best work, and acknowledge that few people do their best when they are constantly under pressure.

Reforms or red herrings? Will we talk about things that matter or things that just make us mad? The incredibly divisive dialog on American education that exists today carries with it an opportunity cost we have been paying for far too long.

In matters of great concern, we are all drawn from time to time to the darker side of our nature. I am no less guilty of this than anyone else. Outrage rises up in us all when we allow ourselves to act out of fear, anger, and confusion. But just as we have the power to choose one way of being, we have the power to choose another.

Reforms or red herrings? Innovations or innuendo? Capacity-building or consequences? Let’s move forward guided by the better parts of our nature and hew to the former a little more than the latter.

Since it’s campaign season, I figured it might be fun to respond to this question using an extended metaphor, with teacher evaluation policy playing the role of Gov. Romney’s Irish Setter, Seamus, and policy makers (including Pres. Obama’s EdSec Arne Duncan) playing the role of Gov. Romney.

In reading on, please remember that I’m trapped here in a “swing state,” subjected to a barrage of distorted photos of candidates overlaid with announcers’ voices portending our collective doom should we vote for the other guy. So bear with me for a bit; hopefully this will resonate even with the non-brain-addled in the non-swing states.

The Seamus story is well-known, at least to regular readers of Gail Collins’ column in the New York Times. The Romneys went on a family vacation, which included a 12-hour drive to Canada (Lake Huron). Seamus, the family dog, was put in his crate and strapped to the roof of the station wagon. The trip was carefully planned, down to specified rest stops. But Seamus fouled up the plans a bit when he expressed his displeasure in liquid fecal form, thus soiling himself and his surroundings. So Mitt Romney had to stop and hose down the dog, crate and car. They all then continued on their way. Seamus survived and, according to Gov. Romney, he “loves fresh air” and continued to like car rides, even up there in his crate.

In writing this, I can’t help but note that this all took place in the summer of 1983—the same year as “A Nation at Risk.” Coincidence?? (I’ve really got to get away from these campaign commercials…)

So how is teacher evaluation akin to Seamus? Just as the Romney family and Seamus needed to get to Canada one way or the other, we can all agree that we need good systems of teacher evaluation. The question is how we get there. Our “reformer” friends have come up with an efficient plan: use statistical growth models based on students’ test scores. Let’s strap teacher evaluation to the kids’ tests! What could go wrong?

Plenty, it turns out. This option comes with many serious weaknesses and unintended consequences. The research tells us that “lawmakers should be wary of approaches based in large part on test scores: the error in the measurements is large—which results in many teachers being incorrectly labeled as effective or ineffective; relevant test scores are not available for the students taught by most teachers, given that only certain grade levels and subject areas are tested; and the incentives created by high-stakes use of test scores drive undesirable teaching practices such as curriculum narrowing and teaching to the test.”

But since nobody can come up with an alternative that is as efficient in generating concrete numerical rankings, we stumble (or drive) forward. Even when the brown muck starts to drip down the windows, we merely perform a quick clean-up and continue on our way.

Gov. Romney’s car trip was well-planned and was executed with an unyielding emphasis on efficiency. And at the end of the day, he and his family made their way to Lake Huron. But, notwithstanding Gov. Romney’s protestations to the contrary, it seems unlikely that Seamus or any other dog in that situation would come back wanting more. Yes, the careful planning and efficiency of the trip were remarkable, but there are less stressful and unpleasant ways for a dog to make that 12-hour trip—ways that aren’t as likely to lead to undesirable, unintended consequences.

This, lord help me, is what I’m thinking about when I consider the current push for more effective teacher evaluation systems. My conclusion is we should indeed go on that trip. But let’s invite our teachers and their evaluation systems inside the station wagon, and let’s plan the trip with a complete understanding of how best to get from Point A to Point B.

Last week, the NEPC published a 3-page brief explaining the importance of balanced evaluation approaches that include all stakeholders in decision-making about evaluation systems. Not easy. Maybe not even efficient. But we won’t have to stop mid-way through to get out the hose.

You are absolutely correct. The vast majority of teachers want evaluations. They want to be assessed based on their hard work and their success in the classroom. Yes, they want these measures to be fair, just as we all should desire.

But most teachers, those who work hard and are committed to ensuring every child achieves, are done a disservice by a small, vocal minority who seek to take a political stand against evaluations of any type. Just last week, Karen Lewis, the president of the Chicago Teachers Union, said that “there is no way to measure the effectiveness of an educator.”

Lewis said this, of course, despite knowing that her AFT brothers and sisters in New Haven, Connecticut have adopted a highly regarded teacher evaluation and development system in partnership with the New Haven Public Schools. The New Haven model is now a national goal for those looking for effective teacher evaluation efforts. So despite Ms. Lewis’ political posturing, it seems we can measure the effectiveness of an educator.

That doesn’t mean it is an easy endeavor. Earlier this year, ConnCAN issued a groundbreaking study that offered a detailed look at 10 teacher evaluation models, looking at how they tackled key evaluation components such as student achievement measures, classroom evaluation, and non-academic measures.

The study was titled Measuring Teacher Effectiveness: A Look “Under the Hood” of Teacher Evaluation in 10 Sites. Conducted by Public Impact and released in partnership with 50CAN with the support of the H.A. Vance Foundation, Measuring Teacher Effectiveness examines teacher evaluation efforts in three states (Delaware, Rhode Island, and Tennessee), five large districts (Hillsborough County, FL; Houston, TX; New Haven, CT; Pittsburgh, PA; and Washington, DC), one charter management organization (Achievement First), and the Relay Graduate School of Education.

The report paid particular attention to the design and implementation challenges that many states and districts face in putting an evaluation system into place. These challenges focused on five areas: 1) student achievement measures; 2) classroom observations; 3) other non-academic measures; 4) accuracy, validity, and reliability; and 5) reporting and using evaluation results.

We know there are few factors as important to student success than that of an effective educator. To ensure that every child has that effective educator, we must implement comprehensive evaluation models. Measuring Teacher Effectiveness is an important tool in understanding what teacher evaluation leaders are doing and what components must be factored into a meaningful evaluation model.

Each site we studied is working to continuously improve their evaluation systems with the belief that the challenges they encounter can be overcome. As Measuring Teacher Effectiveness reported, “None of these systems claims to have cracked the code for teacher evaluation. Nonetheless, we consistently heard that the perfect should not be the enemy of the good.”

There is no magic bullet when it comes to effective educator evaluation. But there is also no need to reinvent the wheel. By taking a close look at many of our evaluation trailblazers, we can see the necessary components for evaluation, the challenges our states and districts face in doing it right, and the unanswered questions we must still pursue if we are to provide all students with exemplary teachers.

At the risk of being labeled a blasphemer, I would argue that focusing on teacher effectiveness is not the best use of our efforts.

Don’t get me wrong: I understand that good teachers are really, really important. I know it from reviewing macro data; I know it from micro data (seeing how some really great teachers work with my two sons). But while it’s important, it is not the most important issue in education, and it’s also not the easiest issue to fix. Not by a long shot.

To explain why, let’s use the idea of a soldier as an example. There’s no question that a better soldier – better trained, better physical condition, more experience, etc. – will fare better than one who lacks those features. But now take that great soldier and, instead of an M-16 rifle, give him a feather duster. And for good measure, let’s send him up the wrong hill.

No matter how good that soldier is, what are his chances of success without the right tools and the right mission?

That’s what we’re facing in K-12 education. No matter how good we make teachers, if we don’t give them the right tools, and if we don’t point them at the right outcomes, they will not produce the results we need as a society.

So let’s have that conversation about desired outcomes; right now, the outcomes we’re focused on aren’t relevant to the lives students are going to be living once they leave our doors, and that needs to change. And once we know what we’re trying to accomplish, let’s focus on the tools that can actually get us there.

Until we do those things, let’s not worry as much about maxing out teacher quality – it’s not the most important thing.