Addressing the teacher/student data firewall is fair and a needed change to help support broader education reforms. The intent is to use this data to drive conversations and strategies around the best ways of preparing, recruiting, rewarding, and retaining excellent teachers. Without this data, it is difficult, if not impossible, to understand what is working and what is not working within an education system.

What isn’t entirely clear is how all the criteria in the Race to the Top applications will be evaluated relative to one another. The Notice of Proposed Priorities listed more than 19 criteria and requirements, but didn’t assign any weights suggesting which criteria may be more important than others. I’ve also heard some concern about the proposed use of NAEP as a selection criteria and whether that conflicts with current law that bars any “rewards or sanctions” to states or schools based on NAEP results.

There is one omission that surprises me. There is no disputing that the Race to the Top will significantly advance and accelerate education reform, particularly with creating more favorable state policies for charter schools and teacher effectiveness. But this program wasn’t authorized under a typical education or appropriations bill. It was created as part of the American Recovery and Reinvestment Act of 2009 in response to an unprecedented economic crisis. Supporters of the Recovery Act argued that it was needed to preserve and create jobs, promote economic recovery, stabilize state and local budgets and assist those most impacted by the recession. That’s part of the reason why the legislation requires projects funded with Recovery Act dollars to provide an “estimate of the number of jobs creates and retained” among other reporting elements. Yet the draft notice doesn’t include a single criteria or question related to job creation or any other economic measure.

Implementing the Recovery Act was always going to be a challenge because it meant carefully balancing two competing (sometimes opposing) goals of advancing specific reforms (such as Health IT adoption, smart grid research, education reform) while also getting the funds out quickly to "shovel ready" projects in order to stimulate the economy and accelerate the recovery. It’s surprising the Administration wouldn't include at least a few questions related to the broader economic goals of the Act. Not doing so just bolsters the critics who argued that the Recovery Act wouldn't have a stimulative effect.

It’s clear, as some have said, that all this federal lawmaking is, without more specific requirements, likely to yield little fruit in the end. Regardless of who is president and which party runs Congress, it is like making sausage. No matter what the ingredients are or the flavor you are looking to create, in the end, it gets jumbled up and tastes pretty much the same (along with some pretty pernicious things thrown in along the way).

Teacher evaluations, however, are not ever limited to a one moment-in-time test score, as Monte Neil opines. We all know it is much more complicated than that. We also all know that a teacher in the 4th grade who inherits students whose third grade teachers failed to teach them, say, grade level math, should not be held “liable” for another’s mistakes. She can, however, demonstrate that she added value to those students. If over time, each of her 4th grade math students shows success compared to the previous teacher, clearly something is working with one teacher and not with the other. We don’t need an empirical study to make the link for us. It happens daily in effective schools. Schools that continually succeed have teachers that are successfully imparting their lessons, lessons that are clearly well organized and based on good content. Schools that continually fail to grow student achievement have deficiencies in teaching which result in deficiencies in learning.

The answer to more great schools is only partially addressed by removing state barriers to student-teacher evaluation. There is a more serious problem that the federal guidance does not address at all: when a teachers union contract forbids a school to use student achievement in evaluating teachers then it doesn’t matter what the state or federal government does or says. And yet, such a clause exists in most big urban district contracts. There is also the so-called anti-NCLB clause, which in the Little Rock, Arkansas teachers contract, requires any displaced teacher “be given the opportunity to choose from the positions vacant and the same grade level.” So when a state does allow teachers to be fired for whatever reason, they are still guaranteed a job in that district.

As UNCF’s Dr. Lomax puts it, “When students take tests to determine what they learned, which teachers’ students are passing with flying colors, year after year? And by the same token, which teachers’ classes aren’t learning the math and reading and science they will need to succeed in college? Which teachers’ classes show year-to-year improvement, and which are showing little or no progress, or even regressing? “

We know the answers. We need to address it head on, and no federal funds should touch a district that allows this kind of adult-protection program to continue when kids lives are at stake. That is only fair.

"Sunshine is the best disinfectant,” Supreme Court Justice Louis D. Brandeis so famously said.

Successful organizations, be they public or private, all need a good dose of sunshine. How many pleas do we see for better data about prison conditions, executive compensation, and physician error rates? Is education any different? Should we be allowed to hide in the shadows? No institution has a more important charge than the education of our children. We urgently need the ability to see when schools are succeeding and when they’re failing.

The most important part of any school, and the most important factor in student learning and achievement, is the teachers. We should be doing everything in our power to identify our star teachers and replicate their approaches and practices. Not only does sunshine expose the best and worst of an entity, person, or practice, but it lets us use this information for overall improvement. And teachers themselves should welcome this. How often have we heard --rightly--that good teachers get neither the recognition nor the pay they deserve? It can only help, to be able to recognize their excellence. But when teachers are not succeeding, the public also has a right to know that.

It should be self-evident that we cannot evaluate a teacher’s efficacy without a sense of how his or her students are actually doing. Here, let me be quick to say that student achievement data should serve as only one – not the sole – piece of teacher or principal evaluation. Any real measure of teacher or principal performance should include a comprehensive set of quantitative and qualitative factors. But student learning and achievement is the very purpose of teaching, and to a great extent student learning can be measured. It simply doesn’t make sense to evaluate teachers without looking at the progress of their students. Today, however, nearly every district has regulations in place prohibiting any discussion of student achievement data as part of an evaluation. (See Lisa Graham-Keegan’s post for further detail.) How can that practice be doing anyone – whether student, parent, teacher, principal, or concerned citizen – any good?

Every student, regardless of race, class, or geography, deserves to be in a classroom with an effective teacher. But we can’t ensure that this happens unless we can evaluate teachers and their students concurrently. The Race to the Top guidelines bring us one step closer to ensuring all students have access to effective teachers.

Ted Hershberg, Professor of Public Policy and History, University of Pennsylvania, submitted the following:

The re­quirements are fair, rational and long overdue. The long-standing legitimate argument against using student achieve­ment data has been successfully addressed by using student growth data. Achieve­ment (a point on a vertical scale at a single moment time) is best predicted by family income. But this bias is eliminated by the use of growth data (the annual progress of in­di­vi­dual students or schools), which is best predicted by the quality of instruction.

More rigorous stand­ards and better assessments are both necessary and welcome, but recent research demonstrates the accuracy of sophisticated value-added models in accurately identifying the most and least effective performers. Results from tests currently in use can be relied upon to identify these “tails” of the teach­er distribution as long as they meet the re­quirements of statistical accuracy, are strongly correlated with the stand­ards taught in the school, and have sufficient “stretch” at each end of the scale to permit differentiation among the highest- and lowest performing students.

Teach­ers shouldn’t be judged by a test scores alone, but there is no sound reason why test scores should not be a component of an eval­u­a­tion system consisting of multiple measures, sophisticated observation protocols, and peer review.

Moreover, inclusion of an empirical component in evaluation makes it possible to accurately and fairly differentiate per­form­ance, which is the basis for differentiating compensation, creating workable career ladders, and attracting and retaining the best and brightest as class­room teach­ers. All these issues are addressed in Theodore Hershberg and Claire Robertson-Kraft, eds., A Grand Bargain for Education Reform: New Rewards and Supports for New Accountability (Harvard Education Press: August, 2009).

Eliza's question deserves responses from the people who actually run some of the most successful schools in the country so they can explain if and how they use student achievement data to evaulate their teachers. I know of not a single school that does not do that, and if one exists I would like to learn about it...then ask them how they know they are successful.

The point here is not to create some state or district sponsored, one size fits all teacher review process. The point is to lift the prohibitions that currently exist. The point is to allow school leaders to discuss and evaluate teacher effectiveness as it relates to student achievement. Not to base an evaluation solely on a single test outcome, but to consider student progress as one indicator of teacher effectiveness. It is a pretty basic request.

And that fact that we are not allowing that discussion to take place is part of the "conspiracy of silence" that Joe Williams rightfully alludes to. Setting up various straw men about whether or not a particular formula for linking student test scores to teacher effectiveness has merit ignores a more basic problem.

All of our data says the teacher affects student achievement. I simply don't understand how you look at everything we know, and come away saying teacher effectiveness is not an influence. If a teacher cannot be held responsible for a lack of achievement, then she cannot be credited either.

It is not an easy equation, however, and I personally do not believe in the strength of one magic formula that must be applied to all teachers. The fact that we have those is a testament to the influence of collective employment agreements more than anything else.

I believe great educators know a great deal about the achievement of students and how to relate teacher effects to the progress of students on a case by case basis. They don't guess...they measure. In all sorts of ways. I would love to believe that one day we will let them do that. Our children will profit greatly from it.

In our discussion of Race to the Top eligibility requirements, one area we have only touched upon lightly is evidence for or against the effectiveness of linking student performance data to teacher and principal evaluation. In an article for the Teacher’s College Record, education professor Kevin Welner states, “little or no research actually supports policies linking teacher compensation to student test scores.”

Is this true?

As I understand, the rules for Race to the Top emphasize charter schools and collecting data to implement pay for performance plans for teachers. I have no problem with the federal government establishing rules for handing out money. I wish they gave out more than the 6 to 8% of local budgets given the scope and detail of rules they seem to think are necessary. More gold to go with the ruling would be good. But 4 to 5 billion dollars is a hefty chunk. I have no problem with charters although think the data suggests they have mixed results. And more information (not just data) is always helpful in decision making.
Neither of these strategies will be very useful in low income rural districts where I have spent most of my time in the past few years. Even worse, most of them won’t be successful applicants in this competition for the usual reasons. Emphasis on scale will mean small places will be dismissed. To get together with a cluster of districts takes a lot of time and many districts lurch from crisis to crisis without time to plan and collaborate. And charters and pay for performance just are not relevant in their locations. So many rural districts won’t even try.
As part of a White House Rural Tour, Secretary Duncan will join Secretary of Agriculture Tom Vilsak to visit Hamlet, NC soon. That’s the Richmond County School District—8000 students, 40% African American, 2/3rds on free and reduced price lunch, in the final year of sanctions under NCLB. Yet this district with a higher concentration of poverty than Chicago gets $1000 less per pupil in Title I funds than Chicago does. That’s the screwy Title I formula that penalizes states with lower school spending (almost all of which can be explained by lower tax bases) and penalizes places for being poor and small. If the Secretary wants to do something about failing schools in these rural places he needs some better arrows in his quiver than charter schools and pay for performance. See the July Rural Policy Matters at www.ruraledu.org for a more detailed analysis of Richmond County.
The Secretary needs to support efforts to recruit and retain teachers like the “grow your own” program in Bertie County, NC or the statewide Grow Your Own Illinois program operating in both urban and rural places. Bertie County Schools recruit up to 20 of their best high school graduates into real jobs in the district for all the years they are in good standing working toward a teaching degree. The jobs allow them to be mentored by the best teachers in the district, take courses offered on line or at nearby Shaw University, have a salary and health insurance. The local school board has committed its own resources to support this effort. Let’s take it to scale.
The Illinois program partners community organizations with local university teacher training programs. Parents and other community activists are recruited into the program and supported both by the organizations and the university. These are people who have made a commitment to live in a place and now need the skills to be teachers; they are likely to stay. Logan Square neighborhood in Chicago has graduated 7 or 8 new bilingual teachers from this program. This one is paid for by the state.
If we think pay scales are the problem let’s do something about the enormous gap in pay between rural teachers and urban and suburban teachers. Rural districts are tired of being a farm system for higher paying districts. Use some of this magic data and create an index to identify hard to staff schools and shovel some money that way.

Just to perhaps get some discussion going among colleagues on this list:

Lisa Graham Keegan is correct that evaluation of teachers and principals is generally quite poor, a point on which a wide variety of analysts and organizations agree. Union contracts may be an obstacle in some places, but that hardly explains the problem in states where there are essentially no union contracts. The reasons are no doubt multiple, from lack of willingness to spend on good evaluation to lack of structures in which to carry them out to lack of training for evaluators, etc. Causes will have to be considered in planning solutions, but it would be best to start from strong conceptions of good evaluation.

Good principals don't just pull out test scores. They know, as do good teachers, how little information is actually conveyed by those scores. They may look at them for the modest information they can convey, but what is far more important is the ongoing evidence of learning from the classroom. And good teachers and principals know that schools are places children live in many hours most days for most of the year - and the quality of those days matters greatly to parents, the children and the educators. The range of important areas should be included in evaluations.

There are strong approaches to comprehensive, useful evaluations of teachers. ARRA is too likely to perpetuate the ways NCLB and state accountability systems have ignored vast areas of learning and children's well being by focusing on test scores. That need not be the case. Rather than emphasizing the elimination of barriers to linking student scores to specific teachers and principals, or emphasizing using scores for salary, tenure and firing decisions, "Race to the Top" should focus on improving teaching. That possibility is there, but it receives too little attention relative to test-based data. Improvement includes professional learning opportunities and the evaluation of educators. Success with students should be part of evaluations, but test scores should only be a small part of the definition of success, and not given enough weight to add more pressure to skew the curriculum and instruction toward narrow test prep.

I love the suggestion that providing evidence that the young children in American classrooms can read is a statist restriction on their creativity. Let’s grow up.

Today’s reality: when a school principal sits down for a one on one evaluation with a teacher, she is allowed to review precisely NONE of the academic progress information for that teacher’s students as part of the formal evaluation. In some states, special interests have been able to insert this into state law, in most others; it sits cozily embedded into school district policy. Bottom line, ANY inclusion of student achievement as a factor in evaluating teacher performance is forbidden.

I challenge all of us who have never asked a principal point blank about this, to do so. Most of them will tell you that they complete their “evaluation” by discussing the legally acceptable issues…attendance, cooperation, dress code, professional development. Perhaps it seems too obvious to point out that schools don’t exist so that teachers can dress well or show up on time – they exist so that students can learn. But nothing that relates to that core mission may be used to assess those most responsible for its accomplishment. So principals get their marginally relevant evaluation completed and signed off on, they put it away – and the great principals then say, “Alright, this evaluation is over, and now I would like to have an informal conversation.” And they pull out the test scores.

Secretary Duncan is right, we should be grateful for his willingness to act on this, and those of us who can should stand firmly behind the effort. Because from what I am reading here, it is clear that his attempt to shine light where it needs shining has only encouraged some to go running in search of a lampshade.

Is it fair to children to further narrow their education and encourage even more teaching to predominantly multiple-choice tests, as is now happening under "No Child Left Behind"? No – but that will be the primary consequence if federal "Race to the Top" guidelines link test scores to teacher and principal evaluations and create a new cycle of standardized tests.

The guidelines say that tests should be a "significant factor" in teacher evaluations. "Significant" is not defined, but it is safe to assume it means weighty enough to affect educator behaviors, and hence intensify teaching to the test.

Secretary Duncan says that "achievement" should be factored into evaluations, then defines student achievement as test scores. That is an impoverished definition because the tests measure far too little of any reasonable curriculum. The Secretary then calls for "going forward" to include other measures such as classroom observations. That could be positive. "Race" money should be used to overhaul state assessment programs to include performance assessments and classroom-based evidence, but it is not clear it will be. So we are far more likely to get increased pressure to turn schools, particularly those serving low-income children, into test-preparation centers.

Supporters of the “Race to the Top” guidelines make many leaps without evidence. Many assume that "payment for results" will produce improved outcomes. The historical evidence, as Madaus, Russell and Higgins sum up in their new book "The Paradoxes of High Stakes Testing," is that the corruption of teaching, learning and the indicators overwhelms any positive results. Eli Broad has maintained pay for performance is common, but Adams, Heywood and Rothstein, in "Teachers, Performance Pay, and Accountability," show it is not common, especially for professionals, and reinforce Madaus, et al.'s conclusions on the harmful consequences. If government proposes to support data-based decision-making, then surely we should see real evidence justifying this proposal.

The push for national standards and new tests also stems from great leaps of logic. The assumption is that tougher tests produce better results. South Carolina's tests are as tough as Massachusetts', yet South Carolina has not shown much progress on the National Assessment of Educational Progress (NAEP) while Massachusetts has. Could it be that the greatly increased funding to low-income districts in Massachusetts is what really made the difference? In a similar vein, why are Montana, Idaho and Wyoming so similar in NAEP scores and gains, while the level of difficulty of their tests is so different?

Proponents also point to academic performance in other nations to justify an increased emphasis on test results (hence the call for "international benchmarking"). However, in elementary schools, high-scoring Finland uses only a few sampling tests; Singapore relies on classroom-based evidence other than its national tests in grade six; and Hong Kong uses matrix sampling. None of the higher-scoring nations mandate standardized testing across six consecutive grades as NCLB does.

Taken as a whole, the "Race" is far too test-centric, in its push for tougher standardized exams, its insistence on linking student test score changes to teachers and principals (and from that to making high-stakes decisions), and the primacy it gives to test scores in the conceptualization of "data." The results of such a focus will not be fair or healthy for children's learning or well being.

The Secretary of Education is both right and wrong to insist on breaking down any alleged firewalls anywhere in state data systems. He writes, "Test scores alone should never drive evaluation, compensation or tenure decisions. But to remove student achievement entirely from evaluation is illogical and indefensible." Few would argue with that as a principle, and he is right on the principle. Beyond the starting-gate requirements of "Race to the Top," the details matter, and he is wrong when he glosses over those details. When Secretary Duncan writes that "analyzing students’ test scores is one of the best ways to measure the quality of teachers," I can't help but think that's damning with faint praise.

Many states such as Tennessee or Florida link a single teacher to a student's test scores in each subject, as if children compartmentalize their lives and their learning. If your chemistry or physics teacher never helped you understand how to solve equations, then you should be proud of systems that link one teacher and only one teacher to a student's performance in math. If you and your friends never had a friend, a parent, or a tutor help you with your work, then you'll feel comfortable assigning all credit or blame to teachers. In my best professional judgment, the research on using test scores for evaluation is mixed, and the use of student outcome data spans a broad range from Long Beach's collaborative efforts to several crashed-and-burned attempts at imposing performance-based pay in Florida. It would be much smarter to acknowledge that we need a system that encourages the right decisions even (and maybe even especially) when the data are messy.

Understanding and working with the limits of messy data require experimentation and much better evaluative research than we have now. Legislators and superintendents often would like to impose a specific mechanism for rewarding or punishing teachers. Call it the "pet idea" obsession, along with pet ideas for a variety of untested education nostrums. The problem is that if the federal government encourages experimentation in using test scores without also pushing for transparency and research, we will end up with dozens of systems based almost entirely on single test scores (despite the protestations of many of my fellow commentators that of course they don't want teachers evaluated based on a narrow set of test scores), with no guarantee that the systems improve schools. I could propose a number of specific ways to address the limits of test scores, but the goal should not be to impose Tom Vander Ark's, Ariela Rozman's, or Sherman Dorn's pet idea. The goal of this Race to the Top requirement should be the development and rigorous evaluation of new teacher evaluation systems. As states build the data systems that Secretary Duncan and many others (including me) want them to develop, they should also allow researchers entry into those systems to evaluate their use, the consequences of technical choices, and options for improvement. There is no such requirement for transparency and research collaboration in the Race to the Top, and that is an unfortunate omission.

In releasing the draft guidelines for awarding Race to the Top funds to states, the Obama Administration acknowledged a crucial truth: the current system is failing many students. We know it’s necessary for teachers to have timely, accurate data about students to help guide their practice in the classroom. But the proposal to link student achievement data to teacher and principal evaluation misses the mark.

A "Race to the Top" can quickly turn into a "Race to Judgment." I’ve explained to Secretary Duncan that educators have been burned by NCLB – where the results of one high-stakes test were used in a punitive manner. We’re concerned about the effectiveness and reliability of requiring states to link data on student achievement to individual teachers for the purpose of teacher and principal evaluation. Teachers who work with disadvantaged students shouldn’t be “evaluated” based on whether their students hit a particular test target on a particular timeline. And we certainly shouldn’t base additional compensation on whether students meet particular testing targets on a particular day. We need to offer incentives so that our best teachers teach the students most in need of assistance, not incentives to teach students most likely to score highest on a standardized test. As with NCLB, good intentions can lead to unintended—and unacceptable—consequences.

NEA and this Administration share the same goal: to dramatically transform public education so that every public school is a center of excellence. And we share the same urgency. If nothing changes, up to half of America’s school children who are poor and minority will not graduate from high school—a situation that is not only deplorable, but criminal as well. We’re going to use the 30-day comment period to work with the Obama Administration to find common ground because innovation, and winning this race, can’t happen without collaboration.

The President and the Secretary are right to insist that there be teacher-student links for states to receive rttt funds. They're not saying teachers should be fired solely on the basis of student scores. They're simply saying that until we understand the patterns in teaching and learning and do something on the basis of those patterns to improve learning there is no such thing as real reform. It's that simple. And they're right on this point. i hope they really follow through on it. We'll see.

What worries me, though, about fairness and rttt is how all the criteria will sort out together in the end. I have a list that I look at frequently of 20 or so states that consistently do very poorly on the NAEP and that have bad records in terms of performance by poor kids and kids of color on the NAEP. Most of these states have done little or nothing even as of this very year to change what they do in any fundamental way. I'm sure several of them will write very pretty applications, promising to be "good boys and girls" in the future. Some of these states are very significant politically as well. And virtually all of them have signed the nga/ccsso letter, which commits them to nothing.

I hope very much to see few, if any, of these states on the final list of states that get funding. I suspect that higher achieving states that have taken, or are taking, the painful steps of real reform, as well as the general citizenry, will cause a firestorm of protest, if they do. This is the real fairness issue. Is real reform going to be recognized and rewarded, or not. If not, this promising strategy will die in the crib.

Updated at 11:16 a.m. on Aug. 6.

Are the "race to the top" requirements fair to whom? The spotlight shouldn't be the teachers and principals, but on the kids.

The "race to the top" dollars--for that matter, every penny that goes to educating kids--ought to be spent in a way that's fairest to them. That means doing whatever it takes to improve America's dismal track record in student achievement (have a look at the NAEP scores), while closing the achievement gap.

To be sure it's hard to design an accountability scheme that really works--that gets the attention of those who teach in and run the schools; that doesn't create perverse incentives; that shows results. Nor is accountability the only reform a school district needs to adopt. From Raleigh NC to Long Beach CA, Montgomery County MD to Union City NJ, school disticts have demonstrated that a "continuous improvement" strategy--what successful businesses have been using for years-- can serve the needs of children. A smart accountability system has to be part of that strategy. If the race for the top" funds move school districts in that direction, that's a smart investment.

I think Secretary of Education Arne Duncan makes a lot of sense in proposing to include student achievement data as one of the factors that can be used to evaluate teachers.

At UNCF—the United Negro College Fund—college is our middle name. But students who have been given an inadequate education before college face long odds of succeeding in—or even being admitted to—college. That’s what makes this issue our concern, and the concern of our students and of the 39 historically black colleges and universities that are members of UNCF.

As I noted in my last post to this blog, just under one in every three American college students have to take at least one remedial course; the number is significantly higher among black and Latino students. In other words, they have to spend scarce family and scholarship and student-loan dollars on learning what they should have been taught in high school. That makes no sense and it’s not fair.

It would be easy to say that our schools and our teachers are failing our students. But that wouldn’t be fair either. Many of our students arrive on campus ready to learn. Clearly, the education provided by many schools and many teachers is working, while the approach taken by others isn’t. We need to identify those that are getting the job done, so we can emulate and reward them. We also need to identify those that are falling short, so we can help them do what their students need them to do or, failing that, entrust those students’ education to schools and teachers who will give them the education they need—and that the nation desperately needs them to have.

How do we determine who’s getting the job done and who isn’t? At least part of the answer needs to be results. When students take tests to determine what they learned, which teachers’ students are passing with flying colors, year after year? And by the same token, which teachers’ classes aren’t learning the math and reading and science they will need to succeed in college? Which teachers’ classes show year-to-year improvement, and which are showing little or no progress, or even regressing?

This is an essential part of rule-of-thumb accountability—or it ought to be. It’s why companies keep sales figures, to keep track of which salespeople consistently sell enough widgets as the company needs in order to show a profit, and which don’t. It’s why baseball teams calculate batting averages.

Because results matter. And if results matter on the sales floor or the baseball diamond, shouldn’t they matter even more in the classroom, where students’ futures are at stake?

The question is not fairness (it is of course fair to require something of states for signficant resources), the question is whether states will have the political will to aggresively implement bold reforms which offend some keepers of the status quo. Today, there is lots of rhetoric around the primacy of the interests of kids but still too little action in that regard. Today, there are no impediments for states to adopt rigorous curriculum standards; sadly, few have and few have linked teacher pay to student performance, aggressively moved to close persistently failing schools, or remove poor teachers who aren't effective with kids. Let's hope this race to the top removes the age old excuse of inadequate resources and makes "kids first" rhetoric a reality at last.

Over and over, education policy has been decided not on the basis of what’s best for kids, but what’s “fair” for adults. It’s the chief symptom of a system that always seems to have its priorities backwards. The Race to the Top requirements, on the other hand, mark an ambitious and frankly courageous effort to knock some sense back into the education system. It’s clear that President Obama and Secretary Duncan recognize that our schools are only as good as their teachers, and that any meaningful attempt to improve the quality of education our children receive must center on the goal of a great teacher in every classroom.

The fact is that if we want to improve teacher effectiveness, we need ways to measure it. That demands honest evaluation systems that focus on a teacher’s impact on student learning. That’s not to say that the measure of a teacher is a single set of test scores – it isn’t. Our nation’s best teachers are extraordinary individuals whose true talents and impact can be assessed only with multiple inputs. But at the end of the day, no teacher can be deemed effective if his or her students show no measurable evidence of strong academic growth.

Connecting student achievement outcomes to individual teachers hardly seems revolutionary, but it’s here that Secretary Duncan’s deep understanding of the way things work in actual school districts shines through. As Joe Williams notes in his post, good ideas all too often die in political backrooms and committees. By making this connection between student and teacher data a prerequisite for Race to the Top funding, the administration has the chance to circumvent that pattern while ending policies that serve no purpose but to put the interests of adults before the interests of kids—because that’s not fair.

“He who has the gold, makes the rules.” Most school districts in California are using assessment data for information to improve instruction in the classrooms. To be honest, it’s not a data point for the evaluation of its teachers or principals. Yet many would agree that good and great teachers/principals achieve results far beyond a single test measure. The care and nurturing, the emotional support, the development of the whole child cannot be measured with assessment data. So is it fair?

Fair is where one goes to see prize winning livestock and enjoy a day of rides and cotton candy. The high stakes of testing should not determine whether or not schools or school districts have access to “incentive” funding. If this narrow definition on the use of assessment tied to teacher/principal performance is upheld, then all of California will be ineligible for “Race To The Top” funding. Sure, in addition to non-access to these funds, we in California still have to contend with our well-known budget crisis. Yet when school bells ring in the fall, our doors will open to the masses and teaching will begin again in classrooms with the renewed enthusiasm coupled with mild anxiety that greets every child.

My hope is that the Obama administration will not punish California’s children because of laws and practices established by unions and legislators (the adults). All children, and ultimately schools and school systems, should have access to these funds. California is not an island.

President Obama and Secretary Duncan are serious about improving educational outcomes for the nation’s school children. Through Race to the Top (RTT), the federal government will invest $4.3 billion to drive reform of the nation’s schools. An additional $5.6 billion will be invested through other federal programs. Recognizing that effective teachers and principals are key determinants of student success, the RTT initiative stipulates that recruiting, developing, rewarding, and retaining effective teachers and principals are key components of the plan. Further, the initiative calls for building data systems that measure student success and inform teachers and principals how they can improve practices.

Obviously, any prohibition on linking student achievement data to teacher and principal evaluation would also prohibit important aspects of the initiative, in particular developing, rewarding, and retaining effective teachers and principals. How can teachers and principals be informed about how they can improve practices that increase student achievement if the evaluations are prohibited? So, the real question is whether RTT is fair. Or maybe a better question to ask is whether our current practices are fair.

In my view, we do not have a fair system when students in low performing schools do not have an equal chance of success as others in high performing schools. If RTT is successful in developing a culture of evidence and improvement in all schools, we will have the possibility for a more equitable and more effective education for all children.

So, is it fair to teachers and principals to have the achievement of their students used to evaluate them? The only way to argue that it is not is if one assumes that teachers and principals do not have an influence on the achievement of their students and that they cannot benefit from feedback and development—or if one believes that the highest performing teachers and principals do not deserve recognition for the good work they do. I would argue, on the contrary, that educators make all the difference, that everyone can learn and improve, and that society does not recognize and value effective educators enough.

Race for the Top is all about fairness.

Are the requirements fair? Sure they are. The federal government can put just about any requirements it wants to in a competitive grant. The bigger question, it seems to me, is one of possibility: Can eligible states actually come up with credible applications that meet the grant’s many criteria? Or, to put it another way, what if someone threw a multi-billion dollar party and nobody came?

Currently, there are a total of 19 selection criteria, any one of which could require major changes in a state. Take, for example, the following: Fully implementing a statewide longitudinal data system; Ensuring equitable distribution of effective teachers and principals; Turning around struggling schools; Making education funding a priority; Raising achievement and closing gaps. Is any state currently even close to achieving any of these goals? And, more to the point, would anyone evaluating a Race to the Top application believe that any state could reach these goals over the life of the grant?

What troubles me here is something I’ll call the “Reading First Phenomenon.” It goes like this: The government sets out some very logical standards; programs are created to meet those standards; money is doled out; programs end up being ineffective – probably because of poor implementation on the ground.

When I look at the Race to the Top from a Washington D.C. perspective, it looks bright and shiny and new – just what we need to kick recalcitrant states into high gear on ed reform. But when I look at it from a classroom or single school perspective, I ask, “How’s all this new stuff gonna get done?”

Take, for example, the criteria from the Great Teachers and Leaders section:

1. PROVIDING ALTERNATIVE PATHWAYS FOR ASPIRING TEACHERS AND PRINCIPALS. A fine idea in theory but it hasn’t worked out so well in practice. I can see some pretty poor programs being funded here.

2. DIFFERENTIATING TEACHER AND PRINCIPAL EFFECTIVENESS BASED ON PERFORMANCE. This’ll take some union battling to be sure. And that takes time. How’s a state gonna work this out in time to meet grant application deadlines?

3. ENSURING EQUITABLE DISTRIBUTION OF EFFECTIVE TEACHERS AND PRINCIPALS. This one requires that #2 get done first, then comes the hard part: involuntary transfers of effective teachers and principals to ineffective schools. That’s not exactly the reward most effective educators are looking for. And then there’s the problem of sending less effective educators involuntarily to more effective schools. I don’t think that’s gonna go over very well either.

4. REPORTING THE EFFECTIVENESS OF TEACHER AND PRINCIPAL PREPARATION PROGRAMS. This sounds smart. But the devil is in the details. How are we going to rate these programs in a way that would be useful to both the programs and their participants?

5. PROVIDING EFFECTIVE SUPPORT TO TEACHERS AND PRINCIPALS. This would be wonderful. But does anyone realize how much support “effective” support is? Forget about the money. Most states don’t have the manpower to provide support that is truly effective.

The more I study the Race to the Top, and really read the fine print, I’m starting to realize how ambitious it is. Ambition is good. But too much ambition causes people to do strange things – like take money for programs they’re not sure can actually be implemented.

I believe the Race to the Top would be more successful if it concentrated on fewer goals that were more likely to be achieved over the life of the grant. Focusing on teacher and principal quality, for example, is likely to yield the best bang for the buck. And the idea of "effective support to teachers and principals" is the best language in the entire grant. Why not start there? States that learn to do a few things well will be closer to reaching the top than states that do many things poorly.

What's fairer than judging performance on the basis of student outcomes, and what could be less fair than prohibiting it?

The typical argument against using student achievement to determine teacher effectiveness is that it depends on the luck of the draw -- the teachers who get the smart kids look better on paper. But if states do create longitudinal data systems that permit a real look at classroom-level progress (not just status), then that argument goes away.

Of course, there are plenty of teachers whose effectiveness will have to be measured some other way, since they teach subjects where no "longitudinal" measure is available. But those judgments also ought to be grounded in data and not popularity or tenure.

Arne's said repeatedly that he's not talking about evaluating teachers on the basis of a single test, leaving plenty of room for developing additional strategies. So it's dismaying that the anti-accountability crowd immediately posits the worst-case scenario about achievement-based evaluation (evil principals gaming the system to play favorites, and so on). Let's instead begin from the right premise -- that student achievement should be the real test of effectiveness for teachers, principals, and schools -- and add whatever safeguards and protections and technologies are needed to make it work.

That's the approach Race to the Top takes, and it makes good sense.

Fairness has nothing to do with it. He who pays the piper calls the tune. States are competing for a finite amount of federal funds disbursed for specific purposes. If they want the money, they should meet the requirements.

It bears repeating that the guidelines do not require states to link student data to teachers and principals, only that they do not prohibit the practice.

Fair to whom? The question reflects how so many K-12 education policies and practices are focused on what’s best and what’s fair for adults, rather than students. Is it in the best interest of students, to evaluate educators based on student performance data? Of course. The best schools across the country, particularly those that improve outcomes for at-risk children, focus relentlessly on student data. The educators in those schools are brutally honest about performance – both their students’ performance and their own. Federal funds should support nothing less. That’s what’s fair for students.

It is fair for The Race to the Top eligibility criteria to require that states remove prohibitions on linking student achievement data to individual teachers and principals for the purpose of evaluating their performance. But there also needs to be fairness in the evaluation process that states and districts establish. Hopefully, the Department will consider the quality of evaluation processes as they review state applications for funding.

The Race to the Top program is a competitive grant program—the largest ever for pre-K-12 education, so it's appropriate to have specific criteria that states must meet in order to win the competition. Its mission is to provide an incentive and support for transformative change in public education systems. Its intent is to push along states are that leading the way on specific reforms so they can serve as models for other states.

Teachers and principals are responsible for engaging students in learning and should be evaluated on their success in doing this. In what other field are you not even able to consider an outcome that is of primary importance in judging employees’ accomplishments? Student achievement data should not be the sole criterion for evaluating teachers. They obviously can't be for the many teachers teaching in untested subjects, and fairness dictates that it is necessary to consider a variety of measures of success. But where possible, student achievement must be among them.

The guidance issued by the Department of Education on the Race to the Top Fund is remarkable for its whole-hearted support of reform principles. Arne Duncan is to be commended for standing firm in endorsing principles of linking teacher and principal compensation to student achievement, support of charter schools, and requiring that states adhere to the reform standards enunciated in ARRA. If states have laws that prohibit taking student achievement into account in setting teacher and principal compensation, they should not be eligible for RTTT money. The same should apply to states that limit charter school growth through caps or other means.

The right principles are set forth in the guidance. The challenge now is for the Department to apply these principles in the face of likely political and interest group pressure. The Secretary should know that the business community and other reform-minded groups are backing him to the hilt in this effort.

Thank goodness that Jon Schnur had the foresight to sneak RTT into ARRA. Otherwise, we’d be in an ugly and unproductive reauthorization battle that would make health care look tame.

RTT gives Duncan a preauthorization blast of capital aligned with an aggressive reform agenda. As I noted last week on HuffPost, RTT will feed the rabbits (the states ready to move) and won’t be held back by the rebel, laggard, and the complacent states.

Two of the difficult 19 grant criteria include fully implementing the Data Quality Campaign elements and using “differentiating teacher and principal effectiveness based on performance.” It’s a game changing and long overdue requirement.

After 15 years in business, I was shocked in my first year as superintendent in Washington State to find that we were flying blind—almost no performance data and none linking individual performance to outcomes. In the retail business I came from we knew sales by store by item the next day.

This grant criteria makes good sense. It’s essential to create quality at scale. It’s not a difficult technical requirement, it’s just something the unions have successfully avoided for several decades. RTT will allow a handful of states to show the way by linking data to placement, evaluation, professional development, compensation. We’ll be able to improve schools and the equitable distribution of teachers. Good teachers will earn more money. And, perhaps most importantly, we’ll learn a great deal about equity and excellence.

Research has found that great teachers are a critical factor in improving student achievement. Great teachers help students who are behind catch up with their peers, and they help close the achievement gap. We need to recruit great teachers, develop them, reward them, and retain them. It’s one of the most important things we can do to reform our schools. Our challenge is to make sure every child in America is learning from an effective teacher.

To accomplish this, we must identify great teachers who are working in classrooms today. Right now, our method of evaluating teachers is broken. As I’ve noted in recent speeches, the New Teacher Project issued a report in May finding that 99 percent of teachers were rated as satisfactory or better by their principals. If 99 percent of our teachers were satisfactory, our country wouldn’t have a pronounced achievement gap and our student achievement would measure up against our international competitors.

I believe that educators need to consider the achievement of students when evaluating teachers. Test scores alone should never drive evaluation, compensation or tenure decisions. But to remove student achievement entirely from evaluation is illogical and indefensible.

While many other factors are important to a state’s reform agenda, we must hold ourselves accountable for student achievement and outcomes.  At the end of the day, the only critical measure of success is whether children are learning, and test scores are an important measure of that.  Going forward, we need to improve the quality of tests and create accountability systems that integrate test results with other indicators, including graduation rates, classroom observation and peer evaluation of teachers.

In the recently released notice of proposed criteria for the Race to the Top Fund, my department proposed that any state that prohibits the link between student performance data and teacher evaluations will be ineligible for grants from the $4.35 billion fund. We believe that analyzing students’ test scores is one of the best ways to measure the quality of teachers. With a prohibition on using test scores evaluation of teachers, a state won’t be able to reach the goal of dramatically improving the quality of their workforce – one of the primary objectives of the Race to the Top Fund.

My staff and I look forward to hearing comments on the proposed criteria from people in the field, and we will carefully consider all comments submitted before the Aug. 29 deadline before publishing the final criteria and application later this year.

Today, more than ever, better schooling provides a down payment on the nation's future. The Race to the Top program marks a new federal partnership with states, districts, and unions to accelerate reform and provide our nation’s children the schools they deserve.

It’s totally fair. This is a race. If you show up at a race all flabby and out of shape and you don’t even have a pair of running shoes, all you are doing is crowding out the other runners who are serious about winning.

It's still too early to tell, but there is a chance that this Race to the Top could live up to its billing as an historic effort to advance education reform in the U.S. But the politics here are tricky – President Obama and Education Secretary Duncan are running head first into the blob.

One of the most important things the president has done is to break the code of silence that exists among political insiders. The issues he and Duncan have raised around charter school caps, equitable school funding, teacher accountability, and student rather than adult-driven reforms have been on the school reform agenda for years. They do well in the light of day, or at summits attended by policy wonks. But everyone reading this blog understands that good ideas are frequently killed in political back rooms by special interests, outside the glare of public scrutiny. Those who want to do the right thing are discouraged. What Obama and Duncan have done is give these reformers hope, and empower them with new levers with which to affect change.

This is not to say that this is going to be easy. Even though those who oppose Obama and Duncan's reforms have been very quiet, you can bet your mortgage they are already figuring out which levers THEY can try to pull to stop all of this reform talk in its tracks.

At this point, what Obama and Duncan have done is pick low-hanging fruit. Charter schools caps that prevent thousands of kids on waiting list from going to schools of their parents' choice are indefensible. So politicians who have carried charter opponents' water on this issue have backed down, and some cases reversed their positions publicly and emphatically. There have been huge victories over the past few months in Tennessee, Connecticut, Rhode Island, Indiana, and Massachusetts as a result of these efforts.

In my home state (New York) we’ve got big problems with a law on the books that prevents student performance from being used in making teacher tenure decisions. The law is so bad, that when reporters wrote about it last year, not even the teachers union would claim credit for having pushed it through the legislature. So yeah, New York should have to adios this law before it can apply for Race To The Top funding, especially since nobody seems to want to defend it anyway. If New York doesn’t lift a finger and it gets funding in the Race, Obama and Duncan will essentially be lifting a finger (you know which one) to all of the other states that are doing everything they can to show they are ready for change and reform.

We at Democrats for Education Reform are pleased that thousands and thousands of children will get more resources and better instruction this coming school year as a result of President Obama’s efforts. But we won't be satisfied until all 50-plus million school-age children in the U.S. have the same opportunities.

President Obama already has demonstrated his knowledge that a state’s laws on the use of student data do not always determine what occurs in practice at the school district level. He demonstrated this knowledge when, in his first major policy speech on education, he singled out the Long Beach Unified School District in California for its effective use of data to improve instruction and close achievement gaps. How could this be, when California also has been singled out, in a negative way, for its law that prohibits the state from using student data to evaluate teachers?

The answer is simple. The California law only prohibits the state as a whole, not individual school districts, from using data to measure achievement student by student, teacher by teacher, and school by school. The reality is that here in Long Beach, we’ve been using data in this manner for 15 years. Our approach has not been punitive, though. It has been collaborative. We have custom built a sophisticated web-based system that electronically gathers, sorts and displays data in a way that supports the day-to-day work of our teachers, many of whom were instrumental in providing us the feedback we needed to develop this system over the past several years.

Today, our teachers can custom sort student test results with the push of a button. Teachers here can, for instance, sort students by their level of performance so that instruction can be properly differentiated for small-group settings.

In determining who receives Race to the Top funds, the Obama Administration would do well to continue its practice of shining a spotlight on effective school districts, rather than expending energy trying to change state laws that have little practical impact on student learning. To do otherwise might unfairly write off entire states and, by extension, many large, urban school systems that are showing great promise and want to build upon their efforts to close achievement gaps. Secretary Duncan has called Race to the Top an opportunity to turn islands of excellence into systems of excellence. If that is the goal, then individual school systems, not states, should be the ones applying for the funds.

For more on this matter, including how Long Beach is using data to support its classroom teachers and to improve learning, see this week’s article in the Los Angeles Times.