For-profit, or career, colleges, play an important role in the post-secondary education environment by ensuring access to higher education for millions of Americans and working towards President Obama’s goal to lead the world in college graduates by 2020. For-profit colleges should be treated as equal partners and an essential cost effective way of assisting the President in meeting those goals. Should the Gainful Employment regulation be imposed, it would incur disproportional harm to low-income and minority populations by discriminating against students who must borrow the needed tuition to attend college.

Studies show that by offering career-focused curriculum to many students, who may be working a full-time job, a single parent, minority or lower income, career colleges do a much better job at graduating low-income students than their four-year and two-year public and private non-profit counterparts. And despite similar enrollment factors, the graduation rate for two-year career college is more than double the rate for community colleges. What’s more, at four-year institutions, with more than 40 percent of their students receiving Pell grants, the graduation rate at career colleges is 38 percent as compared to 33 percent at public institutions.

What should not be lost in this debate is the financial contribution that career colleges deliver to the economy. Career colleges are the only institutions that pay state and local taxes, contributing nearly $1 billion to tax revenues in 2008. The proposed Gainful Employment rule creates a false perception that career colleges cost American taxpayers significantly more for student education than their non-profit college counterparts. In fact, total federal and state grant subsidies for four-year public colleges are over twice that of career colleges. Clearly, during these difficult economic times, career colleges are helping to improve the sustained 10% unemployment that our country has endured for over two years.

The Department of Education should follow President Obama’s commitment to providing opportunity to millions of low-income Americans by rescinding the proposed Gainful Employment Rule.

My answer to this question draws upon my testimony to the Senate HELP Committee from April 2010 as well as a recent op-ed from New Teacher Center (NTC) policy director Liam Goldrick:

The NTC’s philosophy on teacher effectiveness rests on an understanding that great teachers are made, not born. We believe that every teacher deserves rich professional support and collaborative opportunities, including focused induction during their initial years in the profession. Efforts to improve teacher effectiveness also must address teaching and learning conditions— including the critical role of supportive school leadership, opportunities for leadership and collaboration, and individualized professional development—that greatly impact teachers’ chances of success.

Improving teacher effectiveness requires a systemic approach. Federal and state policies should aim to provide a continuum of support—from initial preparation through induction and into career-long professional development—to strengthen the skills and abilities of all teachers to impact student learning. Such a developmental focus is too often missing in current conversations about teacher effectiveness. New teacher induction and job-embedded professional development are key strategies to help transform good teachers into excellent ones and average ones into great ones, and assist those who might otherwise fail, leave, or soldier forth in isolation to strengthen their skills and abilities on behalf of their students.

Accelerating teacher effectiveness is where our greatest opportunity lies. Measuring teacher impact without providing opportunities for educators to strengthen their practice will ultimately fail. A working definition of teacher effectiveness must meet the professional needs of individual teachers. If it doesn’t reach into the classroom, it won’t matter. Such a definition narrowly aimed only at the so-called ‘best’ or ‘worst’ teachers will be a missed opportunity to strengthen teaching effectiveness in every classroom throughout our nation.

NTC Op-Ed: http://www.newteachercenter.org/pdfs/goldrick-fedpolicy_effective_teaching.pdf

U.S. Senate Testimony: http://help.senate.gov/imo/media/doc/Moir.pdf

Mr. Finn is correct in calling this issue a “half-brainer”. But the other half of my brain is marveling at the fact that no one here is disputing the basic concept of results-based evaluation, only the way results are defined. When I started working in ed reform fifteen years ago, this concept wasn’t even on the table. Now it is the table.

But what kind of table have we constructed?

When it comes to tying teachers to numbers, we’re going to use test scores when and where we can. Why? Because that’s just what we’re going to do. This is a train that long ago left the station. As inappropriately reductive as it may seem to some of us, it simply makes too much sense to too many others.

Teaching may be turned into numbers but at least some of those numbers will come form sources other than tests. Mr. Kress points out that while most of us are evaluated by our results on the job, we’re also evaluated by other things. We can argue all day about what those other things should be. But I’d put them simply in the category of “values”.

To the extent that a given state or district values test scores, it will evaluate teachers with test scores. To the extent that it values other things, it will use those, too. Personally, I would prefer a qualitative approach based on human judgment and organizational values that go beyond test scores to the well-being of children. But I know this will not happen. Our dominant data-driven decision making culture favors a quantitative approach. To that end, teaching will be quantized and teacher evaluation will become more reductive over time. It will also become more consistent and, as a result, there will be less variance among teachers. And herein lies the first of several interesting implications of this discussion.

Mr. Vander Ark makes the simple but powerful observation that in the not-too-distant future, educational technology will give us more and better information about what and how kids are learning. It also follows that we can expect the percentage of time kids spend learning from machines to increase in the years to come. At a certain point, kids may spend a lot of time learning from machines. When teaching quality is quantized, variance shrinks, and machine learning time increases, the human component will become more standardized and less important.

This is not a position, or even much of a prediction. It's just a logical extension of dominant trends in instruction, assessment, and teacher evaluation. In the aggregate, teaching quality will probably improve, as will the quality of machine learning. But as machine learning takes over more of the instructional day, and as human beings are evaluated through quantitative systems that reduce the variance in their abilities, teachers risk commoditization.

If teachers become commoditized, there will be no rationale for pay differentials, and overall pay will drop dramatically. At some point, we may educate large numbers of children during significant parts of the school day on computers using no teachers at all, employing instead part-time hourly wage workers acting as monitors. In fact, this may happen in the coming school year in one large urban district.

It would be ridiculous to make the claim that some day computers will take over all teaching—or even half. But what if computers took over merely a third?

This already happens in situations like Scholastic’s popular Read180 program. In a 90-minute Read180 block, kids spend half an hour learning from computers, half an hour reading independently, and half an hour in a tightly scripted small group lesson with a teacher. Most places that have implemented Read180 for more than two or three years realize that the closer their teachers stick to the publishe-mandated structure and curriculum, the more consistent their results are. By definition, therefore, fidelity to the model is highly valued. It follows, then, that this is how Read180 teachers should be evaluated. It just makes sense to evaluate them this way as the program itself cannot be evaluated if it is not accurately implemented.

So how would we disaggregate a Read180 teacher’s contribution to the learning of his or her students? The teacher is only spending 33% of class time teaching each learner. And the teaching doesn’t vary much because each teacher must follow the defined scope and sequence curriculum with great accuracy. So who or what is doing the teaching here? The computers? The workbook? The kids (through their own silent reading)? Optimally implemented, Read180 removes all need to evaluate teachers for anything but fidelity to the program. And even a little infidelity can be tolerated because the teacher plays such a small role in the instruction and assessment students receive.

The future of teacher evaluation is already here. Teachers will be evaluated partly on the test scores of their students and partly on their fidelity to the standardized curricula and materials that dictate how they will achieve those results. At the same time, technology-based instruction will increase. Computers will gather rich assessment data. Teacher evaluation will become easier because teachers will become more alike and because they will be responsible for less of the instruction their students receive.

It isn’t possible to know how fast this scenario will develop. But the fact that it will is inescapable. Indeed, in forms like Scholastic’s Read180 program, it's already here.

If we reach the conclusion that programs like Read180 are more successful at helping kids learn than human beings alone, it will seem silly for us not to use them. It will also seem silly not to expand their use or to avoid updating to newer and more powerful versions of these systems as they improve. Even sillier than that will be the notion of paying teachers more when they are doing less.

Next, imagine adding a math equivalent. Or a writing equivalent. We don’t have to imagine these things because they already exist. Social studies is easy to teach via computer, too. Only science presents a real challenge because of its hands-on nature. Of course, we want our kids to learn technology, too, and we already put them on computers for that. Computers are great for visual arts. And I worked on music education software over twenty years ago. I doubt P.E. will ever go digital. But I wouldn’t be surprised if somehow it did. The exercise machines at my local health club look more and more like the computer I have at home all the time.

We can talk all day about how teachers should be evaluated. And while we do, the discussion will become less important. What we should be talking about is not evaluating teachers but evaluating the entire learning environment—of which the teacher is but one part. Personally, I like to think that the teacher is the most important part. But in some cases, especially as technology develops, this may not always be true.

Again, Mr. Finn is right when he calls teacher evaluation a half-brainer. Mr. Kress is right when he points out that most of us are evaluated partly on results and partly on other values that matter to our jobs, our co-workers, and our organizations. Finally, Mr. Vander Ark is right as well when he reminds us that educational technology is getting better by the minute, and that machines will be taking on more and more of the responsibility for instruction and assessment.

Take these trends and play them out to their logical conclusions. Consider also other powerful trends like the increased use of curriculum standards, increases in standardized testing, increases in adoptions of publisher-supplied programs along with stricter administration of these programs, and the future seems clear. What isn’t clear is whether this is the future we want. And this is what we should be talking about when we talk about evaluating the teaching our children receive.

If all of us would think about the best teachers in our lives, common characteristics would surely emerge. The most passionate teachers set themselves apart. They were excited about their craft. They celebrated when students finally grasped a concept or subject. They challenged students to do their best and chided students when they did not. They knew which students came to school hungry and, in some cases, fed them.

The really good principals were strong leaders who seemed to have earned the respect of their staff, their peers and their students.

The things we associate with the best teachers and principals haven’t changed; our learning environments and our expectations have. Clearly, it can be difficult to define an effective teacher or principal when the nature of their work is so subjective. But it’s not impossible. As Katie Walsh aptly points out, the so-called “superstar” teachers are rare. Today’s environment calls for more collaboration, and that could be a cornerstone of new ESEA teacher and principal assessment.

The highly regarded National Board for Professional Teaching Standards notes, “The conventional image of the accomplished teacher as solo performer working independently with students is narrow and outdated. Committed career teachers assume responsibility in cooperation with their administrators for the character of the school's instructional program. They are team players willing to share their knowledge and skill with others and participate in the ongoing development of strong school programs.”

Indeed, high-performing schools need good teachers and principals, but they don’t become great places to learn until those teachers and school leaders join forces to build a collaborative learning culture that improves student achievement beyond what even the best of them can accomplish alone.

So the national debate can be about the conditions that are necessary to support effective teaching –not just individuals, but learning teams that include teachers, administrators and students who take responsibility for their own learning.

Again, collaboration.

Tom Carroll, president of the National Commission on Teaching and America’s Future, and Hanna Doerr, a program manager at the commission, wrote an excellent commentary on learning teams in the June 28 edition Education Week.

They note that high-performing learning teams have a game plan (curriculum and instructional strategies) that they develop together and work to improve through collaborative practice and professional coaching. They use authentic, real-time assessments to improve their performance – with real time feedback on the impact of their practice on student learning.

Of course, the teachers who are the Jedi masters among us remain unsung heroes, inspiring, prodding and pushing learners to exceed their own expectations. But imagine the impact a team of them would make.

If we are to treat this solely on semantics, "effectiveness" is about outcomes and producing a desired effect. We all know that the current and most easily drawn measures of outcomes are standardized student achievement test scores. Many of us also strongly agree that these are a weak single measure of a world-class education, and of teaching more specifically. The field is tying itself up in knots right now trying to come up with a comprehensive, consensus-based definition of the desired outcomes of teaching.

The Common Core standards focus primarily on important academic outcomes and are a worthy start, but they leave out the social and emotional outcomes. Trying to build in measures for all the components of a successfully educated student would be quite a feat and probably impossible to measure with any sort of validity or reliability. But, if we focus solely on the academic outcomes in the evaluation of teacher and principal effectiveness, we will lose the ability to be fair and consistent.

The myriad influences on student outcomes -- from family to television, curricula to other teachers, to neighborhoods, genetics, and germs -- makes precisely measuring the individual contributions of teachers and principals more than a little “ tricky.” One way to deal with this, the administration and most reformers have conceded, is through classroom observations, which assess teachers' behavior, knowledge, and skills—i.e., inputs and processes. I agree with this impulse and disagree with Checker Finn. We cannot afford to fail to understand, monitor, and improve the "intentions, resources, and efforts" that go into producing “the learning that comes out.”

As states engage in this work, I hope they resist the temptation to go the "highly qualified" definition route. They must resist falling into the trap of developing narrow definitions of effectiveness that serve as minimal standards and satisfy the bureaucrats, but are more or less ignored when it comes to the real work of leading, teaching, and learning. We need standards-based tools and flexible frameworks that allow districts to construct and re-construct their own more specific definitions of effectiveness that are useful and meaningful to practitioners.

However, while we’re engaging in the necessary work of creating measures that assess effective teaching, I hope we can avoid the history of incrementalism in school reform in America. Real danger lay in building evaluation systems that reify the factory model of schooling. I hope principal and teacher assessment systems are built that allow for innovative staffing arrangements, including the flexible assignment of students to teachers and other support personnel, the specialization of teachers’ and principals’ work -- whether in terms of subject matter expertise or pedagogical skill or leadership strengths, the use of uncertified personnel to take on limited teaching tasks, and the increased implementation of team teaching.

Along with our colleagues in other national policy organizations such as Public Impact and Ed Sector, Learning Point Associates is convinced that there are more innovative and effective ways to organize teaching and learning (see our Innovations in Staffing report for more detail). This approach might require broadening the focus from measures of individual effectiveness to organizational effectiveness, not to sidestep individual accountability, but to account for the unalterable fact that it takes a community to produce the gamut of desired effects we expect from our schools.

The impact on student learning in broad and comprehensive ways is a key element of teacher effectiveness. Yet we should reject the silver-bullet approaches that reduce the complexities of teaching to a numerical score that does nothing to inform practice or contribute to improvements in teaching standards.

Instead, we should promote a systemic approach to defining, measuring and supporting teaching excellence. Teachers should be assessed by how they practice the craft of teaching according to a respected, comprehensive set of professional standards. These standards should strengthen the processes we use to prepare, license, induct, develop and evaluate educators so that there is a seamless, interconnected and logical system that guides and supports educator learning, growth and performance—from the first day of a teacher’s career to achievement of National Board Certification and beyond.

Today there is great promise for teachers to resume their path toward “professionalization” and the accompanying responsibility for student learning it implies. Common standards are again emerging, and a new generation of performance assessments is on the horizon.

I recently announced the creation of the NEA Commission on Effective Teachers and Teaching. I hope the commission will guide us to determine what we must do to exercise more control over areas such as teacher training, induction and licensure, professional development and evaluation of educators. Accountants, nurses, doctors and lawyers all have a say in the professional standards, processes and procedures that govern their practice, and educators should have that same influence over their own profession. We must give teachers, principals, and the full complement of staff who support students the professional attention and respect they deserve to advance our profession and better serve our students.

During his talk at last week’s Aspen Ideas Festival, “Precious” screenwriter Geoffrey Fletcher took my breath away by illustrating the importance of teachers with a quote from the Talmud. The passage, which begins the novel on which the movie was based, reads: “Every blade of grass has its angel that bends over it and whispers, ‘Grow, grow.’”

Fletcher aptly described effective teachers as people whose spirits stay with us for the rest of our lives, quietly offering nudges of encouragement. Even after their students leave school, the voices of these teachers linger in their ears saying, “You can do this. You really can.”

Great teachers may come in different packages and work in various settings, but they all share three qualities: a commitment to clear and high expectations, an unwavering conviction that their students can meet those standards, and the dedication to help their students every step of the way.

The truth is that kids work hardest for their strongest teachers—the ones who personally engage them and help them succeed. When you ask students—especially those growing up in challenging circumstances—about their most influential teachers, they’ll say things like, “Mr. Wilson treats us with respect.”

The harder students work, the more they grow. And student growth needs to be at the heart of the new teacher-evaluation systems that states and districts throughout the country are creating.

Building a strong system of assessments won’t happen overnight, but better teacher evaluations don’t have to wait until the systems are perfect. Classroom observations by principals and specially trained peers can help identify strong teachers as well as those who need to improve. In addition, students and parents can offer helpful feedback for school and district leaders seeking insights into how well students are engaged.

A functional teacher-evaluation system is a beginning step in overhauling American education. Less effective teachers won’t magically get better based on evaluations alone. But creating effective systems is crucial.

Any organization that strives to improve takes performance evaluation seriously. They know that employees need clear standards, an honest appraisal of their performance in relation to those standards, and guidelines to help them improve. When school systems treat all teachers as equally effective—which is what most districts today actually do—they seriously undercut any improvement efforts.

Mounds of research show that teacher quality is the most important factor in growing student achievement. Kids with three strong teachers in a row can soar, regardless of family background. Kids who have even two weak teachers back to back rarely recover. We know we can’t compromise on teacher quality. Our kids deserve teachers who can foster growth.

A more honest teacher-evaluation system will help us restore fairness to an American education system that has grown increasingly inequitable. Instead of identifying strong teachers and assigning them to the children who enter school less prepared and behind their peers, we do exactly the opposite—and this increases the gaps between poor and rich, minority and white.

If we want to spur learning gains for low-income students and students of color, we must provide them with effective teachers whose support and encouragement will follow them throughout their lives.

Kate Walsh, president, National Council on Teacher Quality, submitted the following in response to this week's question:

An effective teacher, to go back to the original question, is someone who can move students successfully from one grade to the next. A highly effective teacher is someone who can make incredible progress closing big gaps in student learning. Unfortunately, those teachers of superstar status who are so critical in our battle with the achievement gap are a relative rarity, roughly one in seven teachers.

Teachers who are this effective generally display a host of skills which may not be measurable and which do not look the same from one teacher to the next. But all great teachers know how to nurture, prod, challenge, and motivate children to not only achieve success in school but as future citizens.

While fully recognizing the importance of these immeasurable attributes, NCTQ concentrates on those that can be measured and which policy makers can act upon. Some of what we advocate is pretty old news, though you’d never guess it by many current practices or policies. For example, no one denies that a teacher needs a solid foundation of content knowledge in order to be effective, yet all too many institutions (as well as states that license teachers) shortchange content knowledge. In the next few years, districts and states will struggle to put teachers in classrooms capable of teaching the new national standards.

More recently, we have learned from research about the damage to student progress caused by first-year teachers. New teachers rarely help students progress, no matter how they’ve been prepared. Children who are poor or in a minority group are more apt to be assigned a first-year teacher due to high turnover. We need much better preparation, more bell-to-bell supervision of first-year teachers, and strategies to retain the most promising teachers if we hope to diminish this negative impact—because, of course, there’s no escaping the need for new teachers.

Training can and should matter, but the current ways in which we train teachers seem to add very little value. As an organization that is a strong champion for alternative certification, NCTQ also knows that this route is not the only answer if we are to fill 200,000 new positions each year. Certainly a strategy that relies only on our ability to recruit the best and brightest is by no measure a complete solution. Combined, Teach For America and The New Teacher Project will only be placing some 7,500 teachers this upcoming school year.

Wherever those teachers end up, they’ll be best served by an effective principal. That’s someone who gives teachers enough room to be creative while providing the guidelines necessary for the entire staff to produce results. That principal should have a say, as we’ve stated in our “human capital” reports, on whom he or she can hire, and then follow up with an evaluation system that provides first-hand knowledge of each teacher’s work, so that the principal can make any staff changes necessary to maintain high achievement levels school-wide.

Combine an effective principal with a staff comprised, at the very least, mostly of effective teachers, and you have the foundation of a successful school.

Other than Diane’s, the responses thus far minimize the problems of both the definition and the measures. The new national standards that may be adopted by most states might be partially adequate definitions of 'what' for two subjects. Even then, the standards will be of limited use. For example there will be no mandated curriculum and students presumably will read a great many different things and write with equal diversity across schools and districts, meaning a great deal of what is important is not in the ELA standards. We don't have them for other subjects, other than voluntary and some state ones. So, what to measure in academics is a long way from settled, though proponents of national standards and tests want us to believe otherwise.

At least as important is that a wide range of studies over the years has found that parents, teachers, community people and even legislators have a wide range of goals for schools (one example is in Grading Education, by Rothstein et al). Academics are essential, but not sufficient. If we don't bother to determine "effectiveness" in these other areas while measuring and holding accountable in a few aspects of academics, inevitably schools will ignore those other important outcomes, as has happened with NLCB.

Then we come to the question of how to measure. True, no one has proposed test-score only. But when states decide that 35-50% of a teacher's evaluation will be based on student test scores, they are creating a situation in which teachers will have even more incentive to narrow and skew curriculum and instruction to fit the tests. The tests themselves are heavily skewed toward rote learning. Thus, the incentives for teachers and principals will be to undermine the quality of instruction in the name of accountability - all too familiar. And teachers should not be blamed for doing what their bosses insist they do, boost the scores.

Contrary to assertions here, “growth” models are not ready for prime time - at least not according to the National Academy of Sciences which carefully reviewed the extant literature. Further research has added to awareness of the inaccuracies and flaws, as Diane mentioned. Among the flaws is an inability to fairly attribute "growth" to one teacher independent of other teachers, schools, families and communities. That's aside from the fact that the models all rely on the narrow standardized tests that have so damaged curriculum and instruction.

However, good evaluation does make sense. (A recent Wall St Journal article, however, suggests that annual reviews are actually counterproductive, a point worth considering.) There are good ideas available for building good evaluation and improvement systems. Student learning should be a meaningful part of that, but standardized test scores only a small fraction of the part. Evaluating the work students actually do should be the primary component of the student academic achievement part of the teacher or principal evaluation.

Finally, this discussion, like most of the points argued by the self-styled "reformers," reinforces the wrong-headed notion that externally imposed accountability is the way to improve schools. It likely does have a modest role. As I explained in a recent Education Week Commentary, we should be designing a multi-pronged approach to evaluation. But far more important is the work of building better schools through the ongoing work of educators. Collaborative self-reflection along with learning and planning by teachers is a critical component of that effort. In its discussion of turnarounds, the Forum on Educational Accountability outlines the areas on which improvement should focus, but points to extensive research saying how to engage in that improvement should not be standardized.

This is not leaving schools alone to do what they want. It is to recognize the very sharp limitations of accountability, including of educators, as an improvement tool; and to also recognize the need to really focus on those things that can improve schools.

Checker is right that defining effective teaching is the easy part. Hopefully we can all agree that effective teaching means helping students learn what they’re supposed to learn—a teacher’s most important responsibility. So, student academic growth should be the predominant factor in determining effectiveness (although not the only one).

Measuring teacher effectiveness is harder, but it’s absolutely critical. If we don’t know how effective our teachers are, how can we help them improve or ensure that we’re retaining the best? Fortunately, measuring effectiveness is also eminently doable. It requires combining data from many different sources, including standardized tests, teacher-generated assessments and classroom observations. Each of these measurements has its own drawbacks, but together they can paint a clear picture of how successfully each teacher is helping students make academic progress and master material. I certainly don’t believe we should evaluate teachers based solely on standardized test scores, and I haven’t heard of a single credible proposal to do that, even though I’ve seen plenty of shadowboxing with that argument.

Some people say that we shouldn’t connect teacher evaluations to measures of student learning until we’ve developed a method that will be perfect right out of the gate. But that’s an impossible standard. The tools to measure student growth that we have right now may not be perfect, but they’re very good. The only way they’ll improve is if we actually start using and perfecting them, instead of throwing our hands up and walking away from the challenge. If we do that, we’ll be left with the useless teacher evaluations we have right now, which rate all teachers good or great and deny teachers the feedback they need to grow as professionals.

I have little to add to Checker's fine comment.

I would point out, however, that, as hard as we work at it, there will never be foolproof metrics or "decision rules" for all teachers and schools. There are no such foolproof metrics for doctors or lawyers or accountants or small business men and women or even journalists at the National Journal or staffers at the Fordham Institute!

Yet, most enterprises in our society make an effort at judging the effectiveness of their people. And those efforts typically have a great deal to do with the results their people achieve. So, sales matter. As do successes for clients and prizewining stories AND improved test scores.

Should helping the team matter? You bet. Should growth matter? You bet. Should hard-to-measure factors be taken into account? You bet. Enterprises all across the spectrum find ways to make these sorts of assessments of professional effectiveness all the time. They're never foolproof or perfect. Nor should they be perceived as doing great harm because they're not perfect.

Indeed the stronger case is that far greater harm comes from systems that, out of fear, inertia, or resistance to improvement, do little or nothing to measure and reward their people's effectiveness.

Great opening Checker.

Teachers rightly worry about using a single source of data to judge performance but we'll soon have a vast array of content-embedded assessment as well as teacher-scored assessment that we'll be able to consider when judging student growth and teacher performance. That's why I'm a fan of Terry Grier's approach in Houston which seeks to use "all available information" when making performance judgements. We'll need to build flexible frameworks that year after year incorporate new data elements into increasingly more well rounded pictures of performance.

Which teachers are "effective" and which are not? And what is meant by "effectiveness"? At present, most policymakers and foundations define effectiveness as the ability to produce higher test scores.

But this is a very limited and ultimately not helpful definition. For one, most teachers do not have annual test scores by which they can be judged. For another, very bad teachers may produce higher test scores by drilling students on the questions likely to appear on state tests. Thus, the scores may misidentify effective teachers while not being available for the majority of teachers.

Many research studies show that teacher "effects" are highly unstable. A teacher who produces high scores one year, may not do the next. This is because scores reflect many things other than the teacher's contribution, such as the composition of the class (a willing class one year, a class with 2 or 3 disruptive students the next), and scores are influenced by student effort, student attendance, and family support, as well as resources and curriculum.

It also is very difficult to isolate the influence of any one teacher, since student scores may reflect the work of last year's teacher or other teachers in other subjects.

Teachers should be judged by professional standards, not by politically designed formulas written by legislators who have not set foot in a school since their own student days.

We must take care to do no harm. At present, the political fads of the day have great potential for harm to students and to the teaching profession and to the quality of education.

Diane Ravitch

The easy part of this question is the definition. An effective teacher is one whose pupils learn what they should while under his/her tutelage. An effective principal is one whose school attains state standards, pupils learn what they should, etc. In both cases, it's a results-based definition. The hard part is developing and fairly applying a reliable set of metrics by which to gauge success, hence effectiveness. All sorts of experiments and pilot programs are under way along these lines (e.g. the Gates Foundation's "deep dive" districts) and America needs plenty more such. We obviously don't yet have foolproof metrics or "decision rules" for all teachers and schools. But we obviously need to develop them. In principle, I think, four and a half decades after the Coleman Report, I think American education has finally accepted the key reality that what matters about a school (or a teacher) is not the intentions, resources, efforts or activities that go in but, rather, the learning that comes out.