NCTQ is deeply committed to high-quality formal teacher preparation. Improvement in the performance of teachers—especially during their first few years in the classroom—is essential to improving the performance of students.

Both program approval standards set by states and accreditation standards set by private organizations provide no indication of the quality of one institution's preparation relative to another. Admittedly, this is by design; both the state and the accreditation agencies do not rate or rank schools because accreditation is a pass/fail system that determines whether preparation providers meet national professional standards. But, unfortunately, this leaves consumers—aspiring teachers and schools which hire teachers—in the dark.

Further, a relatively small number of the more than 1,400 schools of education in the country have ever lost program approval by their states or had their regional accreditation withdrawn. This may lead consumers to the conclusion that the roughly 1,400 schools are performing at a satisfactory level, which is just not the case. While accrediting agencies have expressed a commitment to utilize the student performance data to drive program improvement from state data systems now under construction, not a single study has found that an accredited school of education is of higher quality than non-accredited schools.

As a nonpartisan research and policy organization committed to ensuring that every child has an effective teacher, NCTQ is stepping into this vacuum to help consumers distinguish between good, bad, and mediocre education schools. We do so by setting the bar higher than it has been set traditionally.

Here's just one example (from the many that can be pulled from our various ed school reports) of the difference between an NCTQ evaluation and the standard accreditation process. The typical accreditation agency says that, when it comes to admissions, education schools must have multiple standards that are clearly described and well-advertised. The agency does not specify, however, what these standards must be. NCTQ says that the institution should only accept teacher candidates who are in the top half of the college-going population.

The quality of teachers an institution produces is critically important. One of NCTQ's standards is whether education schools are using data on the performance of their graduates to improve their programs. Unfortunately, only a few states currently have data systems designed to provide this information—Louisiana and Florida, for example. In states where it is not yet possible to apply this standard, we are evaluating whether education schools are at least using other sources of data on the performance and retention of their graduates from surrounding school districts.

Once output data on the effectiveness of a program's graduates becomes more available, however, it's not clear that we'll learn as much as is hoped. We'll certainly know more about the effectiveness of one education school relative to others. We'll know who is doing a better job in reading or mathematics relative to others. But we won't learn what value education schools should be adding relative to the highest possible performance standards.

In other words, such a situation calls to mind the difference between norm-referenced and criterion-referenced tests. A norm-referenced test compares students only to one another, without indicating what students should be able to do. NCTQ's studies, like criterion-referenced tests, measure programs against a set of clearly defined performance standards.

And, unlike outcome data, our standards address more than reading and mathematics. NCTQ looks at all aspects of teacher preparation.

This argument, heard often from education schools we studied in Texas, does not hold water. NCTQ is suggesting that institutions exceed state standards, which is certainly not a violation. In our Texas study, in fact, many education schools earn the highest rating on a variety of NCTQ standards.

The one exception is our exit standards. Meeting those would require that schools establish a higher standard on exit tests than what states require on tests for teacher licensure. So they would either have to come up with different tests or demand higher scores.

State regulations, rules and standards generally do not encourage education schools to meet a higher standard, but states certainly do not prohibit education schools from doing so.

NCTQ has a strong track record of producing fair and objective studies. While we do not defend the status quo, we are deeply committed to high-quality formal teacher preparation. In fact, we consider education schools to be necessary, whereas many education reformers dismiss teacher preparation, arguing that the only solution to improving teacher quality is to attract more talented people to the profession.

Unfortunately, studies that compare teachers who enter the classroom with little or no training with those who have gone through undergraduate or graduate-level teacher preparation find no aggregate difference in performance. Only one conclusion can be reached: Most education schools—but not all—are not adding value. NCTQ believes that teacher preparation, if it were selective and rigorous, can and should add real value.

The standards were developed over five years of study and are the result of contributions made by leading thinkers and practitioners from not just all over the nation, but also all over the world. To the extent that we can, we rely on research to guide our standards. However, the field of teacher education is not well studied.

Therefore, when we can't rely on research to guide our standards, we look to the practices of higher performing nations; where relevant, the practices of other professions; and the best consensus-thinking.

NCTQ's mission—executed in an independent, nonpartisan manner—is to ensure that all children have effective teachers. It is with a focus on students that we undertake this work. We are only one organization devoting the time and effort necessary to evaluate teacher preparation programs. Others are welcome to do the same.

First, NCTQ never looks at just a syllabus when rating a course. We also have experts read and analyze every text that is required for a course, as well as any "reading packets" put together by the instructor. In college reading courses alone, NCTQ has reviewed over 700 texts.

Second, the concept of using syllabi to judge quality is certainly not unique to NCTQ. When professors develop their syllabi, they do so not just for the benefit of their students, but also to provide assurances to their departments about the material they intend to cover. In many cases, syllabi must be turned over to states and/or accrediting bodies for approval purposes.

In effect, a syllabus serves the same purpose as a restaurant menu. It tells you, for example, that chicken is being served but sea bass is not.

It is only because NCTQ is looking for evidence of basic, essential topics that we can make fair use of syllabi for rating course quality.

Here's a useful way to better understand NCTQ's methodology: If a syllabus for an early American history course contains no mention of topics associated with the American Revolution, one might rightfully suspect that the course is deficient, because the Revolution is considered a basic, essential topic. But it wouldn't be as troubling to discover that Benedict Arnold was omitted from the syllabus. The professor might not have thought to list Arnold, and in any case he might end up talking about Arnold in a lecture—not unlike adding a "special" dish to a menu one night. But even if the professor doesn't, it would be unfair to assume that the course is deficient as a result of the omission, because Arnold is not a basic, essential topic.

When NCTQ rates a reading course, for example, we look for references to the five components of scientifically based reading instruction, considered to be no less fundamental to the teaching of reading than the American Revolution is to early American history.

When we're not certain, we say so. We mark the syllabus as unclear. If the text is equally ambiguous, the course is rated as unclear. We never give a rating without sufficient evidence.

Another point about syllabi is important. Syllabi tend to provide a more ambitious picture of the content of a course than the professor is usually able to achieve. Professors generally overestimate what they will be able to accomplish, not underestimate. For that reason, a methodology relying partially on syllabi could end up rating a course higher than it might actually deserve.

NCTQ's analysis of an education school is not a minor undertaking. For each school, the entire preliminary rating requires approximately 40 hours. This estimate does not include the time dedicated to refining the ratings after receiving feedback from institutions.

Each analysis starts with approximately 18 hours of an initial examination of catalogues, websites and syllabi collection. This sets in motion a process for verifying and double-checking scores; the use of expert consultants to examine syllabi and textbooks (with requisite checks on their work); and the work of separate teams conducting survey work.

NCTQ examines institutional admissions standards and a program's own admission policy; general education course requirements and course descriptions; course requirements for secondary teachers in their subject area; professional course requirements and descriptions; syllabi and textbooks for selected coursework; student teaching policies and practices; graduation requirements; course schedules and teaching assignments; faculty listings; and a program's record of interaction with area school districts.

We are using course descriptions only to determine whether an education school devotes a course or part of a course to essential areas of study—for example, special education, reading across the content areas and methods. These broad, elemental areas can reasonably be expected to appear in a course description.

We never use course descriptions to judge the quality of a course, but these descriptions are nonetheless quite revealing. For instance, some courses focus much more than others on the topics that will help teachers walk into their classrooms equipped to do their jobs. In the two sample course descriptions below, which would appear to better equip a prospective high school social studies teacher to understand adolescent development issues?

We think the former is better than the latter and would award it full credit, with the latter receiving only partial credit for trying to cover too great an age span in a single course.

Or consider the following required coursework from two different programs attempting to familiarize teacher candidates with issues regarding diversity that are relevant to their teaching. The first program requires three courses, each in a different department, while the second requires only one. Which of the two programs appears the most coherent approach?

If we evaluated these programs on descriptions alone, both would receive full credit, but we would definitely note the efficiency of Program 2.

Lastly, we always provide the school of education an opportunity to enhance or correct any conclusions that we have reached based on course descriptions.

Absolutely. About 15 percent of our preliminary ratings from analysis dependent on syllabi, texts and catalogs change based on institutional feedback. For analyses related to more complicated standards, notably those on student teaching, an even larger percentage of our preliminary ratings may change.

It is important to remember that NCTQ gains nothing and loses plenty by being wrong. If our ratings are repeatedly shown to be inaccurate, they will offer little value.

We began by asking the professors themselves to send to us syllabi, but most of our requests were denied. So, during our earliest studies we mostly visited each campus, to collect the syllabi in person, either from the appropriate offices or students. Recently, we have requested syllabi directly from education school administrators, sometimes through freedom of information requests. We continue to refine the method of collection and employ multiple strategies. One thing NCTQ does not do is use syllabi posted on websites, as they are often outdated.

By choosing to prepare public school teachers, education schools and the institutions in which they are housed, whether public or private, are operating in the public arena. By extension, any coursework offered can be evaluated by the public as to whether or not it furthers the public interest.

With that in mind, NCTQ believes that the tradition of "my way in my classroom" is antithetical to learning all we should about educator preparation. As with all sciences, the sharing of data and techniques is a necessary condition for creating an evidence-based consensus and extending the boundaries of knowledge. In fact, many colleges already post syllabi on publicly accessible websites. An excellent example is the Massachusetts Institute of Technology's visionary OpenCourseWare, where materials including syllabi, course notes, assessments, projects and videos are available for 1,900 courses.

It is also standard operating procedure for state departments of education, higher education bodies and accrediting bodies to collect these syllabi, for much the same purpose as our own.

Finally, we only use the syllabi that we collect for analysis. If we want to publish a syllabus as an example, we always ask for permission from the professor.

Research involving individuals properly does rely on their choice. However, education schools are not individuals. They are institutions we've included in a study designed to serve the public's interest.

No, it is not. Initially, we did not officially announce the start of a study, but we did let education schools know about our work as each study progressed. Currently, we inform each institution when a study has begun.

Then, after we've collected the necessary materials, we continue to communicate with schools throughout the study. These communications involve correspondence to confirm coursework; to comment on preliminary ratings and/or provide additional material relevant to those ratings; and to provide a comment for publication in our report that is edited only to meet the word limits we have established.

Before we publish any study, we send our final ratings to education schools in time for them to submit any last-minute revisions or make last-minute comments.

Some standards are more important than others. Some have been field tested longer than others. Some articulate goals which we know are not widely practiced in education schools, but they represent a direction which we believe education schools should move. For these reasons, not all standards receive the same weight in a study, and some may not be weighted into programs' overall performances at all.

Absent sitting in on courses for an entire year, NCTQ does not believe that campus visits would provide a measurably better perspective than our examination of formal course materials and surveys do. NCTQ is looking at the critical foundations upon which the quality of a program is built, and no amount of engaging classroom interaction can fill in the gaping holes we often find in the content, requirements and sequence of a program.

Campus visits may be seen as essential by education schools to assessing quality because they are a customary feature of the accreditation process. Undoubtedly they do serve a useful purpose in assessing programmatic features that are relevant to accreditation. However, the link between accreditation and program quality is by no means evident, and, by extension, there is no evidence that these site-based visits offer a definitive measure of program quality.

In contrast, our analysis, coupled with the institution's opportunity to provide critical feedback, is more likely to yield an accurate measure of program quality than are short visits orchestrated by the objects of our study.

Our democratic philosophy seems much more alive at the doors of education schools than at the doors of our K-12 schools, with the result that it doesn't have a democratic effect at all. You won't find high performing school districts willing to take in teachers who were themselves poor students and have demonstrably low academic performance.

Where are those teachers teaching? It is poor and minority children who are assigned the teachers with the weakest academic backgrounds. The notion that academic background shouldn't matter all that much has had disastrous consequences for poor and minority children, the ones who are most in need of a high-quality education. We tend to be okay with letting low performing teachers into the profession as long as they don't teach our own kids.

That's why we point to other countries. Those which have better education systems than our own, and are more successful at educating minority students who live in poverty, have much higher standards than the United States does for entering the teaching profession. Finland's education programs, for example, only admit the top 10 percent of their high school graduating classes, Singapore's the top third. In many of our nation's teacher preparation programs, by way of contrast, there are virtually no admission standards at all.

Less than five percent of all education schools are housed in institutions considered to be most selective, but at least one quarter of all education schools are housed in institutions which have no standard for admission. We're advocating that the floor be no lower than standards admitting the top 50 percent of the college-going population—still a long way off from the practices of other countries.

Actually, if you look at the body of research on such tests, its overall conclusion is that teachers with higher SAT/ACT scores appear to be relatively more effective. That doesn't mean that an applicant who didn't get a 1400 on the SAT or a 23 on the ACT might not become a great teacher one day. But it does suggest that there should be a floor on the SAT/ACT admission standard. To the extent that SAT and ACT measure intelligence and verbal and academic ability, being smart is generally a benefit in the classroom, as it is in any profession.

There's little evidence from states that have raised their standards that big teacher shortages ensue. In truth, raising standards makes the profession more attractive to academically talented individuals who are otherwise put off by the profession's low standards. Massachusetts ignored warnings about shortages when it raised its standards to among the highest in the nation and has not experienced any shortages. Likewise, England found that teaching became the most popular profession among undergraduates and graduates after program standards were raised.

In fact, because many people getting teaching degrees never intend to teach and seem to be getting teaching degrees only because it's an easy major, some drop-off in admissions may not mean that fewer teachers are available in classrooms.

While a far lower proportion of the most talented minority students choose to become teachers than do talented white students, highly selective education programs that heavily recruit talent of all colors do succeed in attracting minority teacher candidates. For example, 30 percent of Teach For America teachers are of color.

The long-term strategy to achieve a teaching force that better mirrors the student population is to immediately improve the educational prospects of every child by putting an effective teacher in every classroom. Those effective teachers will be produced by education schools with higher, not lower, admission standards. Among many other things, the fruit of more effective instruction will be many more minority high school graduates qualified to enter teaching, no matter how selective admissions become.

The evaluation of Reading First did not in any way, shape or form show that scientifically based reading instruction is ineffective. The impact study only showed that Reading First schools did not experience greater gains in student reading achievement than control-group schools. One of the primary theories for why this is so—promulgated by, among others, one of the contractors that conducted the impact study—is that there was a high level of spill-over of scientifically based reading instruction into the control schools. The study, in other words, was inconclusive.

Unfortunately, evidence suggests that programs touted as "balanced" are frequently little more than whole language approaches with, at best, some cursory coverage of the five components of scientifically based reading instruction.

A program that attempts to balance a scientific approach with a literature-based approach—that is, one that isn't a whole language program in disguise—would meet NCTQ's reading standard. As long as the program adequately covers the five components without including non-research-based strategies (such as three-cueing systems), NCTQ would not question this approach.

Both basic skills tests and content tests used for state licensing have many weaknesses. First, the basic skills are too easy (testing elementary and middle school level proficiency) and may not be required until after program completion, meaning that programs devote valuable time to remediating teacher candidates who are deficient.

Content tests can be deficient both substantively and structurally. At the elementary level, the tests used for licensing are too easy to pass. At all levels, subjects are tested together and do not have separate passing score requirements. (For example, the typical elementary content test includes reading pedagogy, English/language arts, science, social studies and mathematics, and the typical middle and high school social studies tests cover history, government, geography and economics.) This allows a high score in one subject area to compensate for a low score in another. At the elementary school level, one result is that candidates who have little to no skills in mathematics—typically the subject area with the lowest performance—can still be licensed. Lastly, many states have loopholes that allow teacher candidates who have not yet passed a licensing exam to teach one, two, or even three years.

For specific courses, such as classroom management and special education techniques for general educators, NCTQ recommends a focus on the gradespan appropriate for the certification level of the prospective teacher. For example, a high school mathematics teacher candidate should learn management and special education techniques for adolescents, and not have to sit by while early childhood teacher candidates discuss how best to handle the temper tantrums or learning disabilities of a kindergarten student.

There are likely valid topics from each gradespan that can be included in the other spans, but the focus of each course should be on the teacher's specific age group.

Other coursework need not be focused on a gradespan. For example, a course dealing with education policy challenges that addresses topics such as the achievement gap and the rationale for charter schools need not be focused on any one gradespan.

Frankly, we think there is a problem of priorities or basic program viability that needs to be pointed out if required courses aren't offered at least once a year. Talented individuals will not pursue a program they are prevented from completing on time.

There has been a consensus for decades that teachers, especially secondary teachers, need to have the same subject-matter education as anyone else on campus. And the fact is, we still see plenty of evidence that institutions are allowing their teacher candidates to pursue a less rigorous major than other students on campus. This is a problem because we think it's especially important for the field of education to not be seen as an academic backwater into which less talented students drift.

While we can conceive of coursework that mediates subject matter in ways that would be very useful for the secondary teacher, as a rule, we do not think that such courses should substitute for the courses required for a major. Not only is such an approach more rigorous, but we believe that it would attract to the profession only those candidates truly serious about teaching—and the work involved in becoming a professional. For all of these reasons, it is important for all teacher candidates to have non-education majors that are indistinguishable from the majors of non-teacher candidates.

The three-course requirement is not arbitrary, but based on a calculation of how much class time is required to cover the 12 essential topics of elementary mathematics. A two-course requirement means that the institution is either skipping essential topics or treating topics too lightly. A four-course requirement is likely to be inefficient.

Further, the lack of any clear consensus among American institutions about what aspiring elementary teachers need in mathematics is quite problematic. The diffuse requirements compare poorly with the more consistent standards found in other countries. In the United States, institutions require anywhere from no math courses to five. Some institutions require that these courses be designed specifically for the teacher; others insist that teacher candidates take the same courses as other students on the campus. Many institutions fail to delineate the type of course that a teacher should take, viewing all courses as equally valid.

Clearly this absence of a standard must be resolved in order for the United States to improve its standing in mathematics performance with other nations.

We should note that we do make an exception to our three-course standard. Institutions that have a highly selective admission process can probably cover all the topics needed in two courses.

In a word, no. Teachers cannot rely on the procedural knowledge of mathematics that they acquired as elementary students themselves to be able to successfully teach those topics. Even first grade teachers need to acquire a much deeper understanding of mathematics than is reflected in their ability to get the answers right on a basic mathematics test. They must know how the number system is put together; why the procedures for adding, subtracting, multiplying and dividing are what they are; and that algebra is just generalized arithmetic. Teachers must know elementary concepts at a depth sufficient for instruction, an understanding that can best be acquired through classes specifically designed for future teachers.

They need this knowledge because, in the past, teachers have made mistakes—by failing to understand why students make common errors, or that there are multiple, equally legitimate, ways to arrive at an answer.

For example, American elementary teachers sometimes have trouble teaching children who have emigrated from Mexico. The teachers can't understand the students' procedures, labeling them incorrect even though Mexican schools' methodologies are simply different, not incorrect. But many American teachers have memorized algorithms without understanding how and why they work. As a result, they think the Americans method of computation is the only correct one.

This argument isn't new, and the dichotomy drawn between content and 21st century skills misses the point. Critical thinking is impossible without acquiring substantive knowledge. Someone who knows a lot about baseball is able to think critically about baseball, but that does not mean that he can think equally critically about nuclear physics. Critical thinking is largely associated with depth of knowledge of a specific domain of knowledge. It is not, as is commonly understood, a skill unto itself.

Therefore, elementary and special education teachers need to have a broad understanding of many subjects in order to field and pose questions about the myriad topics that come up day to day in the classroom. They also should know and love a particular subject well, inspiring students to share that passion.

We believe that providing teacher candidates with a broader knowledge base is one of the best ways to prepare teachers to handle the long-term demands of their complex jobs.