Testimony of Kati Haycock, President, The Education Trust Before the U.S. House of Representatives Committee on Education and the Workforce Subcommittee on 21st Century Competitiveness
Mr. Chairman and members of the Committee, thank you for this opportunity to testify before you today on the number one ingredient of high achievement: quality teachers. This Committee has already exhibited great leadership in the effort to improve teacher quality by including important new teacher-related provisions in the Higher Education Act of 1998 and, more recently, by including expansive teacher-related provisions in No Child Left Behind. These were very important first steps.
My purpose here this afternoon is to remind you why this subject should remain high on your agenda as you reauthorize the Higher Education Act and to suggest some ways in which you might build on the momentum you created in the earlier laws.
First some context.
Getting Students to Meet Standards: The Importance of Schools, Teachers
For many years, most Americans believed that what children learned was largely determined by their family background. They believed that, no matter what schools did, children who came from low-income families with low levels of parental education wouldnt learn very much, while those who came from more affluent and better educated families would excel.
Research undoubtedly fed this view. Studies like the so-called Coleman Report issued in 1966 indicated that schools accounted for very little in the equation of academic achievement.
More recent research has, however, turned these understandings upside down. It turns out that some things that schools do matter hugely in whether students learn, or whether they dont. And the thing that matters most is good teaching.
Teacher Impact on Individual Children
The impact of teachers on children is clearest in the research of statisticians and economists who are studying the relationship between individual teachers and the growth students achieve in their classrooms during the school year. This approach is called value-added measurement.
William L. Sanders, who founded the Value-Added Research and Assessment Center at the University of Tennessee, Knoxville, has studied the teacher and student data extensively. In examining data in the state of Tennessee, he found that low achieving students gain about 14 points each year on the state test when taught by the least effective teachers, but gain more than 53 points when taught by the most effective teachers. Teachers make a difference for middle- and high-achieving students, as well. On average, high achieving Tennessee students gain only about 2 points a year when taught by low-effectiveness teachers, but more than 25 points a year when under the guidance of top teachers.
In summarizing available research, Eric Hanushek, an economist at Stanford University, estimated the difference in annual achievement growth between having a good and having a bad teacher can be more than one grade level equivalent in test performance.
Moreover, these teacher effects appear to be cumulative. For example, Tennessee students who had three highly effective teachers in a row scored more than 50 percentile points above their counterparts who had three ineffective teachers in a row, even when they initially had similar scores. An analysis in Dallas found essentially the same pattern there: initially similar students were separated by about 50 percentile points after three consecutive years with high- or low-effectiveness teachers.
As in the case of annual impact, the cumulative impact of teacher quality is biggest for initially low-achieving students. A recent study in Tennessee suggested that students who fail the states 4th grade examination are six times more likely to pass the graduation examination if they have a sequence of highly effective teachers than if they have a sequence of low-effectiveness teachers.
In other words, students whose initial achievement levels are comparable have vastly different academic outcomes as a result of the sequence of teachers to which they are assigned. Differences of this magnitude50 percentile pointsare stunning. They can represent the difference between a remedial label and placement in the accelerated or even gifted track. And the difference between entry into a selective college and a lifetime working at McDonalds.
The View from Americas Classrooms
My colleagues at the Education Trust and I spend a lot of time in classrooms. Over the past ten years, we have spent thousands of hours working with teachers around the country.
In that work, we sometimes see absolutely wonderful teachingin all kinds of schools. In fact, even the lowest performing schools always have at least some quite terrific teachers. But, especially in the highest poverty schools, we often see teaching that is quite dreadful.
These tendencies are clear in the data, as well. No matter which measure of teacher qualifications you usecertified vs. uncertified, in-field vs. out-of-field, experienced vs. inexperienced, high scoring on licensure exams vs. low-scoring, effective vs. ineffectivepoor children end up with less-qualified teachers.
This pattern should have been unacceptable even when we didnt know how much difference teachers make. Now that we are certain of the difference, it is unthinkable that we allow poor children to continue to be taught by more than their share of our least well qualified teachers.
To give you some idea of why it is so important that we find the courage and creativity to turn this pattern around, consider these findings from a recent study of Texas schools:
By our estimates from Texas schools, having an above average teacher for five years running can completely close the average gap between low-income students and others. ( Rivkin, Kain and Hanushek, 2002.)
Six Core Problems
Looked at as a whole, there are at least six central problems in this arena that Title II of the Higher Education Act can affect:
- We dont have enough high-quality teachers in the categories and jurisdictions we need them in (e.g. math, physical science, special education)even as we continue to produce more teachers in categories we dont need;
- The teachers we have are unequally distributed across different kinds of schools and students;
- Fears about supply, distribution and diversity have prevented most states from raising standards for teachers to align with recent increases in standards for students. Academic departments outside schools of education have not been adequately engaged in setting standards for teachers or ensuring that teacher candidates get a strong college-level education in the subjects they will teach;
- Current patterns of transfer and resignation result in our losing disproportionately large numbers of the teachers we most want to keep (including high-end teachers, and teachers in high-poverty schools);
- Because of the ways in which the current Title II accountability provisions were crafted, too many institutions that prepare teachers have been able to avoid real accountability and, even within institutions where there is new-found accountability, those who do the academic side of teacher preparation are off the hook; and,
- Woefully inadequate data systems interfere with both reporting and action on these issues, and hamper the efforts of those who insist that teacher quality should be judged not on proxy measures of their qualifications but on what matters most: their ability to grow student knowledge and skills.
No Child Left Behind and HEA: The Need to Align
In general, the Higher Education Act needs to be more aligned with No Child Left Behinds focus on raising student achievement and closing gaps. The Higher Education Act should reinforce the priorities of NCLB and enlist higher education in a more aggressive effort to address both teacher quality and distribution.
Following are some of the options you may want to consider as you seek to realign the Higher Education Act with the goals and priorities of NCLB. Weve organized those options according to the six critical problems listed above.
Problems 1 and 2: Inadequate supply of high quality teachers in some categories, jurisdictionsand uneven distribution of those we have.
A. Funding streams in Title II should be much more highly targeted to categories and jurisdictions in which there are genuine shortages of quality candidates. We would suggest, in particular:
- A massive focus on the biggest problem of all: the inadequate production of secondary mathematics teachers (and, for that matter, of the mathematics majors from which those teachers are drawn).
- Dedicated funds to make much-needed improvements in programs that prepare early childhood educators to ensure they get high quality preparation in both content and pedagogy.
- Better targeting of resources toward helping produce teachers for the schools and districts most in need. Grants should be targeted to the districts having the most difficulty recruiting and retaining highly qualified teachers and the districts educating the most students living in poverty.
Moreover, grant applications that seek to address teacher recruitment/retention should be required to refer to the applicable states report, required pursuant to NCLB, on the disproportionate assignment of inexperienced, out-of-field, and unqualified teachers to teach poor and minority students.
B. These funding streams should be accompanied by much tougher evaluation of results and by strengthened accountability systems, as described below.
C. Congress should expand loan forgiveness to teachers in high-need academic subjects, in schools with the greatest shortages of highly qualified teachers, and to individuals working in pre-kindergarten programs. To maximize the impact on attracting teachers to high-need subjects and high-need schools, loan forgiveness should begin immediately upon entering the profession and increase over time.
Problem 3: Standards for teachers that are not always well matched with standards for students.
In the rush to put their accountability systems into place after the 1998 reauthorization of the Higher Education Act, most states did not pause to reexamine the extent to which existing certification requirements align with current standards for students. While Congress recognized the need to focus on the subject-matter knowledge of secondary teachers in particular, there is still inadequate involvement of departments outside of education schools in preparing teachers. In addition, teachers are not adequately prepared to use challenging State academic standards and assessments to improve teaching. In the new Higher Education Act, Congress should:
A. Ask each state to examine current certification requirements, eliminating those that are unnecessary to classroom effectiveness and strengthening those that are necessary. This reexamination should be informed by an analysis of data on the characteristics and qualifications of teachers who are effective in producing student learning, as contrasted with those who are less effective.
B. Require state officials submitting the required annual reports to the Secretary to attest that their standards for teachers are fully aligned with what it takes to get students to state standards and to summarize the evidence behind that conclusion.
C. Because most coursework completed by future teachers takes place outside schools of education, whole universities and colleges need to be involved in improving teacher preparation.
1. Faculties outside of education schools should participate in setting standards for the subject-matter knowledge needed to enter teaching.
2. Faculties in and out of education schools should collaborate on limiting barriers to students in other academic disciplines earning teaching credentials.
3. Special emphasis should be placed on improving programs that train early childhood educators.
D. Direct the Department of Education to prioritize grants that reform teacher preparation programs to ensure that teachers:
- Integrate challenging state academic standards into development of curriculum, assignments, and instructional practice;
- Align classroom-based assessments with challenging State academic standards; and
- Regularly analyze assessment data to improve teaching and instruction.
Problem 4: How to hold on to our strongest teachers
No Child Left Behind asks state and local education officials to make certain that low-income and minority children are no longer taught by disproportionate numbers of underqualified teachers. As they confront this challenge, however, most education leaders have very little to base their plans onother than their own hunches.
Although Title II authorizes grants to address these issues, research on the effectiveness of available options (for instance, increased pay, reduced student load, or extra support) is sorely needed. State and local grant monies should be directed to this purpose. Grantees should be required to report consistent and comparable data in order to compare the effectiveness of various approaches. Grant applications should be required to include a detailed description of evaluation plans, including the identification of an appropriate control group, and hard data on outcomes by which to measure the success of grant activities.
Problem 5: Strengthening our accountability systems for higher education.
Accountability for what matters most will only be possible when Congress insists that institutions of higher education conduct value-added analyses that truly account for teachers abilities to impact student learning. Until then, we are left with the best available proxies for what we really care about raising student achievement.
The Committee might want to consider the following amendments to existing reporting requirements under Title II:
- State reports should be signed by the Governor or the state official responsible for teacher certification, who would attest to the accuracy of the data.
- The submitting official should not sign just for accuracy of the data, but should be asked to attest that state standards for teacher certification are adequate to ensure that teachers have the knowledge and skills to teach students up to state standards.
- Pass rates for licensure exams should be reported for all test takers who have been enrolled in a teacher preparation program for at least two semesters, not just program completers. Reporting pass rates only for program completers has enabled institutions to withhold the label completer from students who have not passed the exam(s) and has undermined the value of this data for evaluating the quality of teacher preparation programs. Pass rates should also be reported for participants in alternative certification programs.
- Institutions that use licensure examinations of some sort for entry should also be required to administer exit examinations that enable judgments to be made about institutional value-added;
- Licensure exam pass rate data should be disaggregated by race/ethnicity and states should be asked to set institutional pass rate goals by group;
- Accountability systems should include consequences not only for education schools, but for the academic departments that co-produce teachers. (For example, if secondary mathematics teachers fail the mathematics portion of the exam, but have had all or most of their mathematics instruction in the math department, there must be consequences for that departmentrather than just the education school. Similarly, if the numbers of mathematics teachers produced by a campus goes down, rather than up, the mathematics department must also be held to account.)
- Institutions should also be held accountable for producing increased numbers of teachers in shortage fields and for increasing the number of their graduates who teach in hard-to-staff schools;
- Reporting of teachers on waivers/emergency credentials should be based on a consistent definition so the data is comparable across states and the definition should be aligned with NCLBs definition of highly qualified teachers. Teachers who are participating in an alternative route to certification should be separately identified and reported by number of years teaching.
Problem 6: Inadequate data systems.
Discussions of teacher quality are inadequately informed by data regarding what makes the biggest difference in student learning. As the U.S. General Accounting Office has reported, the U.S. Department of Education has granted almost half a billion dollars ($460 million) in HEA Title II funds but there is no consistent, reliable way to evaluate the effect of these grants on raising student achievement.
A. State Grants should be limited to states that are willing to establish data systems to evaluate the efficacy of teacher training programs and professional development activities on improving teacher effectiveness. Ultimately, states should be in a position to evaluate their success in terms of hard data indicating whether various activities helped raise student achievement. Effects on student achievement should be measured by state assessment data (and if that data would be insufficient for programs focused on high schools, then graduation rates and entry/success in college).
B. If they dont already have the capacity, Congress should allow states to apply for grants to connect data systems that provide analysis of individual teacher effectiveness based on student gains on academic assessments.
Indeed, this latter matterputting systems in place that will finally allow for the evaluation of teachers and the institutions that produce them at least in part upon ability to produce student learning gainslays at the very heart of the needed changes in our education reform strategy. Without good, solid data, virtually all of our effortsto know what matters, to analyze distribution, to evaluate either individuals or the institutions that produced themare seriously handicapped. Rather than being informed by hard evidence, we make decisions based on proxies and hunches. If we have one highest priority in this reauthorization, it has to be to tackle that problem head on.