Ed Trust Press Room

FOR IMMEDIATE RELEASE

February 15, 2007

CONTACT:

Claire Campbell (202) 293-1217, ext. 351

Kati Haycock’s Testimony before the House Appropriations Committee,  Subcommittee on Labor, HHS, and Education

(Washington, D.C.) - Mr. Chairman and Members of the Committee, thank you for allowing me to testify this morning.

I have been asked to testify about achievement and demographic patterns in elementary and secondary education, as well as about educational spending and program effectiveness.  I will try to provide the most succinct overview that I can.

Overall achievement

The National Assessment of Educational Progress (NAEP), also known as the ‘Nation’s Report Card,’ gives us a detailed look at student achievement in the United States.  There are two primary administrations of the NAEP:

• The Long-Term Trends Assessment has been administered periodically since the early 1970s.  It provides long-term achievement trends for the nation as a whole.
• The Main Assessment has been administered since the early nineties.  It provides short-term trends and an examination of the knowledge and skills demonstrated by students nationwide and in each state.

While it’s impossible to pull a single story from all of the NAEP results, we can summarize national achievement patterns like this:  there have been improvements in our elementary schools, with bigger gains in math than reading.  As our students get older, however, achievement has stagnated in some areas and actually declined in others.  So while we’re making progress in educating our youngest students, we are actually falling behind in preparing the students who are closest to entering the world of college and work.

The best news comes in fourth-grade mathematics, where scores have increased significantly over the past fifteen years.  We’ve made particularly good progress in reducing the number of fourth-grade students performing at the lowest levels in math.  The percentage of elementary students not demonstrating even basic math achievement has decreased from more than half (52 percent) 1990 to about one in five (21 percent) today. 

Fourth-grade reading scores were flat over the 1990s, but have increased over the past five years.   While modest, there has been an increase in the percentage of students demonstrating at least basic reading skills, from 57 percent in 2000 to 62 percent in 2005. 

Eighth-grade math scores have also increased over the last 15 years, and here, too, we’ve increased the percentage of students demonstrating at least basic skills, from 51 percent in 1990 to 68 percent in 2005.

In eighth-grade reading, however, scores have remained flat over the past eight years, and there has been no significant change in the percentage of students demonstrating at least basic skills.

Among our 17-year-olds, math scores have been flat over the past 15 years.  In reading, scores have actually gone down during this timeframe.

Underneath the Averages

Those, of course, are the averages, which mask very different levels of performance for different groups of young Americans, with affluent, White and Asian students typically performing above these averages and low-income, African American, Latino and American Indian students typically performing below average.

What do we know about group performance over time?  Though the patterns are somewhat different in different subjects and at different grade levels, we generally saw substantial improvements in the achievement of underperforming minority groups during the seventies and eighties and a substantial narrowing of the gap between those students and others.  During the nineties, however, that progress mostly leveled off, and gaps actually widened.

The good news is that this pattern has turned around, at least at the elementary level.  In both reading and math, we’re now seeing record performance for all groups of students.  Between 2000 and 2005, growth was stronger for African American and Latino 4^th graders than for other groups in both reading and math, and the gaps separating them from other young Americans are smaller than at any time in our history.  This pattern of increased achievement and gap narrowing in the elementary schools is confirmed by results on states’ own assessments.

As with the overall scores, the story for student groups becomes more troubling in the later grades.  In eighth grade, progress for minorities has been somewhat stronger than for whites, with some gap small gap narrowing.  Gaps at the 12^th grade level, however, remain wide and unchanged.  Indeed, by the end of 12^th grade, Latino and African American students are performing about the same as white 8^th graders in both reading and math.

A sobering note, though: Because the NAEP Long-Term Trends test was given in 1999 and 2004, it is impossible to look closely at rate of change in pre- and post-NCLB years. Main NAEP, however, was administered several times during this period.  The latter test results suggest that, while progress in the first two years of the 21^st Century continued after NCLB was passed, there was some slowing in the rate.  Scores are still up, but unless we pick up the pace of improvement, we will remain far from where we need to be.

High School Completion

The numbers for African American and Latino 12^th graders would be frightening enough if I didn’t have to remind you that those were the students who remained in high school through 12^th grade to be assessed.  Here, though our various data systems nationally produce somewhat different estimates of school completion rates, the overall patterns are clear:  disproportionately few student of color earn a diploma within four years of entering high school.  The best national estimates put on-time graduation rates at about 70 percent overall; among African Americans the on-time rate is 52 percent, among Latinos 56 percent.

These numbers should not be interpreted as final, because significant numbers of these students eventually earn either a diploma or a GED.  Unfortunately, however, longitudinal data from the U.S. Department of Education show that outcomes for these late graduates and GED recipients are not comparable to outcomes for on-time diploma recipients.  Almost half (46 percent) of on-time high school graduates go on to earn some type of postsecondary credential, while just 16 percent of late diploma recipients and 8 percent of GED recipients do so.

We have long known that not earning a postsecondary credential lowers an individual’s lifetime earnings.  New analyses of census data show that low levels of postsecondary attainment will, in addition, hurt national economic growth.  It is projected that by 2012, we will have a surplus of almost 3 million individuals without a high school diploma in the workforce and a shortage of over 3 million individuals with BAs. 

To meet the demands of the workforce, we need to significantly increase the number of young people earning Bachelors’ degrees.   Demographic trends show that these BA recipients will have to come from our ‘minority’, and especially Latino, populations.

Changing Demographics

Understanding achievement and attainment patterns for different groups of students is especially important given the fast changing demographics of our schools—and, indeed, of our country.  Between 1993 and 2003, the percent of White students enrolled in public elementary and secondary schools declined from 66 to 59 percent while the number of Latino students increased from 13 to 19 percent and the percent of Asian and African-American students remained essentially the same.

Between 2000 and 2020, the Census Bureau projects that the national population will grow by about 54 million people, almost two-thirds of them Hispanic or Black.  Much of this population growth is being driven by immigration, which is occurring at rates that we have not seen since the early 1900’s. 

The impact of this new immigration is truly national in scope.  Before 1995, more than three-fourths of the immigrant population was concentrated in 6 states.  Since then, many other states have seen substantial growth in their immigrant populations, with 22 states having greater than 100% growth.

This growth in the Latino population is mirrored in our public school enrollment.  Between 1994-95 and 2004-05, Latino enrollment in our schools grew by almost 3,500,000, representing a 60% increase.  Over the same time period, the enrollment of English Language Learners (ELLs) grew by 1.8 million, representing a 56 percent increase.

Our low-income student population also is growing.  Between 2000 and 2004, the percentage of students eligible for the free and reduced lunch program grew from 34% to 37%.

The bottom line, frankly, is very clear:  if we don’t turn around the performance patterns of the groups of students whom we have traditionally thought of as “minorities”—racial minorities, English language learners, and children from low-income families—our performance as a country will decline.  Why?  Because these groups, together, are fast forming a new “majority.”

International Comparisons

The patterns described above —including both the pattern of more growth in the elementary grades than in secondary education and the pattern of underperformance among significant groups of young Americans—are at least part of the reason why the United States lags on many international comparisons of student achievement, especially those focused on math and science and conducted near the end of formal compulsory education. 

There are three major sources of international math and science assessment data.  These include:

• PIRLS, the Programme in International Reading Literacy Study (4^th grade)
• TIMSS, the Trends in International Math and Science Study (4^th, 8^th and 12^th grade); and,
• PISA, the Programme for International Student Assessment (15 year olds).

The U.S. generally does best in reading.  In grade 4, for example, American students are among the highest performers, statistically exceeded only by England, the Netherlands, and Sweden (PIRLS).  By grade 10, however, that performance dropped to about average among developed countries in the last PISA focused directly on literacy (PISA 2000)—and there are hints in the sampling done for the 2003 PISA that our relative performance in reading might be declining.

In mathematics, our fourth-graders perform just below the average for the twelve industrialized nations that participated in both the TIMMS and PISA.  Our international ranking drops when we analyze both eighth- and tenth-grade results.

These data make clear that there are problems at both ends of the scale.  The U.S. has proportionally fewer students who reach the highest level of mathematics achievement than many other developed countries, and proportionally more students at the lowest level of math achievement.  And both our high-SES and low-SES students perform toward the bottom of the pack when compared to international students of similar socioeconomic background.  (PISA 2003.)

In fourth-grade science, our scores have declined and we have lost ground among the fifteen countries that participated in TIMMS in both 1995 and 2003.  In eighth grade science, our scores improved as did our standing among the seventeen countries that participated in both 1999 and 2003.  Our tenth-graders, however, perform below the international average in science literacy—19^th among 29 participating OECD countries, significantly lower than 15 of those countries (PISA).

Not only are we behind other industrialized nations in academic achievement, we are also losing ground in terms of academic attainment.  While we once ranked first among industrialized nations (according to OECD data) in high school graduation, we are now 17^th out of 24.  There is a similar story in higher education, where the United States’ historical lead has greatly diminished. While the United States currently ranks second (to Norway) in the percent of the population with a B.A., we are one of only two countries where our older generation is more likely to have BA than our younger generation.

While the rest of the world is learning from our historic emphasis on education and implementing aggressive and effective improvement strategies, the United States is making only modest gains in some areas and stagnant in others.

Educational Equity

The gaps in achievement that separate low-income and minority students from other young Americans begin before children arrive at school, and it is well established that poverty creates obstacles to success in school. But even if poverty had no impact on their achievement at schools, it is shameful that so many children in this wealthy nation are growing up in poverty. We must all work relentlessly to improve the lives of these children and their families.

As we work to redress the burdens and barriers that poverty creates for these children, however, we can not for one minute use it to excuse the educational malpractice that occurs day in and day out in their schools. The question that educators and policy makers must face NOW is not whether poverty affects student achievement, but what public education can do to help students rise above their outside-of-school circumstances.

First, we can and should do more to ensure low-income children are ready for school. Some intensive parent/child education programs targeting very young children in poverty have shown strong results for children. Yet we only make these services available to a relative handful of the families who could benefit from them.

An even more compelling research base supports early childhood education. In study after study, high quality pre-k programs, for example, have shown enormous power to improve school readiness among low-income children who participate.  Still, however, only some low-income children receive any pre-k education at all, and much of what they do receive is low-quality. Children from high-income families are more than twice as likely as students from the lowest-income families to attend pre-school, in part because of very limited public investment in relation to the needs of low-income children.

As troubling as our neglect of pre-school opportunities, however, is the fact that our schools are not organized to respond to the thousands of children who, every year, enter school behind.  Rather than ameliorating the problems that poverty poses, we’ve actually organized our school systems to exacerbate these problems. 

How do we do that?  By taking the kids who arrive with less and then giving them less in school, too.  In fact, we give these children less of everything that both research and experience tells us makes a difference.

Some of the “lesses” flow from choices that policymakers make.  Included here is the choice that the majority of states have made to actually spend less in school districts serving concentrations of low-income students and students of color than they do on districts serving more affluent and white students.  Nationally, those differences really add up.  We’re spending, on average, $825 less per-pupil, per-year in the quarter of school districts with the most poor children than we do in the quarter of districts with the fewest such children—and  $908 less per pupil in the quarter of school districts with the most minority children than we do in the quarter of districts with the fewest.

Congress has some culpability here, as well.  Title 1 dollars flow disproportionately to high spending states, exacerbating pre-existing differences between high- and low-wealth states. The state expenditure factor in Title I tilts Title I dollars toward those states that are most able to pay for the education of their students and away from those states—the poorer states—that need the most help. Elimination of the state expenditure factor (along with the addition of an adjustment for geographically based cost differences) would help level the playing field between wealthy and poor states -- and would bring the Title I formula into line with the formulas used for special education, English language instruction and child nutrition programs.

While it is important that policymakers make funding more fair and equitable, it would be a mistake to believe that more or fairer funding alone would fix all that ails our public schools. Many of the most devastating “lesses” in the education of poor and minority students flow not from the choices that policymakers make, but, rather, from the choices that educators make. Choices educators make about what to expect of whom.  Choices educators make about what to teach to whom.  And perhaps the most devastating choice of all:  the choice of who teaches whom. The result of that choice is a disproportionately large proportion  of our strongest teachers teaching our most affluent children and a disproportionately large proportion of inexperienced and poorly educated teachers teaching our most vulnerable children.

The results of these lesses are crippling for our low-income and minority students, their communities and our society as a whole.

What works?

We know about the corrosive effects of poverty on children, but we also know that schools that are squarely focused on instruction and that are relentless in their focus on growing the knowledge and skills of their students can enable even the poorest students to achieve at high levels. So what are they doing that works?

It turns out that most of what works to help low-income students learn is more common sense than magic. Throughout their education, low-income and minority students need what our most affluent children get—rigorous curricula, taught by expert teachers, in adequately equipped buildings—and more.  Real-time diagnosis of student needs and highly focused extra instruction for children who are behind—like that provided by the KIPP schools—really helps as well.

Teachers: Highly effective teachers are the most important ingredient of all.  Some teachers produce very large learning growth in their students every year, while others produce almost none.  Low-income and minority children—who currently get less than their fair share of the first kind--need far more. Low-income students, with so little to fall back on at home, simply can not afford teachers of limited knowledge or ability. Ensuring that the students who need the most effective teachers actually get them is probably the single most important element of a gap closing agenda. This means changing the incentives about where to teach and rewarding richly those teachers who choose to work—and are successful in—the most challenging schools. This will require a substantial investment for salary incentives, as well as other investments in these schools.

Extra Time to Learn:  Children who start behind need time to catch up. That can mean longer school days and more days of school. BUT that time must be used wisely and well: more bad instruction is the last thing these students need. The current approach to remediate students who are falling behind is to slow them down, but I ask you to consider this: In what other endeavor would we slow someone down to help them catch up? Students who are behind need intensive, accelerated instruction.

Another way to win extra time for children who need it is to start sooner.  All of our low-income 3- and 4-year olds should have access to high quality pre-kindergarten programs that have the explicit goal of preparing children to do well in school and are lead by teachers with baccalaureate degrees. 

Supports for Struggling Schools: It doesn’t help much to hammer on educators to improve if you can’t also give them a hand.  But too few of our teachers get the curricular support and coaching they need, and the same is true of schools.  Recent research suggests that the more explicit the intervention, and the more careful its implementation, the more likely we are to improve results. We’ve as yet done precious little to ensure that schools get the right sort of help. 

Roles for Appropriators

While ensuring that appropriations levels for NCLB and other critical K-12 education programs match their authorization levels should be among our priorities, this cannot be our only -- or even our highest -- priority.

If Title I were “fully funded” it would increase the federal government’s share of the nation’s total education investment to approximately 12 cents of every dollar spent on education. And while 12 cents is better than 8 or 9 cents, federal spending alone cannot raise overall achievement or close the achievement gap.  Given the federal government’s relatively limited investment in education, the federal priority should be the design of an investment strategy that:

• maintains the federal government’s historic and appropriate focus on disadvantaged students

• Demands information and transparency in exchange for federal funds;

• Leverages systemic change in the allocation of state and local education resources such that these resources are more effectively used and more equitably distributed;

• Recognizes the role of schools, school districts, states and intuitions of higher education as incubators of innovations in policy and practice and provides the resources needed to evaluate the relative effectiveness of such innovations; and 

•  Enhances the federal government’s role as one of identifying success and supporting—even insisting on as a condition of funding-- its replication.

Information and transparency

A policy of strategic decision making demands good information. And while NCLB has, through its assessment requirements, produced more information than we have ever had on student achievement, the fact remains that educators, policymakers, and taxpayers still cannot use this information to inform decision making about education investments. We need to take at least two steps in order to help facilitate information-based decision making at the state and federal levels:

• Create robust state-level data systems that include information of student achievement and teacher effectiveness which are linked to each other and to information on higher education and labor market outcomes; and,
• Increase the transparency of information about education. Student achievement data, data on school effectiveness in growing student learning, and information about the use and distribution of education resources need to be made publicly available annually.

Despite the growing demand for “growth models” for NCLB accountability purposes, the truth is that only a handful of states now have the data capacity to support such systems. The federal government must—through a matching grant program—establish the data infrastructure to allow states to build and operate growth models. The administration’s request for $ 50 million to extend a competitive grant program for the creation of such data systems is wholly inadequate. The pot needs to be richer and the demands on states higher. Establishing such a data system, and ensuring the personnel to maintain and use the data, should be a prerequisite for continued participation in Title I.  Further, regular publication and distribution key data elements and analysis produced by such data systems should be required.

Along with greater transparency on school and teacher effectiveness, Congress should require states and districts to make clear how education funds are used and distributed. There is a growing body of research documenting how education funds intended for low-income students and other disadvantaged groups are siphoned off to boost general revenues, but this practice usually escapes scrutiny because of opaque and out-dated budgeting practices that ignore spending differences between schools.

Leveling the Playing Field: Using Federal Dollars to Leverage Funding Equity at the State, District and School Level:

I have already mentioned that, even if “fully funded,” Title I would still represent a relatively small share of public education funding. And, while Title I could be better targeted to low-wealth states, the truth is that we cannot achieve funding equity without ensuring equity in state and local funding decisions. At the state level, Congress could do more to encourage states to reduce the gaps between their high and low-wealth school districts by expanding the Education Finance Incentive Grant program.

Most important, the federal government could leverage enormous change in the inequitable funding patterns within school districts by amending the “comparability” provisions of Title I. The intent of these provisions is to ensure that Title I funds are added on top of an equitable funding base within each district receiving funds. But the manner in which the law requires the demonstration of equity actually works to mask vast school-to-school inequities that in some case amount to hundreds of thousands of dollars per school.

The nub of the problem is the law’s allowance for the exclusion of teacher salaries in the comparability calculation. Teacher migration patterns, through which more experienced (and higher paid) teachers leave high-poverty schools—taking their higher pay checks with them—shortchange students in high poverty schools not only in terms of access to stronger teachers but also in funding. This practice also provides an unwarranted subsidy to more affluent schools. Requiring school districts to account for the actual amount of money spent in different schools would make the comparability provisions more effective.

R & D

We know more than we ever have about what works-- and what doesn’t work—to boost student achievement. And while it is certainly true that we are not yet applying effectively all that we know and that doing so would reap rich rewards for students, teachers, and schools, there is still a lot that we don’t know. Investing in increasing our knowledge in this area is entirely appropriate for the federal government.  We need to increase our investments, in particular, in the design of better curriculum and assessments; in engineering and testing school designs and school reform models; in strategies for accelerating the learning of children who arrive behind; and in instructional models for enhancing learning among English Language Learners and students with disabilities.   

Along with research and development, we need to spend more, not less, evaluating the success of investments that the federal government is already making in education. While adding strong evaluation components to our education programs will certainly—in the short term—add to their cost, it will in the long run save funds, so that we stop investing in ineffective strategies that waste taxpayer money and the value time of educators and students.

For example, since NCLB’s passage we’ve invested nearly $15 billion in Title II funds. I would not for a minute say that we should invest even a penny less in raising teacher quality. However, after five years we have little information on what the most—and least—effective uses of those funds have been. In part that’s because we haven’t have the data systems needed to evaluate them, but its also because the authorizing legislation required little reporting or evaluation on the use of the funds.  

The uses of Supplemental Services dollars should also be carefully evaluated.  Here, perhaps more than any place else, we made a big federal investment in an unproven strategy.  Congress deserves to know how those dollars are being used, who is being served and, most of all, whether those dollars are helping improve achievement among participating students.

At the other end of the spectrum is Reading First, a program that was based on a strong record of evidence, that is showing promising results, but has had significant problems in its administration. We cannot afford to have personal or political biases infect the implementation of education programs, and the misdeeds that have been exposed regarding Reading First have undermined the program’s credibility and hampered its effectiveness. But educators and researchers alike credit Reading First with taking advantage of significant consensus on the science of how children learn to read to improve reading instruction, especially for students who start school with weak reading skills. We need to build on this model, and build in safeguards for competent administration and aggressive oversight. But, please, do not throw out the baby with the bathwater – we need more federal funds to be invested in evidence-based instructional improvement practices, not less.

Conclusion

Yes, we need better administration and oversight of some federal programs. And yes, we need to comb the data and evaluate the results of various innovations to refine our approach to gap closing. And yes, poverty does pose challenges.

All of that said, we already know a lot about what it takes to get low-income students to high levels of achievement—strong teachers; rich, coherent curriculum; high expectations; extra learning time and carefully crafted assistance to their teachers and their schools. We also know that there are pervasive, significant inequities in our school systems that make it impossible to get these things in place and that effectively stack the deck against certain students.

Over the last seven years, we’ve begun to turn the corner.  All across the country, schools are more focused than ever on student achievement.  Reading and math skills are rising among our youngest children.  And there are no more hidden kids.

But we face increasing challenges in our classrooms, and increasing competition from abroad.  Now is not the time to back down from those challenges or, as some have suggested, to ratchet down the goals.  Rather we need to ramp up both what we ask of schools and the amount and quality of the help we give them.  The decisions that this sub-committee makes are absolutely central to assuring that those who need that help—both students and schools—actually get it.

###