By Michal Smith-Mello, Michael T. Childress, Amy Watts, Mark Schirmer, and Billie S. Dunavent(*)
From Foresight, No. 51
published 2008
In 1983 the latest in a series of presidential commissions on education issued a landmark report to President Ronald Reagan. Unlike its predecessors, A Nation at Risk sounded a dire warning: the postwar prosperity we had presumed was ours to keep was in danger of being lost if “the rising tide of mediocrity” produced by our educational system were not stemmed and reversed. Other nations, this body of educators and policymakers agreed, were matching or surpassing our educational attainments.(1) Gradually, states responded with a flurry of initiatives aimed at reforming our school systems. But, in spite of billions of dollars spent in the pursuit of improved academic performance, national and international testing suggest that we remain awash in mediocrity. Our nation remains at risk, perhaps more so than ever.
Dramatic advances in global communications and rising education levels around the globe present new competitive realities. The promise of electronic communications many saw for strapped rural U.S. economies is being seized by entrepreneurs in underdeveloped nations where U.S. services are migrating in vast numbers. In the absence of structural change in our educational system, some argue, the forces of the global economy will depress wages at all levels and ultimately erode our national standard of living.(2) Others contend that wages have already been depressed by the exportation of many of the better-paying U.S. jobs.
Here in Kentucky, we have achieved laudable educational progress, moving out of the bottom tier of states during an era when virtually every state was bringing renewed attention to education. Nevertheless, we still lag the national average on many measures, and the national average is not one of international distinction. Relative to other nations, academic performance in the United States grows poorer the longer students remain in our schools. The failure to prepare Kentucky’s children for what lies ahead jeopardizes nothing less than our future. Hence, ensuring that our children achieve broadly shared, high-level academic skills is critical to Kentucky’s social and economic progress. Moreover, the enormity of the potential consequences demands that we leverage the highest possible returns from the sizeable public investment we are making in education, and, if need be, invest more.
Here, we briefly explore the context of “the flat world”(3) in which we now live, where rising educational achievement and global communications are creating opportunities for highly skilled workers at a fraction of the cost of U.S. workers. In light of this, we assess the academic progress of U.S. students relative to our international peers and Kentucky students relative to the nation. Finally, we look at some of the strategies that research suggests will yield the highest academic returns on public investment in education and thus offer us the best hope for broadening and sustaining prosperity.
Once a cornerstone of small town economies, the sewing factory and its ilk have all but disappeared from Kentucky’s rural landscape. Most of these operations and the low-skill jobs they offered have been moved “offshore,” to countries where labor costs are a mere fraction of those in the poorest regions of our state and nation. Over the course of a decade, the apparel industry went from being Kentucky’s leading manufacturer of nondurable goods to being a virtual footnote. Apparel industry employees plummeted from 32,200 in 1990 to 12,800 in 2000. By January 2008, just 3,600 Kentuckians worked in the industry.(4) As a result, rural communities in particular weathered substantial job losses, as well as the broad economic impact of shuttered factories.
But the global market for labor has brought the good with the bad. Foreign-direct investment, which creates higher-paying jobs, has increased markedly in Kentucky. Lower offshore manufacturing costs have also led to lower prices for many consumer goods, including essentials like clothing.
Because a factory closing grabs headlines and wrenches family emotions and community economies, the image of a blue collar worker locked out of the factory where he or she has spent their adult work lives is, for many, the personification of the vagaries of the global economy. The world, however, has changed. The firebreak separating low-skill, low-wage jobs from higher-skill service and professional jobs clearly has been breached. As education and skill levels rise in the developing world, businesses are tapping into a vast pool of highly skilled, lower-cost talent. Engineers, accountants, architects, radiologists, and computer programmers from India, China, and the Philippines, for example, now perform these jobs for U.S. entities for a fraction of the cost of their American counterparts. Estimates of the movement of these jobs offshore are, at best, sketchy since the government keeps no official count, but Forrester Research predicts that roughly 3 million high-tech jobs will be relocated overseas by 2015.(5)
At the same time, technology continues to relentlessly eliminate jobs. Marc Tucker, one of the principal authors of Tough Choices or Tough Times, estimates that for every job that is being “offshored,” ten are being automated. In short, jobs that involve “routine” tasks are simply disappearing from our economy.(6) Though the United States continues to rank as the most competitive nation among the 133 countries examined by the World Economic Forum (WEF) for its 2007-2008 report, labor market efficiency or the low cost of eliminating jobs figures prominently in the analysis. As Time notes, the report focuses on productivity without attention to the collateral damage.(7)
By the WEF’s assessment, the U.S. economy remains innovation-driven.(8) In this arena, the so-called “conceptual economy,”(9) ideas and innovations create high-cost products and services that, in turn, provide a foundation for expanded and enduring economic opportunity. Its health ultimately will determine whether U.S. workers enjoy some level of economic security.
Arguably, the diversity and freedom of American culture is the ultimate Petri dish for innovation, but even brilliant ideas can languish in the absence of the skills needed to bring them to fruition. The richer our understanding of the underlying mathematical and scientific concepts, the more likely our economy will continue to give rise to invention and innovation. As the 19th century French chemist and microbiologist Louis Pasteur observed, “Chance favors the prepared mind.”
Thus, it is widely acknowledged that more U.S. students need to pursue careers involving the STEM skills—science, technology, engineering, and mathematics. But, according to the General Accounting Office (GAO), the percentage of students graduating with STEM degrees decreased from about 32 percent of total graduates in 1994-95 to 27 percent in 2003-04, even as the raw number of these degrees rose.(10) A National Science Foundation survey found that only 16.8 percent of first university degrees in the United States were in the STEM fields compared to 33.3 percent for Asia, 64 percent for Japan, 52.1 percent for China, and 40.6 percent for South Korea (see Table 1). As a consequence, the nation’s supply of the highest-paying job skills, those with the greatest potential for yielding broad-based economic benefits, is surprisingly weak. As shown in Table 1, the United States trails major regions of the world, as well as leading competitor countries.(11)
Clearly, educational attainment, the development of STEM skills it enables, and the benefits it extends to our economy will depend on the educational foundation formed in our elementary and secondary schools.
International standardized tests not only offer a basis of comparison with peer students around the world, they are also indicative of the efficacy of the educational systems of which these students are products as well as the likely influence of cultural norms.
The United States participates in several international assessments of student achievement that permit comparisons with other nations. Prominent among them are the Trends in International Mathematics and Science Study (TIMSS) and the Program for International Student Assessment (PISA). The TIMSS, which assesses math and science achievement at grade levels that have the highest number of 9- and 13-year-olds, the equivalent of U.S. 4th and 8th graders, has been conducted every four years by the International Association for the Evaluation of Educational Achievement, an organization of research institutions and governmental agencies, since 1995.(12) The more recently implemented Program for International Student Assessment (PISA) is conducted by the Organization for Economic Cooperation and Development (OECD), an intergovernmental organization of 30 industrialized nations. PISA tests 15-year-olds on reading, mathematics, and science literacy, for mastery of curriculum and the ability to use these skills in adult life. Problem-solving skills are also tested.(13) Results from 2007 testing will be released December 9, 2008.(14)
Here, we examine mathematics and science testing results from the 2003 TIMSS, which was conducted in 46 countries, and the 2006 PISA, which was conducted in 57 jurisdictions and countries, including 30 OECD members.(15) U.S. students do not exhibit international leadership on any of these tests, but by high school, U.S. students fall behind most of their peers in industrialized as well as what many still regard as developing nations.
The Longer They’re in School … The academic performance of U.S. students on international testing mirrors outcomes on national testing. As U.S. students move through the grades, their performance lags. “We seem to be the only country in the world whose children fall farther behind the longer they stay in school,” ERIC Digest editorialized on the occasion of the 1998 publication of A Nation Still at Risk.(16) By the time U.S. students reach early adolescence and approach the critical transition to higher education or entry to a demanding labor market, academic performance declines considerably. For example, the difference between the OECD median and the U.S. score in science moved from a 15-point advantage for U.S. 4th graders to a 12-point deficit by the 10th grade in 2003. Figure 1 shows the decline in mathematics and science as students advance. In mathematics in 2003, the point differential was marked: U.S. 4th graders scored at the OECD median while 10th graders scored 24 points below it.
Figure 1: Declining Performance of U.S. Students Relative to OECD Countries, 2003
Tables 2 and 3 illustrate international comparisons of average student scores on the TIMSS (4th- and 8th-grade students) on science and mathematics in 2003, the most recent year for which these data are available, and Table 4 shows the 2006 PISA test results for 15-year-olds. In the critically important subject of science, elementary and middle school students in U.S. public and private schools perform well on the TIMSS, but by the time they reach high school, U.S. students lose ground, falling well short of their international peers. Here we show the most recent results from both of these international tests, 2003 for the TIMSS and 2006 for the PISA, illustrating participating nations and average student scores for each. Scores within the middle range as illustrated in Tables 2 and 3 are not measurably or statistically different from one another.
Table 2: International TIMSS Results, 2003 (Fourth Grade)
Table 3: International TIMSS Results, 2003 (Eighth Grade)
Table 4: International PISA Results, 2006
As shown in Table 2, U.S. 4th graders scored, on average, well ahead of students in the other 24 participating countries in science; students in just three countries measurably exceeded the performance of U.S. 4th graders. Fourth-grade performance on mathematics, however, falls more closely to the median, with students in 11 other countries scoring, on average, measurably higher than U.S. students, whose scores exceed their peers in 13 countries. When compared to the other 10 participating OECD countries (Australia, Belgium-Flemish, England, Hungary, Italy, Japan, the Netherlands, New Zealand, Norway, and Scotland), U.S. 4th graders outperformed their peers in 7 countries on science and 5 on mathematics.
Students in 44 countries and the United States participated in the 2003 8th-grade TIMSS testing in science and mathematics (see Table 3). U.S. students continued to excel relative to their international peers in science, with only seven countries registering measurably higher scores. The average score of U.S. 8th-grade students was higher than that of peer students in 32 other countries. In mathematics, 8th-grade performance remained relatively strong, but U.S. students were outperformed by their peers in nine other countries, five of which are OECD members. Among the other 12 participating OECD countries, U.S. 8th graders had higher average scores than their peers in just two countries, Italy and Norway.
By the time U.S. students reach age 15, they lose significant ground when measured against their international peers. Here we illustrate the most recent 2006 test results, which found no measurable gains from 2003. As results from the 2006 PISA show (see Table 4), students in 16 of 30 OECD nations, as well as 6 other international jurisdictions, outperformed their U.S. peers in science. Students in 31 jurisdictions, 23 of which are OECD countries, had higher mathematics literacy scores than U.S. students. The average 2006 science literacy score for U.S. students, like that of their OECD peers, remained virtually unchanged from 2003 scores. Likewise, no measurable change in either the U.S. score or position relative to the OECD average in math literacy occurred between 2003 and 2006.(17)
While not illustrated here, the 2003 PISA assessment of problem-solving skills is also noteworthy. U.S. students significantly lagged their international peers as well as those in OECD nations. The differential between U.S. and OECD students alone was a disturbing -30. U.S. students scored lower than their peers in 25 out of 38 participating countries.
In general, the performance of the typical Kentucky student on national testing has steadily improved in each subject area. However, in spite of these gains, the academic performance of Kentucky students merely parallels and at times lags the national average, an unenviable benchmark, international testing suggests.
The Nation’s Report Card. The National Assessment of Educational Progress (NAEP), oft referred to as the “Nation’s Report Card,” gauges student progress in mathematics, reading, science, the arts, civics, economics, geography, U.S. history, and, more recently, writing. NAEP data, similar to international data, reflect the results of testing at the 4th, 8th, and 12th grades. As shown in Table 5, the percentile ranking of Kentucky students who demonstrate proficiency or higher on the NAEP illustrates the challenge of achieving parity or, better yet, excelling nationally and internationally at a time of intensifying attention to and investment in education.
The bright spot for Kentucky students again is seen principally at the 4th grade where the percentage of students scoring at or above proficient levels on the NAEP tests ranked in the 81st percentile in science in 2005, a significant improvement over their 2000 ranking. After making a leap in their ranking between 1992 and 1998, 4th graders have lost ground in reading. In math, 4th graders made modest gains between 1992 and 2000, only to lose significant ground in subsequent testing years.
Eighth graders, likewise, made significant gains in science relative to their national peers between 1996 and 2000 but remained at the 51st percentile on 2005 testing. Relative performance on reading tests seesawed in recent years but fell to the 35th percentile in 2007. On math, 8th graders made gains on their U.S. peers only in 2000, then lost ground in subsequent testing years.
A recent American Institutes for Research study attempts to convert state-level, 8th-grade NAEP scores into a scale comparable to that of TIMSS testing in the key areas of math and science. Based on these comparisons of levels of proficiency, the same portion of Kentucky students are proficient as at the national level in both subjects but trail their Asian counterparts by wide margins in mathematics in particular. Compared to a high of 73 for Singapore, Kentucky students scored 27 in mathematics. The comparison is more favorable for science with Kentucky students registering just one point below the national average of 30, compared to a high of 55 for Singapore.(18)
CATS Testing. Kentucky began conducting its own testing soon after passage of the landmark 1990 Kentucky Educational Reform Act. In 1999, the Commonwealth Accountability Testing System (CATS) was implemented. Since that time, Kentucky’s students have shown steady progress at all levels; however, the same pattern of lost ground at the middle and high school levels is observed. As shown in Figure 2, Kentucky’s 4th graders bested older students by nearly 10 points on the total academic index in 2006.
Figure 2: CATS Total Academic Index Scores, 1999-2006
Importantly, the CATS tests students in more subject areas and more frequently than either international tests or the NAEP. Thus, students who clearly face significant challenges due to their economic status are being challenged to achieve higher levels of performance across a broader spectrum of subjects, including arts and humanities, social studies, language arts, practical living, and writing.
While many continue to regard Kentucky as ranking among the poorest educational performers in the nation, they are misinformed. Things have changed. National indices, as well as our own, show that the state’s overall profile is one of steady, measurable, and broadly recognized improvement. Nevertheless, just as U.S. students do not distinguish themselves internationally, Kentucky students, with the notable exception of 4th-grade science students, have yet to distinguish themselves nationally.
Overall, however, indicators of educational progress show across-the-board improvement here between 1992 and 2005. As shown in Table 6, educational attainment measures such as the percentage of the population with a high school diploma, a two- or four-year degree, and the high school dropout rate all steadily improved. Likewise, each measure of educational achievement, from ACT scores to the percentage of students scoring at a proficient or higher level on the NAEP exams, was at or near its highest level in 2005.
Table 6: Selected Education Indicators for Kentucky, Selected Years
We created an index of Kentucky’s educational progress over time relative to other states by combining available data into a single number.(19) Based on this index, Kentucky has generally progressed since the early to mid-1990s. Based on our calculations, Kentucky has improved its national education ranking from 43rd in 1992 to 34th in 2005, a finding consistent with Education Week’s Quality Counts 2007 Achievement Index, which also ranks Kentucky 34th, and the Morgan Quitno 2006-07 Smartest State Index, which ranks Kentucky 31st.(20) This improvement has been driven by 4th- and 8th-grade science scores, 4th-grade reading, and the steady decline in the dropout rate from 5.6 percent in 1996 to 3.3 percent in 2004.
Education Kentucky’s gains have come in spite of the considerable and broadly recognized liability of educating children who are at a profound economic disadvantage and lifting the educational status of a population that has historically been undereducated and disproportionately poor.
Since the 1990 passage of KERA, Kentucky has made important gains. Nevertheless, vast territory must be closed if academic achievement here is to rise to a level that will permit today’s youth to compete effectively in the future global workforce and, by doing so, lift prosperity in our state. Given the challenge that remains to be met, it will be necessary to maximize returns on the huge public investment we are making in education. We estimate that, in state funding alone, Kentucky spent nearly $50 billion (in constant 2006 dollars) from 1990 to 2006 on educating children, more than 44 percent of total general fund expenditures.(21) Thus, spending limited resources as efficiently as possible is not only vital to our future economic well-being but also to our ability to sustain other critical state government responsibilities. Here we consider the importance of education spending in general and state spending relative to the nation.
Money Matters. Quantity is seldom a measure of quality, but research clearly shows that resources matter to education. For example, a RAND study of student performance on components of the NAEP and the effects of state-level spending identified key areas that exerted the most influence on test scores. While higher levels of investment per pupil clearly yielded higher results, strategic investments may matter more, suggesting that the wise use of even limited resources can improve academic performance.(22) What’s more, RAND researchers conclude, increased investment in public education matters far more to less-advantaged students. While the effects of increased investment are negligible or nonexistent for more-advantaged students, they clearly affect minority and less-advantaged students, and the outcomes “can be large and significant if properly allocated and targeted.”(23)
Here and nationally, per pupil spending has risen steadily over the past 20 years. As shown in Figure 3, however, the gap between spending in Kentucky and the average for surrounding states and the nation has widened since 2000. While surrounding states have caught up with the national average in per pupil spending, Kentucky continues to lag behind.(24) Kentucky now spends just 82 percent of the U.S. average per pupil compared to 99 percent among surrounding states.
Figure 3: Per Pupil Spending, Kentucky, the U.S., and Surrounding States, 1983-2005
How Efficient Is Our Spending? In spite of significant barriers that strongly influence educational outcomes—high poverty rates, low educational attainment levels, and lagging investment—two separate indices suggest that the efficiency of Kentucky’s investments is high. In a 2004 study using 2001 data, the Manhattan Institute concluded that based on the relative “teachability” of students and state investment in education adjusted for cost of living, Kentucky ranked fourth in the nation for school efficiency.(25) Using similar criteria, the U.S. Chamber of Commerce’s Education Report Card gives Kentucky a B for its “solid return on investment,” which it deems “strong relative to state spending.”(26) Importantly, however, the Chamber’s Education Report Card gives Kentucky a “D” for academic achievement, and a “C” for the academic achievement of low-income and minority students, the rigor of its standards, and the quality of its teaching force.(27)
We calculate that Kentucky’s national ranking of 37th on NAEP proficiency is equivalent to a ranking of 25th on NAEP returns per dollar spent on students, and a ranking of 8th nationally when we control for factors such as poverty and the low level of educational attainment in the state. In short, our state has achieved measurable gains relative to our rather significant fiscal and cultural limitations.
However, the substantial gaps between academic performance here and at the national and surrounding state levels suggest that we can and must do much more if we are to make the kind of academic gains that will prepare today’s youth for a future of rising expectations. In some critical areas that research suggests help leverage higher returns, doing more will likely require spending more. In other areas, however, strategically focused investments could yield higher returns.
Increasingly, researchers insist that the yardstick by which education inputs must be measured is the academic gain realized from them. Beyond higher spending per pupil, research suggests that some investments in education are more likely to yield higher academic returns than others. In short, the strategic use of limited resources is key, particularly in a state such as ours which has considerable ground to close given limitations on resources and the enormous challenge of overcoming the legacy of enduring and disproportionate undereducation and poverty.
Here we discuss some of the spending strategies that research suggests will yield the highest returns. While it may have become a policymaking cliché, our findings, as well as those of national research, show that we can ill afford to continue leaving children behind. The costs and consequences of doing so, our own state economy attests, are incalculable.
Start at the Beginning. States with higher levels of achievement, RAND finds, also have more children in public prekindergarten programs,(28) underscoring the critical role played by early childhood education. From birth through age three, an estimated 90 percent of physical brain development occurs. Consequently, educators have long emphasized the importance of children coming to school “ready to learn,” that is, healthy in every sense of the word, from their physical and mental well-being to the age-appropriateness of their motor, social, and language development, their cognition and knowledge, and their approach to learning.(29) Deficits in any of these areas can frustrate learning and consign children to a life of unrealized potential before they ever arrive at the nearest elementary school door.
A significant body of research has focused on model preschool programs, tracking children’s performance as they advance through their school years, and, in some cases, following them into adulthood. From this research, a number of general conclusions have emerged that underscore the importance of high-quality preschool education. Though results vary from program to program, children who receive an excellent preschool education perform better academically in subsequent years, though some of these gains may fade over time. Besides being better prepared for school, upon reaching adulthood, individuals who participated in high-quality prekindergarten programs tend to be healthier, more economically self-sufficient, and less likely to engage in criminal activity than their cohorts.(30) Moreover, such gains prove even greater among children deemed by some criterion to be “at risk.”(31)
Importantly, research shows that young children from low-income families consistently perform more poorly on reading and math, even at the prekindergarten level, and that this achievement gap persists as children progress through school.(32) Thus, the importance of attending to the academic needs of our most vulnerable children at the earliest possible age may be tantamount to correcting the very course of their lives.
That said, Kentucky ranks highly among states for its preschool efforts which are free to at-risk children. However, in light of research about early childhood development, they appear to be too little too late. While the National Institute for Early Educational Research ranks Kentucky highly for access, 5th and 11th nationally for enrollment of three- and four-year-olds, respectively,(33) eligibility criteria for preschool enrollment here is limited. Participation in state-funded preschool programs is restricted to three-year-olds with a disability and four-year-olds with a disability or from low-income families.
On the measure of state spending per child in preschool, Kentucky’s ranking rose from 31st in the nation to 19th between 2006 and 2007, as funding increased from $52 million to $75 million.(34) Eligibility criteria were also expanded to include children from families with incomes up to 150 percent of the federal poverty rate, compared with 130 percent in 2006. Nevertheless, enrollment of income-eligible children remained virtually unchanged and actually declined when district- or tuition-funded participants are included. The latter fell by 1,569 or 51 percent. Because enrollment changed little, per pupil spending increased by nearly $1,000 between 2006 and 2007.(35)
Importantly, enrollment in state preschool programs (exclusive of special education programs and federally funded Head Start) remained unchanged between 2006 and 2007 at 11 percent of the state’s three-year-olds and 29 percent of its four-year-olds.(36)
Given the lifelong impact of early development and the need for nurturing, intellectually rich environments during early childhood, the American Association of Colleges of Teacher Education recommends programs for infants and toddlers be led by teachers with BAs.(37) Kentucky falls egregiously short of this standard, requiring only that child care workers be 18 or older, have no criminal record, and be free of tuberculosis. Moreover, assistants to state-run preschool teachers are not required to be credentialed, and facilities are not monitored.(38)
Not surprisingly, in a comparison of state child care standards that also included the District of Columbia and the Department of Defense, Kentucky ranked 50th out of 52.(39) Were the hallmark of quality child care merely the protection of children’s physical safety, Kentucky’s standards might be sufficient. However, early childhood experts call for moving beyond this “babysitter” model, emphasizing the necessity for child-focused educational activities. Kentucky’s shortcomings are numerous:
A six-year series of annual evaluations conducted by the Kentucky Preschool Program (KPP) found a consistent, significant gap between the chronological age and developmental age of children eligible for the program due to disability or low family income.(40) A gap also existed between eligible children and ineligible students, but follow-up testing indicated that eligible children who participated in KPP performed about as well in school as their ineligible cohorts and outperformed KPP-eligible children who did not participate in the program.(41) Unfortunately, no evaluations have been conducted since 1997, in spite of the growing focus on early childhood education.
Focus on Children at the Social and Economic Margins. Disadvantaged children arguably hold the key to Kentucky’s future. Their academic success will ultimately determine whether our state’s future remains one of disproportionate poverty or gives way to rising prosperity. An extensive and longstanding body of research clearly shows that economic disadvantage has a significant negative drag on academic performance.
Without doubt, the sheer breadth of the population of economically disadvantaged students in Kentucky adversely affects overall performance on both state and national tests. Kentucky is home to the nation’s ninth highest population of students who are eligible for free (44 percent) or reduced-price (9 percent) lunches, a reliable proxy for poverty and need. This significant disadvantage in head-to-head competition with students from far wealthier states makes the gains Kentucky has made all the more remarkable. In view of the circumstances of children’s lives in Kentucky, the state would be well served to become a laboratory for testing the efficacy of educational strategies designed to neutralize the effects of poverty, racial disparities, social isolation, and other factors that adversely affect academic performance.
The gap created by disadvantage is significant. For example, Figure 4 illustrates the performance gap on the Total Academic Index of the CATS between economically disadvantaged or high-needs elementary school students and their more advantaged peers. Since the adoption of the CATS in 1999, the gap had closed by 2.6 points in 2006. A 16.6 point differential remained. In 2007, the gap widened to 18.1 points under the new CATS test for which nonadjusted and, therefore, incomparable scores are shown. Figures 5 and 6 illustrate the persistence of this performance gap among middle school and high school students. Thus, our own state exams show that, in spite of concerted efforts to address inequities in spending and extend support to economically disadvantaged children, we have not done enough or enough of the right things.
Figure 4: Total Academic Index, Elementary Schools, by Free and Reduced Lunch Status, 1999-2007
Figure 5: Total Academic Index, Middle Schools, by Free and Reduced Lunch Status, 1999-07
Figure 6: Total Academic Index, High Schools, by Free and Reduced Lunch Status, 1999-07
The NAEP results are even more stark. As shown in Figure 7, the percentage of students scoring at or above proficiency is consistently and markedly lower for less-advantaged students in every subject area tested at the 4th and 8th grades. State-level data are unavailable for high school students.
Figure 7: Kentucky NAEP Results (2005 and 2007), by Free and Reduced-Lunch Eligibility
Were we to close the substantial academic gaps associated with inequities, Kentucky students would be performing at dramatically higher levels relative to their national peers. As shown in Table 7, Kentucky’s 4th and 8th graders would rank among the nation’s leaders in reading and science and move into the top tier in mathematics. In short, our goals for education would be nearly realized.
Misallocating Our Most Important Resource. Some of the causes of the performance gap that researchers point to are practices that, with notable exceptions, have become engrained in school systems nationwide. How we pay and reward teachers and where the best teachers are typically found have emerged as central issues. At present, we “reward” our veteran and, research suggests, most competent frontline educators by permitting them to choose from available slots based on seniority, rather than utilizing their talents where they are most needed. As a consequence, the most experienced and possibly the most qualified teachers generally gravitate to the best working environments, that is, the more affluent schools where performance is already high and need is relatively low. In public school system after system, research finds an illogical allocation of resources: teachers in high-needs schools are paid considerably less on average than those in low-needs schools,(42) presumably because they are less experienced.
In Kentucky, we find evidence of this disparity in the state’s two largest school districts, Jefferson and Fayette, the only districts sufficiently large to afford a comparison of schools within a given district. As shown in Figure 8, teacher salaries rise with the affluence of the student body in these districts. For all grades, we find a pay differential of $4,200 in Fayette County and $6,100 in Jefferson County between teachers in schools where the highest percentage of students are from families in the lowest income quartile and those in schools where most students come from families in the highest income quartile. Because we know that many high school students never apply for the free or reduced-cost lunch to which they are entitled, we also looked at elementary schools alone. The pay differential remained substantial. Teachers in elementary schools with high populations of poor children were paid, on average, $3,200 less in Fayette County and $5,500 less in Jefferson County.
Figure 8: Average Teacher Salaries in Kentucky’s Largest School Districts, by Poverty Quartile, 2006
Teacher Quality Critical to Academic Achievement. Intangible though their qualities may be, good teachers can literally help shape a child’s future, overcoming demographic factors such as family poverty or undereducation. For Kentucky, these very factors continue to frustrate efforts and undermine educational gains, making the quality of instruction every child receives critical. Consequently, it is imperative that ways of elevating teacher quality—from the path of entrance to the profession to the culture and system within which teachers must perform—be systematically identified and advanced.
Inarguably, teachers are key to rising levels of achievement. A landmark mid-1990s University of Tennessee (UT) study that grouped teachers by their longitudinal effects on student performance on standardized tests found teacher quality to be “the single most dominant factor affecting student academic gain.” Its effects were both additive and cumulative. Those 3rd- to 5th-grade mathematics students who received successive assignments to high-quality teachers enjoyed as much as a 50 percentile point difference in test scores. Students with like abilities could have vastly different outcomes in just two years as a consequence of consecutive assignments to poor teachers, but later assignments to higher-quality teachers did not compensate for the lost ground. Importantly, lower-achieving students “were the first to benefit, followed by average students and, lastly, students considerably above average.”(43) In short, high-quality teachers were found to make a dramatic difference in the academic performance of the children most in need of highly skilled instruction.
A growing body of more recent quality research, Haskins and Loeb observe, further refutes a longstanding assumption that demographics are too powerful to overcome. These studies show that good teachers can raise student performance, regardless of socioeconomic backgrounds.(44) In 2005, for example, Hanushek, Kain, O’Brien, and Rivkin reported findings from a study of gains in the scores of 4th through 8th graders in Dallas schools. Good teachers, they found, made a difference at all student ability levels.(45)
Characteristics of a High-Quality Teacher. Even as more research links high-quality teaching with measurable student achievement, few readily measurable teacher characteristics are associated with quality teaching, and the difference these characteristics appear to make is quite small.(46) While the question of what makes a good teacher remains largely unanswered, some teacher characteristics have been linked to higher student achievement.
As with most occupations, experience, particularly in the early years of teaching, appears to matter. In a December 2007 report to a state task force, researchers from Washington State’s Institute for Public Policy released findings from an analysis of 15 studies that found large gains in student achievement on standardized mathematics and reading tests as teacher experience increased from one to five years. Student gains continued after the first five years but were more modest.(47) The Dallas study drew a more narrow conclusion: “The largest impact is the first year of experience, and experience effects disappear quickly after the first year.”(48) However, recent research from the Urban Institute concludes that “close to half of the achievement returns to experience arise during the first few years of teaching but returns continue to rise throughout most of the experience range.”(49) In the same vein, a 2006 RAND analysis of NAEP results in Tennessee found that a higher portion of inexperienced teachers was associated with lower achievement.(50) However, the Urban Institute’s Goldhaber noted in a 2003 review of literature that more senior teachers may in fact opt to teach higher-level classes and higher-performing students. If so, “the magnitude of the experience effect, should it exist, is not terribly large.”(51)
In addition to seniority, Kentucky teachers are paid for gains they make in pursuit of a master’s degree, regardless of the institution, the content of the degree program, or its relationship to subjects taught. But research in this area is, at best, inconclusive. However, little evidence can be found of positive academic outcomes that are linked to teachers with advanced degrees. The recent Washington state meta-analysis, for example, concludes that graduate degrees have little or no effect on student achievement.(52) While some evidence suggests that advanced degrees in mathematics and science are associated with higher student achievement, no such link was found in other subject areas.(53) A 2007 study by Urban Institute researchers Clotfelter, Ladd, and Vigdor utilized a detailed longitudinal data set that permitted them to link student achievement to specific teachers in North Carolina. They found that a master’s degree obtained after five years of teaching was “associated with negative effects on student achievement.”(54)
On the whole, the conclusions of researchers do little to recommend Kentucky’s current teacher pay structure, which rewards tenure and increments of progress towards a required master’s degree. The latter policy has resulted in a highly educated teaching force. As shown in Figure 9, 70.6 percent of Kentucky teachers held master’s degrees or higher in 2003-04 compared with 48.1 percent nationally. Only five states––Connecticut, Indiana, Massachusetts, New York, and West Virginia––had higher percentages of teachers with master’s degrees, according to the National Center for Education Statistics. Scant evidence, however, suggests that this accomplishment and the considerable investment it requires of teachers yield returns in the classroom. Hence, careful analysis of the relationship of graduate-level teacher education to student achievement is needed, as well as analysis of the costs teachers incur for education relative to the benefits to students. The ratio of costs to benefits may become an increasingly important consideration as we confront the likelihood of teacher shortages in the coming years. To that end, it is important to learn what, if any, role these professional requirements and their attendant costs play in discouraging otherwise strong candidates from entering the profession.
Figure 9: Highest Degree Earned by Teachers, Kentucky and U.S., 2003-2004
Finally, some researchers now suggest that certification, the cornerstone of entry into the teaching profession, should be reconsidered and revamped. While some research has shown that students of regularly certified teachers achieve more than those of teachers who are not certified or hold emergency certifications,(55) a growing body of research now questions the value and the consequences of current certification practices. Goldhaber and Brewer, for example, found that students taught by certified teachers had higher math and science performance on average, but the gains of their students were no greater than those of noncertified teachers.(56) In a 2001 study for the Maryland-based Abell Foundation, Walsh reviewed the full body of studies (150) that were routinely cited as justification for the state’s teacher certification process. She concluded that research used to advocate for the certification process was characterized by numerous deficiencies, that certified teachers were no more effective than those without certification, and that the state had turned away a number of highly skilled individuals because they did not meet its strict certification guidelines.(57) In 2004, Goldhaber urged a reconceptualization of licensure along with a focus on mentoring and training opportunities that would encourage more promising prospects to enter the teacher labor market.(58) Led by a former Chair and Chief Executive Officer of IBM, the Teaching Commission also asserts in a 2006 report that certification paradoxically creates cumbersome procedures that discourage many talented would-be teachers while permitting relative mediocrity among those who gain entrance to the profession.(59)
In general, research appears to recommend that we reexamine our assumptions about how to recruit, train, and compensate teachers. Policy clearly must be brought into alignment with our larger goals of elevating student achievement, which will require us to focus intently on the needs of those who are being left behind even as we strive to maintain achievement levels among those who are performing adequately. Indeed, the human resource system for teachers may require restructuring if we are to ensure that all students, particularly poor and minority children and those from undereducated families, are given every opportunity to excel.
A New Paradigm for Teacher Pay? Not only do teachers exert the most significant effect on academic outcomes, their pay represents the largest portion of the largest public investment we make. Thus, how we compensate and reward teachers has become the focus of increased scrutiny and broad experimentation at the school, district, city, and state levels. At present, Kentucky and 21 other states have a statewide salary schedule that sets minimum standards for compensation based on seniority and education.(60) Nationally, these same criteria are the principal determinants of teacher raises, rather than performance in the classroom or student outcomes.
The historic efforts of teachers’ unions to reform education gradually led to what one researcher refers to as a “codification” of education practices(61) that have effectively compromised the flexibility needed to meet diverse student needs. Current pay structures have also hamstrung efforts to recruit and retain teachers for critical and competitive subject areas, direct teaching talent where it is most needed, and reward excellence.
The rising demands of the global workforce and fiscal pressures are fueling a movement to restructure teacher pay. The movement has been bolstered by a growing body of research about factors that influence student performance and advanced by institutions of every political persuasion. About 30 states already offer incentives such as housing, loan forgiveness, bonuses, and salary increases to address geographic and subject-area teacher shortages.(62) The most controversial and possibly the most reliable differential pay mechanism, however, is the value-added model or VAM, which statistically links individual teachers to the test scores of individual students from year to year.(63) A growing number of state and federal political leaders now openly advocate a pay structure that, at least in part, links pay to academic performance,(64) and several states, including Minnesota, Florida, Texas, and Colorado, are implementing them.
But early indications from research suggest that pay for performance in its current forms garners small, if any, results. A review of international studies found outcomes sufficiently positive to recommend further experimentation but not to prescribe pay structure design.(65) Combining data from the National Education Longitudinal Survey and a 2000 survey on incentives, Florida economists found a link between test score gains and pay targeted at a few teachers, but “no association between test performance and indiscriminate merit pay.” Importantly, the positive relationship was strongest in schools serving low-income populations.(66)
At a recent conference held by the National Center on Performance Incentives at Vanderbilt University, findings from studies of various pay-for-performance plans were mixed. Education Week reported, “Some researchers concluded that positive effects of performance-based pay were small or none, while others found varying levels of increases in student achievement.”(67) The largest gains were found in a Little Rock, Arkansas, pilot program that gives bonuses to teachers who raise test scores, an approach widely criticized for its teach-to-the-test implications.(68) A University of Arkansas study found a 3.5 percent improvement in math and science in these Little Rock schools.(69)
At the Vanderbilt conference, Koppich reported that the most comprehensive approaches to pay for performance were found in the cities of Denver, Minneapolis, New York, and Toledo. Each incorporates various measures of teacher performance, including a VAM component, disdains punitive measures, and retains elements of the single-salary structure. Importantly, all were products of joint union and management efforts that engaged teachers in their design and implementation.(70) Early results from Texas also suggest that this voluntary statewide system is having a positive effect on teacher attitudes, as they report working more collaboratively with their colleagues and encouraging students more.(71)
Lower Pupil-Teacher Ratios. The RAND study of NAEP scores also concluded that “other things being equal ... lower pupil-teacher ratio[s] in lower grades ... show positive, statistically significant effects on [student] achievement.”(72) While some debate remains about the efficacy of class size reduction (CSR) as a tool to improve student achievement, an extensive body of research tends to support these findings. From his literature review, Hanushek concludes that smaller class size does not “systematically” improve student achievement, but Krueger, who analyzes the same studies using an alternative weighting scheme, refutes this conclusion. Regardless, both researchers, as well as many others, generally conclude that smaller class size benefits specific groups of students, subject matters, and teachers.(73)
Research has shown that CSR has the most impact on student achievement when intervention is early, starting in kindergarten or first grade, and students experience small classes all day, every day. While smaller classes should be maintained for at least two years, three to four years yield the most enduring benefits, these studies suggest. The ideal class size may differ with circumstances, but studies generally suggest that classes of 15 to 18 students offer optimal benefits. Importantly, much of the research about CSR finds it most effective for at-risk, low-income students.(74)
New British research echoes the potential benefits for “low-attaining students,” but concludes that reductions in class size increase student engagement at all grade levels. Moreover, these researchers found no “threshold effect,” or an optimum class size at which academic gains emerged. Rather, CSR appeared to help “at any end of the class-size spectrum.” Specifically, these researchers found that low-performing students were nearly twice as likely to be disengaged in classes of 30 students as compared with classes of 15 students. The addition of just five students to a class was found to decrease the odds that students would remain on task by nearly a quarter.(75) Results from a much larger Hong Kong study, however, found no differences in student engagement. However, teachers in smaller classes were less likely to rely on textbooks for instruction, more likely to speak with students more often, and students were more likely to ask for help outside of class.(76)
As shown in Figure 10, Kentucky’s K-3 class size is above the range researchers recommend, above the national average, and above many of our peer states, according to our analysis of findings from the Schools and Staffing Survey. Moreover, while the majority (61 percent) of Kentucky teachers express satisfaction with their class size, 39 percent indicate that they are somewhat or strongly dissatisfied with class size. By contrast, just 26 percent of teachers nationally express dissatisfaction.(77)
Figure 10: Average Number of Students in K-3 Classes, Kentucky, U.S., and Peer States, 2003-04
CSR numbers among those strategies that clearly will require additional funding for salaries, classroom construction, and professional development to avoid the unintended consequence of further misallocation of teaching resources. But even relatively poor states are investing in this research-driven strategy. For example, since 1987-88, Alabama has lowered average class size in the early grades by 7 students and increased its per pupil expenditures for all students by 58 percent, compared to a national average of 31 percent. Spending in Kentucky has increased by 56 percent.(78)
Adequate Resources for Teachers. RAND researchers also found that academic achievement is higher in states where higher percentages of teachers report having adequate resources for teaching. Our analysis of the Schools and Staffing Survey shows that levels of teacher satisfaction in this area have declined steadily since 1987-88. As shown in Figure 11, levels of satisfaction paralleled the national average on the most recent survey for which data are available in contrast to preceding years when they were much higher in Kentucky.
Specifically, 20 percent of teachers in Kentucky disagree somewhat or strongly with the statement, “Necessary materials such as textbooks, supplies, and copy machines are available as needed by the staff.”(79) Thus, it is reasonable to conclude that in the broad design of resource allocation, Kentucky has lost ground in an area that has been linked to academic gains.
Clearly, the context of labor market expectations has changed dramatically. Workers, as well as companies, now face global competition. Without a sufficient intellectual foundation to bring ideas to fruition in the marketplace, many fear that our nation’s standard of living will decline. To avoid continued economic stagnation, incomes that year after year lag the national average, Kentucky must continue to vigorously pursue its ascent in educational status. In some areas, research suggests the need for new investment, but a more informed, focused use of the significant resources we now deploy to public education may yield gains in the classroom, the workplace, and our larger economy.
Pasteur aptly articulated a profound and egalitarian view of knowledge and the fruits of scientific discovery, “Science knows no country, because knowledge belongs to humanity, and is the torch which illuminates the world.” By using the knowledge available to us even as we cultivate a deeper understanding of factors about which we know too little, we can begin to make informed investments in education that will yield greater, more broadly shared benefits. In turn, children from all walks of life can come to know the gift of learning, perhaps the greatest promise of democracy, and grow into men and women who will light the way for our state and nation.
* Michal Smith-Mello is a senior policy analyst, Michael T. Childress is Executive Director, Dr. Amy L. Watts is an economist, Mark Schirmer is a policy analyst, and Billie S. Dunavent is a senior research assistant at the Kentucky Long-Term Policy Research Center. Return to text.
1 National Commission on Excellence in Education, A Nation at Risk: The Imperative of Educational Reform (Washington: U.S. Department of Education, 1983). Return to text.
2 National Center on Education and the Economy (NCEE), Tough Choices or Tough Times (Washington: NCEE, 2006). Return to text.
3 Thomas L. Friedman, The World is Flat: A Brief History of the Twenty-first Century (New York, NY: Farrar, Straus and Giroux, 2007). Return to text.
4 Bureau of Labor Statistics, “State and Area Employment, Hours, and Earnings,” U.S. Department of Labor www.bls.gov 14 April 2008. Return to text.
5 Peter Pae, “Some Firms Replace Offshoring with Onshoring,” Los Angeles Times latimes.com, 21 Oct. 2007. Return to text.
6 Marc Tucker, “The Dynamic Global Labor Market,” presentation, “Making Money Matter,” Kentucky Long-Term Policy Research Center (KLTPRC) annual conference, Lexington, Kentucky, Nov. 13, 2007. Return to text.
7 World Economic Forum (WEF), The Global Competitiveness Report 2007-2008 http://www.gcr.weforum.org/ and Barbara Kiviat, “Getting to the Top,” Time 26 Nov. 2007: 60. Return to text.
8 WEF, “United States Country/Economy Profile,” The Global Competitiveness Report 2007-2008 at http://www.weforum.org/pdf/Gcr/profiles08/UnitedStates.pdf. Return to text.
9 Daniel H. Pink, A Whole New Mind (New York, NY: Riverhead Books, 2006). Return to text.
10 General Accounting Office, Higher Education: Science, Technology, Engineering, and Mathematics Trends and the Role of Federal Programs, GAO-06-702T (Washington, DC, 3 May 2006). Return to text.
11 National Science Foundation, Science and Engineering Indicators, 2006, Vol. 1, NSB 06-01, Table 2-37, Arlington, VA, Jan. 2006. Return to text.
12 National Center for Education Statistics (NCES), “Highlights from the Trends in International Mathematics and Science Study (TIMMS) 2003,” U.S. Department of Education, Dec. 2004. Return to text.
13 Organization for Economic Cooperation and Development (OECD), OECD Programme for International Student Assessment, “What PISA Assesses,” 17 April 2008 http://www.pisa.oecd.org/pages/0,3417,en_32252351_32235918_1_1_1_1_1,00.html. Return to text.
14 NCES, Institute of Education Sciences, “Trends in International Mathematics and Science Study (TIMSS): Overview,” undated, 16 April 2008 http://nces.ed.gov/timss/. Return to text.
15 NCES, “Highlights from PISA 2006: Performance of U.S. 15-Year-Old Students in Science and Mathematics Literacy in an International Context,” Dec. 2007. Return to text.
16 ERIC Clearinghouse on Assessment and Evaluation, ERIC Digest (ED429988) College Park, MD, 1999. Return to text.
17 NCES, “Highlights …”. Return to text.
18 Gary W. Phillips, Chance Favors the Prepared Mind: Mathematics and Science Indicators for Comparing States and Nations (Washington: American Institutes for Research, Nov. 2007). Return to text.
19 Each indicator was standardized (mean=0 and standard deviation=1) by a z-score transformation. The index is the average of the probability values, using a cumulative standard normal distribution of these z-scores and ranges from 0 to 1, with 0 being the worst performing state and 1 the best. Return to text.
20 The Education Week Quality Counts 2007 Achievement Index is available online at http://www.edweek.org/ew/toc/2007/01/04/index.html; the Morgan Quitno 2006-07 Smartest State Index is available at http://www.morganquitno.com/edpress06.htm. Return to text.
21 This included the Kentucky Department of Education, the Teachers’ Retirement System, School Facilities Construction Commission, and the Education Professional Standards Board. Based on the Center’s collection of data from budget documents. Supplementary Information to the Kentucky Comprehensive Annual Financial Report, various years. The reports are prepared by the Division of Statewide Accounting, Office of the Controller, Finance and Administration Cabinet. Return to text.
22 David W. Grissmer, Ann Flanagan, Jennifer H. Kawata and Stephanie Williamson, Improving Student Achievement: What State NAEP Test Scores Tell Us (Santa Monica, CA: RAND, 2000). Return to text.
23 Grissmer, et al. Return to text.
24 Education spending per pupil is calculated from total primary and secondary enrollment and current expenditure data available online from the NCES Common Core of Data http://nces.ed.govccd. Return to text.
25 Manhattan Institute, “The Teachability Index: Can Disadvantaged Students Learn?” Education Working Paper No. 6, Sept. 2004, 14 Dec. 2007 http://www.manhattan-institue.org/html/ewp_06_app13.htm. Return to text.
26 U.S. Chamber of Commerce, “Leaders and Laggards: A State-by-State Report Card on Educational Effectiveness,” 2007, 14 Dec. 2007 http://www.uschamber.com/icw/reportcard/default. Return to text.
27 U.S. Chamber. Return to text.
28 Grissmer, et al. Return to text.
29 National Governors Association (NGA), Building the Foundation for Bright Futures (Washington, DC: NGA, 2005). Return to text.
30 Lynn A. Karoly, Peter W. Greenwood, Susan S. Everingham, Jill Hoube, M. Rebecca Kilburn, C. Peter Rydell, Matthew Sanders, and James Chiesa, Investing in Our Children: What We Know and Don’t Know About the Costs and Benefits of Early Childhood Interventions (Santa Monica: RAND, 1998) 65-68. Return to text.
31 Karoly, et al. Return to text.
32 Valerie E. Lee and David Burkham, Inequality at the Starting Gate (Washington, DC: Economic Policy Institute, 2002). Return to text.
33 National Institute for Early Education Research (NIEER), The State of Preschool 2007 (New Brunswick, NJ: Rutgers Graduate School of Education and The Pew Charitable Trusts, 2008): 64-65. Return to text.
34 NIEER 64. Return to text.
35 NIEER 64. Return to text.
36 NIEER 65. Return to text.
37 Linda Jacobson, “Scholars Split on Pre-K Teachers With B.A.s,” Education Week 27 March 2007 <http://www.edweek.org>. Return to text.
38 National Institute for Early Educational Research, The State of Preschool 2006 (Edison, NJ: Rutgers, 2006) 72-73. Return to text.
39 National Association of Child Care Resource and Referral Agencies, We Can Do Better: NACCRRA’s Ranking of State Child Care Center Standards and Oversight, eds. Linda Smith, Grace Reef, Susan Perry-Manning, Mousumi Sarkar, Caitlin McLaughlin, Mary Beth Salomone, and Kelsey Kurth, 1 March 2007 http://www.naccrra.org. Return to text.
40 College of Education and College of Human Environmental Sciences, “Third Party Evaluation of the Kentucky Education Reform Act Preschool Programs,” University of Kentucky, 1997, 10-12. Return to text.
41 College of Ed. 18-24. Return to text.
42 Education Trust. Return to text.
43 S. Paul Wright, Sandra P. Horn, and William L. Sanders, “Teachers and Classroom Heterogeneity: Their Effects on Educational Outcomes”; Sanders, Wright, and Horn, “Teachers and Classroom Context Effects on Student Achievement: Implications for Teacher Evaluation,” Journal of Personnel Evaluation in Education 11.1: 1997, 57-67; Sanders and June C. Rivers, “Cumulative and Residual Effects of Teachers on Future Student Academic Achievement,” University of Tennessee Value-Added Research and Assessment Center, Knoxville, Nov. 1996. Return to text.
44 Ron Haskins and Susanna Loeb, “A Plan to Improve the Quality of Teaching in American Schools,” The Future of Children, Policy Brief, Spring 2007. Return to text.
45 Eric A. Hanushek, John Kain, Daniel O’Brien, and Steven Rivkin, “The Market for Teacher Quality,” Working Paper 11154 (Cambridge, MA: National Bureau of Economic Research, 2005). Return to text.
46 Dan Goldhaber, “The Mystery of Good Teaching,” Education Next Spring 2002: 50-55. Return to text.
47 Associated Press (AP), “Teacher Experience Matters,” Teacher Magazine 3 Dec. 2007. Return to text.
48 Hanushek, Kain, O’Brien and Rivkin. Return to text.
49 Charles T. Clotfelter, Helen F. Ladd, and Jacob L. Vigdor, “How and Why Do Teacher Credentials Matter for Student Achievement?” Working Paper 2, National Center for Analysis of Longitudinal Data in Education Research, Urban Institute, Washington, D.C., March 2007. Return to text.
50 David Grissmer and Ann Flanagan, “Improving the Achievement of Tennessee Students: Analysis of National Assessment of Educational Progress,” Technical Report, RAND Education, 2006. Return to text.
51 Goldhaber with Emily Anthony, “Indicators of Teacher Quality,” ERIC Digest 184 (2003). Return to text.
52 AP. Return to text.
53 Goldhaber and Dominic Brewer, “Evaluating the Effect of Teacher Degree Level on Educational Performance,” in William Fowler, ed., Developments in School Finance 1996 (Washington, D.C.: U.S. Department of Education, NCES, 1997) 197-210. Return to text.
54 Clotfelter, Ladd, and Vigdor 38. Return to text.
55 Linda Darling-Hammond, Teacher Quality and Student Achievement: A Review of State Policy Evidence (Seattle: University of Washington Center for the Study of Teaching and Policy, 1999). Return to text.
56 Goldhaber and Brewer, “Does Teacher Certification Matter? High School Teacher Certification Status and Student Achievement,” Educational Evaluation and Policy Analysis 22.2 (2000): 129-145. Return to text.
57 Kate Walsh, Teacher Certification Reconsidered: Stumbling for Quality (Baltimore: The Abell Foundation, 2001). Return to text.
58 Goldhaber, “Why Do We License Teachers?” in Frederick Hess, Andrew Rotherham, and Kate Walsh, eds., A Qualified Teacher in Every Classroom: Appraising Old Answers and New Ideas (Cambridge: Harvard Education Press, 2004): 81-100. Return to text.
59 The Teaching Commission, Teaching at Risk: Progress and Potholes, 2006. Return to text. Return to text.
60 Education Commission of the States, “Teaching Quality/Compensation,” State Notes, July 2005. Return to text.
61 Leo Casey, “The Educational Value of Democratic Voice,” in Jane Hannaway and Andrew J. Rotherham, eds., Collective Bargaining in Education (Cambridge: Harvard Education Press, 2006) 181-201. Return to text.
62 Linda Jacobson, “Teacher Pay Incentives Popular But Unproven, Education Week 27 Sept. 2006. Return to text.
63 Henry I. Braun, “Using Student Progress to Evaluate Teachers: A Primer on Value-Added Models,” Policy Information Perspective, Educational Testing Service, Sept. 2005. Return to text.
64 Vaishali Honawar, “Merit Pay Gets Bipartisan Support,” Teacher Magazine 26 July 2007. Return to text.
65 Michael J. Podgursky and Matthew G. Springer, “Teacher Performance Pay: A Review,” Journal of Policy Analysis and Management 26.4 (2007): 909-950. Return to text.
66 David N. Figlio and Lawrence W. Kenny, “Individual Teacher Incentives and Student Performance,” Journal of Public Economics 91.5 (2007). Return to text.
67 Honawar, “Performance-Pay Studies Show Few Achievement Gains, Education Week 12 March 2008: 7. Return to text.
68 Honawar March 2008. Return to text.
69 Doug Thompson, “Merit Pay Data Encouraging, Lawmakers Told,” Arkansas News Bureau, 12 Sept. 2007. Return to text.
70 Honawar March 2008. Return to text.
71 Honawar and Bess Keller, “Performance-Pay Setup in Texas Shows Promise,” Education Week 12 March 2008: 1. Return to text.
72 Grissmer, Flanagan, Kawata and Williamson. Return to text.
73 Alan B. Krueger, Eric A. Hanushek, and Jennifer King Rice, The Class Size Debate, Lawrence Mishel and Richard Rothstein, eds. (Washington, DC: Economic Policy Institute, 2002). Return to text.
74 Research summaries can be found at www.nea.org/classsize/research-classsize.html, www.centerforpubliceducation.org, and www.reduceclasssizenow.org. Return to text.
75 Debra Viadero, “Student Engagement Found to Rise as Class Size Falls,” Education Week 25 March 2008 www.edweek.org. Return to text.
76 Viadero. Return to text.
77 Amy Watts, “Class Size in Kentucky,” Policy Notes, KLTPRC, 25 (Nov. 2007). Return to text.
78 NCES, Common Core of Data, “National Public Education Financial Survey (State Fiscal),” selected years. Estimates are for total current expenditures, public primary and secondary education, in constant 2006 dollars. Return to text.
79 KLTPRC analysis of NCES Schools and Staffing Surveys, U.S. Department of Education. Return to text.