By Phillip W. Roeder(*)
From Foresight, Vol. 7, No. 2
Editors Note: The following article is an abbreviated version of an electronic book that was published in 1999 and updated recently to incorporate new data. The full-length version of the following article is available online at: www.uky.edu/~proeder/kerabkupdate.pdf. The e-book can be ordered by contacting the author at: RDSpub@aol.com or email@example.com.
April 11, 2000, marked the 10th anniversary of the signing of the Kentucky Education Reform Act (KERA) of 1990, a significant milestone for the development of Kentucky’s system of public education and for the broader history of the state, for the overall progress and well-being of the Commonwealth. Reaching this anniversary means not only that Kentucky initiated one of the most comprehensive and substantial reforms of public education, but that it managed to sustain that reform over a decade of significant changes in the political and economic life of the state. Likewise, this anniversary is not only a time to celebrate all that has been accomplished by KERA, but also a time to recognize that much still needs to be done to achieve a world-class system of public education in the Commonwealth.
This article summarizes a recent study of Kentucky’s public educational system under KERA, published by the author as an electronic book in 1999 and updated in February 2000. The study examines and attempts to explain the performance of Kentucky schools and school districts over this decade of reform. The initial study, published in early 1999, found that school performance in the period 1993 to 1997 was determined to a large extent by a set of related factors—social, economic, and organizational conditions that existed prior to the implementation of KERA. Using multivariate models, the 1999 study found that the degree to which a school was characterized by poverty and disadvantage largely determined how well it performed over the five-year period. When controlling for factors, including school size (enrollment) and type (elementary, middle, or high school), teaching resources (student-teacher ratio), extent of bureaucracy (school and district), competition from private schools, district revenue and district achievement prior to reform, change in revenues due to KERA, and whether the schools were early adopters of family resource/youth services centers and school councils, underperforming schools were much more likely to have higher proportions of poor and minority children and to be located in school districts with lower levels of previous academic achievement and fewer financial resources prior to implementation of KERA. It was found that differences between schools in proportion of poor children could translate into substantial gains or declines in performance when controlling for these other factors. These findings were consistent for both yearly cross-sectional models and for models assessing changes in performance over time.
The February 2000 update of this research project focused primarily on Chapter Four of the earlier book and assessed the performance of Kentucky schools using additional data from the 1998 KIRIS and 1999 CATS accountability systems. The findings of the 1999 book do not change with two additional years of school accountability data; and in fact, the conclusions are now supported even more strongly with the addition of the 1998 and 1999 data. After controlling for several school and district factors, not only does disadvantage continue to be the strongest determinant of school performance, but the entire system now appears to have evolved into a relatively stable pattern of accountability behavior.
Throughout the decades-long struggle to improve Kentucky’s public education system, primarily through the reinvented system enacted in 1990 called KERA, most policymakers, reform advocates, and educators have maintained a consistent emphasis on the goal of increasing excellence in schooling. Although much disagreement exists over defining educational quality and measuring school performance, there appears to be an overwhelming consensus that achieving excellence is the paramount goal of the educational system. Parents, educators, business leaders, policymakers, and others want high-quality schools, even if they do not always agree on definitions of this complex concept.
Despite the dominant emphasis on quality, American public education has always faced a dilemma: Is it possible to provide the highest quality schooling for all children, all regions of a state or nation, all schools, and all communities? Can disadvantaged schools with high proportions of poor children and fewer educational resources perform at the same high levels as schools with less poverty and more resources? The 1999 study found that some disadvantaged schools did perform well (or better than would be predicted based on their poverty), but most did not. Eight of these disadvantaged, high-performing schools were examined and assessed in Chapter Five with several interrelated factors seeming to explain how these schools managed to perform at high levels despite their disadvantages. Regardless of the lessons that might be learned from these special cases, it is important to emphasize that the strongest and most consistent predictor of school performance in Kentucky from 1993 through 1997 was child or family poverty.
To highlight the continuing problem of equity and disadvantage in Kentucky’s schools, this paper will begin at the end, that is, the end of Chapter Four from the 1999 book. An update of Table 4-10 in that chapter (see Table 1) compares school performance using accountability scores (KIRIS and CATS) for the 1993-99 period for the top 10 percent of schools that were most advantaged and bottom 10 percent that were most disadvantaged prior to the initiation of KERA. Disadvantage is measured as a composite index (factor score) that combines proportion of children in the school eligible for subsidized meals, as well as state/local revenues per student and academic success in the school district prior to the initiation of KERA. The comparison shows that after almost a decade of comprehensive and sustained reform, Kentucky is still a long way from achieving equitable excellence in the public schools. The performance gap between advantaged and disadvantaged schools increased dramatically from 1993 to 1999, but especially from 1996 to 1999. The gap widened from 10.2 points in 1993 to 16.0 points in 1999, but the key year was 1997, when after three years of relatively improved performance and a closing of the performance gap to less than 8 points, disadvantaged schools began to lag even further behind the advantaged schools.
Table 1: Accountability Scores For Advantaged and Disadvantaged Schools
These data suggest that schools in poorer communities have yet to overcome disadvantages developed over decades and generations of poverty and underperformance. A crucial question for those interested in equitable excellence is why after several years of a modest but steady narrowing of the performance gap did this already large differentiation double in three years (from 8 to 16 points)? To try to answer this and related questions about school performance, data from the 1999 book are updated and examined in more detail.
Table 2 summarizes accountability scores for all Kentucky schools for 1993 through 1999 and, similar to Table 1, it shows uneven progress in performance. The mean score for all schools increased from 1993 to 1994 and from 1994 to 1995, but declined slightly in 1996. After another modest gain from 1996 to 1997, scores declined again slightly in 1998 before increasing substantially in 1999. As an indicator of the overall progress in school performance under KERA, the improvement in average school scores from 1993 to 1999 was an impressive 23.7 points or an increase of 66 percent. However, because the state initiated a new system of accountability in 1999 (CATS) and because improvement in the transition year from 1998 to 1999 was almost 12 points or 25 percent, this substantial increase in performance over the seven-year period may be misleading. Although the system of KIRIS testing and scoring changed in several important ways throughout the 1993-97 period, the accountability system was changed significantly in 1998 (CATS) and used for the first time in 1999. Because of this major system change, it seems likely that the large increase in average score in 1999 is more an artifact of the new testing system than a real increase in student learning or school performance.
Table 2: School Accountability Scores, 1993-1999
In addition to describing performance for all schools, Table 2 compares the highest-scoring schools to the lowest-scoring schools beginning in 1993 and follows these same two groups of schools through 1999. These data show a pattern of performance somewhat similar to the advantaged and disadvantaged schools in Table 1. The worst-performing schools in 1993 (bottom 10th) made substantial improvements in 1994 and closed the performance gap from 22 to 14 points, but then leveled off through 1996 before falling further behind in 1997, closing the gap slightly in 1998, and then falling behind once again in 1999. Except for an initial spurt of improvement in 1994, the poorest-performing schools in 1993 have not closed the gap with the highest performers in 1993; in fact, they now appear to be falling further behind the initial highest performing schools, at least since 1994.
An issue that educators and reformers often fail to consider concerns stability or unpredictability in student achievement or school performance. An unacknowledged assumption of many advocates of school reform is that school performance is and should be somewhat unpredictable. This seemingly counter-intuitive and perhaps unusual assumption follows directly from the more commonly accepted assumptions that all students can learn at high levels, and organizational and curricular changes can have significant impacts on student learning. Despite research studies that find student learning and achievement to be affected to a large extent by various mental, emotional, and physical characteristics of individual children; by the backgrounds, structures, and processes of their families; and by long-standing social and economic structures in their communities; education reformers believe that certain educational strategies, teaching practices and methods, and curricular innovations will greatly improve student achievement and school performance. If not, why continue to develop and implement reforms in search of educational excellence? When the right combinations of effective programs and practices are found by policy and program experts and when these programs are implemented successfully by skilled educators, even previously low-achieving students and poorly performing schools will perform at higher levels. This optimism, rationalism, and belief in individual as well as social progress and perfectibility are basic components of American culture and affect most aspects of society and the economy as well as public education.
In addition to examining and comparing changes in school accountability scores over time to assess the impact of reforms as found in Tables 1 and 2, another way to assess impact would be to measure the degree to which the entire system after KERA has stabilized. Too much stability or predictability might suggest little or no impact of the extensive reforms. Such stability would suggest that whatever was taking place in the successful schools prior to the 1990 reform remained in place after reform (these schools would likely be successful regardless of how extensively the system was changed); and conversely, whatever was taking place in the less successful schools was not improved by the reformed system (something other than the reforms was keeping them from improving).
The questions are: how stable has school performance been under KERA, and how much stability is too much? Tables 1 and 2 show uneven progress over time and suggest at least modest instability in school performance. Although the implication of this uneven progress would be that schools can and do change substantially from one year to the next, further analysis of these data in Table 3 instead finds increased stability in school performance. Fewer schools are experiencing such large relative gains or declines that they leap ahead or fall far behind in the rankings of schools from one year to the next, especially in the latest years of accountability. Examination of bivariate correlations for each succeeding year of accountability shows a steady increase in strength from .56 for 1993-1994 through .68, .71, .75, .82, and .87 for 1998-1999 (the diagonal of Table 3).
Table 3: School Accountability Score Correlations, 1993-1999
It is important to note that the highest correlation in Table 3 of .87 for scores in 1998 and 1999 is for the last year of KIRIS and the first year of CATS—the restructured accountability system. Despite this major restructuring and despite the extremely large increase in average scores from 1998 to 1999, school rankings remain quite stable in this transition. Almost 76 percent of the variation in 1999 scores can be “explained” or predicted by 1998 scores and almost 55 percent of variation in 1999 scores can be predicted by scores from two years previously in 1997. Whatever can be said or concluded about school performance, whether measured by the old KIRIS or the new and improved CATS, schools now appear to be in a relatively stable system of performance and accountability.
The increased stability in performance shown by the correlational analysis in Table 3 complements the data in Tables 1 and 2, and may help explain why disadvantaged schools are falling further behind advantaged schools or why initial poor performers have not been able to catch up to the high performers; and most importantly, why both groups of laggards may have a very difficult time making substantial performance gains or improvements in the future. After the initial “shock” of KERA in the period of early implementation (early to mid-1990s), the complex patterns of behavior of people and systems appear to have stabilized or solidified in such a manner that very few schools are now making the sizable improvements in performance necessary to move up in the school performance rankings. If experience is starting to tell, and if patterns of resource allocation, curricular emphasis, school leadership, teaching practices, and other important components of the complex educational system are becoming “locked in,” along with quite stable economic and social conditions of families and communities, it may take another comprehensive and significant shock to the system similar to KERA to have any substantial impact on the distribution of excellence across Kentucky’s school districts.
Since the adoption of KERA in 1990, disadvantage continues to be the most substantial and significant determinant of the performance of Kentucky schools. In addition, school performance is stabilizing, indicating that the large gap between the haves and the have-nots is not likely to decrease or be reversed without significant policy changes. What are the implications of these findings for policymakers concerned with providing an effective and efficient system of public schools prepared to meet the challenges of the 21st century? What actions should be taken or adjustments made to address the performance gap and continue to strive for equitable excellence? There are many possible reforms in policies or programs that offer some possibility of achieving more equitable excellence in Kentucky’s primary and secondary school system, but two that seem particularly relevant are more equitable funding and more equitable distribution of teaching resources.
More Equitable Funding. The first and most obvious policy tool to deal with inequalities across Kentucky’s local school systems would seem to be more equitable funding across school districts—the primary impetus for the original suit that challenged Kentucky’s system of school finance and resulted ultimately in KERA. To continue and even accelerate improvements in the performance of Kentucky’s primary and secondary schools, more financial resources are needed. But the need is greatest in the poorer schools and districts. Increased salaries for instructional staff, more teaching resources, and improved physical facilities are but a few of the demonstrated needs for increased funding. Some argue that poorer schools with more social and educational disadvantages do not need equal resources; they need more resources than schools serving children from well-off families living in well-off communities. To accomplish such a goal, more new dollars would have to be invested in the K-12 education system, and the SEEK program would have to be revised so that poorer districts acquire even more state resources. Money is certainly not the only resource needed by schools, and money alone cannot solve all the problems faced by our less successful public school systems, but resources do matter.
More Equitable Distribution of Teaching Resources. Another important focus for possible improvements in school performance is teaching. Some experts now claim that not only does good teaching improve learning and performance significantly for all students, it has especially large and positive impacts for disadvantaged children. If the key to successful learning, especially for poor children, is teaching, how can underperforming and disadvantaged schools improve and increase their teaching resources? Answers to this question must assume that:
(1) good teaching can be defined and measured with reasonable precision
(2) teacher training programs in our colleges and universities know how to produce high-quality teachers (regardless of the types of individuals entering these programs either at the undergraduate or graduate levels), and
(3) disadvantaged and underperforming schools can acquire their fair share of good teachers.
Since these three assumptions about good teaching are somewhat questionable, in order to try to bring more and potentially better teaching resources to disadvantaged schools, I propose that an experimental program called the Kentucky Teaching Corps be created. Such a program would target the bottom 10 percent of poorest-performing districts or schools and pay new or recent college graduates to serve a two-year apprenticeship in these underperforming school systems. It would assign two or three apprentice teachers per school to work under the supervision of a master teacher from another successful system and a supervising teacher on site.
Further, it would pay these apprentice teachers salaries and benefits comparable to new teachers in the system, but as an incentive to locate in these more disadvantaged schools, also pay off their education debts accumulated while seeking their college degree (or if they have no college debts offer them bonuses if they stay with the program for the allotted two years). This apprenticeship period might help attract more people to the teaching profession, enable prospective teachers to become more effective in the classroom, especially working with disadvantaged children, and assist the schools and children they will work with during this apprenticeship.
To achieve greater levels of excellence in Kentucky schools and overcome the apparently stable patterns of inequality that have developed in the late 1990s, some type of educational Marshall Plan may be necessary. A significant, comprehensive, and long-term investment of educational resources in the poorest-performing school systems may help achieve higher levels of equitable excellence. Another shock to the system like KERA may be necessary, but since KERA II seems unlikely any time soon, the two proposals on funding and teaching resources presented here may be more feasible and acceptable alternatives. Some may characterize proposals for increased investments in the public schools, especially in poor communities, as financially and political unfeasible or even undesirable, but what could be a higher priority than equal educational opportunity and a better future for all Kentucky children, not just those fortunate enough to be attending successful or resource-rich schools? �