Student Achievement: Gaps, Sources, Possibilities
|In most school districts desegregating under court order, the achievement of students has been a major concern. Achievement has been a topic of court consideration in the Little Rock School District case. One of the major goals of the current desegregation plan is to close the achievement gap between African-American and white students. This goal has clashed with some hard realities.|
The desegregation plan of the Little Rock School District (LRSD) approved by the Eighth Circuit Court of Appeals in 1990 proposed that the achievement gap between African-American and white students be closed and stated that standardized tests should be used to measure that closure. Under the settlement agreement negotiated in 1989, the State of Arkansas has provided funding to reduce the achievement gap to each of the three school districts in Pulaski County and has loaned additional money to the LRSD. The terms of the settlement agreement provide that if African-American achievement scores rise to 90 percent of white achievement scores, the loan will be forgiven.
Standardized test scores provide a comparable measure across grades, classes, and schools. Grades given by teachers do not allow such comparison since there is no assurance that different teachers teach or test or grade the same way from class to class. Comparisons of standardized scores allow us to review student mastery of specific skills based on a test that all took in common.
Standardized test data from the LRSD show clearly the achievement gap between African-American and white students. Figures 8-1, 8-2, and 8-3 show the results for African-American and white students between Normal Curve Equivalent (NCE) scores on the basic battery on the Metropolitan Achievement Test in 1991 and the Stanford Achievement Test in the other years. Data are available for all grades, but only grades one, six, and eleven are shown for comparison purposes.
While the national standard score on these tests is 50, it will be noted that first grade white students started their formal educational programs with scores approximately 10 points higher than that norm and African-American students approximately 10 points lower, a disparity of approximately 20 points between the two races. By grade six both white students and African-American students had shown higher levels of achievement against the national norm. The gap between the races, however, stayed relatively consistent. By grade eleven the improvement for both races had been erased, and achievement returned to approximately what it had been in grade one.
This pattern has been generally repeated on the subtests in reading, mathematics, English, science, and social science. Reading is generally regarded as a crucial skill in learning in all subject areas. A closer look at reading is shown in Figure 8-4, which indicates the percentage of African-American and white students reading below grade level from 1989 to 1996. At any time during this period, in excess of 60 percent of African-American students and 30 percent of white students were reading below grade level, as measured by nationally-normed, standardized tests.
The LRSD should be given credit for the improvement noted in test scores for both races in the early years of school. Although an increase of approximately two points from grades one to six may not seem particularly significant, the fact that it is the average for nearly 17,000 students and has been consistent over the years is very meaningful. That the gap between the races did not diminish in those years may be criticized, but it may also represent a success for the LRSD. Since so many African-American students come from low income backgrounds, the mere fact that the gap did not widen may, in fact, represent a significant achievement.
Why the achievement levels of both races by grade eleven reverted to first grade levels, after improving through sixth grade, cannot be answered without additional careful research. For white students it may be that the exodus of white students starting in the junior high years may be a partial answer. If it is assumed that a disproportionate share of those leaving the LRSD are more capable students, that could begin to explain the reduced achievement. That does not answer the results for African-American students, however, since they do not appear to be leaving in significant numbers.
Improving the achievement of students of all races must be the primary goal of the LRSD. In addition, reducing the gap between the scores for African-American and white students must be achieved. The pattern being experienced in the LRSD should be the subject of a major research investigation to determine those factors contributing to improvements in average achievement levels in the early grades, and the loss of that advantage in the later grades. It is evident at this point that continued reliance on past practices will not achieve these goals, and that the efforts expended under the desegregation plan now in force have not been successful.
In addition to the data that related to the desegregation plan's commitment to narrowing the gap between African-American and white achievement scores, we reviewed test score data in math and reading for a five-year period at magnet schools and incentive schools, which also are elements of the desegregation plan. In comparing these two groups of schools, we found variations among schools, as one might expect. As a whole, magnet school averages were above the district averages while incentive school averages were below--findings one would predict, given the more selective magnet school student body and the more educationally-disadvantaged incentive school student body.
Incentive school students have tended to score lower on standardized achievement tests such as the Arkansas Minimum Performance Test in the third grade. By grade six, however, students in some incentive schools score as high as, and in some cases higher than, the magnet students at the comparable grade level. Schools in which this occurred include Ish and Rockefeller. When such achievement occurs, it suggests that the district should study and document the practices of successful teachers. Then those practices should be followed in other incentive schools and at area schools as well.
What can be done to increase achievement in the LRSD, particularly among minority students? There are two places the challenge must be met. One is outside the school and the other is inside. One is prevention, the other is intervention. This section on pre-school education needs will address a significant source of the achievement gap--a place where preventive action could be taken.
African-American children in the LRSD typically start the first grade with achievement test scores that are approximately 65 percent of those of their white classmates. Table 8-1 shows the first grade results for 1991-92 through 1994-95 on the complete battery of the Stanford Achievement Test.
LRSD achievement data indicate there may be a slight closing of the gap until the fourth through sixth grade range, but those gains are lost after grade six, and the relationship of the first grade is duplicated at high school graduation. In other words, measured from first grade to high school graduation, there has been no closing of the achievement test gap between African-American and white students.
This situation is not unique to the LRSD. It is a situation generally observed throughout the country. Three desegregation experts brought to Little Rock to testify before the federal court by U.S. District Judge Susan Webber Wright agreed that minimal progress in closing these gaps had been experienced throughout the country. Although all believed that specialized intervention programs in the schools have the potential for some improvement, such programs are unlikely to eliminate the gap by themselves.
A child who begins school with a significant achievement deficit is like a runner who has to start a 100-yard dash 35 yards back from the starting line. He is likely to still be far behind at the end of the race. Instead of being told to run faster and catch up with the others, what the runner needs is to be brought up to the starting line before the race begins.
When African-American and white children start first grade with an achievement test score gap of the magnitude present in the LRSD, gains can be expected only through extraordinary efforts by African-American children, accompanied by extremely supportive homes or communities. Demographic information suggests that the necessary home situations will not be present for a significant proportion of African-American children.
Information received from the Arkansas Department of Health shows that a large number of children are being born throughout Arkansas to unmarried women, unmarried teens, and mothers with low levels of education themselves. Although the Pulaski County numbers are generally lower than the state average, they are high enough to demonstrate that a societal problem exists that deprives children of supportive home influences and has a direct and significant impact on the ability of the public schools to achieve their goals. These circumstances are being experienced for both races at alarming levels but appear to have a disproportionate impact in the African-American community.
Table 8-2 shows that approximately 68 percent of African-American children in Pulaski County are being born to unmarried women, 25 percent to unmarried teens, and 26 percent to women with less than a high school education. Among white children, 15 percent are being born to unmarried women, 5 percent to unmarried teens, and 13 percent to women with less than a high school education.
The statistics for white births are sufficient to cause concern about how they may influence the educational achievement of a sizable number of white students. The statistics for African-American children, however, are so high that they represent a crisis that must be acknowledged as a community problem over which the public schools have no control and for which they cannot totally compensate starting from age five.
Perhaps the question needs to be asked, why do poor children start school educationally behind their middle-class and more well-to-do classmates? We would not pretend that there is a definitive answer, based in research, to this question. However, research and common sense both suggest that low income homes are more likely to place a low value on education. Parents of such homes, usually with limited education themselves, are less likely to exhibit or encourage behavior that helps to get a child ready for school. For example, there will likely be few books, magazines, puzzles, games, and toys that stimulate the early physical development of the child's brain. The child is less likely to see parents read books or hold children and read a book to them. For example, there are children who start school not knowing the colors. They are not color blind, but their lives are devoid of the kinds of nurturing interactions which enable other children to learn the colors.
In short, many of the early mentally nurturing experiences which help make other children ready for school simply are absent in the homes of many poor children. As a result, they are behind when they start to school; and since the other children continue to learn, too, the gap is likely to continue. Further, as such children grow older, their parents are not likely to encourage them to believe they can go to college and become physicians or lawyers or engineers. So they do not even consider the possibility.
There are existing pre-school programs that are aimed at this problem. One program, Home Instruction Program for Pre-School Youngsters (HIPPY), is aimed at enabling parents of disadvantaged children to play an increased role in their children's education. In Little Rock, HIPPY serves some 250 families, and there are similar programs in 137 school districts in the state. The LRSD also provides a Four-Year-Old Program in 20 schools, in which participation is voluntary. The Arkansas Department of Education partially finances the Four-Year-Old Program with a special-purpose federal grant. But space is limited and this program has a long waiting list. These two commendable programs and others aimed at pre-school needs have been limited by uncertain funding, inattention, and lack of careful evaluation. At the moment they appear to be only a drop in the bucket compared to the need.
Sometimes children may succeed academically even though born into single parent homes where nearly 25 percent of the parents are mere children themselves and an equal proportion have low levels of education; but the odds are certainly not in their favor. It is beyond the scope of this study to deal with the community crisis the statistics suggest, but it must be recognized. The crisis is a reality, not a theory. It must be addressed in the proper arena if long-term solutions, especially for African-American children, are to be realized. For the purposes of this study, it can only be identified as a primary source of the educational challenges which show up in achievement score gaps and for which compensating efforts must be attempted in the schools.
Achievement disparities that exist between African-American and white children in the LRSD at first grade indicate that concentrated efforts must be directed toward achievement and readiness to learn in the 0-5 age range. Historically, parents and families have assumed these responsibilities with the public schools assuming responsibility only when children entered kindergarten at age 5. Although limited pre-school programs may be available, the public schools are generally funded to provide educational programs for children from age 5 until age 17 or 18 when they complete high school. The evidence is that such arrangements must be altered in the future to promote the academic achievement of all children and reduce disparities between African-American and white children.
Incentive School Funding
The foregoing discussion of pre-school education needs provides a context for comments on one element of the current desegregation plan that has been controversial. In 1996-97, there are now five incentive schools, all elementary schools, in operation in the LRSD: Franklin, Garland, Mitchell, Rightsell, and Rockefeller. Incentive schools receive double funding because they enroll the most educationally-disadvantaged children in the city.
Double funding each incentive school seems unnecessarily arbitrary. Some schools may need less money, others more. However, the underlying concept of larger funding for such schools is reasonable. Every successful organization--whether a business, an army, or a ball team--determines where its greatest challenge is. Then it concentrates and focuses the resources necessary to meet the challenge. In any school district, children from poorer homes will in general be the most educationally disadvantaged. It will take more time, effort, and resources to get them up to par. It makes sense to approach the distribution of funds in a way that takes account of this fact, just as the distribution of funding must also take account of the fact that science programs, because laboratories are expensive, require more dollars than social science programs.
The relationship between being economically disadvantaged and being educationally disadvantaged has been noted. They tend to go together. Sometimes there is a misperception that poverty programs and other government-support programs primarily benefit African Americans. However, there are more whites than African Americans receiving food stamps in Arkansas. (The number in fiscal year 1995 was 254,666 whites to 177,448 African-Americans, according to the Division of Research and Statistics of the Arkansas Department of Human Services.) The same is true of other government support programs. In Little Rock, however, the poor are disproportionately, but not exclusively, African-American. Disadvantaged children show up at school from families of all colors, and the schools must do their best to educate them regardless of the circumstances out of which they come. In the LRSD, these children are disproportionately African-American and constitute a very high percentage of the children in incentive schools.
While the achievement problem appears to be fundamentally rooted in pre-school circumstances outside the school system, the schools must, nonetheless, strive to eliminate the achievement deficits of their pupils at all levels. There is no easy solution, but there is a key to greater success in intervention strategies once children have entered school and are significantly behind their classmates: an attitude of experimentation within the school. We believe an attitude of experimentation is essential if the LRSD is to rise to the challenges of the day and recapture the admiration and respect of the public.
When the United States became concerned that the Soviets were winning the space race in the 1950's, the emphasis on regaining the leading position was placed on education. The lead was regained, and it gave the public schools a level of visibility and a position of importance in society that was probably unprecedented. In part as a result of that experience, the public schools have often been seen as the avenue through which other problems of society could be corrected or at least lessened. Some of those objectives are appropriate, but America may have imposed so many of them that the primary objective of education, student achievement, has often been overlooked or at least has taken a secondary role to other objectives.
It cannot be said often enough or loudly enough that the main goal of any public school district must be student achievement. This is not to say that the public schools cannot or should not join with their communities in helping to solve other problems; but other goals should not be pursued at the expense of student achievement.
In placing student achievement first, educators in the public schools are faced with more complex societal circumstances than in the past. Family structures have changed, respect for authority has changed, students have changed, economic opportunities have changed, and the world community has changed. As a result, the traditional methods of teaching and learning must be constantly challenged and re-evaluated. School boards and superintendents should encourage an attitude of experimentation, and principals should hold teachers accountable for the achievement of their students in a supportive environment conducive to such experimentation.
American education generally does not reward teachers and administrators who are willing to take risks in attempting to improve student achievement. All too often, rigid organizational structures and bureaucratic requirements not only do not reward experimentation--they actively discourage it. Teachers are not taught to engage in experimentation and evaluation as a means for improving achievement, and school districts rarely provide competent and available assistance. It is a situation that must change--because the successes of isolated examples where people were willing to take the personal risk of experimenting are so convincing.
One example is in Dayton, Ohio. In 1989 Mr. Rudolfo Bernardo was appointed as Principal of the Allen School in Dayton. It enrolls 540 students and has a staff of 28 certified teachers. When Mr. Bernardo arrived he found that 78 percent of the students were from families receiving Aid to Dependent Children (ADC), 59.5 percent were from single parent homes, and 60 percent were African-American. By 1995 the student body remained 60 percent African-American; students coming from single-family homes had increased to 70 percent; and those from ADC families had increased to 86 percent.
Considering the demographic information, one would conclude that the situation at Allen had worsened. However, between 1989 and 1995:
What happened? Mr. Bernardo took the risk of experimenting with a back-to-basics character education program. He enlisted the involvement of the teachers, students, parents, and the community in the program. This had a profound impact not only on achievement but also on the general attitude of students toward school. These things would not have happened for the students at Allen School had it not been for an attitude of experimentation by Mr. Rudolfo Bernardo. (The Allen School example is discussed further in Chapter Eleven.)
Another example of an education program with a success rate that cannot be ignored is called Reading Recovery. In the 1994 book entitled Redesigning Education, which was written by Nobel Prize winner Kenneth Wilson and journalist Bennett Daviss, an entire chapter is devoted to this program.
Reading Recovery is a program that resulted from more than two decades of research, experimentation, and evaluation by Dr. Marie Clay, a professor of child psychology at the University of Auckland in New Zealand. It concentrates on reading deficiencies of first grade students. Phenomenal success has been documented in study after study. As stated by Wilson and Daviss, "In poor urban and posh suburban schools alike, the program consistently rehabilitates more than four of every five first grade readers in just 12 to 20 weeks of daily half-hour lessons."2 They also stated, "Reading Recovery offers U.S. education its first real demonstration of the power of a process combining research, development (including ongoing teacher education), marketing, and technical support in an orchestrated system of change."3
In 1982, the New Zealand government made Reading Recovery a national program, and it now reverses reading failure for more than 90 percent of New Zealand first graders headed for illiteracy. It was introduced in the United States in 1984 when a group of educators in the College of Education at Ohio State University started a pilot program in Columbus, Ohio. Ten years later Reading Recovery was being used successfully in 700 schools in 48 states. Reading Recovery would not be available had it not been for an attitude of experimentation by Dr. Marie Clay.
The LRSD should make analysis, experimentation, and evaluation priorities. Competent analysis will identify the problem areas, experimentation will try out different ways of correcting the problems, and evaluation will demonstrate whether or not efforts were successful. Public education does not have a history or structure which encourages teachers and administrators to consider creative approaches to solving problems. Criticism is an inevitable result of taking risks, and the majority of public school officials and teachers appear to have retreated to the safe haven of making no waves.
Increased academic achievement and a reduction of the gap in standardized test scores between white and African-American students will not result without trying new ways and possibly making some waves. With carefully designed programs created by teachers and school administrators to solve achievement problems, with thorough evaluation measures specified as a part of the planning process, and with honest consideration of the results, the chances of improving achievement levels and reducing the achievement disparity significantly improve.
When student achievement--specifically remediation and the closing of the gap in standardized test scores--became a part of the prevailing paradigm in desegregation cases, it turned out to be a double-edged sword for African Americans.
On the one hand, it was a very demanding standard (closing the achievement gap reflected in standardized test scores) and, therefore, a source of leverage to African-American lawyers and plaintiffs. It became a basis for courts to mandate policies, programs, and expenditures with the goal of helping African-American students achieve equality in performance.
On the other hand, the primary tool of measurement of school district success or failure was the standardized test, which many African Americans believe to be culturally biased and which routinely yields higher white and lower African-American average scores. Standardized tests have the prestige of social science surrounding them. Because of the major place of standardized tests in desegregation cases, the media prominently report the results of each test cycle. In short, widespread use of standardized tests in desegregation cases has tended to reinforce racial superiority/inferiority stereotypes.
In short, the achievement goals prominent in desegregation cases have produced a lose-lose situation. School districts have lost because the required goals have proven largely unachievable. African Americans have lost because of the enormous prominence given to standardized test score averages. Too often the point has been lost that, regardless of the different group averages and the reasons for them, many African-American individuals are extraordinarily able and score very high on standardized tests, just as the white race provides many citizens at the low end of the continuum.
In any event, in Little Rock and elsewhere there is little reason to expect any appreciable narrowing of the test score gap in the foreseeable future. Experience clearly suggests now that either an achievement mandate imposed by a court on a school district, or an achievement commitment voluntarily accepted by a school district, is problematical. School districts in fact control their employees, their facilities, their budgets, and their curricula; and they can deliver on commitments involving these matters. But school districts do not control the decisions of parents or students. Nor do they control the motivation and learning of students--at best they can work to influence them.4
Pre-School Education Needs
The limited success of schools in closing the achievement score gap strongly suggests that the attack on the problem is focused on the wrong place. The data in Tables 8-1 and 8-2, above, must surely send one of the strongest messages in this report: The achievement gap already exists when children begin school. The achievement problem needs to be attacked, therefore, before age five. The number of children born to unmarried women, many of whom are teens, and many of whom lack a high school education, can only be described as a community crisis. It would unquestionable be in the community's best interest to devise and carry out strategies to increase the readiness for school of such educationally-disadvantaged children. Such strategies, however, should be subject to analysis, experimentation, and evaluation.
Attitude of Experimentation
Perhaps the charter school legislation will make it easier for some schools to embody an attitude of experimentation.5 In any event, we believe that freedom to experiment and an eagerness to try to develop new programs will do more to enhance student achievement than will intervention programs mandated by a school board or administration.
While schools as a whole are not responsible for the gap in achievement scores and will have limited success at best in narrowing the gap, teachers must nonetheless accept individual accountability for the progress of each student in their classrooms. All students can learn. Teachers should examine skills and competencies mastered and not mastered, as revealed by individual student achievement test results, and ensure that those skills and competencies are taught within their classes, with all students held to high expectations. This is where standardized test scores are most useful and most appropriately used.
Finally, good data feed an atmosphere of experimentation. The LRSD needs to invest in and develop an ongoing student database which facilitates analysis of student achievement. This requires a capacity to pull data, by individual student, year after year. As simple as this might sound in the computer age, few school districts have such capability. The availability of such data on children in school is also essential to research on the long-term effects of pre-school programs.
1. In a focused analysis of scores from the Stanford Achievement Test, Eighth Edition, 1991-92 and 1993-94, the Office of Desegregation Monitoring found that African-Americans and white students in the LRSD had both made gains, but white gains had been greater. As a result, the gap had widened. See the report, "Status Report on Achievement Disparity," issued by the Office of Desegregation Monitoring, United States District Court, Little Rock, June 21, 1995.
2. Kenneth Wilson and Bennett Daviss, Redesigning Education (New York: Henry Holt and Co., 1994), 50.
3. Wilson and Daviss, 76.
4. The experience in Little Rock suggests that a desegregation idea that was discussed but not embraced or implemented--a college scholarship program for high school graduates--might have been much more effective in motivating learning than have been the various intervention programs which have been tried.
5. For information on the Arkansas law providing for charter schools, see the Appendix D: Governance Questions at the end of this report.