Losing Generations: Adolescents in High-Risk Settings (1993)

Education is widely viewed in the United States as the means by which individuals from economically or socially disadvantaged backgrounds can build the skills and credentials needed for successful adult roles in mainstream American life. For many students, however, schools do not now work this way, despite two decades of public debate and reform. This chapter focuses on academic schooling, particularly on those schools that are the educational setting for students from low-income families and neighborhoods, and those who are labeled "low achievers." Adolescents from low-income families and neighborhoods are at much higher risk of educational failure than their more affluent suburban counterparts.¹ Because of residential stratification, most of these adolescents attend schools with the fewest material resources and the least well-trained teachers. Their schools use instructional methods that are not conducive to learning challenging tasks. Compounding these disadvantages are generally lower expectations for student achievement. "The low expectations in our suburban high schools are high in comparison to

expectations in urban schools and rural schools with concentrations of children in poverty … this absence of challenge, of rigor, is dulling the minds and dashing the hopes of millions of America's children" (Commission on Chapter 1, 1992:1).

The income stratification that concentrates large numbers of low-income students into poorly funded schools is followed by instructional stratification, most often on the basis of prior performance. Low-achieving students are likely to be exposed to instructional practices—tracking and grade retention—that deny them educational opportunities, stigmatize them, and contribute to their sense of uncertainty and alienation. Many disadvantaged adolescents are unable to overcome these conditions. Students from low-income families are far more likely to receive bad grades or be held back, and as much as three times more likely to drop out before completing high school, than the children of more affluent families: "consigning them to lives without the knowledge and skills they need to exist anywhere but on the margins of our society, and consigning the rest of us to forever bear the burden of their support" (Commission on Chapter 1, 1992:1).

The educational system has not adequately addressed the conditions of schools in low-income neighborhoods, which face special challenges. Not only are their students more likely to have significant academic deficits and behavioral problems, but the schools in low-income neighborhoods face challenges for which they have not been prepared and receive little public understanding or support (see Chapter 4). Moreover, recent reforms generally ignore the organizational and instructional practices that might promote better academic achievement (Braddock and McPartland, 1992), and existing compensatory programs are deeply flawed from an instructional perspective. The net result is that many students continue to find education boring, instruction lacking in relevance, and schools inhospitable places for learning.

The school performance of adolescents is typically assessed through measures of individual achievement and grade attainment. Studies have produced a wealth of data on national trends in individual achievement. Over a range of measures, adolescents in 1990 generally performed at about the same level as adolescents in 1970. This stability, however, is at relatively low levels of demonstrated ability. For example, only 42 percent of all 17-year-olds (still in school) can be considered "adept" readers, and a

majority are performing at only an eighth grade level in math. In contrast, graduation rates have shown a slow increase (about 3.5 percent) increase since 1970. In 1990, 82.3 percent of all 18- to 24-year-olds had earned a high school diploma or a general equivalency diploma (GED) (Carter and Wilson, 1991; Mullis and Jenkins, 1990; Mortenson and Wu, 1990; Bean and Tienda, 1987; De La Rosa and Maw, 1990).

Trends in achievement show differences among racial and ethnic groups. For whites, achievement test scores have remained virtually unchanged for 20 years. In contrast, the performance of black adolescents has steadily improved, as has that of Hispanics (to a lesser degree), thereby narrowing the achievement gap between whites and blacks. Nonetheless, the average achievement scores of blacks tend to be two to three grade levels lower than those of whites, and the difference in graduation rates is approximately 5 percent. Despite the average gains in achievement on National Assessment of Educational Progress assessments, questions about the distribution of those gains remain, especially in light of evidence from some urban school districts that show a widening race gap in student reading performance (Entwisle and Alexander, 1992). Average trends appear to mask continuing, perhaps worsening, problems at the lower end of the distribution and in large school districts with many poor children (Braddock and McPartland, 1992).

Graduation rates similarly show differences among racial and ethnic groups. For whites, they have remained constant over the past 20 years. Blacks have made substantial gains in completing high school (by diploma or GED), rising from 59.5 percent of 18-to 24-year-olds in 1970 to 77 percent in 1990. Relatively few Latinos complete high school: over the past 10 years, the proportion of 18- to 24-year-olds who had a diploma or GED has fluctuated from a low of 54.1 percent in 1980, to a high of 62.9 percent in 1985, and back to 54.5 percent in 1990 (Carter and Wilson, 1991).

Family income and occupational background are the strongest predictors of school performance. From early adolescence, it is evident that schools are unable to capture the interest or facilitate the achievement of many low-income students. For example, fully 11 percent of eighth graders from low-socioeconomic-status (SES) backgrounds were absent more than one-quarter of the 1989 school year, a rate double that of high-SES students. Low SES is strongly and consistently associated with poor academic performance, and children from low-income families are three times

more likely to drop out of school than are children from middle-income families and nine times more likely than students from high-income families (Barro and Kolstad, 1987; Smith and O'Day, 1991; National Center for Education Statistics, 1990, 1991). Graduation rates also vary by income: in 1989, 65 percent of unmarried young adults from the lowest quartile of family income had earned a diploma or GED, compared with 93 percent of those from the highest quartile. The detrimental effects of low income are most powerful for racial and ethnic minorities: among Hispanics, for example, only 39.6 percent of young adults from the lowest income quartile had a diploma or GED, compared with 85 percent from the highest quartile (Mortenson and Wu, 1990).

These trends, which are based on individual scores, mask other differences. For example, although the national dropout rate is estimated at 11.2 percent, the average for cities is substantially higher, with many cities losing more than 15 percent of their students (U.S. Department of Education, 1992). Another measure of the system is the frequency of "worst cases" among schools: 25 percent of all poor urban high schools have dropout rates of 50 percent or higher (Braddock and McPartland, 1992). By comparison, less than 1 percent of all other high schools have such extremely high dropout rates. Truancy is another issue for city schools. Data for Maryland, for example, show that more than 67 percent of students in urban high schools are absent for more than 20 days each year, a rate more than twice the state average (Maryland School Performance Program, 1991).

More important, few studies examine the distribution of student performance for individual schools. Thus, little is known about the frequency and location of worst cases or about schools that may be effective in helping students from minority or disadvantaged backgrounds catch up in achievement.

Economic and social stratification influence many key aspects of the educational system. The homogeneous composition of many schools stems directly from neighborhood stratification on the basis of family income, race, and ethnicity. Public expenditures for education, when dependent largely on local wealth, serve to further stratify the educational experiences of adolescents simply on the basis of their family background. Consequently, as noted above, students from low-income families usually attend schools

in poor neighborhoods where they confront conditions not experienced by students from more advantaged backgrounds. These conditions, such as a relative lack of safety and the lowest level curriculum and performance expectations, have independent effects on school achievement. As a result, many students whose lives are rooted in family or neighborhood poverty simply do not have the kind of day-to-day experiences that would stimulate their intellectual development and complement the mission of schools. This web of disadvantage also confronts schools as they seek to improve student achievement, and schools in poor neighborhoods confront these challenges with fewer resources than schools in more affluent neighborhoods.

There are large disparities in expenditures among schools in different states, schools in different districts within states, and schools within individual districts. The amount of available funding is largely dependent on the distribution of wealth in the form of property (Educational Testing Service, 1991). One result is that property-poor districts usually have low expenditures per pupil, even with high tax rates, while property-rich districts usually have high expenditures per pupil, even with relatively low tax rates (Odden, 1991). In Maryland, for instance, the one inner-city district (Baltimore) and five rural districts each spend less than $4,500 per student, compared with three suburban districts that spend $6,000-$7,500 per student. Thus, schools in the poorer Maryland districts have about $45,000-$60,000 less each year for each classroom of 30 students than schools in affluent districts.

In 1991, per pupil expenditures in the 47 largest urban school districts averaged $5,200; in suburban school districts, they were $6,073; in rural districts $5,476. When estimates take into account other measures of student need (such as the number of poor, limited-English-proficient, and disabled students), urban school per pupil expenditures are about 7 percent below the national average, although in a number of urban districts per pupil expenditures exceed the state average (Council of the Great City Schools, 1992). Although an $873 per pupil funding gap may not appear significant, in an average class of 25 students the difference is $21,825—enough to employ a teacher's aide, pay higher salaries, offer special instructional assistance, or improve dilapidated classrooms. Differences in funding of this magnitude could make a clear qualitative difference in the total educational experience.

Despite increases in inflation-adjusted educational spending per pupil during the 1970s and 1980s, inequalities in per pupil expenditures did not change much across the country from the mid-1970s to the mid-1980s. This is because the mechanisms of school finance did not change substantially, and in some cases efforts to equalize disparities were overcome by recessionary and inflationary trends (Odden, 1991; Berne, 1988). The issue is currently being addressed by courts in several states, including California, New Jersey, Texas, and Virginia (Briffault, 1990). Compensatory education funds from federal and state sources are targeted toward low-achieving students, and sometimes to poor schools, but these resources are very limited in comparison with the overall financial disparities (Braddock and McPartland, 1992).

Differences in funding levels may translate into differences in the educational experiences offered to students in low-income and more affluent schools. Per pupil expenditures directly affect the availability of textbooks, laboratory equipment, resource rooms, library books, and a range of other educational resources (Mayer, 1991; Kozol, 1991). In one national survey, for example, in districts with more than one-third of the students from families below the poverty line, 59 percent of fourth grade teachers reported a lack of resources, compared with 16 percent in districts with no students below the poverty line. For eighth grade, the comparable results for math teachers were 40 percent in disadvantaged urban districts and 10 percent in more affluent urban districts (Braddock and McPartland, 1992). These assessments reflect in part the greater demands on urban school districts. A recent analysis, for example, indicated that urban and suburban schools spend about equal percentages (62 percent of the funds available to them) on classroom instruction, but the urban schools spend more per pupil on health care, nutrition, and central office administration, while the suburban schools spend more on student extracurricular activities and maintenance and repair of buildings (Council of the Great City Schools, 1992).

Students in low-income schools tend to have less contact with the best qualified math and science teachers, although this finding holds for many, but not all, measures (Oakes, 1990). Many of the most experienced and qualified teachers use seniority rights to secure assignments in schools with fewer poor or minority students. Salary differences also matter. A recent study of over

900 school districts in Texas showed that the better paying districts were able to attract more qualified (higher test scores) and more experienced teachers. Teacher qualifications and experience, in turn, were the most powerful predictors of differences in student test performance after controlling for family and neighborhood factors (Ferguson, 1991).

Schools also differ on the extent to which parents are involved in school decision making, conferences with teachers, and home-school instructional programs. Over the past decade, studies consistently demonstrate the positive effects of such programs on student achievement, yet parents from low-income neighborhoods, especially racial and ethnic minorities, are least likely to participate. The reasons for this lack of participation include not only the lack of funds, but also different levels of school commitment, cultural and language barriers, and time constraints and stress on working poor families. Even in programs under Chapter 1 of the Elementary and Secondary Education Act (which provides funds to schools with low-income students), which requires poor schools to implement parent-participation programs, the level of involvement remains low (see below). By high school, few low-income or minority parents participate either through traditional or Chapter 1 programs, and school staff often do not encourage them to become involved (Epstein, 1992; Swap, 1990; Comer, 1988).

Differences in financial resources, instructional materials, staff qualifications, and parent involvement ultimately produce marked differences between the climate, norms, and instruction of poor schools and those of more affluent schools (Maeroff, 1988; Kozol, 1991). For example, one study of four urban elementary schools found that pupils in middle-income schools were more likely to be encouraged to be independent and otherwise ''take initiative," while students in low-income schools were rewarded for more passive or deferential behaviors. Academically, the more affluent schools encouraged students to explore ideas, often verbally, while the low-income schools emphasized the retention of facts, often through written drill and practice (Leacock, 1969). Moreover, tracking (discussed later in the chapter) is most rigid in low-income schools, especially those with high proportions of racial and ethnic minorities. Differences also exist in the quality of vocational education between relatively poor and affluent schools (Oakes, 1990; see Chapter 7).

However, evidence regarding the links between per pupil expenditures and student achievement is mixed. In the aggregate, data show little relationship between expenditures and outcomes.

A recent review of 187 past studies concluded that variations in school expenditures are not systematically related to student performance (Hanushek, 1990). Hanushek's review demonstrates the need to conduct studies that focus on the connections between the ways in which funds are expended in a school and student outcomes. If, for example, increased funds simply raise existing teachers' salaries without improving other aspects of the educational environment, such as large class size, or providing better instructional materials, little improvement may be visible. Or it may take time for existing staff to retire before good pay attracts the more highly qualified teachers who would otherwise head for suburban districts (Smith and O'Day, 1991). The Hanushek analysis did not differentiate the results of the small number of studies that examine how increased funding is used from the larger body of studies that do not. This considerably weakens the conclusions we can draw from the analysis, because no distinction is made between schools where higher levels of funding reach the classroom compared with schools where increased levels of spending are absorbed in administration and other nonclassroom expenditures (Slavin, 1987a; Smith and O'Day, 1991; Rosenbaum, 1980).

Students face two major transitions during adolescence—moving from elementary school to a middle grade building and, 2 or 3 years later, moving to a high school. Each transition dramatically changes the education experiences (Boyer, 1983). As they move through the education system, students are faced with many "increases": the size of the school and of the student body, use of competitive motivational strategies, rigor in grading and a focus on normative grading standards, teacher control, and whole-class instruction. These changes can be stressful for adolescents, and data indicate that the experience of transition itself may have an independent negative effect on student attitudes and achievement, especially in large urban schools (Eccles and Midgley, 1989). Transcripts from one urban school district revealed that, among students who ultimately dropped out (35 percent), the most significant declines in performance occurred during the first year of middle school and the first year of high school (Roderick, 1991).

With each transition into a new school, more stratification occurs. Different subject matter is taught by different teachers, and within subject matter, students are often grouped by ability level.

Students placed in the lower "tracks" are at the greatest risk for being retained in grade and vice versa. Both practices are grounded in tradition, and are intended to ensure that instruction is paced at students' ability to learn, and that subjects are mastered before the student advances. An unintended consequence is that students feel stigmatized. They are separated from many of their peers and develop a sense of uncertainty and alienation toward school (Carnegie Council on Adolescent Development, 1989). More significant, both practices involve substantial risks for students' academic achievement.

Ability grouping takes different forms. Usually, students are assigned to one class by ability and move together from subject to subject. Less rigid methods occur when students are assigned by ability to each subject separately. Although it is presumed that different forms of tracking will affect students differently, the little research that has been conducted has not detected consistent differences, and in this discussion, we do not distinguish among them (see Slavin, 1990). Students enter each grade level with a range of competencies, social skills, and behavioral styles. Given this diversity, educators have long questioned the merits of grouping students for instruction on the basis of one or more selected attributes (Dewey, 1938). Traditionally, educators have tended to instruct students in homogeneous groups, though the practice recently has been a focus of much research and debate. Proponents generally assume that low achievers will learn more effectively, and instruction will be more efficient, with students of similar ability. Opponents see tracking as stigmatizing and assert that both social and academic development will be diminished (see Slavin, 1990).²

Ability grouping for academic instruction begins informally during kindergarten and the early elementary years. Grouping decisions, typically for reading and math, are based on teachers' judgments, although research shows that low-income and minority students are often inappropriately tracked to the slower groups (Rist, 1970;

McPartland et al., 1987). In middle and high schools, ability grouping becomes more formalized, based largely on achievement test scores and other measures of prior preparation (see Braddock, 1990a; Slavin, 1990). Tracking is used most rigidly in poorer middle and high schools and in those with sizable proportions of black and Hispanic students (more than 20 percent). Black and Hispanic students are disproportionately represented in remedial tracks both because the schools they attend use tracking systems extensively and because they are likely to score lower on academic achievement tests (Goodlad, 1984; Braddock, 1990b; Oakes, 1990).

Research into the effects of ability grouping on achievement is usually based on one of two approaches, and they produce different results. Comparisons among students in homogeneous and heterogeneous classes have consistently found that ability grouping has little or no impact on overall student achievement (Good and Marshall, 1984; Kulik and Kulik, 1987). In contrast, comparisons between students in different tracks have found that high-track assignment tends to accelerate achievement somewhat while low-track assignment significantly reduces achievement, even controlling for factors such as socioeconomic status (Alexander et al., 1978; Gamoran and Mare, 1989; Oakes, 1982). A recent best-evidence synthesis of the broad literature of ability grouping and student achievement concurs with these assessments, although there may be some differential effects for middle and high schools, and in some classes, such as social studies. The author concludes that "decisions about whether or not to group by ability must be made on bases other than likely impacts on achievement" (Slavin, 1990:17).

Ability grouping has not been shown to improve learning among low-achieving students. Further, the social consequences for students placed in low-achievement tracks are unambiguously negative (Rosenbaum, 1980; Good and Marshall, 1984; Oakes, 1990). Research shows that sorting mechanisms may not substantially change student academic performance per se, but rather reinforce, compound, or exacerbate preexisting differences among students in competency and self-perception. Students placed in lower tracks rarely move into higher tracks (Wheelock, 1992; Dreeben, 1968; Oakes, 1990). The inferior quality of instruction and learning environments in the lower tracks is the principal reason why students seldom emerge from lower tracks into more advanced programs. In addition, instruction in those tracks emphasizes basic skills rather than higher-order learning, and students are less likely to be on-task. In higher grades, as tracks become more fixed and

students more clearly labeled, tracking effectively sorts students according to their future educational and career options. Students become keenly aware of their reduced opportunities, thus contributing to their loss of academic interest and motivation (Braddock, 1990a, b; Gamoran, 1987; Slavin, 1987b; Oakes, 1985; Oakes, 1990; Trimble and Sinclair, 1987).

Grade retention has traditionally been used to ensure that students do not advance unless they have specific skills and basic competencies (such as reading comprehension) to function at the next grade level. The objectives of grade retention are to ensure the integrity of the academic process and to protect students from compounding failure by advancing without mastering prerequisite skills for sequential subjects such as mathematics. However, grade retention has not been uniformly applied, and during the 1960s and 1970s "social promotions" were relatively common. During the 1980s, however, schools responded to calls for higher standards by establishing core curriculums and by requiring students to enroll in more core academic courses.

Although there is little disagreement among educators and others on the need for higher standards, there is concern that higher standards will result in the increasing use of grade retention, and that retention has not been shown to be effective in bringing low-achieving students up to a higher level of performance (Labaree, 1984; Hamilton, 1986). When social promotions fell into disfavor in the late 1970s, the academic competence of high school graduates may have improved, but the performance of low-achieving students did not. As a result, 35.3 percent of all 13-year-old males had been retained once in 1988, compared with 24.1 percent in 1976. Females are less likely to be retained, but by 1988 their rate was 24.8 percent. At all ages, the likelihood of grade retention is far greater for minorities. For example, by age 17, 14.3 percent of all black males have been retained twice, compared with 12.5 percent of Hispanics and 6.2 percent of whites. Among females, the proportions are 14.3 percent for Hispanics, 10.0 percent for blacks, and 3.8 percent for whites (Simons et al., 1991).

The fear of retention is highly stressful for pupils (Smith and Shepard, 1987). Students who are retained often show declines in social adjustment and smaller gains in academic achievement than comparable students who were promoted (Holmes, 1983; Holmes and Matthews, 1984; Reynolds, 1992). Retention also increases

the risks of dropping out of the system before completing high school, by as much as 40 to 50 percent. These findings do not support a return to social promotions as an appropriate response to low-achieving students. However, the policy of grade retention has not proved successful in motivating and assisting students so that they are not alienated from the educational process. Chapter 10 reviews approaches that appear to be more effective in raising student performance levels (Hess, 1987; Grissom and Shepard, 1989; Bachman et al., 1971).

Of the many federal, state, and local programs targeted to low-achieving students, this section discusses Chapter 1 of the Elementary and Secondary Education Act and a range of programs that fall under the rubric of dropout prevention. Both seek to prevent school failure. Chapter 1 programs concentrate on elementary and middle schools and provide funds for academic remediation on the assumption that acquiring basic skills will allow low-achieving students to move more successfully through the education system. Dropout prevention programs, in contrast, have traditionally focused on older students, and although academic remediation is one of many objectives, their main goal is simply to motivate at-risk students to stay in school and earn a diploma.

Chapter 1 (previously Title I), which has existed since 1965, represents the federal government's largest commitment to education for low-achieving students who live in poor neighborhoods. Since 1988, funding has increased 24.7 percent to a level of $6.2 billion (Jennings, 1991). Nationally, the program serves about 14,000 schools. Chapter 1 is also the most extensively researched education program in the country (see Birman et al., 1987; Kennedy et al., 1986a; Turnbull et al., 1990; Stringfield, 1991).

Although its primary focus is low-income students, Chapter 1 serves a range of students of mixed family incomes. For example, about 70 percent of all school districts receive some Chapter 1 funding, and an estimated 58 percent of students who receive Chapter 1 services are not themselves from poor families (Kennedy et al., 1986b). About 23 percent of funds are directed to middle grade students and 10 percent to high school students, with the

remainder being directed to elementary schools (Kennedy et al., 1986b; Birman et al., 1987). This pattern reflects an assumption that early intervention is the most beneficial and efficient way to assist disadvantaged students. Educators also hesitate to implement Chapter 1 in secondary schools because of scheduling difficulties, fear that older students will choose not to participate, and the belief that intervention is too late for high school students (Birman et al., 1987; Zeldin et al., 1991). In fact, the program is moving even further away from directly serving low-achieving adolescents: the current emphasis is on preschool and kindergarten students (LeTendre, 1991).

On a daily basis, Chapter 1 constitutes only a minor intervention designed to complement, not supplant, the regular curriculum. Among middle schools and high schools that offer Chapter 1 programs, 70 percent offer reading and math, in addition to language arts or English as a second language. The remaining schools offer either reading or math only. Approximately two-thirds of all secondary schools use a pull-out or tracking model, in which students leave their regular classroom to receive small-group instruction for 10 to 40 minutes (Birman et al., 1987). Thus, Chapter 1 students do not receive substantially more instructional time in each subject (approximately 10 minutes a day) than other students (LeTendre, 1991).

Evaluations of Chapter 1 programs are mixed. Through Chapter 1 and similar efforts, poor and minority children have made notable gains: school dropout rates have decreased, and graduation rates and mastery of rudimentary skills have substantially improved. National evaluations find improvement for students relative to other "needy" students, but the gains are unimpressive compared with other students in the regular curriculum (Kennedy et al., 1986a, b). The 50 percent reduction in the achievement gap between poor and minority children from other Americans over the past 15 years is thought to be the result of Chapter 1 programs (Commission on Chapter 1, 1992:3).

Chapter 1 is often criticized for its dependence on ability grouping and traditional forms of academic remediation. Other criticisms include the lack of integration with the regular curriculum and a lack of parent involvement (Birman et al., 1987; Commission on Chapter 1, 1992).

The 1988 amendments to Chapter 1 provide educators with new options for designing and implementing programs, including schoolwide programs and instructional approaches that allow for less homogeneous groupings of students. Yet, because of the structure

of the program as well as the status quo orientation of most educators, districts and schools have been slow to implement these reforms or change the basic assumptions underlying the Chapter 1 program (LeTendre, 1991; Turnbull et al., 1990; Plunkett, 1991).

The conclusions of a recent blue-ribbon panel review of Chapter 1 (Commission on Chapter 1, 1992:9) echos these assessments and observes that "the core problem with Chapter 1 is even more basic: its add on design, wherein eligible students get extra help to succeed in the regular school program, cannot work when the regular school program is seriously deficient … if Chapter 1 is to help children in poverty to attain both basic and high level knowledge and skills, it must become a vehicle for improving whole schools serving concentrations of poor children."

However, scholars note that changes will be difficult to achieve. As Slavin (1991:592) observes:

In the early 1980s, national attention focused on the problem of school dropouts and brought it to the fore of education policy concerns (Mann, 1986; Finn, 1989; Wehlage and Rutter, 1986). The response was a rapid increase in dropout prevention programs, many of which were enacted with categorical federal and state funds, often on a demonstration basis (Higgins and Mueller, 1988). Initially, the dropout problem was conceptualized as a risk facing older adolescents, and hence most programs were implemented in high schools; there were at least 1,000 programs by 1987, most of which had been in operation for less than 4 years (U.S. General Accounting Office, 1987). Currently, the dropout problem is considered a school issue, not an individual issue.

Studies of the characteristics of dropouts have found that they are more likely to be from poor families, living in single-parent households, have parents who do not participate in decision making for adolescent problems, and live in urban areas. Dropping

out is also associated with having a handicapping condition, engaging in delinquent behaviors, being retained in grade, being truant from school, being pregnant or a parent, having poor grades, and working more than 15 hours per week. Students who reenter school after dropping out are most likely to be white and to have had better grades and test scores before dropping out than those who do not reenter and had dropped out later in their high school careers (Ekstrom et al., 1986; Kolstad and Owings, 1986; U.S. General Accounting Office, 1987).

The early dropout prevention programs were designed in a manner consistent with these findings. Almost all dropout prevention programs were designed to target specific students for special services aimed at improving academic performance, changing student attitudes, and reducing absenteeism (U.S. General Accounting Office, 1987). Unfortunately, few studies assess the overall effectiveness of dropout prevention models (Grant Foundation, 1991), and the research is inconclusive (Massachusetts Advocacy Center and Center for Early Adolescence, 1988:47):

Educators and researchers have reexamined assumptions regarding dropout prevention. The current consensus is that dropout prevention needs to begin at least by middle school and, ideally, in elementary school or even preschool (see Higgins and Mueller, 1988), but the empirical basis for such judgments is not firm (Ramey and Campbell, 1987; Rohwer, 1971; Zigler and Berman, 1983). A second point of consensus is that effective dropout prevention does not occur through categorical or add-on services, but through the use of schoolwide alternatives to tracking, grade retention, suspension, and expulsion (see Massachusetts Advocacy Center and Center for Early Adolescence, 1988). Indeed, the strategy of preventing dropouts by improving schools has become a practical necessity: with an estimated one-fourth of all urban schools having dropout rates around 50 percent, it is no longer feasible to

provide individualized service, and decisions for targeting those most in need become almost arbitrary.

According to this new consensus, a primary aim of dropout prevention programs should be to implement schoolwide practices that will reduce students' alienation from the educational process and facilitate their interest in the learning process (Wehlage et al., 1989). Some studies suggest that the choice to leave school is tied to the perceived need to move quickly into adult status. After poor grades, the most frequently cited reasons for withdrawing from school are that "school is not for [me]," or because students want to seek and obtain employment or become a parent (Grant Foundation, 1988; Bishop, 1989). Staying in school is even more difficult for minority adolescents who must overcome peer pressures not to excel, and overt or subtle messages communicated by the school curriculum and teachers that excellence is not expected (Felice, 1981; Fordham and Ogbu, 1986; Pine and Hilliard, 1990). Until schools address these perceptions and reduce alienation, it is unlikely that dropout prevention programs will be effective, or that high-risk students will be motivated to learn (Hamilton, 1990:123):

Because of class and residential stratification, students from poor families usually receive their education in the poorest schools. These schools have fewer financial and material resources, and they are often unable to retain the most skilled administrators and teachers. Student achievement levels in these schools are significantly lower on virtually all measures than for students in suburban schools.

Traditional educational practices negatively affect the schooling

of low-achieving students. Historically, schools have addressed the diversity of student achievement by tracking students into homogeneous ability groups and by retaining students who fail courses because of poor attendance, grades, or test scores. These practices have not demonstrated the expected benefits for low-achieving adolescents, and a range of studies show negative academic and social consequences, such as exacerbating existing academic or behavioral problems.

Compensatory education funds from federal and state sources are targeted toward disadvantaged and low-achieving students, but they have shown limited success, particularly among older adolescents. Dropout prevention programs for older adolescents are less effective when implemented as remedial or vocational add-ons to the regular curriculum. It has become apparent that the roots of poor achievement lie not only in the condition of poverty or in individual differences, but also in the use of instructional practices such as tracking and grade retention, and the generally lower achievement expectations for adolescents in concentrated poverty schools.

Alternatives do exist, and a research base is gradually developing to identify promising approaches. However, there are few clear solutions to the problem of how to teach students who lack prerequisite knowledge in courses such as mathematics. Chapter 10 describes some of the approaches that appear promising. The application of these alternative approaches is, however, very limited. Consensus is building regarding the changes that are needed in Chapter 1; however, these are likely to be difficult to achieve politically.

Alexander, K.L., M.A. Cook, and E.L. McDill 1978 Curriculum tracking and educational stratification. American Sociological Review 43:47-66.

Bachman, J.G., S. Green, and I.D. Wirtanen 1971 Dropping out—problem or symptom. In Youth in Transition, Vol. III. Ann Arbor: Institute for Social Research, University of Michigan.

Barro, S.M., and A. Kolstad 1987 Who Drops Out of High School? Findings from High School and Beyond. Washington, D.C.: Center for Education Statistics, U.S. Department of Education.

Bean, F.D., and M. Tienda 1987 The Hispanic Population of the United States. New York: Russell Sage Foundation.

Berne, R. 1988 Equity issues in school finance. Journal of Education Finance 14:159-180.

Birman, B.F., M.E. Orland, R.K. Jung, et al. 1987 The Current Operation of the Chapter 1 Program. Final Report from the National Assessment of Chapter 1. Office of Educational Research and Improvement, U.S. Department of Education, Washington, D.C.

Bishop, J.H. 1989 Why the apathy in American high schools? Educational Researcher 18:6-10, 42.

Boyer, E.L. 1983 High School: A Report on Secondary Education in America. New York: Harper & Row.

Braddock, J.H. 1990a Tracking the middle grades: national patterns of grouping for instruction. Phi Delta Kappan 71(6):445-449.

1990b Tracking: Implications for Student Race-Ethnic Subgroups. Baltimore, Md.: Center for Research on Effective Schooling for Disadvantaged Students, Johns Hopkins University.

Braddock, J.H., and J. McPartland 1992 Education of At-Risk Youth: Recent Trends, Current Status, and Future Needs. Commissioned paper for the Panel on High-Risk Youth, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C.

Briffault, R. 1990 Our localism. Columbia Law Review 90:1-115, 346-454.

Carnegie Council on Adolescent Development of the Carnegie Corporation 1989 Turning Points: Preparing American Youth for the 21st Century. New York: Carnegie Council on Adolescent Development of the Carnegie Corporation.

Carter, D.J., and R. Wilson 1991 Minorities in Higher Education: Ninth Annual Status Report. Washington, D.C.: American Council on Education.

Comer, J. 1988 Educating poor minority children. Scientific American 259(5):42-48.

Commission on Chapter 1 1992 Making Schools Work for Children in Poverty: Summary. Washington, D.C.: American Association for Higher Education.

Council of the Great City Schools 1992 National Urban Education Goals: Baseline Indicators, 1990-91. Washington, D.C.: Council of the Great City Schools.

De La Rosa, D., and C.E. Maw 1990 Hispanic Education: A Statistical Portrait, 1990. Washington, D.C.: National Council of La Raza.

Dewey, J. 1938 Experience and Education. New York: Collier.

Dreeben, R. 1968 On What Is Learned in School. Reading, Mass.: Addison-Wesley.

Eccles, J.S., and C. Midgley 1989 Stage—environment fit: developmentally appropriate classrooms for young adolescents. Pp. 139-186 in C. Ames and R. Ames, eds., Research on Motivation in Education, Volume 3: Goals and Cognitions. New York: Academic Press.

Educational Testing Service 1991 The State of Inequality. Princeton, N.J.: Policy Information Center, Educational Testing Service.

Ekstrom, R.B., M.E. Goertz, J.M. Pollack, and D.A. Rock 1986 Who drops out of high school and why? Findings from a national study. Teachers College Record 87(3):357-373.

Entwisle, D.R., and K.L. Alexander 1992 Summer setback: race, poverty, school composition, and mathematics achievement in the first two years of school. American Sociological Review 57:72-84.

Epstein, J. 1992 School and family partnerships. Pp. 1139-1151 in M. Alkin, ed., Encyclopedia of Education Research, 6th ed. New York: Macmillan.

Felice, L.G. 1981 Black student dropout behavior: disengagement from school, rejection, and racial discrimination. Journal of Negro Education 50(4):415-424.

Ferguson, R.F. 1991 Paying for public education: new evidence on how and why money matters. Harvard Journal on Legislation 28(2):465-498.

Finn, J.D. 1989 Withdrawing from school. Review of Educational Research 59(2):117-142.

Fordham, S., and J.U. Ogbu 1986 Black students school success: coping with the burden of ''acting white." Urban Review 18(3):176-206.

Gamoran, A. 1987 The stratification of high school learning opportunities. Sociology of Education 60:135-155.

Gamoran, A., and R.D. Mare 1989 Secondary school tracking and educational inequality: compensation, reinforcement, or neutrality? American Journal of Sociology 94:1146-1183.

Good, T.L., and S. Marshall 1984 Do students learn more in heterogeneous or homogeneous groups? In P.L. Peterson, L.C. Wilkinson, and M. Hallinan, eds., The Social context of Instruction: Group Organization and Group Process. New York: Academic Press.

Goodlad, J.I. 1984 A Place Called School New York: McGraw-Hill.

Grant Foundation (The William T.), and the Institute for Educational Leadership 1988 The Forgotten Half: Pathways to Success for America's Young and Young Families. Washington, D.C.: William T. Grant Foundation and the Institute for Educational Leadership.

1991 Voices from the Field: 30 Expert Opinions on America 2000, the Bush Administration Strategy to "Reinvent" America's Schools. Washington, D.C.: William T. Grant Foundation, Commission on Work, Family and Citizenship and the Institute for Educational Leadership.

Grissom, J.B., and L.A. Shepard 1989 Repeating and dropping out of school. Pp. 334-363 in L.A. Shepard and M.L. Smith, eds., Flunking Grades: Research and Policies on Retention. Philadelphia: Falmer Press.

Hamilton, S.F. 1986 Raising standards and reducing dropout rates. Teachers College Record 87(3):411-429.

1990 Apprenticeship for Adulthood: Preparing Youth for the Future. New York: The Free Press.

Hanushek, E.A. 1990 The Impact of Differential Expenditures on School Performance . Washington, D.C.: Issue Analysis, American Legislative Exchange Council.

Hess, G.A. 1987 Schools for Early Failure: The Elementary Years and Dropout Rates in Chicago. Chicago: Chicago Panel on Public School Finances.

Higgins, P.S., and D.P. Mueller 1988 The Prevention of Poor School Performance and School Failure: A Literature Review. Unpublished manuscript, Amherst H. Wilder Foundation Prevention Planning Project, St. Paul, Minn.

Holmes, C.T. 1983 The fourth R: retention. Journal of Research and Development in Education 17(1):1-6.

Holmes, C.T., and K.M. Matthews 1984 The effects of nonpromotion on elementary and junior high school pupils: a meta-analysis. Review of Educational Research 54(2):225-236.

Jennings, J.F. 1991 Chapter 1: a view from Congress. Educational Evaluation and Policy Analysis 13(4):335-338.

Kennedy, M.M., B.F. Birman, R.E. Demaline, et al. 1986a The Effectiveness of Chapter 1 Services: Second Interim Report from the National Assessment of Chapter 1. Office of Educational Research and Improvement, U.S. Department of Education, Washington, D.C.

Kennedy, M.M., R.K. Jung, and M.E. Orland 1986b Poverty, Achievement, and the Distribution of Compensatory Education Services. Office of Educational Research and Improvement, U.S. Department of Education, Washington, D.C.

Kolstad, A.J., and J.A. Owings 1986 High School Dropouts Who Change Their Minds About School. Paper presented at the American Educational Association, San Francisco, Calif.

Kozol, J. 1991 Savage Inequalities: Children in America's Schools. New York: Crown Publishers.

Kulik, J.A., and C.L. Kulik 1987 Effects of ability grouping on student achievement. Equity and Excellence 23:22-30.

Labaree, D.F. 1984 Setting the standard: alternative policies for student promotion. Harvard Educational Review 54(1):67-87.

Leacock, E.B. 1969 Teaching and Learning in City Schools: A Comparative Study. New York: Basic Books.

LeTendre, M.J. 1991 Improving Chapter 1 programs: we can do better. Phi Delta Kappan 72(8):576-580.

Maeroff, G.I. 1988 Withered hopes, stillborn dreams: the dismal panorama of urban schools. Phi Delta Kappan 69(9):632-638.

Mann, D. 1986 Can we help dropouts—thinking about the undoable. Teachers College Record 87(3):307-323.

Maryland School Performance Program 1991 School System and Schools Reports. Baltimore, Md.: Maryland State Department of Education and Baltimore City Public Schools.

Massachusetts Advocacy Center and Center for Early Adolescence 1988 Before It's Too Late: Dropout Prevention in the Middle Grades. Massachusetts Advocacy Center, Boston, Mass., and the Center for Early Adolescence, Carrboro, N.C.

Mayer, S. 1991 The Effect of School and Neighborhood Social Mix on Adolescents' Transition to Adulthood. Commissioned paper for Panel on High-Risk Youth, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C.

McPartland, J.M., J.R. Coldiron, and J.H. Braddock 1987 School Structures and Classroom Practices in Elementary, Middle, and Secondary Schools. Baltimore, Md.: Center for Research on Elementary and Middle Schools, Johns Hopkins University.

Mortenson, T., and Z. Wu 1990 High School Graduation and College Participation of Young Adults by Family Income Backgrounds 1970 to 1989. Iowa City, Iowa: The American College Testing Program.

Mullis, I.V.S., and L.B. Jenkins 1990 The Reading Report Card, 1971-88: Trends from the Nation's Report Card. Princeton, N.J.: Educational Testing Service.

National Center for Education Statistics 1990 Dropout Rates in the United States, 1989. Washington, D.C.: U.S. Department of Education.

1991 Conditions of Education. Washington, D.C.: U.S. Department of Education.

Oakes, J. 1982 The reproduction of inequity: the content of secondary school tracking. Urban Review 14:107-120.

1985 Keeping Track: How Schools Structure Inequality. New Haven, Conn.: Yale University Press.

1990 Multiplying Inequalities: The Effects of Race, Social Class, and Tracking on Opportunities to Learn Math and Science. Santa Monica, Calif.: The RAND Corporation.

Odden, A. 1991 School finance in the 1990s. Phi Delta Kappan 73(6):455-461.

Pine, G.J., and A.G. Hilliard, III 1990 Rx for racism: Imperatives for America's schools. Phi Delta Kappan 71(8):593-600.

Plunkett, V.R.L. 1991 The states' role in improving compensatory education: analysis of current trends and suggestions for the future. Educational Evaluation and Policy Analysis 13(4):339-344.

Ramey, C.T., and F.A. Campbell 1987 The Carolina Abcedarian project: an educational experiment concerning human malleability. Pp. 127-139 in J.J. Gallagher and C.T. Ramey, eds., The Malleability of Children. Baltimore, Md.: Brookes Publishing Co.

Reynolds, A.J. 1992 Grade retention and school adjustment: an explanatory analysis. Educational Evaluation and Policy Analysis 14(2):101-121.

Rist, R.C. 1970 Student social class and teacher expectations: the self-fulfilling prophecy in ghetto education. Harvard Educational Review 40:411-451.

Roderick, M. 1991 The Path to Dropping Out: Middle School and Early High School Experiences. Working paper #H-900-13. John F. Kennedy School of Government, Harvard University.

Rohwer, W.D., Jr. 1971 Prime time for education: early childhood or adolescence? Harvard Educational Review 41(3):316-341.

Rosenbaum, J.E. 1980 Social implications of educational grouping. In D.C. Berliner, ed., Review of Research in Education, Vol. 8. Washington, D.C.: American Educational Research Association.

Simons, J.M., B. Finley, and A. Yang 1991 The Adolescent and Young Adult Fact Books. Washington, D.C.: Children's Defense Fund.

Slavin, R.E. 1987a An alternative to meta-analytic and traditional reviews: a best-evidence synthesis. Review of Educational Research 15(9):5-11.

1987b Ability grouping and student achievement in elementary schools: a best-evidence synthesis. Review of Educational Research 57:293-336.

1990 Achievement Effects of Ability Grouping in Secondary Schools: A Best-Evidence Synthesis. National Center on Effective Secondary Schools, University of Wisconsin, Madison.

1991 Chapter 1: vision for the next quarter century. Phi Delta Kappan 72(8):586-592.

Smith, M.L., and L.A. Shepard 1987 What doesn't work: explaining policies of retention in the early grades. Phi Delta Kappan 69(2):129-134.

Smith, M.S., and J. O'Day 1991 Educational equality: 1966 and now. In D.A. Vestegen and J.G. Ward, eds., Spheres of Justice in Education—The 1990 American Education Finance Association Yearbook. New York: Harper Business.

Stringfield, S. 1991 Introduction to the special issue on Chapter 1 policy and evaluation. Educational Evaluation and Policy Analysis 13(4):325-327.

Swap, S.McA. 1990 Parent Involvement and Success for All Children: What We Know Now. Boston: Institute for Responsive Education.

Trimble, K.D., and R.L. Sinclair 1987 On the wrong track: ability grouping and the threat to equity. Equity and Excellence 23:15-21.

Turnbull, B.J., S. Zeldin, and T. Cain 1990 State Administration of the Amended Chapter 1 Program. U.S. Department of Education, Washington, D.C.

U.S. Department of Education 1992 Dropout Rates in the United States, 1991. Washington, D.C.: Office of Educational Research and Improvement, U.S. Department of Education.

U.S. General Accounting Office 1987 School Dropouts: Survey of Local Programs. Washington, D.C.: U.S. General Accounting Office.

Wehlage, G.G., and R.A. Rutter 1986 Dropping out: how much do schools contribute to the problem. Teachers College Record 87(3):374-392.

Wehlage, G.G., R.A. Rutter, G.A. Smith, N. Lesko, and R.R. Fernandez 1989 Reducing the Risk: Schools as Communities of Support. New York: Falmer Press.

Wheelock, A. 1992 Crossing the Tracks. New York: The Free Press.

Zeldin, S., M.C. Rubenstein, J. Bogart, et al. 1991 Chapter 1 Beyond the Elementary Grades: A Report on Project Design and Instruction. Policy Studies Associates, Inc., Washington, D.C.

Zigler, E., and W. Berman 1983 Discerning the future of early childhood intervention. American Psychologist 38:894-906.