Adults as mentors or advocates. A widely publicized approach to providing students at risk with the support of a caring adult during the middle and high school grades is to use volunteers from the community as mentors or advocates (Flaxman, Ascher, and Harrington, 1988; Freedman, 1988, 1991). Although terminology differs in various programs, mentoring is commonly defined as a one-to-one relationship between an adult volunteer and a student who needs support for achieving academic or personal achievement goals. Advocacy is usually defined as a continuing set of relationships between an adult (volunteer or paid) and members of a group of students, in which the adult provides support and services by intervening on a student's behalf, monitoring participation in programs, or brokering additional services (McPartland and Nettles, 1991).
Research indicates that using outside adults as mentors or advocates can have modest positive effects on a limited range of student outcomes. A well-designed program may help some students at risk develop more positive attitudes toward school and related behaviors such as good attendance. However, it seems too much to expect these sorts of programs to achieve reliable short-term effects (i.e., less than one academic year) on student academic achievement as measured by standardized tests, or to see dramatic turnarounds by students who have been failing courses or have serious discipline problems. The successful mentoring relationship usually requires continuing contacts (such as weekly face-to-face sessions) and mutually satisfying activities with some discussion of serious issues. Unfortunately, mentoring and advocacy programs have great difficulty in locating and training large numbers of adult volunteers who have the time and commitment to sustain working mentor relationships with needy students (McPartland and Nettles, 1991; Freedman, 1988, 1991).
Having staff also serve in a mentor or advisor role for middle or high school students overcomes the need to depend upon outside volunteers and can increase the value of in- school relationships. However, resource and scheduling issues are not easily resolved. A frequent approach is to establish a homeroom-advisory period that meets several times each week to discuss a variety of school, character, and career topics in a group setting, with time sometimes allocated for individual adult-student advisory conferences. A few impressive case studies suggest that homeroom-advisory functions may be the keys to creating positive climates in middle or high schools (Maeroff, 1990; Lipsitz, 1984; Lightfoot, 1983). Recent analyses of data from a national survey of middle schools, however, provide no clear evidence that these interventions result in positive student perceptions of teacher-student relations, suggesting that the typical homeroom-advisory period today may be similar to the traditional superficial homeroom period, providing few new opportunities for contacts between individual students and caring adults at the school (McPartland, 1992). Obtaining the desired effects may require that more regular class time be set aside for useful one-on- one adult-student discussions and that more training be given to staff on how to create relationships of trust and support as part of the advisory function.
Race-sex role models. The multiple, serious problems of the young African- American male in American society (e.g., high rates of school failure, imprisonment, unemployment, and crime victimization) are well-documented (U.S. House of Representatives Select Committee on Children, Youth, and Families, 1989; Gibbs, 1988). Such problems have prompted the development of various approaches to providing more positive role models for school-aged African-American male students (Ascher, 1991). These include African-American male classroom teachers for elementary grade classrooms of black male students, mentoring programs using African-American male adults from the community, and peer tutoring approaches using older students to help young students of the same race and sex. These approaches have been widely reported in the mass media (Cooper, 1990; Holland, 1987; Lawton, 1990; Merwin, 1990; Southern Education Foundation, 1990; Tifft, 1990), but no careful research has assessed their impact on students. The important questions raised in Ascher's (1991) descriptive review of 22 programs for African-American male students do represent a start in this direction.
In spite of the lack of evaluation, there is reason to believe that adding positive race- sex role models may be particularly important to school-aged African-American males (Fordham and Ogbu, 1986). However, such programs are the object of controversy on civil rights grounds. Because they often purposely segregate students by race and/or sex, they are sometimes charged with violation of federal and state statutes that prohibit single-sex, single- race enrollment standards. As a result, programs of this type may be forced either to modify such criteria or discontinue operation. Further, these programs have generated considerable controversy among veteran civil rights organizations because of claims they may lead to tensions between African-American males and females and may provide support in the larger society for white supremacy advocates (NAACP Legal Defense Fund, 1991, cited in Ascher, 1991, p. 14).
Other sex and ethnic groups -- African-American females (Butler, 1987), females of all race-ethnic groups (American Association of University Women Education Foundation, 1992), and Hispanic youth of both sexes (Pease-Alvarez and Kenji, 1992) -- have also been the subject of special projects to enhance their chances of school success through the use of role models and other program improvements.
Peer support. A student's peer group will almost certainly be a powerful influence on attitudes and behaviors in school, since status and acceptance from others in the same age group become very important through early adolescence and young adulthood. But the peer influence can be either positive or negative with regard to the school's goals of hard work on classroom learning tasks, depending upon the norms that develop within the various friendship groups to which a student may be attached (Epstein and Karweit, 1983). Indeed, a negative peer influence on school work is often likely to develop because of the pressures against raising the curve in the competition for good grades (Coleman, 1961), tendencies to resist adult authority and goals during adolescence (Elder, 1968), and pressures on young people of color to keep from appearing to adopt the school priorities of the white majority (Fordham and Ogbu, 1986). Several approaches have been encouraged to provide positive role models in schools and to structure classroom tasks and rewards that encourage peer support for academic efforts.
One strategy to assist students during the transition between elementary and middle school or between middle and high school is to pair each entering student with an older student at the school in a peer-mentor relationship that begins the first day of school for the newcomer and lasts throughout the year. In a recent experiment in a racially mixed Baltimore middle school, the older student mentors were trained for their mentor responsibilities -- with communication skills, conflict resolution and community service concepts -- and then were carefully matched with incoming students and scheduled to participate in weekly activities with their mentee, including checks on tardiness and absence patterns, tutoring, and community service projects (Spilman, 1990). In other examples, incoming students with previous attendance or discipline problems are assigned older peer mentors who have been trained to establish and maintain supportive contact throughout the term. Evaluations are not yet available on the impact of these approaches for reducing student problems during transition grades, but the positive evidence cited earlier for other peer-tutoring programs seems relevant.
Cooperative learning is another strategy that uses the peer group to attain academic and prosocial goals. As described earlier, cooperative learning usually involves students working in small teams to accomplish a group goal, such as earning points on classroom tests that count in a classroom competition with other teams. Classroom competition is structured between teams so that each student's individual efforts contribute to a shared group goal, rather than raising the stakes for good grades under the usual classroom competition among individual students. As a result, peer norms are shifted to encourage classroom efforts of individual students rather than discouraging them. An extensive set of careful evaluation experiments confirms the positive effects of these forms of group effort on individual student achievement and peer group acceptance of team members (Slavin, 1990).
Tutoring. One-on-one tutoring is a powerful strategy for providing extra help to disadvantaged youth at all levels. With the recruitment of adult volunteers and various peer- tutoring strategies, school systems are able to provide many underachieving students with the type of one-on-one instruction formerly available only to more privileged segments of society (Cohen, Kulik, and Kulik, 1982).
Reviews of peer-tutoring studies that examine same-age and cross-age strategies show that peer tutoring contributes to the achievement of both tutors and tutees (Natriello, McDill, and Pallas, 1990, pp. 89-90). Cohen et al. (1982) conducted a meta-analysis of 65 studies and concluded that peer tutoring has modest positive effects on both the tutors' and tutees' attitudes toward the subjects being taught and their performance in those subject areas, especially when the programs were highly structured. The advantage of highly structured programs in which student tutors are given explicit instructions is also documented by Slavin (1986).
There has been some controversy over the relative effectiveness of tutoring compared with other types of interventions, such as reduced class size, computer-assisted instruction (CAI), and an extended school day. Levin et al. (1984; 1986) find that peer tutoring is the most cost-effective strategy for reading and math achievement, while Niemiec, Blackwell, and Walberg (1986) argue that CAI is the most cost-effective. Wasik and Slavin (1990) conducted a "Best Evidence Synthesis" of five programs using adult tutors to prevent reading failure in the early grades: Reading Recovery, Success for All, Prevention of Learning Disabilities, the Wallach Tutorial Program, and Programmed Tutorial Reading. These programs were shown to have more positive effects on student achievement than reduction of class size and student/adult ratios.
Technology. The potential of technology to transform, even revolutionize, education has been a source of speculation since the advent of computers in the 1960s (Suppes, 1966). Thirty years later, although many observe that the promise of technology has yet to be fulfilled (Mecklenberger, 1990; Foster, 1990), it continues to be viewed as a catalyst for change in schools (Bell and Elmquist, 1992; Braun, 1990; NFIE, 1991). Numerous studies and reports examining how technology is being integrated into classrooms and schools (OTA, 1988, 1989; Sheingold and Hadley, 1990), its impact on student learning (Becker, 1987), and its importance for educational restructuring (Sheingold and Tucker, eds., 1990) identify technology as a key component of the nation's education reform agenda.
Many practitioners also assert that information-age technologies hold particular promise for educating students at risk. A special issue of Electronic Learning (1988) identifies several ways in which technology can help teachers and administrators meet the challenge of educating students at risk: Technology is "empowering and motivating," a self- paced and ever-patient tutor that provides immediate feedback; it enables students to create high-quality products of which they can be proud; it helps train them for a technology-rich workplace; it can be deployed to create more flexible learning environments that accommodate students who learn in different ways, such as students with mental or physical handicaps or children for whom English is a second language. Technology can give teachers and administrators more time to plan and interact one-on-one with students; advanced information systems can provide teachers with ongoing information about student progress, even when students transfer to another school or district; and telecommunications technology can expand educational opportunities, providing rural and other schools with access to courses and real-world data of which they would otherwise be deprived.
Similar claims of technology's potential are echoed throughout the literature, yet, few studies examine its actual effects on learning outcomes for students at risk. Data from a national survey indicate that the most frequently reported effects of computer use on low- ability students are in behavioral and attitudinal areas such as motivation, self-confidence, and self-discipline (Becker, 1986). An analysis of these same data reports that lower ability students are more likely to use computers for developing basic skills in math, reading, and language, and that students in low SES schools and rural schools are more likely to spend more computer time on drill and tutorial programs than those in high SES metropolitan schools.
In an effort to move away from simple drill-and-practice programs for underachievers, the Vanderbilt Learning Technology Center and its Cognition and Technology Group have investigated the potential of interactive videodisk technology to improve learning for children at risk. This ongoing research is grounded in the knowledge base of cognition and child development and identifies active engagement and the need for a learning context accessible to the child as essential to successful learning (Johnson, 1992). Empirical findings from this research show improved comprehension and ability to make inferences when information is presented to the students through videodisk rather than traditional oral format.
The Higher-Order Thinking Skills (HOTS) program also eschews drill and practice in favor of developing problem-solving and conceptual skills. HOTS combines software with special curriculum and instruction strategies to create a stimulating learning environment for students at risk. For 35 minutes each day, students at risk are challenged by trained teachers to think in more sophisticated ways and to develop hypotheses and strategies for solving problems (Pogrow, 1990b). HOTS has not yet been evaluated with an experimental design, using a randomly assigned or even matched control group. However, there are encouraging signs based on comparing HOTS student gains to tabulated national norms (Pogrow, 1988, 1990a).
One barrier to the effective use of technology for students at risk is that students' exposure to technology-rich learning environments may be cut short when they advance or transfer to schools where technology is not effectively deployed. This problem was identified by researchers evaluating the lasting effects of the Apple Classrooms of Tomorrow (ACOT) program in Memphis. While the ACOT program is not specifically geared toward students at risk, the program in Memphis served a 100 percent minority student population in an inner-city elementary school, most of whom were low achievers. Students were given computers on their desks at school and computers to use at home.
Using small, nonrandom experimental and control groups, Ross, Smith, and Morrison (1991) conducted a follow-up study that examined the achievement and adjustment of 7th- grade students at risk in the school year following their participation in the Memphis ACOT program as 5th and 6th graders. This study showed that ACOT students outperformed control students on fall and spring CAT math and reading tests and in keyboarding skills. However, there were no significant differences between former participants and control students on measures such as grades and teacher ratings. The researchers attribute this to the fact that once the ACOT students left the technology-rich, student-centered environment for a school where computer availability was limited, they were unable to transfer the skills they had learned. They conclude that these students "remain at-risk in middle-school grades, despite their positive accomplishments in the ACOT classes" (Ross, Smith, Morrison, 1991, p. 43).
Though there is a paucity of evidence, the programs and studies described above suggest that technology can be used effectively to improve academic achievement for students at risk and, indeed, for all students. However, it is clear that technology costs and that schools with few resources will have difficulty providing their students with equal access to technology-rich learning environments. Moreover, effects will remain limited so long as programs are implemented only in a few classrooms in the school and are not part of schoolwide or districtwide change efforts (David, 1991). Maintaining access to technology throughout the student's school career, integrating technology so that it is available for all kinds of learning, and deploying uses of technology that move away from traditional teaching and learning methods are necessary components of a successful technology strategy for educating students at risk.