Critique of school- and community-based programs.

Wassef, Adel ; Collins, Melissa Lassiter ; Ingham, Denise 等

INTRODUCTION

That emotional distress and behavioral problems are prevalent, especially in high schools, is well known. The lack of consensus as to who should be classified as "at risk" has been reviewed in Part I of this series. It has been found that students in distress are more likely to approach family, friends, and school personnel for assistance rather than mental health professionals. This article reviews the school- and community-based programs advocated in educational, psychological, and psychiatric journals over the past five years. It excludes studies conducted in traditional mental health settings (for example, outcome studies of inpatient programs and residential treatment facilities).

Classification of the Programs and Studies Reviewed

The programs evaluated in this review are classified by location, focus, and format as shown in Figure 1. Table 1 gives the main characteristics of the larger studies and those which utilized expression of emotions and social bonding and support as intervention modalities.

Proposals to Overhaul the Educational System

DeRidder (1991) advocated changes in the school system to assist at-risk students in a nonpunitive way "since additional punishment for students who are already alienated simply creates hostility and does not adequately address the students' real problems." It was hoped that the proposed changes would "help each student by identifying and remedying the factors that contributed to the discipline and academic [TABULAR DATA FOR TABLE 1 OMITTED] problems. They would then provide personalized and caring assistance at a level appropriate to the student's academic and psychological functioning." Preferably, parents would be involved in developing and carrying out each student's individualized program. The specific proposed changes included: (1) separating potential dropouts or "pushouts" (i.e., at-risk students who are encouraged by their teachers to leave school) from other students and placing them in special programs; (2) a strong vocational component, including practical job-related skills; (3) out-of-class learning connected with paid employment; (4) individualized instruction; and (5) more individual and group counseling.

Cuban (1989) recommended separate programs for at-risk students in urban areas, and reducing the school size (between 50 and a few hundred students) and smaller class size (15-20 students per teacher). Cuban proposed housing these programs as schools-within-a-school or completely separate from the main building. Other recommendations included: (1) employing dedicated and accepting teachers and administrators who choose to work with the "at-risk" population; (2) flexibility of the teachers, administrators, and school board members; (3) program flexibility using nontraditional approaches to nontraditional problems (e.g., increasing the population's homogeneity - avoiding "ability grouping" so that the less able do not feel stigmatized), and providing access to a variety of social services; (4) avoiding the conventional school model, where the teacher's primary concern is academic achievement.

Unfortunately, several factors made these two approaches impractical: the magnitude of the problem, high student-to-teacher ratio, limited school budgets, and poor support available to students at home. The societal inclination to integrate rather than separate at-risk students made implementing these recommendations difficult except for those with severe behavioral problems.

Out-of-School Programs and Those Designed for Students Who Have Dropped Out

Behaviorally and psychoeducationally oriented programs or programs focused on specific behaviors. O'Sullivan and Jacobsen (1992) studied the effectiveness of a special health care program on the repeat pregnancy rate and school re-enrollment of unwed African-American adolescent mothers receiving Medicaid. In addition to routine baby care, the program for the experimental group included rigorous follow-up, discussions with the mother about her plans for return to school and the use of family planning methods, and additional extra health education. The control group received routine baby care only. After 18 months, the dropout rate in the experimental group was significantly less (60%) than in the control group (82%). In spite of the high dropout rate, 91% of the mothers were located for the 18-month follow-up interview. Although there was no significant difference in the percentage returning to school between groups, the repeat pregnancy rate after 18 months in the experimental group was 12% as compared with 28% in the control group.

Bernstein and Haessly (1991) advocated a program called Learning, Earning and Parenting (LEAP) aimed at helping teenage mothers who had dropped out and were welfare recipients to go back to school by linking welfare benefits to school attendance. The program "docked teen mothers $62, or 20% or their monthly welfare check if they did not attend school regularly. However, it also paid a $62 bonus to those who miss fewer than four days a month." No details were offered about the success of the program.

Programs involving emotional expression, social bonding, and support. A community program based on family support and educational models of service delivery to pregnant and parenting teenage females used home visiting to reduce isolation, linked the student and her family to community resources, provided home-based information and support, and connected students to appropriate services (see table - Ruch-Ross et al., 1992). The program also used peer support groups to provide information about child health and rearing and emotional support. The groups met weekly with facilitators who, in most cases, were young and had been adolescent mothers themselves. One-year outcomes for 1,004 participants were evaluated for the statewide program, comparing the results with those for 790 adolescent mothers included in the National Longitudinal Survey of Youth. Controlling for baseline, participants were found to be significantly more likely than the national control sample to be enrolled in school, to be employed, and to have avoided a subsequent pregnancy. However, some methodological limitations confound the study: the participants were from 27 different and nonstandardized community-based programs, and the outcome was compared with that of mothers in the National Longitudinal Survey of Youth rather than a matched control group which did not receive the interventions. Also, the outcome measures were limited in scope.

Haggerty et al. (1989) described Project ADAPT which was conducted in a Washington State correctional facility. The project was a 3 1/2-year demonstration treatment program based on the Social Development Model. It dealt with drug-using juvenile delinquents who were incarcerated for different reasons. The investigators hypothesized that providing opportunities for prosocial involvement, building skills to be successful in such activities and interactions, and providing rewards would increase the chances of maintaining rehabilitation once the incarcerated youths returned to the community. The basic philosophy behind the program was that serious and persistent delinquency and the frequent use of illicit drugs emerge from common etiologic roots and that treatment which targets risk factors of adolescent drug use and crime could be effective in preventing subsequent antisocial behavior. The project combined behavioral skill training (an essential component), supportive network development, and involvement in prosocial activities to facilitate reentry into the community after release. No outcome data were provided.

In School Programs

Behaviorally and Psychoeducationally Oriented Programs and Programs Focused on Specific Behaviors. Several school-based programs were designed to have an impact on students' health-related behaviors. A 20-session risk-reduction intervention provided to all tenth graders in four senior high schools improved the rate of exercising and choosing "heart-healthy" snack items and reduced smoking as compared with a control group. Beneficial effects were observed for resting heart rate, body mass index, triceps skin fold thickness, and subscapular skin fold thickness (Killen et al., 1988). A middle school health promotion program administered by college-age "role models" appeared to be effective in changing certain reported behaviors (Moberg & Piper, 1990). While an immediate post-test revealed only inconsistent results, a 20-month follow-up showed clear positive outcomes regarding cigarette smoking and food choices, and ambiguous but positive results on rates of intercourse. Shortcomings of the methodology included nonexperimental assignment to groups and reliance on self-report data only. A school-based multiple risk factor reduction program for tenth graders proved to be effective in altering behaviors known to increase cardiac risk factors (nutrition/diet, physical activity, lack of exercise, and cigarette smoking) two months after the program was offered. Physiologic changes were also observed (e.g., resting heart rate, triceps skin fold thickness, and subscapular skinfold thickness). The program utilized special 20-session cardiovascular disease risk-reduction intervention (Killen et al., 1989).

Two studies focused on violence prevention. The efficacy of a large-scale wife assault and dating violence primary prevention program was evaluated by Jaffe et al. (1992). All students in four high schools in Ontario, Canada attended a brief intervention session, followed by a classroom discussion facilitated by community professionals. Significant positive changes in attitude, knowledge, and behavioral intention were observed on the project postest. Later observations revealed that the majority of the positive changes had been maintained. Females consistently showed better attitudes than did males. Students reported a high level of awareness of, and experience with, violence in their own and their friends' dating and family relationships, and overwhelmingly endorsed primary prevention efforts. A "backlash effect" was noted among a small number of males after the intervention. The authors hypothesized that this group might already have been involved in abusive behavior and required secondary, rather than primary prevention. A social skills training program for African-American adolescents, facilitated by two African-American doctoral-level clinical psychology students, showed some promise as a means of preventing violence (see table - Hammond & Yung, 1991). The skills training focused on giving and accepting positive and negative feedback, resisting peer pressure, solving problems, and negotiating. An incentive system was developed to reward active participation and appropriate behavior. Those who completed the training demonstrated improvement in all target skills. Observers, teachers, and school records confirmed the improvement. None of the 15 participants who completed the training were suspended or expelled from school because of violence. However, the small sample size, the heterogeneity of inclusion criteria, and the undetermined outcome of those who dropped out of the study preclude generalization of the results.

Several programs proved to be effective in preventing certain hazardous behaviors (e.g., substance use) even after several years, especially with persons who had not started to use at the time of enrollment. These programs measured substance use as the outcome criterion. For example, in the North Karelia Youth Program, five to ten classroom sessions over two years taught 13- to-15-year-old students skills for resisting pressure to start smoking. The program resulted in reduction of smoking immediately after the intervention and in a four-year follow-up survey. At the eight-year follow-up, there was consistent evidence of the preventive effect only among those who had been nonsmokers when the program began (Vartiainen et al., 1990). A peer-led intervention program emphasizing the teaching of social resistance skills was implemented in the seventh grade with a booster intervention in the eighth grade. The program reduced tobacco, alcohol, and marijuana use and had a significant impact on several hypothesized mediating variables. Similar effects were noted in females when the prevention program was implemented by classroom teachers (Botvin et al., 1990). Also, a marked suppression in the onset of both experimental and regular smoking was noted in seventh-grade students exposed to the pressure resistance training with peer leader involvement. Pressure resistance training with videotaped instruction but without peer leader involvement produced a more variable and less powerful effect on students' smoking behavior. Alcohol and marijuana use were also diminished, but the effect was weaker than for tobacco among those students exposed to the social resistance training with peers (Telch et al., 1990). Hansen and Graham (1991) noted that teaching skills to refuse drug offers was inferior to "normative education" designed to correct perceptions about the prevalence and acceptability of substance abuse and to establish conservative norms regarding use.

Not all school-based programs aimed at specific behaviors have been successful; for example, a long-term follow-up study of a school-based suicide prevention curriculum on high school students failed to produce convincing evidence of any program effect (Vieland et al., 1991). The program was offered in a regular classroom by teachers who had previously received approximately six hours of training. The study targeted a predominantly white suburban or rural student body and emphasized support networks in alleviating stress, confronting one's peers, and utilizing community resources. This prevention program was, however, brief, lasting only 1 1/2 hours.

In addition to several programs which aimed at improving communication between students and their families, the school system in El Paso, Texas sought the help of some parents to serve as liaisons between the schools and other parents to reduce dropout rate. The liaisons informed parents about available community services (e.g., food banks), and translated for those parents who were reluctant to go to their child's school because they did not speak English (Celis, 1992). The program was not objectively evaluated.

Fertman and Chubb (see table - 1992) described a short psychoeducational personal empowerment program which aimed at increasing participation in community and school activities and enhancing leadership attitudes and skills. The study included 52 ninth-grade students, but only 19 of them were in the experimental group. The program contained a one-week summer workshop which consisted of small- and large-group activities which focused on leadership, communication, assertiveness, stress management and decision making. Follow-up sessions at the end of the school year focused on skill reviews, team building, and goal setting. The program reportedly maintained the participants' level of involvement during the transition to high school, a time when participation generally decreases, but did not change self-esteem or locus of control. The small sample size was an important limitation of the study.

Leo (1991) described a middle school program which included students in activities regardless of their abilities, reportedly to "reduce competition, boost self-esteem, and create school spirit and student bonding." The program was not objectively evaluated.

Several programs have targeted the involvement of at-risk students in constructive activities since students who work were noted as having higher self-esteem, greater family cohesion, and less family conflict than nonworkers (Hardesty & Hirsch, 1992). Paying at-risk middle school students to tutor and mentor elementary school pupils was effectively used by Wilce (see table - 1992) to provide "endangered teenagers some responsibility and something to feel good about and improve their school performance." In addition to four afternoons of tutoring per week, a debriefing session was held every week, but details about this session were not provided by the author. More than 500 middle school students worked with more than 1,500 elementary school pupils. The student tutors showed a lower drop-out rate (1% compared with 12% in the unspecified control group), improved academic performance, fewer disciplinary problems, and more positive attitudes toward themselves and the school, and were noted to be "happier." Two factors were reported to have been crucial for the program's success: the $17 "wage" the student tutors earned for four afternoons of tutoring, and the weekly debriefing sessions. The program was believed to teach work ethics and habits and more importantly, "the students took pride in being able to make a financial contribution to their households, . . . in having their work tangibly rewarded," and in knowing that they played a major role in helping another student to succeed. The limited data reported, the lack of a clearly defined control group, and the experimental design do not allow for a definitive assessment of the program (e.g., the number of students who actually participated in the study and their outcome are hard to determine from the article).

Since disempowered group members achieved personal growth when helping other people (Riessman, 1965), Allen, Philliber, & Hoggson (see table - 1990) encouraged student participation in community volunteer services to boost interactions which might increase identification with the prosocial values of the adults (at least a half hour per week). This allowed them to experience giving rather than just receiving help. Also, the program linked volunteer work with classroom-based discussions of selected topics, group experience, films, and informational presentations held at least once a week. The topics included understanding yourself and your values, communication skills, dealing with family stress, human growth and development, and issues related to parenting. The two major goals of the program were preventing teenage pregnancy and school dropout. Students in the Teen Outreach Program of the Association of Junior Leagues and matching comparison students in 35 schools nationwide participated in a program. The study compared the experimental and control samples as well as the efficacy of the 35 programs (which were conducted differently) to determine the effective elements. Outcome measures were the dropout, school failure, and pregnancy rates. Those sites which highly utilized a volunteer service component and those which primarily served older students reported lower levels of problem behaviors at program exit. When working with younger adolescents, classroom discussions appeared to have a positive effect. Factors which could be related to motivation (e.g., receiving no credit for attending) were related to characteristics of students at entry but not at exit. Gender and ethnicity of the student did not appear to affect the outcome. The study had some limitations: the inclusion and exclusion criteria were not clearly defined and were not standardized across sites, the intervention program itself was not standardized and varied depending on the site, there were differences between the experimental and control groups at baseline, and the statistical analysis did not permit generalization of the results due to lack of random assignment to the experimental groups.

Programs Involving Emotional Expression, Social Bonding and Support

Unsworth (1990) used art to help at-risk or dropout students reveal self-esteem and identify frustrations. According to that author, these could be revealed through art, regardless of the student's natural artistic abilities. No data were provided to permit evaluation of the impact of the program.

Opie et al. (see table - 1992) conducted a small study (eight children and eight junior high school-age adolescents in two separate groups) to assess the effect of eight one-hour weekly groups on students who had experienced the death of a loved one in the previous two years. Attendance was generally low for both groups (mean = 6.4 and 4.6 sessions). Little difference was noted in the affective distress scores before and after the group sessions, but the somatic symptoms decreased in the children's group. The small sample size, brevity of the intervention, and the variable duration between the event (i.e., the death) and participation in the groups were confounding factors.

Korczynski (see table - 1989) studied the efficacy of weekly groups with students selected on the basis of emotional and/or behavioral problems noted by their teachers. The hypothesis was that positive interaction in such groups would increase self-esteem and enhance development. The groups, which consisted of four to six primary school children, met weekly with a trained child psychiatric nurse "for further observation and/or to work on improvement in peer relationships." Data were not provided.

The influence of "a positive one-to-one relationship with a responsible adult" on students' attitude, behavior, and dropout rate was studied by Wilhoit and Roesch (see table - 1989). Their "ABC Program" identified at-risk students through discipline, attendance, and academic records. In addition, students were referred by "parents, peers, teachers, counselors, and the students themselves." A school psychologist, who acted as a full-time program coordinator, trained teachers to serve as ABC facilitators, monitored involvement of all participants, and evaluated the program on a continuing basis. All facilitators received a $275 dollar stipend yearly. The number of students was limited by number of facilitators. As a result, a waiting list was created (but the students on the list were not used as a control group). There were 47 and 54 facilitators available in the first and second years, respectively. The typical length of participation in the program varied from nine weeks to two semesters. Each program facilitator met individually with the student under his or her care for one hour weekly (range = 9 to 36 sessions) to "talk about concerns, and do whatever in their judgment was needed to effect positive attitude and behavior change." Student participation was discontinued when the "facilitator, student, and coordinator agreed that formal meetings were no longer necessary." Exit procedures included a student exit interview, a dismissal contract, and parent notification. Students could also initiate their own dismissal from the program. If the ABC student failed to "show up after the beginning of the program, the facilitator tried to contact the student daily for one week. If the contacts were unsuccessful, the coordinator was notified and made an appropriate follow-up." Outcome measures included the number of students who remained in school, the number of disciplinary referrals, attendance record, and "any observed positive changes in attitude and behavior noted by teachers." The students were noted to have "a more positive attitude toward adults and authority figures," while the teachers reported "a new perception of students who had previously exhibited negative attitudes and behaviors." The program was reported to have "helped three fourths of the students to stay in school." Unfortunately, the study design had some shortcomings: (1) the specifics of the inclusion and exclusion criteria were not elaborated; (2) Students were admitted to the study mainly as a result of behavioral problems and not merely because they suffered from emotional distress; (3) the program duration was variable and was not correlated with the students' response; (4) only scanty outcome data were provided; (5) the data analysis did not explain how students who dropped out were handled; (6) the outcome measures did not systematically assess changes in the students' emotional status as a result of participation; (7) there was no control group data to provide a more meaningful interpretation; (8) no attempt was made to use a diagnostic classification to indicate if the participants qualified for psychiatric disorders; (9) since individual, rather than group support was provided, the number of students participating had to be limited. The study, however, showed that nonmental health professionals could achieve positive outcomes in this population. Upon completion of the first year's program, the authors made three recommendations: (1) middle school guidance referrals should be used to identify at-risk students so the program could start as soon as possible each fall; (2) a facilitator should not be matched with a student enrolled in his or her classes since some students had difficulty distinguishing the role of the facilitator from that of the teacher; (3) "transition, exit, and follow-up procedures should be strengthened by including the exiting ABC student to serve as a big brother or big sister to a student entering the program."

Eggert, Seyl, & Nicholas (see table - 1990) evaluated the efficacy of a school-based network social support program which took the form of a small-group psychoeducational counseling class - Interpersonal Relations (IPR). The program tested the efficacy of a teacher/peer social support and social influence model as well as the appropriateness of school as an environment for such support programs. Cognitive and behavioral changes were prompted through positive reinforcement principles, skills training, and making learning exciting and experimental. This was done "in the context of a positive teacher-student relationship, peer group belonging, and positive peer pressure and support." The investigators used a quasi-experimental design (n = 264). The program included students with known substance abuse problems, students with a chronic absenteeism problem and low grade-point-average, and those who had dropped out of school in the past. Teacher and peer groups met four days a week. In the first 40 minutes, the students discussed their current psychosocial problems in a group format. Skills training in interpersonal communication, problem solving, decision making, and self-management were also addressed. This was followed by 15 minutes of supervised study and peer tutoring. One day a week there was a review of the students' attendance and progress in other classes, goal setting, journal writing, supervised study and tutoring, and planning for alternative drug-free weekend activities. Educational credit was given to those who attended. The program increased the retention of potential dropouts from 61 to 74%. Daily attendance, grade point average, and school achievement were also improved and drug involvement declined. However, it is not possible to implement the program in, or generalize the results to, an average high school for several reasons: (1) daily sessions may not be feasible in standard public schools; (2) the students were from a "homogenous middle class" which precludes generalizing the results to other groups without empirical evidence (Irwin & Vaughan, 1988); (3) differences in response were not linked to demographics; (4) student participation was not solicited, and as a result, most participants had suffered from problems serious enough to be noted by others. In fact, the majority of the students in the study used drugs daily or several times per week. Thus, the results may apply only to students who are known to abuse drugs extensively or those who are failing in school rather than those with moderate problems or who suffer from emotional distress without behavioral problems; (5) the sampling procedure was defective since rules of stratified randomization and true random assignment to the experimental and the control groups were not followed, which may have biased the results. This is evidenced by the significant differences between the experimental and control groups (e.g., none of the students in the control group remained in the same school in either junior high or high school, which could have seriously compromised their support system and limited their ability to seek or receive support from others; although the mean ages of the two groups were similar, students in the control group were in lower school grades possibly because of chronic school problems; the experimental group contained twice as many students in the twelfth grade whom Eggert considered to be possibly more motivated; in addition, members of one of the four IPR groups studied had participated in drug rehabilitation programs which probably biased the results through some carry-over effect or by the choosing of more motivated students for the experimental group; secondary gain (educational credit) confounded the results; an unreported number of students who failed to complete the IPR course were excluded from the analysis which means that only those with high motivation remained in the experimental group while the control group included both the highly motivated and the nonmotivated students (no "Intent to Treat Analysis" was provided); the study did not attempt to establish the long-term efficacy of the program or the effect of ending participation. It did not systematically evaluate the psychological factors (e.g., depression, self-esteem). The study, despite its limitations, provides strong evidence for the potential of this approach as well as useful information for the design of more definitive studies.

Delayed Outcome of Intervention Programs

Only five of the studies discussed here measured the delayed outcome of intervention even though outcome criteria were limited in scope: O'Sullivan and Jacobsen (1992) measured the 18-month rate of pregnancy and return to school. Vartianen et al. (1990) examined the four and eight-year follow-up of a program designed to teach students to resist pressure to smoke. Jaffe et al. (1992) measured attitude, knowledge, and behavioral intention changes related to violence at delayed follow-up, while Moberg and Piper, (1990) measured cigarette consumption, food choices, and intercourse rate 20 months after the intervention. The four studies showed favorable effects. Only the study evaluating the brief intervention focused on suicide prevention in high school failed to find evidence of efficacy on long-term follow-up (Vieland et al., 1991).

CONCLUSIONS

In summary, school management and the current usage of the mental health system have had limited success in managing emotional distress and behavioral problems in high schools. Overhauling the educational system to handle such problems has been proposed, but economic realities and social pressures have prevented any large-scale implementation of these proposals. Several programs have been tried to supplement the current system or remedy its failures. They differed in basic approach (e.g., behavioral techniques, psychosocial support). They also varied in philosophy, from preventative to curative to palliative, and in format from individual to group support to large audience lectures. The aims of these programs also varied (e.g., to prevent specific behaviors such as teen pregnancy, smoking, dropping out of school). The settings included schools, juvenile detention facilities, and social welfare offices. The limited focus of the programs, ignoring the emotional aspects of the participants and the shortcomings of the design of the relevant studies, prevents generalization of the results of intervention to average high school students and do not permit quantification of their utility as early intervention programs. Preliminary evidence indicates that school-based support groups may be useful as early intervention and management programs when applied in the school setting.

Future school programs should be evaluated through studies designed and conducted in a way which predicts their usefulness when applied in a nonresearch setting in a regular school where, for example, highly trained mental health professionals cannot be employed in the program for reasons of cost or acceptability or where the expected frequency of attendance interferes with educational activities. The same is true in deciding on entry criteria for studies: a DSM III-R diagnosis, which requires a mental health professional, is less likely to lead to generalizable results than will entry criteria based on a simple self-rating scale. Self-referral should be allowed because it offers students with emotional distress and early behavioral problems opportunities to enroll in the program. As discussed in Part I of this article, inclusion criteria should be based on expressed emotional distress or behavioral manifestations, not on being categorized as at risk, until consensus is reached about the operational criteria for this term. Considering the magnitude of the problem and the available resources, a group format may be more suitable than individual support. Outcome measures should include students' perceptions of their emotional status, behavioral measures, and school records simultaneously. The design of the studies should allow student participation in the groups throughout high school in order to estimate the optimal duration of participation. Follow-up should continue throughout high school for those students who complete or terminate their participation in the groups in order to allow for evaluation of the residual effects and long-term efficacy of the programs. This is particularly important since such programs will compete with well-established activities for the financial resources of the school and would require evidence of long-term efficacy to earn priority for funding.

The next part of this article will discuss the efficacy and acceptability of a school program based on volunteer-facilitated peer support groups which has been in operation since 1990 in a high school in the Southwest.

This work was supported, in part, by the Dickinson Independent School District. The authors thank all of the volunteers who donated their time as peer support group facilitators, the Dickinson High School Board, the District Superintendent Bill Borgers, Ph.D., The School Principal Ron Ahlhorn, M.S., Louise Bell, Ed.D., and John VanHaalen, M.S. for their effort and support.

REFERENCES

Allen, J. P, Philliber, S, & Hoggson, N. (1990). School-based prevention of teen-age pregnancy and dropout. American Journal of Community Psychology, 18(4), 505-524.

Bernstein, A., & Haessly, L. (1991, October). The ABC's of coaxing teen mothers back to school. Business Week, 96.

Botvin, G. J., Baker, E., Filazzola, A. D., & Botvin, E. M. (1990). A cognitive-behavioral approach to substance abuse prevention: One-year follow-up. Addictive Behaviors, 15(1), 47-63.

Celis, W. (1992, October 14), Hispanic dropout rate stays high since children work in hard times. The New York Times, Education Section, B9.

Cuban, L. (1989, May). What can be done for at-risk students. Education Digest, 3-6.

DeRidder, L. (1991). How suspension and expulsion contribute to dropping out. Education Digest, 56(6), 44-47.

Eggert, L. L., Seyl, C. D., & Nicholas, L. J. (1990). Effects of a school-based prevention program for potential high school dropouts and drug abusers. International Journal of the Addictions, 25(7), 773-801.

Fertman, C. I., & Chubb, N. H. (1992) The effects of a psychoeducational program on adolescents' activity involvement, self-esteem, and locus of control. Adolescence, 27(107), 517-526.

Haggerty, K. P., Wells, E. A., Jenson, J. M., Catalano, R. F., & Hawkins, J. D. (1989). Delinquents and drug use: A model program for community reintegration. Adolescence, 24(94), 439-456.

Hammond, W. R., & Yung, B. R. (1991). Preventing violence in at-risk African-American youth. Journal of Health Care for the Poor and Underserved, 2(3), 359-373.

Hansen, W. B., & Graham, J. W. (1991). Preventing alcohol, marijuana, and cigarette use among adolescents: Peer pressure resistance training versus establishing conservative norms. Preventive Medicine, 20(3), 414-430.

Hardesty, P. H., & Hirsch, B. J. (1992). Summer and school-term youth employment: Ecological and longitudinal analyses. Psychological Reports, 71(2), 595-606.

Irwin, C. E., Jr., & Vaughan, E. (1988). Psychosocial context of adolescent development. Study group report. Journal of Adolescent Health Care, 9(6 Supplement), 11S-19S.

Jaffe, P. G., Sudermann, M., Reitzel, D., & Killip, S. M. (1992). An evaluation of a secondary school primary prevention program on violence in intimate relationships. Violence and Victims, 7(2), 129-146.

Killen, J. D., Robinson, T. N., Telch, M. J., Saylor, K. E., Maron, D. J., Rich, T., & Bryson, S. (1989). The Stanford Adolescent Heart Health Program. Health Education Quarterly, 16(2), 263-283.

Killen, J. D., Telch, M. J., Robinson, T. N., Maccoby, N., Taylor, C. B., & Farquhar, J. W. (1988) Cardiovascular disease risk reduction for tenth graders. A multiple-factor school-based approach. Journal of the American Medical Association 260(12), 1728-1733.

Korczynski, J. (1989). Socialization groups in a school outreach program. Journal of Child and Adolescent Psychiatric and Mental Health Nursing, 2(4), 166-167.

Leo, J. (1991, December). Two cheers for Plainfield. U.S. News & World Report, 19.

Moberg, D. P., & Piper, D. L. (1990). An outcome evaluation of project model health: A middle school health promotion program. Health Education Quarterly, 17(1), 37-51.

Opie, N. D., Goodwin, T., Finke, L. M., Beattey, J. M., Lee, B., & Van Epps, J. (1992). The effect of a bereavement group experience on bereaved children's and adolescents' affective and somatic distress. Journal of Child and Adolescent Psychiatric and Mental Health Nursing, 5(1), 20-26.

O'Sullivan, A. L., & Jacobsen, B. S. (1992). A randomized trial of a health care program for first-time adolescent mothers and their infants. Nursing Research, 41(4), 210-215.

Reissman, F. (1965). The helper-therapy principle. Social Work, 10, 27-32.

Ruch-Ross, H. S., Jones, E. D., & Musick, J. S. (1992). Comparing outcomes in a statewide program for adolescent mothers with outcomes in a national sample. Family Planning Perspectives, 24(2), 66-71, 96.

Telch, M. J., Miller, L. M., Killen, J. D., Cooke, S., & Maccoby, N. (1990). Social influences approach to smoking prevention: The effects of videotape delivery with and without same-age peer leader participation. Addictive Behaviors, 15(1), 21-28.

Unsworth, J. M. (1990, December). Art and the dropout student. School Arts, 14-15.

Vartiainen E., Falleonen, U., McAlister, A. L., & Puska, P. (1990). Eight-year follow-up results of an adolescent smoking prevention program: The North Karelia Youth Project. American Journal of Public Health, 80(1), 78-79.

Vieland, V., Whittle, B., Garland, A., Hicks, R., & Shaffer, D. (1991). The impact of curriculum-based suicide prevention programs for teenagers: An 18-month follow-up. Journal of the American Academy of Child and Adolescent Psychiatry, 30(5), 811-815.

Wilce, H. (1992, June 26). Dropouts drop in. Times Educational Supplement, 24.

Wilhoit, J., & Roesch, S. (1989, May). Attitude behavior change: A dropout prevention program. National Association of Secondary School Principals Bulletin, 130-134.

Melissa Lassiter Collins, B.A., University of Texas Medical Branch, Galveston, Texas.

Denise Ingham, M.D., Harris County Mental Health Mental Retardation Authority, Houston, Texas.

Gayle Mason, B.A., Dickinson High School, Dickinson, Texas.