Student assessment of school-based support groups.

Wassef, Adel ; Mason, Gayle ; Collins, Melissa Lassiter 等

INTRODUCTION

Perceived emotional distress and behavioral problems are quite prevalent among high school students, the reasons for which were discussed in Part I of this series. The difficulties encountered in assessing the efficacy of several programs that were proposed to remedy the situation were discussed in Part II. Briefly, several of those programs that focused on specific behaviors (e.g., substance abuse), required highly trained personnel or daily sessions. These programs often used inclusion criteria which did not allow for the results to be generalized to the students in the average high school who suffer from emotional distress or milder behavioral problems. Subjective emotional distress was not used as an inclusion criterion, and there was no attempt to assess the impact of proactive intervention prior to the appearance of serious behavioral problems. Little attention was paid to measuring the psychological impact of such groups or the students' perception of improvement, two crucial factors which ultimately determine the utilization rate of the programs and their influence on the students' attitude toward school, peers, and family. In-school volunteer-facilitated peer support groups have not been adequately studied as an early intervention and management program despite some hints in the literature of probable efficacy and economic feasibility.

This article reports the authors' experience with a volunteer-facilitated, in-school student-assistance program based on voluntary participation of students in peer support groups. The focus was on the students' assessment of their improvement, and the program's acceptability at the end of the third year of implementation.

METHOD

In the academic year 1990-1991 a proactive student-assistance program based on volunteer-facilitated peer-support groups was implemented in a Southwestern high school. Students aged 14-19, who experienced emotional distress or behavioral problems were recruited by providing information about group availability to all students at the beginning of the year. Because of the anonymity of the responses, the characteristics of the respondents could not be identified. Table 1 shows the demographic characteristics of the catchment area and the students in this school.

Students who perceived that they had emotional distress or behavioral problems and wanted help referred themselves to be evaluated by the school counselor to decide whether help was needed. In addition, any concerned person (peer, teacher, parent) or an external agency could, anonymously submit the name of a student for evaluation. The self-referred students and those referred by others were then invited for an interview to determine the type and degree of help each of them needed. Students who were in immediate danger to self or others and those whose problems were severe enough to require hospitalization or out-of-school help were referred to the appropriate health authority (approximately 10% were so referred). Those whose emotional distress or behavioral problems were significant enough to warrant intervention were invited to participate in weekly small peer groups (8-12 students). The choice of group was based on the reason given for referral. The different groups available are summarized in Table 2. It was noted that once the reasons for referral and the available groups were discussed, there was very little disagreement between the students and counselor about the most appropriate group, but the final decision was to be made by the counselor.

Participation in the groups was strictly voluntary regardless of the type of referral or the reason for it. Students did not receive academic credit for participation, and those who withdrew from the groups were not penalized. Participants were expected to complete, on their own, any class work missed. Approximately 5% of those who joined the groups declined to continue participation; 60% of the withdrawals occurred in the first three weeks and 40% at a later point. Since requests for participation exceeded the capacity of the program in the third year of implementation, priority was given to the students experiencing greater distress. In the 1992-1993 academic year, the groups accommodated 250 of 350 eligible students.

The groups met for 50 minutes during school hours in a school meeting room once a week. The meetings were held in the presence of the two adult nonmental health professional volunteer facilitators, one from the school and the other from the community. The facilitators had received a two-day training seminar in which they were instructed to take a nondirective role. The students discussed their home and school stressors, their feelings, and different coping strategies as well as any other matters of importance to them. Students who failed to show up were contacted by the facilitators and encouraged to discuss their reasons for not continuing. If they still needed help, they were asked to resume participation or were referred to other sources (in-school or out-of-school). At the end of the semester, the participants who desired to continue in the groups were allowed to re-enlist for the following semester, space permitting. New students who needed the service were added to the existing groups. A control group was not available for comparison, since data were not compiled on students who suffered similar problems and were assigned to a no-intervention group. At the end of each of the two semesters, the students were provided with an evaluation instrument called the Self-Assessment Questionnaire developed specifically for this program. The students were asked to mail the completed instrument to one of the authors. To avoid biasing the results, the instrument did not contain any information that could identify the person who completed it. Results were compiled at the end of the academic year.

Instrument

The first 14 items of the Self-Assessment Questionnaire (Table 3) inquired about changes in three major domains (school, interpersonal, and internal). The possible answers were: Positive Effect, No Effect, Negative Effect, Encounter Too Brief To Tell, and Does Not Apply. The following eight items (Tables 4, 5, 6) dealt with overall program acceptability. Four additional questions (Tables 7 and 8) dealt with two major areas of concern: dropping out of school and substance use. One of the two last items of the questionnaire asked for the reasons for missing groups and the other inquired whether the student sought additional out-of-school help as a result of participation.

Table 3: Student Assessment of the Effect of the Groups in Three
Major Domains as Evidenced by Reported Changes in 14 Areas(a).

Domain Question Positive No Effect
Negative
 Effect Effect

School 1. School Attendance 54.92% 36.89% 0.82%

School 2. School Work 52.46% 40.98% 0.00%

School 3. Attitude Towards 60.66% 31.15% 0.82%
 School

Inter- 4. Communication and 74.38% 19.83% 1.65%
personal Expression of Feelings

Internal 5. Self-worth 76.03% 19.01% 0.83%

Internal 6. Coping with Stress 64.46% 28.10% 2.48%

Internal 7. Physical Health 42.37% 48.31% 1.69%

Internal 8. Mental Health 68.91% 23.53% 2.52%

Inter- 9. Relations with 61.54% 30.77% 3.42%
personal Family

Inter- 10. Helping Family 68.91% 26.05% 0.84%
personal

Inter- 11. Relations with 67.50% 27.50% 1.67%
personal Students

Inter- 12. Making Supportive 70.83% 24.17% 0.83%
personal Friendships

Inter- 13. Supporting Friends 75.83% 20.83% 1.67%
personal

Internal 14. New ways to Deal 77.88% 17.70% 0.88%
 With Problems

a "Brief encounter", "does not apply" and missing answers account
for the difference between the row totals and 100%.

RESULTS

Tables 3-9 show the evaluation of the students who participated in the program in 1992-1993 (the third year of implementation). The percentages reported are based on the number of students who answered the particular question, unless otherwise stated. One hundred and thirty-one completed questionnaires were received after the two semesters of that academic year. Since the questionnaires were anonymous, it is not possible to accurately report the number of students who completed the questionnaires after both semesters, the duration of participation of the respondents, or their demographics. The percentages of not completing a specific item varied from 6.9 to 13.7%.

Table 4: Questions Related to General Program Acceptability by
Participating Students.

Question Yes I Do Not No
 Know

Join In-School Group the Following 69.75% 18.49% 11.76%
Year

Join Out-of-School Group the 18.33% 32.50% 49.17%
Following Year

Recommend Group to a Friend 94.74% 3.51% 1.75%

Asked a Friend to Come to Group 66.10% 0.00% 33.90%

Referred a Friend's Name for 36.75% 2.56% 60.68%
Evaluation.
Table 5: Progress of the Feelings of Comfort in the Groups.

Question Comfor- Was Not Comfortable Still Not
 table From Initially, But Comfortable Comfor-
 the Onset at the End of Group table

Comfort Level 29.31% 68.10% 2.59%
Table 6: Rating of Comfort in the Groups and Perception of the
Degree of Group Success at the End of the Semester.

Degree of Comfort 1-Low 2 3 4 5-High

Comfort Level at End 0.84% 0.84% 10.92% 35.29% 52.10%
of Group

Perceived Group 1.68% 5.04% 7.56% 36.13% 49.58%
Success
Table 7: Effect of Groups on Students Regarding Desire to Drop Out
of School or Use Alcohol.

Question Yes No

Have You Ever Considered Dropping Out of 19.30%
80.70%
School?

Did the Group Help Prevent Your Dropping Out 59.52%
40.48%
of School?(a)

Have You Ever Used Alcohol or Drugs? 62.96%
37.04%

a Only those who considered dropping out of school were asked to
answer the question.
Table 8: Effect of Groups on Alcohol and Drug Use.

Question Stopped Decreased No Effect But No
 Using Using More Awareness Effect

Effect on Alcohol 25.35% 32.39% 22.54% 19.72%
and Drug Use(a)

a Only those who admitted to ever using alcohol or drugs in the
past
were asked to answer the question. Sixty-four completed answers
were
received.

[TABULAR DATA FOR TABLE 9 OMITTED]

Instrument Reliability

Reliability analysis was performed on the fourteen areas of potential change shown in Table 3. Cronbach's alpha was 0.85, indicating acceptable internal consistency.

Effects on the School, Interpersonal, and Internal Domains

Table 3 and Figure 1 show that the majority of students reported positive effects more frequently than no effect on all fourteen items, except physical health. Positive effects were more pronounced in the internal and the interpersonal domains though more than half the students reported improvement in the school domain. Positive effects were reported most frequently on developing new ways to deal with problems (77.88%), followed by self-esteem (76.03%), ability to support friends (75.83%), ability to communicate and express feelings (74.38%), establishing supportive relationships (70.83%), and perception of mental health (68.91%). Very few students reported a negative impact of the support groups on any of the items or domains.

Rotated Factor Matrix analysis for the answers to the 14 items was done after excluding students who left three or more questions unanswered. The analysis showed that three main factors would account for most of the variance (56.1%). The first factor, which accounted for 38.3% of the variance (Eigen value = 5.37), appears to be associated with item 3 (school), items 4, 12, and 13 (Interpersonal) and items 5, 6, 8, and 14 (Internal).

Program Acceptability

Table 4 summarizes the answers to the items which indicate the acceptability of the program. Two thirds of the students intended to attend in-school support groups in the following school year while less than one fifth would opt for out-of-school groups. Students who intended to continue in the program outnumbered those not intending to do so by a 6 to 1 ratio. An overwhelming majority of the students stated that they would recommend the program to a friend (94.7%); two thirds asked friends to participate; and one third submitted the name of a friend about whom they were concerned to the school counselor for an initial assessment.

Table 5 shows that only 29.3% of the students were comfortable with attending support groups when they started, but this percentage grew to 97.4% after participation. Table 6 shows that 52.1% gave the program the highest comfort rating at the end of the semester while less than 2% checked comfort levels below the midpoint of the scale, a remarkable change from the level at group onset. The majority of the students reported that the groups were successful. Only 6.7% (three students) graded it below the midpoint on a five-point scale. Almost thirteen times as many reported success above the midpoint (Table 6, second row).

Effects on dropping out of school and drug and alcohol use. The evaluation further addressed two major areas of concern: dropping out of school and alcohol and substance use (Tables 7 and 8). Seventy percent of the 131 students (or 80.7% of those who answered the question) denied that they had considered dropping out, while 17% (or 19.3% of those who answered the question) admitted they had. The remaining 17 students did not answer the question. Almost 60% of those who considered dropping out reported that the groups helped them change their minds.

With regard to alcohol and substance use, 108 students answered the question about ever using alcohol or drugs in the past. Forty of those students denied and 68 admitted to using. Four of the latter did not answer the follow-up question. Of the other 64 students who answered, a quarter reported that the program helped them stop using altogether, a third reported decreased consumption, and a quarter reported that actual consumption did not change but that their awareness of the problem had increased. There was no effect in the remaining fifth. Thus, groups appeared to be effective in these two particularly vulnerable populations.

Reasons for missing group meetings. The reasons given for missing groups are as shown in Table 9. Noteworthy is that only 3.82% reported that they missed the group because the problems they had on admission were resolved and they needed no further help. When the students were asked whether they had sought help outside of the school as a result of participation in the groups, only 21 of the 131 (18.6% of those who answered this question or 16.0% of the 131 respondents) answered affirmatively, while 92 students or 70.2% of the 132 students answered negatively. The remainder of the students (18) did not answer the question.

DISCUSSION

Eggert, Seyl, and Nicholas (1990) suggested that people in supportive social environments are in better off than those without this advantage. Irwin and Vaughn (1988), in a summary of a study group focused on adolescent development, stated that "the promotion of healthy adolescent development is fostered by a prolonged supportive environment with graded steps toward autonomy." Unfortunately, an increasing proportion of adolescents are likely to be raised in impoverished and disrupted family environments in the 1990s.

Despite the prevalence of emotional distress and behavioral problems in high school students and the expected worsening of the situation, only a limited number of studies have attempted to evaluate the yield and economics of in-school support programs and students' perceptions of their psychological impact and acceptability. Also, little attention has been paid to studying the impact on students who suffer from emotional distress rather than severe behavioral problems.

The program reported here is economically feasible because it operated in the school setting, used volunteer facilitators, and was conducted in a group format. It is encouraging that the students in the study noted robust improvement following participation. Only a few respondents reported a negative impact. Improvement was noted in the school arena, including school attendance, performance, and attitude toward school. Some interpersonal gains were reported, including increased ability to relate to family and friends. Gains in psychological status were noted, too, in the perception of degree of mental health, coping with stress, and finding new ways to deal with problems. Improvement in self-concept was also noted. In fact, the students indicated more improvement in the internal and interpersonal arenas than in school performance. Students' evaluation of the overall efficacy of the program was also quite impressive.

The ratings of the effect on school dropout and alcohol and drug use were encouraging, though not as pronounced as the appraisal of the overall perception of the program. This argues that other factors, in addition to assisting students to stay in school and not use alcohol or drugs, enter into their perception of the program's success and acceptability. Since the response to the questions about perceived improvement in communication and expression, self-worth, general mental health, ability to make supportive friendships, and ways to deal with problems were the most positive, it appears that these factors contributed substantially to the perceived success of the program. It is reasonable to assume that sustained improvement in these areas could ultimately have an impact on the dropout rate and the prevalence of alcohol and drug use. The study shows that measures of program efficacy should assess the emotional impact and not be limited to school dropout rate and substance abuse.

Despite the simplicity of the intervention, the lack of extensive training required for the co-facilitators, and the fact that the budget was minimal, the program was successful and appealed to the students. These results support the observations of Wilhoit and Roesch (1989) and Eggert et al. (1990) that nonmental health professionals may be successful in exerting a positive effect on this population of students. However, caution should be exercised not to generalize the outcome to psychiatric settings which deal with more complicated problems (e.g., severe major depression which may require medication).

Almost all the students stated that they would recommend the group to a friend, yet only two thirds actually asked a friend to do so. It is possible that the remaining third did not have friends who needed help. It is also possible that they did not want their friends to know that they were attending the groups, or that they had no friends who were close enough to discuss such matters. Only 50% of those who talked with a friend about coming to the groups took it upon themselves to submit the friend's name to the school counselor for evaluation of a need for assistance. There are several possible explanations: (1) some adolescents wanted to handle their problems without interference by an adult as noted by Offer, Howard, Schonert, and Ostrov (1991) and Vieland et al. (1991); (2) the emphasis adolescents placed on autonomy; (3) concerns that the friend may perceive the action as betrayal of trust especially when the friend's problems, though severe enough to warrant intervention, may not be considered serious enough to justify notifying a counselor; (4) a perception that assistance will not be helpful unless accepted by the friend. Still, 50% of those who talked with friends and asked them to come to groups reported their concerns to the school counselor. This may suggest that such peer support groups can create an understanding of the value of the support systems, trust in the adult counselor, an appreciation of the seriousness of psychological distress, and realization that suffering and distress are not part of normal adolescence. One might also infer that participation increases the students' group cohesiveness, friendship, and sense of responsibility for each other. Be that as it may, it appears that participation in groups provides the school with properly placed early warning systems that can detect problems which might have been concealed from adults, at least initially. It also appears that help-seeking behavior in adolescents is influenced by the availability of what they consider a positive experience, and that such experiences may initiate positive peer pressure on friends to seek help. Not only does this warning system provide the opportunity to prevent catastrophic events from happening to a friend who is distressed, but it relieves that reporting adolescent from some of the anxiety about the friend's distress and the inherent guilt feelings which would develop should a catastrophic event occur. Unfortunately, this early warning system functions only 50% of the time. Whether and when the problems would have been reported by the students to the counselor had there been no group participation remains speculative.

Comfort with the group grew over the course of participation which indicates that, given the right climate, adolescents are willing to experiment with and change their view about sources of available support. The data shed some light on another aspect of adolescents' patterns of seeking help: the students expressed a strong intent to attend in-school, not out-of-school support groups in the following year; only three students stated that they intended to participate in support groups held outside the school. Cost, availability, stigmatization, and comfort level are possible explanations for this reaction. With almost 95% of the students who attended groups stating that they would recommend the groups to a friend, it is likely that the stigma attached to this in-school program was either limited or outweighed by the benefits. It is important to note, however, that this program has been promoted as a self-help group run by volunteers to help normal students who were in distress. This approach may explain, in part, the popularity of the program and the long waiting list. Preference for an in-school program may also be related to students' unwillingness to be displaced from the school system, and to the comfort of being surrounded by peers and friends in the process.

It could be argued that voluntary participation in the program indicates that psychopathology was mild. Refusal of help may result from the severity of the student's or the family's psychopathology. There is some support in the literature for this argument. Papini et al. (1990) noted that adolescents who perceived their parents as being warm and nurturing engaged in significantly more self-disclosure about a variety of subjects, and adolescents from the most affectionate families were found to disclose more not only with their parents, but with peers. This may indicate that students who are willing to share their feelings in a group setting may have experienced more positive relationships in the past and thus may be more likely to improve. However, three pieces of evidence indicate that serious psychopathology prevailed among the students who enrolled in the groups: (1) approximately one fifth of the students considered dropping out of school; (2) almost two thirds reported drug or alcohol use, a high percentage considering that many of the students were still in the ninth grade; (3) less than 4% of the students who missed groups stated that they did not attend due to resolution of the problem that had led to participation. It could also be argued that the reason for the robust improvement was that the students were strongly motivated to seek help, but the data do not point in that direction: only a third of the students were comfortable with being in the group at the beginning; only a fifth indicated intent to join support groups outside the school system in the following year; and only 16% sought other types of help in addition to the in-school groups.

As with any self-report measure, it is possible that the respondents answered the questions with what they considered the most socially acceptable response. This is not likely, however, since almost two thirds of the students answered questions about using alcohol or drugs affirmatively. This indicates that the answers were honest. Also, the procedure of data collection prevented linking individual students with particular answers.

Another limitation of the results is that a sense of "group loyalty" might have developed which increased the perception of positive effects. Also, a group with an overall positive experience might be inclined to endorse improvement on questions focused on all aspects of the student's life even if some aspects failed to improve (i.e., a halo effect). The data, however, does not support either concern: the ratio between the positive and the sum of "no effect" and "negative effects" varied from 0.85 (for item 7: Physical Health) to 4.19 (for item 14: New Ways to deal with problems). Only 42% of the respondents stated that the groups had a positive effect on their physical health, a question embedded in the middle of other questions (partial positive response in the area of physical health was expected because some groups focused on weight problems and the groups led to reduction of alcohol and drug intake in half of those who used them).

Anonymity necessitated that the demographic criteria not be collected. It cannot be assumed that the experience of students from different grades, ages, sexes, and races was uniform. This limits the generalizability of the results. Also, the fact that some of the respondents completed the evaluation twice (at the end of the first and second semesters) makes the percentage of respondents uncertain. In the worst case scenario, however, at least 66 respondents completed the 131 questionnaires received. Another limitation of the data is that no attempt was made to exclude students who were seeking additional help. These students, however, were a minority of the respondents.

The following need further consideration in future studies: (1) the use of standardized instruments to measure the degree of change in emotions and behavior; (2) expansion of some of the items, for example, including questions about the amount of drinking and drug use, reckless behavior, suicidal ideations, reasons for not joining out-of-school support groups, reasons for asking friends to come to groups, and reasons for not referring them to the school counselor if not done; (3) addition of items to reveal the reasons for initial discomfort with groups and the factors which increased the level of comfort; (4) the need for different management strategies to deal with leaving the groups and to monitor the outcome of the students who withdrew from the groups and those who refused to participate; and (5) to more fully explore exactly what was perceived to be helpful in the groups.

Since these groups were facilitated by nonmental health professionals, it would be of interest to compare the outcome of these groups with traditional groups offered by professionals. Program integrity has been reported to affect outcome by Hanson and Graham (1991) who argued that an experienced mental health professional facilitator may be more effective. However, an "intent-to-treat-analysis," which factors in the outcome for every student referred for assistance, may show that program acceptance and rate of participation are higher in the type of groups used in this study as compared to those offered by mental health professionals. This acceptance may raise the outcome of the less-trained volunteers to match that of the professionals. The issue of in-school mental health professional-facilitated groups vs. volunteer-facilitated groups is also of interest, though budgetary limitations would favor the latter even if the efficacy is somewhat lower.

Answers still need to be found for several important theoretical and clinical aspects: (1) does the effect of these groups last after the groups are discontinued?; (2) do psychological and interpersonal gains correlate with objective changes in academic performance, school dropout rate, disciplinary action, and school attendance? Also, do they correlate with the assessments made by the teachers and parents (especially in view of the discrepancies in reports of the adolescents, teachers, and parents reported by Blackeney et al. (1993) and Meyer et al. (1993); (3) what is the ideal duration and frequency of group meetings?; (4) what is the natural history of the problem in those who refuse to participate, those who dropped out of the groups, and those who could not be served due to limited space?; and (5) what types of problems respond best to this intervention?

In summary, the results of this study indicate that the high school students who participated in the in-school volunteer-facilitated support groups perceived the program to be helpful. Improvement was reported in school attendance and performance, interpersonal skills, and emotional state. Also the program seemed to help students decide not to drop out of school and not to use alcohol and drugs. The students expressed the intent to attend future in-school programs and recommended the programs to friends, but showed little enthusiasm for out-of-school programs. The level of comfort with the groups increased with participation. The results show that this low-cost approach is acceptable to the students and reaches out to those who would otherwise not receive the help they need. Coupled with a safety net of traditional mental health services to assist students with severe problems and those who fail to respond to the peer support groups, such programs may be a step in the right direction toward solving the problem of emotional distress and behavioral problems in high school students.

Table 1: Demographic Characteristics of the School Catchment Area and the Students Enrolled in the School.

* COMMUNITY CHARACTERISTICS:

The students came from three catchment areas:

(1) Area 1: Population: 9,497 (1990 census)

Primarily composed of white, low to high-middle socioeconomic class families. Approximately 10% of the town's population is composed of African Americans and Hispanics who fall in the low and middle socioeconomic groups.

(2) Area 2: Population: 5,549 (1990 census)

Predominantly blue collar families working in the fishing industry. Predominantly white, ranging from low to middle socioeconomic status (mostly in the high-low to low-middle socioeconomic status).

(3) Area 3: Population: 3,325 (1990 census)

Predominantly white population in the low socioeconomic status. Many work in the fishing industry but a significant number receive welfare.

The three communities form a single independent school district with one high school of 1,500 students.

* CHARACTERISTICS OF THE STUDENTS WHO ENROLL IN THE SCHOOL

Ages: 14 to 19 years old

Sex ratio: 51% male and 49% female

Race: 69% White, 17% Hispanic, 12% African American, and 2% Asian (i.e. minorities are somewhat overrepresented in the school compared to the community)

Primary language spoken at home: English was the primary language in 82-85% of the students' homes. Spanish and Vietnamese were the languages for most of the remainder.

Socioeconomic status: Difficult to obtain, but as of 1992, 236 students were receiving free lunches and 49 students were receiving reduced lunches. The two main factors used to determine if a student is eligible for free or reduced lunches are the number of children living at home and the parents' income.

Table 2: Support Groups Available:

1) RECOVERY/SOBRIETY SUPPORT: for students who have realized the consequences of their chemical use and have chosen to remain abstinent.

2) CONCERNED PERSONS: for students concerned about and/or affected by the use of chemicals by a relative or friend (usually parents).

3) FAMILY CHANGE: to help students deal with parental divorce and/or remarriage.

4) GRIEF: for students who experience difficulties after the death of a close person.

5) SHARING GROUP: for students who want to explore feelings and concerns about school, family, relationships, self image and coping skills.

6) METAMORPHOSIS GROUP: for students who have been victims of physical, sexual, or emotional abuse in the past.

7) BODY AWARENESS: for students with concerns about body image.

8) TEEN PARENT: for students who are coping with the dual role of parent and student.

9) DEPRESSION: for students who suffer from feeling depressed.

10) WINNERS: for students dealing with minority issues.

REFERENCES

Blackeney, P., Meyer, W. M., Moore, P., Broemeling, L., Hunt, R., Robson, M., & Herndon, D. (1993). Social competence and behavioral problems of pediatric survivors of burns. Journal of Burn Care and Rehabilitation, 14, 65-72.

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.

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), 414-30.

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

Meyer, W. J., Blackeney, P., Holzer, C. E., Moore, P., Murphey, L., Robson, M., & Herndon, D. (1993). Inconsistencies in psychosocial assessment of children post major burn. Journal of Burn Care and Rehabilitation (in press).

Offer, D., Howard, K. I., Schonert, K. A., & Ostrov, E. (1991). To whom do adolescents turn for help? Differences between disturbed and nondisturbed adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 30(4), 623-30.

Papini, D. R., Farmer, F. F., Clark, S. M., Micka, J. C., & Barnett, J. K. (1990). Early adolescent age and gender differences in patterns of emotional self-disclosure to parents and friends. Adolescence, 25(100), 959-76.

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), 811-5.

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