Positive Youth Development in the United States: Research Findings on Evaluations of Positive Youth Development Programs. Programs in One Social Domain: Community

11/13/1998

Big Brothers/Big Sisters

Tierney, Grossman and Resch (1995) evaluated the Big Brothers/Big Sisters program, a mentoring intervention in the community domain.  The study addressed nine positive youth development constructs, including social, emotional, cognitive and behavioral competencies, positive identity, bonding, resiliency, self-efficacy, and prosocial norms.

Program Description

The intervention model featured positive youth development strategies that did not include a specific skills training component, but rather, targeted systemic change in the child's social domains as a function of bonding with a healthy adult.  The core strategy in Big Brothers/Big Sisters is to have youth use the program structure and resources to establish a mentoring relationship with prosocial adults.  The minimum time commitment required of mentors was several hours, two to four times a month for at least a year.  More than 70% of matches in this study met at least three times a month for more than three hours each time, and nearly half met once a week and had an average total exposure of 11 months.  The one-to-one mentoring was based on careful matching of adult mentors and children on backgrounds, preferences, and geographic proximity.  National standards were implemented for volunteer and youth screening, training, matching, meeting requirements, and supervision.

Research Design

The Big Brothers/Big Sisters evaluation used an experimental study design that randomly assigned participants to the intervention condition or a wait-list control group.  The 18-month wait list corresponded to what is often the usual waiting period for matches in this program.  Each agency implemented random assignment procedures until it met the sample size goals.  Sites were chosen based on large active caseloads, waiting lists, and geographic diversity.  No long-term follow-up data were collected.  All data reflected pre-post measurement from interviews conducted at baseline and at immediate posttest 18 months later.  Attrition analyses showed that the original sample of 1138 youth, ages 10-14, experienced attrition of 179 participants, resulting in a final sample of 959 youth.  Participants were from eight BB/BSA program sites in Phoenix, Wichita, Minneapolis, Rochester, Columbus, Philadelphia, Houston and San Antonio during the study period in 1992-93.  Approximately 60% were minority group members (predominantly African-American and Hispanic).  More than 40% received food stamps or public assistance.

Results of the Intervention

The evaluation reported significant outcomes at the p<.10 level, but this review acknowledged only those results which met the more rigorous p<.05 standard.  The evaluation demonstrated results on behavioral and attitude measures in six impact areas: antisocial activities, academic performance and other school indicators, relationships with family, relationships with friends, self-concept, and social/cultural enrichment.  Overall program and control group differences included decreases in drug use (45%, p<.05), hitting (32%, p<.05), the number of times a youth skipped class (37%, p<.05) or a day of school (52%, p<.05), and the number of times a child lied to his parents (37%, p<.05).  Significant increases were found in the perceived ability to complete schoolwork (71%, p<.01), and improved parental relationships reflected as increases on indicators of trust (64%, p<.05).  Specific sub-group effects showed the program had the greatest impact on substance abuse reductions for minority boys (67.8%, p<.05).  Measures of cognitive or academic competence showed larger impacts for minority girls in the treatment group, whose perceived scholastic competence scores were 10% higher (p<.01) than those of control group minority girls.  Caucasian Little Brothers also significantly increased their scholastic competence scores (7%, p<.05).  Minority male children in the intervention group experienced the highest levels of increased peer emotional support (6%, p<.05) compared to children in the control group.  No overall significant effects on measures of self-concept (differences on global self-worth, social acceptance or self-confidence) were noted, but the sub-group of Caucasian treatment group boys scored significantly higher on the social acceptance scale than their control group counterparts (p<.01).

Bicultural Competence Skills

Schinke, Botvin, Trimble, Orlandi, Gilchrist and Locklear (1988) evaluated the Bicultural Competence Skills program, an intervention based on bicultural competence theory and social learning principles.  The program addressed 11 positive youth development constructs, including social, emotional, cognitive, behavioral, and moral competencies, positive identity, bonding, self-efficacy, recognition for positive behavior, opportunities for prosocial involvement, and prosocial norms.

Program Description

A core component of the program philosophy was helping children develop a positive identity based on "bicultural fluency," or becoming socially competent in two cultures.  The specific skills training strategies used by the program ranged from practice and rehearsal techniques (fostering opportunities, recognition, and promoting competence), such as role-play, to positive peer-based cultural strategies (promoting positive identity, prosocial norms, and bonding), such as homework assignments on communication and coping strategies associated with bicultural efficacy.  Two Native American counselors led the 10-session intervention.

Research Design

The study used an experimental pretest, posttest, follow-up design in which subjects were voluntarily recruited, then randomly divided after pretesting by reservation site into prevention and control conditions.  All subjects completed four outcome measures before, immediately following, and at 6-months post intervention that analyzed culture-relevant peer influences on tobacco, alcohol and drug use, levels of substance abuse, and changes in substance abuse knowledge and attitudes relevant to Native American culture.  The unit of assignment matched the unit of analysis (individual).  The attrition analysis showed that attrition averaged nine percent across the sample at 6-month follow-up with no dropout differences identified between conditions.  The sample of 137 Native American youth, whose average age was between 11 to 12 years old, came from two western Washington reservation sites in a population drawn from tribal and public schools.  The evaluation established group equivalence at pretest on the children's household composition, level of acculturation, and current place of residence.

Results of the Intervention

The evaluation found significant results at immediate posttest measurement and at 6-month follow-up in favor of the program group.  The analysis showed main effects for the intervention (F(1,124) = 8.28, p<.005) and for measurement occasion (F(1,124) = 7.22, p<.01).  At posttest, program students were significantly more knowledgeable about substance use and abuse and held less favorable attitudes about substance use in the Native American culture than their control group counterparts.  The intervention group youth also had significantly higher ratings for self-control, the ability to generate alternative suggestions to peer pressure-based encouragement to use substances, and assertiveness.  The outcomes on substances used in the previous fourteen days showed that intervention youth reported lower levels of smokeless tobacco use (Program group (P) mean = 2.38, Control group (C) mean = 3.77, p<.05), less alcohol (P mean = 3.63, C mean = 4.71, p<.05) and marijuana use (P mean = 2.12, C mean = 3.79, p<.05).  At the 6 month follow-up, intervention participants continued to score higher on measures of knowledge of substance abuse (P mean =17.2, C mean =11.5, p<.05), self-control (P mean =1.89, C mean =.87, p<.05), alternative suggestions (P mean =1.14, C mean =.43, p<.05), and assertiveness (P mean =1.38, C mean =.90, p<.05), and reported significantly less use of smoked tobacco (P mean =1.41, C mean =2.37, p<.05), smokeless tobacco (P mean =2.56, C mean =4.11, P<.05), alcohol (P mean=3.76, C mean = 4.92, p<.05), marijuana (P mean =1.97, C mean = 4.02, p<.05), and inhalants (P mean =.94, C mean =1.32, p<.05) in the last 14 days.