Program Profile: Across Ages

Study Title: Mentoring as a Drug Prevention Strategy: An Evaluation of Across Ages. (which is associated with Outcome 1, Outcome 2, Outcome 3, Outcome 4, Outcome 5, Outcome 6, and Outcome 7)

Research Design

Aseltine, Dupre, and Lamlein (2000) conducted a 3-year evaluation of the Across Ages program in Springfield, Massachusetts. Ten sixth-grade classrooms across three public middle schools participated in the evaluation during the 1997 and 1998 school years. Classrooms were randomly assigned to one of three conditions: 1) the full mentoring condition (n = 85), 2) the curriculum-only condition (n = 135), and 3) a control group (n = 138). Students in the mentoring condition received all three components of the program: the Positive Youth Development Curriculum (PYDC), participation in community service with elderly residents, and weekly one-on-one mentoring with older adult volunteers. Students in the curriculum-only condition received the PYDC and participated in community service but received no mentoring. The control condition received no intervention. Parental consent was required for students to participate in the program. The CrimeSolutions review of this study focused on the differences between students in the full mentoring condition and students in the control condition.

Sample

The total analytic sample included 358 students. Youths completed questionnaires at three points: 1) baseline (before the intervention), 2) immediately after the school year, and 3) six months post-intervention (the CrimeSolutions review focused on the findings from this time period). The study took place in Springfield, Massachusetts, a diverse city of 150,000 with high poverty levels, especially among children and minority groups. As of 1990, 20 percent of residents lived below the poverty line, including one in three children. The population is 18 percent Black and 17 percent Hispanic/Latino, with 58 percent of Latino children younger than 18 and 74 percent younger than 5 living in poverty. In public schools, 72 percent of the 26,000 students are members of minority groups and 77 percent have low income. The study authors did not provide specific information on the demographic characteristics of the students in the sample.

Outcome Measures/Data Collection

Measures were primarily drawn from the National Youth Survey and were supplemented with items assessing attitudes toward the elderly. The primary outcomes included measures of mental and behavioral health (e.g., self-control, cooperation), academic performance (e.g., grades), problem behaviors (e.g., alcohol use), and attitudes toward drugs and older adults. Trained interviewers administered the questionnaires during class time. Summary scales were created by averaging item scores, and minimal missing data were imputed using item means. Additional controls, such as family drug use, neighborhood risk, and family supervision, were included for analyses involving substance use and problem behaviors to strengthen the accuracy and power of statistical tests.

Statistical Analysis

Statistical analyses were conducted using general linear models that controlled for: 1) baseline values of outcome variables; 2) covariates including race, ethnicity, sex, family structure, and language spoken at home; and, for some models, additional risk factors such as family drug use and neighborhood risk. The study did not conduct subgroup analysis. 

Citation:

Aseltine, Robert H., Matthew Dupre, and Pamela Lamlein. "Mentoring as a Drug Prevention Strategy: An Evaluation of Across Ages." Adolescent & Family Health 1(1):11–20.

This study is a high-quality randomized controlled trial.