Review of Three Recent Randomized Trials of School-Based Mentoring and commentaries

Abstract

Between 2007 and 2009, reports were released on the results of three separate large-scale random assignment studies of the effectiveness of schoolbased mentoring programs for youth.The studies evaluated programs implemented by Big Brothers Big Sisters of America (BBBSA) affiliates (Herrera et al., 2007), Communities In Schools of San Antonio, Texas (Karcher, 2008), and grantees of the U.S. Department of Education's Student Mentoring Program (Bernstein et al., 2009).Differences in the findings and conclusions of the studies have led to varying responses by those in practice and policy roles.The results of the BBBSA trial led the organization to undertake an initiative to pilot and evaluate an enhanced school-based mentoring model.Findings of the Student Mentoring Program evaluation were cited as a reason for eliminating support for the program in the FY 2010 federal budget (Office of Management and Budget, 2009).In this report, we present a comparative analysis of the three studies.We identify important differences across the studies in several areas, including agency inclusion criteria, program models, implementation fidelity and support, and criteria utilized in tests of statistical significance.When aggregating results across the studies using meta-analytic techniques, we find evidence that school-based mentoring can be modestly effective for improving selected outcomes (i.e., support from non-familial adults, peer support, perceptions of scholastic efficacy, school-related misconduct, absenteeism, and truancy).Program effects are not apparent, however, for academic achievement or other outcomes.Our analysis underscores that evidence-based decision-making as applied to youth interventions should take into account multiple programmatic and methodological influences on findings and endeavor to take stock of results from the full landscape of available studies.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 65 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.