Assessing the Efficacy of Check & Connect for Improving Outcomes for At-Risk High School Students
Co-Principal Investigators: Jason Snipes (IMPAQ International), Mindee O’Cummings (American Institutes for Research)
Purpose: Because the dropout problem has persisted in the United States over the previous decades, much work has focused on the development and implementation of dropout prevention programs in and outside of school. A review of the research on these programs suggests that some, but not many, dropout prevention programs have positive effects on school persistence and completion. Check & Connect is a dropout prevention program with demonstrated effectiveness for students with disabilities. The research team will carry out an efficacy trial testing the effect of Check & Connect for general education students (i.e., students who do not receive special education services) who show signs that they are at-risk for dropping out of high school. This 4-year study has two specific aims: (1) to generate rigorous evidence about the effects of Check & Connect on engagement, school completion, and academic outcomes of general education students who meet specific criteria for dropout risk; and (2) to document the implementation of Check & Connect with at-risk high school students.
Project Activities: Using school district data, the team will identify general education students in 9th grade who are at risk of dropping out of high school and seek parental consent for participation. Consented students will be randomly assigned to treatment or control conditions. The treatment students will receive Check & Connect services, including pairing with a trained mentor for the next 3 years, until the time of their scheduled 4-year high school graduation. Control students will receive typical counseling and support services offered by the school district. School administrative data will be used to identify at-risk students and to measure student academic and behavioral outcomes. An online survey will be used to measure various aspects of students' engagement with school.
Products: Products include evidence of the efficacy of the Check & Connect dropout prevention program for general education students, and published reports. A cost-effectiveness analysis will provide information for school personnel to guide their choices in adopting particular evidence-based programs.
Setting: Ten high schools in the San Diego Unified School District (SDUSD) will participate in the study.
Population: Participants in the study will include a total of 600 at-risk 9th grade students. In the 2008-2009 school year, SDUSD had a dropout rate of 23.5% overall. That rate was higher than the state average (21.5%) and SDUSD also had a larger dropout rate for ethnic/minority groups (27.4 to 30.5%).
Intervention: Check & Connect is an individualized program that pairs a trained mentor with each participating student. The mentor closely monitors the students' attendance and progress in school and matches them with appropriate supports. The "check" component is designed to facilitate continuous assessment of students' level of engagement with school as reflected in attendance, academic performance, and behavior. The students' level of risk is used to guide the intensity of the "connect" component. Basic intervention includes one-on-one conversations between mentor and student at least monthly. More intensive interventions are individualized, ranging from academic support to direct teaching of coping strategies and problem-solving meetings with the student's family.
Research Design and Methods: The study uses a within-school random assignment design where students who meet established criteria for risk (e.g., low attendance rates, failing courses, a grade point average of 2.0 or lower) will be assigned either to participate in the Check & Connect program or not. Within each school, one-half of students identified in the spring of their 9th grade year as at-risk for dropping out will be randomly assigned to treatment and the rest to a business-as-usual control group. Students assigned to treatment will receive Check & Connect services for 3 years, beginning in the summer after their 9th grade year and ending in spring 2014 (at the time of their scheduled 4-year graduation from high school). Baseline and outcome measures (e.g., academic performance, attendance, disciplinary referrals) will be collected directly from the district. Site visits will occur in the fall of 2011, 2012, and 2013 to conduct focus groups with program mentors and a sample of treatment students to monitor implementation. The cost-effectiveness of Check & Connect relative to other similar programs will be assessed throughout the study. All students will complete a survey in the spring of 2012 and 2014 to measure different aspects of student engagement.
< strong>Control Condition: Students in the control condition will receive the ordinary set of counseling and support services to which these students have access in SDUSD, but will not have contact with the Check & Connect program or mentors.
Key Measures: The study outcomes include engagement (e.g., connection to school and participation in school programs, supports, and activities), attendance, course performance (i.e., credit accrual, grades), standardized test scores, and behavioral measures. Key outcomes include progression through school, persistence, dropout status and, ultimately, graduation status. Implementation will be measured through Check & Connect monitoring sheets, intervention logs, and feedback forms.
Data Analytic Strategy: To carry out the main impact analysis, researchers will estimate multiple regression models that account for the clustering of students within schools. Subgroup analyses will be conducted to test whether Check & Connect is differentially effective for different types of students (e.g., those with different levels of risk). Corrections for student-attrition and mediation analyses will also be conducted. Researchers will conduct a set of implementation analyses to address application of the intervention to at-risk students in varied settings. As part of these analyses, the researchers will examine treatment fidelity, service contrast with the control group, and program costs.