Communities In Schools


Every day, 7,000 students drop out of school. Among Latinos and African-Americans, the dropout rate is nearly 50 percent. Communities In Schools (CIS) works with low-income K-12 students in the nation’s poorest-performing schools. It seeks to reduce dropout rates through preventive support services like short-term counseling or annual health screenings for the entire school, alongside more intensive case-managed services, including tutoring, mentoring, and other services for students at high risk of dropping out. If CIS demonstrates effectiveness, it would show that greater integration of services can help reduce dropout rates.

This program is funded by the federal Social Innovation Fund (SIF), a public-private partnership designed to identify and expand effective solutions to social challenges. The Edna McConnell Clark Foundation, a SIF grantee, is leading a SIF project in collaboration with MDRC and The Bridgespan Group to identify proven programs that can help low-income young people become productive adults. CIS is one of nine subgrantees included in that SIF project.

Agenda, Scope, and Goals

The CIS model operates on two levels: “Level 2,” or case-managed, intensive services for individual high-need students and “Level 1” preventive services for the whole school. This evaluation includes two studies, one focused on the implementation and impact of Level 2 services and the other focused on the impact of the whole-school model, including Level 1 and Level 2 services. The primary focus of this evaluation is the first study, which will answer two main questions:

  • What is the impact of Level 2 services on students’ progress toward graduation — that is, their likelihood of being promoted to the next grade level?
  • Do Level 2 services make students less likely to be chronically absent? (“Chronically” absent means being absent more than 20 percent of the time.)

Researchers will also gather information related to implementation issues that may affect how much of an impact a program has:

  • Variations in the CIS case-management model across affiliates and schools
  • The dosage and duration of services provided
  • The difference between CIS services and other services available to students
  • The circumstances in which CIS services are offered

The second study will answer two additional questions:

  • What is the impact of the CIS whole-school model on graduation rates?
  • What is the impact of the CIS whole-school model on student dropout rates?

Finally, the evaluation will also investigate the costs of the CIS model.

Design, Sites, and Data Sources

The CIS model uses two tiers of prevention and intervention resources and services. Basic prevention and intervention services (Level 1 services) are available for any student in a CIS school. Typically these last for only a few hours or days and address schoolwide needs. Examples would include providing clothing and school supplies, coordinating schoolwide career fairs, bringing in health care professionals to conduct annual screenings, and providing short-term counseling in crisis situations. Level 2 case-managed services are sustained interventions lasting throughout the school year or possibly several years, intended for students at high risk of dropping out: those with poor academic performance, poor attendance records, or frequent misbehavior. These students receive services such as tutoring, mentoring, or individual or group counseling. The idea is that integrating Level 1 and Level 2 services should improve outcomes overall for schools and students.

The evaluation will use two complementary studies to investigate whether or not that is true. The first study is an individual-level randomized control trial in middle schools and high schools, designed to investigate the impact of Level 2 services on individual students. From the pool of students eligible for Level 2 services, a group at each school was selected randomly to receive those services. The other students were assigned to a control group. About 2,000 students have been enrolled from a sample of middle schools and high schools across five participating CIS affiliates in North Carolina, South Carolina, and Texas. Data will come from district school records, project-administered student surveys, interviews with school and program staff, program observations, and a project-administered survey of key school-level staff.

The second study uses a comparative interrupted time-series approach to investigate how the CIS model affects whole schools. It will compare the change in outcomes over time for schools that adopted CIS with the corresponding change over time for similar comparison schools that did not adopt it. The difference between the amount CIS program schools improve and the amount the comparison schools improve provides an estimate of the program’s effect. Elementary schools, middle schools, and high schools will be examined in North Carolina and one to two other states or school districts where CIS has a strong presence. Data for comparisons will come from state data warehouses, school district records, CIS records, and the national Common Core of Data.