The Effect of Mentoring on School Attendance and Academic Outcomes: A Randomized Evaluation of the Check & Connect Program (WP-16-18)
Jonathan Guryan, Sandra Christenson, Amy Claessens, Mimi Engel, Ijun Lai, Jens Ludwig, Ashley Cureton Turner, and Mary Clair Turner
The researchers present the results of a four-year randomized controlled trial evaluation of a structured student monitoring and mentoring program that aimed to increase student attendance. The program, called Check & Connect (C&C), was delivered to 765 students in grades 1 through 8 in 23 neighborhood schools in the Chicago Public Schools (CPS). C&C mentors were full-time employees and had caseloads of between 30 and 35 students. Each student was assigned a mentor for two school years, and the program was delivered to two cohorts of students over the 2011-12 to 2014-15 academic years. Mentors tracked data to monitor the attendance and academic progress of the 30 to 35 students on their caseload. Mentors also met regularly with students and delivered personalized interventions designed to increase students’ attendance and engagement with school. Based on estimates of treatment on the treated (TOT), they find that participation decreased student absences among students who began the program in grades 5-7 by a statistically significant 3.4 days, or 20.2 percent relative to the control complier mean. The researchers do not find statistically significant effects of participating in C&C among students who began the program in grades 1-4. For both cohorts, the effect of participating in C&C was larger in the second year of the intervention than the first, though that difference was not statistically significant, which is at least suggestive of the possibility that the development of relationships between the mentor and student may be an important mechanism through which the mentoring program is effective. The researchers did not find significant effects on grade point average, but did find a statistically significant decline in courses failed. There were mixed results for test scores, but no evidence that test scores increased significantly.