Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The JAG high school program currently serves 200 students from South Bend high schools each year; this would support a study enrollment of 400 students per year. While a sample size of 400 would allow for a minimum discernible effect (MDE) of 10.4 percentage points, the City is exploring the option of increasing the duration of this study to increase statistical power. For example, a sample size of 1200 would support an MDE of 6.3 percentage points. Studies of similar interventions have measured changes in high school graduation rates of 9-10 points. (Wilson et. al, 2011 and MDRC, ongoing study)
Wilson SJ, Tanner-Smith EE, Lipsey MW, Steinka-Fry K, Morrison J. Dropout prevention and intervention
programs: Effects on school completion and dropout among school-aged children and youth. Campbell Systematic Reviews. 2011;2011:8.
“New York City Small Schools of Choice Evaluation.” MDRC, MDRC, 19 Apr. 2018,