Improving schooling and life outcomes of disadvantaged youth is a top priority in the US, but to date not many interventions have been shown to improve the outcomes of these youth, especially males. The core theme of our program project is to assess the most effective (and cost-effective) ways to improve long-term life outcomes for disadvantaged youth, with an emphasis on the closely related outcomes of schooling attainment, risky and criminal behavior, and violence involvement. In the summer of 2013 we identified a study sample of 2,134 males entering 9th and 10th grade in 12 public high schools on Chicago's south and west sides, working in close collaboration with Chicago Public Schools (CPS). These youth were randomized to receive a very promising academic intervention (high-dosage math tutoring delivered by Match Education, which in previous non-experimental studies by our collaborator Roland Fryer have been found to boost math scores by 0.5 to 0.7 standard deviations for 6th and 9th graders), or to a very promising non-academic intervention (a form of cognitive behavioral therapy called """"""""Becoming a Man,"""""""" or BAM, developed and implemented by a non-profit in Chicago, Youth Guidance), or to receive both, or neither. The interventions were launched in August 2013 for the 2013-14 academic year with outside (non-NICHD) support. Projects 1 and 2 help sustain both interventions into a second intervention year (AY2014-15). Project 3 would involve in-person data collection from the same sample of male youth who form the study sample of Projects 1 and 2 to measure key social/behavioral outcomes that are not well captured by administrative data, and, importantly, to measure potential mediating mechanisms, which are crucial to understand to enhance the effectiveness program and to help guide successful scale-up. The project engages a survey subcontractor, Institute for Social Research at the University of Michigan, to carry out an in-person survey during AY2015-16 with a sample frame of 1,800 youth (randomly selected out of the total study sample). These in-person data would help measure behavioral outcomes that are not well captured by the other administrative data sources we access for this project, such as injuries, violence victimization, and risky behaviors that do not result in arrest, which is important for carrying out as comprehensive a benefit-cost analysis of the two interventions as possible. The proposed in-person survey collection also includes a number of novel mechanism measures, including modified versions of two laboratory experiments from psychology that measure meta-cognitive processes, which our theory of change suggestions may be a key pathway through which BAM affects youth outcomes. It will also measure potential mediating pathways through time use and peer affiliations that take advantage of the ubiquity of cell phones among high school youth (which would also help measure the degree to which emotion regulation is a mechanism behind BAM), and would include a real-stakes decision-making experiment to measure graduation expectations, a key mechanism of action that may be behind the Match effects.
|Heller, Sara B; Shah, Anuj K; Guryan, Jonathan et al. (2017) Thinking, Fast and Slow? Some Field Experiments to Reduce Crime and Dropout in Chicago. Q J Econ 132:1-54|