Improving academic outcomes for disadvantaged youth: a randomized experiment in Chicago

JPAL-NA affiliate Jens Ludwig (University of Chicago) is carrying out a large-scale RCT with the Chicago Public Schools and Match Education of Boston to test whether intensive, individuated instruction (two-on-one daily tutoring for an hour a day) can improve academic outcomes of disadvantaged youth. We hypothesize that one key reason few academic interventions work for such youth is they fail to address the problem of “mismatch” between regular classroom content and what youth who can be up to seven years behind grade level actually need. Match high-dosage tutoring uses frequent formative assessments to individualize instruction. A non-experimental study by our collaborator Roland Fryer (2011) and our small-scale pilot RCT in Chicago find gains in test scores of 0.5 to 0.7 standard deviations and 50% in expected graduation rates at a cost of $2,000 to $4,400 per youth. We will measure impacts from the first large-scale RCT of this promising intervention that we are currently carrying out in 12 CPS high schools with 2,560 male 9th and 10th graders, and to help measure impacts on candidate mediators. This work also helps build an administrative dataset that can support future RCTs in Chicago.