Indiana University SPEA Edward J. Bloustein School of Planning and Public Policy University of Pennsylvania AIR American University

Panel Paper: Lotteried Down: The Consequences of Losing in Rankings-Based School Assignment

Friday, November 13, 2015 : 10:35 AM
Japengo (Hyatt Regency Miami)

*Names in bold indicate Presenter

Joshua Cowen, Michigan State University, Jane Lincove, University of Texas, Austin and Jon Valant, Tulane University
The study examines the outcomes of a school choice lottery in the context of a citywide choice system.  Previous studies suggest that students who win school lottery experience school quality improvements and their parents have increased satisfaction with schools (e.g.  Angrist, et al. 2012; Wolf, et al. 2013).  In most cases in the US, students have a default neighborhood they can attend, and school choice is an added opportunity to select into a charter school, magnet school, or private voucher school.  In a citywide system of choice, there is no default option, and parents depend on the school matching mechanism for all school placements.  This adds uncertainty, particularly if a choice district includes schools that vary in quality measures that matter to parents.

In New Orleans, public schools have no geographic attendance zones and the large majority participate in a centralized student assignment lottery (OneApp).  The OneApp system includes two rounds of school lotteries in which students rank schools and are assigned to the top-ranked school in which they win a lottery.  Following the second round, students re-sort in process by which parents can request transfers if additional slots are available when lottery winners do not enroll.  State administrators who oversee the OneApp report that over 90% of students are assignment to one of their top three choices.  But in a system with a large number of failing schools, it is in unclear what is lost between a parents’ first and third choices, or what happens to the remaining 10% of students.

Using parental rankings of schools from the OneApp for the 2013-14 school year, we calculate the difference between a parent’s first choice school and subsequent choices to measure the potential cost of being a lottery loser.  Next, we calculate the difference between the first choice and the actual assignment.  Finally, we estimate the effects of losing a first choice school on parent behavior in the round two lottery and final assignment process.  Preliminary results indicate that a large proportion of parents rank a small number of highly oversubscribed school in top choices.  Subsequently, a large number of students face assignments with considerably lower report card grades.  This leads to a large number of parents attempting to gain a reassignment through the second lottery and large number of bureaucratic transfers after the lottery is complete.