Allowing families to express their preferences for various schools—whether inside or beyond their geographically-zoned building or district—sounds good in theory. Indeed, we’ve been hearing for decades that a zip code should never determine the quality of a child’s education. But does allowing parents to provide their preferences for schools mean that they actually get them? A recent report examines how New York City’s centralized enrollment and school matching system—arguably the largest and most complex in the nation—works in practice, with a focus on which families get their preference in the school system and which don’t.
New York City allows families at all grade levels to apply to any public school in the city, regardless of where they live. Kindergarten parents, for example, can choose from over 700 traditional public schools, not including charter and private school options. The district’s school allocation process—which must accommodate tens of thousands of students in each grade each year—encompasses a complicated set of enrollment deadlines, application priorities, parental preferences, and seat availabilities. The analysts for this study had access to restricted-use data on kindergarten enrollment applications—the only grade they focus on—used to make admission decisions for the 2015–16 through 2018–19 school years. The study excludes youngsters who applied to gifted programs, charter schools, special education schools, and the like. Their sample ultimately includes over 233,000 kindergarteners attending 743 district-run schools across the four years of the study.
The report starts with background on the process. Parents submit their preferred schools in rank order, with no limit to the number of schools that may be requested. Applicants are then assigned a priority rating of 1 through 8 per school requested, with 1 being the highest. That top rating is bestowed upon students who both live in the requested school’s zone and have a sibling enrolled there. That is followed by priority 2 students, who live in the zone but don’t have an enrolled sibling; priority 3 is for kids who live outside the school zone but in the larger assigned district and have a sibling in the requested school; and priority 4 is reserved for students in the assigned district with no sibling. The lower priority ratings (5–7) include students who attend a public pre-K program at the school to which they are applying, and on down to the least priority (8): those living outside the district and with no “prior claim” on the school to which they are applying. They, for all intents and purposes, get the schools that have seats remaining. Additionally, families that apply before the posted deadline get a little bump over families who didn’t meet the deadline.
Analysts examined which students did and did not receive their first-choice school. Given the priority system in place, it is not at all surprising that over-subscribed and high-quality schools enroll smaller proportions of students from disadvantaged and minority backgrounds. Specifically, Black students were 14 percentage points less likely to match with their first-choice school than White students. In fact, the priority ranking system essentially explains the demographic differences observed: A model that controlled for the priority rankings of students reduced the Black/White difference in the probability of a first-choice match from 14 percentage points to less than 1 percentage point. The pattern was similar for Hispanic students.
The analysts primarily fault the prominence of residential location in the priority rankings as disadvantaging Black, Hispanic, and low-income students—just as it happens in traditional address-only assignment with its long history of residential redlining. Not only are these families unlikely to get seats in the most-sought-after schools, many don’t even try since transportation to non-zoned schools poses real challenges. In fact, 87 percent of on-time applicants who chose their neighborhood school as their first choice got it, compared to only 6 percent who ranked a non-neighborhood school first.
The authors acknowledge that getting rid of zoned school, sibling, and/or pre-K priorities would be “politically and logistically fraught.” But they suggest that the city use a “diversity code” for each 4–8 block neighborhood which reflects neighborhood racial composition, income level, and adult education level, and then reserve a portion of the seats in highly-sought-after schools for students from neighborhoods that rank high on the diversity code. Another idea is to expand the number and capacity of high-performing schools!
The findings are disappointing—showing that New York’s school choice efforts are more illusion than reality for most families of young learners—but they’re also important. Any system that is serious about giving parents a choice of schools needs to reckon with the scarcity of good ones and the inability of families to break through barriers of entry, especially when, apropos of the study title, they have “constrained agency.”
SOURCE: Rebecca J. Shmoys, Sierra G. McCormick, and Douglas D. Ready, “Constrained Agency and the Architecture of Educational Choice: Evidence from New York City,” Annenberg Institute at Brown University (March 2024).
