Los Angeles Unified School District’s Zones of Choice (ZOC) program began in 2010 as an effort to provide more high school options for a large swath of district eighth graders, combining historical catchment areas in lower-income and lower-performing neighborhoods into larger choice zones and eliminating the default feeder system from middle to high school. ZOC has proven to be an excellent model for researching the positive academic impact of choice. But as a new report published by MIT’s Blueprint Labs shows us, it is also good for examining how parental preferences are shaped.
University of Chicago researcher Christopher Campos conducts a detailed, multi-stage analysis to go beyond the simple “revealed preferences” model typical of school choice research and really tease out what—and who—parents are responding to when they rank their preferred schools. The initial data come from ZOC information and application periods in fall 2019 and fall 2021. (LAUSD spent most of the 2020–21 school year in either full or partial virtual learning; ZOC was functional but highly disrupted.) Approximately 13,000 eighth graders attended a ZOC middle school in each application year under study. Fully 90 percent of ZOC-eligible students were Hispanic (compared to 64 percent in the district as a whole), and 94 percent were classified as low income. The typical ZOC student was entering high school performing roughly 22 percent of a standard deviation behind their typical non-ZOC peer on both math and reading tests. In short, these are exactly the students who stand to benefit most from having access to better options through ZOC.
Campos’s methodology is dense and multi-faceted and defies easy summary. On the school side, he constructs measures of school quality (based on numerous academic factors) and peer quality (that is, the composition of students in receiving high schools according to their individual academic achievement). On the family side, he surveys parents of eighth grade ZOC students prior to them submitting their ranked choice lists for high school in order to determine their beliefs about school and peer quality compared to the actual rankings he devises. On the information side—the real innovation here—Campos constructs a protocol to vary the type and amount of direct information that families receive about school and peer quality for all the schools from which they can choose. This is compared to information received from other families and through other informal sources. Finally, he compares all that pre-choice information to the final options selected, the high schools ultimately attended, and (icing on the cake) the outcomes for ZOC students based on the new understanding of the choices and how they are made.
The results of all these layers of interrelated analyses are also very detailed. Pre-choice survey data show that families tended to underestimate schools’ academic quality, overestimate their peer quality, and that their estimates were least accurate regarding schools they preferred the least. All of these tendencies combined to induce what the report calls “mistakes” on ZOC applications. That is, in a scenario without these biases—pure information about school and peer quality—his models show that parents would rank schools differently. However, families are motivated to choose by the information they receive, right or wrong. Families who received more information, especially those who received it directly from Campos, chose higher-quality schools; those who did not receive direct information but heard about school quality from their Campos-informed neighbors also chose higher-quality schools. It’s not as simple as a quick summary makes it sound, but overall, Campos writes, “the experiment provides robust evidence that when properly informed, families make choices in a way that is consistent with rewarding effective schools and that social interactions are important mediators governing changes in demand.” In previous research, he has noted that ZOC is nearly ideal at offering enough options without overwhelming parents and at providing them with the highest-quality information possible.
While the ultimate aim of this research project is to determine whether ZOC students fare better in their chosen high school, outcome data are limited at this time. Only the 2019 cohort had reached eleventh grade state testing when this analysis was being done, and Campos believes pandemic-mitigation school closures “severely interfered with [their] educational experience.” As a result, no impact of ZOC choices on their test scores was observed. However, survey data showed significant improvements in students’ reported happiness in school, interpersonal skills, school connectedness, and academic effort for the 2019 cohort. Incidents of bullying were also down. These same improvements were reported—at even higher levels—for the 2021 cohort. Campos specifically notes that the academic effort index he constructed from survey responses is more than ten times higher for the 2021 ZOC cohort than the boost reported by the earlier group. He feels it bodes well for an observable test score increase when those students reach eleventh grade, but only time will tell for sure.
Overall, this research—as in-depth and pathbreaking as it is—is not conclusive except to say that parental school preferences are influenced by the information they receive, both from official and unofficial sources. Schools (and indeed policymakers) who are serious about helping families make the best possible educational choices must provide thorough, clear, and digestible information directly to families prior to any application or ranking of schools.
SOURCE: Christopher Campos, “Social Interactions, Information, and Preferences for Schools: Experimental Evidence from Los Angeles,” MIT Blueprint Labs (April 2024).