The recent New York Times article on New York City’s high school admissions process describes how the incorporation of game theory into an algorithm for matching students with schools has substantially increased the rates at which students are matched to schools of their choosing. As a New York City parent whose children have twice gone through the process and twice been matched with their first-choice school, I’m grateful for this.
But like most parents—even those among the 70-plus percent who get one of their top three choices—I found the larger process to be a most unpleasant game indeed: opaque, overwhelming, and with no tools to help manage the choosing, evaluating, and ranking of options from nearly 1,700 different potential programs. The article was in fact mistitled: what improved was the matching process. The decision process was as difficult as ever if not more so, since the number of choices had increased with the proliferation of new small schools.
As consumers of myriad products and services (including educational services), we have dozens of tools available to help us make those decisions and yet for this stressful, high-stakes process, there are none. Well, there is one: a printed directory as thick as an old-timey phone book that the city prints by the tens of thousands every year at a cost of a million dollars, give or take.
That’s especially odd because the New York City Department of Education (DOE) collects and disseminates a tremendous amount of nominally public information about its schools and their outcomes, all of which resides in digital form. Unfortunately, like most districts, it has never seen fit to make that information usefully public through an open data Application Programming Interface (API). Doing so would enable anyone—community groups, policy researchers, commercial websites, hobbyists, or the DOE itself—to build a whole range of data-based tools to help families make better school choices.
The DOE’s Office of Student Enrollment is always looking for ways to better support families. But—as with many central offices—it is so occupied with managing an array of complex, large-scale processes (generally on fire) that undertaking a new initiative outside its core expertise is unlikely. Similarly, without a mandate from a senior official, the IT division, overtasked in its own day-to-day work, has no reason to undertake this on its own, either.
Since our job in the DOE’s Office of Innovation was to demonstrate how districts could invent new ways to solve old problems, we decided to lend a hand. Working alongside colleagues in the Office of Student Enrollment, we created the API and worked with the Public Policy Lab to help kids, parents, and guidance counselors define their needs. We then invited a half dozen software developers to use the data to build their own versions of useful tools to help kids and parents make these important decisions (you can see a four-minute film about that work here).
There are important discussions taking place about the appropriate gathering and use of student data, and reasonable people can disagree about what constitutes best practice. But in the meantime, we can achieve tremendous benefits with modern policies and systems for non-student data that is already publicly available but not in a truly useful form. CRPE’s new report on school choice, in which we interviewed 4,000 public school parents in eight “high-choice” cities, demonstrates that information systems that help parents and students evaluate their options are a crucial component of school choice success.
Recently, New York City Schools Chancellor Carmen Farina asserted that New York’s charter operators intentionally push out difficult kids in order to artificially boost charter school success rates. Charter supporters angrily denied those claims and editorial boards called on the DOE to release data that would support these charges. In a world of modern data APIs, there would be no “release” necessary because the data would simply be there for anyone to analyze at any time for any purpose. While we may not all draw the same conclusions from it, it is certainly better to start from the same set of facts. Open Data can help with that, too.
Steven Hodas (@stevenhodas) is a veteran of both the New York City Department of Education and the edtech industry. In this blog series, School District Innovation: When Practice Collides with Policy, he provides insights into the challenges, struggles, and opportunities of large-district attempts to reform longstanding practices and change cultural norms. This series is part of CRPE’s ongoing examination of innovative school systems.