Objective

To determine whether New York City Public Schools (NYCPS) adequately managed selected aspects of the admissions process and the student-school matching algorithm used in high school admissions.¹ The audit covered selected high school admissions processes for school years 2018-19 through 2024-25.

About the Program

As the nation’s largest school district, NYCPS (formerly the New York City Department of Education) provides primary and secondary education to approximately 900,000 students, from early childhood through Grade 12, at approximately 1,600 public schools (excluding charter schools).

New York City (NYC) students apply to NYCPS’ high school programs across NYC by submitting one application in which they may select up to 12 programs.² A single school can have multiple school programs (e.g., architecture, business, computer science/technology) to which students can apply. Students and their families complete applications for high school through MySchools, the official online platform for NYC families to search for and apply to public schools. Our audit focused on high school admissions. MySchools is also used to apply to 3K, pre-kindergarten, kindergarten, and middle school. Students receive offers to programs based on four key factors: the applicant’s choices, each program’s seat availability, admissions priorities, and admissions methods. For each school year from 2018-19 to 2024-25, more than 70,000 students applied to high school.

To match student applicants to school programs, NYCPS uses a deferred acceptance matching algorithm (algorithm) that employs a logic-based system based on the Gale-Shapley algorithm. NYCPS uses this algorithm for the majority of its admissions processes to school programs. NYCPS’ Office of Student Enrollment is responsible for implementing and overseeing admissions processes at NYCPS, which includes management of the algorithm.

The algorithm uses preferences to match student applicants to school programs. Each student must have a preference of school programs, and each school program must have a preference of students. In NYCPS’ implementation of the algorithm, student preferences come directly from the order in which students list their school program selections on their application. Students could apply to a maximum of 12 school programs (i.e., admission pathways). As part of the admissions process, MySchools randomly generates and assigns each student a lottery number. The lottery number as well as how the applicant ranks the school programs on their application, the number of applicants in the priority groups, and the number of available seats for the school program are factors in determining student matching to a given program.

Many schools in NYC give priority to certain students over others. Priorities vary and can be based on where a student lives, whether the student met the eligibility requirements for the free or reduced-price lunch program, or their academic performance. How students are prioritized among each other impacts the order in which students are placed in the limited number of seats available in each school program. Additional considerations include ranking students based on academic performance and other evaluations such as essays.

The algorithm considers every student for their first-choice school program. NYCPS considers the applicants who applied to each school program and tentatively matches applicants to the number of available seats based on their order (e.g., lottery number) and priority. Every student who is not tentatively matched is considered for their next choice. This process continues until every applicant is matched or all choices of unmatched students are considered. Students who were not matched can be manually placed in available school programs. In addition, NYCPS will routinely run multiple iterations of the matching algorithm as part of its process.

Key Findings

NYCPS needs to strengthen its management of the student-school matching algorithm to ensure that all students are properly prioritized during the admissions process. We found significant weaknesses such as:

• Prioritization for students in temporary housing – For admissions into the 2023-24 school year, there were 7,000 students in temporary housing (such as homeless students) in NYC and there is no assurance these students were appropriately prioritized in the matching algorithm because the algorithm does not account for both the current and prior address—as required by the Chancellor’s Regulations for such students.
• Identifying and prioritizing low-income students – NYCPS uses data generally collected for a different purpose to identify low-income students. In a limited sample of 39 students for admission into the 2023-24 school year, we found 31 who were likely low-income because they lived with another student who was. However, NYCPS did not identify them as low-income. It is likely more students are low-income but have not been identified. Low-income students are given priority at certain school programs if NYCPS has identified them as low-income.
• Manual placement of students – For admissions into the 2023-24 school year, we found over 200 instances where students should have been manually placed in a school program but were either not manually placed or were not placed appropriately—according to NYCPS’ practices and business rules.

Despite the complexity of NYCPS’ matching process—with its many layers of criteria per school and program and the need to accommodate preferences for tens of thousands of students—we found that, for most of the algorithm processes, NYCPS had no written policies to guide implementation. Formal policies and procedures are essential to ensure the algorithm is working as intended, that all students’ applications are captured and given due consideration, and that errors are addressed and actions taken to correct them. 

Additionally, NYCPS did not publish reports timely, as required by Local Law 72 of 2018, on applications, offers, and available seats, which are key inputs and outputs of the matching algorithm. NYCPS also posted incomplete enrollment reports. Timely and complete information is an important resource for families and those involved in the management and oversight of NYC’s education system. These types of delays hinder the goals of the Local Law that mandated these reports.

Key Recommendations

• Take action to improve management of the student-school matching algorithm to ensure students are appropriately identified to receive the admissions prioritization they are entitled to as noted throughout the report including, but not limited, to:
  • Students in temporary housing
  • Low-income students
  • Students who require manual placements
• Develop and implement written policies and procedures for the matching process to ensure the consistent and appropriate treatment of students across the matching process in line with the Chancellor’s Regulations and other relevant requirements.
• Create policies and procedures that ensure complete and timely reports are publicly posted as required by Local Law 72 of 2018.

1 The audit scope did not cover the specialized high school admissions process.
2 For applications to school programs for the 2025-26 school year, which was outside of our audit testing, there is no maximum number of programs that can be selected.