Understanding Variability in School Food Participation 

Introduction

The rate of childhood food insecurity in New York City is higher than the food insecurity rate for the overall city population (16% in children vs 12% overall, 2016-2018), and can be as high as 30% in some communities [1].   School meals, which are available to all 1.1 million students who attend New York City’s public schools, are important sources of nutrition for food insecure children and help defray the cost of food for low-income households. Since the 2017-2018 academic year, New York City has adopted universal free school lunch throughout the city's more than 1,000 public elementary, middle, and high schools, eliminating financial barriers to participating in school food [2].  This program goes beyond the federal program in which free or reduced-price meals are provided only to eligible students following the completion of an application and verification of eligibility [3].  One of the goals of making school food universally accessible was to reduce stigma associated with free lunch, and to increase access to nutritious meals for all children. In addition to addressing food insecurity, universal school food has been associated with improved academic performance and diet quality [4].  

Despite serving lunch and breakfast at no cost to every student, however, school meal participation rates vary significantly from school to school. The CUNY FEED-NYC prototype dashboard visualizes this variability in participation by location, grade level, and need, based on a vulnerability measure that the New York City Department of Education (DOE) uses to determine the economic need of students. To create the dashboard, we used publicly available data from DOE on average daily participation (ADP) in school meals for each public school and average daily attendance (ADA) per school for the 2018-2019 academic year. We chose to visualize pre-pandemic data in the dashboard prototype because school food participation has changed significantly due to pandemic-related school closures and re-openings. We cleaned and visualized these data and divided ADP by ADA for each school to calculate the percentage of students attending school that participated in breakfast and lunch. We also visualized the percentage of students in each school participating in breakfast and lunch in relation to the students who are from low-income households  (reported by each school and largely based on students who are eligible for Human Resources Administration (HRA) benefits such as SNAP and cash assistance).  This shows participation as a function of the percentage of students in each school deemed most in need of school-based meals because of their low income. 

The school food set of dashboards visualize the level of school food participation during the 2018-19 school year at the city, borough, school district, community district, City Council district, and individual school level for breakfast and lunch. It shows the schools that have high, low, and average participation rates overall and participation as a percentage of the students in each school who come from low-income households. Users can view the data citywide and can drill down to specific school districts, community districts, city council districts, and custom geographic areas. Users can also search for data on individual schools.

Drilling Down to Details: Lunch Participation

To learn more about breakfast or lunch participation, or to search for data on individual schools, click on the dashboard buttons at the bottom right. For example, if you click on the button "Lunch," details about school lunch will be visualized. The top of the Lunch Participation dashboard shows a map of all city schools color coded by average daily participation, ranging from a low of 2.62% to a high of 100%. Using the drop-down menus you can filter the schools by borough, school type, or by three geographic areas (school district, community district, and city council district).  You can also use the slider filter to highlight schools with specific participation levels. 

Below the map we present a scatter plot of the schools selected above showing the percent participation per school and the percent of students in each school from low-income families. A line through the origin of the graph divides the schools into those where percentage participation is greater than the percentage of low-income students (upper left) and those schools with participation rates lower than the percentage of students from low-income households (lower right). The box and whisker plots on the lower right show levels of lunch participation divided by school level, illustrating the median participation rates and the wide variability of participation levels.

Hover over any point to get more information about a particular school. Click on a school in either the map or graphs and that school will be highlighted in the other visualizations.

Potential Use Case

Equipped with synthesized information and visualizations about levels of participation among different schools, parents, advocacy groups and organizations (including groups of youth food advocates), elected officials, and school board officials can identify schools that have low participation rates to explore why students choose not to eat in the school cafeteria, and why some schools have very high participation rates, to learn how to replicate successful school programs and policies and to advocate for system wide improvements to the school food program. Previous research has identified several factors associated with different school food participation rates, including whether multiple schools are co-located in one building, student overcrowding, the type of serving lines used, and higher enrollment levels [5].  Whether these or other factors affect an individual school likely requires additional research among students, cafeteria staff, and teachers. 

1New York City Hunger Report, 2019.   Hunger Free America.  https://www.hungerfreeamerica.org/en-us/survey/2019nyc-nyc-metro-region-and-new-york-state-hunger

2Watts A, Araiza AL, Fernández CR, et al. The Campaign for Universal Free Lunch in New York City: Lessons Learned. Pediatrics. 2021;148(4):e2020049734. doi:10.1542/peds.2020-049734

3Poppendieck J. Free for All: Fixing School Food in America.; 2010.

4Cohen JFW, Hecht AA, McLoughlin GM, Turner L, Schwartz MB. Universal School Meals and Associations with Student Participation, Attendance, Academic Performance, Diet Quality, Food Security, and Body Mass Index: A Systematic Review. Nutrients. 2021;13(3):911. doi:10.3390/nu13030911

5Prescott MP, Gilbride JA, Corcoran SP, et al. The Relationship between School Infrastructure and School Nutrition Program Participation and Policies in New York City. Int J Environ Res Public Health. 2022;19(15):9649. doi:10.3390/ijerph19159649