For decades, free and reduced-price lunch (FRPL) status has been used as a proxy measure for student poverty. Families filled out paper lunch forms, and these were the basis for allocating resources to schools, defining accountability goals, and conducting research. But recent changes to the National School Lunch Program mean that FRPL status is in decline as a measure of student need, and states are turning to alternatives.

The most promising of several options is “direct certification,” which links school rosters with public benefit program data to identify students from low-income families participating in programs like the Supplemental Nutrition Assistance Program (SNAP) and Temporary Assistance for Needy Families (TANF). This approach has many advantages—namely, that it reduces the burden on families and school staff previously responsible for collecting paper lunch forms and can accurately identify many economically disadvantaged students.

But direct certification can miss low-income students and families who do not or cannot take up public benefits or who cannot be easily matched across data systems, leading to undercounts in many schools and districts. States need to be thoughtful in developing their direct certification systems, or pair them with other identification strategies, before they can generate accurate counts of low-income students.

Promises and pitfalls of using direct certification counts of low-income students

In a new brief, I describe the public benefit programs underlying direct certification, how they line up against the National School Lunch Program, and implications for new measures of student poverty.

All states can directly certify students through SNAP and could link to TANF or the Food Distribution Program on Indian Reservations (FDPIR). A few states have also participated in Medicaid pilots designed to recover more information on household income and identify students who may have formerly qualified for reduced-price lunch but do not participate in SNAP, TANF, or FDPIR. And some states include foster care participation, homelessness, and Head Start in their direct certification systems.

As I explore in the brief, these programs differ substantially from the National School Lunch Program in their federal eligibility rules. In some cases, eligibility rules vary across states and localities, and some programs have time limits for children or families with children. They also differ in their application and eligibility determination procedures, which further affects who receives public benefits and who appears in direct certification counts of low-income students.

Direct certification programs are likely to miss key populations, including students from noncitizen families and students in states with more restrictive public benefit programs. Students with varied name spellings may be difficult to match across education and public benefit databases, as well. The result is that direct certification is likely to undercount the true number of low-income students in a given school, school district, or state.

Improving new measures of student poverty

States need to address both technical and conceptual challenges to build direct certification systems that generate accurate counts of economically disadvantaged students. Massachusetts and Delaware provide promising examples of how to overcome these challenges.

In addition, I propose expanding the constellation of public programs included in direct certification systems to provide every possible opportunity for automatic data links to identify students from low-income families. Links to tax records, though burdensome, could provide additional income information for all but the poorest students. Multipliers can be used to adjust counts of low-income students at the school or school district levels, though they may not be able to shed light on which additional students are low income.

Even with improved direct certification systems, we no longer have a single, uniformly implemented measure of student poverty. The new mix of measures has implications for school funding, accountability, research, and beyond. Understanding the precise nature of these implications is critical.