Is Free and Reduced-Price Lunch a Valid Measure of Educational Disadvantage?

Abstract

Students in the United States whose household income is less than 130% of the poverty line qualify for free lunch, and students whose household income is between 130% and 185% of the poverty line qualify for reduced-price lunch. Education researchers and policymakers often use free and reduced-price lunch (FRPL) status to measure socioeconomic disadvantage. But how valid is this measure? Linking IRS income tax data to school administrative records for all eighth graders in one California public school district and Oregon public schools, we examine how well FRPL enrollment captures student disadvantage. We find that FRPL categories capture relatively little variation in household income. However, FRPL captures elements of educational disadvantage that IRS-reported household income data do not.

Disparities and Discrimination in Student Discipline by Race and Family Income

“…We omit sometimes-FRPL students from these comparisons in order to draw a relatively sharp contrast between poor and non-poor students. Students in the sometimes-FRPL group (24% of the student-year observations) might have experienced changes in family income over the period observed-or might simply have administrative data that do not reflect their true financial situations in some years (Domina et al, 2018). However, we retain this sometimes-FRPL group when using family income as a covariate to keep from dropping observations.…”

Does the Measurement Matter? Assessing Alternate Approaches to Measuring State School Finance Equity for California’s Local Control Funding Formula

“…First, we exchange the census-based poverty rate with the percentage of students eligible for FRL. As noted in recent research, the two measures are moderately correlated (.79 in California across all years; see Chingos, 2016;Domina et al, 2018;Harwell & LeBeau, 2010). To estimate resource gaps, we predict spending and funding levels for districts with 0% and 80% FRL students, corresponding to approximately the same percentiles of census poverty rates (roughly the 1st and 90th percentiles, respectively).…”

Critiques of socio-economic school compositional effects: are they valid?

“…Not every educational dataset meets this requirement. For example, some US research can only measure SES by student access to free and reduced-price lunch eligibility (Domina et al, 2018). It is possible to measure a SES factor with two-indicators, but such models require other factors to be identifiable (Bollen, 1989).…”