Objective: To determine if a relationship exists between participation in a school breakfast program and measures of psychosocial and academic functioning in school-aged children. Methods: Information on participation in a school breakfast program, school record data, and in-depth interviews with parents and children were collected in 1 public school in Philadelphia, Pa, and 2 public schools in Baltimore, Md, prior to the implementation of a universally free (UF) breakfast program and again after the program had been in place for 4 months. One hundred thirty-three low-income students had complete data before and after the UF breakfast program on school breakfast participation and school-recorded measures, and 85 of these students had complete psychosocial interview data before and after the UF breakfast program. Teacher ratings of behavior before and after the UF breakfast program were available for 76 of these students. Results: Schoolwide data showed that prior to the UF breakfast program, 240 (15%) of the 1627 students in the 3 schools were eating a school-supplied breakfast each day. Of the 133 students in the interview sample, 24 (18%) of the students ate a school-supplied breakfast often, 26 (20%) ate a school-supplied breakfast sometimes, and 83 (62%) ate a school-supplied breakfast rarely or never. Prior to the UF breakfast program, students who ate a school-supplied breakfast often or sometimes had significantly higher math scores and significantly lower scores on child-, parent-, and teacher-reported symptom questionnaires than children who ate a school-supplied breakfast rarely or never. At the end of the school term 4 months after the implementation of the UF breakfast program, school-supplied breakfast participation had nearly doubled and 429 (27%) of the 1612 children in the 3 schools were participating in the school breakfast program each day. In the interview sample, almost half of the children had increased their participation. Students who increased their participation in the school breakfast program had significantly greater increases in their math grades and significantly greater decreases in the rates of school absence and tardiness than children whose participation remained the same or decreased. Child and teacher ratings of psychosocial problems also decreased to a significantly greater degree for children with increased participation in the school breakfast program. Conclusion: Both cross-sectional and longitudinal data from this study provide strong evidence that higher rates of participation in school breakfast programs are associated in the short-term with improved student functioning on a broad range of psychosocial and academic measures. [This study was made possible through grants from the Kellogg Corporation and the Mid-Atlantic Milk Marketing Association.]