This research employed a matched-pairs randomized field experiment design to evaluate a classroom-based health education intervention for pre-Kindergarten and Kindergarten children in a rural region of the southeastern United States. Schools were matched on demographic characteristics, then one school from each pair was randomly assigned to the treatment group and one to the delayed treatment group. The intervention included a field trip experience and an integrated curriculum designed to increase knowledge about nutrition, physical activity, and sleep. Staff conducted individual assessments of changes in knowledge with a random sample of children from each classroom (252 children from treatment classrooms; 251 children from delayed treatment classrooms). We used a multilevel linear regression with maximum likelihood estimation to incorporate the effects of clustering at the classroom and school level while examining the effects of the intervention on individual assessment change scores. During the intervention period, an estimated 3,196 children (treatment: 1,348 students in 68 classrooms in 10 schools; delayed treatment: 1,848 students in 86 classrooms in 10 schools) participated in the intervention. Children in the treatment group had significantly larger assessment change scores than children in the delayed treatment group. Findings suggest significant beneficial effects of the intervention on health knowledge. (author abstract)