Background: School-based measurement of children's body mass index (BMI) is a useful tool for tracking childhood obesity rates, and may be an effective intervention strategy for reducing the increasing trends in obesity. This article examines the relationship between state law, district policy, and school-level BMI measurement practices. Methods: Data were collected during 3 school years (2006-2007, 2007-2008, 2008-2009) as part of an annual study on health policies and practices in a nationally representative sample of US public elementary schools. Data collected included school-level data via a mailed questionnaire, and district-level policies and state laws from publicly available sources. We examined whether state laws and district policies were linked to school-level BMI measurement, either directly, or via a mediation effect. Results: Schools were most likely to measure student BMI if there was a state law in place (65.0% of schools) than where there was not a state law regarding BMI measurement (38.4% of schools; chi[superscript 2] = 120.91, p less than 0.001). However, school-level BMI measurement did not differ by whether the district had a relevant policy or not (49.8% vs. 49.2%, ns). These effects held up in multivariate logistic regression models controlling for relevant school-level covariates (region, race/ethnicity, location, school size, and socioeconomic status). Schools in the south and those with a majority of White students were most likely to measure students' BMI. Conclusion: State laws are associated with school-level BMI measurement, and therefore may be a helpful tool in monitoring and addressing childhood obesity. (Contains 2 tables and 1 figure.)