AbstractThe theory of human functioning and school organisation informed by Basil Bernstein?s sociology of education suggests that to gain the commitment and promote the health of students, particularly those from disadvantaged backgrounds, schools require radical transformations eroding various ?boundaries?: between and among staff and students; between students? academic learning and broader social development and welfare; and between schools and their local communities. Existing research examining this theory has reported associations between school-level proxy measures of student commitment and lower rates of student smoking, drinking alcohol, use of drugs and violence. But this research has not directly assessed whether reduced school boundaries explain this. We piloted a new scale derived from teacher reports to measure unhealthy school boundaries and examined its inter-item reliability and its criterion validity in terms of associations with various measures of school commitment and smoking. Data on boundaries came from 101 teachers across 40 schools. Data on student commitment and smoking came from 6667 students. We assessed reliability by examining correlations between scale-items and criterion validity in terms of associations with student-reported commitment and smoking. Inter-item reliability was sub-optimal but better within the subscales about boundaries between academic/broader learning and schools/local communities. The scale had good criterion validity, strongly associated with reduced student-reported school commitment and increased student-reported smoking. We reflect on the implications of these findings in terms of critical perspectives on health promotion in schools and the strengths and limitations of quantitative research in examining health behaviours as opposed to practices.