Publication Date




Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Psychology (Arts and Sciences)

Date of Defense


First Committee Member

Patrice G. Saab

Second Committee Member

Maria M. Llabre

Third Committee Member

Annette M. La Greca

Fourth Committee Member

Jennifer C. Britton

Fifth Committee Member

Nicholas D. Myers


This study investigated how individual-level health behaviors as well as school-level socioeconomic status (SES) and school climate impact adolescent health outcomes. Most of the research supporting an association between adolescent health and social contextual factors has been conducted at the neighborhood-level. However, schools are also influential contexts in adolescents’ lives and their consideration in studies of adolescent health outcomes is warranted. The present research expands upon the existing literature in the area of school contextual effects on health, as it addresses these issues in an ethnically diverse sample of 84,165 adolescents using multilevel techniques to properly account for the nested nature of these data. Student-level demographic and health behavior data were collected from 10th-grade students attending traditional public senior high schools in Miami-Dade County during school-based blood pressure (BP) screenings for six academic years, beginning in 1999-2000 and ending in 2004-05. School-level demographic, achievement, and climate data were derived from a publicly available database of Florida School Indicators. Separate hierarchical linear models for boys and girls were used to investigate the contribution of school-cohort effects to the overall variance in adolescents’ BMI, systolic BP (SBP), and diastolic BP (DBP). In support of study hypotheses, it was found that several school-level environmental factors, including student-teacher ratio, SES, and safety of the school, were associated with BMI and SBP, particularly in girls. As predicted, lower school-level SES was associated with increased BMI in both boys and girls, such that boys in lower-SES school cohorts have higher average BMIs by 0.010 kg/m2, while girls in such school cohorts have higher average BMIs by 0.014 kg/m2, adjusting for student-teacher ratio, total student enrollment, and additional school cohort-level variables. However, this relationship was not found for either blood pressure outcome. As hypothesized, student-teacher ratio, a proxy for neighborhood overcrowding, was directly associated with BMI for both boys (gamma = 0.046, p < 0.01) and girls (gamma = 0.049, p < 0.01). However, this relationship was again not found for SBP or DBP. As predicted, FCAT passing rate, a measure of school achievement, was inversely associated with BMI, but only in girls (gamma = -0.015, p < 0.01), while school safety rating was inversely associated with SBP, again only in girls (gamma = -0.719, p < 0.01). Other positive school indicators were not significantly associated with any of the study outcomes. None of the school cohort-level factors were associated with DBP for either gender. In support of Hypothesis 3, more school cohort-level factors appear to influence girls’ health outcomes compared to those of boys, but differences in the effects of these predictors did not reach statistical significance. A number of student-level factors, including age, ethnicity, and health behaviors, were associated with the 3 study outcomes, BMI, SBP, and DBP as well. Findings suggest that, in addition to an adolescent’s health behaviors, where the adolescent attends school is relevant in terms of health outcomes and cardiovascular risk. Poor weight status and elevated blood pressure may result, in part, from school contextual influences beyond the adolescent’s control and not solely individual action. Thus, policy changes to impact school environments on a larger scale, rather than focusing on individual behavior change, may help students more effectively maintain healthy weights and blood pressures. As the present study found associations between adversity at the school level and health outcomes, future work may include identifying those adolescents at highest risk of poor health outcomes as a result of school environment.


Adolescent health; socioeconomic status; school environment; cardiovascular risk; multilevel modeling