Date of Completion


Embargo Period



Socio-ecological models; young adults; neighborhood environments; multilevel analysis; smoking; sedentary behavior

Major Advisor

Crystal L. Park

Associate Advisor

Amy Gorin

Associate Advisor

James Dixon

Field of Study



Doctor of Philosophy

Open Access

Campus Access


Objective: The present research investigated a) the effects of psychological and neighborhood environmental factors on changes in smoking and sedentary behaviors from adolescence to emerging adulthood and b) whether there were any cross-level interactions based on socio-ecological models. Methods: The National Longitudinal Study of Adolescent Health wave I (1995) and wave III (2001) data were used. A series of multilevel analyses (level-1: individual-level; level-2: neighborhood-level) were conducted regarding two aims. In Aim 1, it was investigated whether wave I factors predicted wave III outcomes; in Aim 2, it was examined whether wave III factors predicted wave III outcomes, when wave I outcomes were controlled. Results: In Aim 1, adolescents’ delinquency (but not depressive symptoms) predicted emerging adults’ progression to daily smokers. Adolescents living in neighborhoods with more Hispanics reported being less likely to initiate smoking and progress to daily smoking during emerging adulthood. Education level and the number of physical activity resources in the neighborhood during adolescence predicted emerging adults’ sedentary behaviors. In Aim 2, emerging adults’ delinquency was positively associated with their smoking, even when other demographics and neighborhood environment were controlled. In either Aim 1 or Aim 2, there were no cross-level interactions between individual-level psychological factors and neighborhood-level environmental factors. Conclusion: Not only the psychological factors but also neighborhood environments should receive research attention to explain risky health behaviors in emerging adults. Theory-based research with appropriate characteristics across multiple levels is required to better understand emerging adults’ risky health behaviors.