Beyond the role of physiological changes that coincide with puberty, multiple social and environmental factors perpetuate poor sleep health in adolescents. However, there is limited research examining their relative importance. This Stress, Trauma, and Resilience (STAR) Seminar will explore the investigation that used two popular machine learning approaches to identify the relative significance of key factor indicators of multiple dimensions of sleep, including bedtime, sleep duration, and social jetlag in a diverse sample of adolescents. Participants included 3,381 adolescents aged 15 to 19 years from the Future of Families and Child Wellbeing Study. Variables spanning sociodemographics and neighborhood context, sleep behaviors, activities, psychopathology, family, school, and physical factors were entered into Least Absolute Shrinkage and Selection Operator (LASSO) and random forest machine learning models for variable selection and establishing order of importance. Results and implications for adolescent sleep management will also be discussed.