Background: The increasing rates of physical inactivity and sedentary behaviors in youth (ages 6-17 years) are a significant public health concern. Physical activity (PA) levels decline with age starting in preadolescence, with marked disparities among underserved (racial/ethnic minority, low-income) youth, particularly girls. Factors within the social and physical environments appear to influence youth PA, but few studies have investigated their potential impact on accelerometer-measured activity of underserved youth. Purpose: This study examines how multi-level factors (individual: age, sex, BMI; social: transportation support; environmental: neighborhood characteristics) interact to influence a continuum of accelerometer-measured youth activity behaviors. Methods: Participants wore an accelerometer for 7 days. PA levels were classified as sedentary (<50 mg), light (50-150 mg), moderate (>150-500 mg), vigorous (>500mg), and MVPA (>=150 mg). Average daily minutes spent in PA and sedentary behaviors during waking time, the prevalence of meeting MVPA guidelines, and PA volume were estimated from accelerometer data. Three-way between-groups multivariate analysis of variance examined adjusted differences between age, sex, and weight status groups across activity outcomes. Main moderation analyses explored the effects of parent-perceived neighborhood environment and child BMI percentile in the relationship between transportation support and total child PA. Results: Participants (N=68, Mage= 9.2 years) spent an average of 103 minutes/day in MVPA and 50% of participants met MVPA guidelines. Youth engaged in more overall PA (p< .001) on weekdays than weekends. Participants were also highly sedentary, spending 65% of their waking time at a sedentary level. Older participants spent significantly more time in sedentary behavior (p <.01) and had lower PA volume (p <.01) than younger participants. Girls spent significantly more time in bouted MVPA (p <.01) and had higher PA volume (p= .01) than boys. Parent-perceived neighborhood environment (p= .02), not transportation support (p= .91), influenced total PA. The relationship between transportation support and total PA was not moderated by neighborhood environment nor child BMI. Conclusions: This study adds to the limited literature available on raw accelerometer data of underserved youth in the U.S. More research is needed to understand the underlying causes of PA and sedentary behavior patterns found in this study.