Participation in regular physical activity has been identified as an influential factor in promoting health for people of all ages (USDHHS, 2000). The majority of college students, however, do not engage in sufficient levels of physical activity (Keating et al., 2005). Researchers have studied the relationships between social and cognitive variables and physical activity levels, but the effect of environmental influences on college students' participation in physical activity have not been investigated. Given that physical activity habits developed in college years are likely to persist into adulthood, it is important to develop a clearer understanding of how the physical environment can be structured to facilitate the adoption of an active lifestyle. The purpose of this study was to investigate the effect of the perceived environment on college students' activity levels from a social ecological perspective. Participants were 193 college students (82 men, 111 women) from a southeastern university. They completed questionnaires assessing their perceptions of the environment (Bourdeaudhuij et al., 2003) and self-reported physical activity levels (IPAQ; Craig et al., 2003). Components of the environment that were assessed include residential density, land use mix, access to public transit, walkability, bikeability, neighborhood aesthetics, crime safety, traffic safety, connectivity, neighborhood satisfaction, neighborhood features, and access to equipment and facilities. Data were analyzed using simple correlations and multiple regression analyses to predict levels of moderate and vigorous physical activity. Land use mix, specifically access to local shopping, was positively related to students' moderate and vigorous physical activity. Bikeability, neighborhood aesthetics, perceived safety from traffic, neighborhood satisfaction, and neighborhood features were positively associated with moderate physical activity. In the multiple regression analyses, access to local shopping was a significant predictor (b = .17, p = .045) of vigorous activity, but explained only 2.0% of the variance. For moderate physical activity, neighborhood features and access to local shopping emerged as predictors (b = .23, p= .004; b = .20, p = .011), accounting for 3.8% and 3.5% of the variance respectively. These results are consistent with other studies of environmental influences on physical activity in the general population, in that environmental factors account for a statistically significant, but relatively small portion of the variance in self-reported physical activity levels. The findings provide some support for the notion that individuals' surroundings have some influence on college students' activity levels, but further investigation is needed to better understand environmental influences on physical activity patterns. Keyword(s): active participation, health promotion, research