PURPOSE: To examine associations of the built environment surrounding worksites and of work policies with walking behaviors. DESIGN: Cross-sectional convenience sample survey. SETTING: Workplace. SUBJECTS: Employed adults residing in Montgomery County, Maryland. MEASURES: Four different step measures taken at or near work as recorded using an accelerometer and a fifth measure indicated self-reported walking from work. Participants reported on eight built environment characteristics surrounding the worksite (e.g., the presence of sidewalks, crosswalks, and pedestrian signals) and on four worksite policies (e.g, the presence of exercise facilities and exercise programs). ANALYSIS: Cross-sectional associations of self-reported built environment characteristics surrounding worksites and worksite policies with walking behavior were examined. RESULTS: Although participants reported worksites exhibiting built environment characteristics that were supportive of walking (seven of eight characteristics were reported by >50% of participants), no built environment characteristic was associated with walking more than the median number of average weekday steps (p > or = .05). All four worksite policies were associated with walking more than the median number of average weekday steps (p < .05). In addition, a perception of few cul-de-sacs and of the presence of litter, sidewalks, crosswalks, and pedestrian signals surrounding the worksites was associated with a higher proportion of participants taking at least one walking trip from work in the past month (p < .05). CONCLUSIONS: Locating worksites in walkable environments and implementing worksite policies may favorably influence employee walking. Future studies should consider a prospective design and examine a larger, more diverse employee population and worksite environment to examine these associations.