The distribution of pedestrian movement by street segment in three areas in Atlanta is modeled in relation to measures of street connectivity and land use. Although land use accounts for the pronounced differences in average pedestrian volumes per area, the connectivity of the street network affects the distribution of pedestrians on a street-by-street basis within each of them. The measures of connectivity that are used describe the density of street connections and the extent to which streets are sinuous or aligned. This study enhances previous findings, particularly those using space syntax, by better controlling for the effects of land use as compared to the effects of street connectivity and network layout. Asserting the independent role of street network design is important given that streets act as the long-term framework within which land uses change over time. The measures of street connectivity are easy to implement on a GIS platform to support the evaluation and development of designs and regulatory frameworks that promote walking, whether it be in the interest of public health, in reducing automobile dependence or in supporting vibrant urban communities.