This paper applies regression analysis to cross-sectional data, derived largely from the 2006 New Zealand census, for 318 census area units served by the public transportation (PT) system in Auckland, New Zealand.1 The goal is to ascertain the determinants of PT ridership for the purpose of commuting to work in the region. The paper addresses both the modifiable areal units problem (MAUP) and spatial autocorrelation. The analysis indicates that PT utilization responds positively to an increase in the total number of commuters within a census area unit, an increase in population density, and an increase in service frequency. It responds negatively