Double-crested cormorants (Phalacrocorax auritus) and boats of 2 length classes (≤ 8 m and > 8 m) were counted from a boat along 31 established strip transects in western Lake Erie from 24 April to 1 September 2000. Each transect included only one of the following habitats: (1) offshore of a breeding island or roosting/loafing area for cormorants (“refuge”), (2) reefs or shoals, (3) open water, or (4) offshore of an island shoreline that had evidence of development by humans. Foraging cormorants were recorded most often offshore of refuges and least often on open water. There was no difference between the numbers of foraging cormorants/km² recorded offshore of developed shorelines and on reefs and shoals. More than half of all boats recorded were on transects that were within 1 km of developed shorelines. Among those transects > 1 km from developed shorelines, there were no differences among the habitats for the number of boats of either length class. The respective ranks of the 31 transect means of the numbers of cormorants/km² and the numbers of boats/km² in either length class were uncorrelated. The results suggest that (1) cormorants select foraging habitats based mainly on shoreline type, distance from shoreline, and depth, and (2) the amount of boat traffic is influenced by proximity to port and trip objectives, including sport angling and recreational boating. Although there is overlap in habitat use by cormorants and humans, this overlap is not complete. The perception of cormorants as a threat to fish populations may in part be due to this overlap.