Density, as a concept and metric, is widely used to describe the built environment; however, this complex topic deserves further attention because it is inadequate at describing physical and spatial relationships. In the wake of increased urbanization it will be crucial to merge quantitative and qualitative properties with the discussion of density. Residential suburban communities, in the United States, are often designed to achieve a low dwelling unit/area density with its inhabitants preferring an antidote to the perceived congestion and crowd of the urban core environment. In spite of automobile use and land conversion contributing to increased concentrations of carbon dioxide in the atmosphere; car oriented developments, consisting of single family detached homes on individual lots along wide streets, are ubiquitous. The question this study asks is; how do residential built environment spatial characteristics influence the perception of low density? In other words, can we design an environment which is perceived as low density, while utilizing less land area than its actual low density counterparts? A survey with residential street scenes was used to investigate this question. Three housing typologies (single family homes, row houses, and stacked row houses) and three spatial characteristics (street width, set back distance, and tree coverage) were systematically altered and combined in graphically represented images of the residential street scene. The survey was sent to 400 randomly selected inhabitants of Beaverton, Oregon who were asked to choose the scene they felt was the most spacious and most preferred, from sets of stimuli, using discrete choice modeling.