The investigation of residential preference patterns and the implications of these for population distribution among different sizes of communities and between metropolitan and nonmetropolitan settings drew on a sample of the United States noninstitutional adult population, using quotas based on age, sex, and employment. By distinguishing between locations near to and away from large cities, preferences can be compared for proximity to other cities, as well as for size of place preferred. The primary sampling units were Standard Metropolitan Statistical Areas and nonmetropolitan counties stratified by region, age, and race. The respondents' actual residence was compared with their preferred ones. Almost one-half reported living in cities of over 50,000; one-third within 30 miles of such a city; and 20% in more distant locations. If everyone lived where they preferred, however, only one-quarter would live in large cities, and over one-half would be in easy commuting distance of such places. The proportion of those living in more distant areas would remain the same. Characteristics of respondents, their origins, and reasons for preferences were also given. (KM)