This repeated cross-sectional national telephone survey of households was conducted as part of the evaluation of the Robert Wood Johnson Foundation's Urban Health Initiative (UHI), a long-term effort to improve the health, safety, and well-being of children and youth in five economically distressed cities in the United States: Baltimore, MD, Detroit, MI, Oakland, CA, Philadelphia, PA, and Richmond, VA. The UHI Survey of Adults and Youth (SAY) included a variety of questions, asked of both parents and their 10-18 year old children, regarding children's health, safety, perceptions of neighborhoods and schools, family relations, quality of city services, and other issues. SAY surveyed 3 types of households -- households without children, households with children aged 0-9 years, and households with children aged 10-18 years -- in up to 14 geographic areas, including the 5 UHI program cities, 9 comparison cities demographically similar to the UHI cities, the suburban regions of these cities, the most populous 100 United States cities, and the rest of the country. There were 3 waves of SAY fielded during the course of the UHI project: during the 1998-1999, 2001-2002, and 2004-2005 school years.The vast majority of survey items in SAY are from other national surveys, including the National Longitudinal Survey of Youth, the Youth Risk Behavior Survey, and the National Health Interview Survey. The survey instrument was revised between waves, but about 50 percent of the instrument remained the same across all 3 waves. All adult respondents (both parents and nonparents) were first asked questions about services and general conditions in their city. Parents were then asked additional questions about their children, their children's schools, child-related services in their communities, neighborhood and city conditions, and brief question segments about randomly selected children in various age groups. If one or more 10-18 year olds lived in the household, and if the parent respondent gave permission, up to two 10-18 year olds in the household were then interviewed about their schools, neighborhoods, peer group activities and behaviors, out-of-school time, relations with adults, and other topics. There is a separate data file for each wave, and each record contains all of the data for a given household, i.e., the data collected from one adult and up to two 10-18 year olds.