Presents a comprehensive treatment of survey errors and the costs and benefits of alternative survey designs, integrating statistical and social science perspectives on survey design. Considers coverage error, nonresponse error, sampling error, and measurement error, including the effects of interviewers and respondents, the wording of the questionnaire, and mode of data collection. A review of the social science and statistical literatures on survey errors is included, and the relationships between the different types of errors are explored. Presents cost models designed to reduce the various types of errors.