ABSTRACT
This chapter explores approaches to survey design and analysis that consider the whole camel. W. E. Deming discussed survey errors due to interviewer variability, survey mode, questionnaire design, sampling variability, nonresponse, and other sources considered to be part of total survey error. Quality in surveys draws on many ideas from Deming’s work on quality improvement. Total survey design calls for an interdisciplinary approach. The areas of expertise needed to study and reduce sources of error include statistical theory of complex surveys, design of experiments, statistical process control, mixed models, cognitive psychology, management, and ethnography. The mode of survey administration exerts great influence on the coverage properties, and choice of mode should be influenced in part by the coverage that can be obtained. The sampling frame for a mail or e-mail survey is a list of physical or e-mail addresses. The chapter also presents some closing thoughts on the key concepts discussed in the preceding chapters of this book.
