BIEMER, PAUL P., ROBERT M. GROVES, LARS E. LYBERO, NANCY A. MATHIOWETZ, AND SEYMOUR SUDMAN, EDS. 1991. Measurement Errors in Surveys. New York: Wiley.
This advanced and technically sophisticated book contains chapters that review and document the state of scientific knowledge on measurement errors in surveys associated with the questionnaire, respondents, interviewers, and the survey mode. Attention is also given to the issue of "modeling measurement errors and their effects on estimation and data analysis."
The chapters on questionnaire design cover the order of items within questionnaires and the choice of response alternatives. Recent work on "cognitively designed" questionnaires is included. The sections on respondents and interviewers and the errors they can contribute to surveys include problems with recall, self versus proxy responses, interviewer training, and other topics. In addition to reviewing what is known about methods meant to reduce potential measurement errors, the book also covers techniques that can help estimate the nature and size of various measurement errors and shows some ways for using these techniques to make statistical adjustments to improve the accuracy of estimates.
BRADBURN, NORMAN M., SEYMOUR SUDMAN, AND ASSOCIATES. 1979. Improving Interview Methods and Questionnaire Design. San Francisco: Jossey-Bass.
This scholarly book uses the research findings from a joint research program on "response effects in surveys" that was carried out by NORC at the University of Chicago and the Survey Research Lab at the University of Illinois in the 1970s to describe and explain survey procedures that are likely to yield accurate reporting of answers by respondents. Readers need not be statistically sophisticated to be able to understand the text, but they will need a moderately advanced understanding of research methods for the book to be of value. The authors' intended audiences are survey data collectors and students and researchers of response effects.
The chapters deal with three causes of response effects: interviewer characteristics; respondent characteristics; and "variables that derive from the nature and the structure of the interviewing task." The effects of these factors are considered separately for attitudinal questions, nonthreatening behavioral questions, and threatening behavioral questions. For attitudinal questions, task variables were more strongly related to response effects than respondent-interviewer characteristics, except when the characteristics were related to what was being measured (e.g., racial attitudes). For both nonthreatening and threatening behavioral questions, respondent memory factors were most strongly related to response effects (more so in the case of the threatening questions), whereas interviewer-respondent characteristics were unrelated to effects.
CONVERSE, JEAN M., AND STANLEY PRESSER. 1986. Survey Questions: Handcrafting the Standardized Questionnaire. Newbury Park, Calif.: Sage.
This very brief volume (only 80 pages) in a series on quantitative applications in the social sciences is the standard for brief introduction to designing and evaluating questionnaires. It consists of three chapters that the authors describe as "concentric