Evaluating Survey Data: Making the Transition From Pretesting to Quality Assessment
James L. Esposito, Bureau of Labor Statistics and Jennifer M. Rothgeb, U.S. Bureau of the Census
KEY WORDS: Questionnaire Design, Questionnaire Pretesting, Data Quality
ABSTRACT
The intent of this paper, in general terms, is to describe the problem-solving behavior of survey researchers who engage themselves in efforts to detect and to minimize sources of measurement error. We are concerned specifically with efforts by survey researchers to obtain high quality survey data through improvements in questionnaire evaluation and design. In the first section of the paper, we address the issue of survey quality and provide a brief historical review of some of the advances that have contributed to the quality movement within the field of survey methodology. In section 2, we identify and discuss some of the techniques that have been developed to evaluate the clarity/efficacy of survey items and the overall quality of questionnaires. We review how others have used these techniques to improve the quality of data obtained from interviewer-administered questionnaires. In section 3, we report on how some of these methods were used to measure improvements to the data quality of the redesigned Current Population Survey (CPS). And in section 4, we close with a discussion of some of the advantages and disadvantages associated with quality assessment research.
CITATION: 1995, Monograph paper prepared for presentation at the International Conference on Survey Measurement and Process Quality, Bristol, U.K.
