Abstract
In this article, we demonstrate how establishment-survey questionnaire design can be improved by using different research methods such as focus groups, document-design analysis, pretests, and response-analysis surveys. We discuss the lack of research on establishment-survey methods in the past, compare the tasks of respondents in establishment and household surveys, and highlight the work of the Bureau of Labor Statistics in the designing establishment-survey questionnaires. As part of this discussion, we outline changes made to the questionnaire to reduce potential errors. The results indicate that each research method improves establishment-survey questionnaire design by identifying various types of errors.