Sampling from a universe of items and the De-Machiavellization of questionnaire design
Peter Ph. Mohler · Social Science Open Access Repository (GESIS – Leibniz Institute for the Social Sciences) · 2006
"Up-to-date design and development of survey items is like a medieval rite rather than a quality-controlled scientific undertaking. This holds despite existing evidence-based protocols propagated for many years by methodologists. Most prominent is the creation process of items, which in general is not a systematic selection from a universe of items belonging to a specific construct. A case is made here, to give up the round-table ritual of item creation in favor of a well-defined selection process. A differentiation of work roles is advocated in the light of knowledge accrued in other disciplines to streamline the item design and selection process for future surveys." (author's abstract)