Onderzoeksartikel

Classificatie van survey-vragen

---

Abstract

A classification of questions can be very useful for the formulation of interviews and questionnaires. One can use such a classification to select the relevant concept and the proper question type and generate a question. Such a classification can also be used in research to evaluate the quality of survey questions. On the basis of such a classification one can study whether the results are different if different question types are used to measure the same concepts. The formulation of questions as well as the study of data quality require a classification of the content of the questions and a classification of the form of the questions. Many concepts can be found in the literature, but few systematic accounts (e.g. Oppenheim, 1966; Converse & Schuman, 1984; Bradburn & Sudman, 1988; Smith, 1987). No clear connection is made between the classes of concepts and the type of question which could be used as an indicator. In a recent paper Saris and Gallhofer (1997) have tried to design such a classification of questions for the most commonly used concepts in the literature and they have tested the classification on questions formulated in English. In this new study the coding rules have been made explicit and the coding is tested on a representative sample of opinion questions collected by Molenaar (1986) from the Steinmetz archive. With this test we can determine if the classification fits to the questions which are normally asked in Dutch survey research. It turned out that the classification fitted very well. Besides, we also got a good impression of the problems which arise formulating questions for survey research.

How to Cite:

Saris, W. & Gallhofer, I., (1998) “Classificatie van survey-vragen”, Tijdschrift voor Communicatiewetenschap 26(2), 96–121.