Irmtraud N. Gallhofer

Design, Evaluation, and Analysis of Questionnaires for Survey Research: Saris/Design, Evaluation, and Analysis of Questionnaires for Survey Research

Preface. Introduction. PART I. THE THREE STEPS PROCEDURE TO DESIGN REQUESTS FOR AN ANSWER. 1. Concepts-by-postulation and concepts-by-intuition. 2. From social science concepts-by-intuition to assertions. 3. The formulation of requests for an answer. PART II. CHOICES INVOLVED IN QUESTIONNAIRE DESIGN. 4. Specific survey research features of requests for an answer. 5. Response alternatives. 6. The structure of open ended and closed survey items. 7. Survey items in batteries. 8. Mode of data collection and other choices. PART III. THE EFFECTS OF SURVEY CHARACTERISTICS ON DATA QUALITY. 9. Criteria for the quality of survey measures. 10. The estimation of reliability, validity and method effects. 11. The split ballot MTMM designs. 12. The estimation of the effects of measurement characteristics on the quality of survey questions. PART IV. APPLICATIONS IN SOCIAL SCIENCE RESEARCH. 13. The prediction and improvement of survey requests by SQP. 14. The quality of measures for concepts-by-postulation. 15. Correction for measurement error in survey data analysis. 16. Coping with measurement error in cross-cultural research. References. Index.

Formulation of real life decisions: A study of foreign policy decisions

Abstract A sample of 235 Foreign Policy decisions from 1900 to 1955 was studied with respect to the structures of the argumentations. Although it is assumed in the literature that political decisions are complex value problems with uncertainty, the data did not show any evidence in support of this assumption. However, four groups of simplified structures of decision problems could be detected, based on restrictions specified for characterizing probabilities and/or values. Given these restrictions, it was found that more than two thirds of the decisions were in agreement with the SEU model as a normative criterion for correct decisions. But within the class of “correct” decisions the decision was always made in one and the same way. The decision makers indicated that the preferred strategy led at least to as good a result as the other strategies, but possibly to better results. In this argument probability played only a minor role. Trade offs between utilities and probabilities were not made.