Some problems of policy-related attitude surveys - with examples from the davis-sinnott report

Abstract

This paper offers a criticism of some of the problems involved in policy-related attitude surveys. It argues that the assumptions made in order to generate information useful to policy-makers are difficult to defend. The four particular assumptions examined concern the nature of the problem towards which people are presumed to have attitudes, the nature of attitudes themselves, the belief that attitude research is descriptive and finally the relationship between the nature of public opinion and the choice of a research methodology. The argument is illustrated with examples from the report by Davis and Sinnott (1979) on attitudes to the Northern Ireland problem.