The Measurement of Public Opinion about Public Policy: A Report on Some New Issue Question Formats

Abstract

The ability to measure mass attitudes on policy matters is intimately related to the format in which policy questions are presented to respondents. In this paper we examine a set of alternate question formats and contrast them with the seven-point format used in the more recent election studies of the Center for Political Studies (CPS) of the University of Michigan. The analysis is based on a sample of 280 respondents interviewed by the CPS in 1979. Our results indicate that some modifications of the traditional format seem warranted. Responses to the traditional format are difficult to interpret, produce an excessive number of responses in the middle category, and fail to allow for respondent ambiguity. Two of the new formats offer attractive alternatives: the branching format presents issue questions in a manner analogous to the traditional party identification question and seems to resolve at least partially the interpretation and middle-category problems; the ambiguity format duplicates the seven-point format but allows respondents to answer in ranges as well as at specific points on the issue scale.