The effect of question wording on responses to the Constitutional Council questions

Abstract

The focus of this study was to evaluate the questions used in the advisory referendum on the proposals for a new Icelandic constitution by the Constitutional Council on October 20, 2012. Cognitive interviews (N=60) were used to evaluate if the questions are understood in a consistent manner. Two survey experiments were conducted where three different versions of the questions were used; a) the original version, b) a version estimating the effect of status quo on responses, and finally c) a version where the prefix “are you opposed” was used instead of the prefix “would you like” used in the original questions. A web survey was conducted using both a sample of university students (n=209) and a social media sample (n=528). The first hypothesis was that people would be more likely to agree with the status quo when the question did not involve change. The second hypothesis was that people would be less likely to agree with a cause by disagreeing with a negatively worded question (“are you opposed”) than agreeing with a positively worded question (“would you like”). The results indicated that a status quo effect on responses was found in two questions in the university student sample and three questions in the social media sample and an effect of using a negatively worded prefix was found in two question in the social media sample but not in the university student sample.