Measuring Issue Ownership: A Comparative Question Wording Experiment

Abstract

Issue ownership, or the idea that some parties are considered by the public to be better able or more committed to dealing with specific issues, is increasingly used in studies of electoral choice. Yet, various scholars have argued that if measures of issue ownership are confounded with party choice, this raises concerns regarding their usability to predict electoral choice. This research note examines to what extent various measures of issue ownership are confounded with voters’ party preferences and voters’ agreement with the party's position on the issue. Relying on an online survey‐embedded question wording experiment fielded in two countries (Belgium and Denmark), question wording effects for two dimension of issue ownership are examined: competence issue ownership and associative issue ownership. It is found that, in both countries, the two associative issue ownership measures were less affected by party preference and positional agreement. The most used competence issue ownership measures are most confounded with party choice and positional agreement in the experiment. Results in the two countries are largely similar, the main exception being that one of the two associative measures performs worse in Denmark. The results imply that scholars should take care in using measures of especially competence issue ownership to predict the vote, but that also country differences affect the validity of issue ownership measures.