Measuring issue ownership with survey questions. A question wording experiment

• Issue ownership has two dimensions, an associative and a competence dimension.
• Existing competence measures are endogenous with party preference and position.
• The classic competence ownership measure is performing worst.
• Associative ownership measures are less conflated with party preference and position.

Issue ownership, which has a competence and an associative dimension, refers to the link between issues and parties in voters' minds. Although used frequently in voting research, there remain worries about the validity of its current measures. The measures may be confounded with respondents' (dis)agreement with parties' position and general party evaluations. Through a question wording experiment we compare measures of both issue ownership types and test which are most affected by the two confounding factors. We find that competence issue ownership measures are heavily affected by confounding factors while associative issue ownership wordings are less. Challenging existing research, we find that especially the classic ‘best at handling’-wording tapping competence issue ownership is most conflated with positions and party preference.