PUMA Survey 5.3. Insights in societal changes in Austria

Full edition for scientific use. PUMA Surveys consist of separate modules designed and prepared by different principle investigators. This PUMA Survey consists of two modules: MODULE 1 "Trick of the Traits. An experimental study on trait ownership and mediated leader effects", MODULE 2 "An Experimental Assessment of Approval and Evaluative Voting". Fieldwork was conducted by MARKETAGENT.<br><br/> MODULE 1: Trick of the Traits. An experimental study on trait ownership and mediated leader effects (Loes Aalerding, Sophie Lecheler) <br>
This study tests, by means of a survey experiment, how leader perceptions are affected by media portrayals of party leaders in terms of their leadership traits, and to what extent partisan stereotypes and trait ownership moderates this relationship.
Research has shown that citizens’ subjective party leader perceptions, especially in terms of leadership traits, affect voting behavior (e.g., Bittner, 2011; Aarts, Blais, & Schmitt, 2013). What remains a largely unresolved question, however, is which trait evaluations matter most. The main goal of this study is to test how media messages of party leaders in terms of their leadership traits affects voters’ perception of those party leaders and to what extent trait ownership moderates this relation. The contribution of the study is threefold. First, it takes into account that current political life is highly mediatized by focusing on mediated leader effects. Second, it strengthens the causal claim of (the conditionality) of leader effects by using an experimental research design as opposed to correlational data. Third, it is the first to test the theory of trait ownership in Austria and therefore (completely) outside the two-party context of the US. <br/><br>MODULE 2: An Experimental Assessment of Approval and Evaluative Voting (Philipp Harfst, Jean-Francois Laslier, Damien Bol) <br>
In our PUMA module, we ran an online survey experiment in which we asked a representative sample of the Austrian population to cast a vote. We created a ballot to similar to the one of the 2017 election of the National Council. The respondents saw on their screen the main parties and the main candidates of these parties. Then, they had to indicate their preference for one of the parties and for 15 individual candidates within this party.
The experimental treatment is the type of preference vote the respondents could cast to express their preference for individual candidates. A third of the respondents (randomly selected) could choose to approve each of the candidates or not [0,1]. This binary system is often called Approval Voting (AV). Another third of the respondents (randomly selected) could give 0, 1, or 2 points to each of the candidates. The last third of the respondents could give a positive, a negative, or no points to each of the candidates [-1,0,1]. These last two systems are two different versions to what is usually referred to as Evaluative Voting (EV). The goal of our research is to study the effect of the type of preference voting on voters’ decisions.
The survey was fielded in June 2018 and targeted the population of eligible Austrian voters. The sample size is 700 respondents, and is representative of the Austrian population in terms of gender, age and education. The survey was conducted online, which is the best survey model for this type of study. Unlike telephone interviews, online surveys allow for a visualisation of the ballot, which helps improve the quality of responses. Also, this way of asking for respondents’ vote choice has already been successfully implemented in other contexts (Laslier et al. 2015). <br/>