| Description: |
This study provides novel evidence of how issue-based arguments influenced voting preferences in the campaign for the 2014 Scottish independence referendum. The findings, based on a lab-experiment and a follow-up survey, show that, even in the absence of explicit party cues, the participants evaluated and selected information in line with prior voting intentions, thus displaying three types of biases – an attitude congruency bias, a disconfirmation bias, and a confirmation bias. These correlations, however, did not occur among undecided voters, who processed information in line with the normative requirement that evidence should be evaluated independently from prior beliefs. Secondly, a between-group comparison reveals that reading a mixed set of arguments led to a one-sided persuasion effect, by increasing the support for independence mainly through reduction of indecision. This effect occurred regardless of the possibility to select the arguments and found further confirmation in the actual increase of Yes votes in the referendum. Personal economic expectations significantly moderated the effect of information. In line with prospect theory, these results suggest that risk-based calculations and economic perceptions prove important determinants of voting decisions, especially in a context of asymmetrical vote choice between an uncertain ‘Yes for a change’ and a safe ‘No for the status quo’.
|
