Download data and study materials from OSF
Cindy D. Kam
University of Houston
Sample size: 1000
Field period: 7/1/2010-11/9/2010
The dominant assumptions that emerges from studies of voting behavior (i.e. Bartels 1986; Alvarez 1997; Berger et.al 2000) is that (1) voters behave in a risk-averse manner; and (2) this risk-aversion translates into powerful advantages for incumbents and/or the status quo. Yet despite the popular and theoretical expectations that the incumbent and the status quo should be sufficiently insulated, challengers to the status quo (in the form of candidates and policies) often do garner substantial support. Drawing upon the framing literature (i.e. Tversky and Kahneman 1981), we argue that even in poor economic times (such as those the nation has been experiencing), departures from the status quo can be framed as offering more or less risk compared with remaining with the status quo. Thus, this experiment examines the extent to which vote choice is affected by how departures from the status quo are framed, the risk orientations of voters, and their interaction.
H1: Framing Effects: When departures from the Status Quo (SQ) are framed in a negative light, subjects will prefer the certain, or more experienced, candidate. When departures from the SQ are framed in a positive light, subjects will prefer the probabilistic, or less experienced, candidate.
H2: Risk Orientations and Preference for Probabilistic versus Certain Candidates: The risk averse will generally prefer the certain (experienced) candidate; the risk seeking will generally prefer the probabilistic (inexperienced) candidate.
H3: Risk Orientations, Framing Effects, and Preference for "Risky" versus "Certain" Candidates: The effect of risk acceptance will be accentuated by the framing of the status quo. When compared to the non-framed control condition, framing the vote choice so that departures from the status quo are framed more positively or negatively should accentuate the effect of risk orientations on vote choice.
This is a between-subject post-test-only multiple treatment group experiment. A pre-stimulus questionnaire assessed risk orientations. The stimulus consisted of a mock newspaper article describing a local mayoral race. In all three stimuli, one candidate was be portrayed as the risky candidate (who has little experience but seeks to make dramatic policy changes) and one candidate was portrayed as the certain candidate (who has considerable experience and seeks to "stay the course"). Random assignment determined which of three framing manipulations the subject received: one where departures from the status quo are framed positively, one where departures from the status quo are framed negatively, or the control which is neutral.
Following exposure to the treatment, subjects were asked to indicate which candidate they would vote for and which candidate is better described as "risky."
Consistent with our prior work (Kam and Simas 2010), we find that subjects with greater risk acceptance were more willing to support the novice candidate. However, we find little support for our other two hypotheses. There is some evidence that framing departures from the status quo as safe can increase the likelihood of support, but overall, we find that it is the risk orientation of the individual, regardless of the way the departure from the status quo is framed, that is the significant predictor of vote choice.