I: Compromise Effect. Adding a new, more extreme alternative or candidate to a choice set will increase support for the nearest, more moderate, options.
II: The introduction of an option will most likely increase support for the moderate or compromise alternative when the decision maker finds the choice to be difficult.
I test these hypotheses using a between-subject experiment using profiles of candidates running for “local office.” The control condition includes two candidates with platforms emphasizing their positions on growth and development in the area. I compare behavior in this condition to three treatments that introduce a third candidate to the race, one of whom takes a moderate stance, one is more extreme, and one who stresses an unrelated issue.
Contrary to traditional, spatial theories of voting, support for the adjacent candidate does not fall in proportion to the share of the vote earned by the extreme candidate. Instead, I observe an attraction effect benefiting the adjacent (moderate) candidate. Consistent with literature on compromise effects in consumer behavior, I find that when voters find the choice to be difficult, they are more likely to support the candidate that appears to be a moderate when all the candidates take positions on the same issue. This effect is in addition to changes in how voters perceive the adjacent candidate and strategic considerations about the outcome of the election. When a candidate is added that does not address the same growth issues, the respondents’ own priorities highly influence their choice of candidate.
- Compromise effect,
- menu-dependent preferences,
- multicandidate elections,
- third parties
Available at: http://works.bepress.com/renan/14/