Replication Data for: Unresponsive and Unpersuaded: The Unintended Effects of Voter Persuasion Efforts

Can randomized experiments at the individual level help assess the persuasive effects of campaign tactics? To answer that question, we analyze a field experiment conducted during the 2008 presidential election in which 56,000 registered voters in Wisconsin were assigned to persuasive canvassing, phone calls, and/or mail. We find that persuasive appeals by canvassers had two unintended consequences. First, they reduced responsiveness to a follow-up survey among infrequent voters, a substantively interesting behavioral response that has implications for the statistical analysis of persuasion experiments. Second, the persuasive appeals possibly reduced candidate support and certainly did not increase it. This counter-intuitive finding is reinforced by multiple statistical methods and suggests that contact by a political campaign can engender a backlash.

个体层面的随机实验（randomized experiments）能否助力评估竞选策略的说服效果？为解答该问题，我们对2008年美国总统大选期间开展的一项田野实验（field experiment）展开分析：该实验将威斯康星州的5.6万名登记选民随机分配至说服性拉票（persuasive canvassing）、电话游说及邮寄宣传三个组别。研究发现，拉票人员发起的说服性沟通存在两项意外后果：其一，该类沟通会降低非高频选民对后续跟踪调查的参与意愿——这是一项兼具实质研究价值的行为响应，对说服实验的统计分析具有重要启示；其二，这类说服性沟通可能削弱候选人支持率，且完全未起到提升支持率的作用。这一反直觉的发现通过多种统计方法得到验证，表明政治竞选活动的接触行为可能引发选民反弹（backlash）。