Pre-election opinion polls are an increasingly prominent aspect of political campaigns, yet they often vary in terms of their results, sources, and where they are published. Citizens are therefore increasingly confronted with the proposition of which polls to give more credence to than others in shaping their voting behaviour. This study investigates the relationship between subjective determinations of poll credibility and voting behaviour in the polarised context of Turkish mayoral elections. The theoretical perspective of motivated reasoning is employed to consider how individuals determine and act upon credible opinion polls in mixed information environments. Using an original two-step experimental approach conducted in 2020, this paper establishes that while the effects of opinion poll credibility on party choice are limited to the strategic considerations of the supporters of smaller parties, opinion polls can have considerable demobilising effects when polling environments are conflicting or deemed not credible. The findings produced in this study are more suggestive of accuracy-seeking than directional motivations, and they have considerable implications for how we think about the relationship between polls, politicians, and voters.

中文翻译：

---

民意调查大战：对民意调查可信度和投票行为的看法

选举前的民意调查是政治运动中越来越突出的方面，但它们的结果、来源和发布地点往往各不相同。因此，公民越来越多地面临这样一个命题，即在塑造他们的投票行为时，哪些民意调查比其他人更值得信赖。本研究调查了在土耳其市长选举两极分化的背景下，民意调查可信度的主观决定与投票行为之间的关系。动机推理的理论视角被用来考虑个人如何在混合信息环境中确定可信的民意调查并采取行动。使用 2020 年进行的原始两步实验方法，本文确定，虽然民意调查可信度对政党选择的影响仅限于小党派支持者的战略考虑，但当民意调查环境相互冲突或被认为不可信时，民意调查可能会产生相当大的复员效应。本研究产生的结果更多地暗示了寻求准确性而不是定向动机，并且它们对我们如何看待民意调查、政治家和选民之间的关系具有相当大的影响。