ABSTRACT

Opinion polls suggest that Vladimir Putin has broad support in Russia, but there are concerns that some respondents may be lying to pollsters. Using list experiments, we revisit our earlier work on support for Putin to explore his popularity between late 2020 and mid-2022. Our findings paint an ambiguous portrait. A naive interpretation of our estimates implies that Putin was 10 to 20 percentage points less popular than opinion polls suggest. However, results from placebo experiments demonstrate that these estimates are likely subject to artificial deflation – a design effect that produces downward bias in estimates from list experiments. Although we cannot be definitive, on balance our results are consistent with the conclusion that Putin is roughly as popular as opinion polls suggest. Methodologically, our research highlights artificial deflation as a key limitation of list experiments and the importance of placebo lists as a tool to diagnose this problem.

摘要

民意调查显示，弗拉基米尔普京在俄罗斯得到广泛支持，但有人担心一些受访者可能在向民意调查机构撒谎。使用列表实验，我们重新审视了我们早期关于支持普京的工作，以探索他在 2020 年底和 2022 年中期之间的受欢迎程度。我们的发现描绘了一幅模棱两可的肖像。对我们估计的天真解释表明，普京的受欢迎程度比民意调查显示的低 10 到 20 个百分点。然而，安慰剂实验的结果表明，这些估计可能会受到人为通货紧缩的影响——这是一种设计效果，会在列表实验的估计中产生向下偏差。虽然我们不能确定，但​​总的来说，我们的结果与普京大致与民意调查所显示的一样受欢迎的结论是一致的。在方法论上，