Not Another Politics Podcast Do Politicians Really Have A Conservative Bias?
Aug 21, 2025
In this discussion, Adam Zelizer, a political scientist from the University of Chicago, challenges the idea that politicians are biased towards conservatism. He introduces the concept of "midpoint bias," revealing how legislators often misinterpret public opinion due to survey errors. The conversation dives into the complexities of political perceptions, the impact of polling question framing, and the discrepancies between voters' self-perceptions and their true political beliefs. Zelizer's insights shed light on the messy dynamics of democracy and representation.
AI Snips
Chapters
Transcript
Episode notes
Author's Spark: Noisy Answers From Politicians
- Adam began this work after noticing wide noise in politicians' survey responses from his own surveys.
- He observed extreme answers (0% to 100%) and many mid-50% focal responses suggesting inattentive or expressive answering.
Midpoint Bias Explains Apparent Ideological Error
- Adam Zelizer finds a consistent 'midpoint bias' where politicians' survey answers regress toward 50%.
- That bias can create the illusion of a directional ideological error when surveyed items mostly favor one side.
Survey Mix Drives Perceived Directional Bias
- Zelizer shows the apparent conservative bias in prior studies reflects asking mostly liberal-leaning items combined with midpoint bias.
- On conservative-leaning items politicians display the opposite (liberal) bias, consistent with measurement error.