I wanted to share some background on the recent correction to Hall & Thompson's 2018 paper. This correction has significant implications. The conventional wisdom that moderate candidates perform better in general elections is not supported by the data when properly analyzed.🧵 - ThreadSky

adambonica.bsky.social • 5 days ago

I wanted to share some background on the recent correction to Hall & Thompson's 2018 paper. This correction has significant implications. The conventional wisdom that moderate candidates perform better in general elections is not supported by the data when properly analyzed.🧵

Comments

erinnthered.bsky.social•4 days ago

📌

bradleveck.bsky.social•3 days ago

📌

bellschristmasale.bsky.social•3 days ago

Was there a difference between republicans and democrats?

adambonica.bsky.social•5 days ago

Last year, while working on related research, I found that their reported effects of nominating extreme candidates on vote share and turnout were driven primarily by district partisanship rather than candidate ideology. When controlling for district partisanship, the main results no longer hold.

louisevans.bsky.social•5 days ago

really appreciate your work. I have to admit, maybe it's the jet lag, but I'm having trouble picking apart the different mechanisms you describe here. If you've got the time, could you walk through a quick "spherical cow" district description of the two models?

adambonica.bsky.social•5 days ago

I reached out to Andy and Dan with these findings. After reviewing my analysis, they acknowledged the issue and have now published a correction that addresses these methodological concerns. I appreciate their commitment to scientific integrity throughout this process.

adambonica.bsky.social•5 days ago

In a working paper with Nico Studen and Kasey Rhee, we use a within-precinct design that compares voting patterns for different offices in the same precinct, up and down the ballot. This approach isolates persuasion effects by holding turnout constant, measuring how ideology influences vote choice.

moreacomment.bsky.social•5 days ago

Are voters sufficiently aware of the ideological positions of the candidates down ballot to make the absence of a significant ideology effect at all surprising?

adambonica.bsky.social•5 days ago

Using this approach, we find only a *very* modest persuasion effect for even large changes in candidate ideology. When we consider how ideology affects turnout patterns in a separate analysis, the evidence suggests these tiny gains from moderation are likely counteracted by partisan turnout effects.

adambonica.bsky.social•5 days ago

I hope to write more soon about these results and what they tell us about the relationship between ideology and electoral outcomes. There are implications for candidate positioning and campaign strategy that deserve deeper exploration and can inform debates about electoral strategy.

adambonica.bsky.social•5 days ago

This is how academic discourse should work - we test each other's findings, communicate openly when issues arise, and collectively advance our understanding. Kudos to Andy and Dan for their transparency.

Comments

Posting Rules

Reply