Scott Alexander, curated
← Back to curation

SSC Survey Data On Models Of Political Conflict

Quality
63
Strong
Claude Shift
40
Moderate
RWI
2
of 10

Summary

An empirical companion to his influential 'Conflict vs Mistake' essay, testing whether the distinction captures something real using SSC survey data. The closest item ('is bad politics from intellectual or moral failure?') split 80% intellectual-failure among blog readers vs 60% on Mechanical Turk, and by affiliation tracked theory rather than left/right: Marxists most conflict-theory, libertarians most mistake-theory, with intra-coalition divergence exceeding the left-right gap. He combines several items (the political-disagreement questions plus two clever forced-choice categorization questions -- the Nazi-violence and gay-marriage-civil-disobedience triads that sort respondents by principle-applied vs side-won) into an ad-hoc conflict-theory measure, cross-checks with a self-deprecating factor analysis, and reports demographic correlates (autism didn't matter, against his pre-registered prediction; worse finances, higher self-rated morality, and neuroticism/extraversion correlated slightly with conflict theory). Honest about its limits and posts the data. A competent, genuinely interesting empirical operationalization, but a data-supplement dependent on the parent essay for its framework.

Why this score

Quality 63 · Strong. Low-Strong (63): a competent original-data analysis that operationalizes and partly validates a real framework (conflict vs mistake theory), with interesting affiliation patterns and admirable methodological honesty. Held in the low-Strong band because it's an empirical companion to the main essay rather than a standalone idea, with modest and somewhat inconclusive findings.

Claude’s paradigm shift 40 · Moderate. Moderate (40): turning conflict/mistake theory into measurable survey items is a modest novel contribution, but the framework itself comes from the parent post.

Real-world impact 2 · Minor. An original-data companion operationalizing the conflict-vs-mistake framework via SSC survey data; substantive within the discourse but an empirical adjunct to the main essay, with no material-world reach → RWI 2.