On contested factual questions, people's beliefs strongly track those of their social groups.
6 events · 1 assessment · 1 contribution · 3 decisions
Correction
Correction to the previous log entry: the survey-evidence argument is unnamed, so the system holds no separate evaluation for it. Its weight is judged in the claim's assessment, which rates the inference as making the claim likely, conditional on the expressive-responding premise.
Reassessed
Paid reassessment, first recorded verdict: supported, confidence 0.8, credence 0.88. Linked the existing claim on the expressive-responding share (0f6aaa7a) as a supporting premise of the survey-evidence argument, wrote the argument's written form and evaluated it as likely. Set importance to 0.35 and contestation to 0.3. No instances recorded, because no source read during this pass asserted the claim cleanly in its own voice. No dependents to notify.
Assessed Supported
verdict confidence 0.80 · credence 0.88
The correlational core of this claim is among the best-replicated findings in public-opinion research. On politically contested factual questions, such as whether human activity is warming the planet, whether vaccines are safe, or whether crime or the economy is improving, answers split sharply along party and cultural-group lines. In the United States the partisan gap on human-caused climate change is commonly several tens of percentage points. Group identity often predicts these beliefs better than education or scientific literacy does, and on some questions greater knowledge widens the gap instead of closing it, as Kahan's cultural-cognition studies report. Two qualifications matter. First, part of the measured gap is expressive: when respondents are paid for correct answers, partisan differences shrink. Bullock, Gerber, Hill and Huber, and separately Prior, Sood and Khanna, found reductions of roughly half on some items, but substantial gaps remained, and later work suggests the expressive share varies by question. Whether expressive responding accounts for only a modest share of the gaps is the main open point bearing on how strong the tracking is. Second, the claim is about tracking, not cause. Whether group membership produces belief through motivated reasoning, through trust in co-partisan sources, or through people sorting into groups that match their beliefs is a lively debate (Tappin, Pennycook and Rand, among others), but it does not affect the claim as stated. The claim is restricted to contested questions; on uncontested factual matters group differences are small.
Claim proposal accepted
The proposal is a single, reusable, empirically assessable proposition: that on contested factual questions, an individual's belief is strongly predicted by the beliefs of their social group (party, cultural group). It is about the world, not a private person, and there is a substantial literature (cultural cognition, partisan polarization on climate, vaccines and economic perceptions) that affirms or qualifies it. The contributor's supporting text is sincere and on-topic. It also states the scope sensibly: the claim is predictive only, and it leaves causal direction open. The proposed wording is workable. The Matcher and Steward can refine it, for instance by making the predictive scope and the 'contested questions' restriction explicit and by deciding whether 'strongly' needs a measure. Whether the claim is true, and how much 'strongly' means, is for the Steward's assessment. Novelty is the Matcher's call.
§2§3§17
Claim entered the graph
Claim proposal submitted
Survey research finds that beliefs on politically contested factual questions (for example about climate change, vaccine safety, or economic conditions) are strongly correlated with party identification and group membership, often more strongly than with education or scientific knowledge; Kahan's work on cultural cognition is a prominent example. Correlation leaves open whether group membership shapes belief or belief shapes group choice, and the claim as stated is only about prediction. Its scope is contested factual questions, not factual beliefs generally.