Replication Data for: Model Uncertainty, Political Contestation, and Public Trust in Science: Evidence from the Covid Pandemic

Policymakers rely on scientific models, but the inherent uncertainty of underlying assumptions and data renders models susceptible to political contestation in ways that threaten public confidence in the models and the policies they inform. We investigate the politics and policy implications of Covid models predicting the virus’ course. We conduct five experiments that probe how the partisan identity of the cue-giver and the substance of the cue – particularly how it frames uncertainty – affect public trust in science and support for science-based policy. Democratic criticism undermines trust more than criticism from Republicans, even when it was not the intention of the signaling elite. Emphasizing and exploiting uncertainty in projections can erode public trust. Downplaying uncertainty and exaggerating the dire consequences of failing to heed model projections can raise support in the short-term but backfires when projections prove wrong. Transparently acknowledging and contextualizing uncertainty avoids these backfire effects without threatening public trust.