Panel analysis shows negligible support change among Trump backers, contrasting cross-sectional estimates and highlighting implications for event study literature.
How can we measure the effects of campaign events? We estimate how voters respond to a prominent campaign scandal—Donald Trump being convicted of 34 felony counts of falsifying business records—using data from a large, eight-wave panel study with waves fielded before, during, and after the verdict. We find the trial had virtually no effect on Trump supporters, even those who reported their support was conditional on an acquittal. We compare this precisely estimated null to estimates from popular cross-sectional methods, which fail to replicate it. The “change” question format estimates a 6% increase in support, while the “counterfactual” survey experimental design estimates a 10% decrease. We formalize the estimands each method estimates and discuss implications for the event study literature.
No takes yet. Share an insight, caveat, or question.
Lockhart et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: