They Know What They Know and It Ain’t Much: Revisiting the Dunning–Kruger Effect and Overconfidence in Political Knowledge

Abstract

The difference between what Americans know about politics and what they think they know about politics is described as a classic case of Dunning–Kruger, a cognitive bias characterized by overestimation of competence. That bias has potentially malign civic implications, making interventions that reliably reduce overconfidence an issue of high importance. Here, we report the results of a quasi-experiment testing one such intervention among undergraduate students enrolled in an intro political science course: Providing feedback on levels of political knowledge. This naturally occurring experiment was repeated 11 semesters over a half-decade span with consistent null results. These findings are primarily explained by a failure to observe the expected Dunning–Kruger effect—in contrast to the existing literature we find little evidence of such a phenomenon in the realm of political knowledge. Our results suggest that prior reports of overconfidence in political knowledge may be overstated or an artifact of measurement approach.