Mind the Gap: Why Large Group Deficits in Political Knowledge Emerge—And What To Do About Them

Large group differences in political knowledge raise strong concerns about lasting inequities in U.S. politics. I argue such gaps emerge when factual questions operate unevenly across diverse populations, thereby inflating actual knowledge differences between groups. I illustrate this by revisiting the large knowledge deficit often observed among Latinos relative to Whites. Using a survey with traditional factual questions (e.g., what office is held by John Roberts?) and new Latino-themed items (e.g., what office is held by Marco Rubio?), I show that Whites are more likely than Latinos to correctly answer many conventional questions due to item features that are unrelated to people’s level of knowledge (i.e., item bias). Latino-themed questions, however, do not display these extraneous characteristics. Consequently, Whites and Latinos have equal odds in correctly answering these latter items, provided they have the necessary knowledge level. I also show how such item quality differences matter in practical terms. Accordingly, I establish that using a scale of unbiased items reduces the Latino-White knowledge deficit from 31 to 8 %, with the latter gap more fully explained by individual differences in established correlates of political knowledge. I then show that uneven item performance distorts inferences about knowledge’s influence on mass opinion. I discuss the implications of these results for political knowledge’s conceptualization and measurement in an increasingly diverse U.S. polity.