Theoretical work argues that citizens gain important symbolic benefits when they are represented by gender-inclusive institutions. Despite the centrality of this claim in the literature, empirical evidence is mixed. In this article, I argue that these mixed findings are—in part—because many Americans hold beliefs about women’s inclusion that are out of step with reality. Leveraging variation in survey respondents’ beliefs about women’s representation, I examine how these perceptions influence attitudes toward Congress and state legislatures. In both cases, I find that believing women are included is associated with higher levels of external efficacy among both men and women. Using panel data, I then show that when citizens’ underestimations (overestimations) are corrected, their levels of efficacy increase (decrease), shedding further light on this relationship. The findings presented in this research add new theoretical insights into when, and how, Americans consider descriptive representation when evaluating the institutions that represent them.