Abstract

This study reexamines the foundational claim that demographic composition, particularly racial makeup, shapes legislative behavior in the United States. Drawing on a novel dataset that I helped digitize from a government census almanac book, which includes 2,630 state legislators across 18 racially salient states from 1995 to 2000, I test whether the percentage of African American residents in a legislative district predicts the roll-call ideology of its representative. These findings challenge conventional wisdom. While bivariate models suggest a strong negative relationship, with Blacker districts electing more liberal lawmakers, this association collapses under minimal statistical conditioning. Once party affiliation, state context, and district socioeconomic characteristics are held constant, racial composition has no substantively or statistically signifcant effect on roll-call ideology. This null result holds across model specifications, racial thresholds, partisan subgroups, and robustness checks. The one exception, majority-Black Democratic districts, reflects ideological selection rather than policy responsiveness. These results force a reconsideration of both descriptive representation and democratic accountability. In contemporary state legislatures, race determines who wins office but not how they vote. Party discipline, rather than constituency demographics, governs roll-call behavior. Representation has shifted from a delegate model of voter-legislator alignment to a partisan model of ideological conformity. By tracing the structural erasure of demographic influence, this paper offers a revised account of political representation in an emerging era of nationalized partisanship and cautions against equating electoral diversity with substantive policy inclusion.