
What the Paper Asks
Understanding which candidates voters elect matters for representation and equality. Simon Calmar Andersen and Morten Hjortskov ask why prior studies show mixed effects for candidate gender and race while consistently finding strong effects for party affiliation.
What Lexicographic Preferences Are
The authors propose that voters use lexicographic preferences: they rank candidate attributes and attend first to party affiliation. Attributes like gender and race are treated as second-order criteria — consulted only when party does not break a tie or when party information is absent.
How the Authors Tested the Idea
Andersen and Hjortskov adapt conjoint experiments to diagnose lexicographic decision rules. Using a U.S.-representative conjoint sample and a pre-registered replication, they compare voter choices when party labels differ versus when candidates share the same party or when party information is withheld. Their design shows how choice patterns conditional on party information reveal whether voters apply a strict priority ordering of attributes.
Key Findings
Why This Matters
The paper offers a clear diagnostic for researchers using conjoint designs and a conceptual lens for interpreting inconsistent findings about descriptive representation. For scholars and practitioners, the work implies that candidate strategy and experimental measurement must account for voters' prioritized decision rules if they aim to understand or influence how social identities matter at the ballot box.

| Lexicographic Preferences in Candidate Choice: How Party Affiliation Dominates Gender and Race was authored by Simon Calmar Andersen and Morten Hjortskov. It was published by Cambridge in BJPS in 2024. |