Abstract
What structures political behavior, and which distinctions persist across elections? This study examines what differentiates Latino voters who supported Donald Trump from those who supported other presidential candidates in 2016, 2020, and 2024. Using Collaborative Multiracial Post-Election Surveys, random forests, and SHAP, it identifies which variables most clearly distinguish these voters across three nested models that progressively exclude election-proximate and direct partisan measures. Across all three elections, judgments concerning racial justice and immigration consistently emerge among the most informative distinctions. These racialized judgments appear alongside other important differences involving economic expectations, ideology, and evaluations of political events. The analysis also finds that racialized judgments extend beyond policy preferences to perceptions of political parties. Together, these findings underscore racialization as a sociopolitical process shaping political judgments across elections and political environments, including within ethnoracial groups.

![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://preprints.apsanet.org/engage/assets/public/apsa/logo/orcid.png)