This study presents an empirical investigation of naturalization adjudication in the United States using new administrative data on naturalization applications decided by the US Citizenship and Immi-gration Services between October 2014 and March 2018. We find significant group disparities in naturalization approvals based on applicants' race/ethnicity, gender, and religion, controlling for indi-vidual applicant characteristics, adjudication years, and variation between field offices. Non-White applicants and Hispanic applicants are less likely to be approved than non-Hispanic White applicants, male applicants are less likely to be approved than female appli-cants, and applicants from Muslim-majority countries are less likely to be approved than applicants from other countries. In addition, race/ethnicity, gender, and religion interact to produce a certain group hierarchy in naturalization approvals. For example, the prob-ability of approval for Black males is 5 percentage points smaller than that of White females. The probability of approval for Blacks from Muslim-majority countries is 9 percentage points smaller than that of Whites from other countries. The probability of approval for females from Muslim-majority countries is 6 percentage points smaller than that of females from other countries. This study con-tributes to our understanding of the nature of inequalities present in agency decision-making in the naturalization process.