The use of face-to-face interviews is controversial for college admissions decisions in light of the lack of availability of validity and reliability evidence for most college admission processes. This study investigated reliability and incremental predictive validity of a face-to-face postgraduate college admission interview with a sample of Hispanic applicants and approaches from generalizability theory and multiple regression models. Results showed a complex measurement paradox surrounding face-to-face college admission interviews. (Contains 5 tables and 2 figures.)