Predicting Success in Professional Licensure Examinations Through Discriminant Function Analysis

Abstract

The study developed a data-driven predictive model for forecasting graduates’ success in professional licensure examinations using Discriminant Function Analysis (DFA). It employed a quantitative, predictive-correlational design to analyze institutional records of 933 graduates from five licensure programs in higher education institutions in the Philippines between 2020 and 2025. Cognitive-academic factors, including fluid intelligence, communication skills, reading comprehension, mathematical ability, professional knowledge, internship performance, and mock board exam performance, as well as contextual factors such as study duration and academic difficulty, were examined using descriptive statistics, independent samples t-tests, and DFA. Results revealed that licensure passers significantly surpassed non-passers across all variables (p < .001). The discriminant model was statistically significant (Wilks’ Λ = 0.430, χ² (9) = 781.573, p < .001; canonical r = 0.755) and accurately classified 88% of the cross-validated cases. The strongest predictors of licensure success were professional knowledge, mock board exam performance, study duration, and academic difficulty. The research concludes that cognitive mastery and academic persistence are crucial for licensure readiness. The study recommends integrating data-driven and AI-enabled early warning and academic analytics systems to identify at-risk examinees and support timely interventions aimed at optimizing readiness for professional qualifications.