Workshop’s Topic: Universities allocate significant resources each year to recruit new hires fail to faculty from PhD programs, yet a substantial number of achieve tenure, often due to insufficient research output. This study examines whether the faculty recruitment process can be optimized by using machine learning to identify candidates with higher research potential earlier in the hiring process. Focusing on the accounting rookie recruiting market, we develop a machine learning model leveraging readily available job application materials: (i) candidateresumes, (ii)the reputations of their educational institutions and dissertation advisors, and (iii) their dissertation work. Our results demonstrate that a machine learning model using these materials outperforms traditional human judgment in predicting research success. Notably, resumes are particularly strong predictors of future research performance, while the dissertation’s predictive value appears mixed.

Time and Location: 15:00-17:00 PM (GMT+8), Room A423 (School of Management)

Language: Bilingual (Chinese and English)