AI is transforming student assessment in higher education, offering solutions to limitations of traditional methods. It enables personalized, efficient, and data-driven evaluation through automated grading, adaptive learning platforms, and analytics tools. While case studies show improved outcomes, ethical concerns like bias must be addressed. Responsible AI implementation can enhance assessment practices, promoting educational equity and excellence. Institutions adopting AI should set clear goals, involve stakeholders, and maintain ethical safeguards.
Student withdrawal is a persistent challenge in higher education, influenced by academic, socioeconomic, psychological, institutional, and digital factors. This chapter examines theoretical models of student attrition and reviews key determinants of withdrawal across diverse educational contexts. It further explores data sources and predictive approaches, including statistical and machine learning models, to identify at-risk students and support proactive retention strategies. By integrating predictive analytics with academic, financial, psychological, and technology-driven interventions, the chapter provides evidence-based insights for educators, administrators, and policymakers to enhance student persistence and institutional effectiveness.