The Dynamics of School Dropout in The Turkish Higher Education System

Introduction: Student dropout is not only an individual failure but also leads to the ineffective use of public resources, the worsening of educational inequalities, and disruption in the labor market, thereby preventing long-term development. The aim of this study is to address the methodological gap in the area by utilising a large-scale dataset of approximately 2.9 million students covering the 2016–2024 period. Method: The Fine-Gray Competing Risks Model, extended with time-varying coefficients, was employed, which considered graduation not simply as a censoring process, but as a competing risk with dropping out. Based on the higher education students database, the impacts of gender, age, disability status, parental education level, distance of residence, and entrance exam score type on the risk of dropout were explored. Results or Findings: Students accepted based on quantitative test scores had the lowest dropout risk. Students whose parents were college graduates initially faced a relatively higher risk of dropping out in specific fields; this risk increased dynamically in quantitative and verbal fields, Distance represented a constant marginal risk. The probability of dropping out also increases with age; male students initially had a lower risk, but this difference increased over time. Discussion or Conclusion: These findings may suggest that student dropout rates at Turkish universities may generally result from relatively stronger social factors. The continued education of students with disabilities in areas outside the quantitative fields and the growing advantage in school retention rates among male students presents a challenge to traditional views on the vulnerability of higher education.