Burnout Risk Profiles in Psychology Students: An Exploratory Study with Machine Learning

Area of Science:

• Psychology
• Public Health
• Educational Psychology

Background:

• University students experience high academic workloads and performance expectations, increasing their risk of burnout and psychological distress.
• Academic burnout is a significant concern in higher education, impacting student well-being and academic success.

Purpose of the Study:

• To analyze the relationship between psychological and lifestyle variables and academic burnout in Portuguese psychology students.
• To identify distinct burnout risk profiles among psychology students using machine learning.

Main Methods:

• Cross-sectional study involving 274 Portuguese psychology students (72.6% undergraduates).
• Assessment of psychological well-being, distress, emotional regulation, diet, physical activity, sleep quality, and burnout.
• Application of machine learning algorithms (Random Forest, XGBoost, SVM) to identify burnout risk profiles.

Main Results:

• Psychological distress, emotional regulation difficulties, and poor sleep quality were positively associated with burnout.
• Psychological well-being was negatively associated with burnout.
• Two profiles identified: 'Burnout Risk' (62 participants) and 'No Risk'. The 'Burnout Risk' profile exhibited higher distress, emotional difficulties, lower well-being and sleep quality, pro-inflammatory diet, and less physical activity. Machine learning models achieved high accuracy (93.82%–97.53%).

Conclusions:

• Psychological distress, emotional regulation, and sleep quality are critical factors in student burnout.
• Machine learning effectively identifies distinct burnout risk profiles in university students.
• Health promotion and targeted mental health strategies are crucial for preventing academic burnout in university settings.