Relationship of anxiety scores with gender roles using machine learning techniques

This research analyzes the relationship between cognitive reasoning variables, gender roles, and anxiety levels using machine learning techniques. By applying these methods, we aim to deepen the understanding of this association and provide valuable insights for academic and mental health fields. The study examines a population of 251 university students classified into masculine, feminine, androgynous, or undifferentiated roles using the Wisconsin Card Sorting Test and the Bem Sex Role Inventory. Anxiety levels were measured using Beck’s Anxiety Inventory. A decision tree classifier was implemented in MATLAB, with 70% of the data used for training and validation and 30% for testing. The results indicate that cognitive flexibility, femininity and masculinity scores, and anxiety levels achieved a classification accuracy of 99.4% in training and 98.6% in testing. These findings suggest that machine learning can be a valuable tool for exploring the relationship between gender roles and anxiety, offering new perspectives on how these variables interact across different gender identities.