Binomial Logistic Regression and CatBoost Model of Music on Empathy

We conducted a comprehensive review of previous studies and observed a robust and widespread impact of music training on empathy. However, the majority of these studies primarily relied on qualitative approaches, with limited research incorporating covariates and a lack of studies comparing the magnitudes of these variables’ effects. To bridge this gap, we utilized an online questionnaire to gather pertinent data. The dependent variable in our study was the empathy ability of each participant, while the independent variables encompassed five dimensions of musical literacy, methods of acquiring musical literacy, personality traits, and demographic information. Utilizing Binomial Logistic Regression and CatBoost model, we quantitatively analyzed the factors influencing empathy. Across both modeling techniques, there was a consistent and significant influence on high empathy factors, primarily attributed to gratitude attitude, positive reciprocal motivation, ability to perception of music, ability to identify music emotions, collaborative experiences. In contrast, whether individuals had undergone long-term professional music training or the influence of personality traits on high empathy, it was found that this influence was minimal. Our findings contribute to a more nuanced understanding of the relationship between music training and empathy, emphasizing the importance of quantitative analysis in this domain.