Factors influencing digital literacy among the 9185 elderly in South Korea: A machine learning approach

This study aims to identify key factors influencing digital literacy among the elderly individuals in South Korea using advanced machine learning techniques, focusing on enhancing their quality of life and social participation. Data from the 2018 to 2019 National Survey of Older Koreans were utilized, encompassing responses from 9185 individuals aged 65 and above. Digital literacy was assessed through the ability to use various digital devices and applications. Predictors included demographic, socioeconomic, health, and social support variables. CatBoost, a gradient boosting algorithm, was employed for feature selection and importance ranking, followed by logistic regression analysis to determine the relationships between these predictors and digital literacy levels. The analysis revealed that higher education levels, employment status, economic status, social participation, subjective health, and informal support positively influenced digital literacy among the elderly individuals. Conversely, age and depression were negatively associated with digital literacy. The CatBoost model demonstrated superior performance in identifying significant predictors, with education level emerging as the most influential factor, followed by age and employment status. The findings highlight the multifaceted nature of digital literacy among the elderly individuals, emphasizing the importance of educational attainment, social engagement, and economic conditions. Targeted interventions addressing these factors can effectively bridge the digital divide, ensuring that the elderly population can fully participate in the digital age and enhance their overall well-being.