Assessing the Level and Determinants of Active Aging in China: An LDA-Based Topic Modeling Approach

Area of Science:

• Gerontology
• Public Health
• Sociology

Background:

• China faces unique demographic challenges: a rapidly growing elderly population and aging before widespread economic prosperity.
• Existing active aging frameworks are primarily from high-income countries, necessitating a context-specific approach for China.
• Urgent need for a tailored conceptual and evaluative framework to measure active aging in China's socio-cultural and economic context.

Purpose of the Study:

• To construct a multidimensional indicator system for measuring active aging among older adults in China.
• To evaluate individual-level active aging scores using a quantitatively derived framework.
• To analyze trends and correlates of active aging in China.

Main Methods:

• Utilized Latent Dirichlet Allocation (LDA) topic model for indicator system construction.
• Employed five waves of nationally representative panel data from the China Health and Retirement Longitudinal Study (CHARLS) (2011-2020).
• Developed a measurement system with six dimensions and 21 indicators, assessing internal consistency.

Main Results:

• The active aging measurement system includes physical health, psychological well-being, social security, economic support, social participation, and lifelong learning.
• Average active aging score was 0.4912±0.0907, with consistent improvements observed from 2011 to 2020.
• Significant regional disparities in active aging improvements were noted, with the eastern region showing the most pronounced gains.

Conclusions:

• Active aging levels in China show consistent improvement, but regional disparities persist and are widening in some areas.
• Educational attainment, urban residence, male gender, and marital status are key positive correlates of active aging.
• Older age, chronic conditions, and smoking are negatively associated with active aging; education, urban-rural status, age, and gender are most influential factors.