A study on Internet use and subjective well-being among Chinese older adults: based on CGSS (2012-2018) five-wave mixed interface survey data

Chong Zhang, Yong E. Zhang, Yan WangFrontiers in Public Health 2024 · DOI

This study, published in Frontiers in Public Health, explores the complex relationship between digital engagement and mental well-being in China’s rapidly aging population. By analyzing a massive multi-year dataset, the researchers uncover a "masking effect": while the Internet significantly boosts happiness by fulfilling social and self-actualization needs, it simultaneously exposes older adults to social comparisons that make them feel society is less fair, partially offsetting those emotional gains.

TL;DR

Internet use significantly improves the subjective well-being (SWB) of Chinese older adults, particularly for urban, educated males. However, this positive impact is partially undermined because frequent Internet use reduces perceived "social fairness," creating a "masking effect" that dampens the overall happiness boost.

What Problem Does This Paper Address?

As China faces an aging society and a rapid digital transition, the "digital divide" has become a critical public health concern. While the Internet is often touted as a tool for "active aging," academic consensus on its impact is surprisingly split. Some argue it reduces loneliness and increases social support, while others suggest it leads to social isolation or "income envy" through constant comparison.

The authors set out to resolve this debate by looking specifically at Chinese older adults (60+). They investigate not just if the Internet makes them happier, but how it does so, and for whom the effect is strongest. Specifically, they look at "subjective social fairness"—how fair an individual perceives society to be—as a potential mediator that might explain why the Internet's benefits aren't always straightforward.

Methods

The researchers employed a robust econometric toolkit to move beyond simple correlations:

  • Ordered Probit Regression: Used to model the relationship between Internet use frequency and happiness (measured on a 1–5 scale).
  • Propensity Score Matching (PSM): To address "selection bias" (the idea that naturally happier or more resourceful people might be more likely to use the Internet in the first place), they used PSM to create a balanced comparison between "Netizens" and "Non-netizens."
  • Karlson-Holm-Breen (KHB) Method: A sophisticated technique used to decompose the total effect of Internet use into direct and indirect effects, specifically to test the mediating role of subjective social fairness in non-linear models.

Datasets

The study utilizes the China General Social Survey (CGSS), the earliest nationwide, comprehensive social survey in China. They pooled five waves of data (2012, 2013, 2015, 2017, and 2018), resulting in a substantial sample size of 18,458 valid observations of adults aged 60 and older.

Key Insights and Ideas

  • The "Masking Effect" of Social Fairness: This is the paper's most novel contribution. Internet use has a strong direct positive effect on happiness (0.157). However, it has a negative indirect effect (-0.081) because it reduces the user's sense of social fairness. By seeing the "highlight reels" of others or reading about social inequalities, older adults may perceive their own relative status as lower, which hurts their well-being.
  • Activity Theory in the Digital Age: The findings support "Activity Theory," suggesting the Internet serves as a vital substitute for social roles lost after retirement, helping seniors maintain social contacts and find new ways to participate in society.
  • The Gendered Digital Dividend: Interestingly, the happiness boost was significant for older men but not for older women. The authors suggest this is because women often have more robust offline social networks (friends, neighbors) to begin with, whereas men may rely more on the Internet to replace lost professional social circles.

Key Results

  • General Happiness: Internet use is significantly positively correlated with SWB. The higher the frequency of use, the stronger the well-being.
  • Quantified Impact: After controlling for selection bias using PSM, the "Average Treatment Effect on the Treated" (ATT) showed that Internet use increases well-being scores by approximately 0.070 to 0.072 points.
  • Heterogeneity: The positive effect is significantly stronger for:
  • Urban vs. Rural residents.
  • Educated vs. Unschooled individuals.
  • Residents in Eastern China vs. Central or Western regions.
  • Mediation: KHB testing confirmed that subjective social fairness plays a negative mediating role, meaning the Internet would make seniors even happier if it didn't also make them feel society was less fair.

Limitations and Future Work

  • Data Recency: The data ends in 2018. The authors acknowledge that the COVID-19 pandemic likely fundamentally shifted how older adults use the Internet, making more recent data a priority for future study.
  • Causality: While PSM helps, the cross-sectional nature of the CGSS waves makes it difficult to definitively prove long-term causality compared to a true longitudinal study tracking the same individuals over decades.
  • Content Matters: The study looks at frequency but not content. Future research should distinguish between using the Internet for learning and socializing versus passive consumption of news or short videos, which likely have very different psychological impacts.

Summary updated

Abstract

The findings suggest that Internet use contributes to the improvement of subjective well-being in older adults, but attenuates this effect by reducing the subjective social fairness. Future research should further consider other factors such as the purpose, specific function, intensity of Internet use, etc., to gain a deeper understanding of how the Internet can help promote well-being.
BibTeX
@inproceedings{zhang2024study,
  author = {Chong Zhang and Yong E. Zhang and Yan Wang},
  title = {{A study on Internet use and subjective well-being among Chinese older adults: based on CGSS (2012-2018) five-wave mixed interface survey data}},
  booktitle = {Frontiers in Public Health},
  year = {2024},
  doi = {10.3389/fpubh.2023.1277789},
}
Figure