The 21st century has witnessed a surge in the use of digital platforms for information access and sharing. While new media technologies reduce information exchange costs, they also pose challenges to individual social psychology. This study investigates the evolving relationship between internet use and individual tolerance in China, a context largely unexplored in existing research. China's internet penetration rate has increased dramatically, yet the impact of this on individual tolerance remains unknown. This paper aims to address this gap by examining the impact of internet use on public tolerance in China, exploring its moderating mechanisms and considering the role of institutional transformations and fairness perceptions.

Existing literature on public tolerance broadly falls into two paradigms: typological studies focusing on tolerance types and their social functions, and analyses of micro and macro factors influencing individual tolerance. Scholars have categorized tolerance into political, social, and moral dimensions, with each focusing on different aspects of acceptance of others. Factors influencing tolerance include age, education, cognitive ability, political knowledge, social support, and the social and economic environment. The impact of education, in particular, is highlighted, with higher education associated with greater tolerance. While some studies have explored the impact of internet use on tolerance, these often focus on contexts outside China and have yielded mixed results, with some suggesting negative effects. This study aims to add to this literature by focusing on the Chinese context and considering the influence of institutional transformations.

This study utilizes data from the Chinese Social Survey (CSS) from 2015 to 2021, matched with data from the China Statistical Yearbook. The CSS data includes information on internet usage, tolerance levels, and socio-demographic characteristics of respondents. Tolerance is measured using a multi-item scale, factor analyzed into social tolerance (ST) and moral tolerance (MT). Internet use is measured by frequency of internet use and participation in online discussions. Moderating variables include university education, employment sector (system vs. out-of-system), and Hukou registration (urban vs. rural). Mediator variables are opportunity fairness and outcome fairness. Control variables include gender, age, political affiliation, family assets, location, and year of survey. Ordinary Least Squares (OLS) regression is the benchmark method, with Seemingly Unrelated Regression (SUR) used to address potential correlation between tolerance measures. To handle potential endogeneity, instrumental variables (IV) and propensity score matching (PSM) methods are also employed. The IVs used are provincial internet coverage, community internet user ratio, and regional inclusive financial index. The PSM method is implemented using 1:1 matching, radius matching, and kernel matching.

The study's key findings include: 1. Significant differences in ST and MT exist across respondent groups (education, employment sector, Hukou). 2. Internet use has a significantly positive effect on both MT and ST, robust to OLS, SUR, IV, and PSM estimations. The impact of internet use on MT is stronger than on ST. 3. Market transformation moderates the relationship between internet use and ST, but not MT. Specifically, internet use strengthens ST among university graduates and individuals employed in system sectors, but increases the ST gap between urban and rural residents. 4. Opportunity and outcome fairness exhibit suppressive effects, not mediating effects, on the relationship between internet use and tolerance. Internet use decreases perceived fairness but still increases tolerance. The IV analysis shows that higher internet coverage, community internet usage, and inclusive financial development are positively associated with tolerance, after accounting for endogeneity. PSM results support the positive relationship between internet use and both ST and MT, and the impact is stronger for ST.

The findings demonstrate a positive relationship between internet use and tolerance in China, adding nuance to previous research showing mixed or negative results in other settings. The positive effect could be attributed to the unique context of rapid internet penetration in China, increasing access to diverse information and reducing information gaps. The moderating roles of educational attainment, employment sector, and Hukou highlight the impact of social stratification and resource distribution on tolerance. The suppressive effects of perceived fairness suggest that the exposure to information about social inequalities might not necessarily decrease tolerance in the Chinese context. This might be because the overall effect of internet use in increasing tolerance is stronger than its effect in reducing perceived fairness.

This study provides strong evidence that increased internet use in China is associated with higher levels of social and moral tolerance, even after accounting for several confounding factors. The moderating influence of market-oriented reforms underscores the complex interplay between technological advancements, social structures, and individual attitudes. Future research should explore the long-term effects of internet use on tolerance, examine the specific types of online content that influence tolerance, and investigate the role of government regulation in shaping this relationship.

The study relies on cross-sectional survey data, which limits the ability to make strong causal inferences. Self-reported measures of tolerance and internet use may be subject to biases. The study focuses on China, and the findings may not be generalizable to other contexts. Furthermore, the chosen instrumental variables might not fully capture all potential confounding factors affecting the relationship between internet use and tolerance.