Generational Differences: The Levels and Determinants of News Media Trust in China

Abstract

Generational differences in news media trust are mainly studied in Western contexts and remain underexplored in non-Western societies. Drawing on the theory of generations, this study investigates the potential generational gaps in the levels and determinants of news media trust in China. Three analyses are conducted using data from the World Values Survey (WVS) in mainland China. Findings indicate a significant decline in news media trust levels from the Early PRC Generation (–1961) through the Cultural Revolution Generation (1962–1977), the Reform and Opening-up Generation (1978–1991), to the Globalization Generation (1992–). Notably, generation moderates the relationship between news media trust and its determinants. From older to younger generations, the influence of political trust on news media trust gradually weakens, while the impact of authoritarian personality increases. This study offers explanations for the dynamic changes in news media trust from the perspective of generational shifts.

1. Introduction

News media trust is a key concept that reflects the complex relationship between the state, the media, and the public. Public trust in news media is shaped not only by users’ characteristics and the media’s performance (Tsfati & Cohen, 2012) but also by structural factors, including social context, political systems, and media systems (Fawzi et al., 2021; Strömbäck et al., 2020). In the Western context, news media trust is often rooted in journalistic professionalism and institutional independence (Müller, 2013). However, these conceptual frameworks may not seamlessly translate to non-Western settings, especially where media systems are state-dominated or state-influenced. In such societies, trust in news media may be shaped more by political alignment, institutional legitimacy, and cultural values such as collectivism or deference to authority (Liu & Bates, 2009). Particularly in China, the media serves as a tool of state power and is often perceived as an extension of the government (Chu, 1994). However, existing research is mainly rooted in the Western context and has, to some extent, ignored the characteristics and dynamics of media trust in a diversified media environment (Dinana et al., 2025). Consequently, analyzing news media trust requires considering the media’s role, cultural background, and sociopolitical environment (Liu & Bates, 2009). Against this backdrop, the generational perspective grounded in the framework of political socialization provides critical insights into trust in news media.

Generation remains a debated but still valuable concept in scholarly research. Critics view generational categorization as arbitrary and attribute differences to age or period effects (Costanza et al., 2023). Despite these criticisms, modernization and globalization have, in fact, led to successive generations growing up in distinct environments (Mishler & Rose, 2007). These differences are reflected in social life. Many parents in China, for example, believe that by the age of 30, adults should have settled down and started a family. In contrast, today’s younger generation, even at the same age, may prioritize self-exploration, personal development, or life experiences over these traditional life milestones. That is to say, the thoughts and behaviors of the offspring and parental generations have diverged at the same life stage (Twenge et al., 2012; Zhong & Inglehart, 2024), and these differences seem to go beyond the explanatory scope of a life-cycle perspective. Faced with this situation, a generational analysis captures the lasting impact of formative experiences during youth and helps distinguish specific group patterns (Brosius et al., 2022), especially when individual aging and period influences fail to provide a sufficient explanation. Moreover, generational replacement is recognized as a significant driver of social and political change (Hooghe, 2004). Different generations often set different economic, cultural, and social priorities (Abramson & Inglehart, 2009), while media plays a vital role in this context (Bolin, 2016). It does so by forming generational narratives and movements; also, generational habits affect citizens’ media diets and how they respond to news (Andersen et al., 2021). Unlike the older generation, who experienced drastic digital transformation, young adults have been socialized in a digitalized and fragmented media landscape (Cotter & Thorson, 2022). Research has identified unique media perceptions and behaviors among younger audiences (Cotter & Thorson, 2022; Eddy, 2022), which means that these differences may also extend to news media trust.

China provides an ideal setting for comparing generational differences. Since 1949, Chinese society has transitioned from a planned economy characterized by material shortages to periods of political instability and social movements, followed by the rise of the socialist market economy and, most recently, economic globalization (J. Wu, 2015). These significant institutional transformations within a relatively short period shaped unique generational identities and values (Mannheim, 1952). In addition, socioeconomic development often drives generational shifts in values and priorities, particularly in countries experiencing rapid economic growth (Abramson & Inglehart, 2009). According to the latest world economic outlook (International Monetary Fund, 2025), China’s per capita GDP grew from USD 306.98 in 1980 to over USD 10,530 in 2020, an increase of nearly 35 times. In this process, China’s media environment has transitioned toward a more market-oriented and digitized ecosystem, while younger generations exhibit increasing post-materialist values (Zhong & Inglehart, 2024). More importantly, according to the statistical report on China’s internet development (China Internet Network Information Center, 2025), by December 2024, the number of online news users in China had reached 811 million, spanning a wide age range, as 14% are over 60 and 17% are under 20. The demographic profile of Chinese media consumers supports generational analysis.

This study primarily aims to investigate whether the level and determinants of news media trust in China vary across generational groups. By introducing the generational perspective into the field of media trust, this study moves beyond life-cycle-based research paths that focus solely on age, capturing how generational differences reflect deeper sociopolitical transformations. This dynamic approach offers a more context-sensitive framework for explaining changes in news media trust, particularly in a rapidly evolving non-Western society.

2. Theorizing Chinese Generational Differences in News Media Trust

Despite the lack of a universal definition of trust, three widely recognized commonalities exist: trust is grounded in past experiences that shape expectations about future behavior, it involves risk and uncertainty, and it reduces social complexity (Fawzi et al., 2021). When applied to the media context, media trust is defined as “the willingness of the audience to be vulnerable to news content based on the expectation that the media will perform in a satisfactory manner” (Hanitzsch et al., 2018, p. 5). It is important to recognize that in China, where the media is structurally embedded within state institutions, trust in news media may carry additional institutional or political connotations. This, however, does not mean that media trust is equivalent to or reducible to political trust. Despite contextual influences on its expression, the core concept of media trust, as willingness and judgment about the credibility of information content and journalistic professionalism, remains valid and applicable in the Chinese context. Separately, media trust is often divided into two forms: a generalized form, referring to institutional trust in news media or journalism as a whole, and a specific form, referring to trust in specific media objects such as content or journalists (Fawzi et al., 2021; Strömbäck et al., 2020). Notably, this study adopts a broad perspective, focusing on generalized trust in news media, and the same operational approach was validated in previous research (Tsfati & Ariely, 2014).

2.1. Generational Differences in China

From a theoretical perspective, research on generational differences is guided by two leading paradigms: the life-cycle perspective, which posits that younger generations’ behaviors will align with older ones over time; and the generational perspective, which argues that each generation develops distinct values and behaviors shaped by their socialization experiences (Andersen et al., 2021; Brosius et al., 2022; Sun & Wang, 2010). The former emphasizes individual-level longitudinal analysis, while the latter focuses on cross-sectional individual comparisons. Based on the research objectives, this study adopted a generational approach. Admittedly, only a longitudinal and time-lag design and data could fully distinguish generation, period, and age effects (Brosius et al., 2022).

Generation, distinct from kinship, refers to a group of individuals who share a common historical background. They are shaped by major historical events, socioeconomic conditions, and cultural currents during their formative years, which foster a unique generational consciousness, especially when actively participating (Mannheim, 1952). Socioeconomic development also drives changes in generational value. In advanced societies, materialist priorities such as economic security often give way to post-materialist values like autonomy and self-expression (Abramson & Inglehart, 2009). As noted earlier, Chinese society has rapidly transitioned from a centralized planned economy to a socialist market economy, implying that today’s adult population experienced distinctly different socialization processes during their adolescence. The profound political, economic, and media system transformations that Chinese society has undergone in a relatively short period have significantly shaped generational differences. As a result, generational differences and value shifts have emerged in Chinese society (Sun & Wang, 2010; Yi et al., 2010); older generations are more collectivist, while younger generations tend toward individualism and post-materialist values (Zhong & Inglehart, 2024). Post-materialist values, in turn, are associated with increased skepticism toward media institutions (Tsfati & Ariely, 2014).

In China, the media system and technological paradigm have undergone fundamental changes within a relatively short period of time, objectively causing different generations to grow up in quite different media environments. Media environments are crucial in individuals’ socialization (Brosius et al., 2022). Exposure to different media environments during formative years contributes to gaps in media technology proficiency and leads to varying perceptions and dependencies on news media (Prensky, 2001). China’s media system evolved from a state propaganda tool in the 1950s to market authoritarianism in the late 1970s, with commercialization and digitalization since the 1990s (Chu, 1994). Commercial competition and user-generated content have created fertile soil for the production and spread of misinformation (Guo, 2020). Empirical research has shown Chinese ’70s-born cohorts differed from ’90s-born cohorts in their digital habitus, particularly regarding internet proficiency, information exposure through WeChat, and WeChat perceptions (Q. Hu & Cheong, 2021). More broadly, generational differences in social media engagement in China are reflected in older adults’ limited access, reduced platform diversity, and lower levels of participation (Zhou et al., 2022). A generational decline in support for democracy and active political participation has been observed in the authoritarian tendencies of China (Yin, 2022). These contextual dynamics may contribute to differences in news media trust. Accordingly, the following hypothesis is proposed:

H1. 

There are generational differences in the levels of news media trust in China, with younger generations holding lower levels of trust than older generations.

2.2. Generational Categories of Chinese

In the West, widely recognized generational categories generally include the Greatest Generation, Silent Generation, Baby Boomers, Generation X, Millennials/Generation Y, and Generation Z (Parry & Urwin, 2011; Pew Research Center, 2015). However, since these classifications stem from milestone events in Western history (Parry & Urwin, 2011), Chinese and Westerners do not share similar historical and social experiences, so it is not appropriate to directly transplant these terms and labels into the Chinese context (To & Tam, 2014).

A variety of generational classifications have been proposed in the Chinese context. Some scholars identified four Chinese generations: Great Leap Forward, Cultural Revolution, the beginning of economic reform, and the social transition (Sun & Wang, 2010); Pre-socialist, Socialist, Cultural Revolution, and Reform/Opening-up (Earl Bennett et al., 2001; Y. Hu & Scott, 2016). Others divided people into three generations: the Red Guards, the Modern Realists, and the Global Materialists (Hung et al., 2007); Cultural Revolution, Social reform/Transitional generation, and Millennium generation (Chen & Lian, 2015; Yi et al., 2010). Ten or five years as a generation have also been proposed in some studies (Cui & Ji, 2020; J. Wu et al., 2020). Building on the view that a “generation in actuality” forms when individuals of similar ages share experiences of significant historical events and distinct cultural values (Mannheim, 1952) and often use the core political or economic milestones as nodes, and considering that a generation is typically defined by a birth span of 15 to 20 years (Pew Research Center, 2015), this study identified four Chinese generations based on economic, political, and media environments during their formative years.

The Early PRC Generation (Gen 1: –1961) was born during massive land reforms and industrial nationalization. They faced a highly centralized planned economy and extreme egalitarianism and were deeply influenced by traditional Confucian culture. So, this generation developed a strong collective consciousness and responsibility toward the country (Sun & Wang, 2010). Before 1961, China’s media focused on solidifying the regime’s political and ideological foundations, and the public received state-approved information mainly by radio. The Cultural Revolution Generation (Gen 2: 1962–1977) experienced radical political movements and social chaos, witnessing attacks on Confucianism and Western ideologies. Consequently, they typically avoid overt political involvement, hold negative attitudes toward change, and maintain cautious and conservative attitudes in their daily lives (Hung et al., 2007). During the Cultural Revolution (1966–1976), journalism was distorted into a tool for class struggle and political campaigns. Media content was overwhelmingly dominated by revolutionary narratives, such as glorifying the proletariat class and denouncing so-called state enemies (Chu, 1994). For the public, media consumption may be a highly politicized experience. The Reform and Opening-up Generation (Gen 3: 1978–1991) witnessed the transition of Chinese society from a planned to a market-oriented economy. Remarkable economic expansion and urbanization fostered a more individualistic, realistic, and pragmatic personality than their predecessors (Yi et al., 2010). In this phase, China’s media landscape diversified with economic reforms, expanding beyond strict political propaganda to include economic, cultural, educational, and entertainment content. The Globalization Generation (Gen 4: 1992–) was born into an era of rapid technological advancements. They grew up with digital media and generally hold a broader global perspective, critical thinking, and post-materialistic values (Hung et al., 2007). From this period, Chinese journalism began to be viewed as a cultural industry that empowered individuals to participate in content creation and dissemination, fostering a more interactive and dynamic media ecosystem. The four generations roughly correspond to Baby Boomers, Generation X, Millennials, and Generation Z in the West.

2.3. Determinants of News Media Trust

Institutional and cultural theories are the primary paradigms explaining media trust sources (Tsfati & Ariely, 2014). The former views media trust as an outcome of institutional performance, whereas the latter assumes that trust is exogenous, extending from early-learned social trust to institutions such as media agencies (Mishler & Rose, 2001). Correspondingly, variables related to media trust are generally categorized into two dimensions: media-related features and audience-related characteristics (Tsfati & Cohen, 2012). This study takes a culturalist approach and draws on the framework proposed by Fawzi et al. (2021), which categorizes the individual-level antecedents of media trust into three domains: social, political, and media-related characteristics. This framework reflects an integrative perspective grounded in cultural and institutional theories: social characteristics relate to cultural roots, while political and media-related characteristics reflect institutional theory.

Under the political characteristics, political trust is often regarded as one of the most influential predictors of news media trust. Based on trust transfer theory, trust in one entity often stems from trust in another to which it is connected (Stewart, 2003). Given media institutions’ close ties to political bodies and their role as extensions of the government in China (Chu, 1994), public trust in political institutions often shapes trust in news media. Moreover, the medialization of politics (Hanitzsch et al., 2018) and the lower autonomy of the media system (Ariely, 2015) often lead to a stronger relationship between political trust and media trust. Current studies consistently identified this relationship as a strong positive correlation (Brosius et al., 2022; Fawzi et al., 2021; Tsfati & Ariely, 2014; Xu, 2013). Besides, political interest plays a crucial role in shaping perceptions of the political world and predicting political behaviors (Prior, 2010). According to the Mutz et al. (1996) notion of reception gaps, individuals with higher political interest are more likely to receive news messages and actively seek news that aligns with their interests, which can typically lead to a positive prediction of media trust (Fawzi et al., 2021; Tsfati & Ariely, 2014). Furthermore, political involvement influences individuals’ interactions with news media, as politically active individuals are more likely to consume and evaluate news content (Ardèvol-Abreu et al., 2018; Fletcher & Park, 2017). In this process, the public develops a deeper understanding and evaluation of news media.

Although political characteristics shape media trust across different contexts, the mechanisms vary. In Western democracies, such effects are often shaped by partisan alignment or personal cognition, while in China, they are shaped by the proximity between political and media systems. Political participation such as Communist Party membership in China is essentially institutionalized and tends to reflect alignment with the state’s ideology. Individuals engaged in pro-regime activities are more likely to regard state media as legitimate and trustworthy sources of information. This is because their political identity aligns with the legitimacy narratives presented by official media outlets. Political participation in the West can be either pro-establishment or anti-establishment, so political participation does not always lead to trust in the mainstream media, and sometimes even leads to the opposite (Hooghe & Marien, 2013). These contrasts highlight that the effect of political characteristics on media trust may be context dependent.

As part of social characteristics, personality traits determine how individuals utilize opportunities in a structural environment to build their trust (Sztompka, 1999), and individual-level preferences have different effects on news trust (Peng & Yang, 2022). Among these, social trust, formed through socialization, plays a key role in shaping media trust and is positively associated with it (Ognyanova, 2019; Tsfati & Ariely, 2014). An authoritarian personality, rooted in psychoanalytic theory, is shaped by early interactions with authority figures in childhood (Altemeyer, 1998; Brown, 2004). It has later been refined as a tendency toward obedience and submission to authority. Individuals with this personality tend to trust institutions that maintain the existing order, particularly in China, where the media functions as an extension of authority (Chu, 1994); they are more likely to perceive state-aligned media as credible. Accordingly, the study seeks to clarify this relationship.

The relationship between media use and media trust is complex. People with low levels of trust often choose non-mainstream news sources like social media and digital-born providers (Fletcher & Park, 2017; Tsfati & Ariely, 2014; Xu, 2013). In contrast, exposure to mainstream news media is positively correlated with trust in the media (Hanitzsch et al., 2018; Tsfati & Ariely, 2014). Using mainstream or alternative news sources was linked to a higher trust in news while relying on social media as a primary source correlated with lower trust levels (Kalogeropoulos et al., 2019). However, others find the relationship between media usage and trust weak (Kiousis, 2001) or nonexistent (Earl Bennett et al., 2001). Similarly, sociodemographic factors like gender, education, and income have been identified as key influences on media trust, although the findings remain inconsistent (Andersen et al., 2021; Hanitzsch et al., 2018; Tsfati & Ariely, 2014; Xu, 2013). This study includes them as control variables.

More importantly, a large body of work has found that generational differences have emerged in people’s political and social characteristics, such as political trust, interpersonal trust, social trust, and work values (Andersen et al., 2021; Egri & Ralston, 2004; Harmel & Yeh, 2015; Parry & Urwin, 2011; Sun & Wang, 2010). Notably, news sources and consumption patterns have shown obvious generation gaps (Ghersetti & Westlund, 2018; Leung, 2013; Lyu & Chai, 2024; S. Wu et al., 2024); younger generations access news through social media and digital platforms, while the older stick to traditional media (Eddy, 2022). In addition, television viewing often fosters or maintains generational consensus on non-conventional sexual behaviors, while frequent Internet usage widens generational divergences (Lyu & Chai, 2024). In other words, changes in the media environment and technology may have reshaped the relationship between the public and news (Fletcher & Park, 2017). Given the potential for generational differences in the determinants of news media trust, the following research questions are proposed.

• RQ 1: What are the determinants of news media trust in China?
• RQ 2: Are there generational differences in the determinants of news media trust in China, and if so, which factors account for these differences?

3. Method

This study utilized cross-sectional data from the two most recent waves of the World Values Survey (WVS) conducted in mainland China, Wave 6 (2012/13) and Wave 7 (2018), with a final effective sample size of 5336 (Haerpfer et al., 2022; Inglehart et al., 2014). Both surveys employed the GPS/GIS Assistant Area Sampling method and targeted Chinese residents aged 18 and above. Data was collected through face-to-face interviews using consistent survey questions. Thus, combining the two waves is reasonable and ensures a roughly balanced sample size across generations.

Following the operationalization of previous research (Tsfati & Ariely, 2014), the dependent variable, news media trust, was calculated by averaging responses to “The press” and “Television” from the institutional trust section. Participants were asked: “I am going to name several organizations. For each, how much confidence do you have in them?” Responses were reverse coded on a scale from 0 to 3, with higher values indicating greater confidence. A reliability analysis showed excellent results (WVS6 α = 0.893; WVS7 α = 0.834). Political trust was measured similarly, including seven items: armed forces, the police, justice system/courts, the government, the political parties, parliament, and the civil services. Omega values are close to 0.9 in both the sixth and seventh waves (ω = 0.898/0.887, respectively).

Political interest was coded as 1 (interested, 48.8%) and 0 (not interested, 50.2%) based on self-reports. The importance of democracy was measured on a 10-point scale ranging from 1 (not at all important) to 10 (absolutely important). Political involvement was measured as a dichotomous variable, capturing participation or willingness to engage in four political actions: signing petitions, boycotts, demonstrations, and strikes (WVS6: ω = 0.889; WVS7: ω = 0.795). Respondents who selected “Have done” or “Might do” for at least one of these actions were coded as “1”, while those who answered, “Would never do” to all four were coded as “0”. And 53.4% of respondents classified as politically engaged. Social trust was gauged by asking respondents whether most people can be trusted, coded as “1,” or if they needed to be careful, coded as “0.” Authoritarian personality traits measure was worded, “Do you think whether ‘respect for authoritarian’ would be a good thing, something indifferent, or a bad thing?”. The affirmative answer, “a good thing,” was coded as “1” (authoritarian traits), while all others were coded as “0” (no authoritarian traits). Media usage frequency was measured based on a self-reported use of seven sources, including daily newspapers, TV, radio, internet, mobile phones, email, and either printed magazines (WVS6) or social media (WVS7). Items were reverse coded from 0 (never) to 4 (daily) and averaged. Reliability was acceptable (WVS6: ω = 0.805; WVS7: ω = 0.704). Education was coded as 0 (no higher education, 57.4%) and 1 (upper secondary education, 42.3%). Income was assessed using a 10-point scale, ranging from 1 as the lowest to 10 as the highest. Gender was coded as 0 (men, 46.8%) and 1 (women, 53.2%).

4. Analysis

Three sets of analyses were conducted in this study. First, to examine H1, average scores of news media trust were measured across the four generations. A one-way ANOVA with the Scheffé post-hoc test was then performed. The Scheffé adjustment method, suitable for exploring unbalanced data and all possible contrasts between means, has been used in generational studies (Cogin, 2012; Scheffé, 1953). Second, a multivariable linear regression model was built to examine the key predictors of news media trust in China (RQ1). Third, four subgroup regression models were built based on different Chinese generations to identify generational differences in the predictors of media trust (RQ2). Specifically, after obtaining the regression coefficients of each independent variable within each generational group, Z-scores and corresponding p-values were calculated following the procedure and formula proposed by Clogg et al. (1995) and Paternoster et al. (1998) to test for differences in effect sizes across groups. As a robustness check, the full-sample regression model included interaction terms between generation and key predictors to examine whether generations moderate the predictor effects (see Supplementary Table S1).

5. Results

One-way ANOVA results showed significant generational differences in news media trust in China, F(3, 5067) = 29.197, p < 0.001. Scheffé post-hoc tests further indicated a consistent decline in trust from the oldest to the youngest generation and answered RQ1. Specifically, the mean level of media trust was 1.94 for Generation 1, 1.79 for Generation 2, 1.77 for Generation 3, and 1.69 for Generation 4, showing an apparent generational decline. Also, Model 0 was estimated with generational dummies as the sole predictors. The results showed that all Generation 1, Generation 2, and Generation 3 exhibited significantly higher levels than Generation 4, providing support for H1. Detailed post-hoc results are reported in

Table 1. Multiple comparisons of the levels of news media trust across generations in China.

	Gen 1: –1961	Gen 2: 1962–1977	Gen 3: 1978–1991	Mean	SD	N
Gen 1: –1961	-			1.9448 a	0.653	1378
Gen 2: 1962–1977	0.155 ***	-		1.7901 b	0.633	1827
Gen 3: 1978–1991	0.178 ***	0.023	-	1.7672 bc	0.627	1383
Gen 4: 1992–	0.256 ***	0.102 *	0.079	1.6884 c	0.599	483

Note. Mean differences (I-J) are based on Scheffé post-hoc analysis. Different superscripts (a, b, c) represent significantly different means. Mean trust values range from 0 (lowest) to 3 (highest). Data from WVS6 and WVS7. * p < 0.05; *** p < 0.001.

. Given the relatively small sample size of the youngest group, findings related to this group should be interpreted with caution.

Model 1 addressed RQ 1 by exploring the determinants of news media trust in China, with the regression results summarized in

Table 2. Predictors of new media trust.

	Model 0	Model 1	Model 2	Model 3	Model 4	Model 5
			(G1: –1961)	(G2: 1962–1977)	(G3: 1978–1991)	(G4: 1992–)
Political trust		0.562 (0.017) ***	0.719 (0.032) ***	0.506 (0.029) ***	0.517 (0.031) ***	0.509 (0.049) ***
Political involvement		–0.043 (0.018) *	−0.022 (0.032)	−0.047 (0.029)	−0.059 (0.034)	−0.065 (0.061)
Political interest		0.065 (0.017) ***	0.061 (0.033)	0.085 (0.029) **	0.043 (0.033)	0.050 (0.051)
Importance of democracy		–0.011 (0.005) *	−0.026 (0.009) **	−0.027 (0.009) **	0.007 (0.009)	0.030 (0.015)
Social trust		0.020 (0.018)	0.018 (0.032)	0.020 (0.030)	−0.007 (0.035)	0.085 (0.055)
Authoritarian personality		0.065 (0.017) ***	0.020 (0.032)	0.040 (0.029)	0.099 (0.033) **	0.160 (0.050) **
Media usage frequency		0.008 (0.011)	0.001 (0.021)	0.010 (0.017)	0.030 (0.020)	−0.026 (0.034)
Education		–0.129 (0.020) ***	−0.137 (0.042) **	−0.193 (0.033) ***	−0.026 (0.035)	−0.218 (0.071) **
Gender		0.035 (0.017) *	−0.051 (0.032)	0.062 (0.029) *	0.076 (0.032) *	0.030 (0.050)
Income		0.006 (0.005)	0.008 (0.008)	0.007 (0.008)	0.002 (0.009)	0.013 (0.016)
Gen1 (Ref = Gen4)	0.256 (0.034) ***	0.143 (0.034) ***
Gen2	0.102 (0.032) **	0.079 (0.031) *
Gen3	0.079 (0.034) *	0.113 (0.031) ***
Constant	1.688 (0.029) ***	0.504 (0.067) ***	0.494 (0.113) ***	0.855 (0.103) ***	0.470 (0.112) ***	0.327 (0.189)
F	29.197 ***	111.967 ***	58.122 ***	40.225 ***	34.566 ***	16.398 ***
R2	0.017	0.249	0.336	0.204	0.224	0.273
Adjusted R2	0.016	0.247	0.331	0.199	0.217	0.257
Observations	5071	4399	1158	1583	1211	447

Note. Data from WVS6 and WVS7. * p < 0.05; ** p < 0.01; *** p < 0.001.

. Political trust, political interest, authoritarian personality, and education were strongly correlated with media trust (p < 0.001). Political involvement, perceptions of democracy, and gender showed weaker associations (p < 0.05). Among these, only political involvement, perceptions of the importance of democracy, and education negatively correlated with media trust, while other variables demonstrated positive relationships. Gender gaps were observed, with women exhibiting higher media trust than men. However, no evidence was found to suggest that social trust, media usage frequency, or income affects media trust in this study.

Models 2 to 5 were later used to answer RQ2. The data set was divided into four generational groups for the multiple linear regression analysis, and findings are summarized into five key points. First, political trust is strongly and positively correlated with media trust across all generations, with the oldest generation showing a significantly stronger correlation. Second, Generation 1 and Generation 2 individuals exhibited lower media trust when they placed a high value on democracy or had higher education levels. Third, those with authoritarian personalities in Generation 3 and Generation 4 showed higher levels of media trust. Fourth, women in Generation 2 and Generation 3 demonstrated higher media trust than men, whereas no gender correlation is observed in Generation 1 and Generation 4. Finally, political interest positively correlates with news media trust only in Generation 2. Finally, inter-group comparisons of regression coefficients were conducted to accurately measure the differences observed across generations. Z-scores with corresponding p-values were calculated, demonstrating that the correlation coefficients between media trust and five variables vary across generations, as shown in

Table 3. Z-scores of differences between regression coefficients.

	Gen 1 & Gen 2	Gen 1 & Gen 3	Gen 1 & Gen 4	Gen 2 & Gen 3	Gen 2 & Gen 4	Gen 3 & Gen 4
Political trust	4.574 ***	4.417 ***	3.639 ***	–0.093	0.191	0.259
Political interest	–0.650	0.279	0.214	0.942	0.710	0.000
Political involvement	0.509	0.814	0.827	0.358	0.518	0.272
Importance of democracy	0.079	−2.828 **	−3.214 **	−2.907 **	−3.268 **	−1.253
Social trust	0.274	0.612	–0.959	0.369	−1.165	−1.381
Authoritarian personality	–0.301	−1.610	−2.459 *	−1.389	−2.301 *	−1.202
Media usage frequency	–0.074	−1.207	0.801	−1.257	0.894	1.699
Education	–0.276	−2.346 *	–0.862	−2.536	–0.714	1.356
Income	0.177	0.664	–0.112	0.498	–0.224	–0.545
Gender	−2.617 **	−2.698 **	−1.348	–0.250	0.571	0.734

Note. * p < 0.05; ** p < 0.01; *** p < 0.001.

. In other words, the moderating roles of generations are evident for political trust, the importance of democracy, authoritarian personality, gender, and education. These interaction effects provide robust evidence for H2 and are visualized in Figure 1.

Robustness

To test the robustness of the generational moderation effect, an additional regression model was estimated using the full sample, incorporating interaction terms between generations and key predictors. The results show that similar generational moderation effects also emerge for authoritarian personality, perceptions of democracy, and educational attainment in their relationships with media trust. The detailed regression results are given in the Supplementary Materials.

6. Discussion

This study aimed to investigate the generational differences in media trust in China. Strong evidence indicated that generation was a core influencer and moderator of media trust, decreasing progressively from the oldest generation, the Early PRC Generation (–1961), to the Cultural Revolution Generation (1962–1977), the Reform and Opening-up Generation (1978–1991), and ultimately the youngest generation, the Globalization Generation (1992–). Furthermore, the findings align with European studies (Brosius et al., 2022), showing that generational differences influence how various factors relate to news media trust. The generational perspective partially explains the dynamic changes in news media trust in the non-Western context.

The results show that Chinese people’s trust in news media declines with genera-tonal shift, with the Globalization Generation (1992–) exhibiting the lowest trust levels. This phenomenon may be explained by the following. First, the four generations grew up in different stages of Chinese society, and economic development has driven younger generations toward post-materialist values (Zhong & Inglehart, 2024), which may weaken trust in news media (Tsfati & Ariely, 2014). Second, Confucianism emphasizes avoiding conflict (Li, 2013), leading Chinese people to tend to avoid public criticism and tolerate the status quo for social harmony. This cultural background fosters inherent media trust among those who, primarily older generations, uphold traditional values. Third, the rise of the internet and social media has provided younger generations convenient access to news. However, the high-choice media environment results in an influx of false news (Mourão & Robertson, 2019) and may lead to news quality declining. It also compels younger media consumers to navigate between state narratives and alternative perspectives, influencing their trust in news media. Fourth, compared to the traditional media era, where censorship often occurred within the journalism industry and was not easily perceived by audiences, the social media era allows the public to directly sense government control over media (Tai, 2014) and become aware of information restrictions. Consequently, younger Chinese generations hold lower media trust.

More importantly, there are significant differences in the predictors of news media trust among Chinese generations. The strong positive correlation between political trust and news media trust tends to weaken with the generational shifts. This phenomenon may be attributable to the transformation of the media landscape in China. In the early decades after 1949, Chinese media served the Party’s agenda, and politics dominated news content (Chu, 1994). The public had minimal options for news consumption and almost no alternative sources. As a result, people’s attention to the news often translated into a focus on politics, further strengthening the relationship between the two. Post-1978, market reforms have made the media landscape more diversified and liberalized. The high-choice media environment has led to a relatively declining supply of political information (Van Aelst et al., 2017), and fragmented and personalized content has diluted the presence of political news in the media. Consequently, in the media perceptions and news consumption of younger generations, the media increasingly serves non-political functions, which may further decouple media trust from political trust.

Another interesting finding was that the positive link between the authoritarian personality and news media trust was absent among the older generations (G1 and G2) but emerged among the younger (G3 and G4). The difference in the presence or absence of correlations also reflects the moderating effect of generation. Older generations grew up in a tightly state-controlled media environment. As a result, they were more likely to accept media messages uncritically, and media trust may be more of a systemic and habitual response than a reflection of individual personality traits. The media’s authoritative role in their formative years reduced the impact of personal traits, fostering a collective trust in state-controlled media. In contrast, younger generations have been exposed to a more diverse media landscape, such as digital media platforms, alternative news sources, and competing narratives (Cotter & Thorson, 2022). They have more opportunities to align their media consumption with their inherent psychological preferences (Hart et al., 2009), and affect becomes more important in shaping information evaluation (Cotter & Thorson, 2022). In this process, the effect of authoritarian narratives on media audiences tends to weaken. While some younger folks with high authoritarian values support state media, others with low authoritarian values lean toward more liberal views, which gives authoritarian personalities more room to shape media trust.

Conversely, the positive predictive effect of the importance of democracy on media trust was evident only among older Chinese generations (G1 & G2) but not among the youngest two generations (G3 & G4). This result demonstrated that as generational shifts occurred, the role of democracy in shaping media trust in China diminished. Chinese citizens may support “democracy,” but their understanding differs from Western standards (Shi, 2008). The older generations, having experienced poverty and social instability before the Reform and Opening-up, see democracy as a symbol of modernity and are more sensitive to its ideals. However, when the significance of democracy fails to materialize, it can provoke their dissatisfaction with the government and its spokesperson—the media. Conversely, younger generations influenced by global discourse and shifting societal values, emphasize individual rights and personal freedoms over traditional democratic ideals (Sun & Wang, 2010). The critiques of Western-style democracy and praise for Chinese-style democracy reduce younger generations’ attachment to democratic ideals, weakening the link between democracy and media trust. Additionally, younger people adopt more diverse strategies (Cotter & Thorson, 2022), considering news’s partisan and commercial contexts to evaluate media credibility (Swart & Broersma, 2022).

Moreover, political interest, gender, and educational attainment influenced news media trust in China. Political interest was positively correlated with media trust, but this correlation appeared only among the Cultural Revolution Generation. Education significantly predicted media trust for those born before the Reform and Opening-up era. Similarly, gender gaps in news media trust were mainly observed in certain Chinese generations. The current study found that social trust and media usage frequency did not predict Chinese people’s trust in news media. Despite some studies linking media use and media trust (Kalogeropoulos et al., 2019; Xu, 2013), no such correlation was found here, and generalized media usage frequency may not predict media trust among Chinese individuals. Hence, it is necessary to classify media usage behaviors in a future study.

Theoretically, this study challenges the static, individual-level explanations prevalent in news media trust research by introducing an intergenerational perspective to capture the dynamic interplay between political socialization and trust formation. Rather than treating media trust as a fixed attitudinal outcome, it shows how citizens, shaped by differing media environments and value orientations across generations, internalize systemic structures in distinct ways, ultimately leading to divergent patterns of trust formation.

7. Limitations and Future Work

Several limitations should be acknowledged. Undoubtedly, using secondary data has limitations in the precision and depth of certain measurements, such as news media trust and authoritarian personality being operationalized in a simplified and generalized way, which may not be conducive to a deeper exploration of their theoretical dimensions or multifaceted nature. Future research may benefit from employing more refined and comprehensive measurement methods to better reflect the complexity of these constructions. Moreover, generational groups often lack mutual exclusivity because a certain number of people are born on the cusp or cutoff line of generations’ defining birth years (Arsenault, 2004), implying that they are affected by overlapping experiences. Besides, the Globalization generation represents a smaller sample proportion, so results concerning this group should be approached with caution. Future studies with balanced generational samples may yield more exciting or meaningful insights as generational shifts occur. Notably, several determinants of news media trust, such as populism, nationalism, conspiracy theories, and media literacy (Fawzi, 2019; Mari et al., 2022) were not included in the current study. So, future research could examine whether generational effects on news media trust remain significant after controlling for different variables, or to better assess the dynamics of trust in news media across generations. Besides, studies could explore whether and how the normative connotations of news media trust shift across political systems and cultural environments in the future. Last but not least, as a cross-sectional study, this analysis can only reveal associations rather than infer causal relationships, which limit the interpretation of directionality or mechanisms (Bryman, 2016).

Overall, generation is a core influencer and moderator of news media trust in China. Younger generations exhibit lower trust, and generational differences shape the relationships between news media trust and its determinants.