News trustworthiness and verification in China: The tension of dual media channels
First Monday

News trustworthiness and verification in China: The tension of dual media channels by Yiran Wang and Gloria Mark

From 2006 to 2013, the increasing use of social media in China has provided a stage for citizens to report news and to vocalize viewpoints. News information circulated on social media at times contradicted the highly curated official news sources. In this study, we conducted two online surveys to explore 1) the level of trust that Chinese Internet users place on news from social media versus official media; and, 2) whether and how Chinese Internet users verify news from citizen sources. We found that news from official and citizen sources attracted different audience groups, who relied on different features to assess news trustworthiness. News verification was common and some users shared their findings through social media. We discuss the implications situated in a broader context of an authoritarian society.


Official media vs. citizen media
Research questions
Empirical studies: Methods
Results: Survey 1 — Perceived trustworthiness (RQ1–RQ4)
Results: Survey 2 — News verification of citizen media (RQ5–RQ6)




The media landscape in China is changing dramatically as social media has been introduced alongside of traditional mass media (Esarey and Qiang, 2011; Tang and Sampson, 2012). A number of social media sites comparable to Facebook and Twitter exist within China, such as Renren ( and Sina Weibo ( By January 2013, the number of users of microblogs (MBs) and social networking sites (SNSs) had reached 309 and 275 million, respectively, accounting for 54.8 percent and 48.8 percent of the Internet population (CNNIC, 2013). In addition to a high social media adoption rate, China’s social media users were relatively active rather than being passive spectators (Sullivan, 2012).

However, in China the government strictly controls the media environment by regulating service providers, censoring user-generated content, and often blocking or manipulating information about events of national concern. As of September 2015, around 3,000 Web sites were blocked, including Google, New York Times, and social media sites of Facebook, YouTube, Twitter, and Instagram (Wikipedia, 2015). Yet Chinese citizens are able to consume and participate in the production of news through social media channels as an alternative source to the government-controlled media.

The recent decade has witnessed a fast growing “networked information environment” (Benkler, 2006), in which virtually anyone with Internet access can read and write a blog, comment on a social media post, or edit a Wikipedia entry. With a focus on Western societies, a number of scholars (Benkler, 2006; Jenkins, 2006; Jenkins, et al., 2013; Shirky, 2010) have argued that information consumption has evolved into a dual consumption and production process, thanks to new media tools pooling resources, skills, free time, and interest from a network of users. The result of the collective production — be it news, movies, music, or software — could become an alternative source of media power (Castells, 2007; Jenkins, 2006). With the alternative power, individuals are less susceptible to the manipulation of traditional mass media that exercises political agendas and market strategies (Benkler, 2006). However, imagine a different configuration of society, one that does not have a free press, where only state-owned media agencies can produce original news, that has limited access to Web sites out of the country, and where, despite an Internet penetration of 39.9 percent (as of 2013) and wide adoption of social media (CNNIC, 2013), the government still closely monitors and filters what is being said on the Internet. How do citizens evaluate and trust information produced through both traditional mass media channels and peers on social media channels? Can decentralized individual actions with social media in an authoritarian society like China create an alternative power to traditional media?

To investigate the complex media environment in Chinese society, we focus on news information that includes both news reporting and opinion pieces related to news events. News information is important to study because it is the gateway for members of a society to become aware of and to make sense of the current social, cultural and political status, which oftentimes can spur people to perform civic duties, such as voting. We use the term official media to refer to mass media enterprises that are run or influenced by the state, generally involving professional journalists and editorial procedures. In China, examples of official news media outlets run by the state include Xinhua, CCTV, and People’s Daily. Independent commercial news portals or news sites such as Sina or Tencent do not have the autonomy to produce original news content, and instead can only reprint news articles from state-run news outlets (Esarey and Qiang, 2011; Stockmann, 2011). We thus refer to both state-run and commercial media companies under governmental control as official media. On the other hand, we use the term citizen media to refer to news that is posted and disseminated by citizens using various forms of social media. While social media can encompass a wide range of media types, in this paper we concentrate on MBs and SNSs as users of these two media platforms comprise about half of the Chinese Internet population.




Media control and historical events in China

Chinese citizens generally perceive official media acting as the mouthpiece of the government (Xiao, 2010). Throughout modern Chinese history, mass campaigns were an important feature of daily life. Individuals who represented ideas or practices that counteracted the government’s mass campaign were singled out for criticism. For example, in 1957, Mao called for the Hundred Flowers campaign to welcome constructive criticism. Professors and students around the country started to write memorials and wall posters urging for change. Yet this was soon followed by a series of purges of intellectuals with “dangerous ideas,” now known as the Anti-Righties campaign (Wasserstrom, 2010). Since the 1989 Tiananmen protest, the authorities have gone to extraordinary lengths to use media control to limit public awareness of protests and subsequent government interventions (Wasserstrom, 2010). The role mass media played in historical events like these, even though they occurred before the prevalence of the Internet, set a precedence that could influence how Chinese citizens trust the authorities and mass media in the current digital age (Chen and Shi, 2001). At the same time, historical patterns of quelling public dissent could also affect how Chinese citizens perceive the role of social media.

The media environment in China: 2012–2013

In recent years, as the media environment is evolving with the prevalence of the Internet and social media, so is government control. An illustration of such control is found with the case of Chen Guangcheng. In late April, 2012, Chen Guangcheng, a blind human rights activist from Shandong Province, China, daringly escaped from house arrest and arrived at the U.S. embassy in Beijing, where he negotiated with both the U.S. and Chinese top authorities. Chen’s case immediately attracted national and international media attention given its political sensitivity. Within 24 hours of his entry to the U.S. Embassy, censors had deleted discussions about Chen on Chinese social media Web sites (Jiang, 2012). Official Chinese media outlets, on the other hand, offered a series of brief reports which characterized Chen as a political pawn taken into the U.S. Embassy by Hillary Clinton (Xinhua News, 2012). Official news further accused Chen of cleverly playing the role of a dissident in order to seek benefits from a Western power (Nong, 2012).

Despite the censors, the first author observed over a period of seven days that Chen’s case still covertly circulated on China’s most popular microblogging site Sina Weibo (referred to from now on as “Weibo”). Posts that circumvented the censorship typically used several tactics, such as the insertion of random sequences of punctuation in between characters in Chen’s name, decomposing Chinese characters, or using contextual cues or codenames to refer to Chen (e.g., “the sunglass brother”, “the blind warrior”). These posts described Chen as a “normal citizen,” who reflected the authentic life under an authoritarian system. Posts explained that he should be well respected for his contributions to the community despite experiencing injustice. In short, users of social media sites like Weibo showed empathy, understanding, and positive interest toward Chen.

Of course, this is an extreme case where the government tried to suppress all user-generated content involved. We also do not know which side of the story is more accurate. Nevertheless, it illustrates the challenges Chinese citizens face: trying to grasp the meaning from, and assess credibility of, both media channels.

Chen’s case is not intended to suggest that most Chinese Internet users are political activists online. In fact, according to the 2013 Internet statistics report in China (CNNIC, 2013), most Chinese use the Internet for leisure such as instant messaging, search engines, online music and video, online gaming, microblogs, or social networking. As Wasserstrom (2010) pointed out, most Chinese citizens are neither extremely loyal to the government nor explicit dissidents. Yet, these people “in the middle” remain interested in political topics they find important (Wasserstrom, 2010) and their opinions of them may potentially confirm, complement, criticize, or dispute the government accounts. This “broader range of possibilities” [1] benefit individual citizens, for they now have enhanced autonomy to decide what to read and what to believe (Benkler, 2006).

Given the potential of the Internet and new media, the Chinese government monitors and filters sensitive content (Larmer, 2011) from various domestic social media sites, termed “networked authoritarianism” (MacKinnon, 2011). Sensitive content can include historical and political events, current news events, political sentiments, and even the names of government officials (e.g., the chairman of China). In addition to the regulations that are more or less open to the public’s awareness, Internet commentators hired by the government (yet appearing to be average users) are speculated to post comments favorable to the Party in order to subtly shift public opinion. This false grassroots phenomenon, called “opinion guidance,” guides dialogue on social and politically sensitive topics, oftentimes news.



Official media vs. citizen media

Fundamental differences exist between the news generated and disseminated through professional, official channels and by citizens (Esarey and Qiang, 2011; Reese and Dai, 2009; Wall, 2005). Such differences apply to both Western societies and to China. With traditional mass media production, news is written by a small number of professional journalists. Their practices and values adhere to institutional authority and economic resources (Benkler, 2006; Reese and Dai, 2009). A certain level of standards, accountability, and objectivity is therefore expected due to fact-checking and editorial oversight. However, there are a handful of drawbacks with official media. First, official news sets agendas (McCombs and Shaw, 1972) by choosing what news goes on the front page and what news should be hidden from the public. Second, even if the press was free of the agenda of a political party (which is not possible in an authoritarian society), a small number of professional journalists can only accept a limited amount of material (i.e., facts and opinions relevant to any current event) on which official news is based (Benkler, 2006).

In contrast to official news as “finished goods” (Benkler, 2006), citizen media content is always being produced in the process of spreading to a large number of audience members (Jenkins, et al., 2013). Since production rules are more relaxed, news can be written, filtered, interpreted, mutated, and redistributed in complex chains of dissemination, out of citizens’ simple desire to report news, express commentary, or pass on hearsay. Such evolving content provides a remedy to the drawbacks of mass media because average citizens now have a channel to provide eyewitness accounts. Lasica (2003) called this a “random act of journalism,” which dictates “report what you observe, analyze events in a meaningful way but, most of all, just be fair and tell the truth as you and your sources see it” [2]. Even if only a small proportion of social media users each report a tiny slice of information pertaining to a news event, given the large number of reports, they might potentially provide comprehensive views on the event with sufficient impact, as the long tail theory suggests (Anderson, 2006). Furthermore, unlike official news reports, which strive to be objective or voice one particular view, citizen reports can also convey reflections and interpretations from multiple points of view.

The consumer-turned–into–producer opportunity, however, has its own drawbacks. Since average citizens do not have the power or access to professional resources, user-generated news is often soft news (Thurman, 2008). In addition, people are more likely to interact with those who share the same views, which could result in opinion polarization that obstructs meaningful discussions (Sunstein, 2002; Yardi and boyd, 2010). Further, data represented on citizen media, for instance, the number of times a trending topic is re-posted (shared), can be manipulated by fraudulent accounts (Yu, et al., 2012). Even for well-intentioned users, irrational posting and sharing can easily spread rumors, distortions, and clearly erroneous information, causing public scandals (Liao and Shi, 2013). Thus, citizen news is two-sided: it can be trustworthy because of the citizen-centric values it embodies, but also untrustworthy due to concerns about its informality and information quality.



Research questions

Considering the frequent discrepancies in news from official and citizen media, news trustworthiness has become a pressing issue. While a considerable amount of attention (e.g., Qu, et al., 2011; Xiao, 2010) has been paid to the impact of citizen media in China, research is lacking that explicitly examines the trustworthiness of news (Wang and Mark, 2013). Though recent studies have investigated what makes news credible (e.g., Castillo, et al., 2011; Morris, et al., 2012), we seek to investigate the perception of trust from the audience perspective (Choi, et al., 2006). The reasons are two-fold: first, news is subject to interpretation, especially news commentary, which cannot (only) be assessed for its truthfulness; second, how much a news consumer trusts the news has a direct impact on how the news is interpreted and further disseminated. Thus, in this paper, we define trustworthiness as the perceived believability of information (Castillo, et al., 2011).

We maintain that users’ participation in citizen media news activity is related to their perception of the trustworthiness of the news. One action that users can take to assess the trustworthiness of the news is to verify the news. The actions that one takes on a citizen media news item (e.g., commenting on the news post to refute it or sharing the post to promote it) can further affect other users’ perception of the validity of that news post. Therefore, in this paper we focus on two main research thrusts: 1) factors that affect the perception of trust in official and citizen media news in the context of China; and, 2) the actions that users take to verify citizen media news.

Factors that affect Chinese citizens’ perceptions of news trustworthiness

Our first research thrust on perceived news trustworthiness is broken down into four research questions: R1 concerns both citizen and official media channels; R2–R4 focus on citizen media.

Perspective of news towards the government. Since official media is dedicated to disseminating state policies and promoting Party leadership, opinion guidance in citizen media tries to covertly promote pro-government sentiment, Chinese citizens may therefore trust news that affirms government views differently than that which criticizes them, regardless of the news source. We ask here: To what extent do citizens in China trust news information based on its views towards the government (R1)?

Strength of social ties. The stronger the tie (e.g., family, close friends), then generally the more similar two actors are in terms of information, experience, and values (Krackhardt, 1992). Because of this, people may place high trust on news information written or re-posted from strong ties in citizen media. Weak ties (e.g., acquaintances, friends’ friends), on the other hand, help disseminate novel information and bridge local networks into a larger community (Granovetter, 1973). With SNSs and MBs, news consumers tend to access new information or different viewpoints from weak ties. Therefore, we ask: What is the relationship between trust in news (posted or re-posted) and strength of ties in the source delivering the news (R2)?

Links to information sources. Documenting a source has been found to increase the credibility of news by providing additional contexts and evidence (Morris, et al., 2012). In the context of China, however, the effect of embedding an information source is uncertain because the source can point to official news reports that Chinese news readers may be skeptical of. We therefore ask: How do links to external information sources affect trust in news (R3)?

Collective participation. News posts can easily be re-posted, commented, or “liked.” These features provide aggregate measures of a “crowd effect” on citizen media posts. Posts that receive high attention, such as a large number of re-posts, could imply higher credibility (Castillo, et al., 2011). Yet, the meaning of these information cues is not always transparent. Yu, et al. (2012) found that these aggregated numbers could be artificially inflated. Bernstein, et al. (2013) found that the size of an “invisible audience” is much larger than a user would estimate, and that the actual audience cannot be predicted in any straightforward way based on the visible cues such as likes or comments. Further, in the context of Chinese media, these numbers may be manipulated by a commercial or state agenda. We thus ask: Does the crowd effect (e.g., comments or affirmations) increase or decrease trust in citizen news (R4)?

How do users in China verify news from citizen media?

Besides personal bias and selective exposure with regards to news consumption (Garrett, 2009; Hart and Nisbet, 2012; Liao and Fu, 2013), recent studies (Armstrong and McAdams, 2009; Morris, et al., 2012; Yang, et al., 2013) also suggested that readers spend minimal time to make quick judgments of information credibility, mostly relying on social cues such as author features or social media features (e.g., whether it is a retweet). While assessing the reliability of information is critical, the investigation into the process through which users verify news information is scarce. News verification is of particular interest because 1) why a user chooses to verify a news post can be a result of his/her perception of its trustworthiness; and 2) the actions after verification can further affect how other users perceive the trustworthiness of the news post. We thus ask two research questions related to news verification: What information is verified, and how (RQ5)? What actions do users take after verification (RQ6)?



Empirical studies: Methods

Two survey studies are reported in this paper. Survey 1 (RQ1–RQ4) focuses on factors that could affect Chinese citizens’ perceptions of news trustworthiness and was deployed online in April, 2012, for one month. Survey 2 (RQ5–RQ6) presents a descriptive account of how Chinese citizen media users verify news and was deployed online from 26 March to 9 May 2013. Both surveys were anonymous and in the Chinese language (summarized in the Appendix). They were deployed through snowball sampling via multiple links to the authors’ contacts of family, friends, and professional colleagues by e-mail. In addition, both survey invitations were posted on Chinese social network sites, microblogging sites, online forums, and university bulletin board systems. For Survey 1, all survey respondents were asked for recommendations of other contacts. For Survey 2, additional survey invitations were sent as personal messages to randomly chosen users: 170 from Sina Weibo and 737 from Renren, the most popular Chinese MBs and SNSs, respectively. In total, 578 responses for Survey 1 and 386 responses for Survey 2 were analyzed.

Table 1 summarizes demographic information of both survey samples. The gender distributions for Survey 1 and 2 are very similar to the Chinese Internet population as of January 2012 (CNNIC, 2012) and January 2013 (CNNIC, 2013), respectively. Our samples have a higher percentage of people under 30 years old than the general Internet user population (CNNIC, 2013) and are mostly representative of Chinese people who have higher education. The education distributions are, however, close to that of Weibo users (Sina Weibo Data Center, 2012).


Table 1: Demographic information of sample population of Survey 1 and Survey 2.
 Survey 1Survey 2
Age:18–25: 47.6%
26–30: 30.6%
31–40: 9.1%
41–50: 10.2%
50+: 2.5%
18–20: 3.4%
21–30: 75.6%
31–40: 10.1%
41–50: 6.8%
50+: 4.2%
Middle/High school3.8%4.4%
Graduate school49.2%50.1%




Results: Survey 1 — Perceived trustworthiness (RQ1–RQ4)


In this sample, respondents reported that they deliberately seek out news fairly actively: 36.8 percent of respondents read news several times a day and 27.2 percent read news once a day. Respondents also reported a fair amount of news exposure through citizen media: on a typical day, 77 percent respondents read at least one news item from either SNSs (65.6 percent) or MBs (54.3 percent). About one fifth do not use citizen media and 2.8 percent do not read news. These were excluded from the analyses.

Regarding news participation, 81.6 percent of those who have used citizen media and have read news from this channel have participated in at least one type of news activity. News activities in which people participate more frequently include commenting (69.6 percent), sharing URLs to news articles in citizen media (60.7 percent), posting URLs (48.6 percent), and contributing original news content (44.5 percent).

All questions regarding trust used a seven-point scale, where 1=strongly distrust, 7=strongly trust. This sample of Chinese news consumers place moderate trust in both official (mean=4.07, sd=1.63) and citizen media (mean=3.92, sd=1.22) with no significant difference, (F(1, 388)=2.218, p<.14). To unpack the differences of trust in official vs. citizen media, we created profiles of news readers who have different preferences of trust for these different media. Ratings of 5, 6, and 7 were combined into a variable of ‘high trust’ and ratings of 1, 2, and 3 into a variable of ‘low trust’. From this, we created four news reader profiles: ‘Traditionals’ (T) (23 percent; high trust in official, low trust in citizen), ‘New Generation’ (NG) (20 percent; low trust in official, high trust in citizen), ‘Cynics’ (32 percent; low trust in both) and ‘Gullible’ (25 percent; high trust in both). We did not count mid-ratings of 4; thus not every respondent is associated with a news reader profile. Table 2 shows the differences between Ts and NGs, the news reader profiles with the most contrasting perceptions of the two media channels: Ts have a higher proportion of women than the NGs, but do not differ significantly in age or other demographics. NGs adopted SNSs earlier than Ts, but no significant difference with MB adoption. Additionally, significantly more NGs than Ts contribute original news content and post URLs on SNSs.


Table 2: Differences between New Generations (NGs) and Traditionals (Ts).
 TraditionalsNew GenerationsTest results
Trust in official and citizen media:   
Official media (mean, sd)5.68 (0.67)2.53 (0.58) 
Citizen media (mean, sd)2.30 (0.70)5.39 (0.49)
Female59.3%40.7%chi-sq(1) = 4.41, p<.04
News participation:   
Original news39.6%63%Z-score = 2.3 (<.02)
Post URLs on SNSs41.5%67.4%Z-score = 2.6 (<.01)
Adoption of SNSs:   
Early adopter18.5%37%chi-sq(3)=8.07, p<.05
Early majority48.1%39.1%
Late majority14.8%19.6%


Results for RQ1–RQ4

For each research question, we first present the results based on a repeated measures ANOVA for the entire sample, which includes those classified into all four news reader profiles (see above) and those who were not classified into a news reader profile because they gave mid-point ratings. To best understand reasons for distinguishing trust in official and citizen media channels, we further investigated the research questions comparing between the news reader profiles with the most contrasting ratings, the Traditionals (T) and the New Generation (NG). The survey question response is the within-subjects variable and News Reader Profile is the between-subjects variable. Results for research questions 1–4 are shown in Table 3.


Table 3: Test results for research question 1–4.
Variables in the RQEntire sample (mean, sd)News Reader Profile: Traditional (mean, sd)News Reader Profile: New Generation (mean, sd)Bonferroni post hoc test
Affirms government3.49 (1.33)4.17 (1.16)2.70 (1.17)p<.001
Criticizes government3.99 (1.12)3.81 (1.22)4.57 (1.09)
Neutral4.49 (1.18)4.80 (1.12)4.61 (0.92)
F statistic, p-valueF(2,778)=93.52, p<.001Perspective x News Reader Type interaction:
F(2,192)=23.68, p<.001
Strong ties4.75 (1.29)4.91 (1.19)5.09 (1.12)p<.001
Weak ties4.09 (1.00)3.91 (1.01)4.45 (0.93)
No ties3.60 (1.10)3.33 (1.19)4.02 (0.95)
F statistic, p-valueF(2,696)=176.08, p<.001Tie Strength x Type interaction
F(2,172)=2.50, p<.09
With external links4.43 (1.20)4.85 (1.17)4.70 (1.00)N/A
No external links3.30 (1.04)2.89 (1.04)3.48 (1.05)
F statistic, p-valueF(1,351)= 258.95, p<.001URL x Type interaction
F(1,88)=5.38, p<.03
Posts w/o cmt or “likes”3.58 (1.06)3.74 (1.06)3.66 (1.14)p<.001
Posts w/ cmts4.07 (1.04)3.98 (1.04)4.49 (0.80)
Posts w/ cmts & “likes”4.30 (1.14)4.29 (1.34)4.80 (0.95)
F statistic, p-valuep<.001F(2,172)=5.64, p<.005


RQ1: Perspective of news towards the government and trust. Our entire sample has the lowest trust in news that affirms the government perspective, moderate trust in news that criticizes the government perspective, and highest trust in neutral portrayals of the government. Though a small effect, these three trust levels are significantly different from each other. Focusing only on news reader profiles Ts and NGs, there is a significant Perspective x News Reader Profile interaction: Ts have higher trust in news that affirms the government perspective; NGs have higher trust in news that criticizes the government perspective.

RQ2: Strength of ties and trust. For the entire sample, a significant difference exists between the types of relationships and trust: strong ties are associated with the highest trust, followed by weak ties, and then posts from strangers. There is a weak trend for an interaction of Tie Strength x News Reader Profile: as the tie strength grows weaker, the level of trust declines less for NGs than for Ts.

RQ3: Links to information sources and trust. For the entire sample, respondents reported significantly higher trust in posts with URLs than those without URLs. Comparing news reader profiles Ts and NGs, there is a significant URL x News Reader Profile interaction, indicating that the absence of a URL negatively affects trust for Ts more than NGs.

RQ4: Collective participation and trust. How does the crowd effect, i.e., comments or “likes” from other citizens, affect a consumer’s trust in news? For the entire sample, a significant increase in trust exists among plain posts, posts with comments, and posts with both comments and “likes.” A significant interaction between Crowd Effect x News Reader Profile shows that comments and “likes” increase perceived trustworthiness in news more for NGs than Ts.



Results: Survey 2 — News verification of citizen media (RQ5–RQ6)


Participants in the second study are generally active in news consumption: 72.5 percent reported that they read news several times a day and 22.5 percent read news about once a day. 86.8 percent and 74.5 percent of the sample report that they read news from MBs and SNSs, respectively. Since the focus of Survey 2 is on verification of news from citizen media, people who have never used citizen media for news were excluded from the analysis. In contrast to Survey 1, respondents in Survey 2 revealed higher trust in news from official media overall (mean=4.69, sd=1.57) than citizen media (mean=4.12, sd=1.08), (F(1, 339)= 32.95, p<.001).

Results for RQ5–RQ6

RQ5: Verification of information. When asked whether they have verified news information from citizen media in the past, 79 percent of the sample reported that they have. When asked how, and what information they have verified, we found that the majority of our sample checked the correctness of details (73.5 percent), credibility of the sources cited (72.6 percent), and soundness of argument (69.3 percent); on the other hand, fewer people checked the credibility of the author (24.2 percent) and even fewer the reliability of the number of re-posts and likes (5.1 percent).

When asked about how they verify news items from SNSs or MBs, the majority (79.5 percent) reported using a search engine to search for relevant information, followed by paying attention to relevant information on citizen media (55.7 percent), comparing and contrasting the reports of the same events from different sources (51.8 percent), tracing the source of the information (50.9 percent), and looking up reports from official media (43.6 percent).

RQ6: Reactions to news posts after verification. We use the terms “valid information” and “misinformation,” rather than “true and false,” to describe verified news from citizen media, considering the multiplicity of perspectives of any news events. Specifically, “valid information” refers to information that is accurate factually or through reasonable inference, whereas “misinformation” refers to inaccurate information or groundless speculations. According to self-reports, about half of the participants (46.3 percent) reported that they re-post the news post as-is when it is verified as valid. For news they considered misinformation, a fair amount of respondents (37.9 percent) chose to re-post the item along with the additional information they found (to dispute it, to warn other people, etc.), and about a third (32.2 percent) commented on the post. This is interesting because when asked about who is primarily responsible for news credibility in Survey 2, our respondents on average were more in agreement that it is the responsibility of the site moderators (mean=3.76, sd=1.21; 1=strongly disagree, 5=strongly agree) rather than themselves (mean=3.48, sd=1.25, F(1,204) = 4.84, p<.03). However, for those who have judged the news as misinformation, very few of them (8.4 percent) reported it to site moderators. Instead, the news “verifiers” tended to act on their own and confront misinformation by re-posting it with additional information or commenting.




In Survey 1, we investigated how people trust news in a society where official news is subject to authoritarian government control and where the use of citizen media is rapidly expanding. We found that up to two-thirds of Chinese citizens received news from citizen media on a typical day and over two-thirds of these citizen news consumers were active in news activities. In 2012, Chinese news consumers showed no significant difference in trust in official and citizen news channels, yet they showed particularly low trust in news that affirms governmental accounts. This is possibly because discussions of national news events (e.g., the case of Chen Guangcheng, described above) were strictly censored in citizen media at that time, much like the historical events in modern Chinese history where the voices of the people were silenced while party agenda was promoted. Restricting access to only government accounts of national news events could lead the citizenry to develop suspicions or distrust towards these news that praises the government. So how does the rise of citizen media — SNSs and MBs in our study — come into play in such an authoritarian society?

Citizen media as an alternative news platform

Citizen media provides an alternative news channel to government-controlled channels, one that leverages the advantage of personal relations and crowd reflection. Our results support this by showing that strong ties and the collective interpretation and affirmation from the crowd enhance people’s trust in news significantly. However, only about 10 percent of all Internet users we sampled in Survey 1, the news reader profile NGs, embraced citizen media over official media. They trusted news with comments and likes more than that of the news reader profile Ts; and the absence of information sources (oftentimes linked to an official source) affected them less negatively than Ts. One possible explanation is a familiarity effect (Hart and Nisbet, 2012): NGs adopted citizen media earlier and were more active in online news participation than Ts. We cannot say however that higher involvement caused high trust in citizen media. It could also be the reverse: because NGs generally trust content from citizen media, this could explain their higher participation in citizen media. Another explanation for the disparate levels of trust in the two media channels could be due to different political views, as reflected by the contrasting views of trust that NGs and Ts reported: NGs trust news that criticizes the government while Ts trust news that affirms government views.

Official media and citizen media as complementary force

In comparing trust in official and citizen news channels, Surveys 1 and 2 revealed different results: while the levels of trust were not significantly different in Survey 1, respondents placed higher trust in official media in Survey 2, indicating an on-going dynamic between these two media channels. One possible explanation, as discussed above, is that Survey 1 was conducted during the time period when the government strictly censored news of national interest and policed public opinions. Such forceful information blockage may have hindered people’s trust in official media at that time. Another possible explanation is that in April 2013, the month in which Survey 2 was deployed, the official media site started a new monthly column to uncover the top 10 rumor events on the Internet, most of which originated from, and propagated through, citizen media like Weibo (People, 2013). This could have caused the decrease in overall trust in citizen media between the time periods of these two studies.

Instead of seeing official and citizen media channels as two competing powers, they may actually complement and serve a watchdog function for each other. Several findings from our studies support this argument. First, as shown in Study 1, embedding a link to an original news source (which are predominantly official media sources), boosted trust. People may perceive the official news as more trustworthy when it is passed on by their friends or other peer citizens. This is in line with Katz and Lazarsfeld’s (1955) model of personal influence. The notion of interpersonal relationship as a mediating factor in news dissemination suggests that in an authoritarian society, citizen media can change how people perceive official media reports. Second, Study 2 found that the most common ways to verify citizen news include searching and synthesizing information from both media channels, e.g., using a search engine to search for relevant information, comparing and contrasting information from different sources, and looking up reports from official media. These observations suggest that official media and citizen media intersect in such a way that they can potentially hold the other accountable because citizens can now search across these two media channels for validation or discrepancy.

Implications for citizens in an authoritarian society

Citizens from repressive societies (such as Iraq or Egypt) have started to utilize citizen media to create counter-narratives in times of crises, compete with official stories, and fuel social reform (e.g., Al-Ani, et al., 2012; Mark, et al., 2012). Therefore, who is able to produce news and how users engage with news can be particularly important. In Survey 2, most of our sample reported to have verified news from citizen media, and some further distributed valid information or commented on misinformation after verification. These results reflect “enhanced autonomy,” where individuals independently seek out information and create their own expressions without asking for permission from authorities (Benkler, 2006).

These observations suggest that Chinese citizens are using the new Internet media tools to consume and produce new narratives beyond the limit of a single story published by official media. For example, in 2011, a crash between two high-speed bullet trains near the city of Wenzhou, China, killed dozens and injured almost 200 people. The official news agency claimed that design flaws caused the tragedy and attempted to control the media coverage of the incident, such as the exact number of the casualties, details of the cause, and reasons for the immediate disposal of the wreckage. Users of citizen media traced and synthesized the number of the casualties from different sources and published multiple unofficial narratives, contradicting the official report. These users also pieced together the speeds of the two colliding trains, the distance between them, the signal intervals, and then proposed several alternative causes for the collision. Soon after the alternative stories started to appear, the government enforced a strict censorship on this event, deleting these citizen accounts. As a result, citizen narratives only circulated among a limited number of audience members before the discussion eventually died down. Garret (2009) argued that “news stories are bundled information goods, often including a mix of viewpoints and evidence supporting multiple perspectives” [3]. Even though the officials did not confirm the unofficial narratives, citizen stories presented alternative perspectives opposite to the official story. This is an important milestone because citizen media is becoming a force for citizen empowerment within an authoritarian state where historically only state-owned media agencies can write news stories with a single viewpoint.

In this study, we identified a group of Chinese news consumers who embrace citizen media (NGs), while others still struggle to trust it as a channel for news. We also found that a proportion of users have never engaged in news verification; for those who have, less than half reported to have re-distributed relevant information back to citizen media. Therefore, there are still many challenges for more citizens to engage in exercising autonomy in the citizen media environment.

An update of the Chinese citizens’ media environment in 2014–2015

Since the surveys were conducted back in 2012 and 2013, China’s online media environment has witnessed significant changes. Soon after Xi Jinping assumed office in 2013, new cyber policies were enforced to overhaul the Internet (Xinhua News, 2013). A judicial document published in September 2013 stated that spreading rumor on social media could lead to three years in jail (Xinhua News, 2013). What is considered rumor is uncertain by nature, as rumors are essentially unconfirmed messages (Liao and Shi, 2013). On the one hand, such aggressive law enforcement could help weed out erroneous information. On the other hand, it made average users wonder: “who is going to dare say anything now?” (Blanchard, et al., 2013) By threatening Internet users with jail time and by publicly defaming a few influential users on Weibo, the Internet crackdown could easily turn into a repressive tool for officials to arrest anyone who criticizes government misconduct (Kaiman, 2013). As we discussed earlier, the majority of Chinese users use the Internet for leisure. During the time when citizen media was flourishing, we observed that Chinese users, at least a proportion of them, consumed news posts on citizen media, participated in news-related activities, exercised a level of autonomy to configure their trust in news posts, verified news posts, and contributed back to the citizen channel. Though not everyone was on board, we did see an emerging discourse. But when the cost of being involved in news activities became as high as being charged as a criminal act without a transparent process, not surprisingly, users started to migrate away from these social platforms that once supported a wider range of voices. According to the recent statistic reports of Chinese Internet use (CNNIC, 2014; CNNIC, 2015), the user populations of SNSs and MBs, both Web-based and mobile platforms, had been on a constant decline from mid-2013 to the end of 2014. It suggests that, though citizen media has demonstrated great potential to facilitate counter-power by the citizens during its heightened period in 2012 and 2013, such a user-centric media channel is considered a threat to authority, who can ultimately use explicitly repressive measures to impose its gripping power on the media.

Generalizability and limitations

A limitation of our study is that samples of both studies are most representative of the Chinese population who are of younger age groups and who have higher education. We can thus only generalize our findings to this type of population. In addition, because “opinion guidance” exists in China, we cannot rule out the possibility that some respondents may have been hired by the government.

There are several limitations specifically for Survey 2. A dichotomous yes/no was asked for verification actions and actions afterwards. We therefore do not know how often these actions took place. Also, as the actions related to verifying news are self-reported, we cannot assert that the answers reflect accurate behavior. Further, respondents might have answered the questions hypothetically since we did not specify which news events the survey questions were focused on. We hope to address these concerns in future research.




This paper discusses news trustworthiness and news verification in China before mid-2013. Citizen media has been changing how news is consumed and produced worldwide. This is especially important in societies such as China where the official media is subject to strict governmental control. Chinese citizens consume a fair amount of news from citizen media where they resort to social networks and crowd confirmations to evaluate the trustworthiness of news. The relation between citizen media and official media is evolving as citizens can now utilize both of these two media channels to search for validation or to discover misinformation. China is a special case: on the one hand, citizen media has opened a door for citizen empowerment; on the other hand, citizen media is regulated (in the format of censorship) and manipulated (in the format of opinion guidance) by the government. Situated in the special context, this study provides a first step in examining the on-going dynamics between official media and citizen media, and between the perception of trust in news and the participation in news verification. End of article


About the authors

Yiran Wang is a Ph.D. candidate in the Department of Informatics, Donald Bren School of Information and Computer Sciences at the University of California, Irvine.
E-mail: yiranw2 [at] uci [dot] edu

Gloria Mark is Professor in the Department of Informatics, Donald Bren School of Information and Computer Sciences at the University of California, Irvine.
E-mail: gmark [at] uci [dot] edu



1. Benkler, 2006, p. 9.

2. Lasica, 2003, p. 73.

3. Garret, 2009, p. 267.



Ban Al-Ani, Gloria Mark, Justin Chung, and Jennifer Jones, 2012. “The Egyptian blogosphere: A counter-narrative of the revolution,” CSCW ’12: Proceedings of the ACM 2012 conference on Computer Supported Cooperative Work, pp. 17–26.
doi:, accessed 17 January 2016.

Chris Anderson, 2006. The long tail: Why the future of business is selling less of more. New York: Hyperion.

Cory L. Armstrong and Melinda J. McAdams, 2009. “Blogs of information: How gender cues and individual motivations influence perceptions of credibility,” Journal of Computer-Mediated Communication, volume 14, number 3, pp. 435–456.
doi:, accessed 17 January 2016.

Yochai Benkler, 2006. The wealth of networks: How social production transforms markets and freedom. New Haven, Conn.: Yale University Press.

Michael S. Bernstein, Eytan Bakshy, Moira Burke, and Brian Karrer, 2013. “Quantifying the invisible audience in social networks,” CHI ’13: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 21–30.
doi:, accessed 17 January 2016.

Ben Blanchard, Hui Li, and Paul Carsten, 2013. “China threatens tough punishment for online rumour spreading,” Reuters (9 September), at, accessed 9 September 2013.

Manuel Castells, 2007. “Communication, power and counter-power in the network society,” International Journal of Communication, volume 1, pp. 238–266, and at, accessed 17 January 2016.

Carlos Castillo, Marcelo Mendoza, and Barbara Poblete, 2011. “Information credibility on Twitter,” WWW ’11: Proceedings of the 20th International Conference on World Wide Web, pp. 675–684.
doi:, accessed 17 January 2016.

Xueyi Chen and Tianjian Shi, 2001. “Media effects on political confidence and trust in the People’s Republic of China in the post-Tiananmen period,” East Asia, volume 19, number 3, pp. 84–118.
doi:, accessed 17 January 2016.

Junho H. Choi, James H. Watt, and Michael Lynch, 2006. “Perceptions of news credibility about the war in Iraq: Why war opponents perceived the Internet as the most credible medium,” Journal of Computer-Mediated Communication, volume 12, number 1, pp. 209-229.
doi:, accessed 17 January 2016.

China Internet Network Information Center (CNNIC), 2015. “35th statistics report on Internet development in China” (January), at, accessed January 2015.

China Internet Network Information Center (CNNIC), 2014. “34th statistics report on Internet development in China” (July), at, accessed July 2014.

China Internet Network Information Center (CNNIC), 2013. “31th statistics report on Internet development in China” (January), at, accessed January 2013.

China Internet Network Information Center (CNNIC), 2012. “29th statistics report on Internet development in China” (January), at, accessed January 2012.

Ashley Esarey and Xiao Qiang, 2011. “Digital communication and political change in China,” International Journal of Communication, volume 5, pp. 289–319, and at, accessed 17 January 2016.

R. Kelly Garrett, 2009. “Echo chambers online? Politically motivated selective exposure among Internet news users,” Journal of Computer-Mediated Communication, volume 14, number 2, pp. 265–285.
doi:, accessed 17 January 2016.

Mark S. Granovetter, 1973. “The strength of weak ties,” American Journal of Sociology, volume 78, number 6, pp. 1,360–1,380.
doi:, accessed 17 January 2016.

P. Sol Hart and Erik C. Nisbet, 2012. “Boomerang effects in science communication: How motivated reasoning and identity cues amplify opinion polarization about climate mitigation policies,” Communication Research, volume 39, number 6, pp. 701–723.
doi:, accessed 17 January 2016.

Henry Jenkins, 2006. Convergence culture: Where old and new media collide. New York: New York University Press.

Henry Jenkins, Sam Ford, and Joshua Green, 2013. Spreadable media: Creating value and meaning in a networked culture. New York: New York University Press.

Steven Jiang, 2012. “Chinese censor block news on blind activist’s escape,” CNN (9 May), at, accessed 30 April 2012.

Jonathan Kaiman, 2013. “China cracks down on social media with threat of jail for ‘online rumours’,” Guardian (10 September), at, accessed 10 September 2013.

Elihu Katz and Paul F. Lazarsfeld, 1955. Personal influence: The part played by people in the flow of mass communications. Glencoe, Ill.: Free Press.

David Krackhardt, 1992. “The strength of strong ties: The importance of philos in organizations,” In: Nitin Nohria and Robert G. Eccles (editors). Networks and organizations: Structure, form, and action. Boston, Mass.: Harvard Business School Press, pp. 216–239.

J.D. Lasica, 2003. “Blogs and journalism need each other,” Nieman Reports, volume 57, number 3, pp. 70–74, and at, accessed 17 January 2016.

Brook Larmer, 2011. “Where an Internet joke is not just a joke,” New York Times (26 October), at, accessed 26 October 2011.

Q. Vera Liao and Wai-Tai Fu, 2013. “Beyond the filter bubble: Interactive effects of perceived threat and topic involvement on selective exposure to information,” CHI ’13: Proceedings of SIGCHI Conference on Human Factors in Computing Systems, pp. 2,359–2,368.
doi:, accessed 17 January 2016.

Qinying Liao and Lei Shi, 2013. “She gets a sports car from our donation: Rumor transmission in a Chinese microblogging community,” CSCW ’13: Proceedings of the Conference on Computer Supported Cooperative Work, pp. 587–598.
doi:, accessed 17 January 2016.

Rebecca MacKinnon, 2011. “China’s ‘networked authoritarianism’,” Journal of Democracy, volume 22, number 2, pp. 32–46.
doi:, accessed 17 January 2016.

Gloria Mark, Mossaab Bagdouri, Leysia Palen, James Martin, Ban Al-Ani, and Kenneth Anderson, 2012. “Blogs as a collective war diary,” CSCW ’12: Proceedings of the ACM Conference on Computer Supported Cooperative Work, pp. 37–46.
doi:, accessed 17 January 2016.

Maxwell E. McCombs and Donald L. Shaw, 1972. “The agenda-setting function of mass media,” Public Opinion Quarterly, volume 36, number 2, pp. 176–187.
doi:, accessed 17 January 2016.

Meredith R. Morris, Scott Counts, Asta Roseway, Aaron Hoff, and Julia Schwarz, 2012. “Tweeting is believing? Understanding microblog credibility perceptions,” CSCW ’12: Proceedings of the ACM Conference on Computer Supported Cooperative Work, pp. 441–450.
doi:, accessed 17 January 2016.

Mo Nong, 2012. “One leaf is not the whole forest,” China Daily (4 May), at, accessed 4 May 2012.

People, 2013. at, accessed April 2013.

Yan Qu, Chen Huang, Pengyi Zhang, and Jun Zhang, 2011. “Microblogging after a major disaster in China: A case study of the 2010 Yushu earthquake,” CSCW ’11: Proceedings of the ACM Conference on Computer Supported Cooperative Work, pp. 25–34.
doi:, accessed 17 January 2016.

Kenneth Rapoza, 2011. “In China train crash, Sina’s Weibo breaks news,“ Forbes (1 August), at, accessed 1 August 2011.

Stephen D. Reese and Jia Dai, 2009. “Citizen journalism in the global news arena: China’s new media critics,” In: Stuart Allan and Einar Thorsen (editors). Citizen journalism: Global perspectives. New York: Peter Lang, pp. 221–231.

Clay Shirky, 2010. Cognitive surplus: Creativity and generosity in a connected age. New York: Penguin Press.

Sina Weibo Data Center, 2012. “2012 statistics report on user development in Sina Weibo,” at, accessed 26 October 2012.

Daniela Stockmann, 2011. “What kind of information does the public demand? Getting the news during the 2005 anti-Japanese protests,” In: Susan L. Shirk (editor). Changing media, changing China. New York: Oxford University Press, pp. 175–201.

Jonathan Sullivan, 2012. “A tale of two microblogs in China,” Media, Culture & Society, volume 34, number 6, pp. 773–783.
doi:, accessed 17 January 2016.

Cass Sunstein, 2002. “The law of group polarization,” Journal of Political Philosophy, volume 10, number 2, pp. 175–195.
doi:, accessed 17 January 2016.

Lijun Tang and Helen Sampson, 2012. “The interaction between mass media and the Internet in non-democratic states: The case of China,” Media, Culture & Society, volume 34, number 4, pp. 457–471.
doi:, accessed 17 January 2016.

Neil Thurman, 2008. “Forums for citizen journalists? Adoption of user generated content initiatives by online news media,” New Media & Society, volume 10, number 1, pp. 139–157.
doi:, accessed 17 January 2016.

Melissa Wall, 2005. “‘Blogs of war’ Weblogs as news,” Journalism, volume 6, number 2, pp. 153–172.
doi:, accessed 17 January 2016.

Yiran Wang and Gloria Mark, 2013. “Trust in online news: Comparing social media and official media use by Chinese citizens,” CSCW ’13: Proceedings of the Conference on Computer Supported Cooperative Work, pp. 599–610.
doi:, accessed 17 January 2016.

Jeffery N. Wasserstrom, 2010. China in the 21st century: What everyone needs to know. New York: Oxford University Press.

Wikipedia, 2015. “Websites blocked in mainland China,” at, accessed 3 August 2015.

Qiang Xiao, 2010. “The rise of online public opinion and its political impact,” In: Susan L. Shirk (editor). Changing media, changing China. New York: Oxford University Press, pp. 202–224.

Xinhua News, 2013. “Severe penalties for defamatory retweets in China” (9 September), at, accessed 9 September 2013.

Xinhua News, 2012. “China demands U.S. apology on Chen Guangcheng’s entering U.S. embassy” (2 May), at, accessed 2 May 2012.

Jiang Yang, Scott Counts, Meredith R. Morris, and Aaron Hoff, 2013. “Microblog credibility perceptions: Comparing the USA and China,” CSCW ’13: Proceedings of the Conference on Computer Supported Cooperative Work, pp. 575–586.
doi:, accessed 17 January 2016.

Sarita Yardi and danah boyd, 2010. “Dynamic debates: An analysis of group polarization over time on Twitter,” Bulletin of Science, Technology & Society, volume 30, number 5, pp. 316–327.
doi:, accessed 17 January 2016.

Louis L. Yu, Sitaram Asur, and Bernardo A. Huberman, 2012. “Artificial inflation: The real story of trends and trend-setters in Sina Weibo,” 2012 International Conference on Privacy, Security, Risk and Trust (PASSAT) and 2012 International Confernece on Social Computing (SocialCom), pp. 514–519.
doi:, accessed 17 January 2016.




Summary of survey questions.
RQ1: In general, how much do you trust news with the following perspectives: 1) Perspectives that AFFIRM the official account of the government; 2) Perspectives that CRITICIZE official account of the government; 3) Perspectives that are NEUTRAL.
RQ2: In general, on social network sites or microblogging sites, how much do you trust news items written by: 1) A close friend; 2) An acquaintance; 3) A person you don’t know.
RQ3: In general, how much do you trust news items on social network sites or microblogging sites as follows: 1) Posts with links to an original source; 2) Posts without links to an original source.
RQ4: In general, how much do you trust news items on social network sites or microblogging sites as follows: 1) Post only; 2) Post with comments; 3) Post with ”likes”; 4) Post with both comments and “likes”.
RQ5: 1) What types of information have you ACTUALLY CHECKED when you verified news items from SNSs or MBs? (Please check all that apply.) 2) How did you verify news items from SNSs or MBs? (Please check all that apply.)
RQ6: 1) What did you do after you had verified that the news item was MISINFORMATION? (Please check all that apply.) 2) What did you do after you had verified that the news item was VALID? (Please check all that apply.)



Editorial history

Received 13 August 2015; accepted 13 January 2016.

Copyright © 2016, First Monday.
Copyright © 2016, Yiran Wang and Gloria Mark.

News trustworthiness and verification in China: The tension of dual media channels
by Yiran Wang and Gloria Mark.
First Monday, Volume 21, Number 2 - 1 February 2016

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2019. ISSN 1396-0466.