Alternative framing: The effect of the Internet on political support in authoritarian China

Abstract

This study seeks to identify and test a mechanism through which the Internet influences public support in an authoritarian environment in which alternative information is strictly censored by the state. Through online discussions, web users often interpret sanctioned news information in directions different from or even opposite to the intention of the authoritarian state. This alternative framing on the Internet can strongly affect the political views of web users. Through an experimental study conducted in China, we find that subjects exposed to alternative online framing generally hold lower levels of policy support and evaluate government performance more negatively. This finding implies that even though the access to information on sensitive topics is effectively controlled by the government, the diffusion capabilities of the Internet can still undermine the support basis of the seemingly stable authoritarian regime.

Keywords

China experiment framing Internet effect political support

Introduction

The recent democratic movements in the Middle East and North Africa have re-energised the long-standing debate over the democratic potential of the Internet in an authoritarian environment (Farrell, 2012; Howard and Muzammil, 2011; Lynch, 2011; Tufekci and Wilson, 2012). The evident role of the Internet and its associated social media in the protests of the Arab Spring has strengthened the belief that digital media can help bring down the remaining authoritarian regimes. However, later developments in countries such as China, Singapore, Cuba and Iran suggest that the Internet is still effectively controlled by authoritarian governments and is unable to facilitate any significant political upheavals. Many authoritarian regimes even reacted to the Arab Spring by fine-tuning and reinforcing their censorship programmes. This contradictory picture of the role of the Internet raises questions about the Internet’s independent role in authoritarian countries. Is the Internet merely a tool that can benefit both ordinary citizens and authoritarian states? What is the effect of the Internet in places where there are no Internet-facilitated political protests?

We address these questions by turning attention to the micro-process through which the daily use of the Internet changes individuals’ view towards the authoritarian state. In particular, we argue that ‘alternative framing’, facilitated by the diffusion of the Internet, can strongly affect popular support for an authoritarian regime. Framing refers to a process of selecting ‘some aspects of a perceived reality and mak[ing] them more salient in a communicating text’ (Entman, 1993: 52). Compared with framing of the traditional media, a distinctive feature of Internet framing is the active role of web users in framing news events. Through online comments, blog posts, forum discussions and other interactive features of the Internet, dissident opinion leaders and, more often, ordinary web users have gained venues to express their views on news events and issues. The interpretations and perspectives expressed by web users often construct social realities in directions that are different from or even opposite to the ones propagandised by the authoritarian state. This democratic nature of Internet framing entails particularly important implications in authoritarian countries such as China, where overt news reporting is strictly monitored and media framing has traditionally been dominated by the state.

To examine empirically the effect of alternative framing, we conducted a laboratory experiment in a major Chinese university. The news of China’s recent space programme was chosen to present different frames. While the government framed the space programme as the manifestation of the nation’s technological advancement and the righteousness of the party’s leadership, a significant portion of the Internet discussion framed the event as a ‘performance project’ aimed at boosting the regime’s legitimacy and, more importantly, a misuse of resources that could otherwise have been used to improve the well-being of ordinary citizens. The experimental results show that exposure to the alternative framing embodied in Internet discussions significantly reduces the level of participants’ support for the space programme policy and their evaluation of government performance in improving citizens’ well-being.

The contribution of this study is twofold. First, it constitutes one of the first attempts to theoretically specify and empirically test the mechanism through which the Internet influences political attitudes in an authoritarian environment. It offers empirical support for the independent role of the Internet and suggests that the Internet is not merely a tool for political activists to organise collective actions or for the state to consolidate authoritarian rule but, rather, an independent causal factor that generates the makeover of the public’s mindset. Second, this study bridges the study of the Internet with framing studies in the traditional media. On the one hand, it builds on an established paradigm of traditional media research to understand the Internet effect. On the other hand, it adds a new dimension to traditional issue framing by emphasising that the online framing process has been strongly shaped by the emergence of large numbers of active web users.

Literature review: The Internet effect in authoritarian countries

The conventional optimism about the effect of the Internet in authoritarian countries is built mainly upon three tenets (Diamond, 2010; Howard, 2010; Shirky, 2011; Zheng and Wu, 2005). First, it is believed that the expanded access to alternative information offered by the Internet challenges the information control of authoritarian regimes. The corruption of power, brutality and social injustice can easily be exposed on the Internet due to its anonymity, fast speed and low cost of communication (Best and Wade, 2009: 256; Diamond, 2010: 76). Second, digital information technology is praised for facilitating anti-regime political movements (Chase and Mulvenon, 2002; Lynch, 2011: 304–306; Shirky, 2009). It has been documented that political activists have used the Internet to disseminate information, mobilise forces for change and organise actions in a number of protests (Chowdhury, 2008; Diamond, 2010: 79; McFaul, 2005; Rheingold, 2002), which was notably manifested in the Arab Spring (Lotan et al., 2011). Finally, the Internet can foster the development of civil society by pluralising the flow of information, widening and intensifying public deliberation, promoting alternative channels of social interaction, and facilitating networking among activists (Diamond, 2010: 71–72; Hill and Sen, 2000; Lynch, 2011: 306–307; Mercer, 2004; Murphy, 2009; Yang, G, 2003).

Many other scholars, however, question such utopian views of the Internet. Their scepticism stems mainly from the success of governmental control over the Internet. Through sophisticated regulation and censorship, the people of China, along with those of Vietnam, Cuba, Iran and Singapore, have not received the benefit of the democratising effects of the Internet (Boas, 2006; Harwit and Clark, 2001; Kalathil and Boas, 2001, 2003; Rodan, 1998; Taubman, 1998). By employing technological and institutional means, these authoritarian regimes have managed to use the growth of the Internet to help economic development, technology innovation and globalisation and, at the same time, to reduce its harmful political effects (Bueno de Mesquita and Downs, 2005; Morozov, 2011). Researchers have argued that, particularly in China, by controlling the technological infrastructure, co-opting private Internet companies, documenting the ‘real-name’ access and developing sophisticated censorship systems, the regime has imposed a close grip on cyberspace and has successfully transformed itself into a ‘networked authoritarianism’ (Hachigian, 2001; Harwit and Clark, 2001; Hassid, 2008; Hung, 2010).

The current debate over the potential of the Internet in authoritarian countries focuses primarily on the ‘revolutionary’ role of the Internet in contentious politics and, specifically, its contribution to protests, large-scale upheavals and dramatic cases of regime change (Lynch, 2011: 307; Shirky, 2011: 30). It must be borne in mind, however, that the Internet produces ‘change over years and decades, not weeks or months’ (Shirky, 2011: 30) by eroding the ability of states to monopolise information and arguments and by exacerbating people’s ‘dissatisfaction with matters of economics or day-to-day governance’ of the state (Lynch, 2011: 5). Inadequate scholarly effort, unfortunately, has been spent on theoretically explaining or empirically testing how daily consumption of the Internet can influence one’s views towards the authoritarian regime in an environment in which digital information is severely filtered and censored. As pointed out by Farrell (2012: 38), the question of whether the Internet contributes to democracy is unanswerable because it ‘proposes no specific theory as to the connection’ between the Internet and a political outcome. In fact, the number of quantitative studies on this topic is very limited.

This study explores the causal mechanism of the Internet’s effect on attitudinal change at the individual level by using China as a critical case. Such an empirical test of the micro-process of the Internet effect can help disentangle the causal effects of the Internet and provide empirical support for arguments in regard to the political contribution of the Internet. Specifically, this study argues that even if there is tight state control over the access to deviant information, exposure to online discussions of sanctioned information can still undermine people’s support of the government, and one of the key mechanisms of this undermining effect is user-generated alternative framing.

Issue framing and the Internet

Issue framing has been regarded as an important mechanism through which the traditional media affects public attitudes. Frames refer to the modes of presentation and interpretation that construct social realities. Media framing works by either presenting logically equivalent content differently, casting it in a negative or positive light, or selecting some parts of a news item and making them salient, while ignoring or downplaying other parts (Lecheler and De Vreese, 2011: 961). Media framing involves ‘problem definition, causal interpretation, moral evaluation, and/or treatment recommendation’ (Entman, 1993: 52). Framing is critical to shaping people’s political attitudes because it helps individuals make sense of an otherwise meaningless succession of events (Goffman, 1974; Iyengar, 1991; Kinder and Sanders, 1990; Lecheler and De Vreese, 2011; Nelson et al., 1997; Slothuus, 2008).

The news information on the Internet also contains frames, and some media researchers have examined the framing effect of the Internet (Coleman and Hall, 2001; Kopacz, 2008; Song, 2007; Xenos and Foot, 2005; Zhou and Moy, 2007; Zillmann et al., 2004). Zillmann et al. (2004), for instance, argue that compared with traditional media users, Internet magazine viewers have greater freedom to select articles to read and thus demonstrate a stronger selective-exposure effect of news frames. Kopacz (2008) notes that the Internet provides candidates with unparalleled opportunities to discuss their issue stances and enables them to escape the traditional filtering by news reporters and pundits in the conventional media.

However, an important aspect of Internet framing – the active role of Internet users in framing news events – has not received adequate attention. What makes the Internet distinctive is the change in the relative position of journalists and the audience in constructing news, including frame building. In the traditional media, the role of media users is very limited due to the prohibitive costs of message delivery and the restrictions imposed by government regulations. Accordingly, frame researchers have focused mainly on the influence of internal factors, such as journalists’ political orientation, professional values and organisational constraints (Zhou and Moy, 2007: 81).

The rise of the Internet has tremendously increased the opportunities and the capacities of the public to participate in frame building. Compared with the traditional media, web-based communication is decentralised. Web users can actively contribute to both information dissemination and issue framing. Through blogging, commenting, discussing and other means, almost any individual or organisation that has Internet access can discuss political or policy issues, publicise their thoughts, and promote their ideas, and thus contribute to the framing of news events and issues. The one-way asymmetric model of traditional communication has been changed to a dialogical type of journalism, through which news production becomes a collective endeavour (Benkler, 2006; Lowrey and Anderson, 2005; Mitchelstein and Boczkowski, 2009: 573). The meanings and values developed through online discourse and attached to news events by web users are alternatives to those presented by journalists, politicians and organisations.

Internet framing in China

Accompanied by rapid market reform, the mass media in China have experienced a great deal of change in recent decades. In terms of the structure of the market, the official media are no longer the only information source. Commercialisation has led to the growth of a number of media outlets that are not bankrolled by the government (Tang and Iyengar, 2011; Thomas and Nain, 2005). In terms of news content, stories about the negative side of society, like corruption, poverty and violation of norms, and political uprisings in other authoritarian countries have often come to be seen in TV programmes and newspapers (Hassid, 2008; Shirk, 2010; Stockmann, 2011). However, although the commercialisation of the traditional media contributes to offering alternative information for the public, it has not yet transformed the overall pro-regime effect of the media in China. In fact, recent studies of the media effect have consistently revealed that media consumption is significantly associated with higher levels of political support for the system and its political institutions (Kennedy, 2009; Stockmann and Gallagher, 2011; Tang, 2005; Yang and Tang, 2010). This is largely due to the fact that the official media can still be effectively controlled by the state. As scholars of the Chinese media have observed, even when some commercial media outlets do dare to challenge the official directions on major issues, the state is able to quickly step in and steer the deviant reporting back to the official line (Stockmann, 2011).

Compared with the tools of the traditional media, the Internet proves far more difficult to control given the sheer number of users and their activities on the web. In particular, in this study, we stress that the interactive features of the Internet enable users to participate in news making by framing social political issues as they prefer. This user-participated framing entails especially important implications in China, where, among the traditional media, competing frames of news events were limited or non-existent before the rise of the Internet.

First, user-generated frames on the Internet constitute the most viable, if not the only, alternatives to the official frames provided by the state propaganda machinery. Under the traditional media environment in China, the authoritarian regime monopolised not only the supply of news information, but also the framing of information. The regime has attempted to control the interpretation of important events and sensitive issues through reinforcing the journalists’ mouthpiece role in frame building and/or directly interrupting the deviant framing processes of media coverage (Zhou and Moy, 2007: 84). As a result, news frames used by Chinese journalists have largely been consistent with the official line (Chang et al., 1998; Luther and Zhou, 2005; Pan et al., 1999; Yang, J, 2003). The rise of the Internet enables ordinary citizens to act as the agents of framing in news making for the first time in China. With all the interactive features of the Internet, active web users have gained the most effective platforms to publicise their own understandings of the nature of events and issues, interpretations of the cause or the consequences of social happenings, evaluations of institutions or politicians, and the solutions they prefer. At the very least, this means that in the new media environment in China, not only news facts, but also news frames, no longer have to come from journalists or their political overlords. Ordinary web viewers, on an unprecedentedly large scale, are exposed to those interpretations and views that are not fabricated by propaganda agencies of the state.

Second, the alternative frames that emerge from online discourse are often different from or opposite to the intention of the authoritarian state. This is due to the lack of ways to express alternatives through formal institutional approaches or through traditional media outlets. Overall, in China’s cyberspace, the tone of discussion forums tends to be negative and even cynical; the content of discussion often quickly moves from specific news events to general problems in the political, economic and legal systems (Herold and Marolt, 2011; Sullivan, 2012; Tang and Yang, 2011; Yang, 2009; Zhou, 2009). Either promoted by online activists or generated through spontaneous and grassroots deliberation, a few counter-official frames can be identified in online discourse in contemporary China. For instance, while the official media report a crime that involves officials as a judicial case, netizens tend to frame it as a case about the abuse of power and social injustice. While the state media characterise the corruption of government officials as an isolated case and attribute it to a lack of oversight and self-discipline, dissenting individuals on the web associate it with the undemocratic nature of a political system that lacks checks and balances.

Third, the importance of such alternative framing on the Internet is further augmented by the fact that the access to alternative information sources is limited by the control of the Chinese state and the self-censorship of websites. As widely noted, the Internet in China is severely censored (Boas, 2006; Kalathil and Boas, 2003; Rodan, 1998; Taubman, 1998). However, the focus of state control is on limiting the chance of citizens to have access to the news facts, especially those deemed harmful for regime stability, not on limiting the interpretations of seemingly safe facts. The government’s ban on online discussions has targeted mostly sensitive topics, such as the Tiananmen Square protest, Jasmine Revolution and power struggles of top leaders. A recent comprehensive survey of social media websites in China conducted by King et al. (2012) confirms this pattern and shows that negative, even vitriolic, criticism of the state, political elites and public policies is allowed on the web as long as it has no potential for collective action (Sullivan, 2012; Tang and Yang, 2011; Yang, 2009).

Online discussion of sanctioned issues or seemingly safe topics, however, can go as rogue as do subversive news facts. A case in point is the heated online discussion triggered by the death of Steve Jobs, the former Chief Executive Officer of Apple Inc. One line of discussion asks why China does not have an innovator like Steve Jobs. A general theme that emerged from this discussion is that although China is not short of intelligent minds, there are too many restrictions and interventions, many of which are politically relevant, for these minds to emerge as a force. The government is blamed for its part in providing education that does not value critical thinking, monopolising economic resources into the hands of a few with connections and failing to protect property rights and encourage innovation. A widely shared blog post states, ‘If Apple is a fruit in a tree, its branches are the freedom of thinking and to create, and the root is the constitutional democracy.’

Such examples are abundant in China’s cyberspace. Due to alternative framing enabled by the Internet, in a country where politics is supposed to be a taboo, everything can become politically relevant. As long as websites allow features such as the ability to post comments or to blog, active web users have an opportunity to voice their dissident views, and ordinary web viewers have an opportunity to come across these alternative views. While Chinese rulers may have been successful in reducing the amount of subversive news facts on the web, they are still exposed to the challenges posed by the alternative framing of Internet information.

Experimental design

To study the effect of the Internet’s alternative framing on political support, we conducted a post-test-only randomised experiment. Experimentation has been the most frequently used method in the research of media framing. We choose it to study the Internet because it enables us to empirically identify the causal mechanism we selected for study and to test the Internet effect with greater confidence due to the control of the confounding effect of other variables.

Choosing the news event

The news story that we chose to embody different frames is an event in China’s recent space programme – the docking of the Shenzhou 8 capsule with the Tiangong 1 module.¹ We chose this news story for the following reasons. First, the frames employed by the state and Internet discussants to construct the news are ‘thematic frames’ (Iyengar, 1991). Similar frames are frequently found in China’s political discourse and repeatedly adopted to construct and interpret many government-initiated projects (e.g. the Three Gorge Dam). A study of such generic frames enables us to generalise findings to a wider range of issues. Second, a study of the frames of the space programme can make a strong case for the effect of alternative framing of the Internet. Compared with cases of official corruption, police brutality, social injustice and other high-profile incidents in contemporary China, the success of space docking in itself is a piece of positive information for the government and is more likely to induce public support. It is reasonable to expect that alternative framing of the Internet is less likely to function in this event than in other news events.

Presenting the frames

To present the official framing of the government, we took passages from Xinhuanet.com , the official website of the state-run Xinhua News Agency, and edited them together with news fact using web design software. The edited official framing states that the space programme has been carried out under the rightful ideological guidance of the party leaders and with strong material support from the party centre in Beijing: ‘It proves that the party center is absolutely right in making the important decisions about manned space projects’. The docking of the spacecraft is thus framed as the success of government efforts to advance technology and the righteousness of the party leadership (see details in Online Appendix 1, available at: http://ips.sagepub.com).

To present the alternative framing, we took news comments made by Internet users from two of the most popular commercial websites ( Sina.com and Netease.com) and edited them together. Like in a real forum, most comments point to two themes that are deviant from the state framing. First, they discredit the space programme as a ‘performance project’, a project that the government uses to boost political legitimacy. Second, they do not necessarily value the technological achievement of the space programme. Instead, they stress the urgency to improve the well-being of ordinary citizens (Minsheng), including alleviating poverty, increasing educational spending and providing social welfare to those in need. The huge investment in space programmes is thus framed as a misuse of valuable resources that should have been allocated for improving people’s livelihoods and providing social welfare. To make the Internet discussion realistic, we also placed a few positive but short comments (e.g. ‘Congratulations!’) among the mostly negative ones.

Procedures

We recruited a total of 141 students from a major university in eastern China for our research project. The recruitment advertisement stated that the research project was about new media use and its effects among college students. Interested students were paid a small fee for their time. The student sample raises a concern about the generalisability of experimental findings. While we are aware of the potential problem of external validity, we believe that the student sample employed in this study can provide a reliable test of the mechanism of alternative framing for at least two reasons. First, instead of being homogeneous, our sample (as summarised in Appendix 1) varies considerably on important demographic and political variables. For instance, about half of the participants are female, and half come from a rural family background. The parents’ education ranges from 1 (no formal education) to 7 (with graduate degrees) and varies with a standard deviation of 1.33. Family economic situation ranges from the lowest to the highest on a scale from 1 to 10 with a standard deviation of 1.57. Second, college students are the most reliable pool of netizens in China. Internet users in China cluster among those who are relatively highly educated and young (CNNIC, 2012; Liu, 2011).² Although we do not intend our findings to apply to the whole netizen population, it is reasonable to believe that the effect of alternative framing observed in our sample does not differ drastically from that among most other web users in China.

The experiment was conducted in five evenings in a computer lab in November 2011. A pilot experiment was conducted beforehand to ensure that the experimental set-up and procedures were appropriate. Once seated in the lab, participants were asked to complete a questionnaire probing their demographic backgrounds. They were instructed to carefully read questions and information and assured that their responses would be kept strictly confidential.³ Then, the participants were randomly directed to read one of three types of web pages on computers: dry information of the docking (46 participants, control group), information with official framing (47 participants, official group) and information with online alternative framing (48 participants, alternative group).

Immediately after viewing those web pages, the participants were asked to complete a post-test questionnaire. We first asked them to answer a couple of factual questions in regard to the docking procedure itself. We then asked a set of questions that concerned the level of support for the government in various aspects. Questions in regard to the use of the media and the Internet were also asked. After the survey, the students were debriefed on the purpose of this research and the lab manipulation.

Variables and measurement

We aim to show whether the exposure to online alternative frames alters one’s political support. We compare participants’ political support in three dimensions: (1) the support for the specific policy (i.e. space programme); (2) the evaluation of government performance (advancing technology and improving people’s well-being); and (3) the overall support for the regime (diffuse support). As emphasised by Norris (1999: 9), ‘political support needs to be understood as a multi-dimensional phenomenon … since there are significant theoretical and empirical gradations within different parts of the regime’. In order to reveal the ways in which online framing affects ordinary Chinese people’s support for Chinese Communist Party (CCP) rule, we tailored Easton’s (1965; see also Chen, 2004; Norris, 1999) framework of specific support and diffuse support into three sub-dimensions: policy support, performance support and diffuse regime support. As noted by Easton (1965, 1975), while specific support is formed in response to specific policies and the temporal performance of the incumbent authority, diffuse support tends to be more stable and reflects citizens’ deep-rooted commitment to the political system. Together, they can be seen as a continuum that moves from the most specific support for the space programme to the most diffuse support for the regime. We expect that the negative impact of Internet framing tends to decrease as the objects of political support become less specific.

We measure one’s policy support for the space programme by asking the participants to answer the question: ‘To what extent do you support China’s policy of space programmes like the docking of Tiangong 1 and Shenzhou 8?’ The answers range from ‘strongly oppose’ to ‘strongly support’ on a five-point scale.

We fashion four items to capture the respondents’ evaluation of government performance: promoting economic development, advancing technology, improving people’s well-being and fighting official corruption. For each of the items, the participants were asked to grade government policy performance on a five-point scale, ranging from ‘very bad’ to ‘very good’. Among the four policy domains, we focus on ‘technology advancement’ and ‘well-being improvement’, as these two areas are directly related to either the official framing or the alternative framing of this particular event.

To measure diffuse political support, we average the degree of participants’ agreement to the following two statements: (1) ‘I believe that the current political system truly represents and serves the interest of the Chinese people’; and (2) ‘Although there are faults in the political system of our country, I think that it best suits China’s situation’. The responses range from ‘strongly disagree’ to ‘strongly agree’ on a five-point scale. The summary statistics for these dependent variables and other variables used in this study are reported in Table A1 in Appendix 1.

Analyses and results

We first estimate the effects of the Internet’s alternative framing on political support by comparing the mean scores of the three different groups and conducting independent sample t-tests.

Policy support

Figure 1 displays the mean levels of participants’ policy support for the government’s space programme. It shows that participants who were assigned to read alternative frames were less supportive of space programmes than were those who read only the dry information on the spacecraft docking. Reading alternative frames decreases the mean support level from 4.24 to 3.52. The mean difference between the two groups (0.72) is statistically significant in a t-test (t₉₂ = 4.04, p = .0001). The effect of alternative framing is more evident when the alternative group is compared with the official group. The mean difference between those two groups is 0.86 (t₉₃ = 5.56, p = .0000). This demonstrates that online discussion works in the opposite direction to the government’s intention.

Figure 1.

Average level of support for space programme policy.

Moreover, the exposure to the official frame increases one’s support of government policy, from 4.24 in the control group to 4.38 on average. This difference, however, is relatively small in magnitude and not statistically significant (t₉₁ = .92, p = .39). This indicates that government propaganda has failed to get the support of the audience for the policy that it advocates.

Government performance evaluation

As for the two dimensions of government performance, different patterns are observed. First, in the policy area of ‘technology advancement’, as shown in Figure 2(a), the mean values of the responses of both the official group and the alternative group are higher than that of the control group (control group, 3.48; official group, 3.77; alternative group, 3.69). The mean difference between the control group and the official group is statistically significant at a less strict statistical level of p = 0.10 (t₉₁ = 1.77, p = .08). This implies that the effort to associate the success of spacecraft docking with government leadership makes the participants appreciate, to some extent, the work of the government. The difference between the control group and the alternative group is not statistically significant (t₉₂ = 1.22, p = .23). Negative framing does not make people think less of the government’s achievement in advancing technology. Given the extraordinary progress of technology, as evidenced by the Tiangong and Shenzhou programmes, it is, in fact, not reasonable to expect that Internet users would assess the government less favourably in this area.

Figure 2.

Average level of evaluation of government performance.

Interestingly, in the policy area of ‘well-being improvement’ that is directly related to the alternative framing, as displayed in Figure 2(b), we observe a pattern identical to the pattern of the effect of the alternative framing on specific policy support. The alternative framing of online discussions decreases participants’ evaluation of government performance. Exposure to an alternative frame leads to a lower level of evaluation compared with both the control group (2.27 vs 2.63) and the official group (2.27 vs 2.72). The two pairs of mean differences are statistically significant (t₉₂ = 2.19, p = .031; t₉₃ = 2.73, p = .0077, respectively). The mean difference between the control group and the official group is not statistically significant in a t-test (t₉₁ = .55, p = .58).

This finding indicates that alternative framing works in a direction opposite to the intention of the state. The success of the spacecraft docking is supposed to be a piece of absolutely good news for the government. However, a seemingly unrelated policy area becomes relevant due to online discussion; the alternative framing based on the problems of that policy area decreases one’s subjective evaluation of government performance. The participants do not deny that the government had done a great job in advancing technology, but they also believe that the resources were misplaced and the government should have done a better job in improving the well-being of ordinary citizens.

Diffuse (or affective) support

We did not observe any significant difference between the three groups in regard to diffuse support (Figure 3). This suggests that although online framing can strongly affect one’s support of a specific policy or evaluation of the government’s performance in certain policy areas, it does not necessarily lead to immediate attitudinal change towards the regime as a whole. This finding is consistent with our expectation, as discussed earlier. Diffuse support, as a key dimension of regime support, is shaped mainly by prolonged socio-psychological forces and, thus, is only weakly associated with people’s spontaneous responses to specific policies and the performance of incumbent authority (Chen, 2004; Easton, 1965).

Figure 3.

Average level of diffuse support.

This finding, however, does not necessarily dismiss online framing as an important force in shaping ordinary Chinese web users’ political support. After all, for the participants in the alternative group, the online framing embedded in our experimental study is only a one-time and short-term exposure. Ordinary web users in China are frequently exposed to various alternative framings that can strongly reduce their support of the government’s specific policies and performance. As emphasised in earlier studies (e.g. Li, 2008; Seligson and Muller, 1987), such long-term accumulation of negative attitudes towards a government’s policies and performance can finally spill over to the regime itself.

We further regress each of the attitudinal variables on the two treatment variables (ordered logistic regression), controlling a set of variables (in Table 1). We first include only two dichotomous treatment variables and then control socio-economic status variables, media use variables and pre-existing attitudinal variables. The regression results confirm the pattern in regard to the effect of alternative framing presented so far. Consistently, exposure to online discussion is negatively associated with policy support and evaluation of government performance in well-being or welfare improvement (Models 1–4), and the association is statistically significant. It does not lead to significant changes in general support of the regime (Models 5–6). Official framing, in contrast, demonstrates no significant effect of interest on political support, although those who are guided to read government interpretations hold slightly more positive attitudes towards the government.

Table 1.

Estimation of policy support, evaluation and diffuse support.

	Policy support		Government evaluation (well-being)		Diffuse support
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Online framing	−1.42*** (.37)	−1.73*** (.42)	−.81** (.35)	−1.03*** .39)	.062 (.33)	.076 (.37)
Official framing	.53 (.37)	.56 (.41)	.31 (.36)	.38 (.39)	.23 (.35)	.11 (.37)
Gender		−.25 (.37)		−1.12*** (.37)		−.74** (.36)
Family economy		.28** (.13)		.027 (.12)		.023 (.12)
Family residence		−.43 (.38)		−.53 (.39)		−.16 (.35)
Parents’ education		−.43*** (.16)		.039 (.14)		−.28** (.14)
Frequency of media use		.16 (.14)		.092 (.14)		−.063 (.13)
Frequency of Internet use		.055 (.20)		−.16 (.20)		0.02 (.19)
Political efficacy		.42*** (.16)		.28* (.15)		.26* (.15)
Political interest		.054 (.18)		−.30* (.17)		−.13 (.15)
Democratic values		−.37* (.22)		−.51** (.21)		−1.28*** (.22)
Cut 1–4 (omitted)
Log likelihood	−156.84	−144.05	−165.04	−149.90	−225.04	−197.94
N	141	141	141	141	141	141
Prob > Chi²	.0000	.0000	.014	.0001	.80	.0000
Pseudo R²	.074	.16	.03	.11	.001	.12

Notes: Coding and summary statistics of all variables are reported in Online Appendix 1 (available at: http://ips.sagepub.com)

Model: Ordered logistic regression, analyses of unmatched data. *** p = .01; **p = .05; *p = .10.

Source: Authors’ collection, available upon request.

The effect of the Internet might be conditional upon other variables, such as gender, hometown residence, family economic situation, habits of media use or pre-existing values. Our analyses suggest, however, that none of the interaction terms between the information treatment and these factors is statistically significant. This indicates the significance of the framing effect of the Internet across different socio-demographic groups. That is, regardless of participants’ socio-economic background or pre-existing values, once exposed to Internet framing, they will hold a more negative view towards government policy and performance.

To our knowledge, our study is the first experimental study of Internet effect in an authoritarian environment, particularly in China. We address the robustness of our findings in several ways. First, to correct potential errors caused by imperfect randomisation, we introduce the matching procedures suggested by Imai (2005). We then conduct a mean comparison analysis using the matched data and find that all the results are highly consistent with findings that emerged from the original data.⁴ Moreover, the results that emerged from this experimental study are highly consistent with those based on survey data of a representative sample (Lei, 2011; Tang et al., 2011).

Our results differ from those of other research in that we have identified a causal mechanism through which such results are generated, and submitted that mechanism to an empirical test. Moreover, the existing studies of the Internet in China did not focus directly on political support.

Discussion and conclusion

Given the rapid diffusion of the Internet in authoritarian countries, it is urgent to study its political consequences. In China, a country with half a billion Internet users, in particular, we need to know whether and how the daily use of the Internet reshapes the public’s view towards the authoritarian regime. In this study, we argue that even in an authoritarian environment where Internet use is tightly censored by the state, the diffusion of the Internet can undermine political support for the regime. The Internet is able to do so because it enables the public to interpret issues as they prefer and, thus, to contribute to forming news frames. This democratic nature of the Internet challenges authoritarian rule in that the public is exposed to alternative interpretations of news events and social political issues on an unprecedentedly large scale. The results of this experimental study imply that the diffusion of the Internet erodes the public’s support of the CCP regime by influencing its users’ political views towards directions unfavourable for the party-state. For this reason, we should not regard the Internet as ineffective in making political changes in China just because of the lack of Internet-facilitated popular uprisings there.

We have limited the scope of this research to attitudinal change at the individual level, which is an important subject in itself. But the findings can shed new light on the understanding of the relationship between the Internet and collective actions. Protests are only the end product of prolonged societal changes (Lynch, 2011: 5). For large-scale protests and mass mobilisations to take place, the public needs to be dissatisfied with, alienated from and agitated by the state through a long and gradual process of attitudinal makeover. Although the Internet may at the moment have failed to generate large-scale political uprisings in China, this study suggests that it may still be contributing to that process by alienating citizens from the regime.

Therefore, a study of individual-level attitudinal change in China where uprisings like the Arab Spring have not yet happened helps identify the independent role of the Internet in political changes and thus avoids either overestimation or underestimation of its contributions. On the one hand, the observed role of the Internet in facilitating collective actions like those in the Arab Spring does not necessarily indicate that the new media directly caused any of the outcomes with which they have been associated (Lynch, 2011: 302). Any major political changes could have been facilitated by a large range of factors, like economic crisis, poverty, corruption and external pressure. For this reason, even with the apparent role played by the Internet in the Arab Spring, many are still circumspect about its real contribution (Farrell, 2012: 45). On the other hand, as suggested earlier, the lack of large-scale uprisings does not mean the absence of the Internet effect. The contribution of the Internet does not necessarily happen at the moment of uprising; it precedes it, with the makeover of the underlying attitudinal basis, and thus constitutes a necessary, although not sufficient, condition for political changes. Without examining the underlying attitudinal changes, we cannot explain why the public ‘suddenly’ change their side to support the opposition movement and how the apparently stable regime lost public support before the crisis.

Given the ostensibly high level of general political support in China, we do not suggest that the legitimacy of the Chinese state is in imminent danger or that the Internet framing will definitely lead to uprisings. The relationship between attitudinal transformation and political actions is nothing but natural or definitive. A body of social movement literature suggests that there is a wide gap between critical attitudes and protest activity. For attitudinal changes to finally turn into political actions, the Internet needs to be further spread to generate a critical mass and increase the cost of repression for the state, which in turn induces regime concessions and encourages collective actions. The anti-system sentiment on the Internet has not led to political actions in China partly because alternative information and, especially, alternative framing on the Internet has not reached a significant portion of the population due to a digital divide and partly because the state has been able to effectively identify and repress online calls for action through a cyber-police system (King et al., 2013). But the ability of the state cannot be taken for granted. With the ever-increasing speed of assimilation of the Internet and mounting dissatisfaction among the public facilitated by the online alternative framing, it becomes more difficult, if not impossible, for the state to censor all online activities or forestall Internet-facilitated collective actions.

We have been careful in addressing the external validity issue. But the findings of this study are not definitive, given that we conducted only one experiment in one place at one time. We do not intend our findings to be boldly generalised. Further studies are needed to investigate the political effect of the Internet among other segments of the population. Less educated people, for instance, are less exposed to Internet information or discussion. But with fast assimilation of digital technology, more and more people with a lower education or in rural places will begin to gain access to the Internet in China. It is important to know how the alternative information influences the political views of these people, who are presumably more prone to the influence of external information. In contrast, some other people, like young professionals, lawyers and writers, are more politically active in reading alternative framing and information on the Internet than college students. More importantly, these active users are usually the ones who frame social political issues by making comments, publicising opinions and initiating debate. Studies are needed to explore the evolution of cyberspace discussion of social and political issues and the role of these active users in that process. Lastly, in addition to experiment and survey-based quantitative methods, a comprehensive understanding of the Internet also requires qualitative research. Qualitative methods are especially useful to studying the process of frame building and the development of online discourse. In any event, the study of the Internet effect should aim to uncover the more specific mechanisms under authoritarian rule, instead of just focusing on political uprisings at the dawn of regime demise.

Footnotes

Appendix 1

Table A1.

Coding and summary statistics.

Variable	Coding	Obs.	Mean	Std Dev.	Min.	Max.
Support for space programme	1 = strongly oppose to 5 = strongly support	141	4.04	0.89	2	5
Government performance evaluation: technology development	1 = very bad to 5 = very well	141	3.65	0.82	1	5
Government performance evaluation: social welfare provision	1 = very bad to 5 = very well	141	2.54	0.82	1	4
Diffuse support	0 = lowest to 5 = highest	141	3.05	0.67	2	4
Gender	0 = female; 1 = male	141	0.55	0.50	0	1
Family residence	0 = rural; 1 = urban	141	0.55	0.50	0	1
Parents’ highest education level	1 = no formal education to 7 = doctoral	141	4.09	1.33	1	7
Family economic situation	1 = worst to 10 = best	141	5.33	1.57	1	9
Democratic attitudes	Average of the responses to statements:	141	2.17	0.81	1	4
	1. There will be chaos if there are several political parties competing.
	2. Political leaders are like the head of a family; we should obey their decisions.
	3. Political gathering and protest will cause instability and should be banned
	Recoded: 1 = lowest level to 5 = highest level
Political efficacy	People like me can influence government policy	141	2.58	1.15	1	5
	1 = strongly disagree to 5 = strongly agree
Political interest	1 = completely no interest to 5 = very interested	141	3.50	1.03	1	5
Frequency of media use	1 = rarely use to 5 = daily use	141	2.94	1.29	1	5
Frequency of Internet use	1 = rarely use to 6 = use more than three hours a day	141	5.06	0.94	3	6

Acknowledgements

The authors would like to thank Dean Lacy, Lianjiang Li, Yuan Ting, Dwayne Woods and Shiping Tang for their comments.

Funding

This research was supported by National Natural Science Foundation of China (No. 71203130) and Shanghai Pujiang Program (No. 12PJC054).

Notes

Author biographies

Min Tang’s research focuses on comparative democratisation and political communication and his recent publications appear in Democratization and Journal of Chinese Political Science.

Narisong Huhe’s research interests include Chinese politics, particularly political attitudes and grassroots governance. His recent work appears in Social Science Quarterly.

References

Benkler

Yochai

(2006) The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven, CT: Yale University Press.

Best

Michael L.

Wade

Keegan W.

(2009) The Internet and democracy. Bulletin of Science, Technology & Society 29(4): 255–271.

Boas

Taylor C.

(2006) Weaving the authoritarian web: The control of Internet use in nondemocratic regimes. In: Zysman

John

Newman

Abraham

(eds) How Revolutionary Was the Digital Revolution? National Responses, Market Transitions, and Global Technology. Stanford, CA: Stanford Business Books.

Bueno de Mesquita

Bruce

Downs

George

(2005) Development and democracy. Foreign Affairs 84(5): 77–86.

Chang

Tsan-Kuo

Wang

Jian

Chen

Chih-Hsien

(1998) The social construction of international imagery in the post-Cold War era: A comparative analysis of U.S. and Chinese national TV news. Journal of Broadcasting & Electronic Media 42(3): 277–296.

Chase

Michael

Mulvenon

James

(2002) You’ve Got Dissent! Chinese Dissident Use of the Internet and Beijing’s Counter-Strategies. Santa Monica, CA: Rand.

Chen

Jie

(2004) Popular Political Support in Urban China. Stanford, CA: Stanford University Press.

Chowdhury

Mridul

(2008) The role of the Internet in Burma’s Saffron Revolution. Berkman Center for Internet and Society, September, 14.

CNNIC (China Internet Network Information Center) (2012) Statistical Report on Internet Development in China. Available at: http://www1.cnnic.cn/IDR/ReportDownloads/201302/P020130312536825920279.pdf

10.

Coleman

Stephan

Hall

Nicola

(2001) Spinning on the Web: E-campaigning and beyond. In: Coleman

Stephan

(ed.) Cyber Space Odyssey: The Internet in the UK Election. Hansard Society. Available at: www.hansardsociety.org.uk

11.

Diamond

Larry

(2010) Liberation technology. Journal of Democracy 21(3): 69–83.

12.

Easton

David

(1965) A Systems Analysis of Political Life. New York, NY: John Wiley and Sons.

13.

Easton

David

(1975) A re-assessment of the concept of political support. British Journal of Political Science 5(4): 435–457.

14.

Entman

Robert M.

(1993) Framing: Toward clarification of a fractured paradigm. Journal of Communication 43(4): 51–58.

15.

Farrell

Henry

(2012) The consequences of the Internet for politics. Annual Review of Political Science 15: 35–52.

16.

Goffman

Erving

(1974) Frame Analysis: An Essay on the Organization of Experience. New York, NY: Harper & Row.

17.

Hachigian

Nina

(2001) China’s cyber-strategy. Foreign Affairs 80(2): 118–133.

18.

Harwit

Eric

Clark

Duncan

(2001) Shaping the Internet in China: Evolution of political control over network infrastructure and content. Asian Survey 41(3): 377–408.

19.

Hassid

Jonathan

(2008) Controlling the Chinese media: An uncertain business. Asian Survey 48(3): 414–430.

20.

Herold

David

Marolt

Peter

(2011) Online Society in China: Creating, Celebrating and Instrumentalising the Online Carnival. Abingdon: Routledge.

21.

Hill

David T.

Sen

Krishna

(2000) The Internet in Indonesia’s new democracy. Democratization 7(1): 119–136.

22.

Howard

Philip

(2010) The Digital Origins of Dictatorship and Democracy: Information Technology and Political Islam. New York, NY: Oxford University Press.

23.

Howard

Philip

Muzammil

Hussain

(2011) The role of digital media. Journal of Democracy 22(3): 35–48.

24.

Hung

Chin-Fu

(2010) China’s propaganda in the information age: Internet commentators and Wen’an Incident. Issue and Studies 46(4): 149–181.

25.

Imai

Kosuke

(2005) Do get-out-the-vote calls reduce turnout? The importance of statistical methods for field experiments. American Political Science Review 99(2): 283–300.

26.

Iyengar

Shanto

(1991) Is Anyone Responsible? How Television Frames Political Issues. Chicago, IL: University of Chicago Press.

27.

Kalathil

Shanthi

Boas

Taylor C.

(2001) The Internet and state control in authoritarian regimes: China, Cuba and the counterrevolution. First Monday 6(8). Available at: http://firstmonday.org/ojs/index.php/fm/article/view/876

28.

Kalathil

Shanthi

Boas

Taylor C.

(2003) Open Networks, Closed Regimes: The Impact of the Internet on Authoritarian Rule. Washington, DC: Carnegie Endowment for International Peace.

29.

Kennedy

John James

(2009) Maintaining popular support for the Chinese Communist Party: The influence of education and the state-controlled media. Political Studies 57(3): 517–536.

30.

Kinder

Donald

Sanders

Lynn

(1990) Mimicking political debate with survey questions: The case of white opinion on affirmative action for blacks. Social Cognition 8(1): 73–103.

31.

King

Gary

Pan

Jennifer

Roberts

Margaret

(2013) How censorship in China allows government criticism but silences collective expression. American Political Science Review 107(2): 326–343.

32.

Kopacz

Maria

(2008) Framing immigration online: Online position statements of 2006 candidates for Congress. The Electronic Journal of Communication 18(1). Available at: http://www.cios.org/EJCPUBLIC/018/1/01813.HTML

33.

Lecheler

Sophie

de Vreese

Claes H.

(2011) Getting real: The duration of framing effects. Journal of Communication 61(5): 959–983.

34.

Lei

Ya-Wen

(2011) The political consequences of the rise of the Internet: Political beliefs and practices of Chinese netizens. Political Communication 28(3): 291–322.

35.

Lianjiang

(2008) Political trust and petitioning in the Chinese countryside. Comparative Politics 40(2): 209–226.

36.

Liu

Fengshu

(2011) Urban Youth in China: Modernity, the Internet and the Self. New York, NY, and London: Routledge.

37.

Lotan

Gilad

Graeff

Erhardt

Ananny

Mike

. (2011) The revolutions were tweeted: Information flows during the 2011 Tunisian and Egyptian revolutions. International Journal of Communication 5: 1375–1405.

38.

Lowrey

Wilson

Anderson

William

(2005) The journalist behind the curtain: Participatory functions on the Internet and their impact on perceptions of the work of journalism. Journal of Computer-Mediated Communication 10(3). Available at: http://jcmc.indiana.edu/vol10/issue3/lowrey.html

39.

Luther

Catherine A.

Zhou

Xiang

(2005) Within the boundaries of politics: News framing of Sars in China and the United States. Journalism & Mass Communication Quarterly 82(4): 857–872.

40.

Lynch

Marc

(2011) After Egypt: The limits and promise of online challenges to the authoritarian Arab state. Perspectives on Politics 9(2): 301–310.

41.

McFaul

Michael

(2005) Transitions from postcommunism. Journal of Democracy 16(3): 5–19.

42.

Mercer

Claire

(2004) Engineering civil society: ICT in Tanzania. Review of African Political Economy 31(99): 49–64.

43.

Mitchelstein

Eugenia

Boczkowski

Pablo J.

(2009) Between tradition and change. Journalism 10(5): 562–586.

44.

Morozov

Evgeny

(2011) Whither Internet control? Journal of Democracy 22(2): 62–74.

45.

Murphy

Emma C.

(2009) Theorizing ICTs in the Arab world: Informational capitalism and the public sphere. International Studies Quarterly 53(4): 1131–1153.

46.

Nelson

Thomas E.

Clawson

Rosalee A.

Oxley

Zoe M.

(1997) Media framing of a civil liberties conflict and its effect on tolerance. The American Political Science Review 91(3): 567–583.

47.

Norris

Pippa

(ed.) (1999) Critical Citizens: Global Support for Democratic Government. Oxford and New York: Oxford University Press.

48.

Pan

Zhongdang

Lee

Chin-Chuan

Chan

Joseph Man

. (1999) One event, three stories. International Communication Gazette 61(2): 99–112.

49.

Rheingold

Howard

(2002) Smart Mobs: The Next Social Revolution. Cambridge, MA: Perseus Publishing.

50.

Rodan

Gary

(1998) The Internet and political control in Singapore. Political Science Quarterly 113(1): 63–89.

51.

Seligson

Mitchell

Muller

Edward N.

(1987) Democratic stability and economic crisis: Costa Rica, 1978–1983. International Studies Quarterly 31: 301–326.

52.

Shirk

Susan

(2010) Changing Media, Changing China. Oxford: Oxford University Press.

53.

Shirky

Clay

(2009) Here Comes Everybody! The Power of Organizing Without Organizations. New York, NY: Penguin.

54.

Shirky

Clay

(2011) The political power of social media. Foreign Affairs 90(1): 28–41.

55.

Slothuus

Rune

(2008) More than weighting cognitive importance: A dual process model of issue framing effects. Political Psychology 29(1): 1–28.

56.

Song

Yonghoi

(2007) Internet news media and issue development: A case study on the roles of independent online news services as agenda-builders for anti-US protests in South Korea. New Media & Society 9(1): 71–92.

57.

Stockmann

Daniela

(2011) Race to the bottom: Media marketization and increasing negativity toward the United States in China. Political Communication 28(3): 268–290.

58.

Stockmann

Daniela

Gallagher

Mary E.

(2011) Remote control: How the media sustains authoritarian rule in China. Comparative Political Studies 44(4): 436–467.

59.

Sullivan

Jonathan

(2012) A tale of two microblogs in China. Media, Culture & Society 34(6): 773–783.

60.

Tang

Lijun

Yang

Peidong

(2011) Symbolic power and the Internet: The power of a ‘horse’. Media, Culture & Society 33(5): 675–691.

61.

Tang

Min

Jorba

Laia

Jensen

Michael J.

(2011) Digital media and political attitudes in China. In: Anduiza

Eva

Jensen

Michael

Jorba

Laia

(eds) Digital Media and Political Engagement Worldwide: A Comparative Study. Cambridge: Cambridge University Press, pp. 240–253.

62.

Tang

Wenfang

(2005) Public Opinion and Political Change in China. Stanford, CA: Stanford University Press.

63.

Tang

Wenfang

Iyengar

Shanto

(2011) The emerging media system in China: Implications for regime change. Political Communication 28(3): 263–267.

64.

Taubman

Geoffry

(1998) A not-so World Wide Web: The Internet, China, and the challenges to nondemocratic rule. Political Communication 15(2): 255–272.

65.

Thomas

Pradip

Nain

Zaharom

(2005) Who Owns the Media: Global Trends and Local Resistance. Penang, Malaysia: Southbound.

66.

Tufekci

Zeynep

Wilson

Christopher

(2012) Social media and the decision to participate in political protest: Observations from Tahrir Square. Journal of Communication 62(2): 363–379.

67.

Xenos

Michael A.

Foot

Kirsten A.

(2005) Politics as usual, or politics unusual? Position taking and dialogue on campaign websites in the 2002 U.S. Elections. Journal of Communication 55(1): 169–185.

68.

Yang

Guobin

(2003) The co-evolution of the Internet and civil society in China. Asian Survey 43(3): 405–422.

69.

Yang

Guobin

(2009) The Power of the Internet in China: Citizen Activism Online. New York, NY: Columbia University Press.

70.

Yang

Jin

(2003) Framing the NATO air strikes on Kosovo across countries. Gazette 65(3): 231–249.

71.

Yang

Qing

Tang

Wenfang

(2010) Exploring the sources of institutional trust in China: Culture, mobilization, or performance? Asian Politics & Policy 2(3): 415–436.

72.

Zheng

Yongnian

Guoguang

(2005) Information technology, public space, and collective action in China. Comparative Political Studies 38(5): 507–536.

73.

Zhou

Yuqiong

Moy

Patricia

(2007) Parsing framing processes: The interplay between online public opinion and media coverage. Journal of Communication 57(1): 79–98.

74.

Zhou

Xiang

(2009) The political blogosphere in China: A content analysis of the blogs regarding the dismissal of Shanghai leader Chen Liangyu. New Media & Society 11(6): 1003–1022.

75.

Zillmann

Dolf

Chen

Lei

Knobloch

Silvia

. (2004) Effects of lead framing on selective exposure to Internet news reports. Communication Research 31(1): 58–81.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB