Master’s Thesis
Malaysia’s “social media” Election 2013:
Are voters becoming less cynical?
Nurul Huda Jamaluddin 10583440
Graduate School of Communication Universiti of Amsterdam
Erasmus Mundus Master’s in Journalism, Media and Globalisation Jonas Lefevere
27 JUNE 2014
Abstract
The emergence of new media in Malaysia’s “first social media election” last year has brought about questions of social media effects via electoral discussions on citizens’ political behaviour and attitude. Extant literature had shown decrease in cynicism due to social media use. Following this, an online experimental survey was carried out to test if the exposure to electoral discussions on social media has any impact in decreasing political cynicism. Participants were exposed to either control or one of three treatment groups of social media discussion with different partisan slants. However, the results showed that the exposure to social media discussions did not contribute to the decrease in political cynicism but rather increases them. The findings also indicate that heavy social media users were more prone to be cynical. Other interesting finding includes strong influence of partisanship on voter’s cynicism, especially when exposed to dissonant information. Despite its limited sample size, this study provides future directions for related studies and has vital contribution in the realm of social media effects on citizenry.
Keywords: Social Media, General Election-13, Malaysian, Political Cynicism, electoral discussion
2
Acknowledgement
“Learn from yesterday, live for today, hope for tomorrow. The important thing is not to stop questioning.” – Albert Einstein
This is the main takeaway of this rather hectic and strenuous three-month thesis writing. Being a former journalist and from way back from journalism school, I was taught to question everything and that “assumption is the mother of all disasters”. Throughout this whole process, I realize that this is a fitting principle in research too. You have to ask yourself at every step and question everything in order to construct your arguments. You have to get your answers through a rigorous mental exercise – of theory and application – until it makes sense to you and your potential readers. After all the sweat and tears, I could not be more proud that this thesis is reaching its completion. It is like building a house, you layered brick-by-brick and seeing end product is like seeing the house you built coming to life.
Still, I humbly realize that this Master’s thesis would not possible without the tremendous support that I got throughout writing this academic piece. So, first of all, I would like to thank God and all His mercy. To mom and dad, who both have been my pillar of strength and whose presence grounds me. And although they sometimes couldn’t stand this stubborn daughter, they stand by my side anyway. Special thank you to Jonas Lefevere, who has been really patient with my errors and delayed emails (you have been a wonderful thesis supervisor and I couldn’t be more grateful for the support!). Finally, I am thankful for the help from family and friends, especially Andrew Demetriou who has been a great help.
3 Table of Contents List of Tables………..………….4 List of Figures………...5 Introduction……….……….…....6 Theoretical Background………..………...11
Defining political cynicism………11
Traditional versus social media and cynicism………..……….13
Case Description………17
The waning power of the ruling alliance………...17
The paradox of Malaysian media………..18
The rise of social media……….18
Methods………...19
Results………..…..25
Discussion and Conclusion………....31
References………..…38
4
List of Tables
Table 1. Descriptive Statistics for Political Cynicism by groups………..26
Table 2. Descriptive Statistics for Political Cynicism between
Control and Treatment groups...27
Table 3. Regression Statistics for Political Cynicism between
Groups and Social Media Usage………...28 Table 4. Summary of Regression Analysis for Political Cynicism Scores………29 Table 5. Mean Level of Political Cynicism by Groups and Vote Choice………..30
5
List of Figures
Figure A. News Article………..…………22
Figure B. Control Condition (Example)………..…..22
Figure C. Balanced Condition (Example)………..23
Figure 1. Distribution across control and treatment groups………...26
6
Introduction
In the recent Malaysian General Election of 2013 (GE13), social media like Facebook and Twitter became the major battleground between the ruling alliance National Front (BN) and the opposition People’s Alliance (PR) in what was seen as the closest election since the country’s Independence in 1957 (“Malaysia’s Social Media Election”, 2013). In the election aptly called as the country’s “first social media election” by Prime Minister and BN leader Najib Razak(Zahiid, 2013), social media has played an important part in the increase of voter’s turnout from 76.0% to 84.8%, the highest in Malaysian electoral history (Gomez, 2013). However, despite reports by several national and international mediathat touched on the crucial role of social media in this election, there is limited scholarly knowledge on its influence on citizen’s political behaviour and attitudes.
Nevertheless, in a country with tight control over mainstream media such as Malaysia (Sani, 2005), it is certainly not surprising that social media took centre stage in GE13, especially in electoral discussions. In their case study of Malaysia, Smeltzer and Keddy (2010) have
predicted the potential of Facebook as a political tool in the country, especially with its restrictive media environment. Moreover, citing Nisbet, Stoycheff and Pearce’s(2012) multinational study in 28 countries, Howard and Parks (2012, p. 360) concluded that social media may have pertinent roles in bringing social change in countries that have “pent-up demand for democratic governance but relatively few freedoms” such as Malaysia.
In addition, in countries with a restricted media environment, people try to seek
alternative forms of media to state their opinions and engage with political issues (Raoof, Zaman, Ahmad & Al-Qaraghuli, 2003). This is the case in Malaysia, as political analysts lamented that more Malaysians are turning to the Internet for information, which led to politicians to setting up
7
Facebook and Twitter accounts to reach out to voters (Zahiid, 2013). Moreover, research by
Mou, Atkin and Fu (2011) onChinese online political participation saw that due to highly censored mainstream media, “politically sensitive” content is more likely to spread online, prompting citizens to go online instead. Lower trust in mainstream media has not really shown any influence on social media trust per se but the study did suggests that “distrust toward the traditional media in a censored environment could raise trust toward alternative sources” (Mou et al., 2011, p.346). This suggests that citizens in countries with restricted press freedom (like Malaysia), might be more open to seek political information in social media.
Besides stringent control over mainstream media, the popularity of social media in GE-13 can be attributed to the ‘relaxed’ Internet regulation, as Malaysia has a ‘no censorship’ guarantee of the Internet through policies such as Section 3(3) of the Communication and Multimedia Act (CMA) and the Multimedia Bill of Guarantees (Freedomhouse.org, 2012), in its effort to promote Malaysia as a regional hub for information technology investment (George, 2007, p. 900). Online media allows interested voters to obtain critical and relevant political information including information on opposition candidates and dissenting opinions that could influence their voting decision (Willnat, Wong, Tamam, & Aw, 2013).
Evidently, the burgeoning influence of social media in electoral discourse may have an impact on the election outcome. However, with its ever increasing usage, it is certainly not far-fetched to argue that its popularity also has a significant influence on the electorates’ behaviour and attitudes. While scholars have observed that traditional media consumption causes people to be more politically cynical (Cappella & Jamieson, 1996; Adriaansen, van Praag & de Vreese, 2010), several researchers have found that Internet and social media usage contributed to decreasing political cynicism. For example, Hanson, Haridakis, Cunningham, Sharma and
8
Ponder (2010, p.599) found that “the use of the medium of social networking sites (SNS) predicted a lower level of cynicism”, contrary to findings of early media research. Although the study suggests a correlation between the use of SNS and self-reported voter cynicism, it failed to demonstrate a causal relationship between social media and political cynicism.
Even though Hanson’s study exemplifies the impact of social media usage on political cynicism rather than exposure, several studies have looked at how news media exposure (in mainstream media) can affect political behaviour and attitude positively (de Vreese and Boomgarden, 2006). In light of this,this study aims to examine whether exposure to electoral discussions in social media influences political cynicism. As this will be an experimental study, the research is relevant because it will be able to showcausality between social media and political cynicism, which is lacking in the current scholarly pieces.
Besides that,the findings could shed light on the impact of social media exposure on political discourse and democratic functioning. A study by de Vreese (2005, p.283) suggest that cynicism is not always a “bad” thing, that it may in fact be an indication of “an interested and critical citizenry”. In relations to that, this study could also contribute to future studies on social media’s role and influence in between elections and its political impact in the upcoming
Malaysian general election (‘Malaysia’s social media election is already over - James Gomez’, 2013). Furthermore, “long-term electoral implications of social networking tools remain unclear in Malaysia” (Weiss, 2014, p. 101), which warrants further investigation. The research could be expanded by scholars to further understand the contribution of digital media, particularly social media to democratic expectations (citizens’ expectations of a democratic government) in less democratic countries (see Howard & Parks, 2012).
9
In addition, several studies including the mentioned studies have seen the potential of social media as a political tool in Malaysia (see Smeltzer & Keddy, 2010; Raoof et al., 2003), but these studies as mentioned have not really demonstrated causality or shown the impact of social media exposure on political behaviour and attitude, and none particularly on political cynicism, with regards to the Malaysian case.To fill this void, this study will examine whether exposure to electoral discussions in social media contributes to the decrease of political cynicism. Based on the research cited above, the first research question is as follows: RQ1: How does exposure to
electoral discussions in social media influence political cynicism among Malaysian voters?
The expectations are tested through an experimental design, which manipulates content of electoral discussions on Facebook to test whether the presence or absence of electoral
discussion will have any impact on the decrease of political cynicism. Based on the selective exposure theory, the content that will be manipulated are the electoral discussions, specifically the comments on a news article about the Malaysia’s general election posted on the mentioned social media pages. There is very little research that ties selective exposure and political
cynicism. However, in the realm of selective exposure, Sunstein (2001) has warned that we are witnessing an explosion of ideologically motivated selectivity in exposure to news and political information precipitated by Internet technologies especially exposure to consonant views as he fears it will result to less tolerance and more extreme views. It was found that citizens with higher political knowledge and stronger partisan attachments are more likely to select like-minded news (Stroud, 2008) and hold more extreme attitudes (Meffert et al., 2006).
Furthermore, the most notable scholars in selective exposure research, Bennett and Iyengar concluded in 2008 that as media users devolve into smaller, like-minded subsets will “rarely find themselves in the path of attitude-discrepant information” (p. 724). In the context of
10
election, de Vreese (2005) found cynicism to have a negligible impact on citizen participation. However, in studies such as one by Agger et al. (1961), partisanship has been found to be an important variable in determining political cynicism. People nowadays can easily select media outlets sharing their political predispositions; a phenomenon by Stroud (2008) called partisan
selective exposure. She found strong evidences that partisan selective exposure leads to
polarization with cross-sectional results and via over-time analyses.
Nonetheless, if and when citizens are exposed to counter attitudinal messages, their acceptance of the message is dependent on their partisan and ideological orientation (Zaller, 1993). Therefore, in the situation where a particular subject was forced exposure, their acceptance of the message they are exposed to could be affected partisanship. Looking at Zaller’s theory that explained how enlightened citizens most likely to receive change-inducing messages, it is worth to ask, (RQ2): How does partisan slant in electoral discussions in social
media influence political cynicism among Malaysian voters?
It is important to note that although this study intends to investigate whether partisan
selective exposure (PSE) has any influence on political cynicism; it is not an extensive study that
uses several partisan media to measure the said variable. Instead, the study used a pro-government article from ABNnews, which is an online news portal for ABNxcess cable TV, a fairly new media group in Malaysia with pro-government political leanings. Besides that, as I am looking at partisan selective exposure, for the purpose of this study, there will be one control group (posted news article without comments) and three treatment groups. This design considered the divide of political leanings of Malaysian citizens, where people are either pro-government, the ruling alliance/incumbent Barisan Nasional or pro-opposition, opposition alliance Pakatan Rakyat. The treatment groups are: are pro-government (BN), pro-opposition
11
(PR) and balanced comments (which included both pro-government and pro-opposition comments).
Theoretical Background
The experimental study presented to measure political cynicism while also looking at its relationship with partisan selective exposure theory. This section presents a brief introduction to the concept of political cynicism, which will be quite general, with more exposition of particular aspects of the theory as they become important to the discussion. It will also briefly look at how selective exposure theory influences political cynicism. Then, based on the extant literature, we will draw hypothesis from the research questions and arguments.
Defining political cynicism
Political cynicism is recognized as an important political sentiment with serious implications for democracy (Goldfarb, 1991). Still, the modern definitions of cynicism and political cynicism are varied. Political cynicism has been defined as a distrust of the political system (Dennis & Webster, 1975) while other definitions of political cynicism include: “mistrust generalized from particular leader or political groups to the political process as a whole”
(Cappella and Jamieson, 1997, p.166) and the degree of negative affect towards the government (Miller, 1974).Besides multiple definitions, the concept has been interchangeably used like disaffection (Austin & Pinkleton, 1995).
However, some scholars did try to explicitly distinguish cynicism, for example Cappella and Jamieson (1997), whom differentiate “unhealthy cynicism”, which they believe to be created by media, and a healthy scepticism of government and Eisinger (2000) who noted that cynicism is “more than mild distrust”. For the purpose this study, I will take on the definition by Schyns,
12
Nuus and Dekker (2004, p. 3), which after empirical and conceptual consideration defined political cynicism as “an individual’s attitude, consisting of a deep-rooted conviction of the inherent evilness of politicians, political institutions, and/or the political system as a whole”. Political cynicism, according to their concept comprised of three major “ingredients”: subject, orientation and object. Their concept suited this study, not only because it is more explicitly defined, this definition also considers cynicism as an attitude, how individuals think and feel about politics (p.2). Arguably, the root of political cynicism is the lack of trust, a cognitive and affective component. Besides that, the definition by Schyns et al. (2004) also entails the object of political cynicism that is not just focused on politicians but also political institutions and political system (e.g., democracy). This is lacking in some definitions, such as Dennis and Webster’s (1975) that only focus on the political system.
The common consensus among scholars in political communication research is that political cynicism has a negative impact on democracy (see Goldfarb, 1991; Eisinger, 2000). One of the reasons is because cynics fundamentally believe that political leaders and the groups that they represent cannot be trusted even in the absence of evidence, either pro or con (Cappella, 2002). Some researchers concluded that cynical individuals are less likely to participate in politics as they distance themselves from the political process (e.g., Pinkleton & Austin, 2002).
Research has looked into several factors influencing political cynicism. For example, Agger et al. (1961) look into how partisanship, social class, age, and personality influence
cynicism. Their study found that Democrats are slightly less trusting and more neutral than either
Republicans or Independents and that people who are from lower class background and lower education group retain presumably high level of political cynicism than their lower class parents. They also found that people who are highly educated (education being the index of social class,
13
rather than income) are much more politically trusting than the least educated and lastly, aging are discovered to produce more political cynicism.
Traditional versus social media and cynicism
Recent studies have shown that different media influence political cynicism differently. Before the “explosion” of social media, research on political cynicism has revolved around media coverage particularly on strategic frame in mainstream news, where candidates are portrayed as “self-interested actors whose issue position are politically expedient, not motivated by a sincere desire to identify and solve important societal problems” (Valentino, Buhr and Beckmann, 2001, p.93). Cappella and Jamieson (1996) have demonstrated that strategic, manipulative news frames are more likely to create and enhance political cynicism than issue focused frames.
There is no doubt that in strategic news and political advertising research, framing is an essential concept in relation to political cynicism. Valentino et al. (2001) research on strategic news coverage has shown that strategic frames produced more negative reactions than sincere frames while Adriaansen et al. (2010) did not find similar effects when comparing the impact of strategic and substantive news. Without contention, strategic news has a significant impact on political cynicism, as past research has shown that it reinforces participants’ strategic
interpretations of an election campaign, resulting in cynicism (Rhee, 1997). The prevalence of
strategic framing in election news is worrisome, as Patterson (1993, p.52) concluded: “election
news, rather than serving to bring candidates and voters together, drives a wedge between them” (see Schuck, Boomgarden & de Vreese, 2003).
Nevertheless, recent studies have shown that different media influence political cynicism differently. While studies on traditional media mentioned above indicate a positive correlation
14
between mainstream news content and political cynicism - especially strategic news coverage (e.g.: Cappella & Jamieson, 1996; Adriaansen et al., 2010; Shuck, Boomgarden & de Vreese, 2003) and negative political advertising (Pinkleton, Um & Austin, 2002), studies on general
Internet and social media usage has proven the opposite. In Lin and Lim’s (2012) study on
young Korean voters that investigates the relationships of their media behaviour (use of traditional media and the Internet) to political cynicism and political efficacy, they show evidence that the usage of Internet for political information reduces cynicism.
More importantly, research on social media usage in election campaigns has made voters less cynical (see Hanson et al., 2010; Lariscy et al., 2011). For example, Hanson et al. (2010, p.601), found that the use of social networking sites (SNS) predicted lower level of cynicism. Their findings do not demonstrate a causal relationship too but they do suggest that the use of SNS and political blogs correlated to lower levels of self-reported voter cynicism. The research also illustrates that the newer media use and effects is perhaps similar to how audiences use traditional media. However, they found that media use is not the only factor that affects cynicism. Instead, the most significant predictors of cynicism may reside in the individual, namely background characteristics and motives for seeking political information.
Besides that, in recent years, the impact of social media on political behaviour and attitudes has been receiving more attention in the scholarly community. Social networking sites (SNS) like Facebook have been used for various political purposes as through the medium , the site’s members can support a political candidate and show their stance on a particular issue - as well as “participate and engage in the political process freely” (Westling, 2007). By the end of 2013, Facebook, which is undoubtedly the most popular social network site worldwide, has 1.23 billion monthly active users worldwide (Sedghi, 2014). Based on the nature of this research, we
15
will take on Howard and Parks’ (2012, p. 359) definition of social media, defining it as “the content that takes the digital form of personal messages, news, ideas, that becomes cultural products”.
At any rate, the research on exposure (rather than usage) to social media content on political cynicism is scarce. Although it might be true that the latter (usage) has been linked to decreases in political cynicism, past studies on the “predecessor” of social media, online public forums, have looked at the phenomenon of selective exposure to like-minded opinions in this forums (Choi, 2014) while a previous study on the use of imagery connected with written and spoken reports did establish a relationship between frames and selective exposure (Tankard, 2001).This study could find a missing link: if social media usage engenders decreases in political cynicism, mere exposure to social media discussions could very well produce similar results. Based on the extant literature and case description below, I expect that (H1): Exposure to
electoral discussions in social media decreases political cynicism among Malaysian voters.
However, exposure to social media discussions alone is perhaps not strong cause to influence one’s cynical attitude towards politics. As extant literature mentioned indicated that social media usage decreases political cynicism, we can also infer that that (H2): Exposure to
electoral discussion in social media decreases political cynicism among Malaysian voters, especially if they are heavy social media users.
Messing and Westwood (2012) suggest that social media should be expected to increase users’ exposure to a variety of news and politically diverse information - through “socialization of Internet news”. This illustrates that mainstream media and social media differ in terms of how audiences are exposed to news content. Political observers and scholars like Johnson, Braima, & Sothirajah (1999) argued that different from mainstream coverage (e.g.: TV), the Web provides a
16
large amount of substantive information related to candidates’ stances on issues, which will transform the campaign process besides increasing civic engagement (Bucy, D'Angelo, & Newhagen, 1999). This in turn, may result in the reduction of individuals’ cynicism (see Lin & Lim, 2012).
Research has shown that personally relevant beliefs are more likely to influence exposure decisions including political partisanship (Donsbach, 1991). However, Messing and Westwood (2012, p.4) demonstrate that that people rarely encounter counter-attitudinal content in the media and this is especially less severe for online media. On the one hand exposure to mainstream media news content increased political cynicism and on the other, audience is also exposed to this content on social media. Underlying selective exposure theory is the idea of cognitive
dissonance, put forward by the theory’s proponent Leon Festinger (1962) who assumes that
selective exposure prevents dissonance. Basically, he suggests that people has a predisposition hold their attitudes and beliefs in harmony and avoid disharmony (or dissonance).
Donsbach (1991) lamented that “selectivity according to one’s political predisposition is only practiced, if at all, to a very limited extent towards negative information might indicate that the motivation of readers to keep up with negative developments may mask the role of cognitive
dissonance as a selectivity factor” (p. 169). On the flipside, to illustrate how partisanship is an
important predictor of cynicism, some observers have suggested that after eight years of Republican control of the White House, Democrats are more cynical about politics and
politicians (Agger et al., 1961). Partisanship can manifests in political discussions as people tend to hear and see information favourable to their predispositions, thus they are more likely to hear and see congenial information rather than neutral resources (Berelson & Steiner, 1964).
17
Levendusky (2012) explained how slanted presentation of the news on partisan outlets leads viewers to perceive other party more negatively, to trust them less and to be less supportive of bipartisanship. His findings demonstrates that when people watch like-minded media, it made viewers feel more negatively toward the other party and rate them less positively along various dimensions. In short, messages that clashes with one’ beliefs may have an adverse effect and make them more cynical, when they are encounter or exposed to them (messages).
Hence, it can be argued that partisan selective exposure does have a significant effect on political cynicism. Still, partisanship is an overarching concept, which goes beyond media outlets and one’s political leaning, it could also exist in partisan slant in electoral discussion. It is
therefore can be inferred that (H3): Partisanship moderates the impact of exposure to electoral
discussions in social media on political cynicism amongst Malaysian voters. Exposure to dissonant electoral discussion increases cynicism, and exposure to consonant electoral discussion decreases cynicism.
Case description
The waning power of the ruling alliance
Malaysia gained Independence from Britain in 1957 and became Malaysia in 1963. It has a multi-party system, with two major coalitions: the ruling coalition National Front (Barisan
Nasional or BN) and the opposition coalition People’s Alliance (Pakatan Rakyat or PR). The
ruling alliance BN, has been ruling the country since Independence (Liew, 2012). The alliance is made up by 13 parties, led by United Malays National Organization (UMNO). Meanwhile, the opposition alliance is a fairly new coalition formed after 2008 general election by the People’s Justice Party (PKR), Democratic Action Party (DAP), and Pan-Malaysian Islamic Party (PAS).
18
Since Independence, BN won at least two-thirds majority up until 2008 general election and has not been able to return to that position ever since.
The paradox of Malaysian media
The media system in Malaysia is made up by two very opposite components: a tightly-controlled
mainstream media and a relatively free alternative media made possible by looser Internet
regulations. The control over mainstream media is imposed by the State directly through various laws (including Internal Security Act 1960 and Official Secrets Act 1972) and indirectly through
shared ownership of media (Sani, 2005, p. 360).
Meanwhile, at the other end of the spectrum is an almost “no holds barred” Internet, which created a lively alternative media scene in Malaysia. George (2007) made a distinction between mainstream and alternative media, where he characterized mainstream media as
structurally tied to the power centre and comprises of large commercial or state-run organization while alternative media are usually small and linked to weak opposition parties.
Through online platforms, political opponents are able to spread campaign messages without much government interference and consequently mobilized the public against the ruling coalition (Willnat et al., 2013), especially through “alternative” news portals like Malaysiakini and The Malaysian Insider. A number of scholars have observed that the Internet helped the opposition attain electoral success in 2008 (see Ndoma & Tumin 2011; Moten, 2011; Weiss, 2012; Willnat et al., 2013).
The rise of social media
The prominence of social media in 2013 general election campaign was augmented by the growing number of Internet users. Since 2008, there is an exponential growth of social media
19
users: increasing from a mere 800,000 Facebook and 3,429 Twitter users to more than 13 million
Facebook and two million Twitter users (Gomez, 2013). Consequently, the political parties have
developed a Facebook presence to reach voters (Smeltzer & Keddy, 2010, p. 430). Facebook is undoubtedly the most important social network in Malaysia, as almost half of the population are on the medium (The WTM Social Media Report Malaysia, 2014), which is why it was chosen for the study.
With the omnipresence of social media in everyday life of Malaysians and its increasing usage in politics and elections, Malaysia is a fitting case to test the effect of exposure to exposure to electoral discussions in social media on political cynicism. More details of this experimental study will be provided in the methods section below.
Methods
This study employed a survey-embedded experiment and focuses on a single case study, specifically the Malaysian case. Generally, basic case study looked at the in-depth investigation of a single case, or one unit of analysis (in this case a single country) thus provides a context-dependent knowledge which is more practical.
Among the downside in a single case study is the external validity or generalizability of such research (Bryman, 2012, p. 69), as data is often unique to the particular approach, event or process. Malaysia is an excellent choice for case study based on the media usage alone.
According to Nielsen and comScore, the average Malaysians spends 19.8 hours online each week, and a third of that on social networking sites (The WTM Social Media Report Malaysia, 2014). However, it will be probably difficult to generalize it may be could be compared only to those in the same region (e.g., Indonesia), or with similar media environment. This study took on classic experimental design, which is regarded the best for testing causal relations due to
20
experimental control. This method is useful as the study attempts to prove the causality between exposure to social media and political cynicism.
The experiment-embedded survey is a randomized 1 x 4 post-test (between groups) to answer the research question and test the hypotheses. The survey-embedded experiment was carried out via online survey platform, Qualtrics. Participants were recruited via social media platforms, namely Facebook (researcher’s ownaccount and Facebook groups), Twitter (researcher’s own account) and LinkedIn groups by posting a link to the survey in these platforms.
All the participants for the study were Malaysian citizens, of voting age of 21 years (M=21, SD=2.5). A total of 56 participants received one of four randomly assigned treatments or control activities. The experimental group received experimental treatments, in this case different electoral discussions on Facebook. Basically, the experiment was designed to test the presence or absence of electoral discussions: a control group (without comments) and three treatment groups (with comments). Besides that, the three treatment groups were also based on comparison/control group. The control group omitted the key independent variable (electoral discussions), each with different partisan slants.
Procedures
The experiment is a single-blind design, with two main parts: a single exposure and a post-test. The whole experiment did not exceed more than 30 minutes on average. The
participants who got through the screening pages were then asked to answer basic demographic questions and questions indicating their partisan preference, vote choice, propensity-to-vote (PTV) and media usage. Then, participants were asked to read one of the four stimuli or control news articles. During the exposure to the stimulus, hidden time measures were included to
21
control for exposure. The exposure was set for 30 seconds, where the next button was disabled until the said time had passed.
After the exposure, participants filled out post-test scales. Besides that, they were asked to again indicate their vote choice and PTV. The survey ended with an information page that informed participants that a debriefing email will be sent to them. The debriefing procedure took place after the entire field work has been completed, where the researcher sent an email to participants to inform them about the true purpose of the study, the manipulations involved and more information about the study.
Stimulus
Given the popularity and influence of social media in GE-13, which was held on 5 May 2013, the screenshots of electoral discussions on Facebook were manipulated, particularly the comments on a selected news article about the Malaysia’s general election posted on these sites.
There were four groups: a control group (posted news article without comments) and three treatment groups (posted news article with comments); with balanced, pro-government (BN) and pro-opposition (PR) comments. In three treatment groups, the constancy of conditions, aside from the experimental manipulations (partisan slant) are maintained as we include the same pro-opposition/ pro-government comments as in the balanced condition. All screenshots are included in Appendices (Appendix 1) but examples of stimuli are as illustrated in Figure A, B and C below:
22
Figure A. News Article
23
Figure C. Pro-Government Condition
Variable measurements
Political Cynicism
The dependent variable (DV) and main post-test measurements is political cynicism. To assess participants’ political cynicism we used five items adapted from Pinkleton, Um and Austin (2002) and five items from Adriaansen et al. (2010), a fixed measurement for the posttest scales to measure the extent to which they distrust politicians. The dependent variable was
24
measured by the following items using seven-point Likert scales with “strongly disagree” and “strongly agree” as anchors.
Five items from Pinkleton et al. (2002) political cynicism measurements were taken: Politicians lose touch with the people once elected; candidates for office are only interested in peoples' votes, not in their opinions; too many politicians only serve themselves or their special interests; it seems our government is run by a few big interests who are just looking out for themselves and politicians lie to the media and the public. These are common measurement items for political cynicism (e.g.; Austin and Pinkleton, 1995; Mou et al., 2011).
Meanwhile, the remaining five items from Adriaansen et al. (2010) were adopted; politicians consciously promise more than they can deliver; in enabling someone to become Member of Parliaments and State assemblymen, friends are more important than abilities;
politicians do not understand what matters to society; politicians are capable of solving important problemsand most politicians are competent people who know what they are doing. To test whether these scales are reliable, especially the political cynicism scale (N=10), we run internal consistency reliability analysis Cronbach’s alpha scoring α = .884, which indicate that the scale is reliable, above the minimum 0.6 required.
Independent Variables (IVs): Demographics, Social Media Usage and Vote Choice
Overall, there werea few basic demographic questions (besides the age and nationality) including gender, race and education level and a few questions to determine partisan line and vote choice. In line with partisanship hypothesis, questions on vote choice were asked before the experiment (which political party/ alliance/ candidate did you vote for in the last general election
25
(GE-13)?) and again asked in post-experiment (if general election were to be held now, which party would you most likely vote?).
In order to test H1, the researcher looked at the respondents’ political cynicism score after exposure of electoral discussion on social media by either their control and treatment groups. To test H2, respondents were asked answer a few question to determine the volume and extent of their media usage, especially social media. Among the questions, for example, is the frequency of their social media usage by scales and by hours spent on social media. Whether social media usage moderated the effect of exposure were tested and investigated. Meanwhile, to test H3, both vote-choice questions before the exposure and after the exposure (as mentioned above) were the focus to see whether partisanship moderates the effect of exposure to social media discussion on political cynicism for each control/treatment groups.
After the exposure, respondents were asked to answer a few questions for manipulation checks which are mostly general questions about the article and the social media discussions (e.g.: Thinking of the newspaper article that you just read, please indicate to what extent you agree with the following statements – The comments mostly supported the views portrayed in the article). This is vital to determine whether or not the manipulation of the independent variable has had the intended effect on the respondents. It also provides evidence for the construct validity of the manipulation (Cozby, 2009).
Results
As seen in Figure 1, there are four groups in which political cynicism was measured: the Control group (N=9), Pro-Government group (N=18), Pro-opposition group (N=16), Balanced group (N=13). Meanwhile, Table 1 shows that the Control group (N=9) is associated with lower
26
level of political cynicism M=48 (SD= 9.14)than the Pro-Government group M=49.61 (SD=11.27).
Figure 1. Distribution across control and treatment groups
Table 1
Descriptive Statistics for Political Cynicism by groups
Control Pro-Government
Variables M SD M SD
Political Cynicism 48.00 9.14 49.61 11.27
To test the Hypothesis 1, whether the exposure to electoral discussion in social media decreases political cynicism among Malaysian voters, an independent t-test was performed. In the analysis, we compared the Control (N=9) and Treatment groups (N=47) to see whether the said exposure has any effect. All four groups (control and treatment) distributions were
27
sufficiently normal for the purpose of conducting a t-test (e.g.: skew < |2.0 | kurtosis < |9.0|; Schmider, Ziegler, Danay, Beyer and Bühner, 2010). Additionally, the assumption of homogeneity of variance was tested and satisfied via Levene’s F Test, F(54) = .07, p = .79.
There was no significant different between the Control group and the Treatment groups
t(54) = .774, p = .443 (two-tailed). Thus, the null hypothesis is maintained. The effect size,
measured by the eta squared 0.105. The magnitude difference in means was 2.77 with 95% confidence interval of -4.402 to 9.934. Table 2 depicts the descriptive statistics of -4.402 to 9.934.
Table 2
Descriptive Statistics for Political Cynicism between Control and Treatment groups
Control Treatment
Variables M SD M SD
Political Cynicism 48.00 9.14 50.77 9.94 Notes: coded 1.00 = Control, .00 = Treatment for groups
Although the null hypothesis for H1 is still maintained, when the control and treatment groups are use Facebook or are heavy social media users, the correlation between the exposure to social media exposure and political cynicism are statistically significant. A multiple regression analysis was performed to test whether social media usage moderated the effect of exposure to social media discussion to political cynicism. The reason why linear regression was used instead of ANOVA is because the variables were computed from continuous variables (the sum of social media related usage).
28
Levels of political cynicism differed significantly across both groups, who are exposed or not exposed to electoral discussions when moderated by social media usage. In order to see the interaction between social media usage and exposure/non-exposure groups in their influence on political cynicism, we created two models: one without and one with interaction effect. Table 3 shows the regression statistics between independent variables social media usage and
control/treatment groups towards political cynicism.
Table 3
Regression Statistics for Political Cynicism between Groups and Social Media Usage (N=56)
Variables DV: Political Cynicism Model 1 B Sig. (p) Model 2 B Sig. (p) Social Media Usage .57 .08 .65 .07 Groups - .95 .80 2.64 .73 Social Media Usage* Groups - .48 .60 Notes: coded 1.00 = Control, .00 = Treatment for Groups
In Model 2, the coefficient (parameter estimates) for social media usage index is approaching significance, p < .07, indicating that it significantly moderated the impact to electoral discussion on social media towards political cynicism. Furthermore, the coefficient for social media usage is .65. So for every unit increase in social media usage, .65 unit increase of political cynicism is recorded. This means people who have higher social media usage tend to be more cynical than those who are not avid users of social media. This is significantly different from 0. In sum, social media usage rather significantly predicted political cynicism score
,
beta = 0.65, t(56) = 1.8, p < .07. However, the result predicted the increase rather than the decrease of29
political cynicism when exposed to social media discussion, depending on their social media usage. This is the opposite of what was hypothesized in H2.
In addition, the interaction between social media usage and control and treatment groups showed insignificant, beta = -.48, p < .60 (as seen in Table 3). Hence, the null hypothesis is maintained and H2 is rejected. Nevertheless, it is important to note for future research that social media usage could have significantly moderated impact of exposure to social media discussion on political cynicism perhaps in different circumstances or context. This is because even without considering the exposure to social media discussion across control and treatment groups, social media usage significantly influenced political cynicism, beta = .04, p < .01 (see Table 4).
Table 4
Summary of Regression Analysis for Political Cynicism Scores (N=56)
DV: Political Cynicism B SE(B) T Sig. (p) Variable
Social Media Usage .04 .02 .84 .014
To test hypothesis 3 (H3), a one-way ANOVA was performed to look at whether vote choice moderates the impact of exposure to electoral discussion on social media towards political cynicism among Malaysian voters. In the hypothesis, we infer that exposure to dissonant
information increases cynicism while exposure to consonant electoral discussion decreases cynicism. When considered the cumulative index of political cynicism for all respondents (not considering exposure), partisanship has a more significant impact on cynicism, F (5, 56) = 4.871,
30
However, after the exposure, the researcher found that partisanship still significantly moderates impact of exposure to electoral discussion in social media on political cynicism, F(16, 56) =1.621, p < .01. As Figure 2 shows, when respondents indicated (post- exposure) that they will vote for the opposition Pakatan Rakyat, which is the opposition alliance, when exposed to electoral discussion that has a pro-government slant, they tend to be more cynical. Thus, we can reject the null hypothesis and proved Hypothesis 3. More details are provided in Table 5.
Table 5
Mean Level of Political Cynicism by Groups and Vote Choice
BN PR Control M SD .a -51.50 6.372 Political Cynicism Pro Government M SD 39.50 4.51 58.17 3.68 Pro Opposition M SD 45.80 4.03 59.00 9.011 Balanced M SD 48.25 4.51 55.833 3.679
31
Figure 2.Mean Level of Political Cynicism by Groups and Vote Choice
Discussion and Conclusion
Despite widespread concern about the effect of social media usage to political cynicism, the effect of exposure to social media discussions towards political cynicism are poorly
understood. Political information exposed in social media is usually either in the form of announcements, articles or comments in posted status that has a partisan slant or bias. This is because people tend to directly and indirectly reveal their partisan lines when exposed to such information that are either consonant or dissonant of their own political view. Therefore, this study tested the effect of exposure to social media discussion on political cynicism. This study also tested whether these effects were moderated by social media usage and partisanship.
32
The results indicate that, contrary to the expectations, exposure to electoral discussions in social media increased political cynicism. To illustrate the case, those who are not exposed to social media discussion (Control group) and those who are exposed to a media discussion with partisan slant towards the government (Pro-Government) became more cynical after being
exposed to mentioned control/treatment conditions. This is the opposite of our initial expectation, where it was inferred that similar exposure decreases political cynicism. The findings also
differed from the extant literature, which found that the usage of Internet and social media has the most probability to decrease rather than increase political cynicism (Hanson et al., 2010; Lariscy et al., 2011; Lin & Lim, 2012).
Our first hypothesis is rejected be due to a few reasons. One, in general, the level of existing distrust and cynicism towards the government among are quite high. This negative attitude towards the government is well placed. This is because as aforementioned, the government tries to control public discourse by obstructing public assembly and heavily regulating mainstream media through restrictive media laws, annual license renewal
requirements and concentrated ownership linked directly to the ruling (BN) coalition (Smeltzer & Keddy, 2010). Although several scholars were hopeful of the potential of social media especially popular social networking sites in its potential in circumventing the government “politically-motivated control mechanism” (Smeltzer and Keddy, 2010), some researchers were a lot cautious and sceptical.
This scepticism gives a base for the second reason: the popularity of social media has caused the media to be utilised by both sides: the ruling coalition Barisan Nasional and opposition coalition Pakatan Rakyat as rigorously as they would the mainstream media and alternative media, causing people to be cynical after the exposure to social media discussions.
33
Tapsell (2013) confirms this calculated assumption with his paper on media trends surrounding the Malaysia’s general election 2013 (GE13). He concludes that as online and social media have been increasingly influenced by political parties, especially UMNO (the leading party of the ruling coalition), it has created “greater polarisation and sensationalism rather than accurate and fair reporting”.
Tapsell (2013) reiterates that despite the hope that the “Fifth Estate” – an expression that has been referred to online and social media (Dutton, 2009) – is giving more autonomy and voice to the public, political parties gained greater control over this space and further manipulating this space during GE13. This manipulation is arguably similar to strategic framing, which has been found to have caused people to be more political cynical (Schuck et al., 2003), which can be further tested in future research. The third reason could be attributable to general cynicism and distrust towards politics, politicians and political institutions due to corruptions. The findings of 2013 Transparency International survey in Malaysia showed that Malaysians “are losing confidence in the government’s effort to fight corruption” and found that political parties and police were perceived to be the two most corrupted institutions (each scored 3.8 and 4.0 in the corruption scale, out of a five) (Leong, 2013).
When moderated by their social media usage, the effect of exposure to electoral
discussion on social media towards decreasing political cynicism is not significant. As discussed in result section, the second hypothesis is also rejected. The result instead indicates that heavy social media users are more cynical. The impact of exposure to social media discussion in
increasing rather than decreasing cynicism, especially among heavy social media users could be
further dissected. This too goes against the documented extant literature, for example Pinkleton et al. (1998), which found that cynicism is negatively associated with the active media use.
34
One possible explanation of the reverse result is because social media usage increased users’ exposure to variety of news and politically diverse information (Messing & Westwood, 2012) and this massive flow of information could make people to select channels, for example, following online news sites that fit their predisposition and partisan lines on social media to receive update. However, how would one explain the moderating effect of more frequent social media usage on increasing political cynicism?
One justification could be content-related negativism (Pinkleton et al., 1998). The use of social media from an avenue for public voice has been reduced to the “increase mobilisation of supporters through Facebook and Twitter, followed by what seems like a ‘cyberwar’ involving ‘cyber troopers’, the prevalence of fake Twitter accounts, deliberate misinformation and increase paid mobilisation of online participants” (see Tapsell, 2013) – looking more and more like an organised political campaign. Therefore, content-related negativism (the feeling of disgust towards mediated campaign coverage), could apply here causing people to be more cynical after the exposure of social media discussion. As Pinkleton et al. (1998) suggests, these voters, as they use social media more and more, they maybe more negative towards the content in the media and leading them to draw conclusion that can increase political cynicism.
Whereas H1 and H2 are rejected, hypothesis 3 (H3) is accepted. Partisanship has a strong effect on moderating the impact of exposure to electoral discussion on social media towards political cynicism. These findings mirrored Levedunsky (2012) argument that when exposed to messages that clashes with their own beliefs, it might make them more cynical when they are expose to the message or encounter them. Our findings confirmed this as when people who select opposition alliance Pakatan Rakyat as a party they would vote for, when they are exposed to social media discussion that is pro-government/ruling alliance (treatment groups), they tend to be
35
more cynical. The findings also confirmed the researcher’s theory based on Zaller’s acceptance model in which states “if and when citizens are exposed to counter attitudinal messages, their acceptance of the message is dependent on their partisan and ideological orientation”.
On the other hands, Prior’s (2013) study that found mixed evidence for a causal link between more partisan messages and changing attitude and behaviours should be further investigated, as the result proved a link between them.
This study, however, has several limitations. The first obvious limitation is the small sample size (N=56). This could potentially be caused by the flaw in the design instrument. After the field work is completed, the researcher realized a slight glitch in the manipulation. The stimulus, in the form of graphics of news article and posted article in Facebook (with or without comments) took longer to load in the online survey platform Qualtrics. This might discourage respondents to continue with the survey. This is reflected in the high drop-out rate of the survey and low completion mean. Among 154 people who started the survey, there are only 61 who completed the survey, and only accounted for 56 valid surveys. Besides that, researcher runs the study using personal expense and thus not able to attain for respondents using crowdsourcing platforms like Amazon MTurk and clickworker.com. Furthermore, as the link to survey is distributed in researchers’ social media pages, the prospect of getting respondents are only limited to researchers’ social media contacts. Future research should see if there is indeed impact of exposure to social media discussion on political cynicism with bigger sample size to get a more significant relationship and robust findings.
The second main limitation is the nature of the research. The study focused only on the Malaysian case, thus it could not produce a generalizable result. Moreover, as this an
36
ecological validity (Bryman, 2012). It is interesting to see if this could be expanded to other
context and whether the hypothesis could be the opposite in other countries and political systems.
The third limitation is the nature of political discussions used as stimuli in this study. Most of the comments in the social media discussions are emotional and bias (especially pro-government and pro-opposition. Different stimulus and different content or messages could yield different results. Choi’s (2014) study on social-media-based public forums showed that although political discussions on social media are more emotional than cognitive, cognitive discussions are more influential than emotional ones. More objective and factual comments could alter effects on political cynicism, which should be further investigated. Still, this study illustrate Zaller’s acceptance model or the concept of partisan selective exposure provides a good base for future research in determining influence of partisanship in different media effect study. In addition, future research should explore whether strategic framing exist in electoral discussion on social media. Besides that, researchers should test the different effects of exposure to social media discussions, and how this differs according to their age, education and gender. Finally, researchers should really looked at partisanship influence in a different context and environment to see whether there are stronger relationship between the two and find out the reason behind the media effect.
The effect of social media towards political cynicism is an important topic to discuss because it has implications for the practice of citizenship and democracy as a whole. In the local context (Malaysia), social media is instrumental in influencing voters’ choice, as first-time voters made up approximately 30 per cent of the 13.3 million registered voters in Malaysia for GE13 particularly in urban areas (Leong, 2013). The ruling coalition has realized this and has been
37
trying to have greater presence in the realm of social media. With the growing users of social media, the study could contribute to further understanding of voters’ preferences and usage of social media for future electoral campaign.
Most importantly, the findings saw that Malaysian voters are not less cynical when they used or are exposed to social media content contrary to previous findings signalling a much larger systemic issue of the political system in Malaysia. As David Brin says, “only a
knowledgeable, empowered and vocal citizenry can performed well in democracy”. Social media could be the answer to a more empowered society and strengthened democracy only if it is not “shackled” or muddled with certain ideology and partisan interests and is allowed to be the tool it could really be.
38
References
Adriaansen, M. L., van Praag, P., & de Vreese, C. H. (2010). Substance matters: How news content can reduce political cynicism. International Journal of Public Opinion
Research, 22(4), 433-457.
Agger, R. E., Goldstein, M. N., & Pearl, S. A. (1961). Political cynicism: Measurement and meaning. The Journal of Politics, 23(03), 477-506.
Austin, E. W., & Pinkleton, B. E. (1995). Positive and negative effects of political disaffection on the less experienced voter. Journal of Broadcasting & Electronic Media, 39(2), 215-235.
Bennett, W. L., & Iyengar, S. (2008). A new era of minimal effects? The changing foundations of political communication. Journal of Communication, 58(4), 707-731.
Berelson, B., & Steiner, G. A. (1964). Human behavior: An inventory of scientific findings. Bryman, A. (2012). Social research methods (4th edition). Oxford: Oxford University Press. Bucy, E. P., D'Angelo, P., & Newhagen, J. E. (1999). The engaged electorate: New media use as
political participation. In L. L. Kaid & D. G. Bystrom (Eds.), The Electronic Election: Perspectives on the 1996 campaign communication, (pp. 335-348). Mahwah, NJ: Lawrence Erlbaum.
Cappella, J. N. (2002). Cynicism and social trust in the new media environment. Journal of
Communication, 52(1), 229-241.
Cappella, J. N., & Jamieson, K. H. (1996). News frames, political cynicism, and media cynicism. The Annals of the American Academy of Political and Social Science, 71-84.
39
(p. 22). New York: Oxford University Press.
Choi, S. (2014). Flow, Diversity, Form, and Influence of Political Talk in Social‐Media‐Based Public Forums. Human Communication Research,40(2), 209-237.
Cozby, P. C. (2009). Methods in behavioral research. 9. utg.
Dennis, J., & Webster, C. (1975). Children's Images of the President and of Government in 1962 and 1974. American Politics Research, 3(4), 386-405.
de Vreese, C. H. (2005). The spiral of cynicism reconsidered. European Journal of
Communication, 20(3), 283-301.
de Vreese, C. H., & Boomgaarden, H. (2006). News, political knowledge and participation: The differential effects of news media exposure on political knowledge and
participation. Acta Politica, 41(4), 317.
Donsbach, W. (1991). Exposure to political content in newspapers: The impact of cognitive dissonance on readers' selectivity. European Journal of Communication, 6(2), 155-186. Dutton, W. H. (2009). The fifth estate emerging through the network of
networks. Prometheus, 27(1), 1-15.
Eisinger, R. M. (2000). Questioning cynicism. Society, 37(5), 55-60.
Festinger, L. (1962). A theory of cognitive dissonance (Vol. 2). Stanford university press. Freedomhouse.org. (2012). Malaysia | Freedom House. [online] Retrieved from:
http://www.freedomhouse.org/report/freedom-world/2012/malaysia [Accessed: 23 Mar
2014].
George, C. (2007). Media in Malaysia: Zone of Contention.Democratization, 14(5), 893-910. Goldfarb, J. C. (1991). The cynical society: The culture of politics and the politics of culture
40
Gomez, J. (2013). Malaysia’s 13th general election: social media and its political impact. [online] Retrieved from:
http://www.academia.edu/4446983/Malaysias_13th_General_Election_Social_Med ia_and_its_Political_Impact [Accessed: 18 Jan 2014].
Hanson, G., Haridakis, P. M., Cunningham, A. W., Sharma, R., & Ponder, J. D. (2010). The 2008 presidential campaign: Political cynicism in the age of Facebook, MySpace, and YouTube. Mass Communication and Society, 13(5), 584-607.
Howard, P. N. & Parks, M. R. (2012). Social media and political change: capacity, constraint, and consequence. Journal of communication, 62 p. 359–362.
doi:10.1111/j.1460-2466.2012.01626.x
Johnson, T. J., Braima, M. A., & Sothirajah, J. (1999). Doing the traditional media sidestep: Comparing the effects of the Internet and other nontraditional media with traditional media in the 1996 presidential campaign. Journalism & Mass Communication
Quarterly,76(1), 99-123.
Lariscy, R. W., Tinkham, S. F., & Sweetser, K. D. (2011). Kids these days: Examining differences in political uses and gratifications, Internet political participation, political information efficacy, and cynicism on the basis of age. American Behavioral Scientist,
55(6), 749-764.
Leong, T.(2013). Malaysians see cops, politicians as most corrupt. [online] Available at:
http://www.themalaysianinsider.com/malaysia/article/malaysians-see-cops-politicians-as-most-corrupted [Accessed 20 June 2014].
Levendusky, M. (2013). Partisan media exposure and attitudes toward the opposition. Political
41
Liew, C. (2012). International community take note – regime change looms - Centre Tricontinental - CETRI. Cetri.be. Retrieved 1 May 2014, from
http://www.cetri.be/spip.php?article2759&lang=fr
Lin, Y., & Lim, S. (2002). Relationships of media use to political cynicism and efficacy: A preliminary study of young South Korean voters.Asian Journal of communication,
12(1), 25-39.
Malaysia’s Social Media Election. (2014). [online] Available at:
http://stream.aljazeera.com/story/201305012242-0022719[Accessed: 18 December
2013].
Malaysia’s social media election is already over — James Gomez. (2013). The Malaysian
Insider Retrieved from
http://www.themalaysianinsider.com/sideviews/article/malaysias-social-media-election-is-already-over-james-gomez/
Meffert, Michael F., Sungeun Chung, Amber J. Joiner, Leah Waks, and Jennifer Garst. 2006. “The effects of negativity and motivated information processing during a political campaign.” Journal of Communication 56: 27-51.
Messing, S., & Westwood, S. J. (2012). Selective exposure in the age of social media: Endorsements trump partisan source affiliation when selecting news online.
Communication Research, 0093650212466406.
Miller, A. H. (1974). Political issues and trust in government: 1964–1970.American Political
Science Review, 68(03), 951-972.
Moten, A. R. (2011). Changing political culture and electoral behavior in malaysia. Asian
42
Mou, Y., Atkin, D., & Fu, H. (2011). Predicting political discussion in a censored virtual environment. Political Communication, 28(3), 341-356.
Ndoma, I., & Tumin, M. (2011). Virtual Civil Society: Malaysia’s 2008 General Elections Revisited. Global Journal of Human-Social Science Research, 11(8).
Nisbet, E. C., Stoycheff, E., & Pearce, K. E. (2012). Internet use and democratic demands: A multinational, multilevel model of internet use and citizen attitudes about
democracy. Journal of Communication, 62(2), 249-265. Patterson, T. 1993. Out of Order. New York: Knopf.
Pinkleton, B. E., Austin, E. W., & Fortman, K. K. (1998). Relationships of media use and
political disaffection to political efficacy and voting behavior. Journal of Broadcasting &
Electronic Media, 42(1), 34-49.
Pinkleton, B. E., Um, N. H., & Austin, E. W. (2002). An exploration of the effects of negative political advertising on political decision making.Journal of Advertising, 31(1), 13-25. Prior, M. (2013). Media and political polarization. Annual Review of Political Science, 16, 101-
127.
Raoof, J. K., Zaman, H. B., Ahmad, A., & Al-Qaraghuli, A. (2013). Using social network systems as a tool for political change. International Journal of Physical Sciences,
8(21), 1143-1148.
Rhee, J. W. (1997). Strategy and issue frames in election campaign coverage: A social cognitive account of framing effects. Journal of Communication, 47(3), 26-48.
Sani, M. A. M. (2005). Media freedom in Malaysia. Journal of Contemporary Asia, 35(3), 341-367.
43
impact of election news on political cynicism in comparative perspective. Journal of
Communication, 63(2), 287-311.
Schyns, P., Nuus, M. & Dekker, H. (2004). "A Conceptual and Empirical Evaluation of Political Cynicism", paper presented at Politicologenetmaal. Antwerp, Belgium, May 27-28.
Sedghi, A. (2014). Facebook: 10 years of social networking, in numbers. Retrieved from http://www.theguardian.com/news/datablog/2014/feb/04/facebook-in-numbers-statistics Smeltzer, S. & Keddy, D. (2010). Won't you be my (political) friend? The changing face
(book) of socio-political contestation in Malaysia.Canadian Journal of Development
Studies/Revue canadienne d'études du développement, 30(3-4), 421-440.
Stroud, N. J. (2008). Media use and political predispositions: Revisiting the concept of selective exposure. Political Behavior, 30(3), 341-366.
Sunstein, C. R. (2001). Republic.com. Princeton, NJ: Princeton University Press
Tankard, J. W. (2001). The empirical approach to the study of media framing. Framing public
life: Perspectives on media and our understanding of the social world, 95-106.
Tapsell, R. (2013). The Media Freedom Movement in Malaysia and the Electoral Authoritarian Regime. Journal of Contemporary Asia, 43(4), 613-635.
The WTM Social Media Report, Malaysia, Issue 3. (2014). [online] Available at: http://www.wtmlondon.com/library_2/3056144_assocPDF.pdf.
Valentino, N. A., Buhr, T. A., & Beckmann, M. N. (2001). When the Frame is the Game: Revisiting the Impact of “Strategic” Campaign Coverage on Citizens' Information Retention. Journalism & Mass Communication Quarterly, 78(1), 93-112.
44
Contemporary Asia, 43 (4), 591612. doi:10.1080/00472336.2012.759332
Westling, M. (2007). Expanding the public sphere: The impact of Facebook on political communication. The New Vernacular.
Willnat, L., Wong, W. J., Tamam, E., & Aw, A. (2013). Online Media and Political Participation: The Case of Malaysia. Mass Communication and Society, (just-accepted).
Zahiid, S. J. (2013). Najib: Election 2013 first social media election. [online] Retrieved from: http://www.themalaysianinsider.com/malaysia/article/najib-election-2013-first-social-me
dia-election/ [Accessed: 15 Mar 2014].
Zaller, J. (1993). The converse‐McGuire model of attitude change and the gulf war opinion rally. Political Communication, 10(4), 369-388.
45
Appendices
Appendix A: Stimulus (News Article)
46
Appendix B: Stimulus (Electoral Discussion)
47
Figure iii. Treatment Condition: Pro-Government
Translation
a) Posted comments with article
Original Text: Tindakan terdesak PR supaya boleh satu menjadikan ianya satu alasan
untuk mencetusi provokosi
48
b) Comments Comment 1
Original Text: BN tak terdesak. BN steady. Orang PR je terdesak. Boleh saja kempen
dalam media tapi kenapa tak nak letak logo parti? Malu? Takut? Setakat maki hamun dalam laman social berani la. Inilah PR. Pengecut.
Translation: BN (Barisan Nasional) is not desperate. BN steady. PR (Pakatan Rakyat) is
the desperate one. It is fine that they want to launch a media campaign but why don’t they own up to it and included their logo. Embarrassed? Scared? They only dare to insult and voice up their dissatisfaction in social media. This is PR. Coward.
Comment 2
Original Text: Ingat senang k kalau kita daa bertukar kerajaan, terimah jak laa yg sda ada
depan mata, Negara kita jagk telah diiktiraf negara yg ke 19 teraman di seluruh negara hidup BN <3
Translation: Do you think it is easier if the government changed hands, just be grateful
for what we have now. Malaysia has been internationally recognised as the 19th most peaceful country in the world.
Comment 3
Original Text: BN and PR supporter 2x5=10 Translation: BN and PR supporters are the same.
Comment 4
Original Text: Semua tidak telus
Translation: All (both political parties) are not transparent
