How mainstream and alternative media shape public attitudes toward social change: Evidence from two panel studies during Malaysia’s democratic transition

Abstract

Mainstream and alternative media often frame key political events in divergent ways. The present research examined how mainstream and alternative media consumption was linked to public support for Bersih, a pro-democracy movement in Malaysia. We conducted a two-wave panel study before and after the Bersih 5 protests demanding electoral reform (N = 422), and another two-wave panel study before and after the 14th General Elections which were historic in unseating the ruling government (N = 386). Against mounting corruption, the two events were important to Malaysia’s democratic transition. Across both studies, alternative media consumption was linked to more positive attitudes toward the Bersih movement, especially among people who were strong supporters of the previous ruling government. Mainstream media did not play a consistent role in shaping attitudes toward the movement, nor was there evidence of backlash among government supporters. Thus, alternative media may have legitimized the cause for social change.

Keywords

Alternative media Bersih collective action elections mainstream media Malaysia media framing public opinion social change social movements

Malaysia has undergone rapid social and political change over the past few years. In November 2016, thousands of Malaysians participated in a mass street protest to demand clean and fair elections. Known as Bersih 5, it was the fifth round of protests organized by Bersih, a political reform movement named after the word “clean” in Malay language. Media portrayal of the Bersih 5 protest widely differed between mainstream and alternative media outlets (Kee et al., 2013). Consistent with the idea that mainstream media tends to legitimize the status quo, whereas alternative media tends to challenge it (e.g. Downing, 2003; McLeod, 1995), Utusan Melayu, a prominent mainstream news outlet, labeled the Bersih protest as an illegal assembly, whereas Free Malaysia Today, a popular alternative news outlet, highlighted the festive spirit of the same protest. Less than 2 years after the Bersih 5 protest occurred, in May 2018, the 14th Malaysian General Elections went down in history for voting out the regime that had ruled the country since independence, in favor of a multi-racial coalition that championed democratic reforms. For many observers, this election outcome was in part due to the Bersih movement’s decade-long mobilization for social change (Chan, 2018). Once again reflecting the divergence between mainstream and alternative media, in the aftermath of the elections, market share of the mainstream newspaper Utusan Melayu plummeted, whereas online readership for the alternative news outlet Free Malaysia Today soared (Azman, 2018; Nain, 2018). This recent history of Malaysia reveals the tension between mainstream and alternative media in relation to social change.

Mirroring these anecdotes from Malaysia, extant research on the content of news media has shown that mainstream media outlets tend to frame movements for social change in less favorable ways compared to alternative media outlets (e.g. McLeod, 1995; Smeltzer and Lepawsky, 2010). One key distinction between mainstream and alternative media is that the former tends to reflect the interests of dominant groups including the government and large corporations, whereas the latter tends to reflect the interests of marginalized groups in society including those seeking social change and challenging the status quo (Couldry and Curran, 2003; Downing, 2003). Media portrayal of social change efforts (such as protests) are consequential because one of the main ways the public comes to learn about social movements is through the media, so much so that social movement scholars have long relied on newspaper data to study protest events (see Earl et al., 2004). Since achieving long-term social change involves mobilizing support from the general public, social movements tend to rely on the media to garner widespread support for their goals.

The present research examined whether mainstream and alternative media consumption might have disparate outcomes on the public’s subsequent attitudes toward movements for social change. Specifically, consuming mainstream media may be linked to greater opposition toward a movement for social change, whereas consuming alternative media may be linked to greater support for the movement. However, there remain critical questions about the extent to which media consumption merely reflects – rather than shapes – public opinion. Little is known about the degree to which supporters of the status quo (whose attitudes are typically reflected in and reinforced by mainstream media framing) may be shaped by alternative media (with framing that might generally challenge those with favorable attitudes toward the status quo). To examine these questions, we conducted two panel studies situated around important events in Malaysia’s recent history. The first panel survey was conducted before and after the Bersih 5 protests in November 2016. The second panel survey was conducted before and after the 14th General Elections in May 2018.

The Malaysian context

As a society that has lived under the rule of soft authoritarianism (sometimes known as a semi-democracy), Malaysian elections over the past few decades have been regularly marred by corrupt practices including gerrymandering, vote-buying, and bribery (Case, 2009; Means, 1996). In response, the Bersih movement emerged as a grassroots effort to mobilize the broader public against political corruption and the need for electoral reform (Khoo, 2014). To raise public support for its cause, a key challenge for a movement like Bersih is that major mainstream media outlets can be monopolized by the state, which generally aims to suppress dissent. Up until June 2019, the United Malays National Organization (UMNO)—which is the dominant political party in the prior ruling coalition (Barisan National)—owned Media Prima, the corporation that runs mainstream television stations and newspapers in Malaysia (Brown, 2005; Nain, 2018). Furthermore, at the time of writing, media freedom remains largely restricted in Malaysia, for example, through the implementation of laws that give the government control over media licenses (Brown, 2005). Since the government has the power to suspend any press outlets that are believed to be a threat to public order or national security, journalists and editors are often encouraged to self-censor (George, 2007; Nain, 2018).

In response to political monopoly of mainstream media, alternative media operated by civil society organizations, independent businesses, or individual actors has a growing presence in Malaysia especially on online platforms (e.g. blogs, social media, online news networks), where government censorship and control is more difficult (Abbott and Givens, 2015; Smeltzer, 2008). Malaysian alternative media sources tend to be trusted as more independent news outlets, and they often function as government watchdogs that provide a platform for airing critical and dissenting views (Lumsden, 2013; Nain, 2018; Sanawi, 2014). Thus, the content of mainstream and alternative media in Malaysia tends to reflect its ownership structure. Against the backdrop of governmental corruption and the Bersih movement pushing for electoral reform, ordinary Malaysians are often exposed to both mainstream and alternative media framings of political issues (see Nain, 2018). The media and political landscape in Malaysia therefore allowed us to examine potential effects of media framing on public attitudes toward an ongoing movement for social change.

Media framing effects

The media tends to set a reference point for its consumers and organizes everyday events into a discernible idea or storyline (Entman, 2007). By doing so, the media makes certain elements of social reality more important or relevant than others, which can subsequently shape social judgments on a range of outcomes—including policy support, voting patterns, and intergroup attitudes (Scheufele, 1999). How media coverage shapes public opinion is referred to as a framing effect. Although there is rich debate about the theoretical nuances underlying framing effects and how it is different or similar to other social and psychological processes (Borah, 2011; Chong and Druckman, 2007), scholars generally agree that framing effects are about how social reality can be (directly or indirectly) constructed by the media to such an extent that it shapes public opinion. The subtle decisions journalists make about how to cover news stories—through the words, images, and perspectives they highlight—can portray the same issue or event in different ways, which can subsequently influence public attitudes. In the present research, we focused on mainstream and alternative media frames.

Mainstream media is typically embedded in, and reflective of, the dominant power structures within a society. For this reason, mainstream media tends to delegitimize and marginalize social movements that challenge the status quo—a phenomena that media scholars refer to as the “protest paradigm” (Chan and Lee, 1984). According to this paradigm, protesters demanding social change are typically portrayed as dangerous and deviant in mainstream news outlets (Boyle et al., 2004; McLeod and Hertog, 1992) which can in turn shape public perception of those protests (McLeod, 1995). The negative portrayal of protests may be further compounded when the media is not free and is instead controlled by the government, as has been the case with some elements of the media in Malaysia. In semi-democratic or semi-authoritarian nations like Malaysia, the ruling government can exploit mainstream media to foster public support for the system, infringe upon freedom of speech, and suppress dissent (Anuar, 2000). Malaysian mainstream media has widely favored the Barisan National (BN) ruling government during past election campaigns and has attempted to silence or tarnish government critics including the Bersih movement (Abbott, 2011; Manaf and Sedu, 2015; Rajaratnam, 2009; Sani, 2014).

To circumvent mainstream media that largely serves the interests of the dominant groups in society, alternative media sources have been developed to provide a platform for low-status or marginalized groups seeking to challenge the status quo (Couldry and Curran, 2003; Downing, 2003). Alternative media outlets tend to be owned by independent groups or organizations that are not linked to the ruling government and feature content from various actors including freelance reporters, political opponents, activists, and academics (Anuar, 2000; George, 2007). In an attempt to bypass government scrutiny or surveillance, alternative media typically proliferate on the Internet and are widely circulated on social media, blogs, and online news networks. Scholars have noted how the Internet and social media in particular have played a key role in mobilizing collective action since it can provide an avenue for public discourse critical of the status quo (McGarty et al., 2014). Malaysian alternative media tend to portray collective action pushing for social change in a more positive manner, such as by emphasizing racial solidarity during Bersih protests (Sanawi, 2014). Alternative media outlets in Malaysia are also important during elections because they provide information about the opposition coalition’s manifesto and campaign efforts, which are under-reported in mainstream media (Lumsden, 2013; Rajaratnam, 2009; Smeltzer and Lepawsky, 2010).

Besides research on the content of mainstream and alternative media, scholars have also begun to investigate the downstream consequences of consuming alternative and mainstream media on people’s political attitudes and behaviors. Boyle and Schmierbach (2009), for example, found that alternative (but not mainstream) media consumption was linked to increased self-reported participation in a range of political behaviors including protests. Similarly, Chan (2017) found that consuming alternative media predicted greater identification with, and participation in, the pro-democracy movement in Hong Kong. Prior research therefore supports our hypothesis that mainstream and alternative media exposure may have divergent downstream effects on public opinion toward social change. Using a two-wave panel design allowed us to isolate the effects of habitual and specific media consumption around key political events. By focusing on media consumption before and after real-world political events, our design also adds a temporal element with increased external validity, which complements past research that has largely been cross-sectional.

Although we aimed to examine how mainstream and alternative media influences public attitudes, it is important to note that the link between media consumption and public attitudes may be bidirectional. It is very likely that people’s pre-existing attitudes influence their decisions and habits around media consumption. It has been found that people tend to selectively consume certain media sources based on pre-existing political leanings (Stroud, 2008). For instance, Leung and Lee (2014) showed that people who supported democratic principles were more likely to consume alternative media and be more critical toward mainstream media. It is therefore possible that Malaysians who are more supportive of the former ruling government would be more likely to consume mainstream media, whereas people who are more positive toward the Bersih movement would be more likely to consume alternative media. However, if it is the case that mainstream and alternative media consumption can shape subsequent public attitudes, we hypothesized that beyond habitual patterns of media consumption, exposure to mainstream and alternative media should shape subsequent attitudes toward the Bersih movement. Specifically, we hypothesized the following:

Hypothesis 1. Mainstream media consumption will predict more negative attitudes toward the Bersih movement, even after controlling for habitual media consumption and pre-existing movement attitudes.

Hypothesis 2. Alternative media consumption will predict more positive attitudes toward the Bersih movement, even after controlling for habitual media consumption and pre-existing movement attitudes.

Public attitudes toward the status quo

Beyond the effects of consuming mainstream and alternative media on attitudes, audience characteristics also play a role in shaping media framing effects. Druckman (2001) in particular has argued that how people understand, interpret, and react to media frames are filtered through their prior attitudes. Building on research that has shown the potential moderating role of audience characteristics in media framing effects (see Chong and Druckman, 2007), we expected that people’s prior attitudes toward the ruling government will shape how much mainstream and alternative media consumption predicts Bersih movement attitudes. However, we had competing predictions about the pattern of results. On the one hand, the more people supported the ruling government, the more they might be influenced by mainstream media rather than alternative media—since mainstream media could potentially reinforce their pre-existing attitudes. On the other hand, the more people supported the ruling government, the more they might be influenced by alternative media rather than mainstream media—because alternative media might legitimize a movement for social change among members of society who are most resistant to supporting social change. Specifically, we hypothesized two opposing patterns:

Hypothesis 3a. The link between mainstream media consumption and negative attitudes toward the movement will be stronger among government supporters.

Hypothesis 3b. The link between alternative media consumption and positive attitudes toward the movement will be stronger among government supporters.

We tested these divergent predictions across both studies.

Study 1

Study 1 was conducted around the Bersih 5 mass street protests, the fifth round of protests organized by the Bersih movement to demand electoral reform in Malaysia. The protests drew thousands of Malaysians to the streets on 19 November 2016. We fielded a two-wave longitudinal survey of Malaysians; Time 1 surveys were completed at least 2 weeks before the Bersih 5 protests occurred (24 October–6 November 2016) and Time 2 surveys were completed from the day after the protests and continued for about 2 weeks (20 November–6 December 2016). We examined whether mainstream and alternative media consumption, specifically consuming protest-related information, would be related to attitudes toward the movement. We expected that mainstream media consumption around Bersih 5 would predict more negative attitudes toward Bersih (Hypothesis 1), whereas alternative media consumption around Bersih 5 would predict more positive attitudes toward Bersih (Hypothesis 2). We also examined whether these effects differed depending on people’s pre-existing attitudes toward the then-ruling government. Here, we had competing hypotheses: those who strongly supported the then-ruling government may be more influenced by mainstream media as it would reinforce their pre-existing attitudes (Hypothesis 3a) or they might be more influenced by alternative media since it would challenge their pre-existing attitudes to be more in line with the movement (Hypothesis 3b).

Participants and procedure

A convenience sample of Malaysians were recruited to complete online surveys via TurkPrime. At Time 1, 612 participants completed the survey, and of those, 422 participants returned at Time 2.¹ Demographic characteristics of the sample are reported in Table 1, along with the national population estimates for comparison. Descriptively speaking, compared with the general population of Malaysia, our sample on average had a higher level of education. We also sampled more Chinese Malaysians, and as a result, our sample had more Buddhists and Christians compared with the national population.² Measures relevant to the present study were included as part of a larger survey on Malaysians’ socio-political attitudes.³ Variables measured with more than one item were averaged to obtain a composite score. The data, analysis code, and complete study materials are available on the Open Science Framework (OSF): https://osf.io/q3264/

Table 1.

Demographic information of Study 1 sample (N = 422), and demographics of the national population for comparison.

Criteria	Study 1 sample	National population
Sex ratio (ratio of males to females)	108:100 (12 did not report sex)	102:100 (in 2015–2019)
Median age	32.0 years (range: 18–68 years)	28.0 years (range: 0–65+ years; in 2016)
Median household income level per month	RM4000–RM6000 (25.59%)	RM5228 (in 2016)
Median education level	Graduated from college or university (58.06%)	Gross graduation ratio of tertiary programs: 23.35% (in 2017)
Distribution of three primary races
Bumiputera (includes Malay and Indigenous groups)	51.66% (all Malay)	68.6% (in 2016)
Chinese	41.94%	23.4% (in 2016)
Indian	6.40%	7.0% (in 2016)
Distribution of four primary religions
Islam	51.42%	61.3% (in 2010)
Buddhism	23.22%	19.8% (in 2010)
Christianity	13.27%	9.2% (in 2010)
Hinduism	3.32%	6.3% (in 2010)

National population estimates are taken from the Department of Statistics Malaysia and the World Bank. When estimates are not available for 2016, the closest available estimate is used.

Measures

Measured only at Time 1

Habitual media consumption

Participants reported how much they generally paid attention to five different media sources, on a scale from 1 (no attention at all) to 7 (a lot of attention). Mainstream media was measured using the following two items: “Mainstream news reports (e.g. Utusan, News Straits Times)” and “Local public TV channels (e.g. TV3, Bernama TV)” (r = .73, p < .001). Alternative media was measured using the following four items: “Alternative news reports (e.g. Malaysiakini, Free Malaysia Today),” “Local private TV channels (e.g. Astro Awani),” “International news reports (e.g. CNN, New York Times),” and “Social media (e.g. Facebook, Twitter)” (α = .73).

Government support

Participants indicated how they felt toward “the Malaysian government” (which at the time, referred to the BN government) on a scale from 1 (very negative) to 7 (very positive).

Measured only at Time 2

Protest-specific media consumption

Participants reported how much they paid attention to the same mainstream (two items; r = .72, p < .001) and alternative media outlets (four items; α = .76) used to measure habitual media consumption at Time 1, but specifically for information about the Bersih 5 protests, on a scale from 1 (no attention at all) to 7 (a lot of attention).

Measured at Time 1 and Time 2

Movement identity

Participants described the extent to which they were an active supporter of the Bersih movement, on a scale from 1 to 7, whereby higher scores indicated a more supportive movement identity (adapted from Dunlap and McCright’s [2008] measure of movement identity).

Movement efficacy

Participants responded to three statements about the Bersih movement (e.g. “The movement can be successful in their goal to change the country for the better”) on a scale from 1 (strongly disagree) to 7 (strongly agree) (α_T1 = .94; α_T2 = .96).

Support for social change

Participants responded to four statements on the demands of the Bersih movement (e.g. “The political system in Malaysia needs to be reformed”) on a scale from 1 (strongly disagree) to 7 (strongly agree) (α_T1 = .80; α_T2 = .86).

Results

Table 2 displays descriptive statistics on key variables. As seen in Table 2, mainstream media consumption was related to more government support, but alternative media was not significantly related to government support—suggesting that people were not necessarily avoiding nor selectively consuming alternative media due to their pre-existing attitudes toward the government. In line with this, when we examined the distribution of media consumption scores for participants at different levels of pre-existing government support, we found that participants at both high and low levels of pre-existing government support consumed relatively high levels of alternative media (means above 4 on a scale from 1 to 7).⁴ Furthermore, in line with expectations, mainstream media tended to be negatively related to movement attitudes, whereas alternative media was positively related to movement attitudes.

Table 2.

Means, standard deviations, and bivariate correlations between media consumption and attitudes in Study 1.

Variable	M (SD)	Mainstream media_T1	Alternative media_T1	Mainstream media_T2	Alternative media_T2
Government support _T1	3.40 (1.71)	.43***	.07	.34***	−.06
Movement identity_T1	4.45 (.94)	−.19***	.24***	−.09	.30***
Movement efficacy_T1	4.65 (1.67)	−.14**	.33***	−.04	.38***
Support for social change_T1	6.36 (.83)	−.13**	.21***	−.12*	.23***
Movement identity_T2	4.61 (1.06)	−.18***	.24***	−.07	.42***
Movement efficacy_T2	4.99 (1.64)	−.12*	.30***	−.05	.43***
Support for social change_T2	6.27 (.99)	−.18***	.18***	−.14**	.36***

SD: standard deviation.

p < .05, ** p < .01, *** p < .001.

Average changes over time

First, as we did in our previous work (Selvanathan and Lickel, 2019), we tested for changes in average levels of movement attitudes from before to after the protests. Overall, movement attitudes appeared more positive after the elections. Specifically, there was an average increase in movement identification (M_T1 = 4.45, SD_T1 = .94; M_T2 = 4.61, SD_T2 = 1.06; t(419) = 3.89, p < .001, Cohen’s d = .19) and in movement efficacy (M_T1 = 4.65, SD_T1 = 1.67; M_T2 = 4.99, SD_T2 = 1.64; t(421) = 5.65, p < .001, Cohen’s d = .27). However, there was an average decrease in support for social change (M_T1 = 6.36, SD_T1 = .83; M_T2 = 6.27, SD_T2 = .99; t(421) = –2.37, p = .018, Cohen’s d = .11). It is important to note that support for social change was strongly skewed, such that most participants reported very high scores. This would suggest that the relative decrease in support for social change still reflected strong levels of support for social change.

Media consumption predicting movement attitudes as a function of government support

Our main analyses tested whether beyond habitual mainstream and alternative media consumption, as well as people’s movement attitudes before the Bersih 5 protest occurred (at Time 1), mainstream and alternative media consumption for information about the Bersih 5 protest would predict subsequent movement attitudes after the protests occurred (at Time 2). In addition, we also tested the extent to which protest-specific media consumption predicted movement attitudes (both reported at Time 2) as a function of people’s pre-existing support for the government (reported at Time 1).⁵ See Figure 1 for a visualization of the interaction effects reported below.

Figure 1.

Simple slopes of media consumption around the Bersih 5 protests and movement attitudes at different levels of government support. Analyses controls for habitual media consumption and pre-existing movement attitudes at Time 1.

Movement identity

As shown in Table 3, inconsistent with Hypothesis 1, mainstream media consumption for protest-specific information (measured at Time 2) did not significantly predict movement identity at Time 2 (b = –.06, p = .061). However, consistent with Hypothesis 2, alternative media consumption for protest-specific information (measured at Time 2) significantly predicted a more supportive movement identity at Time 2 (b = .25, p < .001). These effects were qualified by a significant interaction between protest-specific mainstream media consumption and government support (b = –.08, p < .001). In line with Hypothesis 3a, simple slopes analyses revealed that mainstream media consumption predicted a less supportive movement identity when (Time 1) government support was high (b = –.13, p < .001), but not when government support was low (b = .02, p = .542). There was also a marginally significant interaction between protest-specific alternative media and government support (b = –.06, p = .051). In line with Hypothesis 3b, simple slopes analyses revealed that the strength of relationship between protest-specific alternative media and a supportive movement identity was somewhat weaker when government support was low (b = .19, p < .001), compared with when government support was high (b = .31, p < .001). Thus, overall, there was a stronger effect of both mainstream and alternative media among government supporters. Importantly, among government supporters, consuming alternative media for coverage about the Bersih 5 protest had a strong positive effect on subsequent support for the movement.

Table 3.

Media consumption behaviors predicting movement identity at Time 2 as a function of government support at Time 1 (Study 1).

Variable	b	SE	t	p	η²
Movement identity_T1	.59	.04	13.29	<.001	.303
Mainstream media_T1	−.02	.03	−0.52	.599	.001
Alternative media_T1	−.01	.05	−0.12	.908	.000
Government support_T1	−.06	.14	−0.46	.647	.001
Mainstream media_T2	−.05	.03	−1.88	.061	.009
Alternative media_T2	.25	.04	6.24	<.001	.088
Mainstream media_T2 × Government support_T1	−.08	.02	−3.50	<.001	.029
Slope of mainstream media_T2 when government support_T1 is high	−.13	.04	−3.63	<.001
Slope of mainstream media_T2 when government support_T1 is low	.02	.04	0.61	.542
Alternative media_T2 × Government support_T1	.06	.03	1.96	.051	.009
Slope of alternative media_T2 when government support_T1 is high	.31	.05	6.41	<.001
Slope of alternative media_T2 when government support_T1 is low	.19	.05	3.70	<.001

Model statistics: F(8, 406) = 52.12, p < .001, R² = .507.

Movement efficacy

As shown in Table 4, consistent with Hypothesis 1, mainstream media consumption for protest-specific information predicted lower movement efficacy at Time 2 (b = –.11, p = .011). Also consistent with Hypothesis 2, alternative media consumption for protest-specific information predicted greater movement efficacy at Time 2 (b = .29, p < .001). These effects were qualified by a significant interaction between pre-existing government support and protest-specific mainstream media consumption (b = –.09, p = .005). In line with Hypothesis 3a, simple slopes analyses revealed that protest-specific mainstream media consumption predicted lower movement efficacy when government support was high (b = –.20, p < .001), but not when government support was low (b = –.02, p = .731). There was also a significant interaction between pre-existing government support and protest-specific alternative media (b = .13, p = .003). In line with Hypothesis 3b, simple slopes analyses revealed that protest-specific alternative media consumption predicted greater movement efficacy when government support was high (b = .41, p < .001), and to a lesser extent, when government support was low (b = .16, p = .033). Thus, similar to the findings for movement identity, consuming alternative media coverage of the Bersih 5 protest predicted higher perceptions of movement efficacy after the protests, particularly among people who had high levels of support for the government.

Table 4.

Media consumption behaviors predicting movement efficacy at Time 2 as a function of government support at Time 1 (Study 1).

Variable	b	SE	t	p	η²
Movement efficacy_T1	.62	.04	16.98	<.001	.414
Mainstream media_T1	.03	.05	0.58	.565	.001
Alternative media_T1	−.02	.07	−0.29	.771	.000
Government support_T1	−.39	.20	−1.98	.048	.010
Mainstream media_T2	−.11	.04	−2.57	.011	.016
Alternative media_T2	.29	.06	5.01	<.001	.058
Mainstream media_T2 × Government support_T1	−.09	.03	−2.81	.005	.019
Slope of mainstream media_T2 when government support_T1 is high	−.20	.05	−3.77	<.001
Slope of mainstream media_T2 when government support_T1 is low	−.02	.05	−0.34	.731
Alternative media_T2 × Government support_T1	.13	.04	3.03	.003	.022
Slope of alternative media_T2 when government support_T1 is high	.41	.07	6.01	<.001
Slope of alternative media_T2 when government support_T1 is low	.16	.07	2.13	.033

Model statistics: F(8, 408) = 71.47, p < .001, R² = .584.

Support for social change

As shown in Table 5, we found that consistent with Hypothesis 1, mainstream media consumption for protest-specific information predicted lower support for social change at Time 2 (b = –.09, p = .002), whereas consistent with Hypothesis 2, alternative media consumption for protest-specific information predicted greater support for social change at Time 2 (b = .24, p < .001). There was no significant interaction between pre-existing government support and protest-specific mainstream media consumption (b = –.02, p = .260). Nevertheless, in line with Hypothesis 3a, simple slopes analyses revealed that mainstream media consumption predicted lower support for social change when government support was high (b = –.11, p = .001) but not when government support was low (b = –.06, p = .069). There was a significant interaction between pre-existing government support and protest-specific alternative media consumption (b = .07, p = .017). In line with Hypothesis 3b, simple slopes analyses revealed that protest-specific alternative media consumption predicted greater support for social change when government support was high (b = .31, p < .001), and to a lesser extent when government support was low (b = .17, p < .001). Thus, consumption of alternative media coverage of the Bersih 5 had a particularly strong effect on people relatively high on support for the government.

Table 5.

Media consumption behaviors predicting support for social change at Time 2 as a function of government support at Time 1 (Study 1).

Variable	b	SE	t	p	η²
Support for social change_T1	.58	.05	12.20	<.001	.267
Mainstream media_T1	.00	.03	0.12	.907	.000
Alternative media_T1	−.04	.04	−0.97	.331	.002
Government support_T1	−.35	.13	−2.63	.009	.017
Mainstream media_T2	−.09	.03	−3.15	.002	.024
Alternative media_T2	.24	.04	6.38	<.001	.091
Mainstream media_T2 × Government support_T1	−.02	.02	−1.13	.260	.003
Slope of mainstream media_T2 when government support_T1 is high	−.11	.03	−3.21	.001
Slope of mainstream media_T2 when government support_T1 is low	−.06	.04	−1.83	.068
Alternative media_T2 × Government support_T1	.07	.03	2.39	.017	.014
Slope of alternative media_T2 when government support_T1 is high	.31	.05	6.75	<.001
Slope of alternative media_T2 when government support_T1 is low	.17	.05	3.52	<.001

Model statistics: F(8, 408) = 48.17, p < .001, R² = .486.

Discussion

Study 1 examined how media consumption of mainstream and alternative outlets about the Bersih 5 protest predicted public attitudes toward the Bersih movement. As expected, we found divergent patterns of media consumption on movement attitudes. Whereas consuming mainstream media was related to less positive movement attitudes, consuming alternative media was related to more positive movement attitudes. Movement attitudes were measured in a variety of ways—identification with the movement, perceptions of movement efficacy, and support for social change. These findings were obtained over and above people’s habitual consumption of mainstream and alternative media, as well as their pre-existing attitudes toward the movement (all reported at Time 1, before the Bersih 5 protest occurred). Furthermore, and particularly important for our purposes, we found that the positive link between alternative media consumption and positive attitudes toward the Bersih movement was strongest among government supporters. This suggests that since alternative media portrays movements for social change in a more positive light, alternative media consumption may have played a role in legitimizing public attitudes toward the Bersih movement, especially among people who were likely antagonistic toward the movement from the outset.

In Study 2, we aimed to replicate and extend the findings from Study 1 around another major political event in Malaysia: the 14th General Elections. Although we obviously did not know it when we conducted the first wave of data collection, these elections turned out to be a turning point in Malaysian history because the ruling government regime lost its political power for the first time, giving rise to a new government led by the pro-reform coalition, Pakatan Harapan. This setting allowed us to assess how the polarized outcomes of alternative versus mainstream media consumption on movement attitudes will play out in the backdrop of a real-world political transformation. Elections are also a time in which the public largely relies on mass media to obtain information about the election, both before and after the elections. Thus, this high-stakes political situation provided another ecologically valid and robust test of our hypotheses.

Study 2

Study 2 was conducted around the 14th General Elections in Malaysia which took place on 9 May 2018. This election was particularly important in Malaysia because it was the first time that the BN—which had been governing the country since independence—lost to the opposition coalition, Pakatan Harapan. For many activists and scholars, this outcome reflected the power of the people in peacefully voting out a corrupt government through democratic means (Chan, 2018). The election also in part reflected the success of the Bersih movement in raising public awareness of government corruption and the need for political change. Although we could of course not know the outcome of the election before it occurred, we designed a study (similar to Study 1) with a longitudinal design centered around the date of the election. The Time 1 data were collected 1 week before the elections (23 April–30 April 2018), and Time 2 data collection started the day after the elections and continued for almost 3 weeks (10 May–29 May 2018). We assessed the link between media consumption of election-specific information and attitudes toward the Bersih movement. We also examined whether the outcomes of media consumption would differ based on people’s pre-existing attitudes toward the outgoing government.

Participants and procedure

A convenience sample of Malaysians (separate from the Study 1 sample) were recruited to complete online surveys via TurkPrime. At Time 1, 646 participants completed the survey, and of those, 386 participants returned at Time 2.⁶ Demographic characteristics of the sample are reported in Table 6, along with the national population estimates for comparison. Descriptively speaking, compared with the general population of Malaysia, our sample on average had higher levels of education. We also sampled more Chinese Malaysians and as a result, our sample had more Christians compared with the national population.⁷ Measures relevant to the present study were included as part of a larger survey on Malaysians’ attitudes around the elections.⁸ Variables measured with more than one item were averaged to obtain a composite score. The data, analysis code, and complete study materials are available on the Open Science Framework (OSF): https://osf.io/q3264/

Table 6.

Demographic information of Study 2 sample (N = 386), and demographics of the national population for comparison.

Criteria	Study 2 sample	National population
Sex ratio (ratio of males to females)	147:100	102:100 (2015−2019)
Median age	35.03 years (range: 18–71 years)	28.6 years (range: 0−65+; in 2018)
Median household income level per month	RM4000−RM6000 (22.86%)	RM5228 (in 2016)
Median education level	Graduated from college or university (50.78%)	Gross graduation ratio of tertiary programs: 23.35% (in 2017)
Distribution of three primary races
Bumiputera (includes Malay and Indigenous groups)	63.99% (59.33% Malay)	69.1% (in 2018)
Chinese	31.61%	23.0% (in 2018)
Indian	4.40%	6.9% (in 2018)
Distribution of four primary religions
Islam	56.22%	61.3% (in 2010)
Buddhism	17.10%	19.8% (in 2010)
Christianity	13.47%	9.2% (in 2010)
Hinduism	2.59%	6.3% (in 2010)

National population estimates are taken from the Department of Statistics Malaysia and the World Bank. When estimates are not available for 2018, the closest available estimate is used.

Measures

Measured only at Time 1

Habitual media consumption

Participants reported how much they generally paid attention to five different media sources, on a scale from 1 (no attention at all) to 7 (a lot of attention). Mainstream media was measured using the following two items: “Mainstream news reports (e.g. Berita Harian, The Star, Sin Chew Daily, Tamil Nesan)” and “Public TV channels (e.g. TV3, NTV7, 8TV, Bernama TV)” (r = .80, p < .001). Alternative media was measured using the following four items: “Alternative news reports (e.g. Malaysiakini, Free Malaysia Today, Aliran),” “Private TV channels (e.g. Astro Awani),” “International news (e.g. CNN, BBC news, New York Times),” and “Social media or social networking platforms (e.g. Facebook, Twitter, Reddit, Youtube, Whatsapp)” (α = .78).⁹

Government support

Participants reported their attitudes toward the ruling coalition (which, at the time was the BN government) on a scale from 1 (do not support at all) to 7 (very strongly support).

Measured only at Time 2

Election-specific media consumption

Participants reported how much they paid attention to the same mainstream (two items; r = .71, p < .001) and alternative media sources (four items; α = .71) used to measure habitual media consumption at Time 1, but this time for information about the recent general elections, on a scale from 1 (no attention at all) to 7 (a lot of attention).

Measured at Time 1 and Time 2

Movement identity

Participants responded to the same item used in Study 1, and three additional items (e.g. “I feel committed to the goals pursued by the Bersih movement”) on a scale from 1 (strongly disagree) to 7 (strongly agree) (α_T1 = .92; α_T2 = .92).

Movement efficacy

Participants responded to the same three items from Study 1 (α_T1 = .95; α_T2 = .96).

Collective action intentions

Participants reported their willingness to engage in five actions related to institutional reforms (e.g. “Sign petitions”) on a scale from 1 (not at all willing) to 7 (very willing) (α_T1 = .90; α_T2 = .87).

Results

Table 7 displays descriptive statistics on key variables. As seen in Table 7, mainstream media consumption was linked to more government support, and alternative media was related to more government support at Time 1 but not at Time 2. This pattern of correlations suggests that media consumption of alternative media around the elections was not necessarily skewed by pre-existing attitudes toward the government. In line with this, when we examined the distribution of media consumption scores for participants at different levels of pre-existing government support, we found that participants at both high and low levels of pre-existing government support consumed relatively high levels of alternative media (means above 4 on a scale from 1 to 7).¹⁰ Further, we found that mainstream media consumption was largely uncorrelated with movement attitudes and sometimes even negatively correlated, whereas alternative media was consistently positively correlated with all measures of movement attitudes.

Table 7.

Means, standard deviations, and bivariate correlations between media consumption and attitudes in Study 2.

Variable	M (SD)	Mainstream media_T1	Alternative media_T1	Mainstream media_T2	Alternative media_T2
Government support_T1	3.63 (1.86)	.45***	.11*	.35***	.03
Movement identity_T1	4.20 (1.24)	−.10*	.38***	−.11*	.26***
Movement efficacy_T1	4.30 (1.64)	−.03	.40***	−.12*	.27***
Collective action intentions_T1	4.42 (1.52)	.03	.49***	−.09	.30***
Movement identity_T2	4.48 (1.23)	−.07	.34***	.02	.42***
Movement efficacy_T2	4.87 (1.53)	−.04	.30***	.05	.44***
Collective action intentions_T2	4.68 (1.34)	−.01	.37***	.05	.49***

p < .05, *** p < .001.

Average changes over time

First, as we did in Study 1, we assessed whether there were changes in average levels of movement attitudes. Overall, we found that movement attitudes toward Bersih were more positive after the elections. Specifically, there was an increase in movement identification (M_T1 = 4.20, SD_T1 = 1.24; M_T2 = 4.48, SD_T2 = 1.23; t(385) = 5.62, p < .001, Cohen’s d = .28), movement efficacy (M_T1 = 4.30, SD_T1 = 1.64; M_T2 = 4.87, SD_T2 = 1.53; t(385) = 7.80, p < .001, Cohen’s d = .12), and collective action intentions (M_T1 = 4.42, SD_T1 = 1.52; M_T2 = 4.68, SD_T2 = 1.34; t(385) = 4.25, p < .001, Cohen’s d = .21).

Media consumption predicting movement attitudes as a function of government support

Mirroring our analyses from Study 1, our main analyses for Study 2 tested whether beyond habitual mainstream and alternative media consumption as well as people’s movement attitudes before the elections occurred (all measured at Time 1), mainstream and alternative media consumption of information about the elections would predict subsequent movement attitudes after the elections occurred (at Time 2). In addition, we also tested whether the extent to which election-specific media consumption predicted movement attitudes (reported at Time 2) would differ as a function of people’s pre-existing support for the prior government (reported at Time 1)—that is, support for the previous ruling government which was subsequently defeated in the elections.¹¹ See Figure 2 for a visualization of the interaction effects.

Figure 2.

Simple slopes of media consumption around the 14th General Elections and movement attitudes at different levels of government support. Analyses controls for habitual media consumption and pre-existing movement attitudes at Time 1.

Movement identity

As shown in Table 8, inconsistent with Hypothesis 1, mainstream media consumption for election-specific information did not significantly predict movement identity at Time 2 (b = .01, p = .673), whereas consistent with Hypothesis 2, alternative media consumption for election-specific information significantly predicted a more supportive movement identity at Time 2 (b = .28, p < .001). There was no significant interaction between pre-existing government support and election-specific mainstream media (b = –.04, p = .099). Contrary to Hypothesis 3a, simple slopes analyses revealed that mainstream media consumption did not significantly predict movement identity when government support was high (b = –.03, p = .538), nor when government support was low (b = .06, p = .157). However, there was a significant interaction between pre-existing government support and election-specific alternative media (b = .12, p < .001). In line with Hypothesis 3b, simple slopes analyses revealed that alternative media consumption predicted a more supportive movement identity when government support was high (b = .40, p < .001), and to a lesser extent when government support was low (b = .16, p = .010). Thus, consuming alternative media about the election had a particularly strong positive effect for movement identity among people who were relatively high in support for the (outgoing) government.

Table 8.

Media consumption behaviors of election-specific information predicting movement identity at Time 2 as a function of government support at Time 1 (Study 2).

Variable	b	SE	t	p	η²
Movement identity_T1	.61	.04	15.29	<.001	.383
Mainstream media_T1	−.07	.03	−1.88	.060	.0097
Alternative media_T1	.02	.05	0.39	.699	.000
Government support_T1	−.42	.21	−2.03	.043	.011
Mainstream media_T2	.01	.03	0.40	.688	.000
Alternative media_T2	.28	.05	5.85	<.001	.083
Mainstream media_T2 × Government support_T1	−.04	.03	−1.65	.099	.007
Slope of mainstream media_T2 when government support_T1 is high	−.03	.04	−0.66	.508
Slope of mainstream media_T2 when government support_T1 is low	.06	.04	1.44	.152
Alternative media_T2 × Government support_T1	.13	.04	3.34	<.001	.029
Slope of alternative media_T2 when government support_T1 is high	.41	.06	6.68	<.001
Slope of alternative media_T2 when government support_T1 is low	.15	.06	2.48	.014

Model statistics: F(8, 377) = 57.35, p < .001, R² = .549.

Movement efficacy

As shown in Table 9, inconsistent with Hypothesis 1, mainstream media consumption for election-specific information did not significantly predict movement efficacy at Time 2 (b = .04, p = .395), whereas consistent with Hypothesis 2, alternative media consumption for election-specific information predicted greater movement efficacy at Time 2 (b = .42, p < .001). There was no significant interaction between pre-existing government support and election-specific mainstream media consumption (b = –.05, p = .129). Contrary to Hypothesis 3a, simple slopes analyses revealed that mainstream media consumption did not significantly predict movement efficacy when government support was high (b = –.02, p = .802), nor when government support was low (b = .09, p = .085). There was also not a significant interaction between pre-existing government support and election-specific alternative media consumption (b = .09, p = .066). Contrary to Hypothesis 3b, simple slopes analyses revealed that election-specific alternative media consumption predicted greater movement efficacy when government support was high (b = .52, p < .001), and also when government support was low (b = .32, p < .001). Thus, consumption of alternative media about the election had a positive effect on movement efficacy, both among people high and low in support for the (outgoing) government.

Table 9.

Media consumption behaviors of election-specific information predicting movement efficacy at Time 2 as a function of government support at Time 1 (Study 2).

Variable	b	SE	t	p	η²
Movement efficacy_T1	.49	.04	11.86	<.001	.272
Mainstream media_T1	−.07	.05	−1.46	.146	.006
Alternative media_T1	−.02	.06	−0.38	.703	.000
Government support_T1	−.28	.28	−0.99	.322	.003
Mainstream media_T2	.04	.05	0.85	.395	.002
Alternative media_T2	.42	.07	6.31	<.001	.096
Mainstream media_T2 × Government support_T1	−.06	.04	−1.52	.129	.006
Slope of mainstream media_T2 when government support_T1 is high	−.02	.06	−0.25	.802
Slope of mainstream media_T2 when government support_T1 is low	.09	.05	1.73	.085
Alternative media_T2 × Government support_T1	.10	.05	1.84	.066	.009
Slope of alternative media_T2 when government support_T1 is high	.52	.08	6.12	<.001
Slope of alternative media_T2 when government support_T1 is low	.32	.09	3.77	<.001

Model statistics: F(8, 377) = 37.50, p < .001, R² = .443.

Collective action intentions

As shown in Table 10, inconsistent with Hypothesis 1, mainstream media consumption for election-specific information did not significantly predict collective action intentions at Time 2 (b = .03, p = .491), whereas consistent with Hypothesis 2, alternative media consumption for election-specific information predicted greater collective action intentions at Time 2 (b = .40, p < .001). There was no significant interaction between pre-existing government support and election-specific mainstream media consumption (b = .01, p = .790). Contrary to Hypothesis 3a, simple slopes analyses revealed that mainstream media consumption did not significantly predict collective action intentions when government support was high (b = .03, p = .507) nor when government support was low (b = .02, p = .684). There was also no significant interaction between pre-existing government support and election-specific alternative media consumption (b = .02, p = .586). Contrary to Hypothesis 3b, simple slopes analyses revealed that protest-specific alternative media consumption predicted greater collective action intentions when government support was high (b = .43, p < .001) and also when government support was low (b = .38, p < .001). Thus, consumption of alternative media about the election had a positive effect on collective action intentions, regardless of the level of support for the (outgoing) government.

Table 10.

Media consumption behaviors of election-specific information predicting collective action intentions at Time 2 as a function of government support at Time 1 (Study 2).

Variable	b	SE	t	p	η²
Collective action intentions_T1	.50	.04	13.17	<.001	.315
Mainstream media_T1	−.06	.04	−1.43	.154	.006
Alternative media_T1	−.06	.05	−1.14	.256	.003
Government support_T1	−.14	.23	−0.61	.544	.001
Mainstream media_T2	.03	.04	0.69	.491	.001
Alternative media_T2	.40	.05	7.38	<.001	.126
Mainstream media_T2 × Government support_T1	.01	.03	0.27	.790	.000
Slope of mainstream media_T2 when government support_T1 is high	.03	.05	0.66	.507
Slope of mainstream media_T2 when government support_T1 is low	.02	.04	0.41	.684
Alternative media_T2 × Government support_T1	.02	.04	0.54	.586	.001
Slope of alternative media_T2 when government support_T1 is high	.43	.07	6.14	<.001
Slope of alternative media_T2 when government support_T1 is low	.38	.07	5.41	<.001

Model statistics: F(8, 377) = 48.43, p < .001, R² = .507.

Discussion

Consistent with findings from Study 1 conducted around Bersih 5, Study 2 showed that in the context of the 14th General Elections (GE-14) in Malaysia, alternative media consumption of election-related information promoted more positive attitudes toward the Bersih movement, in terms of greater identification with the movement, stronger perceptions of movement efficacy, and higher intentions to engage in collective action for pro-democratic reforms. Also consistent with findings from Study 1, in Study 2, we found some evidence that government support moderated the effect of alternative media on movement attitudes such that the outcomes of alternative media were often strongest among supporters of the previous ruling government.

Interestingly, the outcomes of mainstream media did not replicate in Study 2. In Study 2, mainstream media consumption around the elections was not clearly linked to subsequent attitudes toward the Bersih movement. Whereas in Study 1, mainstream media consumption around the Bersih 5 protest was linked to lower perceptions of movement efficacy and lower support for social change, in Study 2, mainstream media consumption election-related information was not predictive of any measure of movement attitudes. Although it is difficult to interpret null findings, it is possible that the regime change observed during GE-14 in 2018 may have temporarily increased the ability for alternative media to challenge the influence of mainstream media. For example, in Tapsell’s (2020) analysis of GE-14, he noted that increased smartphone and Internet usage in Malaysia was important in circumventing the mainstream media’s discourse, further calling smartphones “weapons of the weak” (Scott, 1987) because of its ability to circulate alternative news that resisted predominant narratives favoring the previous ruling coalition. Scholars have also noted that the popularity of alternative media over the years helped set the stage for the historic election outcome in GE-14 (Chan, 2018; Nadzri, 2019). However, since Study 2 only captured a relatively small timeframe (i.e. weeks before and after GE-14), we revisit this point in the General Discussion to consider the possible long-term implications of a growing alternative media platform in Malaysia.

General discussion

One way social movements gain broader public support is through the media. However, media portrayal of social change movements tends to radically differ between mainstream and alternative media outlets. Whereas mainstream media coverage tends to depict movements challenging the status quo in a more negative light, alternative media coverage tends to be more sympathetic to, or even overtly supportive of, collective challenges to the status quo (e.g. Downing, 2003; McLeod and Hertog, 1992). In the present research, we moved beyond analyzing the content of media coverage of protests (which was largely the focus of prior research) and toward examining the perspective of the broader public who are consumers of this media. In particular, in the context of the Bersih electoral reform movement in Malaysia, we examined whether and how public consumption of both alternative and mainstream media shapes subsequent attitudes toward the Bersih movement.

We report findings from two panel survey studies conducted around two events that have been transformative in Malaysia’s recent political history. The first survey was conducted before and after the fifth mass street protest organized by the Bersih movement demanding electoral reform (Study 1). The second survey was conducted before and after the 14^th general elections in Malaysia (GE-14) which led to an unprecedented change in government (Study 2). Findings from both studies showed that alternative media consumption predicted more positive public attitudes toward the Bersih electoral reform movement, specifically in terms of greater movement identification, stronger beliefs in movement efficacy, and more support for social change (either by endorsing electoral reform in Study 1, or by expressing willingness to engage in collective action for electoral reforms in Study 2). We observed these outcomes after controlling for habitual media consumption of alternative and mainstream media, which means we were able to isolate the effects of event-specific media consumption around the Bersih 5 protests in Study 1 and around the general elections in Study 2. Furthermore, we found that alternative media consumption predicted more positive attitudes toward the Bersih movement, especially among supporters of the prior ruling government—who likely had the least favorable attitudes toward social change from the outset. Thus, our findings suggest that alternative media may have legitimized a pro-democracy movement during a period of major democratic transition in Malaysia.

While there were consistent positive links between alternative media consumption and movement attitudes, contrary to hypotheses, we did not find clear or consistent evidence that consuming mainstream media, even among government supporters, was linked to backlash toward the Bersih movement. Specifically, in Study 1, exposure to mainstream media news around the Bersih 5 protest predicted less favorable attitudes toward the Bersih movement, as expected. However, in Study 2, we unexpectedly found that consumption of mainstream media news around the general elections did not relate to attitudes toward the Bersih movement. This might reflect a weakening effect of mainstream media on Malaysian public attitudes over time; however, the political upheaval during the elections also makes it difficult to assess the outcome of mainstream media that was owned by the outgoing government. Nevertheless, the consistent outcomes of alternative media consumption on positive attitudes toward the movement suggests that alternative media may have played an important role in raising public support for a movement demanding democratic reforms. It is also interesting to note that regardless of pre-existing levels of support for the prior ruling government, we found that people consumed high levels of alternative media. By contrast, those who were more supportive of the prior government consistently reported higher levels of mainstream media consumption. This pattern of findings suggests that those who were supportive of the previous government may have been actively in the process of seeking information or perspectives that were lacking in mainstream media.

Contributions to research and theory

There have been recent calls to incorporate new forms of media in our understanding of social change (see McGarty et al., 2014). New media includes alternative outlets that tend to thrive online, where government control can be circumvented more effectively. In a similar way, alternative media can be used to spread narratives that counter the status quo, and has therefore been hailed as an important source for understanding civil resistance (Downing, 2003). The handful of studies examining outcomes of alternative media have found that the more people consumed alternative media, the more likely they were to participate in protests challenging the status quo (Boyle and Schmierbach, 2009; Chan, 2017). Relatedly, experimental work on outcomes of different media framings of political events has shown that subtle differences in media framings can sway subsequent political attitudes (McLeod and Detenber, 1999; Terkildsen and Schnell, 1997). By using two-wave longitudinal surveys, our approach builds on prior research by examining whether mainstream or alternative media frames around major political events in the real-world can shape subsequent public attitudes (measured at Time 2), while controlling for habitual consumption of mainstream and alternative media (measured at Time 1).

Furthermore, social psychological research has traditionally focused on the predictors of collective action, and less so on its outcomes (for a discussion, see Louis, 2009). Recent work has moved toward investigating how collective action can influence public opinion (e.g. Saguy and Szekeres, 2018; Selvanathan and Lickel, 2019). In the present contribution, we build on this literature by focusing on how media consumption might legitimize public attitudes to support a movement for social change. We highlight the importance of considering how broader institutional structures can constrain mobilization for social change. Our findings are also consistent with the observation that the advancement of information technology has promoted a more politically engaged society (e.g. Smeltzer and Lepawsky, 2010; Willnat et al., 2013). Among Malaysians today, online news portals and social media are major sources of information and provide avenues for political discussion (Salman et al., 2018). As a case in point, between 2016 and 2018, the percentage of Internet users in Malaysia increased from 76.9% to 87.4% (Malaysian Communications and Multimedia Commission, 2018). More generally, country-level Internet penetration has been linked to greater public demand for good governance (Khazaeli and Stockemer, 2013). In a multinational study of 28 countries including Malaysia, it was found that country-level Internet usage predicted greater support for democratization (Nisbet et al., 2012). Our research supplements the aforementioned findings by focusing on the general public’s lived experiences.

Applied and practical contributions

When mainstream media is monopolized by the state, alternative media may provide an effective platform for healthy expression of public criticism of the government and discourse around democratic reforms. Activists, non-governmental organizations, and grassroots organizers may be able to leverage alternative media outlets especially in young and growing democracies like Malaysia. More generally, our findings suggest that even though mainstream media may advance a biased view of social change efforts in favor of the ruling government, such media control may not always be effective in maintaining a regime’s grip on power, especially when competing perspectives are presented on alternative media (Chinnasamy and Griffiths, 2013). As Malaysians broadly consumed both mainstream and alternative media outlets regardless of political leanings (see Nain, 2018), in the long term, it is possible that alternative media consumption may narrow the gap between government supporters and dissenters in terms of their openness to social change.

Building cohesion and healthy debate across political divides is part and parcel of democratic ideals. However, so long as mainstream media is manipulated by a political regime, it is possible to think of alternative media itself as a tool of resistance, given that it challenges prevalent narratives of social change and the status quo depicted in mainstream media (see Downing, 2003). Although alternative news outlets may sometimes brand themselves as independent sources with a balanced news coverage, our findings suggest that alternative media functionally advances a critical social change perspective compared with mainstream media—not only in terms of the content of news coverage as past research has shown (e.g. McLeod and Hertog, 1992; Sanawi, 2014; Sani, 2014), but also in terms of news coverage shaping subsequent social attitudes of media consumers as our findings suggest.

The null findings of mainstream media consumption on public attitudes in the aftermath of GE-14 was surprising given that the literature thus far has shown how the mainstream media tends to be biased toward the previous ruling government (e.g. Manaf and Sedu, 2015; Sani, 2014). A transition in power during an election could mean that mainstream media coverage would subsequently become more balanced and there could be progress made toward media freedom. There were initial signs of this in Malaysia under the Pakatan Harapan government. For example, in September 2018, the Prime Minister vowed to reduce the stakes held by political parties in media companies (Tan, 2018), and in July 2019, the ownership of Media Prima—the largest media group in the country—changed from the UMNO political party to a Malaysian business tycoon (Teoh, 2019). Although more needs to be done in Malaysia (and other countries) to ensure press freedom, these small steps reflect a move away from politically controlled media, which could have downstream consequences on how mainstream media consumption shapes public opinion.

However, the future of media and democracy in Malaysia now remains uncertain. The progress toward democracy was thwarted because of an unprecedented political crisis and impasse that ensued in early 2020 (Ostwald, 2020). The Pakatan Harapan coalition that won the historical election in 2019 broke down, and there was a political coup which gave rise to a “backdoor government” that included many former politicians who were ousted during GE-14 (Head, 2020; Walden, 2020). Our findings, which were focused on important events in 2016 and 2018, demonstrate a clear link between alternative media consumption and social change attitudes during that time period. Given that the political situation has changed drastically in 2020, it is difficult to say whether the advances made toward press freedom over the past 2 years will continue into the future. But it is clear that alternative media will remain an outlet for critical voices to push for social change.

Limitations and future research directions

Since the present research utilized longitudinal two-wave surveys, the correlational nature of the data means we are unable to concretely say that alternative media exposure had a causal effect on public attitudes toward Bersih. There might be confounds or third variables involved. For example, people who consume alternative media might also have social networks of people who support reforms and democracy, which can exert a normative influence on their attitudes toward the Bersih movement. Based on prior research and theorizing, we favored the causal direction of media consumption promoting changes in public attitudes. However, our approach and findings do not preclude the possibility that changes in public attitudes can influence the decision to consume different forms of media. It is likely that the link between media consumption and public attitudes toward the movement is bidirectional: media framings can influence public attitudes, but public attitudes can also inform choices about which media to consume.

We also acknowledge the sampling limitations of our studies. Since we only had access to convenience samples of the Malaysian public, our findings are not necessarily generalizable to the entire population. As shown in Tables 1 and 6, our samples tended to have higher levels of education compared with the Malaysian population at large. Furthermore, since our methodology was limited to online surveys, it is likely that participants were Internet savvy and therefore more likely to seek alternative news outlets online compared with the general population. Nonetheless, it is important to note that the Internet is an increasingly important news source for many Malaysians. The Digital News Report in 2018 showed that 89% and 71% of Malaysians obtained their news online and on social media, respectively, compared with only 57% and 41% who said that their news is from television or print media (Nain, 2018). In this respect, our sampling strategy was likely able to reach urban-dwelling Internet users, and our findings could be limited to explaining the attitudes of this group. By design (i.e. use of online surveys), we unfortunately were not able to reach non-Internet users who tend to live in rural areas, have lower levels of education, and be older in age. Malaysians in rural areas are also commonly thought to be a stronghold of the former ruling government in Malaysia (e.g. see Rahman, 2018)—a group for whom alternative media exposure is likely to be low and therefore unlikely to be linked to favorable attitudes toward social change.

Although the present research applies media framing theory—which has focused on the content of news media—we did not directly assess the coverage of mainstream and alternative media outlets during Bersih 5 and the 14th General Elections. A more fine-grained approach such as time-series models would allow us to track attitude change over time and estimate the effect of media coverage of specific issues on shifts in public opinion. For example, in the context of Germany, Boomgaarden and Vliegenthart (2009) examined how frequency of news coverage on immigration issues was linked to macro-level fluctuations in anti-immigrant attitudes as reported in public opinion polls over a period of 12 years. A similar approach might be used to examine whether media coverage on issues of democracy and institutional reforms might correlate with changes in public opinion on social change across the span of a few years. Such a design would also allow researchers to identify subtle differences across different modes of media outlets (e.g. print, radio, online), which the present research overlooks because we aggregated across various news sources.

Concluding remarks

Long-term social change toward democratic ideals often involves shifting the attitudes of the broader public to be open to such changes. Since the media can shape social reality by framing political issues through a certain lens, the present research examined how mainstream and alternative media consumption is linked to public attitudes toward social change. Using longitudinal datasets that captured social change as it unfolded in Malaysia’s recent history, our results suggest the importance of alternative media outlets in promoting public support for dissent in a context in which mainstream media was routinely used by the ruling regime in the past to foster complicit support for the status quo. Our findings underscore the importance of free and accessible media in supporting democratic transitions powered by the people.

Supplemental Material

Online_supplementary_material_10_August_2020 – Supplemental material for How mainstream and alternative media shape public attitudes toward social change: Evidence from two panel studies during Malaysia’s democratic transition

Supplemental material, Online_supplementary_material_10_August_2020 for How mainstream and alternative media shape public attitudes toward social change: Evidence from two panel studies during Malaysia’s democratic transition by Hema Preya Selvanathan and Brian Lickel in New Media & Society

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by an internal grant from the University of Massachusetts, Amherst.

ORCID iD

Hema Preya Selvanathan

Supplemental material

Supplemental material for this article is available online.

Notes

Author biographies

Hema Preya Selvanathan received her PhD in social psychology with a concentration in the psychology of peace and violence from the University of Massachusetts Amherst, USA, in 2019. She is now a Postdoctoral Research Fellow at The University of Queensland, Australia.

Brian Lickel received his PhD in social psychology from the University of California Santa Barbara, USA in 2000. He is now a Professor of social psychology at the University of Massachusetts Amherst, USA, in the Psychology of Peace and Violence Program in the Department of Psychological and Brain Sciences.

References

Abbott

(2011) Electoral authoritarianism and the print media in Malaysia: measuring political bias and analyzing its cause. Asian Affairs: An American Review 38(1): 1–38.

Abbott

Givens

(2015) Strategic censorship in a hybrid authoritarian regime? Differential bias in Malaysia’s online and print media. Journal of East Asian Studies 15(3): 455–478.

Anuar

(2000) Malaysian media and democracy. Media Asia 27(4): 183–199.

Azman

(2018) Utusan shares plunge to five-year low after UMNO’s defeat at election battle. The Edge Markets, 14 May. Available at: https://www.theedgemarkets.com (accessed 14 May 2018).

Boomgaarden

Vliegenthart

(2009) How news content influences anti-immigration attitudes: Germany, 1993-2005. European Journal of Political Research 48(4): 516–542.

Borah

(2011) Conceptual issues in framing theory: a systematic examination of a decade’s literature. Journal of Communication 61(2): 246–263.

Boyle

Schmierbach

(2009) Media use and protest: the role of mainstream and alternative media use in predicting traditional and protest participation. Communication Quarterly 57(1): 1–17.

Boyle

McCluskey

Devanathan

, et al. (2004) The influence of level of deviance and protest type on coverage of social protest in Wisconsin from 1960 to 1999. Mass Communication and Society 7(1): 43–60.

Brown

(2005) The rough and rosy road: sites of contestation in Malaysia’s shackled media industry. Pacific Affairs 78(1): 39–56.

10.

Case

(2009) Electoral authoritarianism in Malaysia: trajectory shift. The Pacific Review 22(3): 311–333.

11.

Chan

(2018) Democratic breakthrough in Malaysia: political opportunities and the role of Bersih. Journal of Current Southeast Asian Affairs 37(3): 109–137.

12.

Chan

Lee

(1984) Journalistic “paradigms” of civil protests: a case study in Hong Kong. In: Arno

Dissanayake

(eds) The News Media in National and International Conflict. New York: Routledge, pp. 183–202.

13.

Chan

(2017) Media use and the social identity model of collective action: examining the roles of online alternative news and social media news. Journalism & Mass Communication Quarterly 94(3): 663–681.

14.

Chinnasamy

Griffiths

(2013) Looking back at Malaysia’s GE2008: an internet election and its democratic aftermath. The IAFOR Journal of Media, Communication and Film 1(1): 59–74.

15.

Chong

Druckman

(2007) Framing theory. Annual Review of Political Science 10: 103–126.

16.

Couldry

Curran

(2003) Contesting Media Power: Alternative Media in a Networked World. Lanham, MD: Rowman & Littlefield Publishers.

17.

Downing

JDH

(2003) Audiences and readers of alternative media: the absent lure of the virtually unknown. Media, Culture & Society 25(5): 625–645.

18.

Druckman

(2001) The implications of framing effects for citizen competence. Political Behavior 23(3): 225–256.

19.

Dunlap

McCright

(2008) Social movement identity: validating a measure of identification with the environmental movement. Social Science Quarterly 89(5): 1045–1065.

20.

Earl

Martin

McCarthy

, et al. (2004) The use of newspaper data in the study of collective action. Annual Review of Sociology 30: 65–80.

21.

Entman

(2007) Framing bias: media in the distribution of power. Journal of Communication 57(1): 163–173.

22.

George

(2007) Media in Malaysia: zone of contention. Democratization 14(5): 893–910.

23.

Head

(2020) How Malaysia’s government collapsed in two years. BBC News, 5 March. Available at: https://www.bbc.com/news (accessed 5 March 2020).

24.

Kee

Ismail

Hashim

, et al. (2013) Framing Bersih 3.0: online versus traditional mass communication. Procedia Technology 11: 161–168.

25.

Khazaeli

Stockemer

(2013) The Internet: a new route to good governance. International Political Science Review 34(5): 463–482.

26.

Khoo

(2014) The rise of constitutional patriotism in Malaysian civil society. Asian Studies Review 38(3): 325–344.

27.

Khoo

Selvanathan

Lickel

(in press) Rights consciousness in Malaysia: an analysis of the electoral reform movement and political efficacy. Asian Studies Review.

28.

Leung

Lee

(2014) Cultivating an active online counterpublic: examining usage and political impact of Internet alternative media. The International Journal of Press/Politics 19(3): 340–359.

29.

Louis

(2009) Collective action—and then what? Journal of Social Issues 65(4): 727–748.

30.

Lumsden

(2013) How independent? An analysis of GE13 coverage by Malaysia’s online news portal coverage. Jurnal Komunikasi: Malaysian Journal of Communication 29(2): 1–30.

31.

McGarty

Thomas

Lala

, et al. (2014) New technologies, new identities, and the growth of mass opposition in the Arab Spring. Political Psychology 35(6): 725–740.

32.

McLeod

(1995) Communicating deviance: the effects of television news coverage of social protest. Journal of Broadcasting & Electronic Media 39(1): 4–19.

33.

McLeod

Detenber

(1999) Framing effects of television news coverage of social protest. Journal of Communication 49(3): 3–23.

34.

McLeod

Hertog

(1992) The manufacture of “public opinion” by reporters: informal cues for public. Discourse Society 3(3): 259–275.

35.

Malaysian Communications and Multimedia Commission (2018) Internet Users Survey 2018. Available at: https://www.mcmc.gov.my/skmmgovmy/media/General/pdf/Internet-Users-Survey-2018-(Infographic).pdf (accessed 14 May 2020).

36.

Manaf

AMA

Sedu

(2015) Framing Islam-related issues during GE13: an analysis of Malaysian mainstream newspapers. Intellectual Discourse 23(1): 29–52.

37.

Means

(1996) Soft authoritarianism in Malaysia and Singapore. Journal of Democracy 7(4): 103–117.

38.

Nadzri

(2019) The 14th General Election, the fall of Barisan Nasional, and political development in Malaysia, 1957-2018. Journal of Current Southeast Asian Affairs 37(3): 139–171.

39.

Nain

(2018) Malaysia. In: Newman

Fletcher

Kalogeropoulos

, et al. (eds) Reuters Institute Digital News Report 2018. Oxford: Reuters Institute for the Study of Journalism, pp. 132–133. Available at: http://media.digitalnewsreport.org/wp-content/uploads/2018/06/digital-news-report-2018.pdf (accessed 14 May 2020).

40.

Nisbet

Stoycheff

Pearce

(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.

41.

Ostwald

(2020) Malaysia 2020: The Impasse of Two-Coalition Politics. ISEAS Yusof Ishak Institute. Available at: https://www.iseas.edu.sg/wp-content/uploads/2020/02/ISEAS_Perspective_2020_25.pdf (accessed 14 May 2020).

42.

Rahman

(2018) Malaysia’s General Elections 2018: Understanding the Rural Vote. ISEAS Yusof Ishak Institute. Available at: https://www.iseas.edu.sg/wp-content/uploads/pdfs/TRS9_18.pdf (accessed 14 May 2020).

43.

Rajaratnam

(2009) Role of traditional and online media in the 12th General Election, Malaysia. The Journal of the South East Asia Research Centre for Communications and Humanities 1(1): 33–58.

44.

Saguy

Szekeres

(2018) Changing minds via collective action: exposure to the 2017 Women’s March predicts decrease in (some) men’s gender system justification over time. Group Processes & Intergroup Relations 21(5): 678–689.

45.

Salman

Yusoff

Salleh

MAM

, et al. (2018) Political engagement on social media as antecedent for political support among voters in Malaysia. Jurnal Komunikasi: Malaysian Journal of Communication 34(2): 152–165.

46.

Sanawi

(2014) Ideologies underlying the use of photographs in newspaper coverage of Bersih 3.0 rally: a social semiotic analysis. Jurnal Komunikasi: Malaysian Journal of Communication 30(2): 41–60.

47.

Sani

MAM

(2014) Malaysia’s 13th General Election: political partisanship in the mainstream print media. Asia Pacific Media Educator 24(1): 61–75.

48.

Scheufele

(1999) Framing as a theory of media effects. Journal of Communication 49(1): 103–122.

49.

Scott

(1987) Weapons of the weak: everyday struggle, meaning and deeds. In: Shanin

(ed.) Peasants and Peasant Societies: Selected Readings. 2nd ed. Oxford: Blackwell Publishers, pp. 343–345.

50.

Selvanathan

Lickel

(2019) Empowerment and threat in response to mass protest shape public support for a social movement and social change: a panel study in the context of the Bersih movement in Malaysia. European Journal of Social Psychology 49(2): 230–243.

51.

Selvanathan

Lickel

Jetten

(in press) Collective psychological ownership and the rise of reactionary counter-movements defending the status quo. British Journal of Social Psychology.

52.

Smeltzer

(2008) Blogging in Malaysia: hope for a new democratic technology? Journal of International Communication 14(1): 28–45.

53.

Smeltzer

Lepawsky

(2010) Foregrounding technology over politics? Media framings of federal elections in Malaysia. Area 42(1): 86–95.

54.

Stroud

(2008) Media use and political predispositions: revisiting the concept of selective exposure. Political Behavior 30(3): 341–366.

55.

Tan

(2018) Mahathir moves to free Malaysian media from political parties. Nikkei Asian Review, 7 September. Available at: https://asia.nikkei.com/Politics/Malaysia-in-transition/Mahathir-moves-to-free-Malaysian-media-from-political-parties (accessed 7 September 2018).

56.

Tapsell

(2020) Malaysia in 2019: a change of government without regime change. Southeast Asian Affairs 2020(1): 190–208.

57.

Teoh

(2019) Malaysian tycoon Syed Mokhtar expands media empire with stake in Media Prima. The Straits Times, 3 July. Available at: https://www.straitstimes.com/asia/se-asia/malaysian-tycoon-syed-mokhtar-expands-media-empire-with-stake-in-media-prima (accessed 3 July 2019).

58.

Terkildsen

Schnell

(1997) How media frames move public opinion: an analysis of the women’s movement. Political Research Quarterly 50(4): 879–900.

59.

Walden

(2020) How did Malaysia’s Mahathir go from prime minister to being kicked out of his party in a matter of months? ABC News, 5 June. Available at: https://www.abc.net.au/news/2020-06-05/malaysia-mahathir-prime-minister-kicked-out-bersatu-explainer/12307442 (accessed 5 June 2020)

60.

Willnat

Wong

Tamam

, et al. (2013) Online media and political participation: the case of Malaysia. Mass Communication and Society 16(4): 557–585.

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

0.29 MB