Patterns in Advocacy Group Portrayal

Abstract

This article compares features of protest and non-protest news items across three types of advocacy groups. More specifically, it tests differences in news item duration (length), prominence (lead item), standing (direct quotation), and source selection (balance). Analyzing 2,845 news items of 17 Belgian advocacy groups, this study shows that across all advocacy groups, protest items are less frequently balanced and significantly shorter than non-protest items. Differences between protest and non-protest items are small for unions, detrimental for environmental groups, and beneficial for peace organizations. Discussion centers on the (un)conditionality of the protest paradigm and its implications for advocacy groups seeking social change.

Keywords

advocacy groups television news coverage protest paradigm

Introduction

News media coverage is an important political resource for advocacy groups (Andrews & Caren, 2010; Vliegenthart, Oegema, & Klandermans, 2005; Walgrave & Manssens, 2005). Typically, advocacy groups lack insider positions in the political system and need to go public with their claims (Gamson, 2004; Lipsky, 1968). By going public, groups aim to expand the scope of conflict, which, in turn, might cause existing power balances to shift (Gamson & Wolfsfeld, 1993). Eliciting media coverage is crucial in this process. The public strategies of advocacy groups are only considered effective when they succeed in generating a certain kind and amount of media coverage (Thrall, 2006).

However, gaining media coverage is easier said than done. The relationship between journalists and activists is one of asymmetrical dependency: Advocacy groups need media coverage to a far greater extent than journalists need their quotes (Gamson & Wolfsfeld, 1993). Besides facing great constraints in gaining media attention, the messages of advocacy groups in the news get easily distorted as well (Ryan, 1991). Exemplar in this respect is the protest paradigm. The protest paradigm holds that coverage of protests follows an implicit template that undermines group goals and tends to marginalize, criminalize, and even demonize protestors (McLeod & Hertog, 1998). In sum, it appears that advocacy groups are caught in a catch-22. They need media coverage to be successful, yet they have to fight an uphill battle to make the news. Protest strategies are considered the most surefire way to access the media arena, yet by protesting, advocacy groups seem to do themselves more harm than good (Danielian & Page, 1994).

This study aims to contribute to the growing body of literature on the relationship between advocacy groups and media coverage by comparing features of protest and non-protest items across three types of advocacy groups. More specifically, it tests whether the following news item attributes vary: source selection (balance), duration (item length), prominence (lead item), and standing (group quotation). Two questions take the center stage. First, are there any systematic differences across protest and non-protest items? Second, are these differences generic, or do they apply more to some types of advocacy groups than to others? By comparing protest to non-protest items, this study adds a baseline to the protest paradigm. Is the portrayal of the very same advocacy groups different if one compares items that feature groups when they are protesting with items in which they are not? Such a comparison permits to determine whether it is the tactic of protest that makes a difference or whether aspects of the protest paradigm can be applied to coverage of advocacy groups more generally. Second, by comparing differences across types of advocacy groups, this study further explores the conditionality of the paradigm. Can we speak of robust laws when it comes to protest coverage, or do coverage patterns vary across types of groups?

To answer these questions, this study draws on a large census data set of television news items spanning an 8-year period (2003-2010). All news items that mention at least 1 of the 17 selected advocacy groups (labor, environmental, and peace organizations) belong to the research population. The setting of this study is Belgium. Belgium is a neo-corporatist country with a strong civil society and with unions strongly embedded in the policy-making process (Lijphart, 1999). The Belgian television news market is a textbook example of a duopoly situation with strong convergence between the only commercial and the public news provider (Hooghe, De Swert, & Walgrave, 2007). Taken together, this article contributes to knowledge on the media packaging of advocacy groups and the (un)conditionality of the protest paradigm.

Literature Review

This section proceeds as follows. First, I present literature on the protest paradigm, specifically focusing on its (un)conditionality. Next, I introduce the different types of groups under study and make assumptions about their news-making capability. Finally, the four news item attributes—which are the dependent variables of this study—are presented together with the hypotheses.

The (Un)Conditionality of the Protest Paradigm

Numerous studies have established that protest coverage is subjected to a “protest paradigm” (Boyle, McCluskey, McLeod, & Stein, 2005; Chan & Lee, 1984; Di Cicco, 2010; McLeod & Hertog, 1998). Protest paradigm studies show that journalists cover protests following a routinized pattern. This fixed, implicit script tends to marginalize, criminalize, and even demonize protestors. Media effect studies show how even subtle differences in media portrayal of protests can affect viewers’ perceptions. The more closely journalists follow the paradigm, the more critical the audience will be (Detenber, Gotlieb, McLeod, & Malinkina, 2007; McLeod, 1995; Nelson, Clawson, & Oxley, 1997).

Characteristic of the paradigm is a focus on the act of dissent, instead of a focus on the issue that fueled the action in the first place. Protest coverage emphasizes the action, rather than the grievance. Devoid of such issue context, protest appears senseless (Smith, McCarthy, McPhail, & Augustyn, 2001). Furthermore, protest reports have a tendency to rely on official, elite sources, instead of direct activist quotations. Elite sources add prestige to a story, maintain the illusion of objectivity, and increase the efficiency of news production. As official viewpoints predominate, protest coverage reinforces the status quo (McLeod & Hertog, 1998). Similarly, different ways of depicting public opinion—via bystander portrayals, polls, or sweeping generalizations—maintain social control (McLeod & Hertog, 1992). A final aspect of the paradigm deals with the dominance of negative frames over sympathetic or balanced frames. News coverage tends to frame protests as if they are crimes, carnivals, freak shows, or romper rooms. These negative frames accentuate the deviance of the protestors from the general population (Detenber et al., 2007). In all, coverage of protest that adheres to the protest paradigm undermines a group’s agenda and protects the status quo (McLeod & Detenber, 1999).

Whereas early research established the paradigm and mapped its characteristics, recent research has started to study the conditions that determine how closely the media follow the paradigm. Protest that is more radical is found to be covered in a more critical manner (Boyle, McLeod, & Armstrong, 2012). Group tactics (how activists behave) are particularly important when it comes to mass media’s adherence to the paradigm. Group goals (type of change sought) matter for a far lesser extent (Boyle et al., 2012). Differences were also found between protest topics. Media coverage of war, labor, and social protests differed across and within particular time periods (Boyle, McCluskey, Devanathan, Stein, & McLeod, 2004). Also context matters. A cross-national comparison of protest against the war in Iraq revealed that the U.S. press invoked the paradigm to a greater extent than the U.K. press, reflecting sociopolitical differences between both countries (Dardis, 2006). Finally, the medium matters as well; mainstream press more closely mirrors protest paradigm characteristics compared with alternative press outlets (McLeod & Hertog, 1992). Also the ideological leaning of a medium matters (Weaver & Scacco, 2013).

Besides teasing out the conditionality of the protest paradigm, other studies have begun to add new aspects to the paradigm. A longitudinal study by Di Cicco, for instance, speaks of a “public nuisance” paradigm and adds references to protest as bothersome, impotent, and unpatriotic to the protest paradigm repertoire (Di Cicco, 2010).

In sum, recent studies have started to answer the call of the paradigms’ founding fathers to “continue to specify the characteristics” of the paradigm and to look into the “conditions that regulate how closely media follow” the protest paradigm (McLeod & Hertog, 1998). This article follows in the footsteps of these studies. First, it compares media portrayals of advocacy groups across protest and non-protest items and as such adds a baseline to the protest paradigm. With such a baseline of non-protest coverage, it can be established whether aspects of protest coverage account for coverage of advocacy groups more in general. If the protest paradigm script extends beyond protest coverage, the advocacy groups’ position is even more bleak than previously acknowledged. Second, this study compares three very different advocacy group families. This allows exploration of the extent to which the protest paradigm truly is a fixed script. If findings are stable irrespective of the groups that stage the protest, the paradigm is a robust template. If the paradigm is flexible, however, there are more opportunities for advocacy groups to challenge the status quo: by mimicking the characteristics, adopting the strategies, or understanding the conditions under which particular advocacy group families trigger more favorable coverage, groups can come to understand the factors that guide the role of mass media in hindering or facilitating social change. Finally, with duration, standing, prominence, and source selection, this study adds several general and relatively unexplored characteristics to the protest paradigm repertoire.

Advocacy Group Families

Three types or families of advocacy groups are dealt with in this study: labor, environmental, and peace organizations. These advocacy group families were chosen because they (a) regularly stage protest events and (b) have very different organizational traits known to affect media portrayal. Selecting such different group families presents a tough test for the generalizability of the results. If the differences between protest and non-protest items hold across very diverse types of advocacy group families and in another media, political, and historical context than previous studies on the protest paradigm, chances are likely that the established patterns represent generic mechanisms of protest coverage.

Labor organizations or unions are advocacy groups that protect workers’ rights. In Belgium, unions are present in many government advisory boards. They even have real implementation power as they provide services that would otherwise be provided by the state (Martens, Van Gyes, & Van der Hallen, 2002). In other words, in neo-corporatist Belgium, unions are insiders of the political system. Their political power makes them highly newsworthy (Wouters, Hooghe, De Smedt, & Walgrave, 2011). Belgian unions have a large membership base, and if “insider” negotiations fail, they mobilize their constituents to hold large-scale protests (Martens et al., 2002). A strong and unique weapon of unions is the strike, which is frequently used (Devos & Humblet, 2007). In sum, unions are rich in terms of resources and members, have a strong insider status, can be seen as part of the system and, as such, rarely challenge, but, rather, try to preserve the status quo (Deacon, 1996).

The environmental organizations in the sample are professionalized and have a clear division of labor. Whereas some organizations foremost try to influence policy by doing research and by taking seats in advisory boards (Natuurpunt, Bond Beter Leefmilieu), other organizations have a more militant, activist profile and engage in disruptive or colorful and theatrical actions (Greenpeace, Friends of the Earth; Verhulst & Walgrave, 2005). Environmental organizations foremost have checkbook members whose engagement consists of monthly financial contributions. Therefore, their protest actions rarely are massive mobilizations but are foremost small-scale actions maximally adapted to a media logic and executed by professional activists (della Porta & Diani, 1999). McCluskey found that especially resource-poor, rather than professionalized, environmental groups succeeded in gaining positive coverage (McCluskey, 2008). Results of a study by Sobieraj point in a similar direction: Journalists especially expect activists to be authentic, emotional, and spontaneous (Sobieraj, 2010). Professionalization in this sense is detrimental for positive portrayal in protest coverage. Moreover, environmental organizations clearly work with annual campaigns. Their actions are staged to put these campaigns in the spotlight. As a consequence, environmental protests rarely tie in with the “news of the day” and are somewhat detached from current affairs.

Peace organizations are neither strongly organized nor professionalized in Belgium. They best fit the label of “grassroots” movements (Walgrave & Rucht, 2010). Yet peace organizations were omnipresent in the beginning of 2003, with the imminent war in Iraq. The issue of the Iraq war was high on the media and political agenda as Belgian ports were being used for weapon transports. Belgian political elites took clear anti-war stances, paralleling Belgian public opinion (Walgrave & Verhulst, 2009). Several large-scale marches were organized with approximately 60,000 participants taking to the streets of Brussels on February 15th, 2003; the worldwide day of action against the imminent war in Iraq (Walgrave & Rucht, 2010). In short, both the political as well as the discursive opportunity structures were favorable for the peace movement in 2003. This makes the peace movement an example par excellence of a movement confronted with a window of opportunity.

Taken together, the above review of three organizational families highlighted their different characteristics in terms of insider positions, professionalization, and “momentum.” In the next section, I relate these characteristics of the organizational families to features of the news items in which they appear, and formulate a number of expectations.

News Item Features

Previous studies on advocacy groups and media coverage analyzed media presence, media frames, or tone of coverage. This article looks at specific features or attributes of news items. Such attributes can easily be measured and compared across a wide range of items and a diverse set of issues. Although rarely tested, such attributes clearly fit the rationale of the protest paradigm as a template used by journalists, resulting in a particular type of news item.

Prominence

Is a news item the opening item of a newscast or not? The running order of items in newscasts is not random. A lead news item covers the most important occurrence of the day. In particular, lead news items, controlled for real-world cues, set the public agenda (Behr & Iyengar, 1985). For advocacy groups seeking social support, appearance in lead news items thus is particularly fruitful. Previous research established that especially large protest events result in lead news coverage (Wouters, 2013). Primarily, unions stage large events, as they can easily mobilize their large membership base. Also protests against the imminent war in Iraq resulted in large-scale mobilizations. I therefore expect that union and peace movement protests will be featured more up front in newscasts. However, because of the strong insider position of unions, and because of the perennial newsworthiness of economic issues (Deacon, 1996), protest is especially expected to make a difference for the peace movement. This leads to a first hypothesis:

H1: Protest items of peace organizations are featured more prominently than non-protest items of peace organizations.

Duration

Duration deals with the length of a news item. The longer a news item, the more the occurrence is considered newsworthy and the more information the news item can transmit. There is considerable variation in the length of news items. Some occurrences are treated in depth, whereas others are given only limited airtime. A well-known newscast feature is the so-called “news carousel”: A number of short news items follow each other, accompanied by up-tempo background music and a continuous voice-over (Wouters, 2013). Given the perennial newsworthiness of economic issues, the momentum of the peace protests, and the fact that protest actions of environmental groups primarily are organized around campaigns that do not necessarily align with the news of the day, I expect especially protest items of environmental organizations to get limited time slots in the news. This leads to a second hypothesis:

H2: Protest items of environmental organizations are especially shorter compared with non-protest items of environmental groups.

Standing

Having a voice in the media arena—that is, being directly quoted—is a crucial step to success (Gamson & Wolfsfeld, 1993). By having a voice, a media source can provide its own interpretation of an event or issue. Instead of merely being mentioned, standing offers media sources the opportunity to contribute substantively to an ongoing debate and points toward acceptance of a group and a certain degree of legitimacy (Ferree, Gamson, Gerhards, & Rucht, 2002). I expect standing to vary across groups. Because environmental organizations tend to protest more disruptively compared with the other organizations, the standing of environmental groups is expected to be lower when protesting. Given the public support for the peace protests, the standing of peace organizations in protest news items is expected to be higher. This leads to the following hypothesis:

H3: Compared with non-protest items, the standing of peace organizations in protest items is higher, and the standing of environmental organizations is lower.

Balance

A final news item feature deals with the source selection pattern in news items. Items can be balanced: A report then presents quotes from different (types of) sources. On the other hand, sources can have monopoly positions in news items: In this instance, a source gives his or her opinion uncontested by any other (type of) source. The balancing norm is an important professional journalistic norm (Entman, 2007). The protest paradigm holds that especially official sources are quoted in protest items, which leads to the expectation that protest items are more frequently unbalanced (McLeod & Hertog, 1998). Especially the official perspective would be presented. More recent evidence, however, shows the opposite pattern: protest group sources and not officials dominate protest coverage (Boyle et al., 2004). Taken together, both perspectives claim that protest items are rarely balanced, yet they disagree about whose voice dominates the item. Here, we add the baseline of non-protest items to the equation. If the presence of advocacy groups in non-protest items is considered to be foremost reactive, that is, in reaction to a statement or action of another actor, non-protest items by definition are more likely to be balanced and protest items are less likely balanced. This leads to a fourth hypothesis:

H4: Compared with non-protest items, protest items are less frequently balanced.

Overall, these hypotheses lead to a final overarching hypothesis. Because unions have insider positions and hence are part of the system, I expect little differences across protest and non-protest items for labor organizations. Because peace protests enjoyed momentum with the imminent war in Iraq, I expect peace protest coverage to be especially supportive. Environmental groups, finally, are in a less favorable position when protesting: The combination of disruptive and professional actions, detached from the news of the day, is expected to have undesirable consequences.

H5: Differences between protest and non-protest items are small for unions, beneficial for peace organizations, and detrimental for environmental organizations.

Method

Television news content spanning an 8-year period (2003-2010) was analyzed. Data were retrieved from the Electronic News Archive (ENA).¹ The ENA database is a census database: Every evening newscast of every single day of both the most important public (Eén) and commercial (vtm) Belgian television channels comprise the research population. These evening newscasts last about half an hour and can be considered the equivalents of the evening newscasts of the major U.S. broadcast networks. The ENA database is hosted by the University of Antwerp and the University of Leuven and is made available to the academic community in a standard SPSS file. Coding is done by a team of trained student coders. Newscasts are coded both at the news-item level (thematic content codes, domestic and foreign focus, news item length, position of item in newscast) and actor level (within a news item for every actor: actor name, actor function, actor quoted or not, actor speaking time). Inter-coder reliability of the database is tested by double coding of a constructed week of seven newscasts² and resulted in acceptable Krippendorff alpha values, all ranging between .748 and .997 (Hayes & Krippendorff, 2007).

For this article, which deals with national Belgian advocacy groups, only items in the ENA database that referred to Belgium (α = .919) were taken into account. In total, 17 advocacy groups were selected for scrutiny, belonging either to the labor (n = 3), environmental (n = 8), or peace movement (n = 6). The selected organizations are well-known organizations and appear among the most newsworthy organizations of their respective families. Although these organizations clearly are not representative of the entire population of advocacy groups of a particular family, they account for the lion’s share of media coverage for every group family. As such, by and large, they determine the public image of these families. All news items mentioning these advocacy groups were selected in the ENA database by searching for the full name and abbreviation/acronym of the organization in the ENA database. Protest items were distinguished from non-protest items by performing a word search in the database. The search term used the word “protest,” its synonyms, and different protest forms (e.g., march, blockade, strike) as keywords and automatically generated the protest/non-protest dummy variable.³

In total, the final data set consists of 2,849 advocacy group family appearances in 2,845 unique news items. Unions prove to be far more newsworthy (n = 2,329) than environmental organizations (n = 463) and peace organizations (n = 57). Generally, 4 out of 10 advocacy group items show protest (37.6%). Especially peace organizations make news by staging protest events (66.7%). Unions (39.1%) appear more frequently with protest actions in the news than environmental organizations (26.3%), although most of the protests staged by unions are strikes (26.5% of all union items, 68% of all union protest items). Further information about the descriptives of the selected organizations can be found in Appendix A.

A news item was considered prominent (α = .804) if it was the first item of the newscast of a particular day. Duration (α = .997) was operationalized as the number of seconds that the news item lasted. Standing (α = .761) is a dummy variable, which was coded 1 if at least one source of the advocacy group family was directly quoted in the news item and coded 0 if the advocacy group was merely mentioned. Balance, finally, was operationalized as whether at least one of the following sources was also present in the news item, irrespective of whether the source supported or disapproved of the advocacy group: a politician (α = .977), an expert (α = .748), or a civil servant (α = .866). Descriptives of the dependent variables can be found in Appendix B. The database is a census database, containing all of the news items that mentioned the selected advocacy groups. To test for differences between groups and protest items, t tests and two-way ANOVAs are used. As the database is not using any sample but, rather, contains all of the news items of the selected advocacy groups, absolute numbers and mean differences, irrespective of significance thresholds, tell most of the story. Significance levels are referred to as a matter of completeness and an indication of robustness.

Results

Prominence

Do advocacy groups more easily make lead item news by staging protests? H1 expected this to be the case especially for peace organizations. A t test shows that 11.1% of all protest items are lead news items. Only 8.3% of the non-protest items open a newscast, t(2028) = −2.409; p = .016. At the first glance, protests seem to lead to more prominent coverage. Results of a two-way ANOVA introducing advocacy group families as a factor, prove otherwise. Only the advocacy group family factor significantly explains lead news item coverage, F(2, 2843) = 11.762, p = .000. Unions (10.8%) and peace organizations (10.5%) open a newscast far more often than environmental organizations (2.8%). Whether an item deals with protest or not is insignificant, F(1, 2843) = 1.945, p = .116. A look at the absolute mean differences between protest and non-protest items across groups (Table 1, Panel 1) presents a clear picture: Both for unions (+2.5%) and environmental organizations (−0.7%), mean differences are small. Protests do not greatly affect news item prominence. For peace organizations, however, protests do matter substantially. Peace protest items are far more prominently covered (+10.5%) compared with non-protest items of peace organizations. H1 is corroborated.

Table 1.

Estimated Marginal Means for Four News Item Attributes Across News Item Types and Advocacy Group Families.

	Family	Item	M (SE)	95% CI		M(dif)	Significance
	Family	Item	M (SE)	LB	UB	M(dif)	Significance
Prominence	Unions	Non-protest	.096 (.008)	.081	.111	.025	.042
	Unions	Protest	.121 (.010)	.102	.140	.025	.042
	Environmental	Non-protest	.032 (.016)	.001	.063	−.007	.802
	Environmental	Protest	.025 (.026)	−.027	.076	−.007	.802
	Peace	Non-protest	.053 (.067)	−.078	.183	.105	.197
	Peace	Protest	.158 (.047)	.066	.250	.105	.197
Duration	Unions	Non-protest	121 (1.229)	118	123	−4.714	.018
	Unions	Protest	116 (1.535)	113	119	−4.714	.018
	Environmental	Non-protest	110 (2.509)	105	115	−33.144	.000
	Environmental	Protest	77 (4.198)	69	85	−33.144	.000
	Peace	Non-protest	130 (10.613)	110	151	−8.421	.518
	Peace	Protest	122 (7.505)	107	137	−8.421	.518
Standing	Unions	Non-protest	.914 (.009)	.897	.931	−.004	.763
	Unions	Protest	.910 (.011)	.889	.931	−.004	.763
	Environmental	Non-protest	.698 (.017)	.664	.732	−.198	.000
	Environmental	Protest	.500 (.029)	.443	.557	−.198	.000
	Peace	Non-protest	.684 (.074)	.536	.832	.211	.020
	Peace	Protest	.895 (.052)	.792	.997	.211	.020
Balance	Unions	Non-protest	.367 (.013)	.343	.392	−.099	.000
	Unions	Protest	.268 (.016)	.238	.299	−.099	.000
	Environmental	Non-protest	.437 (.026)	.387	.487	−.150	.003
	Environmental	Protest	.287 (.043)	.203	.370	−.150	.003
	Peace	Non-protest	.474 (.108)	.262	.686	−.132	.320
	Peace	Protest	.342 (.076)	.192	.492	−.132	.320

Note. CI = confidence interval. LB = Lower Bound. UB = Upper Bound

Duration

Are protest items shorter compared with non-protest items? H2 expected especially protest items of environmental groups to be shorter. A simple t test shows that protest items (M = 111.652; SD = 48.410) are significantly shorter than non-protest items (M = 118.696; SD = 46.097); t(2162) = 3.825, p = .000. The absolute difference is small, however, and SDs are considerable. Nonetheless, the finding is robust. The average advocacy group news item lasts 116 s; 53.5% of the non-protest items are longer, 47.8% of the protest items are shorter. Panel 2 of Table 1 presents estimated marginal means of a two-way ANOVA, introducing group family as a factor, which shows that this conclusion is evident across advocacy groups. The group family factor, F(2, 2837) = 45.282, p = .000; the protest factor, F(1, 2837) = 10.890, p = .001; and the interaction term between both, F(2, 2837) = 14.587, p = .000, are significant, with the group family factor having the largest partial eta squared (.031). In sum, both group type and group tactic matter for item length. Both union (M = 118.742 s; SD = 46.870) and peace group (M = 124.807 s; SD = 48.006) news items generally last significantly longer than items about environmental groups (M = 101.424 s; SD = 45.420; p = .000). Protest items are significantly shorter than non-protest items (p = .000). This finding holds across all three types of advocacy groups, yet the difference between protest and non-protest items duration is most substantial for environmental organizations. The average length of an environmental protest item is about 30 s shorter than a non-protest item (−33.14), which is a considerable difference. For unions (−4.67) and peace organizations (−8.42), the average difference in item length is far less outspoken, and for peace organizations, the difference even is insignificant (p = .518). In sum, on top of advocacy group family differences, the tactic of protest results in shorter news items. Especially for environmental groups, this difference is substantial. This confirms H2.

Standing

Standing is a crucial cultural resource for groups willing to come across as legitimate. Being singled out as newsworthy and being allowed to speak makes groups credible players. H3 expected standing for peace organizations in protest items to be higher, and for environmental groups to be lower, compared with their standing in non-protest items. A t test shows no significant difference between protest (M = .863; SD =.344) and non-protest items (M = .870; SD = .336) when it comes to media standing, t(2847) = .574, p = .566. A two-way ANOVA shows that media standing is a matter of group family, F(2, 2843) = 146.886, p = .000, with all three group families significantly differing from each other. Unions have the highest media standing. They are allowed to speak in about 9 out of 10 items in which they appear (M = .914; SD = .283). Environmental organizations have the least media standing: They get quoted far less, in only 6 out of 10 news items (M = .605; SD = .479). Besides the group family factor, only the interaction term between group family and group strategy is significant, F(1, 2843) = 17.422, p = .000. Put differently, the strategy of protest has a different effect across the different groups. Panel 3 of Table 1 shows the estimated marginal means. The standing of unions is always high, which makes that union standing does not significantly differ across protest and non-protest items (p = .763). Interestingly, the share of standing in non-protest items for both environmental and peace organizations is about the same (about 69%). Environmental organizations, however, get less standing in protest items, whereas peace organizations, in contrast, get more standing in protest items. This confirms H3.

Balance

A final news item attribute is the source selection pattern. H4 expected protest items to be less frequently balanced compared with non-protest items. A simple t test shows that protest items (M = .273; SD = .446) are less frequently balanced than non-protest items (M = .382; SD = .486); t(2405) = 6.097, p = .000. In other words, in protest items, advocacy groups more frequently have monopoly positions and are the only sources with a voice. Moreover, the link between protesting and monopoly positions holds across advocacy group families. In a two-way ANOVA, only the protest factor is significant, F(1, 2843) = 7.112, p = .008, which strongly suggests that source selection is primarily a consequence of group strategy (protest), rather than of group family. For unions, environmental organizations and peace organizations holds that protest items are less frequently balanced than non-protest items. Panel 4 of Table 1 shows that absolute mean differences are substantial (about 10%), and except for the low N category of peace organizations, also strongly significant. In sum, the evidence corroborates H4.

Taking stock of the above findings, it is now possible to turn to the final and overarching hypothesis of this article (H5). In H5, differences between protest and non-protest items were expected to vary across advocacy group families. More precisely, differences were expected to be small for unions, whereas protest was expected to have a positive effect for peace and a negative effect for environmental organizations. Overviewing the panels of Table 1, H5 can be confirmed. For unions, mean differences for prominence, duration, and standing are negligible. Protest appears as detrimental for environmental organizations with regard to duration and standing. Peace organizations, on the other hand, get more supportive protest coverage: They get more prominent coverage and receive more standing. The implications of these results are discussed in the next section.

Discussion and Conclusion

This analysis compared across different advocacy group families whether protest and non-protest items differ on particular attributes. As such, this article contributed to the literature on the protest paradigm in three specific ways: It added new features to the paradigm, provided protest coverage with a baseline (non-protest coverage), and teased out the paradigm’s conditionality (comparison across group families).

The results show that two general laws of protest coverage can be distinguished: Protest items are less frequently balanced and significantly shorter compared with non-protest items. These findings held across three very different types of advocacy groups. As such, they passed a tough test. Obviously, both attributes are related. Because protest items feature other sources less frequently, they tend to be shorter. This study showed that by staging protest events, advocacy groups more easily gain monopoly positions in news items, in which they can present their claims uncontested. This finding of (im)balance is the only finding that is exclusively explained by the protest item factor, strongly indicating that it most likely is a consequence of the tactic and not of the group using the tactic. Adding the baseline of non-protest items showed that official sources are less frequently present in protest than in non-protest items. A potential explanatory mechanism could be that the presence of advocacy groups in non-protest items is more likely to be reactive—that is, in reaction to a statement or a decision of an official actor. In protest items, in contrast, the advocacy group itself “makes” the news by staging the event and as such becomes the starting point for the journalist to construct the news item. Future research could more closely focus on the news production process, rather than analyzing its end product, the news content, to provide a well-founded answer on this matter.

Another key finding of this article is that for all news item attributes, except for the feature of balance, the group factor better accounted for differences than the protest factor. How a group is presented in the news depends less on the tactic and more on the group itself. This suggests that how groups are presented in the news is more structural than the protest paradigm expected. The case of the unions is most telling in this regard. In Belgium, unions are insiders of the political system. Media coverage reflects (and reinforces) this powerful position: Unions are featured most prominently, get most frequently quoted, and most frequently obtain monopoly positions. Moreover, the differences in news item features between union protest items and non-protest items are much smaller compared with the other groups. Irrespective of whether unions protest or not, the news items in which they appear tend to be very similar.

This holds true to a far lesser extent for the other groups under study. In fact, protest coverage turned out to be beneficial for peace organizations and detrimental for environmental groups. The fact that peace protests were large, grassroots, and peaceful and that environmental protests were professional, small, and somewhat more disruptive might account for these findings. More direct measurements of event and group characteristics, which were beyond the scope of this study, can give a more definite answer on this matter. More generally, however, this study presented the argument that the peace protest actions were held in a favorable opinion climate (for a similar argument see Dardis, 2006): Both public opinion and political elites in Belgium opposed the war in Iraq. Such a favorable media opportunity structure was not present for the environmental groups under study. Far more than the peace organizations, environmental groups had to make news by fighting an uphill battle for a spot on an already crowded media agenda.

In sum, the results of this study confirm that the “principle of cumulative inequality” is at work in the media arena, making strong organizations stronger, even in the case of protest actions, which are typically considered as the weapon of the weak (Wolfsfeld, 1997). Nonetheless, the evidence also suggests that the principle is relative, rather than absolute. Far less than only a story of gloom and doom, as often spelled out by the protest paradigm, protest strategies can come with advantages (monopoly positions), and movements can grasp momentum. The contours of the “discursive opportunity structure” not only imply stability but also possess chances of variability. In the words of Gamson (2004: 249),

The media arena is not that flat, orderly and well-marked field in a soccer stadium, but one full of hills and valleys [ . . . ]. To make matters even more complicated, the contours of the playing field can change in an Alice in Wonderland fashion in the middle of a contest because of events that lie beyond the control of the players; and players can themselves sometimes change the contours through actions that create new discursive opportunities.

The research reported here showed considerable variation in advocacy group portrayal. Future research could contribute greatly by empirically taking into account conditions that lie in and beyond the control of advocacy groups, such as event characteristics (whether the demonstration was large or small, disruptive or peaceful), organizational characteristics (group goals, resources, membership size, communication strategies), and contextual characteristics (public opinion, government composition), when trying to explain group portrayal in the media arena. Also, the specific media arena under scrutiny might affect the portrayal of the group: Dynamics could differ across media (press vs. television; commercial vs. public station) and across media and political systems. Comparative research is needed to tease out the generalizability of the findings presented here. For instance, it is probable that in countries in which unions are to a far lesser extent insiders of institutional politics, other powerful groups consistently get “successful” media coverage. Or, that in other time periods, not the peace movement, but another movement whose issue at a certain moment is peaking, enjoys momentum in the media arena and experiences supportive coverage for the protest events it stages.

If media attention is a crucial political resource for advocacy groups, further investigation along the research tracks presented above will aid to untangle how media coverage amplifies or cripples advocacy group power, and as such, hinders or facilitates social change.

Footnotes

Appendix

Appendix B

Descriptives of Dependent and Independent Variables.

	Mean/proportion	Minimum	Maximum	SD	N
Protest	.376	0	1	.484	2,849
Prominence	.094	0	1	.291	2,849
Standing	.867	0	1	.339	2,849
Balance	.341	0	1	.474	2,849
Volume	116	14	567	47.095	2,843

Acknowledgements

The author would like to thank Stefaan Walgrave and the reviewers of Journalism & Mass Communication Quarterly (JMCQ) for helpful suggestions on previous versions of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Ruud Wouters is a postdoctoral fellow of the research fund–Flanders (FWO) at the Department of Political Science, University of Antwerp. He is a member of the research group Media, Movements & Politics (M²P; ). His research interests are media coverage of social movements and protest, protest participation, and the political consequences of protest.

References

Andrews

K. T.

Caren

(2010). Making the news: Movement organizations, media attention, and the public agenda. American Sociological Review, 75, 841-866.

Behr

R. L.

Iyengar

(1985). Television news, real-world cues, and changes in the public agenda. Public Opinion Quarterly, 49, 38-57.

Boyle

McCluskey

Devanathan

Stein

McLeod

(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, 43-60.

Boyle

McCluskey

McLeod

Stein

(2005). Newspapers and protest: An examination of protest coverage from 1960 to 1999. Journalism & Mass Communication Quarterly, 82, 638-653.

Boyle

McLeod

Armstrong

(2012). Adherence to the protest paradigm: The influence of protest goals and tactics on news coverage in U.S. and international newspapers. International Journal of Press/Politics, 17, 127-144.

Chan

J. M.

Lee

C.-C.

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

Dissanayake

(Eds.), The news media in national and international conflict (pp.249-276). Boulder, CO: Westview.

Danielian

Page

(1994). The heavenly chorus: Interest group voices on TV news. American Journal of Political Science, 38, 1056-1078.

Dardis

(2006). Military accord, media discord: A cross-national comparison of UK vs. US press coverage of Iraq War protest. The International Communication Gazette, 68, 409-426.

Deacon

(1996). The voluntary sector in a changing communication environment: A case study of non-official news sources. European Journal of Communication, 11, 173-199.

10.

della Porta

Diani

(1999). Forms, repertoires and cycles of protest. In della Porta

Diani

(Eds.), Social movements: An introduction (pp. 165-192). Oxford, UK: Blackwell.

11.

Detenber

Gotlieb

McLeod

Malinkina

(2007). Frame intensity effects of television news stories about a high-visibility protest issue. Mass Communication and Society, 10, 439-460.

12.

Devos

Humblet

(Eds.). (2007). Arbeid vs. Kapitaal: Een Kwarteeuw Staking(srecht). Labour versus Capital. A Quarter Century of (right to) Strike. Gent, Belgium: Academia Press.

13.

Di Cicco

D. T.

(2010). The public nuisance paradigm: Changes in mass media coverage of political protest since the 1960s. Journalism & Mass Communication Quarterly, 87, 135-153.

14.

Entman

R. M.

(2007). Framing bias: Media in the distribution of power. Journal of Communication, 57, 163-173.

15.

Ferree

M. M.

Gamson

Gerhards

Rucht

(2002). Shaping abortion discourse: Democracy and the public sphere in Germany and the United States. Cambridge, UK: Cambridge University Press.

16.

Gamson

(2004). Bystanders, public opinion and the media. In Snow

Soule

Kriesi

(Eds.), Blackwell companion to social movement studies (pp.242-261). Oxford, UK: Blackwell.

17.

Gamson

Wolfsfeld

(1993). Movements and media as interacting systems. Annals of the American Academy of Political and Social Science, 528, 114-125.

18.

Hayes

A. F.

Krippendorff

(2007). Answering the call for a standard reliability measure for coding data. Communication Methods and Measures, 1, 77-89.

19.

Hooghe

De Swert

Walgrave

(2007). De Kwaliteit van het Nieuws. Kwaliteitsindicatoren voor televisienieuws. The Quality of the News. Quality Indicators for Television News. Leuven, Belgium: Acco.

20.

Lijphart

(1999). Patterns of democracy: Government forms and performance in thirty-six countries. New Haven, CT: Yale University Press.

21.

Lipsky

(1968). Protest as a Political Resource. The American Political Science Review, 62, 1144-1158.

22.

Martens

Van Gyes

Van der Hallen

(2002). De vakbond naar de 21ste eeuw. The Union to the 21st Century. Leuven, Belgium.

23.

McCluskey

M. R.

(2008). Activist group attributes and their influences on news portrayal. Journalism & Mass Communication Quarterly, 85, 769-784.

24.

McLeod

(1995). Communicating deviance: The effects of television news coverage of social protest. Journal of Broadcasting & Electronic Media, 39, 4-19.

25.

McLeod

Detenber

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

26.

McLeod

Hertog

(1992). The manufacture of “public opinion” by reporters: Informal cues for public perceptions of protest groups. Discourse & Society, 3, 259-275.

27.

McLeod

Hertog

(1998). Social control, social change and the mass media’s role in the regulation of protest groups. In Demers

Viswanath

(Eds.), Mass media, social control, and social change: A macrosocial perspective (pp.305-330). Ames, Iowa State University Press.

28.

Nelson

T. E.

Clawson

R. A.

Oxley

Z. M.

(1997). Media framing of a civil liberties conflict and its effect on tolerance. American Political Science Review, 91, 567-583.

29.

Ryan

(1991). Prime time activism: Media strategies for grassroots organizing. Boston, MA: South End Press.

30.

Smith

McCarthy

McPhail

Augustyn

(2001). From protest to agenda building: Description bias in media coverage of protest events in Washington, DC. Social Forces, 79, 1397-1423.

31.

Sobieraj

(2010). Reporting conventions: Journalists, activists, and the thorny struggle for political visibility. Social Problems, 57, 505-528.

32.

Thrall

(2006). The myth of the outside strategy: Mass media news coverage of interest groups. Political Communication, 23, 407-420.

33.

Verhulst

Walgrave

(2005). In or out of the picture? On old and new social movements in the Flemish press. Mens & Maatschappij, 80, 305-327.

34.

Vliegenthart

Oegema

Klandermans

(2005). Media coverage and organizational support in the Dutch environmental movement. Mobilization, 10, 365-381.

35.

Walgrave

Manssens

(2005). Mobilizing the White March: Media frames as alternatives to movement organizations. In Johnston

Noakes

(Eds.), Frames of protest: Social movements and the framing perspective (pp.113-140) Lanham, MD: Rowman & Littlefield.

36.

Walgrave

Rucht

(Eds.). (2010). The world says no to war: Demonstrations against the war in Iraq. Minneapolis: University of Minnesota Press.

37.

Walgrave

Verhulst

(2009). Government stance and internal diversity of protest: A comparative study of protest against the war in Iraq in eight countries. Social Forces, 87, 1355-1387.

38.

Weaver

Scacco

(2013). Revisiting the protest paradigm: The tea party as filtered through prime-time cable news. The International Journal of Press/Politics, 18, 61-84.

39.

Wolfsfeld

(1997). Media and political conflict: News from the Middle East. Cambridge, UK: Cambridge University Press.

40.

Wouters

(2013). From the street to the screen: Characteristics of protest events as determinants of television news coverage. Mobilization: An International Journal, 18, 83-104.

41.

Wouters

Hooghe

De Smedt

Walgrave

(2011). Rood en groen op het scherm: De vakbonden in het Vlaamse tv-nieuws. Red and Green on the Screen: The Unions in Flemish Television News. De Gids op Maatschappelijk Gebied: Blad met Mening (m/v), 102(6), 36-40.