Preferences for the Scope of Protests

Abstract

This paper studies a dimension of protest largely overlooked in the literature: protest scope, that is, whether protests seek large, structural, changes for a large share of the population or focus on small-scale improvements for small groups. We argue that this protest dimension is relevant for understanding the political consequences of protests. We show empirically that protests vary substantially in scope and that scope is not collinear with other protest dimensions, such as size, motive, or tactics. We explore drivers of individual preferences for protest scope with a survey experiment in two South African townships. We find that respondents made to feel more efficacious tend to support protests of broader scope. This effect operates via a social psychology channel whereby efficacy leads people to assign blame for their problems to more systemic causes.

Keywords

protest efficacy South Africa

Introduction

Research concerned with the political impact of protests has mainly studied two core dimensions of protest: its incidence/ size, and to a lesser extent, the use of peaceful versus violent tactics. A dimension of protest largely overlooked in the literature is protest scope, that is, whether protests seek large, structural, changes for a large share of the population (e.g., regime change) or focus on small improvements for small groups (e.g., paving a slum). Yet, this dimension is bound to be important for protest impact. Protests targeting systemic change, such as those toppling Arab autocrats in 2010/2011 or those trying to bring independence to Catalonia in 2017, have very different political consequences than localized, narrow protests focusing on the corruption of a ward councilor, bad health care in a specific district, or property rights for shacks in a slum.

This paper focuses on the demand side of protest scope and asks, what drives preferences for protest scope? The scope of protests obviously depends on supply factors, such as the calculations of movement elites and their decisions on how to frame protest narratives. However, the demand side is likely to be crucial as well: previous research demonstrates that protest narratives offered by elites have little impact unless they resonate with people’s preferences and interpretations (Benford and Snow 2000). In the words of Klandermans (2008), effective “mobilization brings a demand for political protest that exists in a society together with a supply of opportunities to take part in such protest.” Thus, understanding individuals’ preferences for narrow versus broad protests can help us explain (1) why protest movements that advocate for a similar scope of change sometimes succeed in mobilizing the population and sometimes not or (2) why under certain conditions social movements with broad (narrow) demands are more successful in attracting followers than movements with narrow (broad) demands.

We define protest scope with reference to previous literature along two dimensions, intensity (how much change is sought) and extensiveness (how many people would be affected by the change). We define a protest to be of broad scope if the change sought is intense and extensive. Building on previous research on protest occurrence and strategies, we suggest a simple theoretical framework for the analysis of protest scope. While we acknowledge that the nature of underlying grievances or the identity of protesters is likely to influence the demand side of protest scope, we focus in this paper on the role of people’s sense of efficacy. Specifically, we argue that efficacy is a crucial determinant of the demand-side of protest scope: the more people believe that it is possible to alter conditions or policies through protest, the more likely will they have a preference for broad as compared to narrow protest scopes.

Our study focuses on one specific African, middle-income, democracy. However, the concept of protest scope is relevant beyond this type of country. Protest movements in wealthy Western countries such as the Black Lives Matter in the United States also oscillate between a narrow focus on the brutality and impunity of certain police officers and a broad focus on social inequalities. Similarly, protest movements in non-Western autocratic states such as the recent demonstrations and strikes in Iran entail both a narrow focus on the economic conditions of the youth in specific regions and a broad focus on the general legitimacy of the Islamic Republic as a whole. These and other protest movements supply both, protest narratives with a narrow and with a broad scope. Which one of these narratives will garner more support and thereby shape the course of demonstrations and their potential sociopolitical impact depends on the demand-side of protest scope.

Our analysis is based on a survey experiment with 1,500 individuals in two South African townships. Our treatments were designed to temporarily affect respondent’s perceptions of efficacy. Our most notable result is that perceptions of efficacy enhance preferences for a broader protest scope. This seems to operate via a social psychology channel whereby higher perceptions of efficacy lead people to assign blame for their problems to more systemic causes.

The paper’s main contribution is an empirical investigation of the drivers of protest scope. While several previous studies have emphasized the role of the breadth of protest issues in terms of explaining the patterns and outcomes of political mobilization, only very few studies have investigated the determinants of variation in protest scope empirically. These few studies have either focused on static/slow-moving individual-level characteristics such as gender or education (Verhulst 2011) or on strategic considerations of protesters (Harris and Hern 2019). Our findings demonstrate how individual-level cognitive dispositions and processes can influence the demand for narrow versus broad protests. This, in turn, can explain how under similar structural conditions the same grievances can lead sometimes to protests with localized small-scale demands and sometimes to protests that demand systemic change.

In addition, our findings contribute to extant research on efficacy perceptions and collective action. Previous studies have highlighted the role of efficacy for protest participation (Gamson 1992; Van Stekelenburg and Klandermans 2013; Van Zomeren, Postmes, and Spears 2008): people are more likely to participate in protests when they believe that collective action is an effective means of redressing their grievances at acceptable costs. Our results extend this research: our findings suggest that efficacy affects protest behavior not only by influencing people’s cost–benefit calculations but also through cognitive processes of blame attribution. This latter finding is in accordance with recent literature on the effect of efficacy/power on motivated social cognition (Johnson and Fujita 2012; Jost et al. 2003; Pellicer, Piraino, and Wegner 2019; Van Der Toorn et al. 2015).

Finally, our results speak to previous research on the emergence of systemic protests. Our findings suggest that people’s awareness of instances of successful protest can increase their feelings of efficacy and thereby increase preferences for broad protest scope. This finding complements previous studies that try to explain how individual protest events or the presence of opposition movements in other countries can ignite unexpected waves of anti-regime protests—as in the case of the so called “Arab Spring,” the Color Revolution or, most recently, in Hong Kong. Whereas previous studies have argued that the observation of protest can make people reevaluate the costs and benefits of protest (e.g., Kuran 1989; Sunstein 2005; Weyland 2009), our results indicate that the observation of protests may also influence people’s efficacy beliefs and thereby increase their preferences for systemic protests.

Protest Scope

What Is Protest Scope?

Research has considered different dimensions of protests. The main focus has been on protest size and protest tactics. We focus on another dimension of protest: its scope. We define the concept of “protest scope” as a function of two features of the change sought by a protest: the intensity of change and the extensiveness of change. Figure 1 illustrates this idea.

Figure 1.

Dimensions of protest scope.

Intensity of change captures how large of a change is sought. The type of change pursued by protesters can be of low intensity (small), such as asking for the removal of a corrupt politician from office, or it can be of high intensity (large), such as seeking the replacement of the whole political class. Extensiveness of change defines how many people would be affected by the change. The affected population may be small, confined to a few people with certain characteristics (e.g., dairy farmers) or to a small geographical area (e.g., a village or shanty town), or it can be large, including all, or the majority of, citizens of a country.

The idea that protest scope matters is not new. Several previous studies have emphasized the role of the breadth of protest objectives. Older collective action literature has introduced the distinction between collective action seeking to address proximate causes of grievances versus actions focusing on more distal and structural causes (Blumer 1969; Carlier 1977). More recent works suggest a related distinction between minimalist and maximalist demands (see, for example, Klein and Regan 2018). These concepts reflect our understanding of the intensity of change, however, but do not incorporate the dimension of extensiveness.

Other authors have distinguished between particularistic versus universalistic protest demands that vary in the size of their beneficiary group. Particularistic protests focus on changing conditions of one clearly specified and relatively small group, while the latter challenge conditions that affect a broader population (Gamson 1992; McCarthy and Zald 1977; Verhulst 2011). This distinction corresponds to our idea of extensiveness but does not incorporate the notion of intensity.

Finally, a recent study by Harris and Hern (2019) discusses the difference between valence protest and system-changing protest: while the first only “seek a resolution to a grievance without any . . . larger claims,” the latter aim at the “removal of the current president/ruling party or regime.” While this distinction is close to our understanding of narrow and broad scope, the notion of “valence protests” incorporates the nature of the issue behind the protest into the distinction.

Conversely, our understanding of protest scope excludes the nature of the actual grievance as a defining part of the concept. As we will discuss in more detail below, we assume that the same grievance or protest issue can give rise to protests of very different scope. For example, the same valence issue such as health care may be as much of narrow scope, such as asking for more resources for the local clinic, as of broad scope, such as asking for massive redistribution in a country.

Thus, while our concept of scope addresses a similar feature of protests as other concepts that have been suggested in the past, we believe that it offers advantages in terms of its comprehensiveness (combining the dimensions of intensity and extensiveness) and in terms of its independence from other dimensions of protests—most notably, the nature of the grievances driving a protest.

Figure 1 visualizes how different values of the two dimensions define degrees of scope. If the intensity of the change sought is high, we can think of it as protest seeking a large amount of change, such as changing political or socioeconomic structures. In broad-scope protests, these changes would potentially affect a large share of the population (i.e., be very extensive). In turn, a narrow-scope protest would seek low key changes affecting a small population. In addition, we can conceive of intermediate types. Protest with intermediate scope can be either those seeking relatively small policy changes (low intensity) that affect a lot of people (high extensiveness), such as demanding food subsidies or more or less funding for some national programs, or those seeking big changes (high intensity) for small amounts of people (low extensiveness). This is a less realistic case, but one could imagine protests targeting a massive overhaul of local structures, perhaps taking over the municipality or deposing the local chief.

The breadth of protest scope is likely to have important implications. Protests with narrow scope, such as a new school in a neighborhood or the removal of one specific corrupt politician from office, are unlikely to affect political and social stability. In contrast, protest centering on inequality of opportunity or the corruption of the entire political class has more potential to shake up the system. Of course, many protests of broad scope will have little or no effect on stability, but it is a sensible assumption that being of broad scope is a necessary condition for protests to have a more fundamental impact.

Previous theoretical and case-study work dealing with protest issue scopes has focused mainly on the supply side of protest narratives, investigating how protest organizers strategically frame protest narratives in order to maximize participation and impact. Some studies have argued that protest movements may be more likely to realize broad participation when they elevate their concerns to the status of national issues and frame protest issues as broadly as possible (McCarthy and Zald 1977; Rootes 2013). Other studies suggest that movements may be more effective when they focus on very specific and rather limited objectives (Alinsky 2013; Ennis and Schreuer 1987; Thomas and Craig 1973). Protest scope is also likely to influence the chances that social movements can realize actual policy change. Because governments are reluctant to concede to protests that demand fundamental change, narrow-scope protests are more likely to succeed in realizing the intended change (Franklin 2009; Klein and Regan 2018).

We take on another perspective. Rather than focusing on why social movements decide to supply certain protest narratives, we focus on why the population demands broad or narrow protest narratives. To the best of our knowledge, only two previous studies have sought to explicitly explain individual-level variation in support for different protest issue scopes. Verhulst (2011) uses observational data to investigate the determinants of preferences for either particularistic and universalistic protest issues. He finds that preferences vary systematically across individual according to factors such as gender, education, and interest in politics. Harris and Hern (2019), in turn, argues that citizens protest over valence issues (i.e., narrow-scope protests) because they serve as a tool to communicate their own policy preferences and priorities to the incumbent regime where voting alone is not effective.

Building on previous work in social psychology (i.e., system justification theory), we suggest an alternative perspective on preferences for protest issues scopes: rather than focusing on more or less stable sociodemographic characteristics or strategic considerations, we highlight the role of more dynamic cognitive processes. The following sections introduce our main arguments.

Protest Scope in South Africa

We first provide some descriptive evidence on protest scope in South Africa, seeking to distinguish it from other dimensions of protest deemed relevant in the literature. We show that the scope of protests is not collinear with the underlying grievances motivating collective action, protest size, or protest tactics.

We use data from the South African police’s crowd control database, the Incident Registration Information System (IRIS) that records events with more than five participants. These data record the event’s location and date, provide a short description, an approximate number of participants, and indicate whether the gathering was peaceful or violent (see Runciman et al., 2016, for an assessment of and De Juan and Wegner, 2019, for a previous study using these data).

We coded a random subset of five hundred protest events in 2011 and 2013 from the database with the objective to identify the scope of protest as well as other relevant dimensions such as grievance type and size. Sampling was stratified by year, time of the year, and province. The scope of protests was operationalized as protest targets, that is whether the protest was directed at the local, province, or national level with higher levels implying a broader scope. Protest targets in terms of institutional level are not a perfect measure of protest scope. However, they are closely related to the extensiveness dimension of protest scope mentioned above. Protests targeted at the national level entail a potentially larger beneficiary group than protests targeted at the local level. Moreover, national-level protests also typically score relatively high on the intensity dimension of protest scope, as they tend to emphasize the “macro,” structural, aspect of problems.

Table 1 shows the distribution of these different dimensions of protest (topic, size, and tactic) conditional on specific values of protest scope (local, provincial, and national). The vast majority of the events (around 85%) are targeted at the local level (narrow scope). A majority of protests concern employment (around 50%) and service delivery (around 20%), are fairly small (protests below five hundred participants make up 90% of the sample), and are peaceful (around 75%).

Table 1.

Protest Scope and Other Dimensions of Protest.

	Local	Provincial	National
Protest motive
Crime	0.05	0.00	0.12
Employment/salaries	0.54	0.56	0.51
Governance/corruption	0.07	0.06	0.09
Service delivery	0.21	0.28	0.16
Other	0.13	0.11	0.12
Number of participants
Below 100	0.58	0.35	0.52
101–500	0.33	0.59	0.36
501–1,000	0.04	0.06	0.02
Above 1,000	0.04	0.00	0.10
Tactics
Peaceful	0.78	0.78	0.98
Violent	0.22	0.22	0.02

Based on a random subset of 2011 and 2013 events from IRIS (Incident Registration Information System) database. Cells show shares by protest scope level. Fisher’s exact test: Protest Motives–Level: Pr = 0.802; Number of Participants–Level: Pr = 0.200; Tactics–Level: Pr = 0.003.

Comparing the share of specific motives, sizes, and tactics across scope level already shows that none of these protest dimensions are identical with protest scope. Topics such as services, employment, or governance/corruption can all lead to protests targeted at local, provincial, or national levels, as can protests with different numbers of participants and using different tactics.

We conduct Fisher’s exact test to determine whether scope is moreover independent from the other protest dimensions. Table 1 shows the p values of the associations between protest scope with protest topic, size, and tactics, respectively. Based on the large p values for the associations between scope and topic and scope and size, we consider these dimensions to be independent from each other. In contrast, protest scope and protest tactics show a significant association. This association is driven by broad scope with our data containing 51 national-level protests using peaceful tactics and only two that use violent tactics. In turn, narrow-scoped protests show fairly similar shares of peaceful and violent tactics, suggesting that there is no mechanical relationship between scope and tactics.

In sum, while protest scope is not always fully independent from other protest dimensions, it is sufficiently distinct to merit separate consideration.

Conceptual Framework

What drives preferences for protest scope? At present, there is no established conceptual framework to guide the study of this question. We thus build on established factors in the literature on preferences for other dimensions of protest to propose our argument on the role of efficacy for protest scope. There are three types of factors that are consistently used to explain attitudes toward protests: grievances, identities, and efficacy. Grievances are considered key determinants of individual-level participation in protest (Chenoweth and Ulfelder 2017; Finkel, Muller, and Opp 1989; Gurr 1970). As Muller and Jukam (1983, 159) explain, “People who take part in acts of civil disobedience or political violence are discontented about something. That is a truism.” Importantly, what matters for collective action is not objective disadvantage but the subjective experience of disadvantage, perceived to be unfair and blamed on someone else (see Van Zomeren, Postmes, and Spears 2008). The second factor emphasized as relevant for protest participation is identity. The more people identify with social, ethnic, or political groups, the more they feel an obligation to participate in protests on behalf of the group (Van Stekelenburg and Klandermans 2013). Consequently, individual-level protest participation is likely to depend on the strength and political salience of people’s collective identities (Van Zomeren, Postmes, and Spears 2008). Finally, participation also depends on perceived efficacy. The more people believe that it is possible to alter conditions or policies through protest, the more likely they are to participate (McCarthy and Zald 1977; Oberschall 1973). The link between efficacy and protest is thought to be direct and straightforward: “the more effective an individual believes protest participation is, the more likely s/he is to participate” (Van Stekelenburg and Klandermans 2013, 3).

Since grievances, efficacy, and identity are relevant for attitudes toward different dimensions of protest, we conjecture that they may matter for attitudes toward protest scope as well. In this paper, however, we focus specifically on the role of efficacy for three reasons. First, as we cannot rely on a well-established body of research on protest scope, considering all three potential explanatory factors would require theoretical and empirical specifications that would go beyond the scope of this rather exploratory paper. Second, whereas grievances and identity are likely to affect attitudes toward the scope of protests, they are unlikely to fully determine these attitudes. Regarding grievances, it seems plausible that individuals with “bigger,” systemic, grievances engage in broad-scope protest, and individuals with narrow grievances engage in narrow-scope protest. However, grievances alone are unlikely to be the sole drivers of protest scope attitudes and behavior. For instance, as argued and shown for the case of South Africa above, associations between grievances and protest scope do not seem to be unequivocal, as similar types of grievances can result in protests of different scopes. As another example, the 2011 Arab uprisings mostly started with specific economic or administrative demands (more jobs, subsidies, removal of individual corrupt politicians), but later turned into radical demands of system change. It is possible that this change in protest scope was driven by a rapid consolidation of grievances, but it is also plausible that the change was driven by other factors, namely efficacy: protesters may have become emboldened by their success, become more efficacious, and shifted to a more systemic focus.

The potential role of identity for attitudes toward protest scope is even less clear. Although one could imagine that people with strong local identities favor narrow protests that ask for local solutions, and vice versa, effects of identities are probably rather ambiguous: very narrow ethnic identities could either create preferences for very narrow protest scopes, like reforms benefiting only their own respective communities, or create preferences for very broad protest scopes, like the partition of the political system. Conversely, while people with strong national identities might be more susceptible to engaging in protest that would affect all citizens, their national identity may also make them less inclined toward protests demanding extensive change of the political system. Third, whereas efficacy is also not likely to be the exclusive determinant of attitudes toward protest scope, it is the factor that flows most clearly from the existing literature on the scope dimension of protest mentioned above. First, as sketched above, previous studies on the supply side of protest narratives have found evidence that narrow-scope protests may be more likely to succeed in terms of realizing actual policy change (e.g., Klein and Regan 2018). This suggests that a certain degree of efficacy is needed in order to support broad-based protests. Second, efficacy also features prominently as a driver of system justification, the social psychology theory mentioned above. The system justification motive is thought to be particularly weak among those who feel highly efficacious; this again implies that efficacy ought to drive support for broad-scope protests. In what follows, we explain these two mechanisms in detail.

Mechanism

As Figure 2 shows, there are two potential mechanisms through which efficacy could lead to a preference for protests of broader scope. The first perspective is rooted in the collective action literature and the supply side of protest scope literature. This perspective highlights processes of rational decision-making: efficacy perceptions affect the cost–benefit calculus of engaging, and remaining engaged, in protest (see Van Zomeren, Postmes, and Spears 2008). More efficacy implies the perception that protests can be more successful, and this makes engaging in protest activities more worthwhile. This argument can be applied to the scope dimension of protest in two steps. First, corresponding to the top-right arrow in the figure, supporting broad versus narrow protests will naturally depend on how likely these protests are thought to be successful. If people believe that broad protests are likely to succeed, they will rationally be more likely to support them. Second, corresponding to the top-left arrow, efficacy may affect whether broad protests are deemed likely to succeed or not. This emerges from the supply-side literature mentioned above. Narrow protests may generally be considered more likely to succeed, and it may require high levels of efficacy to believe that bigger and broader causes can be taken on successfully. If that is the case, then high efficacy would promote beliefs that broader protests can succeed and hence promote preferences for broader protests.

Figure 2.

Channels linking efficacy to protest scope.

A second perspective is rooted in social psychology accounts of protest preferences. Under this perspective, efficacy affects preferences for the scope of protests because efficacy affects who is blamed for a given grievance. Again, the argument has two steps, corresponding to the bottom two arrows. The first step (bottom-right arrow) is a link between blame attribution and preferences for protest scope. If social problems are blamed on the “system,” as opposed to localized/ narrow factors, broad-scope protests are more likely to be supported. What is critical here is that the same grievance, or “objective” problem, can in principle be attributed to narrow or to systemic causes. Social problems are complex and have a variety of causes, and different people may attribute blame for the same problem to different factors. Two people may both have similar grievances regarding, say, electricity but while one person may attribute blame for this problem to the government, someone else may attribute it to inefficiency of the utility company. As Javeline (2009, 32) argues in her study of protest on wage arrears in Russia:

Targets for blame are . . . conceived by individuals in their minds and through conversations with others about the particular issue in question. At any time, possibilities include managers, local executives, local legislatures, national executives, national legislatures, and a variety of others.

It seems sensible to consider that people blaming problems on narrow factors will prefer narrow-scope protests and vice versa.

The second step in the social psychology perspective (bottom-left arrow) is that perceptions of efficacy affect how individuals attribute the blame for their problems. Here, we draw on system justification theory and related arguments. The main argument is that one of the ways people may cope with disadvantage is by “appraising” (i.e., thinking and evaluating) their situation in a way that makes their disadvantage less painful or salient. This is most likely to happen if people feel powerless and lacking in efficacy. Instead of suffering constant psychological pain by considering that their situation is unfair, engaging in system justification can be psychologically palliative (Jost, Banaji, and Nosek 2004). There is indeed evidence that a feeling of powerlessness activates system justification (Johnson and Fujita 2012; Van Der Toorn et al. 2015). And there is also evidence that this type of “re-appraisal” can lead people to attribute blame for their problems and grievances in different ways (Major and Schmader 2001). Putting these arguments together leads to our social psychology mechanism: a high sense of efficacy de-activates the system justification motive, making people re-appraise their social situation in a way that re-directs blame for their grievances toward more systemic causes, and this promotes support for broad-scope protests.

In sum, we conjecture that efficacy may affect protest scope through two distinct channels: a rational cost–benefit channel that emphasizes beliefs on the success of different types of protests, and a social psychology channel that emphasizes who is blamed for given grievances.

Data

We conducted an individual-level opinion survey with 1,500 respondents in two South African townships, Gugulethu and Mitchells Plain between March and May 2016. One hundred thirty-five enumerator areas (EA) were randomly selected from a complete list of residential EAs defined by Stats SA. Within each EA a random GPS coordinate was chosen as the starting point for a Random Walk; respondents inside the household were selected with a Kish grid. Seventeen fieldworkers conducted the interviews with fieldworker population groups matching those of respondents. The data were collected on mobile devices.¹

We selected these two townships because the combination of grievances and protest engagement makes them good places to test our mechanism. While not the poorest South African settlements, Gugulethu and Mitchells Plain have high levels of unemployment (around 40% in Gugulethu) and suffer from crime and service delivery problems. Both experience many protests so that respondents should be able to engage with our treatment (see below). Gugulethu is an African black township, and Mitchells Plain has a “Coloured” population.²

The survey includes an experimental component—discussed in more detail below—that investigates what drives individual preferences for protest scope. The survey focuses on perceptions of efficacy, as well as blame attribution and preferences for protest scope.³ The key outcome, preferences for protest scope, is measured by asking about support for different hypothetical protest marches that would address a problem of health care provision in the community. The marches presented to respondents address the same objective problem but differ in scope. The first march requests more doctors and nurses. This is a narrow-scope protest that seeks direct solutions for the particular neighborhood by increasing health care personnel in that specific place. The second march asks for the removal of corrupt politicians from office, a protest of broader scope. The third march has the broadest scope, asking for wealth redistribution to increase equality in access to high-quality health care.

We use preferences for these marches to construct two measures of preferences for protest scope. We ask respondents how much they would support any of these marches and subsequently ask them to choose which one they would attend if they could attend just one. The first measure is simply the choice of the march they would attend. The second measure is the difference in support for the corruption and redistribution marches, respectively, and support for the narrow-scope march. These latter measures capture preferences for broad versus narrow protests netting out general support for protest. In addition, we propose respondents to endorse a petition supporting each of these claims.

Three points are worth noting regarding our measures of protest scope. First, we focus on preferences for protest scope for a given objective problem. This is because we want to understand the variation in protest scope preferences net of the scope of objective grievances. Focusing on one specific objective problem allows us to explore more precisely the role of efficacy and blame attribution. Second, the corruption march may have been perceived more ambiguously than we anticipated. Our wording (“the healthcare problem cannot be solved unless corrupt politicians are removed from office”) can be interpreted in rather narrow terms (“rotten apples” or local politicians) or in broader terms (national political class). It could also be chosen simply because corruption is a highly salient topic in South Africa. Third, our analyses focus on self-reported preferences for protest scope. These preferences cannot tell us if people would really be willing to incur the costs associated with actual protest activity, for example, in terms of time investment or potential risks.

The survey includes a number of measures capturing perceived efficacy. The first measure focuses on protest efficacy and assesses to what extent respondents believe that they can alter conditions or policies through protest. Political efficacy assesses the extent to which people are convinced that they can make a change through conventional political means. Social efficacy measures beliefs about people’s ability and willingness to stand up for a just society. Last, personal power measures respondents beliefs about power over their environment (Anderson, John, and Keltner 2012). Each index includes three items with answer options ranging from disagree strongly to agree strongly. We perform a principal component analysis of these items and use the first component as measure of the respective concept. All indices are standardized.

We also collect information allowing us to assess the two potential channels linking efficacy to protest scope (blame attribution vs. rational cost–benefit calculation). Blame attribution is measured via two survey items where respondents are asked to attribute the blame for two specific problems, delivery of water and electricity, and crime. The answer options range from blaming the people (i.e., bad parenting for crime, people stealing electricity for services), to state agencies (the police, ESKOM, water authority), the government, or the wider system (poverty and inequality). We construct variables for blaming people, blaming agencies, and so on by combining the crime and the services items; that is, an individual has value one for blaming the wider system if they blame poverty and inequality for both problems. In addition, we combine the two blame variables (for services and crime) into an additive variable (broad blame) to measure more fine-grained blame levels. To assess the plausibility of the cost–benefit calculation channel, we gauge people’s assessment of the effectiveness of different protest types via an item (broad protest effectiveness) asking respondents about their beliefs about which type of protest they think is most likely to be successful. Protests about small (narrow) issues, protest about big (broad) issues, or whether the type of issue has no bearing on the likelihood of success. We use the latter two items relative to the first to measure beliefs about the potential effectiveness of broad versus narrow protests.

Descriptive Statistics of Key Outcome Variables

Table 2 shows descriptive statistics of the protest scope, blame attribution, and protest type effectiveness variables for the control group. The distribution of the preferences toward protest scope and blame attribution targets shows that there is considerable variation in these attitudes that requires an explanation. It also underscores the point that particular grievances—in this case about health care, service delivery, and crime—are not associated with clear-cut protest scopes or blame targets.⁴

Table 2.

Outcome Variables in Control Group.

	Share respondents
March choice
Choice docs/nurses march	0.44
Choice corruption march	0.31
Choice redistribution march	0.25
Attribution crime problem
Blame people	0.19
Blame police	0.14
Blame government	0.33
Blame poverty/inequality	0.34
Attribution service problem
Blame people	0.25
Blame agency	0.11
Blame government	0.32
Blame poverty/inequality	0.33
Protest type efficacy
Protests never work	0.11
Only narrow protests work	0.34
Only broad protests work	0.30
All protest scopes can work	0.25
Observations	1,482

The “march choice” variables show that there is indeed high variation in terms of which protest respondents choose in order to address the same hypothetical health care problem. While about half of the respondents favor a narrow-scope protest—asking for more doctors and nurses—the remainder is split between the corruption and the redistribution march. This supports our contention that different people articulate similar grievances in different ways. In turn, respondents also differ in where they attribute blame for the delivery and police problems. Each type of response has been selected by substantial numbers of respondents. In particular, around 60 percent of respondents attribute crime and public service problems to broader targets, namely the government or inequality, but there is also about one-quarter of respondents who believes the people are to blame. Last, the table also shows the results for beliefs about protest type effectiveness. Again, there is substantial variation. Beyond the 10 percent that believe that protest never works, there are about one-third of respondents each believing either that only narrow- or only broad-scoped protests work, and about a quarter that believe that protest scope has no bearing on how successful a protest will be.

Results

Our main analysis is based on an experiment embedded in the survey where we seek to temporarily affect perceptions of efficacy and study the effects of this manipulation on protest scope. We consider the two potential channels presented above, via processes of rational decision-making and via blame attribution.⁵

Treatments

To affect perceptions of efficacy, we ask respondents to recall a successful (high-efficacy treatment) or unsuccessful (low-efficacy treatment) protest. We then ask a number of questions about that protest in order to make respondents engage with the treatment. For example, we ask what the protest was about, whether they or someone they know participated, why they felt it was successful (unsuccessful), or how empowered (powerless) they felt when they realized that the protest was successful (unsuccessful). In addition, there is a control group which received the identical questionnaire as the treatment groups except for the treatment question.

To make it more likely that respondents recall a successful (unsuccessful) protest event, the categories of success and failure were defined very broadly. In the high-efficacy treatment, we define success not only as having demands met but as also including the increase of awareness about the problem; in the low-efficacy treatment, we define failure not only as “total failure” but also including protest where not all demands were met.

We opted for a personal recall treatment rather than presenting respondents with real existing protests for three main reasons. First, pre-tests revealed that it was difficult to identify protests that would be relevant and known to most of the respondents.⁶

Second, individuals have pre-conceptions about the success level of an existing protest. It was perfectly possible that a protest presented as successful would have been stored in someone’s memory as unsuccessful—perhaps because it did not solve a problem that was personally relevant to the respondent—and vice versa. Third, as Hassell and Settle (2017) discuss, treatments requiring respondents to think about personal issues are more salient. They make respondents “more prone to conceptualize themselves in that state of being because they are more likely to access memories confirming these conceptualizations” (pp. 13–14). In other words, respondents are more likely to feel efficacious if they remember a protest they personally thought was successful compared to an existing one presented to them as successful.

The treatments were designed in an effort to minimize the chances that they affect individual dispositions besides efficacy. The “high” versus “low” condition differ only in one word: whether a successful or unsuccessful protest was to be remembered. The objective was to induce a difference in perceived efficacy as large as possible, but nothing else.⁷

It is important to note that the treatments are not intended to emulate a real-world situation and generate lasting effects. The treatments intend to temporarily manipulate perceptions of efficacy to help understand mechanisms driving the formation of preferences for protest scope.

The experimental randomization was programmed into the survey so that the different treatment groups should not differ significantly from each other. Ordinary least squares (OLS) regressions of demographic and pre-treatment attitudinal variables on the treatments show indeed that all treatment groups are similar ex ante (see Table A.6 in the online appendix).

Manipulation Checks

We first consider whether our treatments have succeeded in manipulating different measures of efficacy perceptions. The first measure is protest efficacy and represents our manipulation check in the narrow sense. We also consider measures of more general political and social efficacy to understand whether a successful manipulation of beliefs about protest efficacy carry over to beliefs about efficacy more generally.

Table 3 shows the treatment effects on indices of protest, political, social efficacy, and power. The top panel compares the high with the low condition of the efficacy treatment, while the bottom panel compares both conditions to the control. The treatment is successful in affecting all types of efficacy perceptions as well as perceived personal power. Individuals asked to remember a successful protest (relative to those remembering an unsuccessful one), and are more likely to believe generally that protests are effective, that politicians can be held accountable and that social groups can improve their situation.

Table 3.

Manipulation Checks.

	(1)	(2)	(3)	(4)
	Protest efficacy	Political efficacy	Social efficacy	Personal power
Efficacy high vs. low
High efficacy	0.293*** (0.069)	0.114 (0.067)	0.181* (0.071)	0.141* (0.060)
Observations	776	776	751	776
All vs. control
High efficacy	0.264*** (0.061)	0.134* (0.059)	0.112 (0.062)	0.158** (0.055)
Low efficacy	−0.026 (0.060)	0.011 (0.059)	−0.054 (0.061)	0.015 (0.054)
Observations	1,478	1,478	1,439	1,478

Results from OLS regressions of the outcome variables in the column on treatment condition. The upper panel restricts the sample to respondents receiving a treatment and shows the coefficient for the high-efficacy treatment; the lower panel includes the control group and shows the coefficient for the high- and low-efficacy treatment. Controls: fieldworker, area, female, age, and completed secondary schooling. Standard errors in parentheses. OLS = ordinary least squares.

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

Moreover, they are more likely to feel powerful in a personal sense. The size of the coefficients is substantial. The high-efficacy treatment (relative to the low-efficacy one) increases protest efficacy by 0.3 standard deviations and increases political and social efficacy by around 0.1 to 0.2 standard deviations.

The high-efficacy treatment generates these results. In contrast, respondents in the low-efficacy treatment are indistinguishable from the control. This might be an indication that the default perception in this population is rather inefficacious and pessimistic so that perceiving protests as generally unsuccessful is the norm (see also Van Der Toorn et al., 2015, who argue that disadvantaged individuals often feel rather powerless). From here onward we display results that compare the high- and low-efficacy treatments to the control group.

Protest Scope

We consider the effect of the efficacy treatments on preferences for protest of broad scope. Columns 1 and 2 in Table 4 show the results for respondents’ choice of a particular march, and columns 3 and 4 display their level of specific support for one of these marches.⁸ Results are broadly as hypothesized regarding preferences for the redistribution march. Respondents in the high-efficacy treatment are more likely to choose a march focusing on poverty and inequality to address health care problems in their neighborhood. In particular, receiving the high-efficacy treatment increases the probability of choosing the redistribution march by 7.3 percentage points relative to the control condition. The effect for support for the redistribution march is not statistically significant at conventional levels. However, it is close (p value = .071) and in conjunction with the significant result for choosing the redistribution march, this suggests that overall, the treatment effect for protest scope is robust.

Table 4.

Treatment Effects on Preferences for Protest Scope.

	(1)	(2)	(3)	(4)
	Choice march corruption	Choice march redistribution	Support march corruption	Support march redistribution
Efficacy high	−0.023 (0.029)	0.073** (0.027)	−0.051 (0.066)	0.139 (0.072)
Efficacy low	0.054 (0.029)	−0.024 (0.027)	0.017 (0.066)	0.048 (0.071)
Observations	1,465	1,465	1,478	1,478

Results from OLS regressions of the outcome variables in the column on treatment condition. Controls: fieldworker, area, female, age, and completed secondary schooling. Standard errors in parentheses. OLS = ordinary least squares.

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

As expected, the low-efficacy treatment, which does not affect efficacy, does not affect preferences for a redistribution march either. These insights hold across the two outcome measures. In contrast, we find no effect of the high-efficacy treatment on preferences for the corruption march. This may be due to the fact that the corruption march may be considered of rather narrow scope, as mentioned above. These results carry over to people’s willingness to endorse a petition targeting either narrow or broad complaints (see Table A.7 in the online appendix).

Channels: Protest Type Effectiveness and Blame Attribution

Our conceptual framework posits two potential channels through which efficacy could affect preferences for protest scope. One channel could operate via rational cost–benefit calculations. Beliefs that broad-scope protests can succeed may generate preferences for protests with broader scope. We refer to this as a “rational choice” channel. A second channel could operate via blame attribution. Higher perceptions of efficacy may allow for blame to be attributed to more systemic targets. We refer to this as a “social psychology” type of channel. In terms of our theoretical discussion above, we specifically test the first step of each mechanism (i.e., the top-left and bottom-left arrows in Figure 2). In other words, we test whether the efficacy treatment affects perceptions of success of broad protests, or systemic blame attribution. Table 5 shows these two treatment effects. We find evidence in line with the “social psychology” channel but not with the “rational choice” channel. Columns 1 to 4 show that respondents in the high-efficacy conditions are indeed more likely to attribute blame for their grievances to systemic factors: the high-efficacy condition is associated with a higher propensity to attribute social problems to the government and to poverty/ inequality and a lower propensity to attribute blame for these problems to the people themselves. Relative to the control group, receiving the high-efficacy treatment, consistently blaming the government for these grievances increases by 6 percentage points, whereas blaming inequality increases by 7 percentage points.⁹ In contrast, there is no effect of the high-efficacy condition on people’s assessment of the effectiveness of broad-scoped protests. Individuals in the high-efficacy conditions are not more likely to believe that either broad-scoped protests are likely to succeed or that any protest can be successful. The low-efficacy treatment, in turn, has no effect on any of these variables.¹⁰

Table 5.

Channels: Blame Attribution and Protest Type Effectiveness.

	(1)	(2)	(3)	(4)	(5)
	Blame people	Blame agencies	Blame government	Blame inequality	Broad protests effective
High efficacy	−0.060*** (0.016)	0.007 (0.009)	0.060** (0.021)	0.072** (0.022)	0.042 (0.029)
Low efficacy	−0.003 (0.016)	0.014 (0.009)	0.013 (0.021)	0.001 (0.022)	0.015 (0.029)
Observations	1,459	1,459	1,459	1,459	1,462

Results from OLS regressions of the outcome variables in the columns on treatment condition. Controls: fieldworker, area, female, age, and completed secondary schooling. Standard errors in parentheses. OLS = ordinary least squares.

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

Our results therefore suggest that the effect of efficacy on protest scope operates via the social psychology mechanism rather than the rational cost–benefit channel. However, this does not imply that our respondents appear “irrational” or do not care about costs and benefits. As mentioned above, the cost–benefit channel implies two steps: first, that higher efficacy makes it more likely to believe that bigger causes (relative to smaller ones) are more likely to succeed and second, that this drives protest scope preferences. Rational cost–benefit calculations have to do with the second step of this mechanism (thinking that broad protests are more likely to succeed makes you more likely to support them). What we find is that the first step is not present: more efficacy does not lead people to believe that bigger causes are increasingly likely to succeed.

This is a somewhat surprising result, but by no means implies irrationality. This step is based on the idea that narrow protests are in general more likely to succeed and it takes high efficacy to believe that broad ones will succeed. It is possible that on average people believe that big causes are as likely to succeed as small causes. In that case, efficacy need not affect perceptions on the relative success of big versus small causes.¹¹

Concluding Remarks

This paper argues that scope is a dimension of protest relevant for understanding a protest’s potential impact—alongside established dimensions such as size or protest tactics. We show that similar grievances can give rise to protests of different scope. We propose a simple conceptual framework to understand drivers of preferences for protest scope focusing on perceptions of efficacy. We provide evidence that individuals induced to feel more efficacious are indeed more likely to prefer protests of broader scope to address a given social problem, namely protests that target systemic social issues such as poverty and inequality. This effect seems to operate via a social psychology channel whereby high efficacy increases the propensity of people to attribute their problems to systemic factors.

Because no pre-analysis plan was registered for the experiment, the results we present in this paper are exploratory. At the same time, we are confident that the results presented above are not the outcome of selecting variables with statistically significant results or of coding choices. The questionnaire included a limited number of outcome questions, and all outcome questions feature in the results presented above. We have aimed to be fully transparent about our coding decisions and where suitable, results from alternative coding are presented in the online appendix.

We believe that the study of protest scope can offer useful new insights into the political behavior of disadvantaged groups and the dynamics of protest movements. Among others, our findings can contribute to previous research that tries to explain how seemingly small-scale protests can unexpectedly evolve into mass movements demanding systematic change. For example, cascade models of protest participation (Sunstein 2005) highlight that the anticipation and observation of protest participation can make people reevaluate the costs and benefits of protest (Kalinin and Vogel 2016; Verhulst 2011). Research on the diffusion of protests highlights how successful protests in one country can trigger protests in other countries by demonstrating that political change is possible (Bamert, Gilardi, and Wasserfallen 2015; Hale 2013; Weyland 2009). Finally, Kuran (1989) argues that revolutions happen when even slight signals of opposition trigger bandwagoning processes among individuals that previously feared to disclose their anti-regime sentiments. We offer a complementary theoretical perspective to this research that has heavily relied on rational choice theory to explain protest dynamics: experiencing or observing protest success can strengthen efficacy beliefs of the population and thereby increase the demand for more far-reaching, systemic protest demands.

Supplemental Material

Online_Appendix – Supplemental material for Preferences for the Scope of Protests

Supplemental material, Online_Appendix for Preferences for the Scope of Protests by Miquel Pellicer, Eva Wegner and Alexander De Juan in Political Research Quarterly

Footnotes

Acknowledgements

All authors contributed equally to this paper. We would like to thank Bert Klandermans, Eva Anduiza, Paul Walsh, as well as participants of the workshop “A Closely Coupled Tango? Interactions between Electoral and Protest Politics” at the 2017 ECPR Joint Sessions in Nottingham, the 2017 Barcelona-Gothenburg-Bergen Workshop on Experimental Political Science in Bergen, the 2017 International Meeting on Experimental and Behavioral Social Sciences in Barcelona, the EPSA 2017 in Milan, the 2017 seminar series of the School of Politics and International Relations at University College Dublin, and the seminar series of the Southern Africa Labour and Development Research Unit (SALDRU) at the University of Cape Town. All errors are ours.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research for this paper was funded by the German Research Foundation for the project “Local Conflict and the Local State” (JU 2979/2-1).

ORCID iD

Eva Wegner

Notes

Supplemental Material

Replication data can be accessed at: . Supplemental materials for this article are available with the manuscript on the Political Research Quarterly (PRQ) website.

References

Alinsky

Saul D.

2013. Rules for Radicals: A Practical Primer for Realistic Radicals. New York: Vintage Books.

Anderson

Cameron

John

Oliver P.

Keltner

Dacher

. 2012. “The Personal Sense of Power.” Journal of Personality 80 (2): 313–44.

Bamert

Justus

Gilardi

Fabrizio

Wasserfallen

Fabio

. 2015. “Learning and the Diffusion of Regime Contention in the Arab Spring.” Research & Politics 2 (3): 1–9.

Benford

Robert D.

Snow

David A.

2000. “Framing Processes and Social Movements: An Overview and Assessment.” Annual Review of Sociology 26:611–39.

Blumer

Herbert

. 1969. “Social Movements.” In Studies in Social Movements: A Social Psychological Perspective, edited by McLaughlin

Barry

, 8–20. New York: The Free Press.

Carlier

Bob

. 1977. “Social Movements as a Type of Reaction to the Minority Situation: A Literature Survey.” Philosophica 20:35–64.

Chenoweth

Erica

Ulfelder

Jay

. 2017. “Can Structural Conditions Explain the Onset of Nonviolent Uprisings?” Journal of Conflict Resolution 61 (2): 298–324.

De Juan

Alexander

Wegner

Eva

. 2019. “Social Inequality, State-centered Grievances, and Protest: Evidence from South Africa.” Journal of Conflict Resolution 63 (1): 31–58.

Ennis

James G.

Schreuer

Richard

. 1987. “Mobilizing Weak Support for Social Movements: The Role of Grievance, Efficacy, and Cost.” Social Forces 66 (2): 390–409.

10.

Finkel

Steven E.

Muller

Edward N.

Opp

Karl-Dieter

. 1989. “Personal Influence, Collective Rationality, and Mass Political Action.” American Political Science Review 83 (3): 885–903.

11.

Franklin

James C.

2009. “Contentious Challenges and Government Responses in Latin America.” Political Research Quarterly 62 (4): 700–14.

12.

Gamson

William A.

1992. Talking Politics. Cambridge: Cambridge University Press.

13.

Gurr

Ted

. 1970. Why Men Rebel. Princeton: Princeton University Press.

14.

Hale

Henry E.

2013. “Regime Change Cascades: What We Have Learned from the 1848 Revolutions to the 2011 Arab Uprisings.” Annual Review of Political Science 16 (1): 331–53.

15.

Harris

Adam S.

Hern

Erin

. 2019. “Taking to the Streets: Protest as an Expression of Political Preference in Africa.” Comparative Political Studies 52 (8): 1169–99.

16.

Hassell

Hans J. G.

Settle

Jaime E.

2017. “The Differential Effects of Stress on Voter Turnout.” Political Psychology 38 (3): 533–50.

17.

Javeline

Debra Lynn

. 2009. Protest and the Politics of Blame: The Russian Response to Unpaid Wages. Ann Arbor: University of Michigan Press.

18.

Johnson

India R.

Fujita

Kentaro

. 2012. “Change We Can Believe in Using Perceptions of Changeability to Promote System-change Motives over System-justification Motives in Information Search.” Psychological Science 23 (2): 133–40.

19.

Jost

John T.

Banaji

Mahzarin R.

Nosek

Brian A.

2004. “A Decade of System Justification Theory: Accumulated Evidence of Conscious and Unconscious Bolstering of the Status Quo.” Political Psychology 25:881–919.

20.

Jost

John T.

Glaser

Jack

Kruglanski

Arie W.

Sulloway

Frank J.

2003. “Political Conservatism as Motivated Social Cognition.” Psychological Bulletin 129 (3): 339–75.

21.

Kalinin

Kirill

Vogel

Sasha

. 2016. “Measuring Propensity of Individual Anti-government Protest Behavior in Autocracies.” https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2767663.

22.

Klandermans

Bert

. 2008. “The Demand and Supply of Participation: Social Psychological Correlates of Participation in Social Movements.” In The Blackwell Companion to Social Movements, edited by Snow

David A.

Soule

Sarah A.

Kriesi

Hanspeter

, 361–79. Hoboken, NJ: John Wiley.

23.

Klein

Graig R.

Regan

Patrick M.

2018. “Dynamics of Political Protests.” International Organization 72 (2): 485–521.

24.

Kuran

Timur

. 1989. “Sparks and Prairie Fires: A Theory of Unanticipated Political Revolution.” Public Choice 61:41–74.

25.

Major

Brenda

Schmader

Toni

. 2001. “Legitimacy and the Construal of Social Disadvantage.” In The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup Relations, edited by Jost

John T

Major

Brenda

, 176–204. New York: Cambridge University Press.

26.

McCarthy

John D.

Zald

Mayer N.

1977. “Resource Mobilization and Social Movements: A Partial Theory.” American Journal of Sociology 82 (6): 1212–41.

27.

Muller

Edward N.

Jukam

Thomas O.

1983. “Discontent and Aggressive Political Participation.” British Journal of Political Science 13:159–79.

28.

Oberschall

Anthony

. 1973. Social Conflict and Social Movements. London: Pearson Education.

29.

Pellicer

Miquel

Piraino

Patrizio

Wegner

Eva

. 2019. “Perceptions of Inevitability and Demand for Redistribution: Evidence from a Survey Experiment.” Journal of Economic Behavior & Organization 159:274–88.

30.

Rootes

Christopher

. 2013. “From Local Conflict to National Issue: When and How Environmental Campaigns Succeed in Transcending the Local.” Environmental Politics 22 (1): 95–114.

31.

Runciman

Carin

Alexander

Peter

Rampedi

Mahlatse

Moloto

Boikanyo

Maruping

Boitumelo

Khumalo

Eunice

Sibanda

Sehlaphi

. 2016. Counting Police-recorded Protests: Based on South African Police Service data. Johannesburg: Social Change Research Unit, University of Johannesburg.

32.

Sunstein

Cass R.

2005. Why Societies Need Dissent. Cambridge, MA: Harvard University Press.

33.

Thomas

Dani B.

Craig

Richard B.

1973. “Student Dissent in Latin America: Toward a Comparative Analysis.” Latin American Research Review 8 (1): 71–96.

34.

Van Der Toorn

Jojanneke

Feinberg

Matthew

Jost

John T.

Kay

Aaron C.

Tyler

Tom R.

Willer

Robb

Wilmuth

Caroline

. 2015. “A Sense of Powerlessness Fosters System Justification: Implications for the Legitimation of Authority, Hierarchy, and Government.” Political Psychology 36 (1): 93–110.

35.

Van Stekelenburg

Jacquelien

Klandermans

Bert

. 2013. “The Social Psychology of Protest.” Current Sociology 61 (5–6): 886–905.

36.

Van Zomeren

Martijn

Postmes

Tom

Spears

Russell

. 2008. “Toward an Integrative Social Identity Model of Collective Action: A Quantitative Research Synthesis of Three Socio-psychological Perspectives.” Psychological Bulletin 134 (4): 504–35.

37.

Verhulst

Joris

. 2011. “Mobilizing Issues and the Unity and Diversity of Protest Events.” Doctoral thesis, University of Antwerpen. https://research.vu.nl/en/publications/mobilizing-issues-and-the-unity-and-diversity-of-protest-events.

38.

Weyland

Kurt

. 2009. “The Diffusion of Revolution: ‘1848’ in Europe and Latin America.” International Organization 63 (3): 391–423.

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

0.23 MB