Authoritarianism,Institutional Confidence,and Willingness to Engage in Collective Action

Abstract

The antecedents of collective action have received considerable attention in psychology, political science, and sociology. However, few studies have addressed the extent to which individual differences in psychological needs, motives, and traits predict collective action tendencies. In the present study, we focus on an especially important individual difference: authoritarianism. We examined three key hypotheses: (1) that authoritarianism would be associated with lower willingness to engage in collective action (net of other factors known to predict protest), (2) that the negative relationship between authoritarianism and collective action would be stronger among the politically engaged; and (3) that the negative relationship between authoritarianism and collective action would be weaker among those who lacked confidence in major social institutions. Using data from three independent waves of the World Values Survey, we find cross-national evidence supporting all three hypotheses.

Keywords

collective action authoritarianism institutional confidence protest

Historically, collective political action has been a force for societal change (Tarrow, 2011). A variety of social movements (e.g., civil rights, environmental, women’s rights, etc.) have influenced political processes through collective action, (CA, hereafter), and these movements have often been effective in shaping policy outcomes (see Amenta, Caren, Chiarello, & Su, 2010). However, although CA provides an avenue through which citizens can seek to redress grievances against social, political, and economic institutions, individuals vary greatly in the extent to which they are willing to participate in CA. Moreover, individuals often opt out of participating in CA, even when such action has the potential to advance their material interests (see Olson, 1965). Given the political impact of social movements, as well as their capacity to hold institutions more accountable to citizens who are otherwise underrepresented and marginalized (see Schlozman, Verba, & Brady, 2012), it is important for scholars to understand the contextual- and individual-level antecedents of participation in these movements and their CA efforts.

In the present study, we focus on individuals’ openness to CA in general, as opposed to their support for particular movements. Although participation in CA depends on the symbolic and instrumental goals of specific social movements (Benford & Snow, 2000), it remains important to understand predictors of the extent to which individuals vary in their general attitude toward CA as a political strategy. Surprisingly, very little work has directly addressed this question, particularly with respect to the predictive power of individual differences in basic psychological needs, motives, and traits. The present analysis attempts to fill this gap. In particular, we focus on role of authoritarianism—a basic tendency to value conformity and deference to established authorities and social norms—as predictor of willingness to engage in CA (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950; Altemeyer, 1996; Hetherington & Weiler, 2009; Stenner, 2005). We examine three key hypotheses. First, we predict that the preference for order and conformity characteristic of those high in authoritarianism will leave them relatively averse to CA. Second, we predict that this relationship will be stronger among those high in political engagement, who should best understand the incompatibility between strong preferences for social order and CA. Third, we predict that the relationship between authoritarianism and CA will be stronger when respondents lack confidence in major social institutions. As prior research has shown, people are more likely to engage in CA when they have low confidence in social institutions, such as legislatures, political parties, the criminal-justice system, and powerful economic actors (Dalton, 2002; Inglehart, 1990; Jenkins, Wallace, & Fullerton, 2008). However, we argue that CA is likely to be differentially attractive in the face of low institutional confidence to those low and high in authoritarianism. When people are generally confident in social and political institutions, both high and low authoritarians should be relatively unwilling to participate in CA, because high institutional confidence suggests a lack of impetus for CA. In contrast, when institutional confidence is low, differences in authoritarianism should be a strong predictor of CA tendencies: Authoritarianism should act as a buffer against engaging in activities that upset the status quo, such as CA. We review the relevant literature and develop these hypotheses in more detail below.

Predictors of Willingness to Engage in CA

Social scientists have long been interested in the antecedents of CA, and they have analyzed the phenomenon at a variety of levels. Some models focus on macro-political variables that facilitate or impede CA. In this vein, proponents of resource mobilization theory (e.g., McCarthy & Zald, 1977) argue that CA is more likely when social movements are more effectively able to mobilize monetary resources, social networks, and communications in the pursuit of political goals. From a different perspective, advocates of deprived actor models predict that CA will be more widespread in the face of high levels of inequality, given its tendency to breed discontent (Gurr, 1970). However, other perspectives suggest that inequality should reduce CA, as larger imbalances in power and resources make it more difficult for those at the bottom to mobilize and easier for those at the top to repress CA (Parvin, 1973).

All things considered, macro-level approaches of this sort have proven limited in their utility, given contradictory predictions and inconclusive evidence. Thus, many researchers have focused instead on the social-psychological bases of CA. For example, drawing on resource mobilization theory’s underlying assumption that individuals make decisions about CA in a strategic fashion, researchers have found that perceived personal or political efficacy predicts increased willingness to engage in protest (Berman & Wittig, 2004; Brunsting & Postmes, 2002; Corcoran, Pettinicchio, & Young, 2011; Klandermans, 1984), though this relationship is not always strong (see Hornsey et al., 2006). Other researchers find that identification with social movements and politicized social groups is a strong predictor of willingness to engage in CA (Drury & Reicher, 2005; Reicher, 2001; Simon & Klandermans, 2001; Simon et al., 1998; Thomas, Mavor, & McGarty, 2012; van Zomeren, 2013; van Zomeren, Postmes, & Spears, 2008). Finally, a third approach argues that CA stems from a perception that established institutions do not function in a legitimate fashion or give rise to a just distribution of resources (Gurr, 1970). Specifically, the perception that major institutions cannot be trusted to operate in a fair fashion or produce fair outcomes may motivate people to engage in CA in an effort to elicit change. Consistent with this basic logic, prior research generally suggests that individuals who express low confidence or trust in key social institutions are more likely to engage in political protest (Dalton, 2002; Inglehart, 1990; Jenkins et al., 2008; see also van Zomeren et al., 2008). Similarly, individuals who believe that their groups are deprived relative to others—a belief that often goes hand in hand with doubts about the legitimacy of major institutions—are also more likely to engage in protest (Abeles, 1976; Pettigrew, 1967; Walker & Mann, 1987; see also Runciman, 1966).

Authoritarianism and CA

While all of these variables are important, we also argue that a complete analysis of the antecedents of CA needs to consider the role of generalized needs, motives, and traits that influence individuals’ willingness to challenge established institutions and ways of doing things. Recent work in political psychology has drawn extensive attention to the relationship between general psychological variables and political attitudes and behaviors (Federico, 2015; Gerber, Huber, Doherty, & Dowling, 2011; Jost, Federico, & Napier, 2009; Mondak, 2010). This literature suggests that psychological needs, motives, and traits that reflect needs for security, order, and certainty are associated with a preference for the political status quo. In many of these studies, the outcome of interest is ideological conservatism as opposed to support for protest (e.g., Carney, Jost, Gosling, & Potter, 2008; Jost, Glaser, Kruglanski, & Sulloway, 2003). However, we argue by extension that many of the same psychological characteristics should predict a reluctance to engage in CA, as CA almost by definition involves a challenge to established systems of authority and ways of doing things.

In offering this prediction, we focus specifically on what is perhaps the individual difference most directly relevant to individuals’ attitudes toward socially disruptive acts of protest: authoritarianism (Adorno et al., 1950; Altemeyer, 1996; Stenner, 2005). Authoritarianism is currently understood as a general preference for conformity and deference to established authorities (Duckitt & Sibley, 2010; Feldman, 2003; Stenner, 2005). The study of authoritarianism has a rich history in social and political psychology. As is well known, early work on the construct was grounded in the psychoanalytic tradition and suggested that authoritarianism develops in early childhood in response to harsh, rigid childrearing (Fromm, 1941). In this conceptualization, authoritarianism reflects rigid adherence to social conventions, combined with a desire to inflict punishment on those who deviate from these conventions (Adorno et al., 1950). After a period of dormancy, Altemeyer (1996) rehabilitated the concept, dispensing with the earlier psychoanalytic framework and developing a new measure of “right-wing authoritarianism” (RWA), which focused on three core tendencies: authoritarian aggression (i.e., punitiveness toward deviants), authoritarian submission (i.e., deference to authorities), and conventionalism (i.e., adherence to norms and traditions).

Although the RWA scale has been used extensively, it has been criticized for including items that tap attitudes that RWA is supposed to predict, such as opinions about homosexuality and religion (Feldman, 2003; Hetherington & Weiler, 2009; Stenner, 2005). Indeed, with respect to our own focus, some items on the RWA scale appear to directly tap attitudes toward CA. For example, one item states the following: “The established authorities generally turn out to be right about things, while the radicals and protestors are usually just ‘loud mouths’ showing off their ignorance” (see Altemeyer, 1996). Of course, because of items like this, using RWA as a predictor of CA would be problematic. Given problems like this—and the deeper theoretical point that authoritarianism has typically been conceptualized as a pre-political antecedent of political attitudes and behavior (rather than as a cluster of political attitudes per se)—recent research on authoritarianism has instead used measures of childrearing values that assesses general preferences for obedience and conformity versus independence and autonomy as important values to instill in children (Federico, Fisher, & Deason, 2011; Federico & Reifen Tagar, 2014; Hetherington & Weiler, 2009; Stenner, 2005). We adopt this strategy in our own analyses, as we describe below.

Even setting aside measurement issues, almost no research has looked directly at the relationship between authoritarianism and CA. There are a few notable exceptions. Lemieux and Asal (2010), for instance, assessed the influence of RWA on support for peaceful and violent forms of political action in response to reading a vignette that manipulated level of grievance. They found that respondents who were high in RWA were generally less inclined to support any form of political action. In addition, Zhang (2012) found that RWA predicted reduced political participation in Singapore. Gutting (2015) offered perhaps the most robust assessment of the influence of authoritarianism on CA in the United States. Using four American datasets, Gutting shows that authoritarianism—measured using childrearing values—predicts reduced participation in CA. Although these findings are suggestive, they are not without limitations. First, they are based solely on data from single nations (i.e., Singapore and the United States), leaving open the question of authoritarianism and CA are related across national contexts and political cultures. Second, the analyses fail to control for a number of other social-psychological factors that have been shown to predict CA, including efficacy, membership in politically relevant groups, and confidence in social institutions. Third, two of the studies—by Lemieux and Asal and Zhang—rely on the RWA scale, introducing the problem of tautological measurement described previously. In light of these issues, we provide a more robust test of the relationship between authoritarianism and CA.

The Interaction Between Political Engagement and Authoritarianism

Importantly, we also argue that this relationship should be moderated by two other variables. The first of these is political engagement (or attention to politics; see Johnston, Lavine, & Federico, in press; Malka, Soto, Inzlicht, & Lelkes, 2014). Expanding on prior research, we hypothesize that the negative relationship between authoritarianism and CA should be especially strong among those high in engagement. It is clear from previous research that not all individuals are equally aware of which political attitudes and behaviors fit with their underlying needs, motives, and traits. In particular, recent work suggests that the relationship between authoritarianism and conservative political identifications are stronger among the politically engaged, who are more likely to understand that conservatism provides a better match to the needs for order and security associated with authoritarianism (Federico et al., 2011; Federico & Reifen Tagar, 2014; Malka et al., 2014; see also Converse, 1964; Zaller, 1992).

Moreover, politically engaged individuals are also more attuned to the symbolic and expressive aspects of their political attitudes and behavior—that is, what the latter signal about their values and identities (Johnston et al., in press). If the engaged are more likely to care that refraining from or engaging in CA potentially says something about how much they value order versus change, then we should expect them to more precisely “select” the level of support for CA that matches their position on the authoritarianism dimension. Given that politically engaged individuals are more likely to understand the socially disruptive implications of CA and care about what CA might symbolically indicate about their own convictions, we predict that the same will be true with respect to the relationship between authoritarianism and CA.

The Interaction Between Institutional Confidence and Authoritarianism

The second moderator we focus on is one of the key antecedents of CA reviewed earlier: confidence in major social institutions. As noted previously, individuals who express low confidence in major social institutions are more prone to CA (Dalton, 2002; Inglehart, 1990; Jenkins et al., 2008; see also van Zomeren et al., 2008). Besides this well-established main effect, we believe that it may also moderate the impact of psychological variables like authoritarianism. Specifically, low institutional confidence may represent the kind of normative threat that leads high authoritarians to default to the status quo (Stenner, 2005). When confidence is high, no one is likely to be motivated to protest, leading to a relatively low level of support for CA that is uniform across the authoritarianism spectrum (i.e., a null relationship between authoritarianism and CA). In contrast, a lack of confidence in major institutions may seem especially threatening to social cohesion in and of itself among those high in authoritarianism. As these threats prompt efforts to restore order among high authoritarians, they should therefore be averse to participating in CA. This implies that the negative relationship between authoritarianism and CA should be stronger to the extent that people lack confidence in social institutions.

Breaking the interaction between authoritarianism and institutional confidence down differently, authoritarianism can also be thought of as a buffer against discontent when institutions lose their legitimacy. In particular, seemingly illegitimate institutions may motivate CA to a lesser degree among those high in authoritarianism. Given that CA has the potential to create social disorder and challenge established authorities and institutions, authoritarians should be more wary of the disruptive aspects of CA even in the face of institutions that fail to inspire confidence. In other words, the threat introduced by potential disorder may outweigh the perceived illegitimacy of the status quo among authoritarians, reducing motivation to engage in CA. In contrast, those low in authoritarianism should feel no such qualms, leading them to see CA as a more appropriate response to low confidence in institutions.

The Present Research

The purpose of the present study is to examine the predictions outlined above. To overcome the aforementioned shortcomings of previous research, we use cross-national data from three waves of the World Values Survey (WVS) and operationalize authoritarianism in terms of respondents’ childrearing preferences. We also control for a number of other predictors theoretically relevant to willingness to engage in CA, including nation-level variables (i.e., human development and democratic rights) and individual-level variables (i.e., efficacy, membership in politically relevant groups). For the sake of clarity, our core hypotheses are restated below:

Hypothesis 1: Authoritarianism should be negatively associated with willingness to engage in CA, and this relationship should be especially strong among those high in political engagement.

Hypothesis 2: Political engagement should moderate the relationship between authoritarianism and willingness to engage in CA, such that the tendency for authoritarianism to be associated with reduced CA will be stronger among those high in engagement.

Hypothesis 3: Confidence in institutions should moderate the relationship between authoritarianism and willingness to engage in CA, such that the tendency for authoritarianism to be associated with reduced CA will be stronger among those who lack confidence in social and political institutions.

Method

Data

Data were taken from Waves 4 (1999-2004), 5 (2005-2009), and 6 (2010-2014) of the WVS; each wave constitutes an independent cross-sectional dataset. Each WVS is conducted by social scientists from across the globe. Each national survey translates the English-language root WVS questionnaire into local languages and reaches a representative sample of that nation’s adults for face-to-face interviews. In each participating nation, the root survey is translated into all languages spoken by at least 15% of the population to ensure that the survey widely represents the nation’s various ethno-linguistic groups. In each wave, we used data only from those national samples in which all relevant variables were assessed, leaving us with samples from 31 nations in Wave 4, 45 nations in Wave 5, and 50 nations in Wave 6. In these nations, complete data for all variables were available for n = 32,214 respondents in Wave 4, n = 45,405 respondents in Wave 5, and n = 56,960 respondents in Wave 6.¹

Measures

We describe our measures below. More detailed information on the variables for each WVS wave can be found in the online appendix, and intercorrelations between the key individual-level variables in each dataset can be found in Table 1.

Table 1.

Correlations Between Key Individual-Level Variables.

Variables	1	2	3	4	5
World Values Survey, Wave 4
1. CA	1.00
2. Efficacy	.06***	1.00
3. Political engagement	.25***	.04***	1.00
4. Confidence in institutions	−.07***	−.02***	.23***	1.00
5. Authoritarianism	−.19***	−.05***	−.11***	.04***
World Values Survey, Wave 5
1. CA	1.00
2. Group memberships	.17***	1.00
3. Efficacy	.05***	.06***	1.00
4. Political engagement	.24***	.19***	.05***	1.00
5. Confidence in institutions	−.05***	.13***	.09***	.23***	1.00
6. Authoritarianism	−.16***	.01**	−.02***	−.10***	−.02***
World Values Survey, Wave 6
1. CA	1.00
2. Group memberships	.20***	1.00
3. Efficacy	.03***	.04***	1.00
4. Political engagement	.23***	.18***	.03***	1.00
5. Confidence in institutions	−.02***	.12***	.05***	.21***	1.00
6. Authoritarianism	−.13***	.02***	.0002	−.09***	−.03***

Note. Entries are Pearson correlation coefficients. N = 32,214, in Wave 4; N = 45,405, in Wave 5; N = 56,960, in Wave 6. CA = collective action.

***

p < .001.

Willingess to engage in CA

The dependent variable was operationalized using WVS items asking about willingness to participate in three different political actions: “signing a petition,” “joining in boycotts,” and “attending peaceful demonstrations.” Responses to each were given on a 3-point scale: have done, might do, and would never do. Scores were reversed, recoded to run from 0 to 1, then averaged (α = .72, in Wave 4; α = .77, in Wave 5; α = .75, in Wave 6). Higher scores indicate a greater willingness to engage in political action (M = 0.33, SD = 0.30, in Wave 4; M = 0.37, SD = 0.31, in Wave 5; M = 0.29, SD = 0.29, in Wave 6).

Self-efficacy

Self-efficacy was assessed with a single item used in previous studies of CA (Acevedo, 2008; Corcoran et al., 2011). This item asks individuals to indicate how much choice and control they perceive themselves to have in life on a scale ranging from none at all (1) to a great deal (10). Scores were recoded to run from 0 to 1; higher scores indicate greater efficacy (M = 0.63, SD = 0.30, in Wave 4; M = 0.69, SD = 0.24, in Wave 5; M = 0.69, SD = 0.24, in Wave 6).

Group membership

Given that group memberships and associated social identities are an important antecedent of CA (e.g., van Zomeren, 2013), we also included a composite measure of membership and activity in a series of politically relevant groups in Waves 5 and 6; the measure was not available in Wave 4. The measure was based on responses to items in which respondents were given a list of voluntary organizations and asked whether they participated in each one. We used responses regarding five organizations: church or religious organizations, labor unions, political parties, environmental organizations, or charitable/humanitarian organizations. There were three response options: active member (coded 1), inactive member (coded 0.5), and don’t belong (coded 0). In both Waves 5 and 6, these items formed a scale (α = .65, in Wave 5; α = .66, in Wave 6), so they were averaged to form a membership scale (M = 0.17, SD = 0.21, in Wave 5; M = 0.14, SD = 0.19, in Wave 6).

Political engagement

This measure was a composite of responses to two items asking (a) how important politics was in the respondent’s life, answered on a scale from 1 (very important) to 4 (not at all important); and (b) how interested in politics the respondent was, answered on a scale ranging from 1 (very interested) to 4 (not at all interested). These items correlated highly (r = .56, in Wave 4; r = .58, in Wave 5; r = .54, in Wave 6), so they were reversed, averaged, and recoded to run from 0 to 1 (M = 0.47, SD = 0.29, in Wave 4; M = 0.48, SD = 0.28, in Wave 5; M = 0.47, SD = 0.28, in Wave 6).

Confidence in institutions

A composite measure of respondents’ confidence in major national social institutions was generated using responses to questions about how much confidence they had in a series of entities: the armed forces, the press, the police, the national parliament, the national government as a whole, political parties, major companies, and the justice system; Wave 4 did not include the item asking about the justice system. Responses were given on a scale ranging from 1 (a great deal) to 4 (none at all). Responses were reversed, recoded to run from 0 to 1, and then averaged (α = .81, in Wave 4; α = .86, in Wave 5; α = .85, in Wave 6). Higher scores indicate greater overall confidence in national institutions (M = 0.50, SD = 0.22, in Wave 4; M = 0.49, SD = 0.21, in Wave 5; M = 0.47, SD = 0.21, in Wave 6).

Authoritarianism

To assess authoritarianism, we follow others by using childrearing values as an index of authoritarianism (Federico et al., 2011; Federico & Reifen Tagar, 2014; Hetherington & Weiler, 2009; Stenner, 2005). In contrast to other measures of authoritarianism (e.g., the RWA scale; Altemeyer, 1996), these measures are largely pre-political and avoid confounding the predisposition to favor group authority and conformity with its political correlates (e.g., intolerance against specific groups, social conservatism, etc.). Specifically, we used an index originally developed by Stenner (2005) for use in the WVS and subsequently employed in a variety of later studies (Brandt & Henry, 2012; Dunn, 2013; Dunn & Singh, 2014; Singh & Dunn, 2013). The index was based on responses to a series of checklist items asking about “qualities that children can be encouraged to learn at home.” Respondents could indicate up to five they considered “especially important.” To generate the Authoritarianism scale, responses to four qualities were considered: obedience, good manners, imagination, and independence. Responses were scored 1 if the quality was endorsed and 0 otherwise. To generate the scale, scores on imagination and independence were reversed, and all responses were summed; the sums were then recoded to run from 0 to 1 (M = 0.49, SD = 0.28, in Wave 4; M = 0.39, SD = 0.24, in Wave 5; M = 0.55, SD = 0.31, in Wave 6).²

Nation-level controls

We also considered two nation-level variables with structural implications for citizens’ willingness to engage in CA. First, as an index of the degree to which a society provides individuals with a positive capacity to act on the world, we used the United Nations Human Development Index (HDI). We used 2001 HDI for the Wave 4 data, 2005 HDI for Wave 5, and 2010 HDI for Wave 6. For each nation, this index is a composite of life expectancy at birth, mean years of schooling, and per capita income; it can be taken as an index of the degree to which social conditions materially allow individuals to be what they want to be and do what they want to do (see United Nations Development Programme, 2006). Second, as an indicator of the degree to which a society provides individuals with the political freedom to engage in CA, we used a composite of each country’s Freedom House rating political rights and civil liberties ratings in 2005. These ratings are based on a “multilayered process of analysis and evaluation by a team of regional experts and scholars” (https://freedomhouse.org/report/freedom-world-2012/methodology). These paired indices were taken from 1999 for Wave 4, 2005 for Wave 5, and 2010 for Wave 6. Both indices are scored on a 1 to 7 scale in their original form, with higher scores indicating that the nation does a poorer job guaranteeing political rights and civil liberties. These indices correlated highly (r = .92, in Wave 4; r = .95, in Wave 5; r = .95, in Wave 6), so they were averaged to form a single index in each dataset. Scores were then reversed and recoded to run from 0 to 1, so higher scores indicate a better record on rights and liberties.

Demographics

We also included several demographic variables: income decile, education (coded as eight categories; recoded 0-1), gender (0 = female, 1 = male), and age (in years; recoded 0-1).

Results

Given the nested nature of our data—with individual respondents nested in national samples—we estimated a series of two-step multilevel linear mixed models in each dataset to examine our hypotheses. In these models, individual respondents were the Level 1 units, and nations were the Level 2 units. In the first step (Model 1 in each dataset), we regressed willingness to engage in CA on the two nation-level controls (HDI and Freedom House rating for the relevant survey years), the demographic controls, self-efficacy, group membership, political engagement, confidence in institutions, and authoritarianism. In the second step (Model 2 in each dataset), we add the Authoritarianism × Engagement and Authoritarianism × Confidence in Institutions interactions. In all models, we allow the intercept and the slopes for self-efficacy, group membership, political engagement, confidence in institutions, and authoritarianism to vary randomly across nations.³ The models were largely identical across all three WVS datasets; the one exception was the absence of the membership variable in Wave 4. The results of these analyses are summarized in Tables 2 through 4. We leave all predictors in their recoded 0 to 1 metric; no centering was used. Given the 0 to 1 coding of all variables, each coefficient can be interpreted as an effect size, that is, the proportion change in the dependent variable associated with moving from the lowest to the highest value of each independent variable.⁴

Table 2.

Willingness to Engage in CA as Function of Political Engagement and Confidence in Institutions (World Values Survey, Wave 4).

	Willingness to engage in CA
	Model 1		Model 2
Predictor	b	SE	b	SE
Country-level predictors
HDI (2001)	0.16	(0.09)	0.14	(0.09)
Freedom House rating (1999)	0.16	(0.10)	0.15	(0.10)
Individual-level predictors
Income decile	0.05***	(0.01)	0.05***	(0.01)
Education	0.11***	(0.01)	0.11***	(0.01)
Gender	0.04***	(0.003)	0.04***	(0.003)
Age	−0.02*	(0.01)	−0.02*	(0.01)
Self-efficacy	0.003	(0.01)	0.003	(0.01)
Political engagement	0.23***	(0.02)	0.25***	(0.03)
Confidence in institutions	−0.07**	(0.02)	−0.12***	(0.03)
Authoritarianism	−0.03**	(0.01)	−0.07***	(0.02)
Authoritarianism × Engagement	—	—	−0.04	(0.03)
Authoritarianism × Confidence	—	—	0.12***	(0.03)
Intercept	−0.003	(0.06)	0.04	(0.06)
Random-effect variance components
Intercept	0.01	(0.004)	0.01	(0.003)
Self-efficacy	0.002	(0.001)	0.002	(0.001)
Political engagement	0.01	(0.004)	0.02	(0.004)
Confidence in institutions	0.02	(0.005)	0.02	(0.005)
Authoritarianism	0.003	(0.001)	0.001	(0.001)
Authoritarianism × Engagement	—	—	0.004	(0.003)
−2 × log-likelihood	2,725.88		2,697.47
Wald χ² (degrees of freedom)	1,227.72 (10)***		1,256.83 (12)***

Note. Entries are coefficients from a multilevel linear mixed model. Number of Level 1 units = 32,214; number of Level 2 units = 31. CA = collective action; HDI = Human Development Index.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table 3.

Willingness to Engage in CA as Function of Political Engagement and Confidence in Institutions (World Values Survey, Wave 5).

	Willingness to engage in CA
	Model 1		Model 2
Predictor	b	SE	b	SE
Country-level predictors
HDI (2005)	0.16^†	(0.08)	0.16^†	(0.08)
Freedom House rating (2005)	0.15^†	(0.09)	0.14^†	(0.08)
Individual-level predictors
Income decile	0.03***	(0.01)	0.03***	(0.01)
Education	0.14***	(0.005)	0.13***	(0.005)
Gender	0.02***	(0.002)	0.02***	(0.002)
Age	−0.07***	(0.01)	−0.07***	(0.01)
Self-efficacy	0.01	(0.01)	0.01	(0.01)
Membership	0.20***	(0.02)	0.20***	(0.02)
Political engagement	0.24***	(0.01)	0.28***	(0.02)
Confidence in institutions	−0.09***	(0.01)	−0.14***	(0.02)
Authoritarianism	−0.08***	(0.01)	−0.09***	(0.02)
Authoritarianism × Engagement	—	—	−0.10***	(0.02)
Authoritarianism × Confidence	—	—	0.13***	(0.02)
Intercept	0.01	(0.06)	0.02	(0.06)
Random-effect variance components
Intercept	0.01	(0.003)	0.01	(0.003)
Self-efficacy	0.003	(0.001)	0.003	(0.001)
Membership	0.01	(0.003)	0.01	(0.003)
Political engagement	0.01	(0.002)	0.01	(0.002)
Confidence in institutions	0.01	(0.002)	0.01	(0.002)
Authoritarianism	0.004	(0.001)	0.004	(0.001)
−2 × log-likelihood	5,031.50		4,997.24
Wald χ² (degrees of freedom)	2,222.09 (11)***		2,280.34 (13)***

Note. Entries are coefficients from a multilevel linear mixed model. Number of Level 1 units = 45,405; number of Level 2 units = 45. CA = collective action; HDI = Human Development Index.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Table 4.

Willingness to Engage in CA as Function of Political Engagement and Confidence in Institutions (World Values Survey, Wave 6).

	Willingness to engage in CA
	Model 1		Model 2
Predictor	b	SE	b	SE
Country-level predictors
HDI (2010)	0.19**	(0.06)	0.19**	(0.06)
Freedom House rating (2010)	0.16**	(0.05)	0.16**	(0.05)
Individual-level predictors
Income decile	−0.01*	(0.005)	−0.01*	(0.005)
Education	0.12***	(0.004)	0.12***	(0.004)
Gender	0.02***	(0.002)	0.02***	(0.002)
Age	−0.06***	(0.01)	−0.06***	(0.01)
Self-efficacy	0.01	(0.01)	0.01	(0.01)
Membership	0.25***	(0.03)	0.25***	(0.03)
Political engagement	0.20***	(0.01)	0.22***	(0.01)
Confidence in institutions	−0.06***	(0.02)	−0.10***	(0.02)
Authoritarianism	−0.04***	(0.01)	−0.05***	(0.01)
Authoritarianism × Engagement	—	—	−0.04**	(0.01)
Authoritarianism × Confidence	—	—	0.06***	(0.02)
Intercept	−0.06	(0.04)	−0.05	(0.04)
Random-effect variance components
Intercept	0.01	(0.002)	0.01	(0.002)
Self-efficacy	0.002	(0.001)	0.002	(0.001)
Membership	0.03	(0.010)	0.03	(0.007)
Political engagement	0.01	(0.001)	0.01	(0.001)
Confidence in institutions	0.01	(0.003)	0.01	(0.003)
Authoritarianism	0.002	(0.001)	0.01	(0.001)
−2 × log-likelihood	336.58		319.59
Wald χ² (degrees of freedom)	2,041.84 (11)***		2,068.15 (13)***

Note. Entries are coefficients from a multilevel linear mixed model. Number of Level 1 units = 56,960; number of Level 2 units = 50. CA = collective action; HDI = Human Development Index.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Standard Predictors of Willingness to Engage in CA

We begin by examining the predictive power of the standard variables that serve as our controls. The relevant coefficients can be found in Model 1 in Tables 2 through 4. Looking first at the nation-level controls, we see that both HDI and political freedom were associated with greater willingness to engage in CA, though the relationships reached significance only in Wave 6 (ps < .01). With respect to demographics, those with higher levels of education (ps < .001) and males (ps < .001) were consistently more willing to engage in CA across all three datasets. Moreover, younger respondents were consistently more willing to engage in CA across the three WVS waves (ps < .01). However, the relationship between income and CA varied across datasets: In Waves 4 and 5, it was positively associated with CA, as expected (ps < .001), but showed a weaker negative relationship with CA in Wave 6. Turning to other variables, self-efficacy failed to predict CA in any of the datasets (ps > .20). The strongest individual predictors of increased willingness to engage in CA were group membership (ps < .001, in Waves 5 and 6) and political engagement (ps < .001), with coefficients ranging from b = .20 to b = .25. Moreover, as expected, individuals who had less confidence in their nations’ major institutions were consistently more willing to engage in CA in all three datasets (bs ranging from −.06 to −.09; ps < .01). Thus, the estimates for our controls are consistent with prior theory and research on the antecedents of willingness to engage in CA.

Authoritarianism and Willingness to Engage in CA

Hypothesis 1 predicted that authoritarianism would relate negatively to willingness to engage in CA (net of various controls). The first part of this prediction is addressed in Model 1 in Tables 2 to 4, which contains the main-effect estimates for all predictors. Consistent with Hypothesis 1, authoritarianism was associated with reduced willingness to engage in CA across all three WVS waves (b = −.03, in Wave 4; b = −.08, in Wave 5; b = −.04, in Wave 6; all ps < .01). Although its effects were weaker than those of group membership and political engagement (mean b = −.05), they were robust across all three datasets and persisted despite numerous controls.

Authoritarianism, Engagement, and Willingness to Engage in CA

Hypothesis 2 predicted that the negative relationship between authoritarianism and CA would be stronger among the politically engaged. Consistent with this prediction, the estimates in Model 2 indicate a significant Authoritarianism × Engagement in two of the three datasets (b = −.04, p < .001, in Wave 5; b = −.10, p < .01, in Wave 6); the interaction failed to reach significance in Wave 5 (b = −.04, p = .178). We illustrate these interactions in Figure 1. The graphs in this figure plot conditional effects and confidence intervals for authoritarianism across the full range of political engagement values. Here, we focus on the significant interactions in Waves 5 and 6. In Wave 5, these conditional effects ranged from b = −.03 (SE = 0.01, p < .05) at the lowest level of engagement to b = −.13 (SE = 0.02, p < .001) at the highest level of engagement. In other words, the magnitude of the relationship quadrupled across the full range of engagement among these respondents. In Wave 6, the conditional effects ranged from a marginally significant b = −.02 (SE = 0.01, p = .07) at the lowest engagement level to a significant b = −.06 (SE = .01, p < .001) at the highest. Although the effects in this sample are modest at all levels (to be sure), the magnitude of the relationship nevertheless triples across the full range of engagement. Again, all variables are recoded to run from 0 to 1, so the coefficients indicate the proportion change in CA when moving from the lowest to the highest level authoritarianism. As such, at the highest level of engagement, respondents at the lowest level of authoritarianism were 12% higher in willingness to engage in CA in Wave 5 and 6% higher in Wave 6. Thus, our results are largely consistent with Hypothesis 2: In two out of three cases, the negative relationship between authoritarianism and CA was stronger among the politically engaged—that is, those most attuned to the implications of the desire for order at the heart of authoritarianism for the desirability of CA.

Figure 1.

Conditional effects of authoritarianism on willingness to engage in collective action at different levels of political engagement.

Authoritarianism, Confidence in Institutions, and Willingness to Engage in CA

Our third hypothesis predicted that confidence in institutions should moderate the relationship between authoritarianism and willingness to engage in CA, such that the tendency for authoritarianism to be negatively related to willingness to engage in CA would be stronger among those with less confidence in major social institutions. A formal test of this hypothesis is provided by the Authoritarianism × Confidence interactions in Model 2 in each wave. As the estimates in Tables 2 through 4 indicate, this interaction was significant in all three WVS datasets (b = .12, p < .001, in Wave 4; b = .13, p < .001, in Wave 5; b = .06, p < .01, in Wave 6).

We illustrate these interactions in the graphs on the left side of Figure 2. These graphs plot conditional effects and confidence intervals for confidence in institutions across the full range of authoritarianism values. Consistent with Hypothesis 2, these conditional-effect plots indicate that authoritarianism is associated with reduced collective-action tendencies only when confidence in institutions is at its lowest level; among those with high confidence, authoritarianism is essentially unrelated to willingness to engage in CA. In WVS Wave 4, those high in authoritarianism shied away from CA at the lowest level of institutional confidence (b = −.09, SE = 0.02, p < .001), whereas the relationship was indistinguishable from zero among those at the highest level of authoritarianism (b = .03, SE = 0.02, p > .10). Similarly, in Wave 5, authoritarianism had a relatively strong negative relationship with willingness to engage in CA when confidence in institutions was at its lowest level (b = −.14, SE = 0.02, p < .001), but was essentially unrelated to collective-action tendencies when confidence in institutions was at its highest (b = −.01, SE = 0.02, p > .50). Rounding things out, in Wave 6, the negative relationship between authoritarianism and openness to CA was much stronger among those at the lowest level of confidence (b = −.07, SE = 0.01, p < .001) than it was among those at the highest level of confidence (b = −.002, SE = 0.01, p > .50).

Figure 2.

Conditional effects of authoritarianism on willingness to engage in collective action at different levels of confidence in institutions (left) and conditional effects of confidence in institutions on willingness to engage in collective action at different levels of authoritarianism (right) for World Values Study Waves 4 through 6.

As a reminder, the 0 to 1 coding of these variables means that these estimates can be interpreted as effect sizes. Thus, when confidence in social institutions is at its lowest, going from the minimum to the maximum level of authoritarianism reduces willingness to engage in CA from 7% to 14% of its full range, but this effect dwindles to a 0.2% to 3% reduction at the highest level of authoritarianism. This pattern suggests that authoritarianism becomes relevant to collective-action tendencies depending on the confidence accorded to major social and political institutions. When confidence in these institutions is high, authoritarianism is irrelevant, as those at both ends of the authoritarianism spectrum find little reason to contemplate CA. However, when confidence in institutions is low, individual differences matter: Those low in authoritarianism express a willingness to rebel, whereas those high in authoritarianism—consistent with their needs for order and stability—are more restrained.

As an alternate illustration of the interplay between authoritarianism and confidence in institutions, we break these interactions down in the opposite fashion in the graphs on the right side of Figure 2. Here, we present conditional effects and confidence intervals for confidence in institutions across the full range of authoritarianism values. As predicted, the graphs on the right consistently indicate that the tendency for those with low confidence in national institutions to be more inclined to CA is strongest among those low in authoritarianism, but drops to zero as authoritarianism approaches its maximum. For example, in Wave 5, low confidence was significantly related to willingness to engage in CA among those lowest in authoritarianism (b = −.14, SE = 0.02, p < .001), but unrelated to CA tendencies among those highest in authoritarianism (b = −.01, SE = 0.02, p > .60). Thus, when authoritarianism is at its lowest, going from the minimum to the maximum level of confidence in institutions reduces willingness to engage in CA by 14% of its full range, but this effect dwindles to a 1% reduction at the highest level of authoritarianism. Thus, high levels of authoritarianism provide a buffer against the tendency for low confidence in institutions to be associated with a greater willingness to engage in CA.⁵

Discussion

In this study, we examined three hypotheses about the implications of authoritarianism for individuals’ willingness to engage in CA: (1) that authoritarianism would be associated with a reduced willingness to engage in CA, (2) that this relationship would be stronger among those high in political engagement, and (3) that this relationship would be weaker among those with less confidence in major social institutions. In analyses of data in three different waves of the WVS, we find strong support for our hypotheses. Authoritarianism was associated with reduced willingness to engage in CA in all three datasets, and this relationship was stronger among the politically engaged in two of them. Moreover, in all three datasets, the usual negative relationship between authoritarianism and CA appeared only among those who lacked confidence in institutions.

These results offer a number of contributions to the literatures on CA, authoritarianism, and the psychological foundations of political attitudes and behavior more generally. Above all, they suggest that psychological needs, motives, and traits are relevant to understanding CA, even after considering the role of other factors more commonly cited as antecedents of CA (such as efficacy or membership in politically relevant groups; see van Zomeren, 2013). Specifically, our results suggest that authoritarianism—a very general preference for social order and conformity, measured in our study without direct reference to protest or politics—is consequential in and of itself as a predictor of CA. All other things being equal, individuals with a general predisposition to favor conformity and deference to authority find CA to be relatively unattractive. This not only reinforces other findings (Gutting, 2015; Zhang, 2012) but also extends them by demonstrating that the relationship between authoritarianism and CA persists even when using a measure of authoritarianism that is not conflated with political preferences, accounting for a wider range of control variables, and drawing on datasets from a wider variety of national contexts. Thus, our results add to extant findings regarding relationships between individual psychological differences and CA (e.g., the Big Five; Brandstätter & Opp, 2014; Gallego & Oberski, 2012; see also Gerber et al., 2011; Mondak, 2010), and they also expand a growing literature on the relevance of authoritarianism for mass politics (e.g., Federico et al., 2011; Hetherington & Weiler, 2009; Stenner, 2005) by showing that it predicts not only ideology and issue preferences but also willingness to directly challenge the status quo via protest.

Tests of our second and third hypotheses add to the literature in one other novel way as well: They suggest that the implications of authoritarianism for CA are not equally visible for all individuals. With respect to Hypothesis 2, our results (with one exception) indicated that authoritarianism was associated with reduced CA only among those who were highly engaged. In other words, motivated acceptance or rejection of CA as a political strategy on the basis of one’s authoritarianism level appears to require that citizens be engaged enough with politics to make a judgment about how consistent the idea of CA is with the degree to which they prefer social order and deference to authority over individual freedom and expression. This finding is consistent with other recent findings suggesting that psychological needs, motives, and traits are only likely to have implications for political attitudes and behavior among those whose knowledge of and interest in politics is high enough to ensure an understanding of how adequately different political positions align with their basic psychological inclinations (e.g., Federico et al., 2011; Malka et al., 2014). It is also broadly congruent with research on social movements suggesting that social causes “resonate” only when their fit with basic needs and values is apparent (Benford & Snow, 2000). To the extent that political engagement facilitates understanding of the correspondence between psychological predispositions and political behavior it is not surprising that we observe a resonance between authoritarianism and reluctance to engage in CA only among the engaged.

With respect to Hypothesis 3, our results suggest that authoritarianism was associated with reduced CA mainly among those who lacked confidence in major social institutions. This finding adds to a burgeoning research literature suggesting that authoritarianism may have some of its most important political consequences in interaction with perceptions of social order. Authoritarianism, according to this view, is a propensity to react to social disruption with efforts to restore order (Feldman, 2003; Feldman & Stenner, 1997). As such, threats to perceived normative agreement in society are more likely to result in intolerance of difference among those who are also high in authoritarianism (Stenner, 2005). Our findings offer a parallel to this: The threat of failing institutions—rather than eliciting protest as it does among others—reinforces the tendency for those high in authoritarianism to respond with an order-protecting reticence. Thus, the threat to social order implied by a loss of institutional legitimacy may simply ramp up the tendency for authoritarians to behave in ways that protect order—in this case, by withdrawing from social action. From a broader institutional perspective, these findings therefore suggest that authoritarianism may serve as a buffer against discontent when social institutions are no longer seen as legitimate. As noted in our introduction, lack of confidence in major social institutions is a robust predictor of support for CA (e.g., Dalton, 2002; van Zomeren, 2013). In our data, this appears to be less true among those high in authoritarianism: For these individuals, the connection between low institutional confidence and collective-action tendencies effectively drops to zero. This finding is consistent with the idea that individuals who prefer order and respect for established authority are unlikely to see protest as a desirable response, even if they feel no particular fondness for major institutions. It also echoes findings indicating that authoritarians are more averse to risk and loss, leading the threat of social disorder to loom larger for them than the potential benefits of changing institutions that are perceived to be defective in some way (see Johnston et al., in press; Lavine, Lodge, & Freitas, 2005; Lavine et al., 1999; Lavine, Lodge, Polichak, & Taber, 2002; see also Hibbing, Smith, & Alford, 2014).

Our findings may also shed light on the so-called CA problem, which occurs when people with common goals and interests fail to act collectively to achieve mutually beneficial outcomes (Olson, 1965). The CA problem has conventionally been explicated in terms of rational economic incentives—individuals can benefit from the collective political actions of others without having to commit to CA themselves, thereby incentivizing them to “free-ride” off the efforts of others. Although the free-rider problem offers one explanation as to why individuals may fail to take action to redress grievances against institutions or achieve desired objectives, the prevalence of authoritarianism may offer another. As we have shown, even when confidence in institutions is low, high authoritarians tend to be unwilling to participate in CA. As such, high authoritarians may sometimes contribute to the CA problem through their general reluctance to take action, even when such action may be needed to improve the functioning of institutions that they themselves have come to distrust. Of course, the flip side of this is that when CA efforts are likely to result in more overall harm than good (e.g., by disrupting the functionality of institutions with few tangible benefits), reluctance to participate in CA among high authoritarians may prevent social disorder even when existing institutions fail to inspire confidence.

Limitations and Future Directions

Of course, our analysis is not without its limitations. Most obviously, our data are from cross-sectional surveys, limiting our ability to make causal inferences about our variables with the confidence that would be afforded by true panel data or an experimental manipulation of institutional confidence. A second and more specific concern has to do with the measures we employ. Although the large, diverse samples provided by the WVS were a distinct advantage for our analysis, we were limited to the variables included in the surveys. For example, this left us with measures of efficacy and politicized group memberships that included fewer items and less-tailored content than those used in other social-psychological studies of CA (e.g., van Zomeren, 2013). Moreover, in the case of authoritarianism, we had to rely on items that were not originally designed to tap authoritarianism, and that did not form scales that are as internally consistent as others used in the literature. Although our WVS measure has been extensively used in the political-psychology literature (e.g., Brandt & Henry, 2012; Dunn & Singh, 2014; Singh & Dunn, 2013; Stenner, 2005) and functions similarly to other authoritarianism measures, it does leave something to be desired from an ideal standpoint. That said, we are reassured that the measure reveals consistent, replicable effects for authoritarianism in spite of this limitation. Future research should nevertheless explore our hypotheses with other measures of authoritarianism.

A final issue has to do with when we might expect authoritarians to show stronger support for CA. In the present study, we focus on general support for CA, as opposed to support for CA on behalf of specific causes or movements. However, as Benford and Snow (2000) argued, people are likely to participate in social movements when these movements are framed in ways that resonate with one’s values and predispositions. We agree with this point, and it implies that when particular CA efforts are framed in terms of values and goals that are consistent with an authoritarian outlook (e.g., making society more orderly or homogeneous), authoritarianism may predict increased participation in these efforts. Thus, additional research is needed to understand the extent to which the relationship between authoritarianism and CA varies as a function of features of particular social movements and their goals.

Despite these limitations, the present analysis offers important contributions to the psychological literature on CA. By assessing both individual difference antecedents and the contexts in which these individual differences are most predictive, we gain a more complete understanding of the psychology underlying CA. Moreover, to the extent that proponents of CA are interested in mobilizing citizens, such knowledge has the capacity to facilitate communications that are likely to resonate with potential CA participants. In turn, this has the potential to make social movements more effective in achieving social change.

Footnotes

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.

Supplemental Material

The supplemental material is available with the online version of the article.

Notes

References

Abeles

R. P.

(1976). Relative deprivation, rising expectations, and black militancy. Journal of Social Issues, 32, 119-137.

Acevedo

G. A.

(2008). The fatalistic imagination and the continued salience of religion in the modern world: A test of resource compensation/resource amplification hypotheses using crossnational data. Sociological Spectrum, 28, 602-624.

Adorno

T. W.

Frenkel-Brunswik

Levinson

D. J.

Sanford

R. N.

(1950). The authoritarian personality. New York, NY: Harper & Row.

Altemeyer

(1996). The authoritarian specter. Cambridge, MA: Harvard University Press.

Amenta

Caren

Chiarello

(2010). The political consequences of social movements. Annual Review of Sociology, 36, 287-307.

Benford

R. D.

Snow

D. A.

(2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26, 611-639.

Berman

S. L.

Wittig

M. A.

(2004). An intergroup theories approach to direct political action among African Americans. Group Processes & Intergroup Relations, 7, 19-34.

Bollen

Lennox

(1991). Conventional wisdom on measurement: A structural equation perspective. Psychological Bulletin, 110, 305-314.

Brandstätter

Opp

K. D.

(2014). Personality traits (“Big Five”) and the propensity to political protest: Alternative models. Political Psychology, 35, 515-537.

10.

Brandt

M. J.

Henry

P. J.

(2012). Gender inequality and gender differences in authoritarianism. Personality and Social Psychology Bulletin, 38, 1301-1315. doi:10.1177/0146167212449871

11.

Brunsting

Postmes

(2002). Social movement participation in the digital age predicting offline and online collective action. Small Group Research, 33, 525-554.

12.

Carney

D. R.

Jost

J. T.

Gosling

S. D.

Potter

(2008). The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology, 29, 807-840.

13.

Converse

P. E.

(1964). The nature of belief systems in mass publics. In Apter

D. E.

(Ed.), Ideology and discontent (pp. 206-261). New York, NY: Free Press of Glencoe.

14.

Corcoran

K. E.

Pettinicchio

Young

J. T.

(2011). The context of control: A cross-national investigation of the link between political institutions, efficacy, and collective action. British Journal of Social Psychology, 50, 575-605.

15.

Dalton

(2002). Citizen politics. Chatham, NJ: Chatham House.

16.

Drury

Reicher

(2005). Explaining enduring empowerment: A comparative study of collective action and psychological outcomes. European Journal of Social Psychology, 35, 35-58.

17.

Duckitt

Sibley

C. G.

(2010). Personality, ideology, prejudice, and politics: A dual-process motivational model. Journal of Personality, 78, 1861-1894.

18.

Dunn

(2013). Preference for radical right-wing populist parties among exclusive-nationalists and authoritarians. Party Politics, 21, 367-380. doi:10.1177/1354068812472587

19.

Dunn

Singh

S. P.

(2014). Pluralistic conditioning: Social tolerance and effective democracy. Democratization, 21, 11-28. doi:10.1080/13510347.2012.697056

20.

Federico

C. M.

(2015). The structure, foundations, and expression of ideology. In Berinsky

(Ed.), New directions in public opinion (2nd ed., pp. 81-103). New York, NY: Routledge.

21.

Federico

C. M.

Fisher

E. L.

Deason

(2011). Political expertise and the link between the authoritarian predisposition and conservatism. Public Opinion Quarterly, 75, 686-708. doi:10.1093/poq/nfr02

22.

Federico

C. M.

Reifen Tagar

(2014). Zeroing in on the right: Education and the partisan expression of authoritarianism. Political Behavior, 36, 581-603. doi:10.1007/s11109-013-9250-4

23.

Feldman

(2003). Enforcing social conformity: A theory of authoritarianism. Political Psychology, 24, 41-74.

24.

Feldman

Stenner

(1997). Perceived threat and authoritarianism. Political Psychology, 18, 4741-4770. doi:10.1111/0162-895X.00077

25.

Fromm

(1941). Escape from freedom. New York, NY: Holt, Rinehart & Winston.

26.

Gallego

Oberski

(2012). Personality and political participation: The mediation hypothesis. Political Behavior, 34, 425-451.

27.

Gerber

A. S.

Huber

G. A.

Doherty

Dowling

C. M.

Raso

S. E.

(2011). Personality traits and participation in political processes. The Journal of Politics, 73, 692-706.

28.

Gurr

T. R.

(1970). Why men rebel. Princeton, NJ: Princeton University Press.

29.

Gutting

R. S.

(2015). Ideological asymmetries in unconventional political participation (Unpublished doctoral dissertation). State University of New York, Stony Brook.

30.

Hetherington

M. J.

Weiler

J. D.

(2009). Authoritarianism and polarization in American politics. New York, NY: Cambridge University Press.

31.

Hibbing

J. R.

Smith

K. B.

Alford

J. R.

(2014). Differences in negativity bias underlie variations in political ideology. Behavioral and Brain Sciences, 37, 297-307. doi:10.1017/S0140525X13001192

32.

Hornsey

M. J.

Blackwood

Louis

Fielding

Mavor

Morton

White

K. M.

(2006). Why do people engage in collective action? Revisiting the role of perceived effectiveness. Journal of Applied Social Psychology, 36, 1701-1722.

33.

Inglehart

(1990). Culture shift in advanced industrial society. Princeton, NJ: Princeton University Press.

34.

Jenkins

J. C.

Wallace

Fullerton

A. S.

(2008). A social movement society? A cross-national analysis of protest potential. International Journal of Sociology, 38(3), 12-35.

35.

Johnston

C. D.

Lavine

H. G.

Federico

C. M.

(2017). Open versus closed: Personality, identity, and the politics of redistribution. Cambridge, UK: Cambridge University Press.

36.

Jost

J. T.

Federico

C. M.

Napier

(2009). Political ideology: Its structure, function, and elective affinities. Annual Review of Psychology, 60, 307-337.

37.

Jost

J. T.

Glaser

Kruglanski

A. W.

Sulloway

(2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375.

38.

Klandermans

(1984). Mobilization and participation: Social-psychological expansions of resource mobilization theory. American Sociological Review, 49, 583-600.

39.

Lavine

Burgess

Snyder

Transue

Sullivan

J. L.

Haney

Wagner

(1999). Threat, authoritarianism, and voting: An investigation of personality and persuasion. Personality and Social Psychology Bulletin, 25, 337-347.

40.

Lavine

Lodge

Freitas

(2005). Authoritarianism, threat, and selective exposure to information. Political Psychology, 26, 219-244.

41.

Lavine

Lodge

Polichak

Taber

(2002). Explicating the black box through experimentation: Studies of authoritarianism and threat. Political Analysis, 10, 342-360.

42.

Lemieux

A. F.

Asal

V. H.

(2010). Grievance, social dominance orientation, and authoritarianism in the choice and justification of terror versus protest. Dynamics of Asymmetric Conflict, 3, 194-207.

43.

Malka

Soto

C. J.

Inzlicht

Lelkes

(2014). Do needs for security and certainty predict cultural and economic conservatism? A cross-national analysis. Journal of Personality and Social Psychology, 106, 1031-1051.

44.

McCarthy

Zald

M. N.

(1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82, 1212-1241.

45.

McCrae

R. R.

(2015). A more nuanced view of reliability. Personality and Social Psychology Review, 19, 97-112.

46.

Mondak

J. J.

(2010). Personality and the foundations of political behavior. New York, NY: Cambridge University Press.

47.

Olson

(1965). The logic of collective action: Public goods and the theory of collective action. Cambridge, MA: Harvard University Press.

48.

Parvin

(1973). Economic determinants of political unrest: An econometric approach. Journal of Conflict Resolution, 17, 271-296.

49.

Pettigrew

T. F.

(1967). Social evaluation theory: Convergences and applications. In Levine

(Ed.), Nebraska symposium on motivation (Vol. 15, pp. 241-311). Lincoln: University of Nebraska Press.

50.

Reicher

(2001). The psychology of crowd dynamics. In Hogg

M. A.

Tindale

R. S.

(Eds.), Blackwell handbook of social psychology: Group processes (pp. 182-208). Oxford, UK: Blackwell.

51.

Runciman

W. G.

(1966). Relative deprivation and social justice: A study of the attitudes to social inequality in 20th century England. Berkeley: University of California Press.

52.

Schlozman

K. L.

Verba

Brady

H. E.

(2012). The unheavenly chorus: Unequal political voice and the broken promise of American democracy. Princeton, NJ: Princeton University Press.

53.

Simon

Klandermans

(2001). Politicized collective identity: A social psychological analysis. American Psychologist, 56, 319-331.

54.

Simon

Loewy

Stürmer

Weber

Freytag

Habig

Spahlinger

(1998). Collective identification and social movement participation. Journal of Personality and Social Psychology, 74, 646-658.

55.

Singh

Dunn

(2013). Veto players, the policy-making environment and the expression of authoritarian attitudes. Political Studies, 61, 1119-1141. doi:10.1111/j.1467-9248.2012.00959.x

56.

Stenner

(2005). The authoritarian dynamic. New York, NY: Cambridge University Press.

57.

Streiner

D. L.

(2003). Being inconsistent about consistency: When coefficient alpha does and doesn’t matter. Journal of Personality Assessment, 80, 217-222.

58.

Tarrow

S. G.

(2011). Power in movement: Social movements and contentious politics (3rd ed.). New York, NY: Cambridge University Press.

59.

Thomas

E. F.

Mavor

K. I.

McGarty

(2012). Social identities facilitate and encapsulate action-relevant constructs: A test of the social identity model of collective action. Group Processes & Intergroup Relations, 15, 75-88.

60.

United Nations Development Programme. (2006). Human development report. Available from http://hdr.undp.org/en/

61.

van Zomeren

. (2013). Four core social-psychological motivations to undertake collective action. Social & Personality Psychology Compass, 7, 378-388.

62.

van Zomeren

Postmes

Spears

. (2008). Toward an integrative social identity model of collective action: A quantitative research synthesis of three socio-psychological perspectives. Psychological Bulletin, 134, 504-535.

63.

Walker

Mann

(1987). Unemployment, relative deprivation, and social protest. Personality and Social Psychology Bulletin, 13, 275-283.

64.

Zaller

(1992). The nature and origins of mass opinion. New York, NY: Cambridge University Press.

65.

Zhang

(2012). The effects of political news use, political discussion and authoritarian orientation on political participation: Evidences from Singapore and Taiwan. Asian Journal of Communication, 22, 474-492.

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

0.30 MB