Accessible Published by De Gruyter August 18, 2021

Perceived injustice, perceived group threat and self-reported right-wing violence: An integrated approach

Harley Williamson, Ann De Buck and Lieven JR Pauwels


The present study seeks to explain individual differences in self-reported politically motivated violence and vandalism, and participation within an extreme right-wing group. While violent extremism is highly debated, few criminological studies explicitly test factors that can trigger violent extremism. The present study addresses this gap by integrating two different frameworks: a perceived injustice and group threat-initiated model and an impulsivity-initiated model. We also investigate several intervening mechanisms. We draw on a sample of 705 adolescents and young adults living in Flanders, Belgium to test the strength of direct and intermediary effects of perceived injustice, perceptions of out-group threat from Jewish populations, ethnocentrism, feelings of superiority, moral support for right-wing extremism, and exposure to racist peers on politically motivated violence and vandalism. Results of structural equation models (SEM) indicate various direct and intermediary effects between both perceived injustice and violent extremism, and between impulsivity and violent extremism. Our model reveals the complex and intricate antecedents of violent extremism. Importantly, we find that feelings of injustice and unfair treatment are a major source of extremist violence, as they easily trigger often debated causes such as high in-group identification and ethnocentrism. Implications of these findings for preventing violent extremism are discussed, given the centrality of perceptions of injustice and threat.

1 Introduction

Political violence presents a significant threat to individual safety and security. It further challenges the ability for governments globally to implement adequate preventative and reactive policies (Pauwels, Ljujic, & De Buck, 2018). Yet, recent acts of right-wing extremism spotlight the growing threat right-wing extremism poses (Blackbourn, McGarrity, & Roach, 2019; Hardyns, Thys, Dorme, Klima, & Pauwels, 2021; Pantucci & Ong, 2021). Moreover, the prominence of far-right political parties espousing anti-establishment and xenophobic rhetoric and calling for racial and/or ethnic assimilation further legitimizes the use of right-wing violence (Jones, Doxsee, & Harrington, 2020). Extremist attitudes may not necessarily extend to condoning or engaging in violence (McCauley & Moskalenko, 2017; Williamson, Murphy, & Sargeant, 2020). However, better understanding how and why people become members of right-wing groups or engage in right-wing violence can inform wider preventative efforts.

Attempts to explain political violence have traditionally drawn on individual theoretical perspectives. For example, criminological theories such as strain theory, control theory and social learning theory are often employed to understand the predicates of political violence (Pauwels & Heylen, 2020). However, the reliance on a single explanatory variable has resulted in theories that explain only a small percentage of the variance in criminal behaviour (Elliott, Ageton, & Canter, 1979; see also Weisburd & Piquero, 2008) and political violence (see Opp, 2009 for a similar discussion on political protest; see also Wikström & Bouhana, 2017 for a discussion on violent extremism). Thus, since the 1990 s scholars have sought to integrate several theoretical perspectives to better understand the prevalence and drivers of (political) violence.

2 An integrative mechanism-based approach

Theoretical integration has not always been embraced in criminology. There are concerns that integrated theories could be developed that are logically inconsistent, as theories explain human nature and the antecedents of crime and political violence in divergent ways. A significant difference between the core criminological perspectives on the causes of delinquency is grounded in different views on human nature. Ultimately, the often-simplistic nature of traditional criminological theories has been criticized for failing to account for the complex and multifaceted antecedents of crime.

Agnew (2011) argues that none of the assumptions underlying the core criminological theories are congruent with empirical findings. Empirical research shows that people differ in terms of social controls and motivations. For example, contemporary contextual studies of youth violence suggest that social structures contextualize social controls and motivational factors on violent behaviour, thereby indirectly affecting violent behaviour. In light of Agnew’s (2011) conclusion that classical theories have departed from one-sided visions of human nature and social order, the most important critique on integration (i.e., the risk of constructing inconsistent theories) seems to be futile.

We therefore suggest that integration built around an internal causal logic distinguishing between proximal and distal factors has merits. This view derives from the analytical tradition in criminology and offers a solid foundation for developing integrated theories. An analytical approach is concerned with understanding why people engage in violent extremism, which can be accomplished by identifying the key social, developmental, and situational processes (i.e., mechanisms) involved in crime causation. A recent focus on the antecedents of political violence have seen studies testing integrated frameworks of risk and protective factors derived from these criminological and psychological theories (Beelmann, 2020; Lösel, King, Bender, & Jugl, 2018; Wolfowicz, Litmanovitz, Weisburd, & Hasisi, 2020). Indeed, an analytical approach to political violence enables an examination of the interaction between causation, human agency, and the person-environment more seriously to advance knowledge on (political) violence, its causes and prevention. This interaction has implications for the study of political violence as it facilitates insight into the ›causes‹ of political violence, and to »the causes of the causes of political violence« (Pauwels, Brion, & De Ruyver, 2014, p. 84).

Against the backdrop of the persistent threat of right-wing extremism in Europe and beyond, our study seeks to contribute to the research base by offering an interdisciplinary explanation of the correlates of extreme right-wing group participation and violence. Specifically, we test an integrated model of political violence that draws on concepts from General Strain Theory (GST) (Agnew, 2011), Perceived Injustice Theory (Gurr, 1970), Integrated Threat Theory (ITT) (Stephan & Stephan, 2000), Social Identity Theory (SIT) (Turner & Tajfel, 1974) and the General Theory of Crime (Gottfredson & Hirschi, 1990). We argue that criminological and social psychological mechanisms may shape individuals’ moral support for the use of right-wing extremist violence, which can influence membership in right-wing extremist groups and the utility of politically motivated violence.

In the proposed model, we hypothesize that membership in right-wing groups and right-wing violence may be influenced by perceived group threat, perceived injustice, and feelings of superiority. Moreover, we suggest that these relationships may be mediated by in-group and out-group processes stemming from SIT, including perceived group threat, exposure to racist peers, and ethnocentrism (Doosje, Van den Bos, Loseman, Feddes, & Mann, 2012). Figure 1 presents the conceptually integrated model of politically motivated violence, vandalism, and right-wing group membership, which is followed by an explanation of the concepts used in the integrated model. The single headed arrows define the theoretically derived effect parameters between latent variables in the model, whereas the double headed arrows represent correlations between latent variables.

Figure 1 
Analysis of direct effects: Specification of the conceptual model explaining self-reported politically motivated violence, self-reported politically motivated vandalism and right-wing group participation

Figure 1

Analysis of direct effects: Specification of the conceptual model explaining self-reported politically motivated violence, self-reported politically motivated vandalism and right-wing group participation

Perceived injustice

Studies of political violence repeatedly draw attention to inequality and injustice. Within criminology, Agnew’s GST stresses the importance of perceived injustice. GST posits that negative feelings may cause strain, which can pressure adolescents into crime by stimulating negative emotions and support for violence (Agnew, 2006, 2011). Sources of strain such as perceived injustice may act as stressors that can trigger one’s participation in violent extremism, especially if it interacts with moral support for violent extremism as an intermediate mechanism.

Prior research consistently highlights how perceived injustice is a significant determinant of support for extremism (Doosje et al., 2012). This is observable in a recent study by Nivette and colleagues (2017) who utilized a sample of Swiss adolescents to examine if strain predicted support for violent extremism. Results demonstrated that collective strain, which the authors conceptualized to include »political, cultural, and economic discrimination; systematic exclusion; and exposure to war and conflict« was a source of support for violent extremism (Nivette et al., 2017, p. 765). These findings suggest that injustice and discrimination perceived at the group level may predict extremist beliefs.

In this study, perceived injustice pertains to the idea that one’s own group is relatively disadvantaged compared to other groups in society by, for example, perceiving that their group is treated unfairly by society. In this sense, it is not objective discrimination or injustice but rather the subjective perception and experience of it.

Perceived group threat

In addition to perceived injustice, researchers often observe strong effects of group threat on out-group prejudice (Quillian, 1995). In social psychology, ITT focuses on the context of threats to members of an in-group and is arguably regarded as the most important integrative threat-related theory to date (Stephan & Stephan, 2000). In ITT, realistic and symbolic threats – the two classic approaches to threat studies – are complemented by the addition of negative stereotypes and intergroup anxiety (for a meta-analytic review, see Riek, Mania, & Gaertner, 2006).

Existing research demonstrates the role of perceived group threat in predicting support for extremism and political violence (Doosje et al., 2012; Pauwels & Heylen, 2020). For example, Doosje and colleagues (2012) drew on a sample of 1086 Dutch youths to examine the antecedents of moral support for right-wing extremism. Findings showed that perceived group threat (which the authors operationalized as realistic threat, symbolic threat, and intergroup anxiety) was a significant predictor of moral support for right-wing extremism. Similar findings were revealed in Pauwels and Heylen’s (2020) sample of 723 Flemish adolescents. The authors focused on participants’ perceptions that Muslims pose a threat and tested the statistical effect of different exogenous variables on right-wing extremist violence: a threat-perception initiated model and a thrill-seeking initiated model. Pauwels and Heylen found that perceived Muslim threat positively shaped moral support for right-wing extremist violence, which in turn was directly related to the use of right-wing violence.

Taken together, these results reinforce the notion that people are more likely to morally support right-wing extremism and engage in violence if they perceive that their group is threatened (Kruglanski, Chen, Dechesne, Fishman, & Orehek, 2009). In the present study, we draw on perceptions that Jewish populations pose a threat as a predicate of support for and engagement in right-wing extremism. We utilise this operationalization of group threat because the core of Nazi ideology supports the anti-Semitic and dehumanizing rhetoric that depicts Jewish populations as the enemy (Steizinger, 2018). Moreover, the contemporary extreme right-wing movement is described as being driven to defend against a purportedly Jewish-dominated threat of world domination and economic imperialism (Grumke, 2017).

Feelings of superiority

Relatedly, those who perceive out-groups as more threatening are also more likely to deem their own group as superior (Doosje et al., 2012). Feelings of superiority are derived from SIT (Turner & Tajfel, 1974). SIT can explain the circumstances that enhance identification with a perceived in-group and the subsequent feelings of superiority one can feel about their in-group as distinct from perceived out-groups (Tajfel, 1969). Such feelings of self-esteem and superiority derived from belonging to a positively valued in-group and subsequently harbouring exaggerated perceptions towards one’s group, can perpetuate out-group derogation. Hence, feelings of superiority (self-esteem) can play a crucial role in the aetiology of extremist violence (van Prooijen & Krouwel, 2019).


The third predictor of extreme right-wing participation and violence is ethnocentrism. Ethnocentrism refers to harbouring a negative worldview resulting from positive attitudes towards the ‘in-group’ and negative views towards perceived ‘out-groups’ (e.g., Willems, 1995). The ‘in-group/out-group’ relationship allows people to break down the barriers to committing violent acts (van Der Valk & Wagenaar, 2010). Theoretically, ethnocentrism is expected to positively predict moral support for right-wing extremism. This is because threats to an in-group’s status are likely to elicit negative attitudes towards the group representing the source of such threats (Bircan, 2012), which may in turn effect moral support for right-wing extremism (see Turner & Tajfel, 1974).

Exposure to racist peers and impulsivity

In addition to the social psychological measures discussed above, we also include two important mechanisms from criminological literature into the model: exposure to racist peers and impulsivity. Firstly, an individual’s association with racist peers may influence their feelings towards those of diverse racial/ethnic backgrounds, and their subsequent support for right-wing extremist ideals that perpetuate such racist notions (De Waele & Pauwels, 2014; Pauwels & Schils, 2016). The role of (racist) peers is prominent in social learning theories (Akers, 1998; Warr, 2002) but highly contested in control theories (Gottfredson & Hirschi, 1990; Hirschi, 1969). However, exposure to racist peers provides two specific contexts of exposure to settings in which the use of general or political violence is supported. Differential exposure is not only important in Social Learning Theory, but also in Routine-Activities/Lifestyle Theory (Pauwels & Svensson, 2013; Sampson & Laub, 2003). From a routine activities/lifestyle perspective, peers are important as they may be responsible for the situational motivation to engage in political violence. Previous studies have documented that exposure to racist peers is related to self-reported violence (De Waele & Pauwels, 2014; Pauwels & Schils, 2016). In the present study we consider exposure to racist peers as an important indicator of exposure to extremist moral settings, thereby shaping the individual’s routines and risky lifestyle.

Secondly, impulsivity is a general risk factor of violence, troublesome youth group involvement and other deviant behaviours such as reported vandalism. Studies reveal the role impulsivity plays in young people’s decisions to partake in (right-wing) extremist groups (e.g., Borum, 2014). For example, ethnographic studies indicate that new members of right-wing extremist groups are often driven by their own opportunistic or impulsive motives (Bjørgo, 2002; Watts, 2001). Although ideological motivations are important for understanding politically motivated violence, ideology is not always a precursor to entering extremist groups. Instead, impulsivity is often mentioned by (former) members of extremist groups to explain their participation in such groups (Schils & Verhage, 2017).

The role of impulsivity in explaining violence has been highlighted in several disciplines. In criminology, arguably the best-known model is the General Theory of Crime, which describes impulsivity as a key aspect of ‘low self-control’.[1] Impulsivity is also a key predictor of violence in psychological theories (Farrington, 2020; Moffitt, 2017) (e.g. Farrington; Moffitt). It must be noted that impulsivity is distinct from risk or sensation seeking. The former refers to the tendency to act without thinking, whereas the latter refers to a preference for novel stimuli (see Burt, 2020).[2] Thus, impulsivity is expected to be both directly related to violent extremism and indirectly through moral support for right-wing extremism and exposure to racist peers; a phenomenon usually termed a selection effect (Bendixen, Endresen, & Olweus, 2006; Decker, Melde, & Pyrooz, 2013).

Moral support for right-wing extremist violence

Finally, we measure individual moral support for the use of violence by right-wing extremist groups to obtain political goals. Violent radicalization is often defined as the process whereby individuals acquire moral support for violence in a means-to-an-ends fashion (Borum, 2011). Studies have found that measures of antisocial values are significantly related to (adolescent) offending (Akers, 1998; Hirschi, 1969; Pratt et al., 2010; Silver & Silver, 2021; Svensson, Pauwels, & Weerman, 2010) and self-reported political violence (De Waele & Pauwels, 2014; Hagan, Merkens, & Boehnke, 1995). In our integrated model, we propose that moral support for right-wing violence is a key mechanism by which perceived injustice, perceived group threat, ethnocentrism and feelings of superiority may translate into right-wing extremist group membership, and right-wing political violence and motivated vandalism.

3 Hypotheses

To test our conceptual model (see Figure 1), a series of hypotheses are tested:

Hypothesis 1 Perceived injustice will be positively related to perceived group threat.

Prior research highlights how feelings of injustice can trigger perceptions of group threat if such injustices are deemed to target a specific out-group (Pauwels & Heylen, 2020). As such, we expect that perceived injustice will predict a sense of out-group threat, which in turn will act as an antecedent for other identity-based predictors of politically motivated violence and vandalism.

Hypothesis 2 Perceived group threat and feelings of superiority are positively related to ethnocentrism.

This hypothesis firstly reflects the idea that favouring high distinctiveness of a certain out-group (in the present study we operationalize the out-group as Jewish populations) may translate into a greater propensity to view the world from the perspective of the in-group (i.e., ethnocentrism) (Turner & Tajfel, 1974). Secondly, this proposition also incorporates the idea of a hostile and competitive worldview whereby feelings of superiority may positively affect ethnocentrism (Duckitt & Sibley, 2007; Pauwels & Heylen, 2020) and subsequently, politically motivated violence and vandalism (Doosje, Loseman, & Van Den Bos, 2013).

Hypothesis 3 There will be a differential effect from ethnocentrism, feelings of superiority, impulsivity, and exposure to racist peers on moral support for right-wing extremism.

Largely in line with the perceived injustice model, we hypothesize that ethnocentrism, feelings of superiority, impulsivity and exposure to racist peers will positively predict moral support for right-wing extremism, which are considered to be positive attitudes towards the use of violence on the basis of extremist views. Contrary to Hypothesis 4 (below), we argue that moral support for right-wing extremism can be nurtured through antisocial factors rather than specific ideas about a particular out-group. For this reason, we hypothesize that ethnocentrism more broadly will influence moral support for right-wing extremism compared to exposure to racist peers.

Hypothesis 4 Exposure to racist peers will be positively influenced by ethnocentrism and impulsivity.

In Hypothesis 4, the finding that impulsivity is often associated with seeking the company of extremist/racist peers is incorporated. Additionally, the idea that holding negative views of a certain out-group (i.e., ethnocentrism) may lead individuals to seek out others with similar views on that out-group. Regarding the latter, we argue that ethnocentrism is an intermediary variable between group threat and feelings of superiority on the one hand, and exposure to racist peers on the other. This is because in ethnocentrism, a clear out-group is depicted as the source of the identity threats that underlie feelings of superiority (van Der Valk & Wagenaar, 2010).

Hypothesis 5 Moral support for right-wing extremism, impulsivity and exposure to racist peers will have a positive effect on participation in a right-wing disruptive group (i.e., membership), and reported politically motivated violence and vandalism.

We hypothesize that the behaviour-based index of extremist violence, vandalism, and participation in a right-wing disruptive group will be positively influenced by tolerance for such acts (i.e., moral support for right-wing extremism), impulsivity and indirect influences of peer pressure (exposure to racist peers). These relationships align with existing theories of the aforementioned constructs. With the inclusion of Hypothesis 5, the model is completed.

4 Methods

Data were collected through a web survey completed by 705 young Flemish adults aged between 18 and 25 who live in Flanders, have Belgian nationality and whose parents are Belgian nationals. The survey consisted of a self-administered questionnaire conducted online between December 2012 and April 2013 to participants residing in the Flanders region of Belgium. A snowball sampling method was adopted whereby participants were recruited via the Internet (extremist websites, forums, Facebook pages, blogs) and word of mouth. The majority of respondents were female (64.3 %), students (94.7 %) and were either Catholic (45.6 %) or atheist (47.3 %). The higher proportion of female participants in the sample may be reflective of the greater proportion of females who use Facebook when compared to their male counterparts (Statista, 2021). Similarly, given the nature of the recruitment strategy and the desired age range of participants, it is not surprising that the majority identified as students. Moreover, Catholicism has traditionally been a significant religion in Belgium, and particularly Flanders (where are sample resides), which supports the breakdown of Catholic respondents in the study. However, in recent years atheism has gained traction in Belgium, which also justifies the proportion of atheist respondents in our sample (see De Waele (2015) for a more detailed account of the study protocol).

4.1 Measurement of constructs

A range of scales were constructed to assess the relationship between exogenous, intermediary, and endogenous variables. A brief overview of the scale constructs and associated Cronbach’s alphas (α) is presented below. A list of all item wordings can be found in the supplementary materials.

Endogenous variables. The self-reported political violence scale was created by summing the scores across five items. Respondents were asked if they ever had done certain things, such as ‘fought with someone because of their political or religious belief’. Each item was measured using a four-point never to more than three times Likert scale (De Waele, 2015) (α=0.88). Higher scores indicate respondents more frequently engaged in politically motivated violence.

The self-reported political vandalism scale was created by summing the scores across three items. Respondents were asked if they ever had done things such as ‘destroyed something on the street because of their political or religious belief’. Items included in this scale were measured on a four-point never to more than three times Likert scale (De Waele, 2015) (α=0.73). Higher scores suggest that respondents more frequently engaged in politically motivated vandalism. An examination of the zero-order bivariate Pearson’s correlations showed a rather high correlation between the reported politically motivated violence scale and the reported politically motivated vandalism scale (r=0.635; p<0.001). However, both variables were treated as separate, but intercorrelated constructs.

Finally, a right-wing disruptive group participation measure was created by using a funnelling technique. Specifically, we combined answers to one filter question and four follow-up questions to measure self-reported violent youth gang participation. Respondents were first asked »Do you consider yourself to be a member of a group of friends (not a formal organization or association) that frequently meets and considers itself as a group?« (1 = yes, 0 = no). The four follow-up questions were (1) How long have you been a member of the group? (2) How big is your group? (3) »Do members of this group get involved in fights with other cliques?« (4) »Are members of this group involved in law-breaking?«. The methodology was inspired by the Eurogang questionnaire but modified to capture the right-wing affiliation of the group (De Waele & Pauwels, 2016). We retained respondents who were members of a group of 3 or more persons for more than three months. The last two questions were dichotomous (1 = yes, 0 = no). Respondents were categorized as participating in a right-wing ‘disruptive group’ if they answered affirmatively to the leading question and the four follow-up questions. Over two-fifths (40.8 %; n=295) of the total sample answered positively to the first question, while 9.8 % (n=69) of the total sample answered positively to the follow-up questions. In addition, the members were asked to position themselves on a left (0) to right (7) scale.[3]

Exogenous and intermediary items. A series of exogenous and intermediary items were included in the model. The perceived injustice scale was created from scores on six items that were measured on a 5-point strongly agree (1) to strongly disagree (5) Likert scale translated from Van den Bos, Loseman, and Doosje (2009) (α=0.88). The group threat scale was measured by presenting respondents with five items on a 5-point strongly agree (1) to strongly disagree (5) Likert scale. These items tap into the extent to which respondents perceive a particular out-group, e.g., Jews, as both a relative personal threat and an in-group threat (α=0.89). The feelings of superiority scale was created by using scores on three items measured on a 5-point strongly agree (1) to strongly disagree (5) Likert scale (Van den Bos et al., 2009) (α=0.73). The ethnocentrism (out-group hostility) scale was created using scores on three items measured on a 5-point strongly agree (1) to strongly disagree (5) Likert scale derived and translated from the SCIF[4] questionnaire (α=0.80). The moral support for right-wing extremism scale was created using participants’ scores on three items measured on a 5-point strongly agree (1) to strongly disagree (5) Likert scale. The items were adapted from the work of Van den Bos et al. (2009) and De Waele and Pauwels (2014) (α=0.87). Exposure to racist peers[5] was assessed by summing participants’ scores on five items measured on a 4-point none (1) to all (4) Likert scale. Each item was derived and translated from Van den Bos et al.’s (2009) work (α=0.85). Finally, impulsivity was created by summing the scores of three items measured on a 5-point strongly agree (1) to strongly disagree (5) Likert scale. The items used in this scale were adapted from the work of Grasmick, Tittle, Bursik Jr, and Arneklev (1993) (α=0.65).

5 Results

5.1 Descriptive statistics and zero-order correlations

A summary of means, standard deviations, minimum and maximum values of the measured variables are presented in Table 1 and Pearson’s correlations are in Table 2.

Table 1:

Descriptive statistics of all variables

Variables N Min Max Mean SD
Perceived injustice 704 6 30 9.11 4.23
Perceived group threat 701 5 25 10.81 4.20
Exposure to racist peers 697 5 20 10.12 3.30
Ethnocentrism 705 3 15 7.91 2.59
Feelings of superiority 705 3 15 5.63 2.34
Moral support for right-wing extremism 705 3 15 4.71 2.57
Impulsivity 700 3 15 7.11 2.33
RW group participation 705 0 1 0.10 0.30
Reported politically motivated violence 671 6 15 6.07 0.53
Reported politically motivated vandalism 672 3 10 3.09 0.52
Table 2:

Zero-order Bivariate Pearson’s Correlations Matrix

N=648 1 2 3 4 5 6 7 8 9 10
Independent Variables
1. Perceived injustice 1
2. Group Threat .196*** 1
3. Racist peers .216*** .211*** 1
4. Ethnocentrism .285*** .391*** .439*** 1
5. Superiority .234*** .230*** .200*** .287*** 1
6. Moral support for right-wing extremism .195*** .220*** .310*** .343*** .282*** 1
7. Impulsivity .136*** .147*** .217*** .072 .178*** .181*** 1
Dependent Variables
8. RW group participation .070* -.005 .080* .004 .077* .082* .146*** 1
9. Reported politically motivated vandalism .165*** .063 .061 -.005 .147*** .154*** .099* .090* 1
10. Reported politically motivated violence .151*** .065 .140*** .092** .117** .216*** .087* .086* .635*** 1

5.2 Confirmatory factor analysis (CFA/Structural equation modelling (SEM)

Confirmatory factor analysis (CFA) and full structural equation models (SEM) were performed utilizing the Mplus statistical package version 7.11 (Muthén & Muthén, 2012). A CFA was conducted to evaluate the measurement model containing six factors: perceived injustice, group threat, ethnocentrism, feelings of superiority, moral support for right-wing extremism and impulsivity. Results showed that this six-factor model had a good model fit: WLSMVχ²(215): 1348.684, p<0.000, RMSEA=0.086 (90 %CI 0.082 ; 0.091), CFI/TLI= 0.964/0.957, WRMR (Weighted Root Mean Square Residual) = 1.872 (Marsh, Wen, & Hau, 2004). The six factors were moderately positively correlated (see supplementary materials). Thus, a six-factor model was retained for further SEM analysis.

From here, a SEM was conducted to test the integrated structural model (displayed in Figure 1). The individual data format was used for input on the analyses. Individual data are raw data in which the scores of all individuals on all variables are preserved (Geiser, 2012). WLSMV (Means and Variance Adjusted Weighted Least Squares) procedure for ordinal Likert scale variables was chosen over other estimation methods. According to Kline (2015), estimation methods for continuous variables are not suitable when utilizing Likert scale variables with a relatively small number of categories (five or fewer) or when the response distributions are severely asymmetrical, as is the case in this study. Simulation research related to the WLSMV estimator has been shown to yield accurate test statistics, parameter estimates and standard errors under both normal and nonnormal latent response distributions across sample sizes ranging from 100 to 1000 (Byrne, 2013).

We did not conduct post-hoc modifications to the original hypothesized model to improve model fit.[6] Model fit was assessed using the following indices: A non-significant χ² is desired. However, χ² statistic is highly sensitive to sample size. As such, the significance of the χ² test should not be a reason by itself to reject a model (Wang & Wang, 2020). Instead, Comparative Fit Index (CFI) (Bentler, 1990), Tucker Lewis Index (TLI) (> 0.95 (Hu & Bentler, 1999)) and Root Mean Square Error of Approximation (RMSEA) are the most popular measures of model fit for SEM. The RMSEA values are often interpreted as: 0 = perfect fit; <0.05 = close fit; 0.05-0.08 = fair fit; 0.08-.0.10 = mediocre fit; and >.10 = poor fit (Byrne, 2012; Hu & Bentler, 1999). In addition, the 90 % CI computed for the RMSEA is reported. Ideally, the lower value of the 90 % CI should be very near zero (or no worse than 0.05) and the upper value should be less than 0.08; weighted mean-square residual standardized (WRMR), a residual-based model fit index. Perfect model fit is indicated by WRMR = 0 and increasingly higher values indicate worse fit (Bentler, 1990; Byrne, Shavelson, & Muthén, 1989; Cudeck & Browne, 1983; Hu & Bentler, 1999; Kline, 2015; Wang, 2019).

Three endogenous variables were introduced: reported politically motivated violence, vandalism, and membership in a right-wing disruptive group. Fit indices indicated fair model fit to the data (WLSMVχ²(304): 1402.543, p<0.000, RMSEA=0.072 (90 %CI 0.068 ; 0.075), CFI/TLI= 0.965/0.960, WRMR = 1.976). The results are shown in Figure 2.

Figure 2 
Regression estimates of the best fitting structural model. Displayed path coefficients are standardized coefficients.

Figure 2

Regression estimates of the best fitting structural model. Displayed path coefficients are standardized coefficients.

Overall, the results largely confirm the conceptual model. As expected, levels of perceived group threat are in part determined by perceptions of injustice towards the in-group (β=0.326, p<0.001). In turn, group threat has a relatively large and positive effect on ethnocentrism (β=0.477, p<0.001). The second determinant of ethnocentrism in this study is feelings of superiority which has a direct positive and significant effect (β=0.350, p<0.001).

A second cluster of effects centres around moral support for right-wing extremism. Ethnocentrism (β=0.352, p<0.001), feelings of superiority (β=0.241, p<0.001), and impulsivity (β=0.193, p<0.001) are positively and significantly related to support for right-wing extremism. A third cluster of effects relates to exposure to racist peers. Both ethnocentrism and impulsivity have a positive effect on exposure, with the effect of ethnocentrism being considerably stronger (β=0.453, p<0.001) than the effect of impulsivity (β=0.244, p<0.001). The final cluster of relationships that relates to the three endogenous variables incorporate politically motivated violence and vandalism and reported membership in a right-wing disruptive group. Findings reveal a relatively strong positive and significant relationship between impulsivity and membership (β=0.334, p<0.001), politically motivated violence (β=0.236, p<0.001) and politically motivated vandalism (β=0.233, p<0.001). Support for right-wing extremism is also positively and significantly related to vandalism (β=0.118, p<0.05) and violence, although the latter effect is rather small (β=0.096, p<0.001). Finally, a small but significant positive relationship is observable between exposure to racist peers and politically motivated violence (β=0.036, p<0.05). The resulting best fitting model is displayed in Figure 3 and displays the standardized regression estimates of the structural model.

Collectively, our results indicate multiple sources of reported politically motivated violence, vandalism, and right-wing group membership. Specifically, supporting right-wing extremism, being exposed to racist peers and being impulsive are direct determinants of all three endogenous variables, albeit in a differential way. These three determinants, however, are triggered by clusters of interrelated factors. Specifically, perceived injustice predicts feelings of group threat which in turn relates to ethnocentrism. Ethnocentrism is an intermediary factor related to both support for right-wing extremism and exposure to racist peers. Moreover, feelings of superiority predicted ethnocentrism and was directly related to support for right-wing extremism.

6 Discussion

The findings of our study are largely confirmed by the theories underpinning the integrated framework proposed at the outset. Resultantly, they support the empirical utility of an interdisciplinary approach to understanding support for and engagement in right-wing extremist groups and offending behaviours. For example, the pathway model identified a positive relationship between perceived group threat and injustice. This finding aligns with the premise of ITT, whereby groups that act in an unjust manner towards the in-group can easily be seen as threatening towards the in-group, both in a realistic or symbolic way (Stephan & Stephan, 2000). Our findings also showed that group threat and feelings of superiority were positively associated with ethnocentrism. Regarding the former, as group threat is operationalized as prejudice towards the out-group, a straightforward interpretation in line with Generalized Prejudice Theory is that threatening groups are usually regarded more negatively compared to non-threatening out-groups (Cunningham, Nezlek, & Banaji, 2004). In terms of the latter result, feelings of superiority can be considered as a form of heightened identification with the in-group and its norms, and may result in negative attitudes towards members of an out-group (Tajfel, 1969).

Our results also revealed a positive association between feelings of superiority and ethnocentrism on moral support for right-wing extremism. These relationships can be explained through the lens of SIT, as people who engage in extremism often display greater feelings of superiority, which can be related to heightened feelings of self-esteem (Turner & Tajfel, 1974). When in-group distinctiveness is low, which downplays the self-esteem derived from belonging to an in-group by making it less different or exclusive compared to the out-group, support for right-wing extremism can result to restore feelings of self-esteem. The larger effect of ethnocentrism in comparison to the effect of feelings of superiority bolsters this claim, as ethnocentrism is specifically located at the group level rather than the individual level, which plays a more subtle role in the aetiology of group dynamics. By virtue of a social learning mechanism, we would expect to find a positive relationship between exposure to racist peers and support for right-wing extremism. However, in this study, the relationship was not significant.

Our model also shows a series of effects related to exposure to racist peers. Specifically, both ethnocentrism and impulsivity were positively associated with exposure to racist peers. Theoretically, this finding is logical given that ethnocentrism is operationalized as negative attitudes towards an out-group (i.e., Jewish populations), and association with racist peers can result in the endorsement of these attitudes, thereby heightening self-esteem. Further, people tend to identify with positively valued groups or group members, so if one has negative opinions of a specific out-group, identification with likeminded people is not at all surprising. Both explanations have been extensively tested by scholars investigating the tenets of SIT (e.g., Doosje et al., 2012; Pauwels & Heylen, 2020; Quillian, 1995; Velasco González, Verkuyten, Weesie, & Poppe, 2008). The effect of impulsivity is considerably lower, which is most likely because it is a sub-dimension of self-control and is usually considered to be a personality trait rather than an attitude. However, as discussed above, impulsivity in itself is consistently identified as a salient predictor of extremist group membership (Schils & Verhage, 2017).

Relatedly, we found a small positive relationship between exposure to racist peers and politically motivated violence. This finding can be explained through the tenets of Differential Association Theory and an individual’s routine activities. Specifically, through a Differential Association Theory lens, an individual’s contact with delinquent social networks can expose them to criminogenic attitudes and behaviours. In the context of the present study, our findings suggest that exposure to racist peers can reinforce attitudes and behaviours that are tolerant of the use of violence to achieve political goals (De Waele & Pauwels, 2014). Resultantly, the reinforcement of such behaviour can create and maintain a context whereby the benefits associated with engaging in violence (e.g., prestige, loyalty) outweigh the risks associated with such acts (Pauwels & Schils, 2016).

We further observed a positive association between impulsivity, right-wing group membership, politically motivated violence, and politically motivated vandalism. Our findings support prior research that finds a link between impulsivity and violence, including violence generally (Krakowski & Czobor, 2014), online (Kaakinen, Sirola, Savolainen, & Oksanen, 2020), and that which is politically-motivated (Schils & Verhage, 2017). Moreover, as impulsivity is also indirectly related to violent extremism through the mediating role of moral support for extremism, our findings are reflective of the selection effect phenomenon identified in gang research (Bendixen et al., 2006; Decker et al., 2013). Specifically, and drawing on the life-course perspective of offending (Sampson & Laub, 2003), our findings support the notion that individuals who are already on a delinquent trajectory are more susceptible to become members of offending groups (e.g., gangs, extremist groups) and engage in violent behaviours that support the group’s motives (Decker et al., 2013).

Finally, we found that moral support for right-wing extremism was positively related to politically motivated vandalism and politically motivated violence. While a small effect was observed, this finding suggests that support for right-wing extremism is the mechanism explaining the association between perceived injustice, group threat, ethnocentrism, impulsivity, exposure to racist peers and feelings of superiority and engagement in politically motivated offending (i.e., violence, vandalism) (De Waele & Pauwels, 2016). Our findings concur with prior research that finds a link between support for antisocial ideals and behaviours, and engagement in crime (see e.g., Antonaccio & Tittle, 2008; Bottoms, 2002; Stams et al., 2006; Svensson et al., 2010) and politically motivated violence (see e.g., De Waele & Pauwels, 2014).

7 Implications

Taken together, our findings are consistent with a small but growing number of studies that empirically document the importance of perceptions of (un)fair treatment and discrimination on support for right-wing extremism (see e.g., Doosje et al., 2012; Pauwels & Heylen, 2020; Pauwels et al., 2018). They further demonstrate the explanatory value of an integrated framework of right-wing extremism that includes criminological and social psychological predicates. Countering violent extremism hinges on the ability for policies to address the differing drivers of extremism. In the context of right-wing extremism, group- and identity-based dynamics, as well as criminogenic actions can influence extremist beliefs and behaviours (Doosje & van Eerten, 2017). Determining these separate drivers of political violence is important not only for developing and evaluating criminological and social psychological theories, but also for informing the prevention of violent extremism. Wikström and Bouhana (2017) argue that a holistic explanation of specific attitudes and behaviours is the bedrock of an efficient preventive strategy. In this context, our findings have two main and interrelated implications that may shape preventative approaches to address violent extremism.

First, the role of individuals’ associations with racist peers and strong in-group identification suggests that identity-based mechanisms are important antecedents of political violence. SIT is an important framework for understanding how out-group prejudice can shape an individual’s attitudinal or behavioural intentions (Turner & Tajfel, 1974). Hence, preventive programs should explicitly incorporate notions of socialization that seek to promote tolerance and intergroup inclusion. Schools play a crucial role in an individual’s socialization process (Bjørgo & Carlsson, 2005). Thus, school-based programs such as the Tolerance Project can promote democratic values and reduce the likelihood for disillusioned youths to be recruited into extremist groups (Skiple, 2018). Other programs such as the Exit Project can also assist individuals to deradicalize (alter attitudes) and disengage (alter behaviors) from extremist groups (Bjørgo, 2002). The principles of these programs have been successfully adapted to counter violent extremism in differing contexts (Canna, Day, & Popp, 2020; Hardy, 2019).

Second, the importance of perceiving injustice and its associated threat should not be underestimated, as the intricate mechanism leading to extremism is rooted in these perceptions. Hence, programs should seek to mitigate perceptions of threat more explicitly. This type of program is likely to be more indirect by, for example, incorporating superordinate goals that seek to reduce inter-group superiority and conflict by promoting cooperation and shared goals (e.g., Sherif, 1958). Such an initative may foster more harmonious contact amongst groups and facilitate a common identity (Gaertner, Dovidio, Anastasio, Bachman, & Rust, 1993). Yet, despite the potential for these programs to promote tolerance and reduce intergroup prejudice, they are often awarded limited attention when addressing extremism. Given that the threat associated with perceived injustice can be deemed a root cause of extremism, this lack of attention seems ill-founded.

8 Study Limitations

While the current study offers an innovative approach to understanding the antecedents of politically motivated violence, there are some limitations to consider when interpreting the findings. First, our conceptual framework only accounts for some variation in Flemish youths’ participation in right-wing groups. The scope of the findings is thus limited to this spatiotemporal region and cannot necessarily be generalized to other regions where other social and cultural values and customs are observed. Items that resonate with Flemish individuals may not necessarily be meaningful amongst other audiences. Hence, future research may wish to adapt the operationalization and measurement of the concepts used in this study to account for potential social and cultural differences in other samples. Doing so may help to generalize our findings to other populations.

This study utilizes an online web survey because of the cost-effective and efficient nature of such a data collection method. However, researchers cannot completely monitor the processes of participant selection using this approach and questionnaire responses cannot be completely verified due to the online nature of the survey. Thus, the impossibility of monitoring response selection, self-selection bias, and internet availability is important to consider. However, these drawbacks are present in traditional survey modes as well. Furthermore, due to the lack of randomization, web surveys contribute more to research investigating the causes and correlates of crime, rather than studying the prevalence of the phenomenon. Yet, scholars should consider that this approach only works if enough respondents indicate activity in the dependent variable.

Relatedly, the study is cross-sectional and therefore causal inferences cannot be made. Cautionary notes have been raised against the misleading use of causal modelling (especially SEM). Although SEM is a powerful statistical tool, it cannot enable causal conclusions to be drawn from correlational data. Hox and Bechger (1998) caution that while a SEM model that has been corroborated by data can empirically support hypothesised relationships, it does not discount the possibility that other competing models may exist (p.15) (see also Bunge, 2004; Cliff, 1983; Wikström, 2007). Furthermore, our model is a recursive model, that is, one where all postulated effects are unidirectional. Feedback loops or reciprocal effects were excluded from our model. Panel data are more appropriate to assess cross-lagged effects. Given the potential bias this method entails – as does any method – cross-referencing our findings with results from longitudinal and experimental designs is necessary and thus strongly recommended.

9 Conclusion

Our study provides further evidence highlighting the value of developing integrated frameworks to understand politically motivated violence. By taking an interdisciplinary approach, we have been able to demonstrate connections between varying drivers of individual differences of both political violence and right-wing extremist group participation. Rarely are both addressed in one study. The inclusion of these predictors may also inform ways that political violence can be prevented. Scant research exists that explicitly examines the indirect, long-term and life-course oriented protective factors that prevent extremism. Hence, further research is warranted to examine the efficacy of such protective factors to complement the picture presented by etiological studies such as ours. Without a thorough overview of the factors pertaining to political violence, efficient preventative strategies and policies can presumably never be conceived of. It is not only the explanation of the behaviour, but equally the explanation of the absence of the behaviour that is required to elicit positive change.


For access to the supplementary materials, please contact Lieven Pauwels at


Agnew, R. (2006). Pressured into crime: an overview of general strain theory. Oxford, UK: Oxford University Press. Search in Google Scholar

Agnew, R. (2011). Chaper 6: Revitalizing Merton: General strain theory. In F. T. Cullen, C. Lero Jonson, A. J. Myer, & F. Adler (Eds.), The Origins of American Criminology (Vol. 16, pp. 137-158). Search in Google Scholar

Akers, R. L. (1998). Social structure and social learning. Los Angeles: Roxbury. Search in Google Scholar

Antonaccio, O., & Tittle, C. R. (2008). Morality, self‐control, and crime. Criminology, 46(2), 479-510. Search in Google Scholar

Beelmann, A. (2020). A social-developmental model of radicalization: a systematic integration of existing theories and empirical research. International Journal of Conflict and Violence (IJCV), 14(1), 1-14. Search in Google Scholar

Bendixen, M., Endresen, I. M., & Olweus, D. (2006). Joining and leaving gangs: Selection and facilitation effects on self-reported antisocial behaviour in early adolescence. European Journal of Criminology, 3(1), 85-114. Search in Google Scholar

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological bulletin, 107(2), 238. Search in Google Scholar

Bircan, T. (2012). Community stucture and ethnocentrism. A mulitlevel approach: A case study of Flanders (Belgium). (PhD dissertation). KU Leuven, Leuven, Belgium. Search in Google Scholar

Bjørgo, T. (2002). Exit neo-Nazism: Reducing recruitment and promoting disengagement from racist groups. Retrieved from Oslo, Norway: Search in Google Scholar

Bjørgo, T., & Carlsson, Y. (2005). Early intervention with violent and racist youth groups (0800-0018). Retrieved from Oslo, Norway: Search in Google Scholar

Blackbourn, J., McGarrity, N., & Roach, K. (2019). Understanding and responding to right wing terrorism. Journal of policing, intelligence and counter terrorism, 14(3), 183-190. Search in Google Scholar

Borum, R. (2011). Radicalization into violent extremism II: A review of conceptual models and empirical research. Journal of strategic security, 4(4), 37-62. Search in Google Scholar

Borum, R. (2014). Psychological vulnerabilities and propensities for involvement in violent extremism. Behavioral sciences & the law, 32(3), 286-305. Search in Google Scholar

Bottoms, A. E. (2002). Morality, crime, compliance and public policy. Paper presented at the I. Paper presented at the Ideology, Crime and Criminal Justice: A Symposium in Honour of Sir Leon Radzinowizs., Cambridge, UK. Search in Google Scholar

Bunge, M. (2004). How does it work? The search for explanatory mechanisms. Philosophy of the social sciences, 34(2), 182-210. Search in Google Scholar

Burton Jr, V. S., Cullen, F. T., Evans, T. D., Alarid, L. F., & Dunaway, R. G. (1998). Gender, self-control, and crime. Journal of research in crime and delinquency, 35(2), 123-147. Search in Google Scholar

Byrne, B. M. (2013). Structural equation modeling with Mplus: Basic concepts, applications, and programming: routledge. Search in Google Scholar

Byrne, B. M., Shavelson, R. J., & Muthén, B. (1989). Testing for the equivalence of factor covariance and mean structures: the issue of partial measurement invariance. Psychological bulletin, 105(3), 456. Search in Google Scholar

Canna, S., Day, J., & Popp, G. (2020). Common Characteristics of« Successful« Deradicalization Programs of the Past. Retrieved from Washington DC, US: Search in Google Scholar

Cliff, N. (1983). Some cautions concerning the application of causal modelling methods. Multivariate Behavioral Research, 18, 115-126. doi: Search in Google Scholar

Cudeck, R., & Browne, M. W. (1983). Cross-validation of covariance structures. Multivariate Behavioral Research, 18(2), 147-167. Search in Google Scholar

Cunningham, W. A., Nezlek, J. B., & Banaji, M. R. (2004). Implicit and explicit ethnocentrism: Revisiting the ideologies of prejudice. Personality and Social Psychology Bulletin, 30(10), 1332-1346. Search in Google Scholar

De Waele, M. (2015). Angry white rebel?: study on the mechanisms and processes of participation in extreme-right groups. (PhD dissertation). Ghent University, Ghent. Search in Google Scholar

De Waele, M., & Pauwels, L. (2014). Youth involvement in politically motivated violence: why do social integration, perceived legitimacy, and perceived discrimination matter? International Journal of Conflict and Violence (IJCV), 8(1), 134-153. Search in Google Scholar

De Waele, M., & Pauwels, L. (2016). Why do Flemish youth participate in right-wing disruptive groups? In C. Maxson & F. Esbensen (Eds.), Gang transitions and transformations in an international context (pp. 173-200). New York, US: Springer. Search in Google Scholar

Decker, S. H., Melde, C., & Pyrooz, D. C. (2013). What do we know about gangs and gang members and where do we go from here? Justice Quarterly, 30(3), 369-402. Search in Google Scholar

Doosje, B., Loseman, A., & Van Den Bos, K. (2013). Determinants of radicalization of Islamic youth in the Netherlands: Personal uncertainty, perceived injustice, and perceived group threat. Journal of Social Issues, 69(3), 586-604. Search in Google Scholar

Doosje, B., Van den Bos, K., Loseman, A., Feddes, A. R., & Mann, L. (2012). »My in‐group is superior!«: Susceptibility for radical right‐wing attitudes and behaviors in dutch youth. Negotiation and Conflict Management Research, 5(3), 253-268. Search in Google Scholar

Doosje, B., & van Eerten, J. J. (2017). ‘Counter-narratives’ against violent extremism. In L. Colaert (Ed.), ‘De-radicalisation’: Scientific insights for policy (pp. 83-100). Brussels, Belgium: Flemish Peace Institute. Search in Google Scholar

Duckitt, J., & Sibley, C. G. (2007). Right wing authoritarianism, social dominance orientation and the dimensions of generalized prejudice. European Journal of Personality: Published for the European Association of Personality Psychology, 21(2), 113-130. Search in Google Scholar

Elliott, D. S., Ageton, S. S., & Canter, R. J. (1979). An integrated theoretical perspective on delinquent behavior. Journal of research in crime and delinquency, 16(1), 3-27. Search in Google Scholar

Farrington, D. P. (2020). The integrated cognitive antisocial potential (ICAP) theory: past, present, and future. Journal of Developmental and Life-Course Criminology, 6(2), 172-187. Search in Google Scholar

Gaertner, S. L., Dovidio, J. F., Anastasio, P. A., Bachman, B. A., & Rust, M. C. (1993). The common ingroup identity model: Recategorization and the reduction of intergroup bias. European review of social psychology, 4(1), 1-26. Search in Google Scholar

Geiser, C. (2012). Data analysis with Mplus. London, UK: The Guilford Press. Search in Google Scholar

Gottfredson, M. R., & Hirschi, T. (1990). A General Theory of Crime. Stanford, CA: Stanford University Press. Search in Google Scholar

Grasmick, H. G., Tittle, C. R., Bursik Jr, R. J., & Arneklev, B. J. (1993). Testing the core empirical implications of Gottfredson and Hirschi's general theory of crime. Journal of research in crime and delinquency, 30(1), 5-29. Search in Google Scholar

Grumke, T. (2017). Globalized Anti-Globalists. The Ideological Basis of the Internationalization of Right-Wing Extremism. Perspective Politice, 10(1), 45-52. Search in Google Scholar

Gurr, T. R. (1970). Why men rebel. Princeton: Princeton University Press. Search in Google Scholar

Hagan, J., Merkens, H., & Boehnke, K. (1995). Delinquency and disdain: Social capital and the control of right-wing extremism among East and West Berlin youth. American Journal of Sociology, 100(4), 1028-1052. Search in Google Scholar

Hardy, K. (2019). Countering right-wing extremism: lessons from Germany and Norway. Journal of policing, intelligence and counter terrorism, 14(3), 262-279. Search in Google Scholar

Hardyns, W., Thys, J., Dorme, L., Klima, N., & Pauwels, L. (2021). A multi-agency approach to prevent violent radicalisation. Radices, 1(1), 22-40. Retrieved from Search in Google Scholar

Hirschi, T. (1969). Causes of Delinquency Berkeley, CA: University of California Press. Search in Google Scholar

Hox, J. J., & Bechger, T. M. (1998). An introduction to structural equation modeling. Search in Google Scholar

Hu, L. t., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural equation modeling: a multidisciplinary journal, 6(1), 1-55. Search in Google Scholar

Jones, S. G., Doxsee, C., & Harrington, N. (2020). The Right-wing Terrorism Threat in Europe. Washington DC, US: Center for Strategic & International Studies. Search in Google Scholar

Kaakinen, M., Sirola, A., Savolainen, I., & Oksanen, A. (2020). Impulsivity, internalizing symptoms, and online group behavior as determinants of online hate. PloS one, 15(4), 1-17. Search in Google Scholar

Kline, R. B. (2015). Principles and practice of structural equation modeling. New York, US: The Guilford Press. Search in Google Scholar

Krakowski, M. I., & Czobor, P. (2014). Depression and impulsivity as pathways to violence: implications for antiaggressive treatment. Schizophrenia bulletin, 40(4), 886-894. Search in Google Scholar

Kruglanski, A. W., Chen, X., Dechesne, M., Fishman, S., & Orehek, E. (2009). Fully committed: Suicide bombers‹ motivation and the quest for personal significance. Political psychology, 30(3), 331-357. Search in Google Scholar

Lösel, F., King, S., Bender, D., & Jugl, I. (2018). Protective factors against extremism and violent radicalization: A systematic review of research. International journal of developmental science 12(1-2), 89-102. Search in Google Scholar

Marsh, H. W., Wen, Z., & Hau, K.-T. (2004). Structural equation models of latent interactions: evaluation of alternative estimation strategies and indicator construction. Psychological methods, 9(3), 275-300. Search in Google Scholar

McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205. Search in Google Scholar

Moffitt, T. E. (2017). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy: Routledge. Search in Google Scholar

Muthén, L. K., & Muthén, B. O. (2012). MPlus: statistical analysis with latent variables–User's guide. Wiley: New York, NY. Search in Google Scholar

Nivette, A., Eisner, M., & Ribeaud, D. (2017). Developmental predictors of violent extremist attitudes: A test of general strain theory. Journal of research in crime and delinquency, 54(6), 755-790. Search in Google Scholar

Opp, K.-D. (2009). Theories of political protest and social movements: A multidisciplinary introduction, critique, and synthesis. Oxon, UK: Routledge. Search in Google Scholar

Pantucci, R., & Ong, K. (2021). Persistence of right-wing extremism and terrorism in the West. Counter Terrorist Trends and Analyses, 13(1), 118-126. Retrieved from Search in Google Scholar

Pauwels, L., Brion, F., & De Ruyver, B. (2014). Explaining and understanding the role of exposure to new social media on violent extremism. An integrative quantitative and qualitative approach. Ghent, Belgium: Academia Press. Search in Google Scholar

Pauwels, L., & Heylen, B. (2020). Perceived group threat, perceived injustice, and self-reported right-wing violence: An integrative approach to the explanation right-wing violence. Journal of interpersonal violence, 35(21-22), 4276–4302. Search in Google Scholar

Pauwels, L., Ljujic, V., & De Buck, A. (2018). Individual differences in political aggression: The role of social integration, perceived grievances and low self-control. European Journal of Criminology, online first, 1-25. Search in Google Scholar

Pauwels, L., & Schils, N. (2016). Differential online exposure to extremist content and political violence: Testing the relative strength of social learning and competing perspectives. Terrorism and Political Violence, 28(1), 1-29. Search in Google Scholar

Pauwels, L., & Svensson, R. (2013). Violent youth group involvement, self-reported offending and victimisation: an empirical assessment of an integrated informal control/lifestyle model. European journal on criminal policy and research, 19(4), 369-386. Search in Google Scholar

Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gottfredson and Hirschi's general theory of crime: A meta‐analysis. Criminology, 38(3), 931-964. Search in Google Scholar

Pratt, T. C., Cullen, F. T., Sellers, C. S., Thomas Winfree Jr, L., Madensen, T. D., Daigle, L. E., . . . Gau, J. M. (2010). The empirical status of social learning theory: A meta‐analysis. Justice Quarterly, 27(6), 765-802. Search in Google Scholar

Quillian, L. (1995). Prejudice as a response to perceived group threat: Population composition and anti-immigrant and racial prejudice in Europe. American sociological review, 60(4), 586-611. Search in Google Scholar

Riek, B. M., Mania, E. W., & Gaertner, S. L. (2006). Intergroup threat and outgroup attitudes: A meta-analytic review. Personality and social psychology review, 10(4), 336-353. Search in Google Scholar

Sampson, R. J., & Laub, J. H. (2003). Life‐course desisters? Trajectories of crime among delinquent boys followed to age 70. Criminology, 41(3), 555-592. Search in Google Scholar

Schils, N., & Verhage, A. (2017). Understanding how and why young people enter radical or violent extremist groups. International Journal of Conflict and Violence (IJCV), 11(2), 1-18. Search in Google Scholar

Sherif, M. (1958). Superordinate goals in the reduction of intergroup conflict. American Journal of Sociology, 63(4), 349-356. Search in Google Scholar

Silver, J. R., & Silver, E. (2021). The nature and role of morality in offending: A moral foundations approach. Journal of research in crime and delinquency, 58(3), 343-380. Search in Google Scholar

Skiple, A. (2018). Youth delinquency or everyday racism? Front-line professionals’ perspectives on preventing racism and intolerance in Sweden. Journal for Deradicalization(14), 52-78. Search in Google Scholar

Stams, G. J., Brugman, D., Deković, M., Van Rosmalen, L., Van Der Laan, P., & Gibbs, J. C. (2006). The moral judgment of juvenile delinquents: A meta-analysis. Journal of abnormal child psychology, 34(5), 692-708. Search in Google Scholar

Statista. (2021). Facebook user share in Belgium 2021, by age group and gender. Retrieved from Search in Google Scholar

Steizinger, J. (2018). The significance of dehumanization: Nazi ideology and its psychological consequences. Politics, Religion & Ideology, 19(2), 139-157. Search in Google Scholar

Stephan, W. G., & Stephan, C. W. (2000). An integrated threat theory of prejudice. In S. Oskamp (Ed.), Reducing prejudice and discrimination (pp. 23-45). Mahwah, NJ: Lawrence Erlbaum Associates Inc. Search in Google Scholar

Svensson, R., Pauwels, L., & Weerman, F. M. (2010). Does the effect of self-control on adolescent offending vary by level of morality? A test in three countries. Criminal Justice and Behavior, 37(6), 732-743. Search in Google Scholar

Tajfel, H. (1969). Cognitive aspects of prejudice. Journal of biosocial science, 1(S1), 173-191. Search in Google Scholar

Turner, J., & Tajfel, H. (1974). The social identity theory of intergroup behavior. Psychology of Intergroup Relations, 13(2), 65-93. Search in Google Scholar

Van den Bos, K., Loseman, A., & Doosje, B. (2009). Waarom jongeren radicaliseren en sympathie krijgen voor terrorisme: Onrechtvaardigheid, onzekerheid en bedreigde groepen. Retrieved from Amsterdam: Search in Google Scholar

van Der Valk, I., & Wagenaar, W. (2010). Monitor Racisme & Extremisme: in en uit extreemrechts. Amsterdam: Amsterdam University Press. Search in Google Scholar

van Prooijen, J.-W., & Krouwel, A. P. (2019). Psychological features of extreme political ideologies. Current Directions in Psychological Science, 28(2), 159-163. Search in Google Scholar

Velasco González, K., Verkuyten, M., Weesie, J., & Poppe, E. (2008). Prejudice towards Muslims in the Netherlands: Testing integrated threat theory. British journal of social psychology, 47(4), 667-685. Search in Google Scholar

Wang, J., & Wang, X. (2019). Structural equation modeling: Applications using Mplus. New Jersey: John Wiley & Sons. Search in Google Scholar

Warr, M. (2002). Companions in crime: The social aspects of criminal conduct. Cambridge, UK: Cambridge University Press. Search in Google Scholar

Watts, M. W. (2001). Aggressive youth cultures and hate crime: Skinheads and xenophobic youth in Germany. American Behavioral Scientist, 45(4), 600-615. Search in Google Scholar

Weisburd, D., & Piquero, A. R. (2008). How well do criminologists explain crime? Statistical modeling in published studies. Crime and Justice, 37(1), 453-502. Search in Google Scholar

Wikström, P. O. H. (2007). In search of causes and explanations of crime. In R. King & E. Wincup (Eds.), Doing research on crime and justice (2nd ed., pp. 117-139). Oxford: Oxford University Press. Search in Google Scholar

Wikström, P. O. H., & Bouhana, N. (2017). Analyzing radicalization and terrorism: A situational action theory. In G. LaFree & J. D. Freilich (Eds.), The handbook of the criminology of terrorism (pp. 175-186). Sussex, UK: John Wiley & Sons. Search in Google Scholar

Willems, H. (1995). Right‐wing extremism, racism or youth violence? Explaining violence against foreigners in Germany. Journal of Ethnic and Migration Studies, 21(4), 501-523. Search in Google Scholar

Williamson, H., Murphy, K., & Sargeant, E. (2020). The grievance-identity relationship: Understanding the role of identity processes and stigmatisation on Muslims’ perceptions of terrorist grievances. Journal of policing, intelligence and counter terrorism, online first, 1-19. doi: Search in Google Scholar

Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2020). A field-wide systematic review and meta-analysis of putative risk and protective factors for radicalization outcomes. Journal of Quantitative Criminology, 36(3), 407-447. Search in Google Scholar

Published Online: 2021-08-18
Published in Print: 2021-09-14

© 2021 Walter de Gruyter GmbH, Berlin/Boston