1
[â¦] the developmental tasks of adolescentsâamong them developing a personal, social, and political identity, redefining bonds and relations, and forming new relationshipsâmake them vulnerable to radical beliefs.
(Pels & de Ruyter, 2012, p. 313)
Political and religious radicalization and extremism pose a threat to the stability of democratic societies (e.g. Ghosh et al., 2017). Both the term extremism and the term radicalization have been defined in various ways (Beelmann, 2020), but a generalized specific terminology has not yet been developed (Lösel et al., 2018). Very simply, political, religious, or other forms of extremism can be described as a âsignificant deviation in attitudes and behaviour from basic legal and political norms and values within a social system (society or state) that seek their (at least partial) abolition and replacementâ (Beelmann, 2020, p. 2). Radicalization denotes the process by which these attitudes (sometimes also referred to as beliefs) and behaviours emerge and develop in individuals and groups, representing a âchange in the individualâs psycho-cognitive construction of new identitiesâ (Ghosh et al., 2017, pp. 6â7). The beliefs adopted in this process serve to justify the use of violence for bringing about political and social change (Doosje et al., 2016; Lösel et al., 2018). Thus, the stability of democracies can be threatened by large groups of people undermining the support of its fundamental values. In the context of the present chapter, we address extremist attitudes and behaviours that are not in line with democratic values.
A growing body of research from various disciplines (e.g. sociology, criminology, political science, psychology) has searched for explanations for how, when, why, and under what circumstances extremism and radicalization emerge and develop, and has explored potential reasons for why they may assume violent forms like terrorism (e.g. Borum, 2011; 2014; Knight et al., 2017). One focus lies on exploring how and why (very) young individuals radicalize themselves. Findings from recent research indicate that social-developmental processes contribute to radicalization and extremism. Therefore, identity problems, prejudice, political or religious ideologies, and antisocial attitudes and behaviours have been described as âcore conditions for radicalization and extremismâ (Beelmann, 2020, p. 1). Further theoretical explanations of radicalization include, for example, the âlone wolfâ trajectory; difficult childhood experiences like rejection, exclusion, or bullying due to a migrant history; association with violent or extremist groups; or identity problems during adolescence (e.g. Beelmann, 2020; Nivette et al., 2017).
Although not all extremism is violent (Knight et al., 2019), there seems to be a danger that, given certain circumstances, violent means might be employed by individuals and groups to achieve their goals, resulting, in extreme cases, in terrorist attacks (e.g. Ghosh et al., 2017). Drawing on Moghaddamâs (2005) metaphor of the âstaircase to terrorismâ, Ghosh et al. (2017) describe a stage model of transformative behaviour with fundamentalism at the bottom, followed by extremism, radicalism and, finally, terrorism in a sequence of continuously narrowing steps. The continuous narrowing of the steps represents the decreasing number of people who can be persuaded to continue âclimbingâ. Preventing (young) individuals from climbing the staircase towards violent and terrorist acts is seen as a core task for both society and education (Ghosh et al., 2017; Moghaddam, 2005).
Whilst the staircase model describes the behavioural side of radicalization with terrorist acts as the culmination of the process, the two-pyramid model by McCauley and Moskalenko (2017) distinguishes between behavioural and cognitive radicalization pathways. The cognitive pathway (i.e. the opinion radicalization pyramid) leads to extremist ideologies and beliefs and must be distinguished from the more dangerous behavioural pathway (i.e. the action radicalization pyramid), which leads to action (see McCauley & Moskalenko, 2017, p. 211). McCauley and Moskalenko underline the necessity for conceptualizing two distinct pathways by referring to the rather weak empirical relationship between extreme (or radical) attitudes and the corresponding behaviours, observing that â99% of those with radical ideas never actâ (p. 212). However, individuals holding radical attitudes may provide moral support, exert moral pressure, and offer (post-hoc) justifications to action-prone individuals, thereby playing an indirect role in supporting radical (and violent) behaviour. Moreover, the endorsement of radical attitudes, including the justification and support for radical behaviours, was found to be a significant risk factor for both radical intentions and behaviours in a recent meta-analysis of 57 studies involving mostly samples comprising (young) adults (Wolfowicz et al., 2020). Accordingly, considering factors which promote or prevent the development
1.1 Developing Extremist Attitudes: Risk and Protective Factors
Various individual, social, and societal/contextual risk and protective factors regarding radical/extremist attitudes and behaviours (perpetration) have been identified, mainly from studies involving (young) adults (see Beelmann, 2020). Meta-analytically confirmed risk and protective factors for developing radical or extreme attitudes were described recently by Wolfowicz et al. (2020). Risk factors include, for example:
- âsociodemographic factors like male gender, a personal status relating to being an immigrant, national/religious identity, low educational attainment, or unemployment;
- ânegative personal experiences like discrimination or exposure to violence;
- âperson-related factors like personality disorders, anxiety, or low self-control;
- âand social (societal) factors like deviant peers or low integration.
Protective factors include, for example:
- âsociodemographic factors like socioeconomic status, education, or marital status;
- âperson-related factors like general trust, depression, school performance, or law abidance;
- âand social factors like school bonding and out-group friendships (Wolfowicz et al., 2020).
The larger number of risk factors reported here reflects the relative dearth of research on protective factors (Beelmann, 2020; Lösel et al., 2018).
More recently, researchers have also begun to investigate risk and protective factors in adolescent populations (Lösel et al., 2018) and have particularly addressed the issue of vulnerability (Harpviken, 2020), the latter being understood as âa state of openness to attack, harm or damageâ (Borum, 2014, p. 291). In his social-developmental model of radicalization, Beelmann (2020) proposes that radicalization occurs in three consecutive stages spanning early childhood to middle adulthood, namely: Stage 1âontogenetical developmental processes; Stage 2âproximal radicalization processes; and Stage 3âpolitical or religious extremism.
Stage 2 of Beelmannâs model encompasses the years from early adolescence to middle adulthood (age fourteen to about age thirty) and is more closely and directly linked to radicalization. It ârepresents the central preconditions for political or religious extremism to emergeâ (2020, p. 8) and covers the life phase during which the radicalization of most extremist offenders occurs. The four proximal processes of radicalization include identity problems, prejudice, political or religious ideologies, and antisocial attitudes and behaviours. These processes both influence and reinforce each other while at the same time exerting distinct influences on extremist attitudes and behaviour. The ontogenetic risk-protection background resulting from Stage 1 forms the backdrop against which current individual, social, or societal problems or events can trigger or reinforce the four proximal radicalization processes, as follows:
- âProximal radicalization process 1: Identity problemsârefers to a thwarted need for appreciation as expressed in feelings of injustice or marginalization which themselves feed as motives into radicalization processes. Intense experiences of rejection (at the individual, social, or collective level) and discrimination as well as low or excessively high self-esteem are the most relevant risk factors contributing to radicalization via identity problems.
- âProximal radicalization process 2: Prejudiceârefers to the social-cognitive conditions of radicalization and extremism. It includes âstrong derogatory schemata regarding members of other social groups (e.g. âforeignersâ, ârefugeesâ, âunbelieversâ)â (p. 8). The resulting negative intergroup attitudes are linked to extremist offences by offering excuses and justifications as well as help condoning derogatory treatment and low sympathy towards victims of such offences. Both the social contexts establishing such structures and a lack of diversity experiences are among the core risk factors for prejudice processes.
- âProximal radicalization process 3: Acquisition of political or religious extremist ideologiesâthese ideologies can be more or less coherent but are used both to justify notions of inequality and to portray as legitimate the use of
illegal and violent means to reach extremist goals. Often, these ideologies are created in the context of in-groups which are also engaged in constructing prejudices. Among core risk factors for adopting such ideologies are certain personality characteristics (e.g. authoritarianism), a societal context which condones these political or religious ideologies, and contact with groups that uphold them. - âProximal radicalization process 4: Antisocial attitudes and behavioursâthese represent the fourth proximal radicalization process and are characterized by a marked tendency to break age-related social rules and norms and engage in antisocial and delinquent behaviours. The strongest risk factors relate to an early onset of aggressive and antisocial behaviours (already at preschool age) as well as to an affiliation with delinquent peer groups in adolescence.
The interaction between the ontogenetic risk-protection history with the âcurrent triggering social context, against the background of social intergroup processesâ (Beelmann, 2020, p. 9) determines the type and strength of the above proximal radicalization processes. In turn, the strength of these proximal radicalization processes is directly related to the risk of developing extremist attitudes and behaviours, i.e. Stage 3: the stronger these processes, the greater the risk. However, protective factors like having a non-deviant partner or stable employment may still buffer the effects from Stage 2. The interplay between the different factors at the various stages also implies that different individual trajectories towards radicalization are possible (Beelmann, 2020). This is also described in Borumâs (2014) worldview model or McCauley and Moskalenkoâs two-pyramid model (e.g. 2017). The multicausality characterizing Beelmannâs (2020) model as well as its developmental perspective render it particularly valuable for attempts to explain how students may develop pre-extremist/fundamentalist political and/or religious attitudes, and how these might contributeâgiven unfavourable social and societal circumstancesâto enhancing their risk of further radicalization. In sum, both âgeneralâ radicalization research targeting mainly (young) adults and developmentally oriented research indicate that radicalization unfolds and intensifies over time and results from an interplay of both individual and social risk and protective factors. This underlines the need for early prevention.
1.2 Fundamentalist Religious Attitudes and Beliefs
Using Moghaddamâs (2005) staircase model as a starting point, we may locate fundamentalism, i.e. the endorsement of fundamentalist attitudes and beliefs, at the lowest step of a potential radicalization trajectory. Alternative definitions of fundamentalism sometimes equate fundamentalism with radicalism
As discussed in the previous sections, however, even the endorsement of extreme attitudes (one step above fundamentalism according to the staircase model) or radical attitudes (one step above extremism) does not automatically lead to radical or terrorist action but rather represents attitudinal propensities and potential psychological vulnerabilities that may become activated in given contexts and based on specific circumstances (e.g. Borum, 2014). According to McCauley and Moskalenko (2017), fundamentalist attitudes must also become linked with a given religious and/or political cause (e.g. the jihadist terrorist cause) to reach higher levels of the opinion pyramid, whereby the highest level represents a perceived moral obligation to take up violence in defence of âthe causeâ.
At this point, we can establish a link to Beelmannâs (2020) social-developmental model of radicalization as described above. Political or religious ideologies represent one out of the four social-developmental (proximal) processes that are triggered by actual conflicts (social, societal, or individual) and that are âmarked by continuous intergroup processesâ (p. 1). The likelihood of developing extremist attitudes and behaviours is positively related to the intensity of these proximal processes. For religious or political ideologies, this means that the more pronounced they are, the greater the likelihood is that they contribute to developing extremist attitudes and behaviours in the context of particular triggering factors. Accordingly, the endorsement of fundamentalist religious or political attitudes may be understood as a potential openness towards adopting more extreme forms of related religious or political ideologies.
From an empirical perspective, behavioural radicalization represents a rather simple, specific, and relatively easy to measure construct, whereas cognitive radicalization according to the attitude pyramid (McCauley & Moskalenko, 2017) constitutes a less specific and more complex construct. Thus, cognitive radicalization is often assessed using a variety of more or less specific proxy variables, whereby measures of the support or the justification for radical violence and terrorism are among the most specific (Wolfowicz et al., 2020). Consequently, attitudes and beliefs as cognitive outcomes of radicalization can be
2 The Role of Socialization and Education at School in Promoting or Counteracting Radicalization
The contribution of socialization and education at school to promoting or preventing radicalization and extremism is not yet clear (e.g. Pels & de Ruyter, 2012; Vergani et al., 2020). Some research identifies lower levels of education as a push factor, indicating that, for example, lower levels of education are associated with strongly dichotomous worldviews and identities that are themselves predictive of radicalization (Vergani et al., 2020). The meta-analysis by Wolfowicz et al. (2020) indicates that higher levels of education represent a protective factor against developing both radical attitudes and radical intentions, although the effect sizes found were small. In the systematic review by Lösel et al. (2018), attending higher education and good levels of achievement in school were also identified as protective factors against extremism and radicalization, with the former playing an important role regarding religious/ethnic ideology. Therefore, education in schools may help to foster protective factors against radicalization. In addition, higher levels of civic knowledge may foster studentsâ ability to understand other groupsâ points of view, and fostering social relationships may promote affective bonds between students of different groups, thereby increasing tolerance (Ziemes & Abs, 2020).
Generally it seems that until recently, radicalization research has only marginally considered the role of childrenâs and adolescentsâ socialization and educational environments (Pels & de Ruyter, 2012). However, as discussed earlier, radicalization processes start at school age (Beelmann, 2020). Youngsters are particularly vulnerable to endorsing radical beliefs during their adolescent years, mainly because of the developmental tasks they are facing, such as âdeveloping a personal, social, and political identity, redefining bonds and relations and forming new relationshipsâ (Pels & de Ruyter, 2012, p. 313). Therefore, in order to better understand radicalization processes, it is important to
2.1 The Quality of Social Relationships in School and Student Outcomes
The quality of social relationships at school is meaningfully related to core academic, social, and health outcomes in students. The quality of the studentâteacher relationship has been linked to studentsâ social, emotional, behavioural, and academic outcomes (e.g. Farmer et al., 2011; see also the review by McGrath & Van Bergen, 2015). Positive studentâteacher relationships are associated with outcomes like an enhanced sense of school belonging (Wong et al., 2019), academic achievement (Hattie, 2009), peer liking (Hendrickx et al., 2016), and prosocial behaviour (Luckner & Pianta, 2011). Moreover, they have been found to prevent peer aggression, specifically bullying and victimization (e.g. Konishi et al., 2010). In addition to their predictive function, positive studentâteacher relationships have been identified as a protective factor for students in general and particularly for those students at risk of both socioemotional and academic difficulties (McGrath & Van Bergen, 2015; Sabol & Pianta, 2012). A positive relationship between teachers and student may also buffer adverse effects of ethnic discrimination by peers. A recent study by Civitillo et al. (2021) found that a high quality of the student-teacher relationship attenuated the negative association between perceived personal and group discrimination and both global self-esteem and emotional school engagement.
Regarding the quality of studentâstudent or peer relationships, research has consistently indicated that positive relationships are important for studentsâ cognitive, academic, motivational, emotional, and social development (e.g. Bukowski, 2001; Deci & Ryan, 2008; Hay et al., 2004), and that they are positively related both to trust in political institutions (Ziemes et al., 2020) and to tolerance (Ziemes & Abs, 2020). Additionally, positive studentâstudent relationships have been found to prevent peer aggression, specifically bullying and victimization (e.g. Konishi et al., 2010).
Bullying (perpetration) and victimization (being the target of bullying) are indicators of dysfunctional, negative peer relations (Lenci & Matuga, 2010). Bullying is characterized by a certain repetitiveness and an imbalance of power between bully and victim (e.g. Olweus, 2003) and manifests a genuine lack of respect (see Gutzwiller-Helfenfinger & Ziemes, 2017). As constructs of âothernessâ lie at the heart of bulliesâ and bystandersâ justifications for their behaviour, they use all possible facets of diversity to justify and âlegitimizeâ their systematic aggression against others as well as the discrimination and marginalization of others. Student characteristics related to migration, ethnicity,
Finally, the generalized quality of social relationships at school (often conceptualized as school climate) has been identified as one essential factor determining core student outcome variables like, for example, academic achievement, school absenteeism, aggressive behaviour, or psychosocial health and wellbeing (e.g. Durlak et al., 2011; Wang et al., 2016). Positive social relationships at school fulfil studentsâ basic need for social belonging, i.e. to feel connected to others, to be accepted, respected, and appreciated, as described by self-determination theory (e.g. Ryan & Deci, 2001).
2.2 Relationship Experiences at School and Radicalization
We may assume that negative relationship experiences at school can contribute to radicalization processes, whereas positive social relationships at school and a feeling of school connectedness or school bonding may represent factors preventing or reducing extremism (Roffey & Boyle, 2018; Lösel et al., 2018; Wolfowicz et al., 2020). Regarding the former, negative experiences with peers and relevant adults (e.g. bullying, social exclusion, discrimination) have been identified as one level of strain that renders adolescents vulnerable to extremism (see the review by Harpviken, 2020). These experiences of violence and abuse seem to operate both at the level of attitudes and behaviours: on the one hand, they heighten the risk that individuals endorse extremist attitudes; on the other hand, they increase the risk that individuals become involved in violent extremism (Knapton, 2014; Knight et al., 2017). However, there seems to be a lack of studies which specifically target the role that the quality of teacherâstudent and studentâstudent relationships plays in an individualâs radicalization.
Beelmannâs (2020) conceptualization of political or religious extremism âas an expression of deviant social developmentâ (p. 5) offers a more integrated view of the role which social relationships in general play in radicalization. Based on his model as described above, we may hypothesize that negative
School-based experiences of victimization, discrimination, marginalization, and rejection all have one factor in common: they indicate to individuals that they do not belong to the school and classroom community. Accordingly, individualsâ basic need to belong is thwarted (see Roffey & Boyle, 2018), rendering them more vulnerable to extremist attitudes and behaviours (Beelmann, 2020). Conversely, positive relationship experiences at school may act as a protective factor by making individuals feel welcome and accepted. However, systematic research is scarce, particularly regarding the relationship between teachers and students, for which no studies seem to exist. In sum, school represents a potentially relevant context for radicalization, regarding both the academic and the social context. Whether studentsâ need to belong is fulfilled or thwarted may thus impact their potential openness to extremism.
3 The Present Study
Radicalization represents a developmental process over time, with proximal radicalization processes spanning from age fourteen (early adolescence) to about age thirty (Beelmann, 2020, p. 8). Therefore, radicalization processes that are potentially proximal and more intensified start to play out during a critical phase. In early adolescence, students attend secondary school and often must adapt to a new school environment. They face challenges in their individual, social, and academic development. The developmental task of forming a personal, social, and political identity renders them especially vulnerable to radical beliefs (Pels & de Ruyter, 2012). In other words, potential proximal radicalization processes coincide with a developmental phase during which students are particularly vulnerable to such processes.
Against this background, we explored the link between the quality of social relationships at school and pre-extremist attitudes in multi-ethnic classrooms.
As core sociodemographic variables have been shown to be related to radical attitudes (some as risk and some as protective factors; see Wolfowicz et al., 2020), we included gender, age, socioeconomic status, and immigration background as additional potential predictors. Another potentially relevant area in studentsâ lives refers to religion. Research suggests that whereas religious practice and adherence seem to play only a marginal role as a risk factor for developing radical attitudes, religious identity seems to have a much stronger influence (e.g. Koopmans, 2015; see Wolfowicz et al., 2020). Accordingly, we included both the religion with which students identified and their religious practice as additional potential predictors.
Furthermore, positive attitudes towards society and political values such as democracy have been identified as an extremism-specific protective factor (Beelmann, 2020, p. 6). However, there seems to be little comprehensive research on whether and to what extent political education (e.g. citizenship education) is effective in preventing the development of political or religious extremist ideologies (Beelmann, 2020). To explore a potential relationship between citizenship education and the endorsement of pre-extremist attitudes, we therefore included studentsâ civic knowledge as a predictor in our analyses.
4 Method
Our analyses use the German data from ICCS 2016. The study was conducted by the International Association for the Evaluation of Educational Achievement (IEA) and aimed to assess how students across the world are being prepared to become citizens (Schulz et al., 2017). Prior studies were conducted in
4.1 Sample
To account for the multilayered educational system and an uneven distribution of students with an immigration background in NRW, a stratified, multistage randomized cluster sample was drawn. Sampling strata were school type and proportion of students with immigration backgrounds (more versus less than 30%). The sample consisted of 1,451 eighth-grade students from 59 secondary schools. Studentsâ mean age was 14.29 (range: 12.08 to 16.75). 52% of students were girls, and about 40% had an immigration background. Students were categorized as having an immigration background if they themselves or one of their parents were born outside of Germany. Distribution of students across school types was as follows: 10.27% attended Hauptschule (lowest academic track), 18.33% Realschule (middle academic track), 15.92% Gesamtschule (comprehensive school), and 55.48% attended Gymnasium (highest academic track). Written consent from a parent or guardian was obtained for all students of the sample. Detailed information on the German sample can be found in Ziemes et al. (2017).
4.2 Measures and Procedure
Data was collected from students, teachers, and headteachers. Our present study focuses on student data only, which was collected using paper-and-pencil questionnaires. Data collection took place in studentsâ regular classrooms and lasted about 135 minutes, with breaks between individual parts of the survey. In the first part, students completed a civic knowledge test covering the areas of civic society and systems, civic principles, civic participation, and civic identities (Schulz et al., 2016). The second part consisted of an international student questionnaire assessing affective-behavioural and contextual factors (including studentsâ experiences of social relationships at school) as well as sociodemographic variables. In the third part, studentsâ attitudes toward civic society and systems in the European context were assessed through the European student questionnaire. Data from the third part is not addressed in our study.
For all scales, the IEA provided authors with values based on weighted likelihood estimations (WLE) of partial credit models (Schulz et al., 2018). In accordance with data preparation in the 2009 cycle, these scales were calibrated to
4.2.1 Sociodemographic Variables
Gender was assessed with one question with a dichotomous answer choice (0 = boy; 1 = girl). The indicator for socioeconomic status (SES) as provided by IEA is a nationally z-standardized indicator which includes the number of books at home, parental occupation, and education.
Religion was assessed with a single choice question which asked students âWhat is your religion?â. The answer choices were âno religionâ (N = 182); âChristianityâ (N = 973); âIslamâ (N = 195); âJudaismâ (N = 5); and âother religionâ (N = 51). Three dichotomous indicators were created for âno religionâ, âChristianityâ and âIslamâ, whereby students self-identifying with these choices were assigned 1 and all others were assigned 0.
Attendance at religious services was assessed with one question where students could indicate if they attended religious services. Answer options were âneverâ (N = 291), âless than once a yearâ (N = 137), âat least once a yearâ (N = 437), âat least once a monthâ (N = 321), or âat least once a weekâ (N = 221). 35 students who indicated âno religionâ still reported that they attended religious services once a year or more often.
4.2.2 Civic Knowledge
Civic knowledge was designed as a competence measure using both multiple choice and open-ended tasks. Based on Rasch analyses, five plausible values were calculated for each student and introduced in the calculations (Schulz et al., 2017).
4.2.3 Quality of Studentsâ Social Relationships at School
The quality of studentâteacher relationships at school was assessed using the studentsâ perspective. Three items measured studentsâ assessment of their own relationships with teachers (e.g. âMost of my teachers treat me fairlyâ). Two items assessed studentsâ evaluation of the quality of the relationship between teachers and the other students (e.g. âStudents get along well with most teachersâ). All items had a 4-point Likert-type response scale ranging from 1 (âstrongly agreeâ) to 4 (âstrongly disagreeâ). Internal consistency was .80.
The quality of studentâstudent relationships at school was also assessed using the studentsâ perspective. Two items described the way students treat each
Studentsâ self-reported victimization at school measured the frequency of self-reported victimization by students within the past three months. The six items included direct, indirect, verbal, physical and cyber-victimization (e.g. âA student said things about you to make others laughâ). The 4-point Likert-type response scale ranged from 1 (ânot at allâ) to 4 (âfive times or moreâ). Internal consistency was .74. For further details, please see Schulz et al. (2018, p. 163).
4.2.4 Pre-Extremist Attitudes
Studentsâ pre-extremist attitudes were assessed as part of a broader scale on studentsâ attitudes on religion in society. Items were answered on a 4-point Likert scale (1 = âstrongly agreeâ to 4 = âstrongly disagreeâ). Three items of the broader scale ascribe religion a dominant role in social and political life (e.g. âRules of life based on religion are more important than civil lawsâ). For the purposes of our study, these items were identified as representing a pre-extremist attitude. The items were combined to form a latent scale which we call hegemony of religion. Internal consistency was .87. Item wording and dichotomous agreement rates for students with and without religion are listed in Table 3.1, as are factor
Items and aggreement rates for hegemony of religion
| Item wording | No religion | Muslim students | Christian students | Factor loading | |||
|---|---|---|---|---|---|---|---|
| % | SE | % | SE | % | SE | ||
| Religion is more important to me than what is happening in national politics. | 5 | 1.68 | 89 | 3.04 | 29 | 1.37 | .825 |
| Religious leaders should have more power in society. | 4 | 1.83 | 61 | 4.35 | 18 | 1.36 | .847 |
| Rules of life based on religion are more important than civil laws. | 5 | 1.22 | 64 | 6.74 | 17 | 1.43 | .835 |
Note: Percentages indicate the percentage of students choosing the âagreeâ or âstrongly agreeâ answer options.
4.3 Results
Weak to moderate bivariate associations were found between the main study variables (Table 3.2). Hegemony of religion was negatively associated with SES and civic knowledge. No significant correlations were found for indicators of the quality of social relationships, i.e. studentâteacher relationship, studentâstudent relationship, and victimization. Therefore, these variables were not included in multivariate analyses.
Bivariate relationships between main study variables
| Hegemony of religion | SES | Civic knowledge | Studentâteacher relationship | Studentâstudent relationship | |
|---|---|---|---|---|---|
| SES | â.228*** | ||||
| Civic knowledge | â.392* | .432* | |||
| Studentâteacher relationship | .021 | .024 | .062 | ||
| Studentâstudent relationship | â.001 | .130*** | .055 | .467*** | |
| Victimization | â.041 | â.024 | â.046 | â.248*** | â.302*** |
Mplus 8.2 was used to predict the latent variable hegemony of religion. The analyses controlled for the clustered nature of the data (type = complex) and employed weights. Model 0 introduced gender, SES, and immigration background. Students with a lower SES and an immigration background were more likely to support a strong role of religion. Model 1 introduced civic knowledge, which shows a negative association with hegemony of religion.
Model 2 introduced studentsâ religion and their attendance at religious services. Only self-identification as Christians or Muslims was used because the other groups were too small. Both religious affiliations predicted support for hegemony of religion. Furthermore, the frequency of attendance at religious services was a much stronger predictor of hegemony of religion than self-identified religion.
Regression models: standardized coefficients for hegemony of religion
| Variables | Model 0 | Model 1 | Model 2 | Model 3 |
|---|---|---|---|---|
| Girl | .036 | .054 | â.005 | â.008 |
| SES | â.113*** | â.015 | â.060â | â.064* |
| IBG | .340*** | .311*** | .165*** | .160*** |
| Civic knowledge | â.246*** | â.184*** | â.190*** | |
| Muslim | .346*** | .544*** | ||
| Christian | .120*** | .132*** | ||
| Religious service | .439*** | .516*** | ||
| Muslim à service | â.263* | |||
| Christ à service | â.058 | |||
| R² | .148 | .197 | .505 | .512 |
| N | 1,382 | 1,382 | 1,360 | 1,360 |
5 Discussion
Our results indicate that a particular constellation of and interplay (as interactions) between factors best predicts hegemony of religion, i.e. our measure of pre-extremist attitudes. Looking at a normative, representative sample of 14-year-old German secondary school students, our findings confirm results from earlier research, including risk and protective factors for radical/extremist attitudes.
Interestingly, indicators of studentsâ relationship quality at school were not related to their endorsement of pre-extremist attitudes. Due to their non-significance, they were excluded from multivariate analyses. The strongest association with hegemony of religion was found for studentsâ self-reported victimization. Those students who reported more victimization also showed a higher endorsement of beliefs supporting hegemony of religion, confirming earlier research (Knapton, 2014; Knight et al., 2017; see also Beelmann, 2020).
There are several possible explanations for the weak results regarding the impact of relationships as found in our study. First, we included a normative sample of 14-year-old students attending regular German secondary schools. We may assume that relationships might have a stronger impact in a sample including at-risk or already radicalized students. As even serious cases of political/religious extremism have very low base rates (Lösel et al., 2020), the variance regarding hegemony of religion in our sample might be too small. Second, our analyses focus on the individual level and do not include shared perceptions of the social climate. This is partly because all relationship variables were assessed at a general level without identifying the respective relationship partners, yet social relationships play out in a given context involving specific relationship partners. Indeed, the quality of relationships has been shown to vary across classrooms and schools, for example in the context of bullying research (Gini et al., 2015). Thus, future research will need to use measurement approaches that allow for the representation of this nestedness and include relationship (and climate) data from multiple informants at different levels (individual, classroom, and school).
Our multivariate analyses indicated that with respect to sociodemographic variables, neither studentsâ gender nor their socioeconomic status were related to the endorsement of beliefs supporting hegemony of religion, whereas an
Civic knowledge proved to be a significant negative predictor of hegemony of religion (Model 1), indicating that students with a higher level of civic knowledge were less likely to endorse beliefs supporting the hegemony of religion. As it remained a significant negative predictor in the subsequent models, we conclude that civic knowledge can be seen as a potentially important protective factor against embracing hegemony of religion. Fostering civic knowledge may be one avenue schools can take to support students in linking their religious identity with democratic attitudes. This interpretation finds some support in the meta-analytic finding by Wolfowicz et al. (2020) that states that anti-democratic values are a risk factor for developing radical attitudes.
We also found that Christians, Muslims, and students who frequently attended religious services were more likely to embrace hegemony of religion. As religious identity and attendance at religious services were also included as interaction terms in Model 3, we will not interpret these main effects here. Furthermore, accounting for these religion-related factors (Model 2) also decreased the predictive power of immigration background. This is in line with previous research showing that the experience of societal rejection is a risk factor for radicalization (e.g. Roffey & Boyle, 2018). Having an immigration background or endorsing a socially less appreciated religion may create feelings of isolation in students. The question whether migration is endangering the political culture of democratic societies is part of an ongoing debate (e.g. Pickel & Pickel, 2018). However, it would be too simplistic to portray immigration as a driving force for the development of pre-extremist attitudes. Our regression models show that the effect of immigrant background decreases continuously with every analytic step. Immigrant background is therefore better perceived as a proximal indicator covering various influences that often
Finally, with regard to the interaction effects of religious identity and attendance at religious services, we found that Muslim students attending religious services were less likely to embrace hegemony of religion. This confirms findings reported in the reviews by Lösel et al. (2018; 2020) which showed that Muslimsâ intense religious practice acted as a protective factor against religious extremism and radicalization. Indeed, Wolfowicz et al. (2020) also found that religious practice and adherence had very small effects on radicalization, whereas religious group identity and religious fundamentalism had larger effects. However, as religious group identity was inextricably linked with national identity in the studies which the authors included in their meta-analysis, and as no interaction effects with religious practice were considered, the relative contribution of religious identity per se cannot be surmised.
Our own findings using pre-extremist attitudes as an outcome in a normative sample of German 14-year-olds suggest that identifying as a Muslim or Christian and practising that religion cannot explain potential openness to radicalization. Still, the main effects for both self-identifying as Muslim and attendance at religious services in Model 2 retained their substantial respective size in Model 3 despite the inclusion of the interaction terms, marking them as potential risk factors. Regarding self-identification as Muslim, a recent study by Goede et al. (2019) including 6,863 German secondary school students (mean age = 14.7) found that 1.5% of Muslim students endorsed extremist/Islamist attitudes and had displayed at least one instance of religiously motivated deviant behaviour. The authors classified these students as at risk for further radicalization. However, unlike in our study, religious practice (including attending religious services) was not identified as a risk factor for extremist attitudes (right-wing, left-wing, Islamist) in the sample as a whole (Goede et al., 2019). Other research (involving mainly adults) also suggests that the effect of religious identity on radicalization by far exceeds that of religious practice (Aly & Striegher, 2012; Wolfowicz et al., 2020; see also the discussion by Koopmans, 2015). As our study used a restricted measurement of religious practice, namely the frequency of attending religious services, further research will need to utilize a broader assessment that includes additional aspects like praying at home, attending religious celebrations, or discussing religious issues (see Goede et al., 2019).
5.1 Limitations
Our study has several limitations. First, we did not study the radicalization process per se but investigated potential predictors of pre-extremist religious attitudes at an age which marks the beginning of the age window during
Second, as already mentioned, our assessment of the quality of social relationships at school focused on the individual level and did not include aggregate measures. This may have blurred potential effects on one of those levels. Future research using measurement approaches that allow for the representation of this nestedness and that include relationship (and climate) data from multiple informants at different levels (individual, classroom, and school) will offer a promising way for exploring the relative contribution of the quality of relationships to young peopleâs endorsement of pre-extremist attitudes. Furthermore, such research might also explore potential mediation or moderation effects of relationship quality and thus account for possible indirect effects.
Third, our rather restricted assessment of pre-extremist attitudes involving only three items does not allow for a fuller representation of such attitudes. Future research might include factors like openness to extremist attitudes or admiration for people endorsing fundamentalist views in order to characterize pre-extremist attitudes more clearly. Capturing the early stages of the radicalization process (Beelmann, 2020) calls for measurements that are adapted to these stages and do not refer to already fully-fledged extremist attitudes and behaviours.
5.2 Practical Relevance and Outlook
Our findings suggest that strategies for the early prevention of (violent) extremism might include addressing pre-extremist attitudes, particularly regarding the role of religion in society. This is relevant for at least two reasons: first, the radicalization process unfolds and intensifies over time (Beelmann, 2020; McCauley & Moskalenko, 2017; Moghaddam, 2005); and second, even if most individuals holding extreme/radical beliefs will never translate them into (terrorist) action, they still indirectly support radical and potentially violent behaviour, for example by offering (post-hoc) justifications to action-prone individuals (McCauley & Moskalenko, 2017).
Educational efforts at school are often designed to help identify young people at risk of radicalization, with a clear security policy focus as, for example, in the context of the Prevent strategy in the United Kingdom (e.g. Mattson et al., 2016; Bryan, this volume). Systematic, evidence-based approaches to prevention seem to be scarce, partly because of a dearth of sound empirical research (Lösel et al., 2018). In their interdisciplinary review of the literature on the prevention of violent extremism, Stephens et al. (2021) identified four core themes
If we map these themes onto Beelmannâs (2020) model of the radicalization process, we see that they can be directly related to proximal radicalization processes, particularly to issues of identity problems, prejudice, and political and religious ideologies. In addition, as the model also considers social and societal influences, it aligns well with Stephens et al.âs (2021) systemic perspective on the promotion of resilience against radicalization. Addressing pre-extremist attitudes in this context might particularly relate to promoting young peopleâs identity search and dialogue. Given early adolescentsâ vulnerability to radical beliefs, our findings that pre-extremist attitudes can be found as early as age fourteen help to make a case for early prevention.
Such early prevention needs to include civic and citizenship education. State school systems were developed within the emerging nation states of the 19th century. They were installed to serve a homogeneous citizenry and to support the idea of a homogenous nation state. But nowadays, it appears that schools must serve ever more heterogeneous societies, with teachers and students being characterized by multilayered or hybrid identities (Abs, 2021). To be a driver for integration, schools need to develop further their potential for integration. Our findings underline the importance of civic knowledge for preventing pre-extremist attitudes that focus on religion. Civic knowledge seems to be an important educational force when it comes to preparing students for developing cohesive perspectives for societies as a whole.
Furthermore, both religious education and citizenship education frequently proceed in our schools without referring to each other (Bacia & Abs, 2017). Our analysis highlights how important it is for both religious and citizenship education to deal with the place of religion in the democratic state (Hahn-Laudenberg & Abs, 2019).
References
Abs, H. J. (2021). Options for developing European strategies on citizenship education. European Educational Research Journal, 20(3), 329â347. https://doi.org/10.1177/1474904121994418
Abs, H. J., & Hahn-Laudenberg, K. (Eds.). (2017). Das politische Mindset von 14-Jährigen. Ergebnisse der International Civic and Citizenship Education Study 2016 [The political mindset of 14-year-olds. Results of the International Civic and Citizenship Education Study 2016]. Waxmann.
Aly, A., & Striegher, J.-L. (2012). Examining the role of religion in radicalization to violent Islamist extremism Studies in Conflict & Terrorism, 35(12), 849â862. doi:
Bacia, E., & Abs, H. J. (2017). Politische und zivilgesellschaftliche Bildung in Kernlehrplänen [Civic and citizenship education in core curricula]. In H. J. Abs & K. Hahn-Laudenberg (Eds.), Das politische Mindset von 14-Jährigen. Ergebnisse der International Civic and Citizenship Education Study 2016 (pp. 27â46). Waxmann.
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), 1â14. doi:
Bhui, K. S., Hicks, M. H., Lashley, M., & Jones, E. (2012). A public health approach to understanding and preventing violent radicalization. BMC Medicine, 10(1), 16. doi:
Borum, R. (2011). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4(4), 7â36. https://doi.org/10.5038/1944-0472.4.4.1
Borum, R. (2014). Psychological vulnerabilities and propensities for involvement in violent extremism. Behavioral Sciences & the Law, 32(3), 286â305. doi:
Bukowski, W. M. (2001). Friendship and the worlds of childhood. In D. W. Nangle & C. A. Erdley (Eds.), The role of friendship in psychological adjustment (pp. 93â105). Jossey-Bass.
Civitillo, S., Göbel, K., Preusche, Z., & Jugert, P. (2021). Disentangling the effects of perceived personal and group ethnic discrimination among secondary school students: The protective role of teacherâstudent relationship quality and school climate. New Directions for Child and Adolescent Development, 2021 (177), 77â99. doi:
Doosje, B., Moghaddam, F. M., Kruglanski, A. W., de Wolf, A., Mann, L., & Feddes, A. R. (2016). Terrorism, radicalization and de-radicalization. Current Opinion in Psychology, 11, 79â84. doi:
Deci, E. L., & Ryan, R. M. (2008). Self-determination theory: A macrotheory of human motivation, development, and health. Canadian Psychology/Psychologie canadienne, 49(3), 182â185. doi:
Driskell, R., Embry, E., & Lyon, L. (2008). Faith and politics: The influence of religious beliefs on political participation. Social Science Quarterly, 89(2), 294â314. doi:
Durkin, K., Hunter, S., Levin, K. A., Bergin, D., Heim, D., & Howe, C. (2012). Discriminatory peer aggression among children as a function of minority status and group proportion in school context. European Journal of Social Psychology, 42(2), 243â251. doi:
Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing studentsâ social and emotional learning: A meta-analysis of school-based universal interventions. Child Development, 82(1), 405â432. doi:
Farmer, T. W., McAuliffe Lines, M., & Hamm, J. V. (2011). Revealing the invisible hand: The role of teachers in childrenâs peer experiences. Journal of Applied Developmental Psychology, 32(5), 247â256. doi:
Ghosh, R., Chan, W. A., Manuel, A., & Dilimulati, M. (2017). Can education counter violent religious extremism? Canadian Foreign Policy Journal, 23(2), 117â133. doi:
Gini, G., Pozzoli, T., & Bussey, K. (2015). The role of individual and collective moral disengagement in peer aggression and bystanding: A multilevel analysis. Journal of Abnormal Child Psychology, 43(3), 441â452. doi:
Goede, L.-R., Schröder, C. P., & Lehmann, L. (2019). Perspektiven von Jugendlichen. Ergebnisse einer Befragung zu den Themen Politik, Religion und Gemeinschaft im Rahmen des Projektes âRadikalisierung im digitalen Zeitalterâ. KFN-Forschungsbericht Nr. 151 [Adolescentsâ perspectives. Results from a survey on politics, religion, and community in the project âradicalization in the digital ageâ]. KFN.
Gutzwiller-Helfenfinger, E., & Ziemes, J. (2017). Qualität der schulischen Sozialbeziehungen [Quality of social relationships at school]. In H. J. Abs & K. Hahn-Laudenberg (Eds.), Das politische Mindset von 14-Jährigen. Ergebnisse der International Civic and Citizenship Education Study 2016 (pp. 279â305). Waxmann.
Hahn-Laudenberg, K., & Abs, H. J. (2019). Religion in research on citizenship education. In D. Käbisch (Eds.), Religion and educational research. National traditions and transnational perspectives (pp. 135â150). Waxmann.
Harpviken, A. N. (2020). Psychological vulnerabilities and extremism among Western youth: A literature review. Adolescent Research Review, 5(1), 1â26. https://doi.org/10.1007/s40894-019-00108-y
Hattie, J. A. C. (2009). Visible learning. A synthesis of over 800 meta-analyses relating to achievement (Reprinted). Routledge.
Hay, D. F., Payne, A., & Chadwick, A. (2004). Peer relations in childhood. Journal of Child Psychology and Psychiatry, 45(1), 84â108. doi:
Hendrickx, M. M. H. G., Mainhard, M. T., Boor-Klip, H. J., Cillessen, A. H. M., & Brekelmans, M. (2016). Social dynamics in the classroom: Teacher support and conflict and the peer ecology. Teaching and Teacher Education, 53, 30â40. doi:
Knapton, H. M. (2014). The recruitment and radicalisation of Western citizens: Does ostracism have a role in homegrown terrorism? Journal of European Psychology Students, 5(1), 38â48. doi:
Knight, S., Keatley, D., & Woodward, K. (2019). Comparing the different behavioral outcomes of extremism: A comparison of violent and non-violent extremists, acting alone or as part of a group. Studies in Conflict & Terrorism, 1â22. doi:
Knight, S., Woodward, K., & Lancaster, G. L. J. (2017). Violent versus nonviolent actors: An empirical study of different types of extremism. Journal of Threat Assessment and Management, 4(4), 230â248. doi:
Konishi, C., Hymel, S., Zumbo, B. D., & Li, Z. (2010). Do school bullying and studentâteacher relationships matter for academic achievement? A multilevel analysis. Canadian Journal of School Psychology, 25(1), 19â39. doi:
Koopmans, R. (2015). Religious fundamentalism and hostility against out-groups: A comparison of Muslims and Christians in Western Europe. Journal of Ethnic and Migration Studies, 41(1), 33â57. doi:
Lenci, M., & Matuga, J. (2010). The lifetime bully: Investigating the relationship between adolescent bullying and depression in early adulthood. Journal of School Counseling, 8(7), 7â7.
Lösel, F., Bender, D., Jugl, I., & King, S. (2020). Resilience against political and religious extremism, radicalization, and related violence: A systematic review of studies on protective factors. In D. Weisburd, E. U. Savona, B. Hasisi, & F. Calderoni (Eds.), Understanding recruitment to organized crime and terrorism (pp. 55â84). Springer International Publishing.
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, 89â102. doi:
Luckner, A. E., & Pianta, R. C. (2011). Teacherâstudent interactions in fifth grade classrooms: Relations with childrenâs peer behavior. Journal of Applied Developmental Psychology, 32(5), 257â266. doi:
Mattsson, C., Hammarén, N., & Odenbring, Y. (2016). Youth âat riskâ: A critical discourse analysis of the European Commissionâs Radicalisation Awareness Network Collection of approaches and practices used in education. Power and Education, 8(3), 251â265. doi:
McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205â216. doi:
McGrath, K. F., & van Bergen, P. (2015). Who, when, why and to what end? Students at risk of negative studentâteacher relationships and their outcomes. Educational Research Review, 14, 1â17. doi:
Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161â169. doi:
Moore, S. E., Norman, R. E., Suetani, S., Thomas, H. J., Sly, P. D., & Scott, J. G. (2017). Consequences of bullying victimization in childhood and adolescence: A systematic review and meta-analysis. World Journal of Psychiatry, 7(1), 60â76. doi:
Nivette, A., Eisner, M., & Ribeaud, D. (2017). Developmental predictors of violent extremist attitudes. Journal of Research in Crime and Delinquency, 54(6), 755â790. doi:
Oliveira, W. A. d., Silva, M. A. I., Mello, F. C. M.d., Porto, D. L., Yoshinaga, A. C. M., & Malta, D. C. (2015). The causes of bullying: Results from the National Survey of School Health (PeNSE). Revista Latino-americana de Enfermagem, 23(2), 275â282.
Olweus, D. (1993). Bullying at school: What we know and what we can do. Blackwell Publishers.
Pels, T., & de Ruyter, D. J. (2012). The influence of education and socialization on radicalization: An exploration of theoretical presumptions and empirical research. Child & Youth Care Forum, 41(3), 311â325. doi:
Pickel, G., & Pickel, S. (2018). Migration als Gefahr für die politische Kultur [Migration as a danger to political culture]? Zeitschrift für Vergleichende Politikwissenschaft, 12(1), 297â320. doi:
Pottie, K., Dahal, G., Georgiades, K., Premji, K., & Hassan, G. (2015). Do first generation immigrant adolescents face higher rates of bullying, violence and suicidal behaviours than do third generation and native born? Journal of Immigrant and Minority Health, 17(5), 1557â1566. doi:
Roffey, S., & Boyle, C. (2018). Belief, belonging, and the role of schools in reducing the risk of home-grown terrorism. In K. Allen & C. Boyle (Eds.), Pathways to school belonging (pp. 149â164). Brill|Sense.
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on hedonic and eudaimonic well-being. Annual Review of Psychology, 52, 141â166. doi:
Sabol, T. J., & Pianta, R. C. (2012). Recent trends in research on teacherâchild relationships. Attachment & Human Development, 14(3), 213â231. doi:
Schulz, W., Ainley, J., Fraillon, J., Losito, B., Agrusti, G., & Friedman, T. (2017). ICCS 2016 International Report. Becoming citizens in a changing world. Springer International Publishing. https://doi.org/10.1007/978-3-319-73963-2
Schulz, W., Carstens, R., Losito, B., & Fraillon, J. (Eds.). (2018). ICCS 2016 Technical Report. IEA.
Stephens, W., Sieckelinck, S., & Boutellier, H. (2021). Preventing violent extremism: A review of the literature. Studies in Conflict & Terrorism, 44(4), 346â361. doi:
Vergani, M., Iqbal, M., Ilbahar, E., & Barton, G. (2020). The three Ps of radicalization: Push, pull and personal. A systematic scoping review of the scientific evidence about radicalization into violent extremism. Studies in Conflict & Terrorism, 43(10), 854. doi:
Wang, C., Hatzigianni, M., Shahaeian, A., Murray, E., & Harrison, L. J. (2016). The combined effects of teacher-child and peer relationships on childrenâs social-emotional adjustment. Journal of School Psychology, 59, 1â11. doi:
Wong, T. K.Y., Parent, A.-M., & Konishi, C. (2019). Feeling connected: The roles of student-teacher relationships and sense of school belonging on future orientation. International Journal of Educational Research, 94, 150â157. doi:
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), 407â447. doi:
Ziemes, J. F., & Abs, H. J. (2020). Welche schulischen Bedingungen sind geeignet um Toleranz zu fördern? Eine Analyse auf Basis der International Civic and Citizenship Education Study 2016 [Which school conditions are suitable for promoting tolerance? An analysis based on the International Civic and Citizenship Education Study 2016]. Die Deutsche Schule, 112(2), 157â177.
Ziemes, J. F., Hahn-Laudenberg, K. & Abs, H. J. (2020). The impact of schooling on trust in political institutionsâDifferences arising from studentsâ immigration backgrounds. Learning, Culture and Social Interaction, 26. https://doi.org/10.1016/j.lcsi.2020.100429
Ziemes, J. F., Jasper, J., Deimel, D., Hahn-Laudenberg, K., & Abs, H. J. (2017). ICCS 2016âDesign und methodisches Vorgehen [ICCS 2106âdesign and methodology]. In H. J. Abs & K. Hahn-Laudenberg (Eds.), Das politische Mindset von 14-Jährigen: Ergebnisse der International Civic and Citizenship Education Study 2016 (pp. 47â76). Waxmann.