Supporting the democratic political organisation of Muslim immigrants: the perspective of Muslims in the Netherlands and Germany

ABSTRACT

Using data from three survey studies, this paper examines the support for the democratic political organisation of Muslims among Muslim immigrants in the Netherlands (Studies 1, 2, and 3) and Germany (Study 3). Using a social psychological perspective, support is examined in relation to religious group identification, Muslim linked fate, perceived discrimination, fundamentalist religious belief, and host national identification. The findings in all three studies show support for the political organisation of Muslims. Furthermore, higher religious group identification and higher linked fate were associated with stronger support. More discrimination and more fundamentalist beliefs were also associated with stronger support, and part of these associations was mediated by linked fate. National identification was not associated with support for the political organisation of Muslims.

KEYWORDS:

• Muslim politics
• immigrants
• linked fate
• discrimination
• group identification

Census projections for countries of the European Union indicate that there will be an increasingly diverse population in the coming years (Eurostat 2010). This raises important and urgent questions, including the question of political participation and representation. In Europe, members of immigrant-origin groups are clearly underrepresented in political institutions and this is not a simple reflection of their recent arrival, delayed acquisition of citizenship, educational qualifications, social experiences, or cultural attitudes (Bloemraad and Schönwälder 2013). This underrepresentation can be seen as a democratic deficit and could increase political alienation among immigrant-origin groups (Pantoja and Segura 2003).

For strategic or principle reasons, mainstream political parties are increasingly recognising the importance to represent immigrant-origin individuals in their membership and candidacy. This can attract new voters from immigrant-origin groups and increase a party’s attractiveness for previous voters (Rocha et al. 2010). In countries with a multi-party political system, immigrants and minorities also have the possibility of participating in the democratic process by organising politically along ethnic or religious lines. For example, in countries such as Spain, Belgium, Denmark, and the Netherlands, there have been several political organisations and parties that try to participate in (local) host society politics by explicitly presenting themselves as Islamic, or that claim to draw inspiration from Islam. In the Netherlands, some of these political parties have secured seats in municipality councils in the last local elections (2014; e.g. ‘NIDA’ in Rotterdam, and ‘Islam Democrats’ in The Hague). This form of political organisation is similar to small Christian-orthodox parties that represent specific sections of the population and try to achieve their particular goals within the democratic political system.

The integration of Muslim immigrants is placed at the heart of European immigration and integration debates, and various studies have examined how natives perceive and evaluate Muslim immigrants and their integration in society, also politically (e.g. Hindriks, Verkuyten, and Coenders 2015; McLaren 2003; Sniderman and Hagendoorn 2007). Yet, relatively few studies have focused on the perception of Muslims themselves, and in relation to political representation and organisation in particular (e.g. Fleischmann and Phalet 2012; Michon and Vermeulen 2013; Verkuyten 2011). In the present research I use existing survey data from three studies to examine the support for democratic political Muslim organisation among Muslim immigrants in the Netherlands (Studies 1, 2, and 3) and Germany (Study 3).

Various sociological studies have examined forms of political organisation and mobilisation among immigrant-origin groups in relation to institutional and discursive opportunity structures in society (e.g. Cinalli and Giugni 2016; Koopmans et al. 2005; Statham and Tillie 2016). Political science research has focused on the identity-to-politics link and it has been argued that research should go beyond investigating the role of demographic ethnic or racial minority labels and consider the underlying processes of group identification, group consciousness, and the choice of political means (see Lee 2008). A focus on these processes is central in the social psychological approach to collective action (see Van Zomeren, Postmes, and Spears 2008). Social psychological research emphasises that internal processes of group identification and perceived group disadvantages are basic requirements for political mobilisation support among minority group members. Taking a social psychological perspective, I therefore look at the role of religious group identification, perceived discrimination, fundamentalist religious belief, and host national identification for the support for democratic political Muslim organisation.

Group identification and politics

In social psychology and in political science, group identification rather than demographic self-labelling is considered an important link between group membership and the support for and engagement in group-based political behaviour (Lee 2008; Van Zomeren, Postmes, and Spears 2008). Identifiers view their group as an important reflection of the self and therefore are motivated to think and act in their group’s best interest (Tajfel and Turner 1979). There is a substantial body of social psychological work in support of this theoretical reasoning (see Ellemers, Spears, and Doosje 1999), including work on political involvement (Huddy and Khatib 2007). Thus, minority members who more strongly identify with their group are more likely to organise politically in order to improve the rights, power, and influence of their group (Becker and Tausch 2015; Van Zomeren, Postmes, and Spears 2008). This means that it can be expected that higher Muslim group identification is associated with stronger support for Muslims organising politically. This expectation is examined in Study 1.

One important aspect of group identification is group consciousness which implies linked fate perceptions (Ashmore, Deaux, and McLaughlin-Volpe 2004). When people identify with a particular community or group, it is the functioning and fate of the group as a whole that counts and not one’s fate as an individual. It is this aspect of group identification that is particularly likely to translate group membership into group-based support and action in the political realm (Dawson 1994; Lee 2008; Miller et al. 1981; Van Zomeren, Postmes, and Spears 2008). The feeling among minority members that their fate is linked makes it more likely that they want to organise politically. This feeling gives a sense of inner obligation to participate in group actions, or at least to support them. Research among ethnic minority members in the US context has shown that higher perception of linked fate is associated with being more likely to participate politically: ‘linked fate, remains the most consistent and powerful cognitive definition of how shared beliefs link demographically defined identities to group politics’ (Lee 2008, 269; e.g. Sanchez 2006; Stokes-Brown 2003). The same process can be expected among Muslim immigrants: those who feel a higher sense of Muslim linked fate can be expected to be more supportive of Muslims organising politically. This expectation will be examined in Studies 2 and 3.

Perceived discrimination

A sense of Muslim linked fate can be based on the perception of shared negative experiences such as social rejection and discrimination. The perception of group discrimination has been found to increase democratic political engagement among ethnic minority groups, including Muslim immigrants in Europe (e.g. Fleischmann, Phalet, and Klein 2011; Just, Sandovici, and Listhaug 2014; Phalet and Güngör 2004; Sanchez 2006; Schildkraut 2005; Valdez 2011). Group discrimination implies unjust disadvantages and unfair treatment of one’s minority group which tells people that they are not equal members of society. Disadvantages and injustices give rise to shared grievances that increase support for group-based action in challenging and changing discriminatory practices (Van Zomeren, Postmes, and Spears 2008). Organising politically is an important form of collective action of immigrant-origin groups, including Muslim immigrants. Thus, in Studies 2 and 3, higher perceived discrimination of Muslims as a group can be expected to be associated with stronger support for political organisation of Muslims.

Part of this expected association might be due to an increased sense of linked fate. Group discrimination presents a threat to in-group interests and identity, causing minorities to increasingly turn towards their minority group (Branscombe, Schmitt, and Harvey 1999; Masuoka 2006). Applied to Muslim immigrants, this represents a form of ‘reactive religiosity’ which implies a positive association between perceived group discrimination and Muslims’ religious group identification (Karlsen and Nazroo 2013). Evidence for this association has been found among Muslim minorities in West European countries including the Netherlands (e.g. Fleischmann and Phalet 2012; Verkuyten and Yildiz 2007), but there is also research that does not support the notion of reactive religiosity (e.g. Maliepaard and Gijsberts 2012). However, this research has not focused on linked fate beliefs and it is likely that discrimination of one’s minority group is more consistently associated with a sense that one’s fate is linked. Thus, the perception of group discrimination may lead to stronger support for Muslims’ political organisation through increased linked fate feelings. In other words, a sense of Muslim linked fate might play a mediating role in the association between perceived group discrimination and support of political organisation of Muslims. Statistically, this implies an indirect effect that will be examined in Studies 2 and 3. 

Fundamentalist belief

It is likely that among Muslim immigrants, a sense of linked fate is not only a form of ‘reactive religiosity’ resulting from perceived group discrimination, but also based on Islamic belief as a form of ‘intrinsic religiosity’. In Islam, there is an emphasis on Muslims forming a single community of believers (‘ummah’). Islam is not just about its five pillars and behavioural rules, but also about the unity of Muslims at local, national, and international levels. A common Islamic community bound by its religion and patterned after the community founded by Muhammad is central to the faith. So in order to examine the origin and role of linked fate beliefs for political organisation support of Muslims, it is important to also consider Islamic belief.

Religious fundamentalism is a term that is commonly, but also variously and loosely used (see Emerson and Hartman 2006; Herriott 2009). One of the most widely accepted definitions is in terms of, ‘The belief that there is one set of religious teachings that clearly contains the fundamental, basic, intrinsic, essential, inerrant truth about humanity and deity’ (Altemeyer and Hunsberger 1992, 118). This definition implies religious norms and ideals that emphasise a single, unchangeable interpretation that is binding for all believers. Importantly, the definition does not include the endorsement or willingness to engage in violent means to achieve fundamentalist aims because these aims can also be pursued in peaceful, democratic ways (Emerson and Hartman 2006; Verkuyten 2011). Fundamentalist believers are strict in their attitude to the text of Qur’an and in following the prescribed religious practices (Almond, Appleby, and Sivan 2003). This makes it important to create and secure conditions where believers can publicly express and practically live by their faith (Just et al. 2014). The implication is that it can be expected that among Muslim immigrants, stronger fundamentalist belief is associated with higher support for Muslims organising politically in order to defend Muslims’ interests and identity in West European societies. I will examine this expectation in the context of the Netherlands (Studies 2 and 3) and Germany (Study 3).

Furthermore, fundamentalist believers consider it important to establish and maintain a unified community of co-believers (i.e. ‘ummah’). This means that it can be expected that fundamentalist belief is associated with a stronger sense of Muslim linked fate that subsequently drives (part of) the support of fundamentalist believers for the democratic political organisation of Muslims. This indirect effect of fundamentalist belief on political organisation through linked fate will be examined in Studies 2 and 3.

Host national identification

In various European countries politicians, the media and commentators have expressed concerns about the assumed lack of national belonging among immigrants, and Muslim immigrants in particular (see Verkuyten 2014). This lack of belonging is assumed to undermine national stability and unity. Research in Western Europe has found that Muslim immigrants do not always feel themselves to be a citizen of the country of settlement (Pew Project 2006). Yet, this does not mean that Muslim immigrants cannot develop a sense of belonging and commitment to the host society or adopt an adversarial stance in which this society is rejected (Reeskens and Wright 2014; Verkuyten and Yildiz 2007). Many immigrants identify not only with their minority group but also with the host society, although often not as strongly (Elkins and Sides 2007; Reeskens and Wright 2014).

For immigrants, a sense of national belonging may diminish the likelihood of defending group-based interests and confronting group-based disadvantages through organising politically (Schildkraut 2005). When immigrants come to see themselves as nationals they might be less concerned with defending interests along religious or ethnic lines. A feeling of national attachment might imply a focus on the interest of society as a whole with a stronger orientation on the normative beliefs and practices of the majority group. For example, it has been found that identifying primarily as American reduces the tendency of Latino and Asian Americans to support ethnic-based politics (Schildkraut 2013). And among Muslim immigrants in the Netherlands, it has been found that host national identification is associated with higher trust in society’s political institutions (De Vroome, Hooghe, and Marien 2013).

Yet, it is also possible that higher national identification is associated with stronger support for group-based political organisation because only as members of society do immigrants feel entitled to societal acceptance of their group-based claims (Simon and Klandermans 2001; Valdez 2011). While minority identity provides the motivation for normative political action, national identification might provide the legitimisation. Empirical evidence for this has been found among Turkish Muslims in Germany (Simon and Ruhs 2008) and in the Netherlands (Klandermans et al. 2008). For the current research, this would mean that stronger national identification is associated with stronger support for the political organisation of Muslims. Thus, there are reasons and empirical findings to expect a positive or rather a negative association between national identification and support for political organisation. Therefore I will explore this relation in all three studies.

The current research

To summarise, my aim is to examine Muslim immigrants’ support for the democratic political organisation of Muslims. I expect higher religious group identification and stronger sense of Muslim linked fate to be associated with stronger support. Furthermore, I expect higher perceived discrimination and more fundamentalist beliefs to be related to stronger support for Muslims organising politically, and that part of these associations can be explained by a stronger sense of linked fate. Additionally, I will explore the importance of host national identification for the level of support. In examining these relations I will control statistically for various demographic variables that might account for the expected relations. For example, age, educational level, length of residence, and citizenship could relate to religious and national identification as well as support for political organisation of Muslims. The data do not contain information on people’s organisational involvement or social networks.

In Study 3, the focus is on Muslim immigrants of Turkish origin and this study also considers possible differences between the Netherlands and Germany. This allows me to examine whether the expected associations are similar or different in both countries. A comparison between the Netherlands and Germany is very useful because in both countries the largest group of Muslims is of Turkish origin which has a similar history of immigration as ‘guest-workers’. Furthermore, Turkish Muslims tend to have strong transnational political ties with Turkey, and the state-related Directorate of Religious Affairs (Diyanet) provides support and services to Turkish Muslims in Europe. This similar migration history and importance of the country of origin could mean that there are clear country similarities in how Muslim immigrants think about political organisations of Muslims. At the same time, Germany, compared to the Netherlands, has a relatively more ‘closed’ opportunity structure for group demands and has been much more reluctant to grant Muslims religious rights (Statham and Tillie 2016). This might mean that there are differences in the ways in which Muslim immigrants in both countries think about the political organisation of Muslims. By looking at the same Muslim immigrant group in both countries, the immigrant group is held ‘constant’ which makes it possible to investigate whether there are country differences.

I will analyse three existing dataset¹ that contain the same measure of support for political organisation of Muslims. In Study 1, data were collected among a national sample that is representative for important demographic characteristics, while Studies 2 and 3 used snowball samples which inevitably implies some selectivity. Yet, by presenting and discussing the findings of three separate studies I try to provide a more generalised picture about Muslim immigrants’ views.

Study 1

Using a national sample, the first study examines the level of support for Muslim political organisation among Muslim immigrants from different ethnic groups: of Turkish and Moroccan origin in particular. People of Turkish origin form the numerically largest immigrant-origin group in the Netherlands and compared to the Moroccans they have a relatively dense organisational network, have taken a more group-based integration approach, and tend to be politically active along ethnic lines (Michon and Vermeulen 2013; Vermeulen, Brünger, and Van de Walle 2010; see also Schönwälder 2013). Their relatively high level of cohesiveness and group-based resources makes it easier to gain access to the political arena and to participate politically. Furthermore, Turkish compared to Moroccan Muslims more often experience religious-based discrimination and there is a strong connection between Turkish group identification and Muslim group identification (Maliepaard and Gijsberts 2012; Phalet and Güngör 2004; Verkuyten and Yildiz 2007). Therefore it can be expected that Muslim immigrants of Turkish origin more strongly support the political organisation of Muslims than those of Moroccan origin. The level of support for political organisation is further examined in relation to religious group identification and host national identification. Statistically, it is possible to control for age, gender, and education because information about these demographic characteristics was available in the survey.

Data and participants

Data were collected using the online panel of TNS/NIPO Consult which is specialised in collecting national population data. Samples were drawn of people for the three largest immigrant-origin groups: of Turkish, Moroccan, and Surinamese background. The questionnaire was in Dutch because most members of the immigrant-origin groups usually speak the language well enough (Huijnk and Dagevos 2012), and the wording of the items used in the questionnaire was rather simple. It is likely, however, that the least integrated were in this way precluded from participating in the survey.

Adult respondents were invited to participate via e-mail and were paid €1.20 in vouchers for completing the questionnaire (or could donate that amount to charity). In total 250 of the 474 respondents indicated Islam as their religion (Moroccans N = 119, Turks N = 119, and Surinamese N = 12). Of these Muslims, 48% was male and 52% female which does not fully correspond with the somewhat higher proportion of males in the Turkish immigrant population in the Netherlands (SCP 2011). The age ranged between 18 and 80 (M = 38.23, SD = 10.61). About 13% of the respondents completed primary education only, 42% completed secondary education, 20% completed lower tertiary education (vocational), 15% obtained a Bachelor’s degree, and 10% obtained a Master’s degree or PhD. This means that the sample is somewhat higher educated than the immigrant population in the Netherlands (SCP 2011).

Measures

The support for Muslim political organisation was assessed with three items (7-point scales): ‘It is important for Muslims that an Islamic political party is established in the Netherlands’, ‘Islam should have its own democratic voice in societal and political issues, similar to other religions’, and ‘Muslims should work more together in order to have more political influence in the Netherlands’. Cronbach’s α for these three items was .85.

Religious group identification was measured with two items (7-point scales) that are commonly used in social psychological research on group identification (see Ashmore, Deaux, and McLaughlin-Volpe 2004): ‘How strong do you identify with your religious community’ and ‘How important is your religion to you’. The correlation between these two items was high (r = .71, p < .001) and the mean score (5.50, SD = 1.44) indicated a relatively strong religious group identification.

National identification was also measured with two items (7-point scales): ‘I identify with the Dutch’ and ‘I feel myself to be Dutch’. The correlation between these two items was high (r = .74, p < .001). The mean score (4.93, SD = 1.55) indicated a level of identification that was significantly above the neutral mid-point of the scale, t(249) = 9.47, p < .001.

Findings

The mean score for the support of Muslim political organisation (M = 4.55, SD = 1.56) was significantly above the neutral mid-point of the scale, t(249) = 5.59, p < .001. This indicates that there was support for organising politically. The support was significantly correlated with religious identification (r = .40, p < .001) but not with host national identification (r = .03, p > .60). Religious and national identification were not related to each other (r = −.03, p > .30).

The findings of a multiple regression analysis indicate that age (in years), gender (1 = female, 0 = male), education (7-point scale from no education to university degree), and host national identification are not independently associated with the support for Muslim political organisation (see Table 1). However, the support is lower among Moroccan compared to Turkish participants, and religious group identification is a relatively strong predictor. Higher religious identifiers are more in favour of Muslims organising politically in the Netherlands.

Table 1. Predicting support for Muslim politics, Study 1.

	β	SE
Female	−.08	.18
Age	−.02	.01
Education	−.03	.05
Moroccan (ref. is Turkish)	−.18**	.19
Surinamese (ref. is Turkish)	.04	.43
Religious identification	.36***	.07
National identification	.04	.06
R^2	.205
F-value	8.873***

The findings of Study 1 lead to three conclusions. First, religious group identification is a clear predictor of the support for the political organisation of Muslims. Second, host national identification appears not to stimulate, but also not to diminish, the level of support. Third, political organisation of Muslims in the Netherlands is more strongly endorsed among Muslim immigrants originating from Turkey than from Morocco. This corresponds with research showing that Turkish compared to Moroccan immigrants have a relatively dense organisational network, take a more group-based integration and political approach (Michon and Vermeulen 2013; Vermeulen, Brünger, and Van de Walle 2010), and more often experience religious-based discrimination (Maliepaard and Gijsberts 2012).

Study 2

The second study builds upon and adds to the first one by focusing on linked fate as a dimension of group identification that is particularly relevant for explaining group-based organisation and action (Dawson 1994; Lee 2008; Van Zomeren, Postmes, and Spears 2008). In addition, the roles of perceived group discrimination and fundamentalist belief are examined. Both these construct are expected to be associated with support for the political organisation of Muslims, and part of these associations is expected to run through linked fate. In Studies 2 and 3, data were only collected among Muslims of Turkish origin. People of Turkish origin form the numerically largest immigrant-origin group in the Netherlands and are best organised and therefore potentially the politically most influential one. Furthermore, Turkish Muslims experience most often religious-based discrimination, and Turkish and Muslim group identification have been found to be strongly associated (Maliepaard and Gijsberts 2012; Verkuyten and Yildiz 2007). Additionally, in Study 2 questions about length of residence and citizenship were asked and therefore could be controlled for in the statistical analysis.

Data and participants

Data were collected among a sample of 217 self-identified Turkish Muslims living predominantly in and round the cities of Utrecht, Arnhem, Deventer, and Apeldoorn. The data collection took place with two interviewees of Turkish origin reaching the first participants through their own contacts and ensuring that these participants introduced them further to their personal network. The questionnaire was again in Dutch. The participants were between 16 and 69 years of age (M = 29.97; SD = 11.52) and their length of residence in the Netherlands varied between 1 and 45 years (M = 20.19, SD = 7.68). Age and length of residence were positively associated but not very strongly (r = .38, p < .001). In terms of gender distribution, 61% of the participants was male and this gender difference corresponds with the Turkish immigrant population in the Netherlands (SCP 2011). Regarding education, all levels were represented in the data, with 15% of participants having no formal degree or only a primary school diploma, while 38% was in a possession of different types of secondary school diploma, and 47% held a tertiary degree. This means that also this sample is somewhat higher educated than the Turkish immigrant population in the Netherlands (SCP 2011). In total, 37% of the participants had only Turkish citizenship and 63% had only Dutch or dual citizenship which is similar for the Turkish immigrant population in the Netherlands (SCP 2011).

Measures

The support for Muslim political organisation was assessed with the same three items (7-point scales) that were used in Study 1. Cronbach’s α was .90.

Perceived Muslim linked fate was measured with four items (7-point scales): ‘What happens with other Muslims determines my own life and future’, ‘I feel strongly connected to the ummah’, ‘my own fate is linked to that of other Muslims’, and ‘Muslims are dependent on each other and should be strong together’. Alpha for this scale was .92.

Perceived discrimination of Muslims was measured with four items (7-point scales). Two sample items are, ‘Muslims in the Netherlands are often being discriminated’ and ‘In general, the native Dutch are negative about Islam’. Alpha was .83.

Fundamentalist belief was measured with four items (7-point scales); ‘The things I believe in are so true that I would never doubt them’, ‘My beliefs give a definite answer about the questions of life’, ‘I have never found a belief that for me convincingly explains everything (reverse)’, and ‘There are no discoveries or new things that could change my beliefs’. Cronbach’s α = .80.

National identification was measured with five items (7-point scales: e.g. ‘I identify with the Dutch’ and ‘I feel myself to be Dutch’) and Cronbach’s α = . 92.

Analysis

Because of a few missings on the common fate (mediator) and endorsement of political organisation (outcome) measures, the analysis was conducted with a sample of 198 participants. In a stepwise multiple regression analysis, a number of socio-demographic characteristics was controlled for: gender, education, age and length of residence (both in seven categories of 5–10 years), and citizenship (1 = Dutch/dual, 0 = Turkish).

Findings

The mean score for the endorsement of Muslim political organisation was very similar to Study 1 (see Table 2), and again significantly above the neutral mid-point of the scale, t(198) = 4.24, p < .001. This indicates that also among this sample of Muslim participants, there was support for organising politically. This support was significantly correlated with perceived group discrimination, fundamentalism, linked fate, and host national identification (Table 2). National identification was weakly associated with fundamentalism and with linked fate (highest shared variance = 3.6%), and not with perceived discrimination.

Table 2. Correlations, means, and SDs for the measured variables, Study 2.

	1	2	3	4	M	SD
1. Fundamentalism	–				3.81	1.26
2. Discrimination	−.06	–			4.50	1.53
3. Linked fate	.24***	.14*	–		4.03	2.04
4. National identification	−.19**	.05	−.17*	–	3.60	1.64
5. Muslim politics	.30***	.19*	.56***	−.17*	4.57	1.96

A stepwise multiple regression analysis was conducted with the demographic variables entered in Step 1, and group discrimination, fundamentalism, and national identification entered in Step 2. In Step 3, linked fate was added to the regression equation. As shown in Table 3, in the first step there was a significant effect for age: older Turkish Muslims were less in favour of organising politically than younger Muslims. In Step 2, higher fundamentalism and higher perceived discrimination were both associated with stronger support of Muslim political organisation. Similar to Study 1, national identification was not an independent predictor. In the third step, linked fate was a relatively strong predictor of political organisation. In addition, fundamentalism and discrimination were more weakly associated with political organisation in Step 3, compared to Step 2. This suggests that part of the effects of fundamentalism and discrimination runs through Muslim linked fate.

Table 3. Predicting support for Muslim politics, Study 2.

	Step 1		Step 2		Step3
	β	SE	β	SE	β	SE
Female	−.17*	.29	−.16*	.27	−.09	.24
Age	−.22**	.08	−.22**	.07	−.13	.06
Length of residence	.02	.10	.04	.09	.04	.08
Education	−.12	.08	−.08	.07	−.04	.07
Citizenship (ref. Turkish)	.11	.31	.06	.29	.05	.26
Fundamentalism			.28***	.10	.18**	.09
Discrimination			.23***	.09	.16**	.08
National identification			−.13	.08	−.08	.07
Linked fate					.44***	.06
R^2	.08		.22		.39
F-change	3.32**		12.00***		49.41***

I used the Preacher and Hayes’ (2008) bootstrapping macro with 1000 iterations for testing the indirect effects of fundamentalism and perceived discrimination on the support for Muslim political organisation through linked fate (controlling for the unique effects of the demographic variables and national identification). In these analyses, an indirect effect is significant if the 95% Bias Corrected Confidence Interval for this indirect (mediated) effect does not include zero. Results showed that the indirect effect (.149) of fundamentalism through linked fate on political organisation was positive and reliably different from zero (lower CI = .0611, higher CI = .255). The indirect effect of discrimination (.138) was also significant (lower CI = .025, higher CI = .268). These findings provide evidence for mediation, whereby stronger fundamentalist belief and higher perceived discrimination strengthen the perception of Muslim linked fate which subsequently increases the support of the political organisation of Muslims in the Netherlands.

The findings of the second study lead to four conclusions. First, a sense of linked fate is a strong predictor of the support for the political organisation of Muslims. Second, higher perceived discrimination of Muslims in society and stronger fundamentalist religious belief are independently associated with stronger support for political organisation. Third, linked fate, in part, mediates the statistical effects of perceived discrimination (‘reactive religiosity’ path) and fundamentalist beliefs (‘intrinsic religiosity’ path) on support. Thus, higher perceived group discrimination and stronger fundamentalist belief are both associated with a stronger sense of linked fate and thereby to higher support for the political organisation of Muslims. Fourth, similar to Study 1, host national identification was not independently associated with support for Muslim political organisation and therefore does not seem to stimulate or diminish the level of support.

Study 3

The third study adds to the previous one by investigating whether the same associations exist among a different sample and by using somewhat different measures of perceived group discrimination and fundamentalist belief. Furthermore, this study examined whether the support for Muslim political organisation is similar in Germany and the Netherlands, and also whether the associations with the different predictor variables are similar in the two countries.

Data and participants

The data used for Study 3 was collected as part of a larger survey among Turkish Muslim minorities in the Netherlands and Germany that covered a range of topics. For the present study, we focus on 464 self-identified Sunni Muslims (N = 260 in the Netherlands, and 204 in Germany).² To ensure comparability, the snowball data collection procedure was the same in the Netherlands and Germany and it took place simultaneously, with four interviewees of Turkish origin reaching the first participants through their own contacts and ensuring that these participants introduced them further to their personal network. The majority of the participants was from the middle and eastern parts of the Netherlands, and from the western part of Germany. The questionnaire was offered only in the official language (Dutch or German).

Participants older than 16 were recruited for the study, and the average age of the sample was 36.54 (SD = 12.95; range 16–84). The mean length of residence in the host country was 22.13 (SD = 9.99), and length of residence was positively associated with age (r = .49, p < .001). In terms of gender distribution, 57% of the participants were males which corresponds with the higher proportion of males in the Turkish immigrant population. In total, 54% had only Turkish citizenship and 46% had host national or dual citizenship. Regarding education, all levels were represented in the data, with 31% of participants having no formal degree or only a primary school diploma, while 40% was in a possession of different types of secondary school diploma, and 29% held a tertiary degree.

Measures

The support for Muslim political organisation was assessed with two of the three items (7-point scales) that were used in Studies 1 and 2. The two items were associated with r = .60 (p < .001).

Perceived linked Muslim-fate was measured with three of the four items (7-point scales) used in Study 2. Alpha for these three items was .75.

Perceived discrimination of Muslims was measured with four items (7-point scales) that ask about the discrimination of Muslims in finding a job, in nightlife, and in everyday life. Alpha was .84.

Fundamentalist belief was measured with two items that explicitly asked about religious belief (7-point scales): ‘There is only one truth and that is the divine truth’ and ‘Islam has a fixed and unchangeable meaning’. The correlation between these two items was .60 (p < .001).

National identification was measured with the same five items as used in Study 2. Cronbach’s α = . 86.

Findings

The mean score for support of Muslim political organisation, again, was significantly above the neutral mid-point of the scale, t(461) = 3.99, p < .001, indicating that also among this sample there was support for Muslim political organisation. This support was significantly correlated with perceived group discrimination, fundamentalism, linked fate and (negatively) with host national identification (Table 4). National identification was weakly associated with fundamentalism and linked fate (highest shared variance = 4.1%), and not significantly with perceived discrimination.

Table 4. Correlations, means, and SDs for the measured variables, Study 3.

	1	2	3	4	M	SD
1. Fundamentalism	–				4.90	1.65
2. Discrimination	.47***	–			4.55	1.46
3. Linked fate	.57***	−.47***	–		4.18	1.64
4. National identification	−.21**	−.10	−.11*	–	3.65	1.43
5. Muslim politics	.51***	.48***	.58***	−.25**	4.30	1.62

A stepwise multiple regression analysis was conducted, similar to Study 2 (see Table 5). In the first step, none of the demographic variables was significantly associated with support for Muslim political organisation. In Step 2, higher fundamentalism and higher perceived discrimination were independently associated with stronger support. In addition, higher host national identification was associated with weaker support for political organisation. In the third step, linked fate was a relatively strong predictor and fundamentalism and discrimination were more weakly associated with political organisation in Step 3 compared to Step 2. This again suggests that part of the effects of fundamentalism and discrimination runs through linked fate.

Table 5. Predicting support for Muslim politics, Study 3.

	Step 1		Step 2		Step3
	β	SE	β	SE	β	SE
Female	.01	.16	.01	.13	.01	.12
Age	.05	.04	.03	.04	.01	.03
Length of residence	.03	.05	−.02	.04	−.03	.04
Education	.03	.04	−.05	.03	−.05	.03
Citizenship (ref. Turkish)	.03	.18	−.06	.14	−.07	.13
Country (ref. Netherlands)	.06	.17	−.16**	.14	−.06	.14
Fundamentalism			.35***	.04	.17**	.05
Discrimination			.35***	.05	.25**	.05
National identification			−.17**	.05	−.15	.05
Linked fate					.35***	.05
R^2	.01		.38		.44
F-change	1.75		85.70***		51.93***

Similar to Study 2, Preacher and Hayes’ (2008) bootstrapping macro with 1000 iterations was used for testing the indirect effects of fundamentalism and perceived discrimination on the endorsement of Muslim political organisation through linked Muslim-fate (controlling for the unique effects of the demographic variables and national identification). Results showed that the indirect effect (.175) of fundamentalism through linked fate on political organisation was positive and reliably different from zero (lower CI = .118, higher CI = .239). The indirect effect of discrimination (.117) was also significant (lower CI = .069, higher CI = .174). These findings are quite similar to Study 2 and provide further evidence for linked Muslim-fate mediating the association between fundamentalist belief and perceived group discrimination, on the one hand, and the support for the political organisation of Muslims, on the other hand.

Country differences

The results of the regression analysis (Table 5) indicate that support for the political organisation of Muslims did not differ between Germany and the Netherlands. In an additional analysis, I examined whether the measured predictor variables were differently associated with support in both countries. In a regression analysis with the four interaction terms (country by perceived discrimination, by fundamentalist belief, by linked fate, and by national identification; centred scores), there was only a significant interaction effect between country and fundamentalist belief. Further analyses showed that stronger fundamentalist belief was related to more support for political organisation in Germany (β = .29, SE = .07, p < .001). In the Netherlands, the association was also positive but not significant (β = .11, SE = .06, p > .10).

The findings of the third study are similar to, and thereby replicate, the results of the second one. Again, it turned out that a sense of Muslim linked fate was a strong predictor of the support for the political organisation of Muslims. Furthermore, higher perceived group discrimination of Muslims in society and stronger fundamentalist religious belief were independently associated with support. In addition, linked fate, in part, mediated these associations. Additionally, there was a tendency for stronger host national identification to be associated with weaker support, but this association was no longer significant when linked fate was taken into account. Finally, Turkish Muslim immigrants in Germany and the Netherlands were very similar in their level of support for the political organisation of Muslims and in terms of the variables that predicted this support.

Discussion

The substantial gap in immigrant political participation and representation across West European countries raises important questions of democratic inclusion and political legitimacy (Bloemraad 2013). These questions are particularly relevant and urgent in relation to Muslims because the integration of Muslim immigrants is central in West European debates. Whereas various studies have examined how the native majority evaluates the integration and political participation of Muslims (e.g. Hindriks, Verkuyten, and Coenders 2015; Sniderman and Hagendoorn 2007), relatively few studies have examined the perception of Muslims themselves (e.g. Fleischmann and Phalet 2012; Michon and Vermeulen 2013). In addition to descriptive representation in mainstream political organisations and parties, Muslim immigrants can also organise politically along religious lines. In different European countries, there have been political organisations and parties that present themselves as Islamic or draw inspiration from Islam, and some of these parties have secured seats in municipality councils. This form of organisation is similar to small Christian-orthodox parties and one-issue parties (e.g. party of the elderly) that represent specific sections of the population and try to achieve their particular goals within the democratic political system.

Using a social psychological perspective, this article examined whether Muslim immigrants support the democratic political organisation of Muslims and how religious and host national identification, perceived discrimination and fundamentalist beliefs play a role in this. The findings of three survey studies are quite similar and indicate support for Muslim political organisation in the Netherlands and Germany. Furthermore, the level of support did not depend on demographic characteristics such as gender, age (except Study 1), length of residence, citizenship, and educational level, and did also not differ between the Netherlands and Germany. This suggests that the support for the political organisation of Muslims is not specific or limited to particular sections of the Muslim immigrant population. Yet, Muslims originating from Turkey were more supportive of political organisation than those originating from Morocco (Study 1). Furthermore, there were clear differences related to Muslim group identification and linked fate, perceived discrimination and fundamentalist religious belief.

In Study 1, it was found that stronger Muslim group identification was associated with stronger support for the political organisation of Muslims. This is in agreement with the social psychological and political science argument that identification – rather than demographic self-labelling – is an important psychological link between group membership and support for group-based action (Lee 2008; Van Zomeren, Postmes, and Spears 2008). Stronger identifiers are more likely to think and act in their group’s interest. They are also more likely to want to express and live by their religious identity and therefore find it more important to secure the societal conditions that enable them to do so.

Studies 2 and 3 went beyond mere group identification by focusing on the sense of linked fate that more easily can develop into a politicised identity (Dawson 1994; Lee 2008; Van Zomeren, Postmes, and Spears 2008). When people identify with a particular community or group, it is the functioning and fate of the group as a whole that counts and not one’s fate as an individual (Ashmore, Deaux, and McLaughlin-Volpe 2004). It was found that a stronger sense of Muslim linked fate was indeed associated with stronger support of political organisation. This is similar to what is found in research among ethnic minority groups in the USA (Dawson 1994; Sanchez 2006; Stokes 2003). The sense of sharing one’s fate and being dependent on each other provides the motivation to support group-based actions.

I additionally examined whether perceived discrimination and fundamentalist religious beliefs underlie a sense of linked fate. Perceptions of discrimination can lead to a form of ‘reactive religiosity’ whereby the feeling of one’s group being excluded or treated unfairly strengthens the sense of linked fate. In addition to this ‘reactivity’ path there might also be an ‘intrinsic religiosity’ path in which the sense of Muslim linked fate is grounded in Islamic faith. This means that people who more strongly endorse the religious teachings that are considered to contain the basic and intrinsic truths about the social and metaphysical world (Altemeyer and Hunsberger 1992) can be expected to have a stronger sense that their fate is linked to and dependent on the Muslim community. There was empirical evidence for both paths in Studies 2 and 3. Perceived discrimination and fundamentalist belief were associated with Muslim linked fate and through linked fate with support for the political organisation of Muslims. This suggests that sense of linked fate can play an important role in translating perceived group discrimination and fundamentalist beliefs into support of Muslim political organisation. However, only partial mediation was found which indicates that this sense does not fully explain the associations. Discrimination and fundamentalist beliefs were also directly and independently associated with support. This shows that both paths are important for understanding why Muslim immigrants support the political organisation of Muslims. This support does not only depend on feelings of linked fate but can also more directly result from perception of discrimination of Muslims in society and beliefs in fundamental religious teachings.

There was no systematic evidence for host national identification being related to support for the political organisation of Muslims. Furthermore, national identification was not associated with religious group identification (Study 1) and only weakly with Muslim linked fate and fundamentalist beliefs (Studies 2 and 3). These are interesting findings in light of the continuing concerns and debates about the importance, and assumed lack, of host national belonging among immigrants, and Muslim immigrants in particular. The findings indicate that there is not a clear and strong contradiction between Muslim identity and host national belonging. Furthermore, national identification does not appear to stimulate nor diminish Muslim immigrants’ support for the political organisation of Muslims.

However, the findings should be interpreted with care. Survey research cannot establish causality and therefore it is possible that there are different forms of influence between the variables. For example, a sense of linked fate can lead to perceptions of group-based discrimination and might also stimulate more fundamentalist beliefs. And while it is less likely that support for political organisation, compared to actual participation, stimulates a sense of linked fate or perceptions of group-based discrimination, this is not impossible. Hence, it is likely that there are mutual influences and what the current findings then show are important directions of influence on the support of political organisation of Muslim immigrants in the Netherlands and Germany. Longitudinal research is needed to further examine the various possible influences.

Another limitation of the research is that only in Study 1, a national sample was used and snowball samples in the other two studies. This limits the possibility to draw more generalisable conclusions. Some form of selectivity is likely, also because the questions were asked in the host language. Although the great majority of Turkish immigrants is able to speak and read Dutch, some subgroups (first generation, older women) are not (Huijnk and Dagevos 2012). This selectivity probably implies that we did not reach the most segregated Turkish Muslims. This can have implications for the mean levels of the predictor variables and the support for political organisation of Muslims, but not necessarily for the associations between the measures. Furthermore, the similar findings for the three different samples suggest that I have presented a more general picture about Muslim immigrants of Turkish origin in these two countries. In addition, despite the fact that somewhat different measures were used across the studies (for perceived discrimination and fundamentalist belief) the results are similar which further supports the robustness of the findings. Also, the similar findings in the Netherlands and Germany (Study 3) indicate that more general social psychological processes are involved and these are not affected by country differences in opportunity structure and the incorporation of Muslims (Statham and Tillie 2016).

Thus, there are reasons to argue that this research has identified some important and more general social psychological processes involved in the support for the political organisation of Muslims. Yet, future studies should improve the sample to investigate this further. These studies could also examine the role of, for example, social networks and organisational involvement for processes of group identification, perceived disadvantages, religious belief and support for Muslims organising politically. This could further improve our understanding of the ways in which social psychological processes involved in the support for Muslim political organisation are shaped by contextual circumstances. Additionally, the support for Muslim political organisation could be examined in relation to political orientation and identification with existing political parties. This would allow, for example, to assess the preference for Muslim political organisation in comparison to descriptive representation of Muslims in existing political parties (Bloemraad 2013; Schildkraut 2013).

The current study focused on Muslims’ support for political organisation and not on actual mobilisation and behaviour. Understanding the level of support is important because it indicates the potential for Muslim political organisation, and support tends to be associated with actual behaviour (Van Zomeren, Postmes, and Spears 2008). Yet, it remains to be seen whether the current results hold for behavioural involvement. For example, the fact that there was no country difference in support does not have to mean that there is no behavioural difference between Germany and the Netherlands. Germany has a different political system and has been much more reluctant to grant Muslims religious rights compared to the Netherlands. Different political contexts and different diversity and integration policies offer different opportunity structures for the political participation and group-based action of immigrant-origin groups (e.g. Buckley 2013; Koopmans et al. 2005). Future studies should examine the democratic behavioural involvement of Muslim immigrants in different national contexts.

In conclusion, immigrant-origin political organisation and representation has important symbolic and normative implications for the political incorporation of minority groups and the legitimacy of the democratic system. In Western Europe, there are strong concerns about the integration of Muslim immigrants and their alleged lack of commitment to the host society. The current study shows that Muslim immigrants in the Netherlands and Germany support religiously oriented democratic organisation of Muslims. This support is especially likely among Muslims who identify with their religious group and who have a sense of Muslim linked fate. In turn, feelings of linked fate can be generated by the perception that Muslims are discriminated in society and by fundamentalist religious beliefs. Discrimination can lead to a form of ‘reactive religiosity’ with the related group solidarity that is translated into support for group-based politics. In addition, the ‘intrinsic religiosity’ of fundamentalist beliefs can stimulate a sense of Muslim interdependence which makes it more likely that Muslim immigrants support the democratic political organisation of Muslims.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. The three dataset were collected by the European Research Centre on Migration and Ethnic Relations at Utrecht University and are safely stored at the faculty data storage. Part of these data have been used in previous research papers but the current research question, focus, and analysis are novel. Thus, the findings have not been reported elsewhere and the paper does not duplicate previously published work.

2. This dataset also contained Turkish Alevis. However, I did not include their responses in the analysis because not all of them consider Alevism as having a religious meaning or being within Islam. Furthermore, Sunni and Alevi Muslims have a very different understanding of what it means to be a Muslim, and they follow different religious practices.