ECPR Joint Sessions of Workshops 2015, Warsaw Paper prepared for the presentation at the workshop: The Civic Turn in European Immigrant Policies

Transcription

1 Dr. Katarzyna Lasinska University of Mannheim ECPR Joint Sessions of Workshops 2015, Warsaw Paper prepared for the presentation at the workshop: The Civic Turn in European Immigrant Policies DRAFT ONLY PLEASE DO NOT QUOTE Social Participation of Immigrants. Does Culture Matter? 1 Introduction Migration to western European countries increased in recent decades from three per cent in 60 to ten per cent nowadays and this dynamic growth becomes an important political and societal challenge for recipient countries. One of the key questions related to the growing migration rates is whether social cohesion can be stimulated by immigration policies and how. An important aspect of social integration of immigrants is social participation, which favours social cohesion through inclusion in social networks. Because immigrants are in principle excluded from elections processes (in some countries they are allowed to vote in communal or local elections), social participation can be seen as a kind of substitute for political exclusion and an important mean to facilitate immigrants integration in political systems. Therefore social participation is in case of immigrants even more important factor of social integration as of native citizens. Furthermore a body of research shows that social engagement related to own ethnic community also stimulates political engagement and the subjective sense of societal inclusion (Berger at al. 2004, Togeby 2004). While many studies of immigrants participation focus on the context features such as legal settings, social structures or political opportunities of the context they live in, we propose to look rather on the normative background of the motivation to join social networks that have its roots in the culture of origin. In this paper we want to address the question to which extend the culture of country of origin matters for social participation of immigrants? The paper is composed of three parts. The first part focuses on theoretical considerations how the country of origin could influence the participatory behaviour of immigrants. In particular it discusses on the one hand the role of cultural socialization for propensity to join social networks, and on the other hand the relevance of the political socialization. Additionally it considers the possible influence of religiosity, too. The second section demonstrates the results of the empirical cross-national examination of the newest European Social Survey data (2012, 6. Round). The third part concludes that the cultural imprint is relevant for explaining participation patterns of immigrants, while the political socialisation 1

2 matters, too but mostly when referring to communist regimes. Furthermore, individual religiosity is a mobilizing factor for participation, but only for those who are actively connected to church communities. 2. Social participation of immigrants why the culture of origin could matter? Whereas there is a wide agreement among researchers concerning low rates of participation of immigrants in voluntary organisations when compared to non-migrants (van Londen et al. 2007), the reasons for these discrepancies are far less clear. When searching for the determinants of immigrants participatory behaviours scholars focus mostly on two kinds of factors: individual-level determinants and context characteristics (Morales, Guigni 2011, 5). Analyzing individual-level factors apart from determinants usually explaining participation such as socio-economic status, family status, gender, political interest scholars also consider individual degree of attachment to the recipient country such as the length of residence or recipient-country language proficiency (Jacobs et al. 2004). Second strand of research addresses the question concerning local (and national) contexts characteristics influencing immigrants inclination to participate. A whole range of context-level variables is considered in this group of studies from the concentration of ethnic groups at the community level, through collective mobilization and political opportunities at the local (also national) levels, to a number of organizational characteristics: the structure (quality, density) of social networks, the supply of ethic-concentrated groups and cross-ethnic groups, the issues the organisations address etc (Morales, Guigni 2011, 262ff). Given the focus on the context characteristics, comparative studies of immigrants are mostly limited to a small number of local communities (or cities). Most of the studies compare several ethnic groups within the same local context, or compare the same ethnic group in several different contexts. Therefore a generalization of the results is somewhat problematic. This paper applies a broader perspective of the problem of immigrants participation and addresses a question that seems to be neglected to a large extent in the research: the relevance of cultural and political legacy of origin country. As many of the case studies on the local levels, there is one striking similarity among them quite often ethnic groups seem to display group-specific determinants of participation, even in the same local and national contexts. For instance, when comparing participation of Turks and Italians in Germany an empirical study shows that gender is an important determinant of social participation of Turks while irrelevant for Italian immigrants (Glatzer 2004, 83). Given this evidence, one should ask whether there is an underlying factor explaining participation of immigrants related to the culture of origin country. In other words, can the culture of country of origin be an explanatory factor for different determinants of participation across different ethnic groups of immigrants, irrespective from the contexts of recipient countries? Why could the culture of country of origin matter for participation of immigrants? The reason is a different socialization: people migrating from one country to another are socialized to different cultures than the one of recipient country, i.e. have internalised different values systems and norms of behaviours. Therefore the first issue they need to deal with before they can be actively involved in social undertakings, is to understand the normative background of social life in recipient society. The organizational and 2

3 structural aspects of voluntary associations reflect to a large extend the culture of a society and this might be quite different from the way voluntary organizations work in the country of origin. A broad study of volunteering patterns pursued by Verba and colleagues provide us with the theoretical background for our research. In their Civic Voluntarism Model Verba and colleagues point to three reasons why people are reluctant to volunteer: because they can't; because they don't want to; or because nobody asked (1995, 269). To formulate it positively recruitment, resources and engagement are the key explanations for the inclination to volunteer, and this theory is applicable to immigrants, too. While the first two factors recruitment and resources are related on the one hand to the structures and context characteristics (intensively explored in other studies, as mentioned before), and on the other hand to individual socio-economic status and skills; the engagement refers rather to the motivation resulting from values and norms. And the wider the gap between norms of social behaviour one is socialized to and the norms widely accepted in society one lives, the more challenging it is to join actively voluntary organisations. Now, we need to address the question what kind of values orientations are of relevance when analyzing patterns of participation. As many studies of cultural diversity, one promising approach offers the study of Inglehart and Baker (2000) conceptualising cultural differences along traditional and secular-rational values orientations. This categorization from traditional to secular-rational values orientations is suitable for the analysis of cultural cleavage possibly explaining reluctance to volunteer among immigrants coming to Western Europe for several reasons. First, the usual direction of migration in general is from traditional oriented societies to secularized-rationalised societies. Most of recipient countries in Western Europe rank high on rational-secular values orientations (though English-speaking and Catholic societies ranking lower than Scandinavian and German-speaking countries), in contrast many of the immigrant countries of origin rank high on the traditional (and low on rational-secular) values orientations (for instance African and most of Latin American societies) (Inglehart, Baker 2000, 29). One of the key reasons for migration is a hope for prosperity and better life, therefore this is a plausible direction of migration from less developed countries to more developed ones, what means with high probability migration from traditional societies to more secularized and rationalised ones. Secondly, secular-rational values orientations are prone to participatory behaviour. A shift from traditional to secular-rational values orientations is related to modernization processes that societies undergo due to increase in economic prosperity (Inglehart, Baker 2000). And with modernisation of societies a more participatory citizenship appears and people s empowerment rises, too (Inglehart, Baker 2000, 30). In consequence, societies ranking higher on secularrational values orientations tend to have higher levels of participation than traditional societies. Thirdly, since we can rank societies on the scale from traditional to secular-rational values orientations, we can also make a gradual distinction of the origin countries according to degree how much different they are from the culture of recipient country. And we assume the more similar the culture of origin to the recipient country culture, the less alienated is the ethic group. 3

4 By taking some examples of cultural gaps between societies on traditional and secular-rational dimensions, we can make our point more clear. For instance, in traditional societies role of women is definitely different than in secular-rational oriented societies. Traditional societies e.g. Arab societies have patriarchal social structures, and institutions of patriarchy determine the social order (Joseph 2005, 195). Woman is protected by her male relatives that have at the same time authority over her (Joseph 2005, 195). Therefore in patriarchal societies female involvement in social undertakings is not supported, differently from the situation in Western Europe where it is not questioned. We can expect that women migrating from a traditional patriarchal culture are more reluctant to engage themselves in social undertakings, because they are socialized to a different social order. Another difference between traditional and secular - rational societies is the centrality of family in the former. The family is the core institution in the traditional systems giving support for individuals, and being the source of their identity. Family members are expected to view the good of the family above the fulfillment of individual wishes and selfsatisfaction (Abudabeh 2005, 427). So kinship relations are key force to create social networks, what is different from the kind of social networks built upon common goals and interests independent from the family ties as it is the case in secular-rational Western societies. Therefore people socialized to traditional cultures with kinship ties as the key integrating factor would be reluctant to join social voluntary associations organized along social norms accepted in Western societies. Another factor crucial to explain the reluctance of immigrants to engage themselves in voluntary work might be political socialization. A study on relationship between democracy and participation provides empirical evidence for mutual reinforcing effects between both (Paxton 2002). Furthermore as widely known participation in post-authoritarian regimes in Latin America is lower than in established Western democracies (Howard ff.). And similar to participation social trust in these societies is lower, too, as effects of massive poverty, inequality and widespread corruption (Diamond 2008, 188). Therefore people socialized to authoritarian regimes in Latin America would be used to trust less and to participate to a lesser degree when coming to Western Europe. Another example is the Central Eastern Europe. Having experienced socialization to communist regimes, which were especially hostile to free engagement of people outside the institutions of political control, immigrants from this region would be also probably more reluctant to participate, simply because they are not used to: a country s prior communist experience has very strong negative impact on the organizational membership of its citizens today (Howard 2003, 90). And in general social participation across post-communist societies is even lower than participation in post-authoritarian regimes (Howard 2003, 80ff.), therefore one could expect communist legacy the most unfavourable political context for participation. Another aspect of socialization which is relevant for explaining participation is religion (Putnam 2000, 69). As it is widely known, social participation and social trust, are higher in Protestant societies when compared to Catholic societies (Curtis et al. 2001, 791, Gabriel et al. 2002, 52). Further studies show also that Orthodox religious tradition as well as Islam are also negatively correlated with participation (Inglehart, Norris 2004, 189) Therefore we could also expect that immigrants socialized to one of these 4

5 religious traditions would be less likely to participate than Protestant immigrants. Especially immigrants that moved to majority-protestant country should be less inclined to get involved in voluntary organization, simply because they are socialized in another religious tradition. Finally, let us systematize the theoretical claims. In general, three aspects of socializations can to be important to explain the patterns of immigrants participation. First, we deal with the cultural socialization, meaning the culture of country of origin consisting in specific social structures and dominating worldview, and we claim that particularly traditional orientations, in contrast to secular-rational values orientations, could an obstacle for participation. Since European recipient societies are mostly imprinted by secularrational values orientations, we expect that the wider cultural gap between recipient and homeland country, the more is an immigrant reluctant to participate in voluntary associations. For instance, we can expect that immigrants originating from another Western European country to be more inclined to volunteer then individuals coming from e.g. traditional oriented society, where social structures and traditional worldviews are more distant from the culture of any European recipient country. Similarly, we could expect that people socialized in Latin America, which culture is over years dominated by Christian European rulers would be more likely to participate when compared to people from South East Asian or African societies. Second aspect is political socialization, and we claim that socialisation in a non-democratic political context results in lower inclination for participation, whereby ex-communist regimes are probably the most unfavourable political context, even if compared to other authoritarian (and post-authoritarian) regimes (Howard 2003). And the third aspect of socialization is religion, whereby Protestants seem to be the more participative, in contrast to Catholics, Orthodox adherents and Muslims. 3. Social participation of immigrants empirical analysis of the influence of country of origin To check whether our claims concerning the ways how culture of origin country can possibly influence participation patterns of immigrants, we pursue the empirical analyses. The examinations are based on survey data of the sixth round (2012) of ESS. We examine social participation of immigrants across eleven Western European societies with the highest immigration rates in Europe: Belgium, France, Germany, Great Britain, Ireland, Italy, the Netherlands, Norway, Spain, Switzerland and Sweden. As immigrants we define the individuals who have no citizenship or who are not born in the resident country. The rates of immigrants as defined in this way range from 25.3% in Switzerland to 7.29% in Italy (Table 1). 5

6 Table 1 Per cent of immigrants in Western European countries per cent of population not born in a country per cent of population which are not citizens per cent of immigrants Belgium France German Great Britain Ireland Italy The Netherlands Norway Spain Switzerland Sweden Note: source ESS 2012, 6.Round. The participation of immigrants is analysed in two steps. First we examine the patterns of participation of immigrants in reference to the whole population, in order to answer the questions whether immigrants are indeed less socially active than native citizens and whether the expected influences of the culture of origin can explain their reluctant behaviour. In the second step we pursue a more rigorous examination of immigrants participation by excluding non-immigrants from the analyses and therefore by concentrating the analyses uniquely to those who fall into category immigrant. In this way we check whether the expected influences of culture of origin country can explain the different participation patterns across different groups of immigrants, as one can conclude from the evidence showed by numerous case studies (mentioned above). In the both steps of analyses we consider threefold way how the culture of origin country can impact readiness for participation: (1) through a broadly understood cultural imprint (cultural socialization), (2) through the socialization in political regime of the country of origin (political socialization) and (3) through religious tradition one adheres to (religious socialization). Social participation is operationalised by a question: How often were you involved in work for a voluntary or a charitable organization in the last 12 month? Respondent can choose the answers on the 5-point scale from 1- at least once a week until 5- never (whereby a reversed scale is used in the regression analyses). We analyse the patterns of participation by running several OLS-regression analyses checking stepwise the influence of the three kinds of socializations, starting from the most simple model to the most elaborated one including all three expected effects. The first Model aims in testing the religious influence on participation, therefore beside a dummy for immigrant we include also the kind of religious traditions one adheres to. Four the most frequently mentioned denominations are considered as independent variables in the regression analyses: Protestant, Catholic, Orthodox, Muslim and the remaining religious traditions are captured in a broader category other religious traditions, while non-religious is the reference category. We expect Protestantism to have 6

7 a positive impact on participatory behaviour (Curtis et al. 2001, 791), while all other religious traditions might have a negative impact (Inglehart, Norris 2004, 189). Irrespective from the adherence to a religious tradition, religious services attendance is usually claimed as a good predictor of social participation, since churches provide a connection to local religious community and mobilise potential volunteers to work in church-related social projects (Putnam 2000, 69ff., Ruiter, de Graaf, 2006). Therefore the frequency of churchgoing 1 added to the Model 1 is expected to have a positive effect on participation. Apart from religious factors we also consider a number of standard socio-demographic measures explaining social participation: age, gender and education. As widely confirmed by empirical evidence individuals with higher socio-economic status are more ready to participate (Almond, Verba 1963, 383ff., Verba et al. 1995, 281), therefore we expect that education 2 affects participation positively. Similarly men participate more frequently than women (Gabriel et al. 2002, 98), while young people are more reluctant than old to participate (Gabriel et al. 2002, 98-99). Besides, we also include an item considering presence of children in household, since we expect that immigrants could get mobilized by local communities through their children attending regularly local schools or daily-care institutions (Putnam 2000). Last, but not least we also control for the influences of national contexts by including country dummies for the eleven countries, since the relevance of various local and national settings is widely evidenced in the previous research. In addition to religious influence the second Model explores how the culture of country of origin impacts social participation. Since there is no item in the dataset indicating directly the country of origin of a respondent, we consider the country father was born in 3 as a good proxy for the cultural socialization of immigrants. The rationale behind this argument is that, in case of immigrants, cultural socialization takes place when living as a child in a different country or, through the culture the immigrant-parents are committed to. In the models we consider only father s country of birth and not mother s, since both are highly correlated, so that we can include only one of them (and both bring obviously very similar results). 4 Furthermore, we assign countries of origin to larger regional clusters according to their similar cultural legacy. The following cultural legacies are included to the analyses: Western Europe, Central-Eastern Europe including Baltic countries, ex-soviet countries (excluding Baltic region, that have to a large extend different cultural legacy as the remaining ex-soviet countries), Arab culture, Chinese culture, Indian culture, African region, Latin America region, South-East Asian region. 5 We expect that culture of origin which is similar to recipient-country culture in terms of predominance of secular-rational worldviews would have less or no impact on likelihood of participation when compared to cultures of origin stamped by traditional worldviews. In specific, we expect that immigrants socialized in Western Europe, Central Eastern Europe and Latin America might have not significant different patterns of volunteering as native 1 For the question wording please consult the web page of ESS: 2 For the question wording please consult the web page of ESS: 3 For the question wording please consult the web page of ESS: 4 The models with the impact of the mother s country of birth are calculated, too and are available upon request by author. 5 The last three legacies Africa, Latin America, South-East Asia - refer to larger regions when compared to the remaining cultures included in the analyses, however due to much lower number of cases we need to compromise the measures of cultural legacies in this way. 7

8 population, in contrast to immigrants from traditional Arab, Indian or African cultures. As far as Chinese and South-East Asian cultures are considered we can expect both kinds of directions on the one hand the hierarchical Confucian philosophy could be an obstacle for participation in some kinds of associations, on the other hand strong emphasis on collective values could result in higher participation of immigrants coming from the Confucian cultures. As far as the ex-soviet region is considered, it is difficult to predict the direction of influence. On the one hand these societies are not particularly stamped by traditional worldviews (Inglehart, Baker 2000, 29) and are mostly Christian, in so far are similar to the Western European countries. On the other hand however Orthodoxy comes together with low levels of participation in this region (Howard 2003, 82, Lasinska 2013, 49). Apart from cultures of origin Model 2 includes the same variables as in Model 1: dummy for being an immigrant, socio-demographic characteristics, variables concerning religious socialization and country dummies. The question addressed by Model 3 concerns the influence of political socialization on participatory behaviour of immigrants. Again, we consider the political regime of the country father was born in as a proxy for political socialization of an immigrant. We assign countries of (father s) origin to one of six categories that arise from two dimensions of political legacy. One dimension is the level of democratization as assessed by the Freedom House in 2012 (three types of regimes are considered: free, partly free, not free). The second dimension is post-communist legacy (or no). We expect that individuals socialized in a democratic regime without post-communist legacy would be similarly likely to work for volunteer organisations as native population, because their socialization would not differ from recipientcountry s citizens, so should have no effect on participation. In contrast, provenience from a democratic country with post-communist legacy could be a negative determinant of participation (Howard 2003). Similarly, partly free regimes with communist legacy might have stronger negative influence on participation than partly free regimes without post-communist legacy (Howard 2003). Whereas in case of not free regimes post-communist legacy probably does not bring any difference, therefore we expect a negative effect of socialization in a non-free regime irrespective from the presence or absence of postcommunist legacy. Beside variables concerning political socialization Model 3 considers the same variables as Model 2 with exception of cultural socialization, because we want to examine first these two kinds of influences political and cultural socialization separately. Finally Model 4 includes the both kinds of socialization together with all remaining variables including religious socialization and country-dummies. This allows us to analyse what kinds of influences are more relevant for explaining the readiness to volunteer of immigrants religious, cultural or political socialization. Table 2 depicts the results of linear regression analyses. 8

9 Table 2 Linear regression analyses of work for voluntary and charitable organisations in Western Europe (unstandardised regression coefficient B, R 2 and N, unweighted) Independent variables Model 1 Model 2 Model 3 Model 4 Immigrant (dummy) *** *** *** *** SOCIO-DEMOGRAPHIC INDIVIDUAL CHARACTERISTICS Education (1 = elementary; 8 = university degree) 0.136*** 0.135*** 0.134*** 0.135*** Age (year of birth) Gender dummy (1= male) Children in household (1=yes) ** ** ** ** RELIGIOUS SOCIALISATION Churchgoing (5 = more than once a week; 1 = never) *** *** *** *** Catholic (dummy) Protestant (dummy) 0.103** 0.095* * * Orthodox (dummy) Muslim (dummy) *** * * * Other denomination (dummy) 0.228** 0.299*** 0.247** 0.301*** Non-religious (ref.) CULTURAL SOCIALISATION Western Europe Central Eastern Europe (incl. Baltic) *** Ex-soviet (excl. Baltic) *** ** Arab culture * Chinese culture Indian culture *** * African region ** Latin American region South-East Asian region POLITICAL SOCIALISATION Democratic regime (excl. post-communist) Partly free regime (excl. post-communist) ** Not free regime (excl. post-communist) ** Post-communist democratic regime *** Post-communist partly free regime Post-communist not free regime ** Constant 4.401** 3.373** 3.446*** 3.555*** N Adjusted R Note: *p<0.05. **p<0.01, ***p< ESS 2012, 6.Round. First of all the analyses bring the expected evidence of lower participation of immigrants when compared to native citizens. Across all four models, being an immigrant is significant and negative determinant of volunteering. Furthermore, though its power becomes lower with inclusion of other variables, even in the last model with all socialization-related variables, being an immigrant is the third most influencial variable 6. This means that the variables included in the model not fully explain why immigrants are reluctant to volunteer in social and charitable organizations. 6 The standardized coefficients β, not shown on the Table, can be sent by author upon request. 9

10 Starting with the first Model 1, religious socialization is relevant for explaining volunteering, as expected. The most powerful religion-related predictor of participation is churchgoing and its power does not diminish with inclusion of other variables across the next models. Adherence to Protestant religious tradition positively affects the propensity of volunteering, as expected, in contrast Muslims and members of other churches volunteer less than non-religious people, what corroborates findings in other studies (Inglehart, Norris 2004). In other words, religion appears a mobilizing factor mostly due to connections to the church structures, while the adherence to a religious tradition brings not consistent results. Altogether, the effects of religious socializations remain unchanged through all four models meaning that subsequently included variables such as culture of origin do not much interfere with religion-related factors. Furthermore, across four socio-demographic characteristic considered in the analyses, only two appear significant. First, education is a good and the most powerful predictor of participation, as expected, while gender and age do not matter. Secondly, individuals living with a child in the household are less likely to volunteer than individuals without children. Therefore children do not forcibly serve as a connection to a local community resulting in volunteering, probably through the lack of time, or more involvement in informal kinds of social interactions. And this finding remains unchanged across all four models. The second Model brings us insights concerning the impact of culture of origin. As expected, immigrants originating from Western Europe and Latin America are not less likely to volunteer than native population, similar to people coming from Chinese and South-East Asia. In contrast, immigrants from Central and Eastern Europe are reluctant to participate despite the relative cultural similarity to the Western Europe, while provenience from Ex-Soviet countries results in less participation, what was expected. Lastly, the three traditional culture circles Arab, Indian and African are significantly negatively related to participation what was anticipated. Altogether the effects of culture of origin corroborate expectations, with only exception of Central Eastern Europe. This indicates, that indeed socialization in a traditional culture, in which family plays a key role in shaping social order in contrast to Western culture, can result in reluctance to volunteer in formal associations. The effects of other variables related to religion as well as socio-demographic characteristics are very similar to previous model. The aim of the third Model is to analyse the influence of political socialization, independently from the culture of origin. According to the expectations, individuals socialized in the democratic context are similarly likely to volunteer as native population, in contrast to those whose parents experienced democratization after the collapse of communism: post-communist legacy has a negative effect on propensity to volunteer, even if the country of origin is eventually successfully democratized. In contrast, post-communist legacy does not matter in case of countries that rank as partly free according to Freedom House, though socialization in partly free countries (without post-communist legacy) is related with less volunteering. Lastly, as expected, socialisation in a non-free political regime results in less participation, irrespective from post-communist legacy. In sum, the influence of political socialization corroborates the 10

11 expectations with one exception (partly free with post-communist legacy), and we can generalize: the regime an immigrant is socialized to can partly explain one s participative behaviour. Last, but not least we turn to the Model 4 that includes effects of country of origin together with the effects of political socialisation. Again, effects of socio-demographic characteristics, child in household and individual religiosity are very similar as in Model 1. In contrast, political socialization irrespective from the type of regime concerned does not play any significant role in explaining participation of immigrants, when cultural socialization is included in the model. In this full Model 4 cultural socialization in ex-soviet republics is significant and negative predictor of immigrants volunteering. Furthermore, individuals socialized in Indian culture are significantly less likely to volunteer. Otherwise, socialization in other cultures included on the model turns insignificant for volunteering. Concluding, it seems that cultural socialization is more relevant for explanation of participation patterns across immigrants than political socialization, however actually only two cultural circles matter: Indian and Russian-speaking ex- Soviet. Now, let us check whether this evidence holds also when we examine the participation patterns only across immigrants that live in the eleven Western European countries. The aim of the examination in this step of analyses is to answer the question how can we explain differences in participation among immigrants and whether we can find consistent determinants related to their culture of origin, political socialization or religion. We pursue the analyses using to a large extend the same four models as in previous step of analyses. The only difference consists in adding a new variable to the models: the length of stay in the recipient country (in years). As previous research shows, political participation among immigrants rises with the length of stay (Gonzales-Ferrer 2011, 77ff.), therefore we can expect similar effects on social participation. Otherwise we exclude for obvious reasons variable immigrant from our models, while include socio-demographic characteristics, religious factors, cultural socialization and political socialization, together with country dummies. This time we expect that political socialization in a democratic regime can have a positive influence on volunteering, since we compare participation in reference to other immigrants groups. The only exception might be socialization in countries with post-communist legacy though nowadays democratic possibly showing negative effects on volunteering. Furthermore, we expect that cultural socialization in Western Europe could influence participation positively, while socialization in traditional cultures such as Arab, Indian and African, negatively. Table 3 presents the results of the linear regression analyses of participation patterns across immigrants in Western Europe. 11

12 Table 3 Linear regression analyses of work for voluntary and charitable organisation of immigrants in Western Europe (unstandardised regression coefficient B, R 2 and N, unweighted) Independent variables Model 1 Model 2 Model 3 Model 4 SOCIO-DEMOGRAPHIC INDIVIDUAL CHARACTERISTICS Education (1 = elementary; 8 = university degree) 0.116*** 0.105*** 0.107*** 0.107*** Age (year of birth) Gender dummy (1= male) Children in household (1=yes) * * Length of stay in resident country (in years) 0.016*** 0.013*** 0.012*** 0.012*** RELIGIOUS SOCIALISATION Churchgoing (5 = more than once a week; 1 = never) *** *** *** *** Catholic (dummy) * Protestant (dummy) Orthodox (dummy) Muslim (dummy) Other denomination (dummy) Non-religious (ref.) CULTURAL SOCIALISATION Western Europe Central Eastern Europe (incl. Baltic) * Ex-soviet (excl. Baltic) * Arab culture Chinese culture Indian culture * * African region Latin American region South-East Asian region POLITICAL SOCIALISATION Democratic regime (excl. post-communist) * Partly free regime (excl. post-communist) * Not free regime (excl. post-communist) Post-communist democratic regime * Post-communist partly free regime Post-communist not free regime Constant N Adjusted R Note: *p<0.05. **p<0.01, ***p< ESS 2012, 6.Round. Analyses presented in Table 3 bring more insight into patterns of immigrants participation. Starting with religious socialization, churchgoing is a positive determinant of volunteering across immigrants, while alone adherence to a religious tradition is irrelevant with one exception being a Catholic is a negative predictor of participation. This finding is quite surprising, since Protestantism is usually related with higher propensity to volunteer, but obviously this is not the case for immigrant Protestants. Among five sociodemographic individual characteristics included in the models two show a consistent and robust influence on participation across all four models. First, higher education is a significant determinant of participation, secondly with the length of stay in a country the likelihood to be mobilized rises, too, what was 12

13 anticipated. Otherwise, immigrants having children are less likely to participate, though these effects disappear in the Model 3 and Model 4. So, the claim that immigrants find connections to voluntary associations through their children is obviously false. The aim of the second Model is to examine the impact of cultural socialization on volunteering across immigrants. Apart from positive effects of education and the length of stay Model 2 shows that indeed for certain ethnic groups the cultural socialisation can partly explain differences in participation, even when controlled for country-specific contexts. In specific, immigrants socialized in Indian culture, Central and Eastern Europe as well as in ex-soviet societies are consistently more reluctant to join voluntary associations when compared to immigrants from other regions. This finding corroborates expectations concerning the influence of cultural socialisation, yet on the other hand the anticipated positive influence of the provenience from a Western European society did not found its empirical confirmation. Similarly, socialization in traditional Arab and African cultures are not related to less participation. Altogether we conclude, that cultural socialization explains to a higher extend the differences in participation between immigrants and autochthon populations, while is less relevant for explaining different participation patterns across immigrants. Last but not least the adherence to a religious tradition is irrelevant for explanation of distinct levels of participation across immigrants, while churchgoing is related to more volunteering, what was expected. The evidence from Model 3 testing the relevance of political socialization for volunteering is even more surprising. Actually socialization in a political context has no effect on propensity to volunteering across immigrants, and this applies to all kinds of political regimes, including post-communist legacy. In contrast churchgoing is positively related to participation, while religious adherence has no effect at all, similarly to the previous Model. Otherwise with higher education and longer stay in a recipient country an immigrant is more inclined to voluntary work, as expected. Last but not least, the full Model 4, including both cultural and political socializations brings somehow different evidence, when compared to Model 3. Across all kinds of cultural socialization included in the Model, only Indian culture has a negative effect on volunteering, meaning that across all different cultures and regions immigrant come from, only the Indian has significantly different impact on participation. Other traditional cultures like Arab or African are not relevant. On the other hand political socialization shows different effects, when cultural socialization is included in the Model. In the full Model 4 socialisation in a democratic regime without post-communist legacy has a positive influence on propensity to volunteer, what was expected. Surprisingly, positive effect on volunteering has also socialization in a partly free regime without post-communist legacy. On the other hand, as expected postcommunist legacy is a negative predictor of participation, though only in case of those countries that successfully democratized. In sum, the evidence from the Model 3 and Model 4 is not robust and consistent. It seems that in order to explain different patterns of participation across immigrants we probably need somehow more specific measures of political and cultural socialisations. 13

14 4. Conclusions The aim of this study was to address the question whether the country of origin can explain the participation patterns across immigrants. In order to answer the research question we pursued several regression analyses using data from ESS (2012). The examination was pursued in two steps. First we explored the impact of different kinds of socializations on participation considering the whole population in eleven Western European countries. This gave us the insight of the differences in participation between immigrants and native population. In the second step of analyses we explored participatory behaviour only across immigrants living in these eleven Western European societies, what allowed us to understand how different countries of origin impact different patterns of participatory behaviour across different immigrants groups. First of all, there is clear evidence that immigrants volunteer less than autochthon population. And secondly, that socialization cultural and political is a relevant factor for explaining the patterns of participation across immigrants. However, the specific way how socialization influences participatory behaviour seems to be very complex and the evidence is not always consistent. At this stage of research two findings seem to be clear: first, the Indian culture is a cultural obstacle for participation that turned relevant in both steps of examination when comparing immigrants volunteering with autochthon population and with other immigrants. Secondly, post-communist legacy left indisputable imprints on social behaviour patterns of its citizens, and even when moving to democratic countries they are consequently reluctant to take part in social and charitable undertakings. As far as other influences are concerned, the findings are not robust enough to make more generalized claims about them. We probably need a more elaborated way to define the specific values orientations related to cultures that can explain their impact on participatory behaviour. A clear deficit of this study is the use of large clusters of countries of origin, with cultural similarities only assumed a priori, without empirical examination of the values orientations. There is a need to define in a more fine-grained way the cultural influence, for instance in a systematic analysis of values orientations, so that we can be more specific about which values orientations enrooted in the culture of origin can be an obstacle for participation e.g. values related to the family, hierarchies, social order etc. But even although the cultural clusters are so broadly defined and need much more specification, it is encouraging that the influence of socialization is significant explanation of immigrants participation. Beside socialization, there are two other determinants of immigrants participation showing a robust impact across all models. First, the length of stay has a positive influence on likelihood of participation, what confirms that social integration of immigrants rises with time. Secondly, churchgoing is a strong and robust determinant of volunteering across all models, too. It means that churches play an integrative role in societies not only in relation to native population, but churches can also serve as bridges for integration of immigrant groups, irrespective from their cultural and political socialization. 14

15 References: ABUDABBEH, Nuha Arab Families. An overview. In: McGOLDRICK, Monica, Joe GIORDANO, Nydia GARCIE-PRETO. Ethnicity and family therapy (New York: The Guilford Press). ALMOND, Gabriel A., Sidney VERBA The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton: Princeton University Press. BERGER, Maria, Christian GALONSKA, Ruud KOOPMANS Political integration by a detour? ethnic communities and social capital of migrants in Berlin, Journal of Ethnic and Migration Studies 30 (3): CURTIS, James E.,Douglas E. BAER and Edward G. GRABB, Nations of Joiners: Explaining Voluntary Association Membership in Democratic Societies, American Sociological Review 66 (6): DIAMOND, Larry The Spirit of Democracy: The Struggle to Build Free Societies Throughout the World. (NY: Henry Holt and Company). GABRIEL, Oskar W., Volker KUNZ, Sigrid ROßTEUTSCHER and Jan W. VAN DETH, eds., Sozialkapital und Demokratie. Zivilgesellschaftliche Ressourcen im Vergleich (Wien: WUV- Universitätsverlag). GONZALES-FERRER, Amparo The Electoral Participation of Naturalized Immigrants in Ten European Cities In MORALES Laura, Marco GUIGNI. Social Capital, Political Participation and Migration in Europe (Basingstoke: Palgrave Macmillan). HOWARD, Marc Morje The Weakness of Civil Society in Post-Communist Europe. (Cambridge: Cambridge University Press). INGLEHART Ronald, BAKER Wayne E Modernization, Cultural Change, and the Persistence of Traditional Values, American Sociological Review 65 (1): NORRIS, Pippa, Ronald INGLEHART, Sacred and Secular. Religion and Politics Worldwide (Cambridge: Cambridge University Press). GLATZER Wolfgang Integration und Partizipation junger Ausländer vor dem Hintergrund ethnischer und kultureller Identifikation. Ergebnisse des Integrationssurveys des BiB (Wiesbaden: Bundesinstitut für Bevölkerungsforschung). JACOBS D., K. PHALET and M. SWYNGEDOUW Associational Membership and Political Involvement among Ethnic Minority Groups in Brussels, Journal of Ethnic and Migration Studies, 30: LASINSKA, Katarzyna Social Capital in Eastern Europe. Poland an Exception? (Wiesbaden: Springer VS). 15

16 van LONDEN Marieke, Karen PHALET, Louk HAGENDOORN Civic Engagement and Voter Participation among Turkish and Moroccan Minorities in Rotterdam, Journal of Ethnic and Migration Studies. 33 (8): MORALES, Laura, Marco GUIGNI Social Capital, Political Participation and Migration in Europe (Basingstoke: Palgrave Macmillan). PAXTON, Pamela Social Capital and Democracy: An Interdependent Relationship, American Sociological Review 67 (2): PUTNAM, Robert D Bowling Alone. The Collapse and Revival of American Community (New York: Simon & Schuster). TOGEBY, Lise It depends how organisational participation affects political participation and social trust among second generation immigrants in Denmark Journal of Ethnic and Migration Studies 30 (3): VERBA, Sidney, Lehman K. SCHOLZMAN, Henry E. BRADY Voice and Equality: Civic Volunteerism in American Politics (Cambridge/Mass: Harvard University Press). 16