Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands This is an updated version of the paper that appeared as: Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands, Journal of Immigrant & Refugee Studies, 10:3, 338358, DOI: 10.1080/15562948.2012.693336. Compared to the published version, the following has been changed: - Recalculation of the weighted country means from the WVS2005-2007 and EVS20082010 data - Figure 2 has been replaced by a graph with 95% Confidence Interval bars. - Addition of robustness checks using only the weighted mean for West-Germany instead of all of Germany (see appendix) - Addition of robustness checks using EVS2008-2010 data (see appendix) Where these changes led to different conclusions, this has been flagged in the text. Model(ling) citizens? Integration policies and value integration of Turkish immigrants and their descendants in Germany, France and the Netherlands Evelyn Ersanilli1 This paper investigates to what extent integration policy models affect i igra ts adoption of residence country attitudes. Data come from an original survey among Turkish immigrants and their descendants in three countries with distinctive policy models; Germany, France, and the Netherlands. Two types of attitudes are investigated; room for Islam in the public sphere and moral liberalism. The opinions of the immigrants are compared both in absolute terms and in terms of the value gap with the average opinion in the residence society populations. The results suggest that the effect of integration policies is modest at best. KEYWORDS: socio-cultural integration; integration policies; Turkish immigrants; Muslims; values; immigrant attitudes; citizenship; assimilation; INTRODUCTION As the immigrant population from nonwestern countries - especially from Muslim countries – in Europe is becoming larger and more established there is growing concern that many of these immigrants do not share the core values of the liberal democratic societies they came to live in. Previous research has shown that non-Western immigrants and their children tend to be more conservative than natives; they have more traditional gender role attitudes and are less supportive of liberal values such as respect for homosexuals (Brouard & Tiberj, 2005; Diehl, Koenig, & Ruckdeschel, 2009; Entzinger & Dourleijn, 2008; Phalet, Vanloteringen, & Entzinger, 2000; Roeder, 2010; SCP, 2003; Teney & Subramanian, 2010). Many Europeans feel that the moral conservatism of immigrants and the demands of Muslim immigrants for more room for Islam in the public sphere pose a threat to their I would like to thank Ines Michalowski, Claudia Finotelli and the other participants in the DFG-Network The heu isti pote tial of odels of itize ship for their constructive comments on previous versions of this paper. 1 1 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands way of life. This fear is so widespread that even parties with a morally conservative background are trying to win voters by claiming to stand up for the protection of liberal values such as gender equality and the rights of homosexuals (Cf. Akkerman & Hagelund, 2007; Fassin, 2010). Failing integration policies are often pointed to as one of the main sources for the persistent traditionalism of immigrants. Particularly multiculturalist policies are blamed for providing too much room for immigrants to retain traditional values and even claiming a place for these in the public sphere, but empirical evidence supporting this claim is lacking. This paper sets out to examine to what extent integration policies affect the degree to which immigrants and their descendants adopt values similar to the native population. It does this by comparing the attitudes of immigrants in three countries that have historically followed different types of integration policies; Germany, France, and the Netherlands. So far there have been few cross-national studies that have explored the value integration of immigrants. The studies that do exist focus on collectivism versus individualism values or on the strength of intergenerational transmission (e.g. Berry, Phinney, Sam, & Vedder, 2006; Phalet & Schönpflug, 2001). Cross-national comparative studies on immigrant integration often suffer from a lack of adequate data. To be able to determine the influence of policy models on immigrant integration, it is important to minimise the effects of other factors that might influence cross-national differences. As will be explained below, the analyses in this paper are therefore based on original survey data on the same narrowly defined subpopulation of Turkish immigrants in each country. On average, people in Turkey hold more conservative values than people in France, Germany and the Netherlands (WVS, 2014), and there is a significant Turkish origin population in each of these three countries. In what follows I will first explain how I conceptualise integration policies and value integration, and develop hypotheses on their possible relations. This is followed by a brief overview of polices in Germany, France and the Netherlands, after which the research design and operationalisations are presented. Two types of attitudes are investigated; liberal values (gender roles, euthanasia, abortion, homosexuality and soft drug use) and attitudes on the place for Islam in the public sphere. The hypotheses are tested in multivariate analyses that control for individual-level characteristics that are known to influence attitudes. Though attitudes in all three countries in this study tend to be more liberal than in Turkey, there are also differences between these three countries. Therefore I will both compare the attitudes of Turkish immigrants and their descendants across countries and vis-à-vis the attitudes of the general population in their respective countries of residence; i.e. the absolute alues a d the alue-gap ith the ge e al populatio a oss ou t ies. To measure the latter, World Values Survey and the European Values Study data are used. After a discussion of the results, the article ends with a number of tentative conclusions. IMMIGRANTS VALUE INTEGRATION AND INTEGRATION POLICIES Past studies have pointed to a number of factors that influence the degree to which immigrants adopt values akin to those of the general population of their countries of residence. With length of residence and over generation, interaction with and socialisation in the residence society, immigrants and their children gradually adopt values that are more similar to those prevalent in the that society (ArendsTóth & Van de Vijver, 2007; Roeder, 2010; Teney, 2009; Uunk, 2003). Immigrant women are more likely than men to adopt liberal values of the residence society, especially gender equality (Diehl et al., 2009; Güngör & Bornstein, 2009; Idema & Phalet, 2007; Teney, 2009). Education and labour market 2 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands participation are associated with more liberal values, partly because they lead to more contact with the majority population (Uunk, 2003). A factor that has thus far received little research attention is the role of the residence country and its integration policies. Several scholars have developed typologies that describe the differences in integration policy regimes (e.g. Brubaker, 1992; Castles, 1995). In this paper I will use the model developed by Koopmans et al (2005) because its two-dimensional nature best captures the differences between the three countries in this study. The two dimensions are the degree to which states grant immigrants individual equality in the form of access to citizenship and protection from expulsion and discrimination, and the degree to which states accommodate the cultural identity of immigrants by supporting ethnic and religious group formation or by granting special rights or exemptions based on cultural or religious group membership. There are several theoretical perspectives that can help shed light on the possible relation between integration poli ies a d i ig a ts alue i teg atio . Follo i g a atio al hoi e perspective (see e.g. Esser, 2008) integration policies can be seen as a set of incentives and disincentives for retention of more conservative origin country values and adoption of more liberal residence country values. It can be argued that policies that accommodate diversity reduce both the costs of maintaining traditional values and the benefits of adopting more liberal values (Barry, 2001; Koopmans & Statham, 1999; Meyer, 2002). In accommodative regimes, value differences can be used as the basis for claiming support for ethnic organizations, ethnic media, and religious schools. Accommodative policies can lower the costs of retention by accommodating the up-keep of traditional values; for instance by organising women-only hours at the public swimming pool, or not firing civil servants who refuse to marry gay couples because of their moral objections to homosexuality. Furthermore, the costs of value retention are decreased if there are limited or no cultural requirements for access to the rights that come with naturalisation or permanent residence permits. It follows that in countries with accommodative policies, immigrants will retain more conservative values (H1). Following Social Identity Theory (Tajfel & Turner, 1986) the degree to which countries grant immigrants and their descendants individual legal equality through easy access to citizenship rights and state protection from discrimination can be seen as an indicator of the permeability of group boundaries (Bourhis, Moise, Perreault, & Senecal, 1997). Policies that grant a low degree of individual equality can make the boundary with the residence society seem impermeable which can lead to immigrants withdrawing into thei o i le ea ti e eth i ity a d i easing their self-esteem by emphasizing the good characteristics of their own group (Padilla & Perez, 2003; Portes & Rumbaut, 2001; Verkuyten & Yildiz, 2009). This in turn can lead to a rejection of the group norms of the residence society and an emphasis on the values of the origin group (Teney & Subramanian, 2010). Policies with a high degree of individual equality on the other hand signal that immigrants are welcome to become full members of the residence society. This can encourage immigrants and their descendants to adopt the behavioural and value pattern of the residence country. It can therefore be predicted that in countries with policies that grant a high degree of individual equality, immigrants will have attitudes more similar to the residence country society (H2a). The value integration of immigrants tends to be measured as the difference in opinion between immigrants and the general population (e.g. Diehl et al., 2009; Teney & Subramanian, 2010), in the remainder of this paper this will be referred to as the alue-gap . Because values differ across countries, the size of the shift in opinion immigrants need to make to share the general values varies. Duyvendak (2004) has a gued that the p og essi e o se sus of Dut h so iety ag ifies the alue 3 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands differences with immigrants; for immigrants in the Netherlands it is much harder to fully adopt the values of the residence society because those values differ more from the values they had when leaving their country of origin than in most other European countries. I will therefore also test a weaker variant of hypothesis H2a looking at absolute instead of relative levels of liberalism 2 ; in countries with policies that grant a high degree of individual equality, immigrants will have more liberal attitudes (H2b). Unlike for liberal attitudes, it is difficult to determine a depa tu e alue - i.e. the attitude at the time of migration - for i ig a ts suppo t fo room for Islam in the public sphere. The attitude people have on rights for minority religions can change when their situation changes from being a member of the majority to a member of an immigrant minority. Musli i ig a ts suppo t fo oe room for Islam in the public sphere is support for rights for their own group. Studies on the political integration of immigrants have shown that immigrant organisations adapt their political claims to the institutional setting of their residence country (Koopmans et al., 2005; Kortmann, this issue). In countries that accommodate diversity, policies reflect the norm that minority religions should have a place in the public sphere. Even though this norm might (no longer) be fully endorsed by the residence country society (Cf. Brubaker, 1992: 184, Helbling this issue), it can still affect the attitudes of immigrants. This leads to the hypothesis that in countries with policies that accommodate diversity, immigrants more strongly endorse room for Islam in the public sphere (H3). Integration policies in Germany, France and the Netherlands The idea that countries can be treated as exponents of a certain integration policy model faces increasing criticism (see e.g. Duyvendak & Scholten, 2009; Freeman, 2004; Joppke, 2007). However, for an investigation of the effects of integration policies, policy models still provide a useful heuristic tool (Cf. Bader, 2007). Several recent comparative policy studies (Banting, Johnston, Kymlicka, & Soroka, 2006; Koopmans, Michalowski, & Waibel, 2012; Koopmans et al., 2005) have revealed that there are still fairly consistent cross-national policy differences. These studies take a more dynamic approach to policy models and also look at change over time. Koopmans et al. (2012) have operationalised the two above mentioned dimensions of integration policies into a range of policy indicators and measured these for four points in time (1980, 1990, 2002, 2008). They found that despite shifts and changes in policies, Dutch policies throughout this period consistently showed a higher degree of accommodation of diversity than French and German policies. They also showed that German policies have offered a lower degree of individual equality than French and Dutch policies – even after the 2000 citizenship reform. Other studies confirm the direction of these cross-national differences (Banting et al., 2006; Geddes & Niessen, 2005). France and the Netherlands have a lower residence requirement for naturalisation and the subsequent legal equality than Germany; five compared to eight (and until 2000, fifteen) years of legal residence. France unconditionally allows the retention of previous citizenship, the Netherlands de facto condone it, but Germany is (much) less lenient. Until 2000 German-born children of immigrants could only obtain German nationality through naturalisation. In France the second generation is attributed citizenship automatically and in the Netherlands members of the second generation since 1985 have an option right to Dutch citizenship (for more information, see e.g. Bauböck et al 2006). France and the Netherlands also have more extensive anti-discrimination legislation than Germany. 2 Since Germany, France and the Netherlands all have on average more liberal values than Turkey, it is justified to ph ase this hypothesis i te s of li e alis . 4 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Integration policies in the Netherlands are best known for their relatively high degree of accommodation of cultural and religious diversity. Partly as a consequence of legislation that was put in place during the time of pillarisation (Lijphart, 1968), the Netherlands has far more publicly funded Islamic schools (currently about 40, compared to 1 in Germany and France), Islamic broadcasting corporations and ethnic organisations and consultative bodies than France and Germany. The integration policy of the 1980s and early 1990s has ofte ee efe ed to as i teg atio ith the p ese atio of ig a ts o ide tity . This i luded oo fo alues that diffe ed f om those of the Dutch mainstream. The Dutch Commission on Equal Treatment (Commissie Gelijke Behandeling) has promoted an expansive interpretation of discrimination on religious grounds and has in recent years several times ruled in favor of Muslim plaintiffs who demanded the right to wear veils or refused to shake hands with members of the opposite sex in the workplace. French policies have been most restrictive in providing room for Islam, especially when it comes to the headscarf. Not only has it barred civil servants from wearing a headscarf, it is the only of the three countries in this study to have banned it for secondary school students. In some German states teachers and civil servants are not allowed to wear a headscarf but students are. In the Netherlands only police officers and members of the judiciary are not allowed to wear a headscarf. For a long time the Netherlands was also more lenient in integration requirements for naturalisation; language requirements were modest and cultural requirements were absent (van Oers, 2008). This stands in contrast to French and German naturalisation procedures that have long demanded a certain degree of cultural integration (Hagedorn, 2001). The pre-2000 German naturalisation guidelines listed o ie tatio to a ds Ge a ultu e Hinwendung zum Deutschtum) as a precondition (Hailbronner & Renner, 1998). In France civil servants not only test applicants language proficiency but also his or her knowledge of French culture and values (Hagedorn, 2001; Zoka, 2002). Practicing polygamy or wearing a headscarf has sometimes been treated as a sign of insufficient assimilation (Hagedorn, 2001; Weil & Spire, 2006). The Netherlands only in 2003 introduced a formal naturalisation test that includes knowledge of Dutch cultural practices and values (see e.g. Michalowski, 2009). By that time however a large share of the immigrant population had already naturalised. DATA AND METHODOLOGY The most comprehensive study on values is the World Values Survey and its European pendant the European Values Study. These datasets however do not provide a good basis for testing the hypotheses in this paper because they contain a very limited number of immigrants, and it is likely that there is a selection bias because the surveys were only conducted in the national languages of the survey country and not in immigrant languages. The European Social Survey dataset contains a larger number of immigrants, however it was also only conducted in the national languages of participating countries, again likely undersampling immigrants who are not fluent in that language (Van Tubergen & Sindradottir, 2011). Since language proficiency varies cross-nationally (Ersanilli & Koopmans, 2011; van Tubergen & Kalmijn, 2005), this bias can confound the cross-national comparison. The analyses in this paper are therefore based on an especially collected dataset on Turkish immigrants and their descendants in France, Germany and the Netherlands that has been designed to minimise confounding variance (Ersanilli, 2010). Germany has been the main destination of Turkish immigrants - currently the estimated size of the Turkish origin population totals almost 2.5 million - followed by the Netherlands and France - each with about 350,000 people of Turkish origin. 5 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Since type and timing of migration can also influence integration, the target group is limited to people who migrated during the guest-worker era - i.e. before 1975 - and their offspring who were born in the residence country or arrived as minors under family reunification regulations. Turkey is a country with large regional differences in wealth, education, ethnic composition and religious life. Because these regional differences might also impact integration, the target group is further limited to immigrants from South-Central and East-Central Anatolia.3 South-Central Anatolia is a predominantly ethnic Turkish and religiously Sunnite region. East-Central Anatolia, by contrast, has more ethnic and religious diversity (Kurds and Alevis).4 The focus on this narrow target group might mean that the results are not representative for the entire Turkish origin populations in Germany, France and the Netherlands; it does however reduce the likelihood of confounding variance and therefore improves the cross-national comparison that is the main goal of this paper. Data were gathered in a telephone survey from November 2005 to June 2006. All interviewers were bilingual and respondents could choose between answering the questions in either Turkish or the language of their residence country. To minimise sampling bias the same combination of sampling techniques was used in all three countries. The main sample was drawn from online telephone books on the basis of stems of common Turkish surnames.5 Variables For liberal values five questions were borrowed from the European and World Values Surveys. The attitude towards gender relations is measured by A pre-school child is likely to suffer if his or her othe o ks . Respondents could answer on a 5-point scale ranging fro 1 o pletely disag ee to 5 o pletely ag ee . Coding was inverted so that a higher score reflects a more liberal attitude. In addition, respondents were asked to what extent they find four types of behaviour justifiable; homosexuality, abortion, euthanasia, and soft drug use. The original WVS question uses a 10-point answer scale, but because data were collected in a telephone survey a 5-point scale was used ranging from 1 e e justifia le to 5 al ays justifia le .6 Again a higher score corresponds to a more liberal attitude. Attitudes o oo fo Isla e e easu ed ith t o uestio s; Tea he s should ot ea a heads a f i side the s hool a d The e should e o e Isla i s hools i [residence ou t y] . Respondents could again answer on a 5-poi t s ale a gi g f o 1 o pletely disag ee to 5 o pletely ag ee . The s o es o the uestio o tea he s i the lass oo e e i e ted so that a higher score shows a higher degree of support for the presence of religion in the public sphere. The multivariate analyses reported below are controlled for several individual level and context variables that are known to influence values and attitudes. For generation, dummy variables for the second generation (born in the residence country) and the in-between or 1.5 generation (born in Turkey but migrated before the age of 18) are included with the first generation immigrants as the reference category. For education, dummies are added for secondary and post-secondary education, South-Central Anatolia consists of the p o i es Afyo , Aksa ay, Ka a a , Kayse i, Ko ya, Ne şehi , a d Niğde. East-Ce t al A atolia e o passes Adiya a , A asya, Elaziğ, Malatya, Tokat, Tu eli a d “i as. 4 Alevism is a humanistic current within Islam. In general the relation between the sexes is different from that prevalent within Sunni Islam, and Alevi women rarely wear headscarves. 5 More information on the sampling techniques used in this study and response rates can be found in Ersanilli (2010) and under http://dare.ubvu.vu.nl/bitstream/1871/19217/1/appendix_appendix.pdf 6 The questions from the EVS and the WVS were only included in the second wave of data collection (AprilJune 2006). This is why the N in the tables is lower than for the other items. 3 6 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands leaving people with no or only primary education as the reference group. Further dummies are added for females (vs. males), for labour market participation, and people who are married (ref. group is single and divorced). To control for the specific design of this study, a dummy variable is added for respondents from East-Central Anatolia, leaving South-Central Anatolia as the reference category, for Alevis with Sunnis and the very small category of non-religious respondents as reference group, and two dummies for the different sampling strategies used with the phone-book sample as the reference category. Though Turkish immigrants form a significant population in all three countries, the relative size of the Turkish origin community varies. To control for cross-national differences in the relative size and within-country distribution of the Turkish origin population the relative size of the Turkish immigrant population in the place of residence of the respondent is included in all regression models. A table with descriptives by country is included in the appendix. Since this study has only been conducted in three countries it is not possible to include quantitative measures for the differences in policies. Instead dummy variables are used to measure the differences between Germany, France and the Netherlands. In the tables shown below, Germany serves as the reference category. Additional analyses have been done to test the differences between France and the Netherlands for significance. Outcomes of these analyses are reported in the tables and text. For the five liberal values the analyses are done in two models. The first model compares the alue gap a oss ou t ies. This was calculated by deducting the residence country average from the score of each respondent. In all countries Turkish immigrants are less liberal than the general population; this means that a negative coefficient for the country dummies indicates a less liberal attitude compared to the general population. Information on the general opinion of the host population7 is taken from the 2005-2007 World Values Survey (WVS, 2014). Not all question were asked in this round. For the question on soft drug use, the country means were calculated from the 1999-2001 WVS and for the question on working mothers the 2008-2010 European Values Survey (EVS, 2011). The scores from the EVS and WVS have been rescaled to fit the scale used in the survey of Turkish immigrants. The second model analyses the absolute score of agreement with each value across countries (i.e. differences in absolute levels of liberalism). RESULTS Liberal attitudes To get a sense of the difference in attitudes between the general population in the three destination countries and Turkey it is helpful to look at the data from the EVS and WVS. Figure 1 shows the results from the 2008 European Values Survey of the question on working mothers (EVS, 2011). Of course Turkey has seen a lot of change since the (parents of the) survey respondents left the country over forty years ago, nevertheless the data give an impression of the diffe e e et ee the depa tu e alue a d the a e age attitude i the eside e ou t y. As Figure 1 shows, the general populations in France and the Netherlands have the most liberal attitude towards working mothers, followed by Germany. The general population in Turkey has the most conservative attitude. 7 For Germany, the country average was used. Robustness checks based on only the mean opinion in WestGermany are included in the appendix. 7 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands 100% 90% 80% 70% 60% disagree strongly 50% disagree 40% agree agree strongly 30% 20% 10% 0% France Germany Netherlands Turkey Figure 1. A swers of the ge eral populatio to A pre-school child is likely to suffer if his or her Source: EVS 2008-2010, weighted other works Figure 2 gives the population averages for the four other indicators of liberalism in the 2006 round of the World Values Survey (WVS, 2014). For homosexuality and soft drug usage, the Dutch population stands out as being the most liberal, for euthanasia they are a bit less liberal than the French population. The French have the most liberal attitude towards abortion. Of the three immigration countries, Germany is most conservative, however the population of Turkey is much more conservative than of any of the three immigration countries. Moving to the attitudes of immigrants and their descendants, following the cost-benefit perspective a higher degree of conservatism in countries with accommodative policies was hypothesised (H1). If this holds true the results should show a higher degree of conservatism in the Netherlands as this is, as described above, the country in this study that most accommodates diversity. Based on Social Identity Theo y it as p edi ted that i ig a ts attitudes a e ost si ila to those of the general population in countries that grant a higher degree of individual equality (H2a). This would mean that the value gap in France and the Netherlands is smaller than in Germany, as the former provide immigrants with more individual equality. Following the weaker variant of this hypothesis on absolute attitude scores (H2b), immigrants and their descendants in the Netherlands and France should be more liberal than those in Germany. 8 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Homosexuality Abortion 5.0 5.0 4.5 4.5 4.0 4.0 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 Netherlands France Germany Netherlands Turkey Euthanasia France Germany Turkey Soft drugs 5.0 5.0 4.5 4.5 4.0 4.0 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 Netherlands France Germany Turkey Netherlands France Germany Turkey Figure 2. Attitudes of the general population in France, Germany, the Netherlands and Turkey (higher scores indicate a higher level of justification) Source: Soft drugs: EVS 1999-2001, all others WVS 2005-2007. Weighted and rescaled from 1-10 to 1-5. Table 1 presents the results of multivariate analyses for each of the five liberalism items. For each item the dependent variable in the first model is the value gap (i.e. relative score compared to the host population) and in the second model the absolute score. In all three countries the majority of respondents agreed that a young child will suffer if the mother goes to work (see appendix). The first column of Table 1 shows that for this item, French Turks are more conservative compared to the general population in their country of residence than Turkish immigrants in Germany. The difference between Germany and the Netherlands is in the same direction but is not significant8. The comparison of the absolute scores in the second column, however paint a different picture and show that Turkish immigrants in the Netherlands and France have a marginally (p<.10) more liberal attitude in this matter than those in Germany. These results are only in line with the weak variant of the hypothesis on the effect of a high degree of individual equality. 8 In the published version from 2012 the difference between the Netherlands and Germany was marginally significant. When only looking at West-Germany – the part of Germany where the Turkish guest-workers moved to - the difference between the Netherlands and Germany is significant (p<.05) (see Table A2 in the appendix). 9 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands For homosexuality there is no significant difference in the value gap across countries9. The second model shows Turkish immigrants in the Netherlands have a significantly more liberal attitude towards homosexuality than Turkish immigrants in France. There is no significant difference between Germany and France. This pattern fits none of the hypotheses. The value gap in attitudes towards abortion is larger for Turkish immigrants in France than in Germany, i.e. French Turks are relatively more conservative than the general population the German Turks10. For both Dutch and French Turks the value gap in attitudes towards euthanasia is larger than for German Turks. Finally, for soft drugs use, Dutch Turks are relatively more conservative than both German and French Turks. None of these three items show cross-national differences when measured in absolute terms. Again none of these results are in line with the hypotheses on policy effects. The found differences in the size of the value gap seem mainly driven by the degree of liberalism of the general population in the countries of residence and not by integration policies. The share of variance explained by the country dummies in the models for the absolute scores is low; adding the country dummies to a model with only the control variables does not significantly improve the model fit.11 This suggests that the residence country context plays only a very modest role in i ig a ts adoptio of li e al alues. The results do show a consistent positive relation between education and liberal attitudes. As shown in previous research women and those active in the labour market have a more liberal attitude on gender roles and homosexuality, however this does not hold true for the other items. Surprisingly, the in-between and second generation are not more liberal than their parents, with the exception of the se o d ge e atio s suppo t fo soft d ug use. This ight ho e e e o e a efle tio of youth culture than a value shift. 9 When the gap is calculated based on the weighted country means from EVS 2008-2010, the gap in the Netherlands is significantly larger than in Germany (p<.001) and France (p<.001). Table A3 in the appendix. Analysis of EVS 1999-2001 shows the same pattern, suggesting that the Dutch WVS 2006 sample might be an outlier. 10 When the gap is calculated based on the weighted country means from EVS 2008-2010, the gap in the Netherlands is significantly larger than in Germany (p<.001) and France (p<.001). Table A4 in the appendix. Analysis of EVS 1999-2001 shows the same pattern, suggesting that the Dutch WVS 2006 sample might be an outlier. 11 It does improve the model fit in the value-gap model but this is an artificial effect because the calculation of the value-gap makes the scores more dependent on the country of residence. 10 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands TABLE 1. Unstandardized coefficients of OLS regression of the value gaps and absolute value scores (standard errors in parentheses) Working mother Homosexuality Abortion Euthanasia Softdrugs gap absolute gap absolute gap absolute gap absolute gap absolute Germany Ref Ref Ref. Ref. Ref Ref Ref. Ref. Ref Ref France -0.34* 0.29+ -0.17 -0.17 -0.36* 0.20 -1.06*** -0.23 -0.03 0.07 (0.15) (0.15) (0.14) (0.14) (0.15) (0.15) (0.17) (0.17) (0.07) (0.07) Netherlands -0.22 0.29+ -0.19 0.13 ª -0.17 -0.03 -0.90*** -0.28 -0.42*** ª 0.11 (0.16) (0.16) (0.15) (0.15) (0.16) (0.16) (0.18) (0.18) (0.08) (0.08) East-Central -0.10 -0.10 0.04 0.04 0.20 0.20 0.06 0.06 0.03 0.03 (0.13) (0.13) (0.12) (0.12) (0.13) (0.13) (0.15) (0.15) (0.06) (0.06) Alevi -0.07 -0.07 0.41+ 0.41+ 0.72** 0.72** 0.92*** 0.92*** 0.04 0.04 (0.23) (0.23) (0.21) (0.21) (0.23) (0.23) (0.25) (0.25) (0.11) (0.11) st 1 generation Ref. Ref. Ref. Ref. 1.5 generation -0.12 -0.12 -0.09 -0.09 0.21 0.21 0.04 0.04 0.06 0.06 (0.18) (0.18) (0.17) (0.17) (0.18) (0.18) (0.21) 0.25) (0.09) (0.09) 2nd generation 0.19 0.19 0.03 0.03 0.05 0.05 0.18 0.18 0.30** 0.30** (0.22) (0.22) (0.20) (0.20) (0.22) (0.22) (0.25) (0.25) (0.11) (0.11) Female 0.27* 0.27* 0.33** 0.33** -0.09 -0.09 -0.15 -0.15 -0.10 -0.10 (0.12) (0.12) (0.11) (0.11) (0.12) (0.12) (0.14) (0.14) (0.06) (0.06) Married -0.07 -0.07 -0.24 -0.24 0.10 0.10 0.00 0.00 0.06 0.06 (0.16) (0.16) (0.15) (0.15) (0.16) (0.16) (0.18) (0.18) (0.08) (0.08) No/primary Ref. Ref. Ref. Ref. education Secondary 0.38* 0.38* 0.43** 0.43** 0.26 0.26 0.53** 0.53** 0.04 0.04 (0.16) (0.16) (0.15) (0.15) (0.16) (0.16) (0.19) (0.19) (0.08) (0.08) education Post-secondary 0.60** 0.60** 1.36*** 1.36*** 1.21*** 1.21*** 0.98*** 0.98*** 0.36*** 0.36*** education (0.21) (0.21) (0.20) (0.20) (0.22) (0.22) (0.24) (0.24) (0.10) (0.10) Working 0.22+ 0.22+ 0.27* 0.27* 0.08 0.08 0.05 0.05 -0.05 -0.05 (0.12) (0.12) (0.11) (0.11) (0.12) (0.12) (0.14) (0.14) (0.06) (0.06) Phonebook sample Ref Ref Ref. Ref. Ref Ref Ref. Ref. Ref Ref Snowball sample -0.14 -0.14 -0.12 -0.12 -0.09 -0.09 0.03 0.03 0.01 0.01 (0.14) (0.14) (0.13) (0.13) (0.14) (0.14) (0.16) (0.16) (0.07) (0.07) %Turkish pop. -0.06 -0.06 -0.04 -0.04 -0.09* -0.09* -0.04 -0.04 -0.00 -0.00 (0.05) (0.05) (0.04) (0.04) (0.05) (0.05) (0.05) (0.05) (0.02) (0.02) Constant -0.92*** 1.81*** -1.98*** 1.45*** -1.06*** 1.81*** -0.90*** 1.83*** -0.36*** 0.96*** (0.27) (0.27) (0.25) (0.25) (0.27) (0.27) (0.30) (0.30) (0.13) (0.13) Adj. R2 0.05 0.05 0.16 0.17 0.11 0.11 0.15 0.07 0.11 0.06 N 567 567 537 537 567 567 545 545 571 571 Two-tailed t-tests, + p<0.10, * p<0.05, ** p<.01, *** p<.001 ª the difference between France and the Netherlands is significant (p<.001) 11 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Support for room for Islam in the public sphere As the policy descriptions above showed, the Netherlands has accommodated Islam to the highest extent, followed by Germany and then France. Following the hypothesis (H3), we should find that immigrants in the Netherlands more strongly endorse room for Islam in the public sphere, than in the two other countries. Table 2 shows the results of the multivariate regression. With the exception of the headscarf question in France, the majority of respondents in all three countries agreed with both statements (see appendix). This shows a preference for (more) room for Islam in the public sphere. The endorsement of the right of teachers to wear a headscarf follows the expected pattern; support for this is significantly higher in the Netherlands than in Germany and France, and significantly higher in Germany than in France. This result is all the more remarkable since civil servants and pupils at state schools and universities in Turkey are not allowed to wear a headscarf. This might explain why the in-between and second generation are more in support of this policy than the generation of their parents; their countries of residence provide more support for this than their origin country. The result is also in line with answers of the general population to the 2005 PEW Global Attitudes survey; 22% of the French, 40% of the German and 46% of the Dutch population felt it is a ad idea to a the ea i g of heads a es y Musli o e i pu li pla es i ludi g s hools. For the second item, however, the results are not in line with the hypothesis; it is respondents in France, not in the Netherlands, who are most in favour of more Islamic schools. Possibly this pattern is caused by a difference in the current number of schools; the 40 or so Islamic schools in the Netherlands might be seen as sufficient whereas the one publicly and two privately funded schools in France are seen as too little. In the months before the survey, Islamic schools in the Netherlands received a lot of bad publicity due to corruption, poor quality of teachers, and low test scores. This ight also ha e affe ted espo de ts a s e s. To y k o ledge the e is o ge e al populatio survey that asks about support for Islamic schools. The first round of the European Social Survey did ha e a elated ite ; Co munities of people who have come to live here should be allowed to educate thei hild e i thei o sepa ate s hools if they ish . I te esti gly he e also o se t as highest in France and lowest in the Netherlands; 37% vs. 12% (ESS, 2002). This difference between policy and attitudes in the Netherlands possibly reflects the wider debate there on the negative consequences of faith-based schools on social cohesion and immigrant integration. Addition of the country dummies significantly improves model fit (p<.001). The variance explained by the country dummies is larger than for the liberal values; at three per cent and two per cent respectively. This suggests that the national context has a larger impact on claims for Islam in the public sphere than for liberal attitudes, though it is still modest. 12 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands TABLE 2. Unstandardized coefficients of OLS regression of attitudes on the position of Islam (Standard errors in parentheses) Teachers allowed to Wants more Islamic schools wear headscarf Germany Ref. Ref. France -0.40** 0.41*** (0.13) (0.11) Netherlands 0.38** ª -0.15 ª (0.13) (0.11) East-central 0.20 -0.09 (0.11) (0.09) Alevi -1.45*** -1.57*** (0.19) (0.16) st 1 generation Ref. Ref. 1.5 generation 0.63*** 0.10 (0.17) (0.14) 2nd generation 0.88*** 0.09 (0.20) (0.16) Female 0.02 -0.18* (0.11) (0.09) Married -0.11 -0.20 (0.14) (0.12) No/primary education Ref. Ref. Secondary education -0.14 -0.40** (0.15) (0.12) Post-secondary -0.19 -0.86*** (0.20) (0.17) Working -0.02 -0.24** (0.11) (0.09) Phonebook sample Ref. Ref Holiday sample 0.12 0.12 (0.17) (0.14) Snowball sample 0.12 0.14 (0.11) (0.09) %Turkish pop. 0.02 -0.02 (0.04) (0.03) Constant 2.86*** 4.23*** (0.23) (0.19) Adj R2 0.14 0.21 N 887 886 Two-tailed t-tests, *p < .05 ** p < .01 *** p< .001 ª the difference between France and the Netherlands is significant (p<.001) 13 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands DISCUSSION AND CONCLUSION This paper explored the influence of integration policies on the extent to which immigrants retain conservative values or adopt liberal values and support a place for Islam in the public sphere. Hypotheses were formulated and tested with a dataset from a subgroup of Turkish immigrants and their descendants in three countries with different policy types; Germany, France and the Netherlands. The analyses showed that Turkish immigrants in the Netherlands are most supportive of the right of teachers to wear a headscarf and those in France least supportive. This is in line with the policy differences between these countries, suggesting immigrants copy the norms of the political system they live. Support for increasing the number of Islamic schools conversely did not follow policy differences. The cross-national difference on this item did however show a pattern similar to the attitude of the ge e al populatio as easu ed i the fi st a e of the E““. Possi ly i ig a ts lai s for a space in the public sphere for their minority religion is influenced more by the general public opinion than by the possibilities offered by the political-institutional system they live in. The overall picture from the analyses of the liberal attitude items is that the value gap between Turkish immigrants and the general population is smallest in Germany, though the crossnational differences are not always significant. Analyses of the absolute scores however suggest that this pattern is mainly due to the higher degree of conservatism of German society. The cross-national differences for liberal attitudes were not in line with the hypotheses on policy effects. For most variables the addition of the country dummies did not lead to significant model improvement. These results suggest that the effects of integration policies on the value integration of immigrants are small or even absent. The (dis)incentives for value integration provided by integration policies in the three countries in this study are apparently not large enough to have an effect. In light of the role that is attributed to policies in most political and societal debates, this is an important finding. It is however possible that no policy effects were found, because the differences in policies between the three countries in this study are simply too small to have a noticeable effect. Though integration policy models do not appear to have an effect, the residence countries do influence the attitudes of their new citizens. While Turkish immigrants in France, Germany and the Netherlands are more conservative than the general population of their residence country, they are more liberal than the general population of Turkey. The more positive outlook on homosexuality in the Netherlands, the country that most strongly treats acceptance of homosexuality as one of its core values, together with results for the items on room for Islam in the public sphere provide additional indications that value adoption is taking place. Care should be taken when generalising the results from this study. I was only able to do a rudimentary analysis of the value gap between immigrants and the native population. Because the data come from different datasets I was not able to control for characteristics such as age, level of education, that might explain (part of) the value gap between immigrants and natives. However, the restricted nature of the sample is also the major strength of this study, since it minimizes crossnational composition effects, which previous comparative studies have not been able to control for sufficiently. It is thus a supplement to large-N studies that include more origin and destination countries but might also suffer from more confounding variance. Subsequent studies should focus on datasets that include both natives and sufficient numbers of immigrants from different origin countries. Ideally the samples should be large enough to also take regional variations within countries into account. Finally, it is possible that the lack of cross-national differences is caused by the selection of dependent variables. It might be that they are too abstract. Subsequent studies should look at less 14 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands abstract questions. The analyses of the value-gaps and the absolute values show different patterns of cross-national differences; hopefully future cross-national studies will profit from this result and also take a dual approach to the operationalization of value integration. REFERENCES Akkerman, T., & Hagelund, A. (2007). ''Women and children first!' Anti-immigration parties and gender in Norway and the Netherlands'. Patterns of Prejudice, 41(2), 197-214. Arends-Tóth, J., & Van de Vijver, F. (2007). Cultural and Gender Differences in Gender-Role Beliefs, Sharing Household Task and Child-Care Responsibilities, and Well-Being Among Immigrants and Majority Members in The Netherlands. Sex Roles, 57(11), 813-824. Bader, V. (2007). The Governance of Islam in Europe: The Perils of Modelling. Journal of Ethnic and Migration studies, 33(6), 871-886. Banting, K., Johnston, R., Kymlicka, W., & Soroka, S. (2006). Do multiculturalism policies erode the welfare state? An empirical analysis. In K. Banting & W. Kymlicka (Eds.), Multiculturalism and the welfare state: recognition and redistribution in contemporary democracies (pp. 49-91). Oxford: Oxford University Press. Barry, B. (2001). Culture & Equality. An Egalitarian Critique of Multiculturalism. Cambridge: Polity Press. Berry, J. W., Phinney, J. S., Sam, D. L., & Vedder, P. (2006). Immigrant Youth in Cultural Transition: Acculturation, Identity and Adaptation Across National Contexts. New Jersey: Lawrence Erlbaum Associates. Bourhis, R. Y., Moise, L. C., Perreault, S., & Senecal, S. (1997). Towards an Interactive Acculturation Model: A Social Psychological Approach. International Journal of Psychology, 32(6), 369-386. Brouard, S., & Tiberj, V. (2005). Francais comme les autres? Enquête sur les citoyens d'origine maghrébine, africaine et turque: Presse de la Fondation Nationale des Sciences Politiques. Brubaker, R. (1992). Citizenship and nationhood in France and Germany. Cambridge: Harvard. Castles, S. (1995). How nation states respond to immigration and ethnic diversity. New Community, 21(3), 293-308. Diehl, C., Koenig, M., & Ruckdeschel, K. (2009). Religiosity and gender equality: comparing natives and Muslim migrants in Germany. Ethnic and Racial Studies, 32(2), 278-301. Duyvendak, J. W. (2004). Een eensgezinde vooruitstrevende natie. Over de mythe van 'de' individualisering en de toekomst van de sociologie. Amsterdam: Vossiuspers UvA. Duyvendak, J. W., & Scholten, P. (2009). Le "modèle multiculturel" d'intégration néerlandais en question. Migrations Societé, 21(122), 1-30. Entzinger, H., & Dourleijn, E. (2008). De lat steeds hoger. De leefwereld van jongeren in een multietnische stad. Assen: Van Gorcum. Ersanilli, E. (2010). Comparing Integration. Host culture adoption and ethnic retention among Turkish immigrants and their descendents in France, Germany and the Netherlands. Vrije Universiteit Amsterdam. Ersanilli, E., & Koopmans, R. (2011). Do Immigrant Integration Policies Matter? A Three-Country Comparison among Turkish Immigrants. West European Politics, 34(2), 208-234. ESS. (2002). Round 1: European Social Survey Round 1 Data: Data file edition 6.2. Norwegian Social Science Data Services, Norway – Data Archive and distributor of ESS data. 15 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Esser, H. (2008). Assimilation, ethnische Schichtung oder selektive Akkulturation? Neuere Theorien der Eingliederung von Migranten und das Modell der intergenerationalen Integration. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 48(special issue), 81-107. EVS (2011). European Values Study 1981-2008, Longitudinal Data File. GESIS Data Archive, Cologne, Germany, ZA4804 Data File Version 2.0.0 (2011-12-30) DOI:10.4232/1.11005. Fassin, E. (2010, 20-12-2010). Pourquoi Marine Le Pen défend les femmes, les gays, les juifs…. Libération, from http://www.liberation.fr/politiques/01012309000-pourquoi-marine-le-pendefend-les-femmes-les-gays-les-juifs Freeman, G. P. (2004). Immigrant Incorporation in western Democracies. International Migration review, 38(3), 945-969. Geddes, A., & Niessen, J. (2005). European Civic Citizenship and Inclusion Index. Brussels: British Council. Güngör, D., & Bornstein, M. H. (2009). Gender, Development, Values, Adaptation, and Discrimination in Acculturating Adolescents: The Case of Turk Heritage Youth Born and Living in Belgium. Sex Roles, 60, 537-548. Hagedorn, H. (2001). Wer darf Mitglied warden? Einbürgerung in Deutschland and Frankreich im Vergleich. Opladen: Leske. Hailbronner, K., & Renner, G. (1998). Staatsangehörigkeitsrecht. M•nchen: Beck. Idema, H., & Phalet, K. (2007). Transmission of gender-role values in Turkish-German migrant families: The role of gender, intergenerational and intercultural relations. Zeitschrift für Familienforschung, 19(1), 71-105. Joppke, C. (2007). Beyond national models: Civic integration policies for immigrants in Western Europe. West European Politics, 30(1), 1-22. Koopmans, R., Michalowski, I., & Waibel, S. (2012, in press). Citizenship Rights for Immigrants: National Political Processes and Cross-National Convergence in Western Europe, 1980-2008, American Journal of Sociology Koopmans, R., & Statham, P. (1999). Challenging the Liberal Nation-State? Postnationalism, Multiculturalism, and the Collective Claims-Making of Migrants and Ethnic Minorities in Britain and Germany. American Journal of Sociology, 105(3), 652-696. Koopmans, R., Statham, P., Giugni, M., & Passy, F. (2005). Contested citizenship. Immigration and Cultural Diversity in Europe. Minneapolis: University of Minnesota Press. Lijphart, A. (1968). The Politics of Accommodation: Pluralism and Democracy in the Netherlands. Berkeley: University of California Press. Meyer, T. (2002). Identitätspolitik. Vom Missbrauch kultureller Unterschiede. Frankfurt am Main: edition Suhrkamp. Michalowski, I. (2009). Citizenship Tests in Five Countries - An Expression of Political Liberalism? WZB Discussion Paper SP IV 2009-702 Padilla, A. M., & Perez, W. (2003). Acculturation, Social Identity and Social Cognition: A New Perspective. Hispanic Journal of Behavioral Sciences, 25(1), 35-55. Phalet, K., & Schönpflug, U. (2001). Intergenerational Transmission of Collectivism and Achievement Values in Two Acculturation Contexts: The Case of Turkish Families in Germany and Turkish and Moroccan Families in the Netherlands. Journal of cross-cultural psychology, 32(2), 186201. Phalet, K., Vanloteringen, C., & Entzinger, H. (2000). Islam in de multiculturele samenleving: Universiteit Utrecht, European Research Centre on migration and Ethnic Relations. 16 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Portes, A., & Rumbaut, R. G. (2001). Legacies; the story of the immigrant second generation. New York: Russell Sage Foundation. Roeder, A. (2010). Gender gap? The divergent gender role attitudes of immigrant men and women. Paper presented at the Trinity Immigration Initiative Conference. SCP. (2003). Rapportage minderheden 2003. Onderwijs, arbeid en sociaal-culturele integratie. Den Haag: SCP. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behaviour. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations. (pp. 7-24). Chicago: Nelson-Hall. Teney, C. (2009). Acculturation and prejudice against sociological minorities among Brussels youth: A multilevel regression approach. Universite libre de Bruxelles, Brussels. Teney, C., & Subramanian, S. V. (2010). Attitudes toward homosexuals among youth in multiethnic Brussels. Cross-Cultural Research 44 (2), 151-173. Uunk, W. (2003). The Cultural Integration of Immigrants in the Netherlands: A Description and Explanation of Modern Attitudes of Turks, Moroccans, Surinamese, Antilleans and the Indigenous Population. In L. Hagendoorn, J. Veenman & W. Vollebergh (Eds.), Integrating immigrants in the Netherlands: Cultural versus socio-economic integration (pp. 199-233). Aldershot: Ashgate. van Oers, R. (2008). From Liberal to Restrictive Citizenship Policies: the Case of the Netherlands. International Journal of Multicultural Societies, 10(1), 40-59. van Tubergen, F., & Kalmijn, M. (2005). Destination-Language Proficiency in Cross-National Perspective: A Study of Immigrant Groups in Nine Western Countries. American Journal of Sociology, 110(5), 1412-1457. Van Tubergen, F., & Sindradottir, J. I. (2011). The Religiosity of Immigrants in Europe: A CrossNational Study. Journal for the Scientific Study of Religion, 50(2), 272-288. Verkuyten, M., & Yildiz, A. A. (2009). Muslim immigrants and religious group feelings: selfidentification and attitudes among Sunni and Alevi Turkish-Dutch. Ethnic and Racial Studies, 32(7), 1121-1142. Weil, P., & Spire, A. (2006). France. In R. Bauböck, E. Ersboll, K. Groenendijk & H. Waldrauch (Eds.), Acquisition and Loss of Nationality Policies and Trends in 15 European States (Vol. Volume 2: Country Analyses, pp. 187-211). Amsterdam: Amsterdam University Press. WORLD VALUES SURVEY (2014). Wave 5 2005-2008 OFFICIAL AGGREGATE v.20140429. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: Asep/JDS, Madrid SPAIN. Version history: 2014-04-29: Official release Zoka, N. (2002). Comment peut-on être francais? [DVD]: Artline Films. 17 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands APPENDIX TABLE A1. Descriptives Germany Mean SD Dependent variables Teachers should not wear headscarf (reversed coded) There should be more Islamic schools Young child suffers if mother works (reversed coded) Justifiable: homosexuality Justifiable: abortion Justifiable: euthanasia Justifiable: soft drugs Independent variables East-Central Anatolia Alevi First generation In-between generation Second generation Female Married No / primary education Secondary education Post-secondary education Working Phone book sample Holiday sample Snowball sample Relative size of Turkish imm. pop. France Mean SD Netherlands Mean SD Range 3.18 1.67 2.87 1.46 3.72 1.38 1 5 3.28 1.47 3.94 1.21 3.31 1.27 1 5 2.12 1.32 2.42 1.36 2.37 1.34 1 5 1.92 2.31 2.38 1.15 1.34 1.29 1.66 0.58 1.70 2.46 2.11 1.20 1.17 1.44 1.44 0.69 2.04 2.18 2.05 1.25 1.36 1.29 1.35 0.71 1 1 1 1 5 5 5 5 0.56 0.17 0.18 0.53 0.29 0.42 0.78 0.20 0.68 0.12 0.53 0.60 0.11 0.29 2.89 0.50 0.38 0.38 0.50 0.45 0.49 0.42 0.40 0.47 0.33 0.50 0.49 0.31 0.46 1.51 0.28 0.05 0.25 0.45 0.30 0.45 0.82 0.31 0.58 0.11 0.50 0.63 0.04 0.32 1.31 0.45 0.22 0.43 0.50 0.46 0.50 0.39 0.46 0.49 0.31 0.50 0.48 0.20 0.47 1.20 0.29 0.06 0.18 0.53 0.30 0.51 0.77 0.24 0.59 0.17 0.46 0.57 0.19 0.24 2.63 0.45 0.23 0.38 0.50 0.46 0.50 0.42 0.43 0.49 0.38 0.50 0.50 0.39 0.43 1.24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.02 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7.33 18 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands Homosexuality Abortion 5.0 5.0 4.5 4.5 4.0 4.0 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 Netherlands Netherlands France Germany(W) Turkey Euthanasia Soft drugs 5.0 5.0 4.5 4.5 4.0 4.0 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 Netherlands France Germany(W) Turkey France Germany(W) Turkey Netherlands France Germany(W) Turkey FIGURE A1. Attitudes of the general population in France, West-Germany, the Netherlands and Turkey (higher scores indicate a higher level of justification) Source: Soft drugs: EVS 1999-2001, all others WVS 2005-2007. Weighted and rescaled from 1-10 to 1-5. TABLE A2. Unstandardized coefficients of OLS regression of the value gaps (standard errors in parentheses) working mother. EVS 2008-2010 West Germany only Germany Ref France -0.49** (0.15) Netherlands -0.37* (0.16) Constant -0.94*** (0.27) Adj. R2 0.06 N 567 Two-tailed t-tests, *p < .05 ** p < .01 *** p< .001 Notes: Controlled for region, denomination, generation, gender, marital status, education, employment, sample and size of co-ethnic community. 19 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands TABLE A3.Unstandardized coefficients of OLS regression of the value gaps (standard errors in parentheses) Homosexuality. WVS 2005-2007 EVS 2008-2010 EVS 2008-2010 West Germany only West Germany only Germany Ref Ref Ref France -0.08 -0.19 -0.17 (0.14) (0.14) (0.14) Netherlands -0.10 -0.76*** ª -0.74*** ª (0.15) (0.15) (0.15) Constant -0.94*** -1.64*** -1.66*** (0.27) (0.25) (0.25) Adj. R2 0.17 0.17 0.17 N 537 537 537 Two-tailed t-tests, *p < .05 ** p < .01 *** p< .001 ª the difference between France and the Netherlands is significant (p<.001) Notes: Controlled for region, denomination, generation, gender, marital status, education, employment, sample and size of co-ethnic community. TABLE A4.Unstandardized coefficients of OLS regression of the value gaps (standard errors in parentheses) Abortion. WVS 2005-2007 EVS 2008-2010 EVS 2008-2010 West Germany only West Germany only Germany Ref Ref Ref France -0.43* -0.42** -0.46** (0.15) (0.14) (0.15) Netherlands -0.24 -0.37* -0.41* (0.16) (0.16) (0.16) Constant -0.99*** -0.79*** -0.76** (0.27) (0.27) (0.27) Adj. R2 0.12 0.12 0.12 N 567 567 567 Two-tailed t-tests, *p < .05 ** p < .01 *** p< .001 Notes: Controlled for region, denomination, generation, gender, marital status, education, employment, sample and size of co-ethnic community. 20 Ersanilli, Evelyn (2012) Model(ling) citizens? Integration Policies and Value Integration of Turkish Immigrants and Their Descendants in Germany, France, and the Netherlands TABLE A5.Unstandardized coefficients of OLS regression of the value gaps (standard errors in parentheses) Euthanasia. WVS 2005-2007 EVS 2008-2010 EVS 2008-2010 West Germany only West Germany only Germany Ref Ref Ref France -1.22*** -1.09*** -1.06*** (0.17) (0.17) (0.17) Netherlands -1.06*** -1.10*** -1.07*** (0.18) (0.18) (0.18) Constant -0.91*** -0.86** -0.90** (0.30) (0.30) (0.30) Adj. R2 0.17 0.16 0.15 N 545 545 545 Two-tailed t-tests, *p < .05 ** p < .01 *** p< .001 Notes: Controlled for region, denomination, generation, gender, marital status, education, employment, sample and size of co-ethnic community. TABLE A6.Unstandardized coefficients of OLS regression of the value gaps (standard errors in parentheses) Soft drug use. WVS 1999-2001 EVS 2008-2010 EVS 2008-2010 West Germany only West Germany only Germany Ref Ref Ref France 0.02 0.12 0.18* (0.07) (0.07) (0.07) Netherlands -0.36*** ª -.20** ª -.14+ ª (0.08) (0.08) (0.08) Constant -0.50*** -0.53*** -0.59*** (0.13) (0.13) (0.13) Adj. R2 0.11 0.09 0.09 N 571 571 571 Two-tailed t-tests, + p<0.10 *p < .05 ** p < .01 *** p< .001 ª the difference between France and the Netherlands is significant (p<.001) Notes: Controlled for region, denomination, generation, gender, marital status, education, employment, sample and size of co-ethnic community. 21