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- Figure 2 has been replaced by a graph with 95% Confidence Interval bars.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