Global Networks 6, 3 (2006) 285–308. ISSN 1470–2266. © 2006 The Author(s)
Journal compilation © 2006 Blackwell Publishing Ltd & Global Networks Partnership 285
Transnational involvement and social integration
ERIK SNEL, GODFRIED ENGBERSEN AND ARJEN LEERKES
Abstract In this article we offer a quantitative examination of the extent to which
migrants from various countries are involved in transnational activities and have
transnational identifications. The study is based on a survey of 300 immigrants (from
the USA, Japan, Iraq, former-Yugoslavia, Morocco and the Dutch Antilles) living in
the Netherlands. The respondents are deliberately chosen to include different categories
of immigrants. Transnational activities constitute a substantial part of their
lives and are to a large extent socio-cultural. Many migrants also transfer money
abroad. Professional economic activities were rare and mainly limited to the American
group. As a whole, our respondents identify more with other compatriots living in
the Netherlands than with people living outside the Netherlands. The research also
found that transnational involvement in general does not impede ‘immigrant integration’.
Migrant groups that are known as poorly integrated into Dutch society are
not more involved in transnational activities and have no stronger identifications with
the country of origin than other groups. However, within the Moroccan and Antillean
groups those respondents with the weakest labour market position identify more
strongly with the country of origin than others. Strong identifications with compatriots
living elsewhere and withdrawal from Dutch society may reinforce their poor
labour market integration
In this article we examine how transnational involvement of immigrants living in the
Netherlands relates to their incorporation or integration into Dutch society. As various
authors have observed, the old concept of immigration, that immigrants settle permanently
and assimilate in the host country, has lost significance. The modern transmigrant
is at home in several different social worlds, speaks several languages,
participates in cross-border social networks and political movements, and sometimes
makes a living with transnational economic activities. But what do these transnational
activities and identifications of modern ‘transmigrants’ imply for their incorporation
or integration into the host society? This is the central research question in this article.
Using survey data of 300 immigrants in the Netherlands coming from six different
Western and non-Western countries we examine their transnational involvement and
whether this impedes their integration into Dutch society.
There are two central concepts in our argument, transnationalism and immigrant
incorporation or integration. The concept of ‘transnationalism’ came up in the early
1990s when anthropologists noticed intense interactions between the sending and
receiving countries of international migrants (Glick Schiller et al. 1992). As Glick
Erik Snel, Godfried Engbersen and Arjen Leerkes
286 © 2006 The Author(s)
Schiller and Fouron (1999: 344) state, ‘transnational migration is a pattern of migration
in which persons, although they move across international borders, settle, and
establish relations in a new state, maintain ongoing social connections with the polity
from which they originated. In transnational migration people literally live their lives
across international borders. Such persons are best identified as “transmigrants”.’ The
result is the genesis of ‘transnational communities’, ‘transnational social fields’ and
‘transnational social spaces’ (Faist 2000; Guarnizo et al. 2003; Vertovec 1999). For
modern transmigrants, primary social connections and identifications are less in relation
to social groups in a national context than to transnational communities.
Concepts such as transnationalism and transmigrants have led to several debates
(Guarnizo et al. 2003; Levitt et al. 2003; Portes et al. 1999, 2002; Waldinger and
Fitzgerald 2004). One debate deals with the novelty of the phenomenon. Various
authors argue that transnational activities and lasting solidarity with sending countries
have always existed (Gerbner 2000). However, as Portes and others convincingly
argue, modern transportation and communication technologies have allowed transnational
activities and movements to reach a scope and intensity that were impossible
in former times (Portes et al. 1999, 2002). A second debate deals with the actors
behind transnationalism. Who are the transmigrants? Some authors want to confine
the notion of transnationalism ‘to a new class of immigrants, economic entrepreneurs
or political activists who conduct cross-border activities on a regular basis’ (Guarnizo
et al. 2003: 1212). Others argue that transnationalism may not be restricted to professional
connections and activities, and that lasting family ties and remittances have
a huge impact on both sending and (immigrant communities in) receiving countries
(Bryceson and Vuorela 2002; Staring 2001). Lastly, some authors argue that bodily
geographic mobility is not a necessary requirement for transnational involvement.
Besides migrants who regularly travel back and forth on a professional or nonprofessional
basis, there are also migrants who live primarily in the country of origin
or the host country, but whose lives and identities are still integrally entwined with the
people, social relations and resources in other countries (Guarnizo and Smith 1998;
Levitt 2003; Vertovec 2001). These distinctions between professional and nonprofessional
activities and between bodily geographic and non-bodily geographic
mobility will be taken into account in our operationalization of forms of transnational
involvement.
Both adherents and critics of the notion of transnationalism draw attention to the
limited empirical basis of these theories. Reflections on transnationalism and transmigrants
are often either theoretical or based on qualitative case studies. As these case
studies tend to focus on migrants who are involved in transnational activities, they
give the impression that transnationalism has become the major form of immigrant
incorporation. Studies of random samples of migrants who may or may not be
involved in transnational activities are rare and limited to the USA (Guarnizo et al.
2003; Portes et al. 2002). Our research focuses on a European country and, what is
more important, an advanced welfare state, the Netherlands. This difference in institutional
context may influence what transnational activities and transnational identifications
imply for immigrants. In the USA, with marginal social security, transnational
Transnational involvement and social integration
© 2006 The Author(s) 287
economic activities may be a crucial alternative income strategy for immigrants. In
European countries like the Netherlands many non-Western immigrants live on social
security. For them, (non-economic) transnational activities and strong transnational
identification may function as a comforting safe haven that in effect impedes their
incorporation in the host society. This is where the second concept, immigrant
incorporation or integration, comes into view.
The question of whether transnational activities and identifications impede or
support the incorporation of immigrants into the host society has, to our knowledge,
not been examined extensively. Before elaborating on the relationship between transnationalism
and immigrant integration, we have to make some remarks about the
latter concept in general. How immigrants adapt to their new environment has always
been important in sociological research. Sociologists refer to assimilation, acculturation,
incorporation and integration (Alba and Nee 1997; Engbersen 2003; Esser
2003; Gordon 1964; Vermeulen and Penninx 2000). ‘Immigrant integration’ in
general refers to the incorporation of new elements (immigrants) into an existing
social system. There is a general consensus that integration is a multi-dimensional
concept, although authors differ on what the relevant dimensions of integration are.
Many authors distinguish socio-economic aspects of integration (‘social positioning’)
from social and cultural aspects (interethnic relations, cultural adjustment and shared
norms). Here we follow a distinction widely used in Dutch migration literature
between structural integration on the one hand and social and cultural integration on
the other. The former refers to the social position of migrants in the host society, particularly
in terms of their level of education and position in the labour market. Social
and cultural integration refers, on the one hand, to informal social contacts of immigrants
with native Dutch people and, on the other, to the extent to which immigrants
endorse the host society’s prevailing moral standards and values (Vermeulen and
Penninx 2000; see also Dagevos 2001). Empirical research shows that structural and
cultural dimensions of integration are strongly related. Migrants with good social
positions (high education, stable job) generally also have more informal contact with
native Dutch people and more often endorse ‘modern’ ideas and values than other
migrants (Dagevos 2001; Odé 2002).
Despite limited knowledge on the relation between transnationalism and integration,
the current prevailing political view is that the two are at odds. The terrorist
attacks on New York and Washington (2001), Madrid (2003) and London (2005),
which can be interpreted as incidents of transnational political activism, have
strengthened this point of view. Since 2001, Western states have increasingly been
raising questions about the allegiance and political bona fides of their immigrant
residents (Waldinger and Fitzgerald 2004). In the Netherlands, for instance, the
government opposes dual citizenship because it deems unacceptable the resulting dual
loyalties of migrants. Migrants should choose their host country (in this case the
Netherlands).
1
Transnational activities (like sending remittances) and a continuing
identification with the sending country or with internationally dispersed migrant communities
(the so-called diaspora) are often perceived as an impediment to immigrant
integration into the host country.
Erik Snel, Godfried Engbersen and Arjen Leerkes
288 © 2006 The Author(s)
In academic literature, one can find two opposing assumptions about the relationship
of transnationalism and integration. One position is that transnational economic
activities, such as international trade, may constitute an alternative income strategy
particularly for underprivileged immigrants (Portes et al. 1999). More generally, in a
study on transnational entrepreneurs in the USA, Portes and others show that transnational
entrepreneurs have strong connections with their compatriots, both in the
USA and the country of origin, but are also very well integrated into American
society. In other words, migrants’ strong transnational involvement and integration
into the host country do not rule each other out (Portes et al. 2002). On the other hand,
this may differ according to class: ‘Those [immigrants] who have income, education,
and language skills are more likely to be able to choose transnational activism, while
those with less social and cultural capital are more likely to be forced into it’ (Levittt
2003: 183). For underprivileged migrant groups, continuing transnational identifications
and retaining the customs of the home country may rather impede adequate
incorporation into the host country (Joppke and Morawska 2003; Levitt 2003;
Morawska 2003).
In this article we examine empirically how various migrant groups’ patterns of
transnationalism relate to their integration into Dutch society. Before doing so, however,
we will describe the transnational activities and transnational identifications of
our respondents in detail. To start, we have to define the concept of transnationalism
more precisely. Concepts like ‘transnational fields’ or ‘transnational communities’ are
rather vague for empirical research. Here we use the term transnational involvement,
which includes both transnational activities and transnational identifications. The
former refers to migrants’ cross-border activities and practices. The latter refers to the
extent to which migrants living in the Netherlands identify with compatriots living
outside the country (in the sending country or in the international diaspora). We
explain the precise operationalization of transnational activities and identifications in
the subsequent section. Next, we describe the transnational activities of our respondents,
their transnational identifications and finally we return to the central research
question about the relationship between transnational involvement and immigrant
integration.
Research, definitions and operationalizations
This article is based on a survey of 300 respondents from six different immigrant
groups living in the Netherlands. The survey consisted of structured questions on the
respondents’ involvement in various transnational activities, transnational identifications,
and various facets of their integration into Dutch society. As we wanted to paint
as varied a picture of migrants as possible, we interviewed migrants from three
different categories: older immigrant groups or ‘ethnic minorities’ in the Netherlands
(Morocco and Dutch Antilles, former Dutch colonies in the West Indies); more recent
immigrants from what count as ‘typical refugee countries’ (former Yugoslavia and
Iraq); and so-called ‘new labour migrants’ and their partners from prosperous Western
countries (the United States and Japan).
Transnational involvement and social integration
© 2006 The Author(s) 289
We interviewed 50 people from each immigrant group. In most cases junior
researchers who came from the respective migrant groups carried out the interviews.
Respondents were selected and approached by the so-called snowball method. In
other words, at the end of each interview respondents were asked whether they knew
of any other potential respondents in their community. The interviewers also looked
for respondents in their private social networks, which guaranteed access to them. The
disadvantage, however, is that interviewed respondents cannot be considered representative
of whole immigrant populations. Another point worth mentioning is that not
every respondent is a typical example of the category to which he or she belongs. For
instance, one in four Moroccan respondents and one in five respondents from former
Yugoslavia were second-generation. The latter group consists of children of the first
wave of Yugoslavian immigrants who came to the Netherlands in the 1960s as guest
workers. There was practically no one from the second generation among the Iraqi,
Japanese and American respondents. The advantage of including children of immigrants
in the sample is that we obtained more variation in the period of residence in
the Netherlands.
Due to their different migration backgrounds, the groups also differed in terms of
social position. For example, more than three-quarters of the American and Japanese
respondents have high levels of education. The same goes for almost half the Iraqi
respondents, but for less than a quarter of the Yugoslavian, Antillean and Moroccan
respondents. More than 80 per cent of the American and Japanese respondents are
formally employed. The share of working respondents is the lowest in the Moroccan
group (only 36 per cent), but this is also because we interviewed a number of students
in this group. In the other groups (Yugoslavs, Antilleans and Iraqis), the share of
working people varies from 50 to 60 per cent.
Transnational activities and identifications
In this article we focus on two dimensions of transnationalism – transnational activities
and transnational identifications. These two dimensions together give us insight
into the transnational involvement of migrants. By transnational activities we mean
cross-border activities of an economic, political or socio-cultural nature. In operationalizing
these dimensions, we draw inspiration from a study by Al-Ali et al. (2001) of
Bosnian and Eritrean refugees, in which the authors develop an empirically founded
typology of transnational activities. This typology distinguishes between transnational
activities aimed at the country of origin and activities aimed at the host country. The
former involve cross-border activities in the true sense of the word, such as money
transfers, or visits to and political participation in the country of origin. Then there are
transnational activities within the host country: visiting cultural events with artists
from the country of origin, participating in meetings that many compatriots attend, or
mobilizing political support for parties or movements in the country of origin. Our
questions on transnational activities were strictly confined to cross-border activities
between the country of origin and the host country, and did not concern activities or
contacts within the internationally dispersed migrant communities or diasporas.
Erik Snel, Godfried Engbersen and Arjen Leerkes
290 © 2006 The Author(s)
A second key concept is transnational identification. This refers to the extent to
which migrants living in the Netherlands identify with compatriots outside the
country (cross-border identifications). Operationalizing the concept of ‘transnational
identification’ is based on some elementary notions from socio-psychological, sociological
and anthropological literature. An essential point of departure is that people
construct their social identities (Bauman 2004; Jenkins 1996; Verkuyten 1999). These
social identities indicate how people define themselves in relation to their social
environment. It is not about what distinguishes one individual from the other, but
about what is shared with others. The social identity of a person refers to two basic
questions in life: (1) to whom do I belong? and (2) how should I behave? These two
key questions relate to the group dimension and the normative dimension of social
identity respectively (Douglas 1978, 1996; Wildavsky 1987, 1989).
Our research, however, focuses on ethnic identifications, namely the extent to
which people feel related to a particular ethnic group and orient themselves towards
the norms and values of that group. We examined the extent to which respondents
identify with native Dutch residents, with compatriots living in the Netherlands, or
with compatriots living either in the country of origin or elsewhere outside the
Netherlands. It is only in the latter two cases that transnational identifications occur.
Strong identifications with compatriots living in the Netherlands indicate strong
ethnic identifications, but not transnational identifications.
In the survey, we used two different tools to measure the respondents’ ethnic
identifications. First, we presented the respondents with 50 statements about to whom
they feel close, are proud of, are occasionally ashamed of, whose norms and values
are taken into account, and with whom they agree on ‘the important things in life’.
For every question the respondents could indicate how they viewed their relations visà-vis
(1) native Dutch people, (2) compatriots living in the Netherlands, (3) compatriots
living in the country of origin, and (4) compatriots in third countries. On the
basis of our respondents’ answers to the statements we distinguished the two dimensions
of social identities that were already mentioned: the group dimension and the
normative dimension. Some questions focused on how close respondents feel to
persons belonging to various other groups, while others referred to on whose norms
and values the respondents orient themselves. It appeared that both dimensions could
indeed be distinguished quite well in the respondents’ answers. That is why we constructed
two scales representing both dimensions of identification (group dimension
and normative dimension).
2
One limitation of using the survey questions method is, however, that it partly
depends on the respondents’ language skills (although many were interviewed in their
own language) and on their own (partly culturally-determined) interpretation of the
questions. To avoid these problems, we added a second, non-verbal method for establishing
ethnic identifications. The respondents were asked to indicate graphically how
close they felt to native Dutch people, compatriots in the Netherlands, compatriots in
the country of origin, and compatriots elsewhere. The respondents could indicate
whether they saw their relationship with another category as two separate circles, as
two completely overlapping circles, or as something in between these two extremes
Transnational involvement and social integration
© 2006 The Author(s) 291
(see Figure 1). The outcomes of this so-called circle score can be interpreted as the
degree to which respondents feel close to various groups of people. Here again, a
strong sense of closeness with compatriots living outside the Netherlands (in the
country of origin or elsewhere) is considered as transnational. Finally, we calculated a
total score of identifications by adding up the outcomes of all measurements.
3
Figure 1: Ethnic identifications (circle score)
As mentioned above, we distinguish a structural and a socio-cultural dimension of
integration. The structural one was measured by using various characteristics of the
respondents (educational level, employment status, living on social benefits). The
socio-cultural one was measured by using the respondents’ answers relating to their
contact with native Dutch people (number of native Dutch in social network) and to
their identifications with native Dutch people.
Transnational activities
Table 1 shows to what extent respondents of various migrant groups are involved in
all kinds of transnational activities. A distinction is made between economic, political
and socio-cultural activities.
Everyday economic activities such as sending money or goods to the country of
origin, home ownership or donations to charities in the country of origin occur relatively
often, although distinctions can be made between the various migrant groups in
this respect. This type of support of family and/or relatives in the country of origin is
found very often among Moroccans, Iraqis and Yugoslavs, but much less often among
Americans, Japanese and Antilleans. The main reason why Americans and Japanese
migrants do not support family and/or relatives in their country of origin is no doubt
that the latter do not rely on the support of overseas relatives.
Professional economic activities such as investments in, business dealings with, or
business trips to the country of origin were relatively rare among our respondents.
Only US and Japanese respondents have regular business contact with their country of
origin. However, since our US and Japanese respondents are predominantly highlyskilled
with good positions in the business world they can hardly be typified as a new
‘class of immigrant transnational entrepreneurs’ in the sense of Portes et al. (2002).
They in part are employees of large transnational corporations travelling around the
world for business.
Erik Snel, Godfried Engbersen and Arjen Leerkes
292 © 2006 The Author(s)
Table 1: Overview of transnational activities per migrant group (in percentage of
the group total)
Morocco
(N= 50)
Dutch
Antilles
(N= 50)
Iraq
(N= 50)
(former)
Yugoslavia
(N= 50)
Japan
(N= 50)
USA
(N= 50)
Everyday economic activities
Transfers money to family 40 16 72 62 02 04
Sends goods to country of origin 28 14 4 54 12 14
Owns house in country of origin 16 08 14 46 24 14
Contributions to charities in country of origin 12 02 30 26 00 26
Total (involved in at least one activity) 54 28 78 82 32 42
Professional economic activities
Invests in companies in country of origin 02 02 06 00 10 24
Conducts trade with country of origin 04 02 00 00 08 02
Visits country of origin for business 00 08 00 02 14 32
Total (involved in at least one activity) 04 10 06 02 22 50
Political activities
Reads newspapers from country of origin 10 58 62 70 54 66
Keeps in touch with politics in country of origin 70 56 80 76 64 96
Member of political party in country of origin 02 06 24 00 06 20
Participates in demonstrations related to
country of origin
04 06 52 58 04 06
Total (involved in at least one activity) 72 76 88 94 82 100
Sociocultural activities in country of origin
Visits family/friends in country of origin 90 78 36 92 72 86
Frequent contacts with family in country of
origin
72 82 94 92 92 98
Member of social organization in country of
origin
02 04 22 00 06 16
Total (involved in at least one activity) 94 96 94 96 96 100
Sociocultural activities in the host country
Member of organization related to country of
origin
16 16 62 08 18 16
Attends meetings with primarily compatriots 50 60 62 60 38 16
Visits cultural events 56 44 28 34 66 56
Total (involved in at least one activity) 86 90 96 84 88 68
The figures in this table indicate the percentage of respondents reporting the activity concerned.
Transnational involvement and social integration
© 2006 The Author(s) 293
The other immigrant groups participate much less in transnational business activities.
Immigrant groups with the weakest labour market positions (Moroccans and
Antilleans) participate least in these activities. There is also a significant positive
relation between involvement in transnational business activities and having a formal
job. We may conclude that ‘immigrant transnational entrepreneurship’ as an alternative
route to social success for migrants is rare in the Netherlands.
Political activities relating to the country of origin occur frequently among all
migrant groups. The majority of respondents read newspapers from their country of
origin, keep up-to-date with the politics of, or are members of, a political organization
in their country of origin and/or occasionally take part in demonstrations against the
politics in their country of origin. All US respondents are engaged in at least one of
these transnational political activities. That Americans, Yugoslavs and Iraqis are quite
up-to-date with political developments in their countries of origin is partly due to
extensive media coverage in the Netherlands. Iraqis and Yugoslavs often take part in
demonstrations relating to the country of origin. More than half of both groups
indicate that they do so occasionally (the empirical research was conducted before the
2003 Iraq war). Moroccans are the least involved in transnational political activities.
Transnational socio-cultural activities are also very frequent. We distinguish
between activities in the country of origin and activities in the Netherlands. The former
involve, for example, visiting and maintaining contacts with family and friends in
the country of origin or being a member of public organizations in the country of
origin. In all groups, at least nine out of ten respondents are involved in this type of
activity. Almost three-quarters of the US respondents have intensive contact with
family and friends at home (at least once a week). Moroccans have the least frequent
contact with their families. Nonetheless, half of all Moroccan respondents (including
a substantial number of second-generation migrants) contact their country of origin at
least once a month.
Last, transnational socio-cultural activities in the Netherlands include going to
meetings attended mainly by compatriots, supporting cultural activities featuring
artists from the country of origin, and joining migrant or other organizations in the
Netherlands with connections to the country of origin. This type of socio-cultural
activity occurs very frequently among our respondents. For example, more than half
the Moroccan, Antillean, Iraqi and Yugoslav respondents mentioned they occasionally
go to meetings that are primarily attended by compatriots. The Japanese and
Americans do this much less frequently. Generally, Americans seem to be the least
involved in this type of transnational socio-cultural activity in the Netherlands.
One conclusion is that transnational activities constitute a substantial part of the
lives of migrants in the Netherlands. This goes for all the migrant groups that participated
in our survey. A large majority of our respondents maintains intensive contact
with family and friends in the country of origin, attend meetings or cultural activities
in the Netherlands related to the country of origin, are up-to-date with, and sometimes
active in, the politics of the country of origin, and send money and goods to the
country of origin on a regular basis. Only the Japanese and Americans do so much
less. However, this should not be seen as a sign of less transnational involvement on
Erik Snel, Godfried Engbersen and Arjen Leerkes
294 © 2006 The Author(s)
the part of these migrant groups as they do participate very frequently in other transnational
activities.
The second finding is that socio-cultural transnational activities prevail among
most of the migrants, followed by political and everyday economic activities. Transnational
entrepreneurship is rather rare. We know transnational entrepreneurship
exists among migrants in the Netherlands, but apparently not on a large enough scale
that it shows up in a limited quantitative survey (Van Tillaart 2001). Third, we
observed considerable differences between the migrant groups with respect to the
nature and scope of transnational activities. These differences seem to be related
partly to the nature of the migration to the Netherlands. For example, migrants from
typical refugee countries (like Iraq and former Yugoslavia) are more strongly
involved in political activities, and migrants from groups with many highly-skilled
labour migrants are involved in international trade. If we add up all the different
transnational activities, we see that former Yugoslavs and Americans participate most
strongly in transnational activities. The Japanese, Moroccans, and especially the Iraqis
and Antilleans are comparatively much less involved in transnational activities (cf.
Table 2). These findings already say something about the supposed relationship
between transnational involvement and integration into Dutch society. It is untrue that
the immigrant groups often mentioned as being poorly integrated into Dutch society
(Moroccans, Antilleans) show more transnational involvement: quite the contrary.
Fourth, we examined whether the differences in transnational activities between
the various migrant groups result from differences in personal characteristics of the
respondents. This indeed proves to be the case to some extent (cf. Table 2). Model 1
in Table 2 again shows the already mentioned differences in transnational activities
between migrant groups. In Model 2, we included personal characteristics in the
analysis that indeed make a difference. Whereas Moroccan respondents in Model 1
were significantly less involved in transnational activities, these differences become
smaller (and no longer significant) in Model 2. That Moroccans are less involved in
transnational activities can therefore be partly explained by personal characteristics of
respondents, especially their age of migration. Migrants who came to the Netherlands
at an older age participate more – ceteris paribus – in transnational activities. This is
especially true for Moroccan respondents. The Moroccan respondents who came to
the Netherlands at a young age (or were born in the Netherlands) are less involved in
transnational activities, and that explains why Moroccans in general participate less in
transnational activities than former Yugoslavs and Americans. Antilleans and Iraqis
remain significantly less involved in transnational activities, even if we take into
account the personal background of the respondents in these groups.
Finally, we would like to point out that the absence of significant differences
might also be an interesting finding. When transnational activities are indeed an
impediment to immigrant integration, one would expect particularly poorly-educated,
unemployed social security recipients to be involved in such activities. Following
traditional assimilation theories (Alba and Nee 1997) one would also expect transnational
involvement to decrease with the length of stay in the host country. However,
this is not the case. Transnational activities occur equally among all migrants, inde-
Transnational involvement and social integration
© 2006 The Author(s) 295
pendent of level of education, social status and length of stay. Only when we look at
the social backgrounds of the actors by type of transnational activity we see some
striking differences. Professional economic activities occur significantly more frequently
among the highly-educated respondents with formal paid jobs. What you see
here is a sort of Matthew effect (‘Ye who have, will be given’), which was also found
in a study on informal labour (Pahl 1987). People who participate in the formal labour
market benefit more from the informal economy than the unemployed do. In our
study we found that immigrants with good jobs in the host country are more involved
in transnational business activities than poor unemployed migrants.
Table 2: Involvement in transnational activities (linear regression: coefficients
are betas)
Model I Model II
Country of origin (compared with USA)
Yugoslavia –0.10** -0.17**
Japan –0.21** –0.20**
Morocco –0.28** –0.11**
Iraq –0.32** –0.27**
Antilles –0.35** –0.24**
Background characteristics
Sex –0.01**
Length of stay –0.01**
Age of migration –0.27**
Education –0.03**
Dutch nationality –0.02**
Formal, paid job –0.02**
Social benefit –0.05**
Explained variation (R) –0.20** –0.25**
* p < 0.05. ** p < 0.01.
Transnational identifications
The second main question in our analysis deals with ethnic and transnational identifications.
With whom do our respondents identify? Transnational identifications are
defined as cross-border identifications, that is if migrants living in the Netherlands
identify more strongly with people outside the country (either in the country of origin
or in some other country) than with other Dutch residents (native Dutch people or
compatriots living in the Netherlands). As explained before we distinguish between
two dimensions of identification: a group dimension (to whom do I belong?) and a
normative dimension (whose norms and values are important to me?). Furthermore,
Erik Snel, Godfried Engbersen and Arjen Leerkes
296 © 2006 The Author(s)
we asked respondents to describe their relationship with other groups graphically (the
circle score) and finally we calculated a total score of identification with various
groups by adding up the outcomes of all these measures of identification. All
outcomes are summarized in Table 3.
Table 3: Ethnic and transnational identification by group of origin
Morocco
(N= 50)
Dutch
Antilles
(N= 50)
Iraq
(N= 50)
(former)
Yugoslavia
(N= 50)
Japan
(N= 50)
USA
(N= 50)
Group dimension
Native Dutch 3.0 2.9 3.1 3.2 2.9 3.4
Compatriots in The Netherlands 3.9 3.6 4.0 4.0 3.5 2.9
Compatriots in country of origin 3.6 3.6 4.0 4.1 3.5 3.3
Compatriots in other countries 3.5 3.5 3.6 3.9 3.4 3.2
Normative dimension
Native Dutch 2.6 2.7 2.4 2.8 2.9 3.2
Compatriots in The Netherlands 2.9 2.9 3.4 3.0 2.9 2.4
Compatriots in country of origin 2.7 2.9 3.1 3.0 2.9 2.6
Compatriots in other countries 2.6 2.8 2.9 3.0 2.9 2.5
Feels close to (circle score)
Native Dutch 2.1 1.8 1.6 2.2 2.0 3.0
Compatriots in The Netherlands 3.1 2.3 2.1 2.9 2.4 2.6
Compatriots in country of origin 2.3 2.6 2.4 3.1 2.6 2.6
Compatriots in other countries 1.8 1.7 1.5 2.1 1.7 1.8
Total score identification
Native Dutch 2.5 2.3 2.2 2.6 2.5 3.2
Compatriots in The Netherlands 3.2 2.8 2.9 3.2 2.8 2.6
Compatriots in country of origin 2.7 2.9 3.0 3.3 2.9 2.8
Compatriots in other countries 2.4 2.4 2.4 2.8 2.4 2.3
Note: The numbers in the table are scores on a scale from 1 up to 5.
The outcomes show several things. First, our immigrant respondents identify more
with compatriots living either in or outside the Netherlands than with native Dutch
people. This is true for almost all immigrant groups, irrespective of how identifications
are measured. US respondents are the only exception here. They identify more
strongly with Dutch people than with fellow Americans living in the Netherlands. Of
all the groups examined, the Americans seem to have assimilated the most in the
sense that they identify more strongly with the native Dutch people than with compatriots
in the Netherlands or in the country of origin.
Transnational involvement and social integration
© 2006 The Author(s) 297
Second, our respondents in general identify more with co-ethnics living in the
Netherlands than with people still living in the country of origin. This is a first indication
that transnational identifications are less common than phrases like ‘transnational
spaces’ or ‘transnational communities’ suggest. There is no cross-border
social space in the sense of migrants identifying more with compatriots living outside
the country than with Dutch residents (either native Dutch or compatriots living in the
Netherlands). On the contrary, respondents who identify strongly with their own
ethnic groups in the Netherlands are also more focused on compatriots in the country
of origin and vice versa.
4
Ethnic identifications are more prominent than transnational
identifications among immigrants.
Third, the weak identification with the diaspora is salient. This is most apparent
from the data in the bottom half of the table (data from the so-called circle scores and
total scores). Identification with compatriots in other countries is not as strong as
identification with compatriots living either in the Netherlands or in the country of
origin. The transnational identifications of most respondents primarily relate to the
country of origin: ‘bilateral’ transnational identifications prevail over ‘multilateral’ or
‘cosmopolitan’ transnational identifications.
We might conclude that transnational identifications are important, but less
important than sometimes assumed. It is certainly untrue that space has lost its meaning
in late-modern society and that contemporary migrants function in transnational
communities rather than in their country of residence, as adherents of transnationalism
sometimes argue (Faist 2000). Migrants primarily identify with their own ethnic
group, living either in the Netherlands or in their countries of origin. Of course, there
are still strong identifications with friends and relatives in the country of origin. But,
as we argued before, this is hardly something new in contemporary society. Furthermore,
identifications with the diaspora are relatively weak.
To what extent are these differences in transnational identifications (identifications
with compatriots living in their respective countries of origin)
5
between migrant
groups the result of different personal characteristics of migrants (including involvement
in transnational activities)? Again, we use regression analysis to answer this
question (Table 4).
Model 1 (in Table 4) shows almost no differences in the extent to which respondents
from various migrant groups identify with compatriots in the country of origin.
Respondents from almost all groups score equally high (or low) as the US reference
category. The only exceptions are respondents from former Yugoslavia. Their
identifications with compatriots in the country of origin are significantly stronger than
with all other groups. The picture changes very little when personal characteristics are
added to the analysis (Model 2). It turns out that transnational identifications are
negatively related to both length of stay in the Netherlands and to having a formal job.
In other words, more recent immigrants and respondents without a formal job have
stronger transnational identifications. It should be noted that the educational level has
no independent effect on transnational identifications. Stronger transnational identifications
are not related to lower levels of education but occur just as frequently among
the higher educated.
Erik Snel, Godfried Engbersen and Arjen Leerkes
298 © 2006 The Author(s)
Table 4: Identification with compatriots in the country of origin (linear
regression: coefficients are betas)
I II III
Country of origin (compared to USA)
Morocco –0.04** –0.03** –0.01**
Antilles –0.03** –0.05** –0.08**
Iraq –0.06** –0.02** –0.05**
Yugoslavia –0.20** –0.22** –0.07**
Japan –0.01** –0.01** –0.09**
Background characteristics
Sex –0.02** –0.00**
Length of stay –0.17** –0.22**
Age of migration –0.13** –0.01**
Education –0.09** –0.06**
Dutch nationality –0.04** –0.00**
Formal, paid job –0.16** –0.18**
Social benefit –0.06** –0.08**
Transnational activities
Professional economic activities –0.04**
Everyday economic activities –0.16**
Political activities –0.18**
Social-cultural activities in country of origin –0.19**
Social-cultural activities in The Netherlands –0.09**
Explained variation (R) –0.04** –0.15** –0.26**
* p < 0.05. ** p < 0.01.
Model 3 shows that transnational identifications are also related to involvement in
transnational activities. The more respondents are involved in transnational activities
(especially transnational political activities and socio-cultural activities in the country
of origin) the stronger their transnational identifications. After adding this factor to
the analysis, the initial stronger transnational identifications among Yugoslav respondents
disappear. In other words, the stronger transnational identifications among the
Yugoslavian group (observed in Model 1) can be fully explained by the Yugoslav respondents
being more involved in the relevant transnational activities (as we saw in the
previous section).
We may conclude that there are few differences in the strength of transnational
identifications between the various migrant groups. This also implies that migrant
groups that are often mentioned as poorly integrated into Dutch society (Moroccans,
Antilleans) have no stronger identifications with the country of origin than members
from other migrant groups. The strength of transnational identifications is primarily
Transnational involvement and social integration
© 2006 The Author(s) 299
related to the length of stay in the Netherlands, the formal job status and to involvement
in (some) transnational activities. Migrants with a shorter residence in the
Netherlands, without a formal job and who are more involved in transnational
activities have stronger transnational identifications.
Transnational involvement and integration
We now return to the central question of whether transnational activities and
identifications impede migrants’ successful integration into Dutch society.
Retaining ties with the country of origin or with internationally dispersed migrant
communities, financial investments in the country of origin and double citizenship
are often perceived as an impediment to integration. The implicit assumption is that
transnational involvement and immigrant integration are mutually exclusive
phenomena. However, one could also assume that migrants’ continuing transnational
involvement goes hand in hand with integration (Berry and Sam 1996;
Verkuyten 1999). In this section we examine the possible relationship between
transnational activities and identifications on the one hand and structural, social and
cultural aspects of integration on the other.
Transnational involvement and structural integration
The structural dimension of integration can be defined as the full participation of
migrants in the central societal institutions (especially the educational system and the
labour market). Structural integration of migrants can be measured in terms of their
level of education and labour market participation. We already observed that there is
only a weak empirical relationship between the transnational activities and identifications
of migrants on the one hand and the level of structural integration on the
other. Respondents with low levels of education are not more (or less) involved in
transnational activities nor do they identify more (or less) strongly with their countries
of origin than highly educated respondents. Working respondents do not participate
more (or less) in transnational activities than non-working respondents and social
security recipients. The only exception was that non-working respondents have
stronger transnational identifications than working respondents (Tables 2 and 4).
A closer look at the type of transnational activity and migrant group in the correlation
between transnational involvement and structural integration reveals interesting
differences. We have already mentioned that highly educated working respondents are
more involved in professional economic activities. Therefore we could say that professional
economic activities are related to (and probably presuppose) successful
structural integration. Migrants in established social positions who work for international
companies are in a position to develop such transnational business activities.
Furthermore, the correlation between transnational activities and identifications on
the one hand and the degree of structural integration on the other differs between one
migrant group and another (Table 5). US respondents, especially those who are well
integrated structurally, participate more in transnational business activities and
political activities. By contrast, Moroccan and Antillean respondents, namely the least
Erik Snel, Godfried Engbersen and Arjen Leerkes
300 © 2006 The Author(s)
integrated migrant groups in the structural sense, appear to have stronger transnational
identifications. The latter seems to confirm the fear that strong transnational
identifications might go hand in hand with (or even impede) structural integration into
Dutch society, at least as far as these groups are concerned. On the other hand, we
saw that poorly structurally integrated Moroccans and Antilleans do not participate
more in transnational activities than their better-integrated compatriots.
Table 5: Correlation between transnational activities and identities and
integration in the labour market by group of origin
Morocco
Dutch
Antilles Iraq
(former)
Yugoslavia Japan USA
Everyday economic activities –0.12** –0.06** –0.19** –0.25** –0.02* –0.11**
Professional economic activities –0.15** –0.14** –0.07** –0.18** –0.18* –0.40**
Political activities –0.28** –0.07** –0.20** –0.01** –0.02* –0.40**
Social-cultural activities in The Netherlands –0.04** 0.08** –0.37** –0.13** –0.09* –0.01**
Social-cultural activities in country of origin –0.17** –0.18** –0.28** –0.00** –0.13* –0.00**
Transnational ties general –0.29** –0.06** –0.32** –0.10** –0.08* –0.25**
Identification with country of origin –0.40** –0.38** –0.22** –0.11** –0.28* –0.27**
Identification with ethnic group in other
countries –0.32** –0.25** –0.14** –0.18** –0.07* –0.04**
* p < 0.05. ** p < 0.01.
To sum up, there is no unequivocal relation between migrants’ transnational
activities and structural integration. Transnational activities occur both among
migrants with good and with marginalized social positions (in terms of educational
level and labour market participation) in the host society. The explanation may be that
transnational activities result from two contrasting factors. First, it should be noted
that transnational activities always cost money and therefore are in part dependent on
the respondents’ financial means (Al-Ali et al. 2001). Unemployed respondents on
social security may well have a great need for transnational activities, but lack the
financial means with which to engage in them. That the least structurally integrated of
our respondents, the Moroccans and Antilleans, identify strongly with the country of
origin but do not develop many transnational activities seems to indicate this.
Conversely, employed migrants with the means to develop transnational activities
may not feel such a strong need to do so. The ultimate effect is that transnational
activities occur equally often among migrants with good economic positions in Dutch
society as among poorly integrated migrants. Second, one should realize that we
interviewed both migrants in marginal social positions who are involved in
transnational activities and employees from international companies (or their
partners). In Portes’s (2000) words, we are dealing here both with ‘transnationalism
from above’ and ‘transnationalism from below’. Both categories of respondents are
Transnational involvement and social integration
© 2006 The Author(s) 301
engaged in transnational activities in their own way. Participation in transnational
activities is not restricted to certain social classes of migrants.
Transnational involvement and social and cultural integration
Two indicators operationalize social and cultural integration: the degree to which
migrants have informal contact with native residents and the degree to which they
approve of typical Western or modern ideas and values on the other. Our survey
contained several questions about the respondents’ social networks. The cultural
aspect of integration is measured by the extent to which migrants identify with native
Dutch people. Are respondents who are involved in transnational activities and have
strong transnational identifications less integrated into Dutch society in a social and
cultural sense?
Table 6 shows the two indicators of social and cultural integration: the number of
native Dutch in the social network of respondents and the degree of identification
with native Dutch people. It shows that Moroccans and to a lesser extent Iraqis and
Yugoslavs have significantly fewer native Dutch people in their social networks than
the US reference category. Antilleans and Japanese hardly differ from the Americans.
In the table the influence of factors such as personal characteristics, involvement in
transnational activities and the strength of transnational identifications is already
taken in account. None of the factors has any influence on the number of native Dutch
in the social networks of migrants. The only factor of influence, again, is the length of
stay in the Netherlands. Migrants who live in the Netherlands longer have more
informal contact with the native Dutch. We may conclude that social integration
(informal contact with native residents) is not related to transnational involvement
(either measured by transnational activities or by transnational identifications).
The last column in Table 6 shows the extent to which the respondents from
various migrant groups identify with native Dutch people. We have already indicated
that US respondents are the most assimilated in Dutch society in the sense that they
identify strongly with native Dutch people. All other migrant groups identify
significantly less with the native Dutch than the US reference category. This is
especially true of Moroccan, Antillean and Japanese respondents, but to a lesser
degree also of Iraqi and Yugoslav respondents. Here also the length of stay in the
Netherlands appears to be an important factor. Migrants who live longer in the
Netherlands have stronger identifications with the native Dutch than migrants who
arrived more recently. All other of our respondents’ personal characteristics (educational
level, labour market participation, living on social benefit and having Dutch
citizenship) appear to be not related to cultural integration.
6
Our question was whether migrants’ cultural integration is related to transnational
involvement. The bottom half of Table 6 shows that this is the case. The more respondents
are involved in social-cultural activities in or with the country of origin
(such as visiting friends and relatives back home), the less they identify with native
Dutch people. More unexpected are the following outcomes. Participation in transnational
economic activities and identifications with compatriots in different countries
Erik Snel, Godfried Engbersen and Arjen Leerkes
302 © 2006 The Author(s)
(the internal diaspora) appears to be positively related to cultural integration. In other
words, respondents who are involved in transnational economic activities and who
identify strongly with compatriots in the international diaspora also identify more strongly
with native Dutch people. Particularly the last outcome indicates that migrants appear
to be quite able to live in two different worlds. Identifications with internationally dispersed
ethnic communities and with native Dutch people do not rule each other out.
Table 6: Determinants of social-cultural integration (number of native Dutch in
the social network and the degree of identification with native Dutch) (multiple
regression in betas)
Number of native
Dutch in social
network
Identification with
native
Dutch
Group of origin (compared to the USA)
Morocco –0.40** –0.35**
Antilles –0.18** –0.41**
Iraq –0.23** –0.47**
Yugoslavia –0.23** –0.23**
Japan –0.10** –0.36**
Background characteristics
Sex (male=1) –0.05** –0.04**
Length of stay –0.32** –0.18**
Age of migration –0.01** –0.08**
Education –0.08** –0.03**
(also) Dutch nationality –0.03** –0.03**
Formal, paid job –0.13** –0.03**
Social benefit –0.03** –0.03**
Transnational activities
Everyday economic activities –0.02** –0.07**
Professional economic activities –0.08** –0.12**
Political activities –0.08** –0.01**
Social-cultural activities in/with country of origin –0.08** –0.19**
Social-cultural activities in The Netherlands –0.04** –0.09**
Transnational identities
Identification country of origin –0.08** –0.01**
Identification Diaspora –0.10** –0.24**
Explained variation (R) –0.28** –0.27**
p < 0.05. ** p < 0.01.
Transnational involvement and social integration
© 2006 The Author(s) 303
Discussion
Many studies on transnationalism or transmigrants are either theoretical or based on
qualitative empirical research, with the quantitative work being currently limited to
the USA (Guarnizo et al. 2003; Portes et al. 2002). We offer a quantitative examination
of the extent to which migrants from a variety of countries participate in
transnational activities and have transnational identifications. Our study is based on a
survey of 300 immigrants from a variety of countries living in the Netherlands. Some
respondents come from the Netherlands’ classical ethnic minorities (Moroccans,
Antilleans), others are migrants from typical refugee countries (Iraq, former Yugoslavia),
and others again are new labour migrants (or their partners) from prosperous
Western countries like Japan and the USA. Our central research question was whether
or not transnational activities and identifications impede the successful integration of
migrants into Dutch society.
We found that transnational activities constitute a substantial part of the lives of
migrants in the Netherlands. Transnational activities are not confined to particular
migrant groups or to a particular type of migrant. It turned out that transnational
identifications weaken the longer the migrants live in the Netherlands. However,
involvement in transnational activities hardly diminishes with increased length of
stay, which suggests that these occur in part for reasons other than transnational
identification (for example, because of familial obligations). Transnational activities
are to a large extent socio-cultural (family visits, contacts with relatives in the
country of origin). Many migrants (especially the respondents from Iraq, former
Yugoslav and Morocco) also transfer money to their countries of origin.
Respondents from refugee countries are also relatively more often involved in
transnational political activities. Professional economic activities were rare among
our respondents and were largely limited to the American group. Transnational
identifications – when migrants identify more closely with friends and relatives
outside the Netherlands (either in the country of origin or elsewhere) than with
Dutch residents (either among their own ethnic group or native Dutch persons) –
seemed to be less important. Our respondents in general identify more strongly with
compatriots living in the Netherlands than with compatriots living abroad.
Americans are a notable exception. They identify more strongly with native Dutch
people and with compatriots back home than with fellow Americans living in the
Netherlands. Generally, involvement in transnational activities goes hand in hand
with strong identifications with relatives/compatriots in the country of origin. The
identifications with migrants living elsewhere were comparatively less important.
No empirical evidence was found for the significance of internationally dispersed
migrants (diaspora). In other words, ‘bilateral’ transnational identifications prevail
over ‘multilateral’ or ‘cosmopolitan’ transnational identifications.
Are migrants’ transnational activities and identifications an impediment to their
integration into Dutch society? Or does strong transnational involvement not necessarily
obstruct immigrant integration? Can migrants retain strong ethnic and transnational
ties and be well integrated into Dutch society at the same time? The
Erik Snel, Godfried Engbersen and Arjen Leerkes
304 © 2006 The Author(s)
outcomes of our survey show that neither position can be maintained as such.
Generally, our findings support the assumption that transnational involvement does
not necessarily impede integration. We saw for instance that, of all the examined
migrant groups, the former Yugoslavs and Americans are most involved in transnational
activities. In the Dutch context, these are certainly not the least integrated
migrant groups. As regards identification with the country of origin, we saw few
differences between the various groups (except that former Yugoslavs identify more
strongly with the country of origin than other groups). If transnational involvement
constitutes an impediment for immigrant integration one would expect the strongest
transnational involvement among Moroccans and Antilleans, the migrant groups that
are often mentioned as being poorly integrated. However, Moroccans and Antilleans
are neither more involved in transnational activities nor identify more strongly with
the country of origin than the other groups.
Second, also at the individual level, there is no indication that transnational
involvement is related to the various factors that together determine the degree of
integration. There is no correlation whatsoever between the transnational activities of
migrants and their social positions. More highly educated respondents and respondents
with jobs engage in just as many transnational activities (though different ones)
as the poorly educated, unemployed respondents on social security. We also found no
correlation between identification with the country of origin and migrants’ levels of
education. What we did find was that working respondents identify significantly less
with the country of origin than non-working respondents. There was also hardly any
relation between transnational activities and identifications on the one hand and the
migrants’ social and cultural integration on the other. Transnational involvement does
not go hand in hand with less (or more) personal contact or identification with native
Dutch people. Respondents involved in transnational business activities and with
strong identifications with international diasporas tend to identify even more strongly
with native Dutch people.
7
We should add, however, that the latter two phenomena do
not occur very often. Only strong socio-cultural activities in or with the country of
origin (like visiting relatives) appear to be related to less identification with native
Dutch people.
All these findings largely support the assumption that transnational activities and
identifications do not need to constitute an impediment to integration. However, one
outcome does point in another direction, showing that the concern expressed by some
authors and politicians that transnational involvement of migrants impedes their
integration into Dutch society is not entirely unjustified. Groups that are (perceived to
be) culturally different from mainstream society appear to have a harder time combining
transnational involvement with cultural integration. For example, among
Moroccans and Antilleans – migrant groups that also have the weakest labour market
positions – we observed that strong identifications with the country of origin go hand
in hand with poor structural and cultural integration. Our analysis does not explain
what the causes are. However, our hypothesis is that migrants in weak labour market
positions tend to withdraw from Dutch society and feel more strongly related with the
country of origin (Engbersen et al. 1993). The availability of modern means of com-
Transnational involvement and social integration
© 2006 The Author(s) 305
munication reinforces this process of social seclusion. Although, generally speaking,
transnational involvement does not constitute an impediment to successful integration
into Dutch society, the situation for specific marginalized groups of migrants may
actually be quite different.
Erik Snel and Godfried Engersen
Department of Sociology
Erasmus University Rotterdam
PO Box 1738, 3000 DR Rotterdam
The Netherlands
(also affiliated to the Amsterdam School of Social Research)
Arjen Leerkes
Amsterdam School of Social Research
University of Amsterdam
Kloveniersburgwal 48
1012 CX Amsterdam
The Netherlands
Acknowledgements
The authors thank Prof. M. Verkuyten for his advice concerning the measurement of ethnic
identifications. We also thank Ms Afke Weltevrede and Ms Masja van Meeteren for their
assistance and comments on earlier versions of this article. We thank Dr Alisdair Rogers and
the anonymous reviewers of Global Networks: A Journal of Transnational Affairs for their
insightful comments on the draft version of this article. We thank Ms Annika Kristof for
improving our English.
Notes
1. The former multicultural orientation of Dutch politics has recently come under pressure.
One example is that the Netherlands has explicitly said farewell to multiculturalism as a
cornerstone of Dutch integration policy (Entzinger 2003; Snel 2003; Snel et al. 2004: 2).
2. The group dimension of identification is measured by five questions: I feel strongly related
with …, I have an emotional bond with …, I am involved with …, I am proud off …, I feel
personally attacked when other people are negative about … . Respondents were asked to
position themselves vis-à-vis native Dutch people (Alpha 0,75), compatriots living in the
Netherlands (Alpha 0,80), compatriots in the country of origin (Alpha 0,81) and
compatriots in the international diaspora (Alpha 0,79). The normative dimension of
identification is measured by four questions: to determine my own norms and values I take
the opinions of … into account, about important things in life I think the same as …, to
know what is good and valuable I look at …, I compare my own ideas about life
(upbringing, male-female-relationships) especially with the opinions of … . Again
respondents were asked to position themselves vis-à-vis native Dutch people (Alpha 0,79),
compatriots living in the Netherlands (Alpha 0,84), compatriots in the country of origin
Erik Snel, Godfried Engbersen and Arjen Leerkes
306 © 2006 The Author(s)
(Alpha 0,86) and compatriots in the international diaspora (Alpha 0,84). The high values of
the alphas legitimate the construction of two scales.
3. The total score is each time the average of the circle score, on the one hand, and the
average score on the basis of the 40 statements, on the other. This score probably gives the
best summary of the strength or weakness of the different group identifications.
4. There is a strong connection between the degree of the migrants’ identification with their
own groups in the Netherlands and their identification with compatriots in the country of
origin (r = 0.64).
5. A similar analysis using the degree of identification with compatriots elsewhere as the
dependent variable does not lead to substantially different outcomes.
6. This outcome contradicts the findings of other studies about integration of migrants and socalled
ethnic minorities in the Netherlands. Relevant studies invariably show strong
relations between structural and cultural aspects of integration (Dagevos 2001; Odé 2002).
Our findings may differ because we surveyed different migrant groups, for instance
Japanese migrants who are integrated in a structural sense (measured by educational level
and labour market participation), but not in a cultural sense.
7. The reason is probably that especially American respondents are involved in these
transnational business activities and they are rather close to the native Dutch.
References
Al-Ali, N., R. Black and K. Koser (2001) ‘Refugees and transnationalism: the experience of
Bosnians and Eritreans in Europe’, Journal of Ethnic and Migration Studies, 27, 615–34.
Alba, R. and V. Nee (1997) ‘Rethinking assimilation theory for a new era of immigration’,
International Migration Review, 31, 26–74.
Bauman, Z. (2004) Identity, Cambridge: Polity Press.
Berry, J. W. and D. L. Sam (1996) ‘Acculturation and adaptation’, in J. W. Berry, M. H. Segall
and C. Kagitcibasi (eds) Handbook of cross-cultural psychology, volume 3, Social behavior
and applications, Boston: Allyn & Bacon, 291–326.
Bryceson, D. and U. Vuorela (eds) (2002) The transnational family: new European frontiers
and global networks, Oxford/New York: Berg.
Dagevos, J. (2001) Perspectief op integratie: over de sociaal-culturele en structurele integratie
van etnische minderheden in Nederland, Den Haag: Wetenschappelijke Raad voor het
Regeringsbeleid.
Douglas, M. (1978) Cultural bias, London: Royal Anthropological Institute of Great Britain
and Ireland (Occasional Paper 35).
Douglas, M. (1996) Thought styles, London: Sage Publications.
Engbersen, G. (2003) ‘Spheres of integration: towards a differentiated and reflexive ethnic
minority policy’, in R. Sackmann, B. Peters and T. Faist (eds) Identity and integration:
migrants in Western Europe, Aldershot/Burlington: Ashgate, 59–76.
Engbersen G., K. Schuyt, J. Timmer and F. van Waarden (1993) Cultures of unemployment: a
comparative look at long-term unemployment and urban poverty, Oxford: Westview Press.
Entzinger, H. (2003) ‘The rise and fall of multiculturalism: the case of the Netherlands’, in C.
Joppke and E. Morawska (eds) Toward assimilation and citizenship: immigrants in liberal
nation-states, Basingstoke/New York: Palgrave Macmillan, 59–86.
Esser, H. (2003) ‘What substance is there to the term “Leitkultur”?’, in R. Cuperus, K. Duffek
and J. Kandel (eds) The challenge of diversity: European social democracy facing
migration, integration and multiculturalism, Insbruck: Studien Verlag, 47–58.
Transnational involvement and social integration
© 2006 The Author(s) 307
Faist, T. (2000) The volume and dynamics of international migration and transnational social
spaces, Oxford: Oxford University Press.
Gerbner, D. (2000) ‘Theories and lives: transnationalism and the conceptualization of
international migrations to the United States’, IMIS-Beiträge, 15, 31–53.
Glick Schiller, N. and G. E. Fouron (1999) ‘Terrains of blood and nation: Haitian transnational
social fields’, Ethnic and Racial Studies, 22, 340–66.
Glick Schiller, N., L. Basch and C. Szanton-Blanc (1992) Towards a transnational perspective
on migration: race, class, ethnicity, and nationalism reconsidered, New York: New York
Academy of Sciences.
Gordon, M. (1964) Assimilation in American life: the role of race, religion and national
origins, New York: Oxford University Press.
Guarnizo, L. E. and M. P. Smith (1998) ‘The locations of transnationalism’, in M. P. Smith and
L. E. Guarnizo (eds) Transnationalism from below, vol. 6: Comparative urban and
community research, New Brunswick, NJ: Transaction, 3–34.
Guarnizo, L. E., A. Portes and W. Haller (2003) ‘Assimilation and transnationalism: determinants
of transnational political action among contemporary migrants’, American Journal
of Sociology, 108, 1211–48.
Jenkins, R. (1996) Social identity, London: Routledge.
Joppke, C. and E. Morawska (2003) ‘Integrating immigrants in liberal nation-states: policies
and practices’, in C. Joppke and E. Morawska (eds) Toward assimilation and citizenship:
immigrants in liberal nation-states, Basingstoke/New York: Palgrave Macmillan, 1–36.
Levitt, P. (2003) ‘Keeping feet in both worlds: transnational practices and immigrant
incorporation in the United States’, in C. Joppke and E. Morawska (eds) Toward
assimilation and citizenship: immigrants in liberal nation-states, Basingstoke/New York:
Palgrave Macmillan, 177–94.
Levitt, P., J. De Wind and S. Vertovec (2003) ‘International perspectives on transnational
migration: an introduction’, International Migration Review, 37, 565–75.
Morawska, E. (2003) ‘Immigrant transnationalism and assimilation: a variety of combinations
and the analytic strategy it suggests’, in C. Joppke and E. Morawska (eds) Toward
assimilation and citizenship: immigrants in liberal nation-states, Basingstoke/New York:
Palgrave Macmillan, 133–76.
Odé, A. (2002) Ethnic-cultural and socio-economic integration in the Netherlands: a comparative
study of Mediterranean and Caribbean minority groups, Assen: Koninklijke Van Gorcum.
Pahl, R. E. (1987) ‘Does jobless mean workless? Unemployment and informal work’, Annals of
the American Academy of Political and Social Sciences, 493, 36–46.
Portes, A. (2000) ‘Globalization from below: the rise of transnational communities’, in D. Kalb,
M. van der Land, R. Staring, B. van Steenbergen and N. Wilterdink (eds) The ends of
globalization: bringing society back, Lanham: Rowman & Littlefield Publishers, 253–70.
Portes, A., L. Guarnizo and P. Landolt (1999) ‘The study of transnationalism: pitfalls and
promise of an emergent research field’, Ethnic and Racial Studies, 22, 217–37.
Portes, A., W. J. Haller and L. E. Guarnizo (2002) ‘Transnational entrepreneurs: an alternative
form of immigrant economic adaptation’, American Sociological Review, 67, 278–98.
Snel, E (2003) De vermeende kloof tussen culturen, Enschede: Universiteit Twente.
Snel, E., J. de Boom and G. Engbersen (2004) Migration, immigrants and policy in the
Netherlands: report for the continuous reporting system on migration (SOPEMI) of the
Organisation of Economic Co-operation and Development (OECD), Rotterdam: Risbo.
Staring, R. (2001) Reizen onder regie: het migratieproces van illegale Turken in Nederland,
Amsterdam: Het Spinhuis.
Erik Snel, Godfried Engbersen and Arjen Leerkes
308 © 2006 The Author(s)
Van Tillaart, H. (2001) Monitor etnisch ondernemerschap 2000: zelfstandig ondernemerschap
van etnische minderheden in Nederland in de periode 1990–2000. Nijmegen: ITS.
Verkuyten, M. (1999) Etnische identiteit: theoretische en empirische benaderingen,
Amsterdam: Het Spinhuis.
Vermeulen, H. and R. Penninx (eds) (2000) Immigrant integration: the Dutch case,
Amsterdam: Het Spinhuis.
Vertovec, S. (1999) ‘Conceiving and researching transnationalism’, Ethnic and Racial Studies,
22, 447–62.
Vertovec, S. (2001) ‘Transnationalism and identity’, Journal of Ethnic and Migration Studies,
27, 573–82.
Waldinger, R. and D. Fitzgerald (2004) ‘Transnationalism in question’, American Journal of
Sociology, 109, 1177–95.
Wildavsky, A. (1987) ‘Choosing preferences by constructing institutions: a cultural theory of
constructing institutions’, The American Political Science Review, 81, 4–21.
Wildavsky, A. (1989) ‘Frames of reference come from cultures: a predictive theory’, in M.
Freilich (ed.) The relevance of culture, New York: Bergin & Garvey Publishers, 58–74.