Content uploaded by Erik Snel
Author content
All content in this area was uploaded by Erik Snel on May 19, 2016
Content may be subject to copyright.
© 2016 The Author(s)
Global Networks © 2015 Global Networks Partnership & John Wiley & Sons Ltd
1
Reactive transnationalism: homeland involvement
in the face of discrimination
ERIK SNEL,
*
MARGRIETHA ’T HART
†
AND
MARIANNE
VA N BOCHOVE
‡
*
(corresponding author)Erasmus University Rotterdam,
Department of Sociology, P.O. Box 1738,
3000 DR Rotterdam, the Netherlands
snel@fsw.eur.nl
†
formerly at Erasmus University Rotterdam,
Department of Sociology, P.O. Box 1738,
3000 DR Rotterdam, the Netherlands
mej.thart@gmail.com
‡
Erasmus University Rotterdam,
Institute of Health Policy and Management, P.O. Box 1738,
3000 DR Rotterdam, the Netherlands
vanbochove@bmg.eur.nl
Abstract In this article we examine whether migrants’ perceived discrimination in
the country of settlement leads to an increase of their transnational involvement. So
far, this so-called ‘reactive transnationalism’ has not been studied extensively. Based
on literature on discrimination and transnationalism, reactive transnationalism is
expected to be most prominent among socioeconomically successful migrants,
particularly among males and those who consider themselves Muslims. Our research
among middle-class migrants in Rotterdam, the Netherlands, indeed shows that the
more respondents experienced discrimination, the more transnationally involved they
are, both regarding transnational identifications and transnational activities. While
no gender difference was found regarding reactive transnational activities, for
women perceived discrimination proves to lead to stronger instead of weaker
transnational identifications than for men. The fact that no difference was found
between Muslim and non-Muslim respondents regarding reactive transnationalism
suggests that, despite heated public debates about ‘Islam’, in the Netherlands, ethnic
divides – being considered as ‘Dutch’ or ‘non-Dutch’ – are even more prominent than
religious ones.
Keywords
DISCRIMINATION, TRANSNATIONAL ACTIVITIES, TRANSNATIONAL MIGRANTS,
TRANSNATIONAL IDENTIFICATIONS
, TRANSNATIONALISM
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
2
© 2016 The Author(s)
I might return to Morocco. It is not always nice to live in the Netherlands if
you are a foreigner, and especially not if you are a Muslim. We are perceived
as second-class citizens today.
Fatiha is a 33-year-old woman who was born in the Netherlands after her parents moved
there from Morocco in the early 1970s. She works in Rotterdam, as the spokesperson
for one of the aldermen. Although she is highly educated, speaks Dutch fluently, and
has a successful career, she does not always feel accepted by mainstream society,
because of her Moroccan background and the fact that she is a Muslim. Since right-
wing politicians such as Pim Fortuyn and Geert Wilders gained popularity in the early
years of the twenty-first century, she has more often felt discriminated against. Because
of this lack of acceptance, she is considering building a new life in Morocco, even
though she has never lived there.
Fatiha was one of the respondents who was interviewed for a project on Trans-
nationalism and Urban Citizenship (the TUC project) in Rotterdam, the Netherlands,
between 2007 and 2008 (van Bochove 2012). She was not the only one who talked
about experiencing discrimination and dreaming about ‘returning home’. Several of the
middle-class men and women with a Moroccan, Turkish, or Surinamese background
expressed similar opinions. Although we have not scrutinized these findings before,
they suggest that (first- and second-generation) migrants who feel discriminated against
become more oriented towards their (parents’) country of origin. This phenomenon of
what Itzigsohn and Saucedo (2002, 2005) called ‘reactive transnationalism’ has not
been studied extensively, in contrast with other types of reactions on perceived
discrimination.
It is a well-established outcome of social science research that (perceived) discrim-
ination and social exclusion often result in enhanced in-group identifications and
affiliations. When members of ethnic minorities feel rejected by the majority popu-
lation, they tend to commit themselves more to their own ethnic community and its
institutions, varying from migrant organizations and religious institutions to virtual
social networks of co-ethnics (Verkuyten 2010; Verkuyten and Yildiz 2007). Portes
and Rumbaut (2001) call this phenomenon ‘reactive ethnicity’. More recently, Dutch
researchers used the term ‘reactive religiosity’ to describe the enhanced religious iden-
tifications in the face of discrimination among young Muslims in the Netherlands and
other Western countries (Maliepaard and Phalet 2013; Voas and Fleischmann 2012).
In this article, we further examine the issue of reactive transnationalism, based on
the TUC survey among 225 first- and second-generation migrants in Rotterdam.
Typical of our survey is that all the respondents have a middle-class social background.
We focus on both transnational activities and identifications. Our main question is to
what extent perceived discrimination by migrants leads to more involvement in trans-
national activities and stronger transnational identifications. The qualitative data of the
TUC project suggest that both males and females are subject to reactive trans-
nationalism. However, Itzigsohn and Saucedo (2005) only found proof for the existence
of reactive transnationalism among male migrants. In this article, we will further
examine the existence of gender differences. Moreover, since our qualitative data – as
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
3
well as other studies on the Dutch case – suggest that particularly Muslims experience
discrimination and develop ‘reactive’ responses, we also take the role of religion into
consideration. In the following, we will first discuss relevant existing literature on
transnationalism and perceived discrimination and explain how our approach adds to
this. We will then explain the data and methods used in our analyses, before presenting
our results. In the closing section, we discuss the implications of our findings for further
theorization and empirical research on reactive transnationalism.
Reactive transnationalism: some theoretical reflections
In their work on second-generation immigrant youth in the USA, Portes and Rumbaut
(2001: 148) coined the notion of ‘reactive ethnicity’. Reactive ethnicity refers to the
heightened group consciousness, ethnic group solidarity and political mobilization
among migrants as a result of perceived prejudices, hostility or social exclusion in the
host society. The notion of ‘reactive ethnicity’ is based on research in social
psychology, which showed that individuals who experience discrimination will identify
with their in-group more strongly to buffer the negative consequences that the perceived
discrimination has for their self-esteem (Branscombe et al. 1999). Portes and Rumbaut
emphasize that a heightened in-group consciousness among migrants does not result
from reluctant assimilation, but from their negative experiences in the host society; it
is ‘made-in-America’ (Portes and Rumbaut 2001: 284).
Building on the work of Portes and Rumbaut, various scholars have recently
studied the phenomenon of ‘reactive religiosity’ among adolescents with a Muslim
background in the Netherlands and other west-European countries. Concurrently,
prejudices and negative stereotypes regarding Muslims are widespread in Dutch
society, and a majority of the Dutch public believes that Islam and the western way of
life are incompatible (Huijnk and Dagevos 2012; cf. Vasta 2007). Several studies show
how perceived discrimination, exclusion and negative stereotypes reinforce young
Muslims’ religious identifications and affiliations (Maliepaard and Phalet 2013;
Verkuyten and Yildiz 2007; Voas and Fleischmann 2012). Verkuyten and Yildiz
(2007) found that perceived discrimination is related to enhanced Muslim identifi-
cation among Turkish–Dutch Muslims. Similarly, Fleischman et al. (2011) concluded
that perceived discrimination is associated with enhanced willingness to engage in
political action to defend Islam among European-born Moroccan and Turkish Muslims.
Maliepaard and Phalet (2013) found that perceived discrimination is related to both
higher levels of religious identifications and, indirectly, to religious practices such as
attending the mosque or participating in Ramadan. These studies confirm reactive
religiosity – prejudices and perceived discrimination do not result in distancing from
the stigmatized identity, but rather in strengthening it (see, for an extensive overview,
Voas and Fleischman 2012).
Our study extends this line of research by examining whether perceived discrimin-
ation of migrants also results in enhanced transnational involvement, conceptualized as
migrants’ involvement in their (parents’) country of origin and consisting of both
transnational activities and transnational identifications (cf. Snel et al. 2006). So far,
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
4
© 2016 The Author(s)
only a limited number of studies have elucidated transnationalism in relation to
perceived discrimination.
Itzigsohn and Saucedo (2002, 2005) tested what they call the ‘reactive trans-
nationalism hypothesis’ among Latin-American migrants in the USA and paid special
attention to gender differences. They found positive effects of perceived discrimination
on participation in both economic and sociocultural transnational activities, but found
these effects only for men and not for women (Itzigsohn and Saucedo 2005). The
authors explain this outcome by suggesting that female migrants are more engaged in
social life in the receiving society. Women invest available resources primarily in their
own households in the country of settlement and invest in transnational activities only
when they have resources to do so. Male migrants, on the other hand, may be more
susceptible to the loss of social status in the process of migration. As a result, they
develop a stronger orientation to the community of origin when experiencing prejudices
and discrimination. In short, men’s transnationalism ‘may be more of the “reactive”
type’ (Itzigsohn and Saucedo 2005: 900, 916).
More recently, reactive transnationalism was also studied in a European context.
While Carling and Hoelscher (2013: 950) found no effect of perceived discrimination
of respondents with an immigrant background on transnational activities (in their case
sending remittances to the home country) in Norway, Green et al. (2014: 97) found a
significant positive association between the transnational political behaviour of
Senegalese migrants in France and Switzerland and perceived discrimination. Existing
research on reactive transnationalism has yielded interesting results, but various aspects
of the phenomenon so far have not been studied extensively and many questions remain
unanswered. Our study adds to existing ones in three respects.
First, our measurement of transnational involvement is more comprehensive than in
previous studies on reactive transnationalism. Instead of only focusing on political
activities (cf. Green et al. 2014) or on a combination of economic and sociocultural
activities (Itzigsohn and Saucedo 2002, 2005), we include all these types. Moreover,
next to activities, we also take into account transnational identifications, defined as the
extent to which migrants identify with the country of origin. As Tsuda (2012: 642)
argues, transnational identities are often mentioned in the literature but there has been
relatively little systematic analysis of them (Snel et al. 2006; van Bochove et al. 2010
are exceptions). Tsuda also relates the topic of transnational identifications to Portes
and Rumbaut’s notion of reactive ethnicity. Identifications with the sending or the
receiving country are often seen as a zero-sum relationship. Immigrants who have
negative experiences in the host country because of marginalization and experienced
discrimination will strengthen their transnational (or transborder) affiliation to their
home country and reduce their possible identification with the home country. Previous
research shows that discrimination experiences indeed lead to reduced affiliation with
the host country (Verkuyten and Yildiz 2007). Here, we examine whether perceived
discrimination also results in stronger transnational identifications.
Second, in addition to the attention that Itzigsohn and Saucedo (2005) paid to
gender, we also study the role of religion in reactive transnationalism. While recent
studies showed that in the context of heated public and political debates about Islam,
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
5
many migrants with a Muslim background develop a stronger religious engagement
(Maliepaard and Phalet 2013), we study whether migrants who consider themselves
Muslims and who feel discriminated against become more strongly oriented towards
their countries of origin.
Third, we focus on migrants with a middle-class status, who are expected to be more
receptive to experiencing discrimination than migrants with a lower socio-economic
status – the so-called ‘integration paradox’. This paradox states that migrants with a
higher socio-economic status often perceive more discrimination because they have
more contact with the native population and expect more acceptance by dominant
society (Ten Teije et al. 2010; van Doorn et al. 2013). This makes middle-class migrants
a strategic group for studying whether perceived discrimination results in enhanced
transnational involvement.
Based on these theoretical reflections, we formulate three hypotheses to be tested in
our analyses. These are hypothesis 1: perceived discrimination leads to enhanced
transnational activities and transnational identifications (‘reactive transnationalism
hypothesis’); hypothesis 2: the reactive transnationalism hypothesis only applies to
male migrants; and hypothesis 3: the reactive transnationalism hypothesis only applies
to migrants who consider themselves Muslims and find religion important in their lives.
Data and methods
Transnational and Urban Citizenship survey
The data are based on the Transnational and Urban Citizenship (TUC) survey
1
conducted
among 225 middle-class migrants in the Dutch city of Rotterdam (van Bochove 2012;
van Bochove et al. 2010). With more than 600,000 residents, Rotterdam is the second
largest city of the Netherlands. Rotterdam has a highly ethnically diverse population, for
instance reflected in the approximately 170 different nationalities that are counted there
(COS 2011). In this study, we focus on members of the three largest minority categories
– first- and second-generation migrants with a Surinamese (9 per cent of the Rotterdam
population), Turkish (8 per cent), and Moroccan background (6 per cent) (COS 2011).
Surinam is a former Dutch colony in South America. After the country gained inde-
pendence in 1975, many Surinamese opted for Dutch citizenship and moved to the
Netherlands. The first wave of Turks and Moroccans came in the 1960s and 1970s as
so-called ‘guest workers’. Migration from Turkey and Morocco continued after the for-
mal recruitment of guest workers ended in the mid-1970s because of extensive family
reunion and marital migration. Although a particularly large number of first-generation
Turks and Moroccans were of low socio-economic status, Dagevos et al. (2006) show
the rise of a middle class among the minorities in the Netherlands. According to their
study, one-third of the Surinamese population in the Netherlands belong to the middle
class. The share of Turkish and Moroccan middle-class migrants is lower (17 and 14
per cent respectively), but their numbers are also growing: between 1991 and 2005, the
middle class among these groups doubled in size (Dagevos et al. 2006: 123). Although
comparable figures are unavailable at the city level, Rotterdam, like other large Dutch
cities, is an important location for the formation of an immigrant middle class.
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
6
© 2016 The Author(s)
Following Dagevos et al. (2006), we consider people as ‘middle class’ when they
meet at least one of the following criteria:
they have a job that requires at least intermediate vocational education;
they have worked as an independent entrepreneur for at least a year; and
they earn a gross annual income above the national median income.
2
In addition, we restricted the second criteria to entrepreneurs who work in producer
services, and we only selected respondents who worked more than 20 hours a week.
In the absence of reliable administrative data on middle-class migrants in Rotter-
dam, we used snowball sampling and chain referral sampling to select respondents.
Interviewers recruited the first respondents within their own social networks (with the
restriction that they did not know them personally), through businesses, governmental
institutions, and community organizations. The initial respondents were asked if they
knew any other potential respondents, up to a maximum of three. This method of
sampling implies that there is no guarantee that the results are representative for the
migrant middle class in Rotterdam. However, the aim of this article is to find a relation-
ship between perceived discrimination and transnational involvement, rather than
presenting representative figures about, for example, the discriminatory experiences of
the migrant middle class in Rotterdam.
The face-to-face interviews took about one-and-a-half hours and were conducted in
Dutch. The TUC survey included both open- and closed-ended questions. The closed-
ended questions, such as ‘how often do you read homeland newspapers?’ or ‘how often
do you feel discriminated against?’ mainly provided quantitative information – that is
the frequency or extent to which the respondents participate in activities or experience
certain feelings. In the open-ended questions, respondents were usually asked to
motivate their answers, particularly in the case of their (homeland) identifications.
Although this article is mainly based on quantitative data, we will turn to the qualitative
material to explain further or illustrate these findings.
Concepts and measures
The main dependent variables of our analyses are transnational activities and trans-
national identifications.
Transnational activities: following Portes et al. (1999), we examined three different
types of transnational activities – economic, political, and sociocultural. Economic
transnational activities are the economic endeavours of migrants that take place in both
the origin and destination countries. The classic example of economic transnationalism
is the immigrant entrepreneur who profits from cheap products or cheap labour from
the country of origin or from co-ethnics in the destination country as guaranteed
customers (Rusinovic 2006). However, since relatively few migrants are actually active
as entrepreneurs (Snel et al. 2006), we look at everyday transnational economic activi-
ties, such as sending remittances or donating to charities related to the origin country.
Political transnationalism refers to the degree to which migrants keep up to date with
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
7
homeland politics and participate in political or civic activities directed at the country
of origin (Snel et al. 2006; van Bochove et al. 2010). The political transnational
activities that we study are reading newspapers, participating in social and political
actions, and performing voluntary work for homeland political or social organizations.
Finally, sociocultural transnational activities include visiting the origin country or
having close contacts with friends and family there (all items in Table A1 in the
appendix). We computed indices of the three different types of transnational activities.
As the range of the answer categories of the items varied, the items were standardized.
As a result, the mean of the items is 0, and the standard deviation is 1. Then, we com-
puted a scale of the means of the standardized items. In addition, we combined all
activities in one index measuring transnational activities.
Transnational identifications: our understanding of transnational identification is
based on elementary notions from literature on identity formation (Phinney 1990;
Verkuyten 2005; Verkuyten 2010). The starting point is that people define themselves
in relation to their social environment: identity is about ‘what you have in common with
some people and what differentiates you from others’ (Weeks 1990: 88). Social or
ethnic identity starts with self-identification (‘to what category do I belong?’), but
crucial is how important it is for individuals to belong to this category and their emo-
tional appreciation of it. We asked respondents about their opinions on several
(positive) statements about their country of origin (cf. Phinney 1990); for instance, we
asked whether they are proud of their country of origin, feel involved in it, and feel hurt
when someone says something negative about it.
3
Respondents could indicate on a 5-
point scale (from 0= totally disagree to 4 = totally agree) to what extent they agree with
these statements. Factor analysis revealed there was only one dimension. The mean of
the answers on these items were combined into a reliable scale for transnational
identifications (Cronbach’s α 0.822).
Perceived discrimination: we are interested in personal perceived discrimination
rather than perceived group rejection. Perceived discrimination was measured in a two-
way step. First, respondents could indicate whether they experienced discrimination.
They could answer (0) never, (1) sometimes, (2) regularly, and (3) often. If they
experienced discrimination, they were asked on what ground they felt discriminated
against most often: (1) gender, (2) origin, (3) religion, (4) skin colour, or (5) other.
Respondents were then asked whether they also felt discriminated on the basis of
another feature (with the same answer categories). We combined these three questions
into one variable with four values: the respondent never, sometimes, regularly, or often
perceived discrimination based on origin, religion and/or skin colour (discrimination
based on gender or other grounds was coded as never feeling discriminated against).
Interaction variables: we expect that the reactive transnationalism hypothesis will
only apply to men (cf. Itzigsohn and Saucedo 2005) and to respondents who consider
themselves Muslims. Therefore, we included gender (0=male, 1= female) in the
analyses. In the questionnaire, respondents were asked about their religion. Further-
more, they were asked how important religion was in their life. They could answer that
it was (1) not important at all to (5) extremely important. For the respondents who had
indicated that they considered themselves Muslims, we used the information on the
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
8
© 2016 The Author(s)
question how important their religion is by combining the answer categories (1) not
important at all and (2) not important, and combining the answer categories (4) very
important and (5) extremely important. In this way, the following variable was con-
structed: (0) not Muslim, (1) being Muslim and considering religion as not important,
(2) being Muslim and considering religion as fairly important, and (3) being Muslim
and considering religion as very important.
Control variables: we included the following control variables in the analyses:
ethnicity (0= Turkish, 1 =Moroccan, 2= Surinamese), generation, and educational level.
Following Kasinitz et al. (2008), we define migrants of the 1.5th generation when they
were younger than 12 years when arriving in the Netherlands, and of the second gener-
ation when they were born in the Netherlands. We define migrants as belonging to the
first generation when they arrived in the Netherlands when they were 12 years or older.
Respondents were asked what their highest obtained educational level was (in either
the origin or the host country). In this article, we distinguish between (1) higher
education, (2) higher secondary education, and (3) lower secondary education and
lower levels of education obtained.
Missing values: Tables 2 and 3 present the descriptive statistics. Certain variables,
such as religion, have some missing values. As the total number of missing values is
less than 5 per cent, we excluded the respondents with missing values from the analyses.
The final N is 223.
Reactive transnationalism among middle-class migrants
Exploring transnational involvement and perceived discrimination
Table A2 (in the appendix) shows various personal characteristics of the respondents.
The majority are relatively young (not older than 40 years) and have finished a higher
education. The majority of them are brought up in the Netherlands (1.5th or second
generation), although we also interviewed a significant number of first-generation
Moroccan migrants. An important difference among the three immigrant categories is
their religious affiliation. The large majority of the Dutch–Turkish and Dutch–
Moroccan respondents identify themselves as Muslims. More Dutch–Moroccan than
Dutch–Turkish respondents indicate that their religion is very important to them. The
religious affiliations of the Dutch–Surinamese respondents are more diverse. The large
number of Hindus in the Dutch–Surinamese group is because a significant part of the
Surinamese population has its ethnic roots in India.
Table 1 shows the means per ethnic group with regard to transnational activities,
transnational identifications, and perceived discrimination. Dutch–Turkish respondents
are more transnationally active than both Dutch–Moroccan and Dutch–Surinamese
respondents. Dutch–Moroccan respondents are most active in everyday economic
transnational activities, but the Dutch–Turkish ones are most active in all other
transnational activities. In addition, the level of transnational identifications differs
among the migrant groups. Dutch–Turkish respondents identify more with their origin
country than Dutch–Moroccan ones, while, on average, the identification with the
origin country is in between for the Dutch–Surinamese respondents.
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
9
Table 1: Description of dependent and independent variables (N=225)
Turkish
N Min. Max. Mean Std. dev.
Transnational identification 75 -0.29 4.00 -2.59 0.79
Transnational activities 75 -0.57 1.04 -0.23 0.39
Sociocultural activities 75 -0.51 1.23 -0.39 0.44
Everyday economic activities 75 -0.83 2.42 -0.01 0.62
Political activities 75 -0.43 2.30 -0.26 0.65
Experiences of discrimination 75 -0.00 3.00 -0.81 0.75
Moroccan
N Min. Max. Mean Std. dev.
Transnational identification 75 -0.43 3.86 -2.24 0.72
Transnational activities 75 -1.29 1.50 -0.10 0.44
Sociocultural activities 75 -2.18 1.23 -0.10 0.61
Everyday economic activities 75 -0.83 2.42 -0.20 0.76
Political activities 75 -0.43 1.32 -0.21 0.48
Experiences of discrimination 75 -0.00 3.00 -0.55 0.72
Surinamese
N Min. Max. Mean Std. dev.
Transnational identification 75 -0.57 4.00 -2.45 0.66
Transnational activities 75 -1.15 1.72 -0.18 0.48
Sociocultural activities 75 -1.95 1.07 -0.29 0.64
Everyday economic activities 75 -0.83 1.86 -0.16 0.57
Political activities 75 -0.43 2.75 -0.05 0.62
Experiences of discrimination 74 -0.00 2.00 -0.82 0.56
Table 1 also shows the means of reported perceived discrimination per ethnic group.
Dutch–Surinamese respondents more often report experiences with discrimination than
the other two ethnic groups, though the mean difference with the Dutch–Turkish respon-
dents is minimal. This is in line with recent Dutch research, which found that res-
pondents with a Surinamese background more often report discrimination experiences
than Dutch–Turkish or Dutch–Moroccan respondents (Huijnk and Dagevos 2012: 37).
These outcomes might seem surprising, since migrants with a Surinamese back-
ground are generally regarded as better integrated and more accepted in Dutch society
(CBS 2014). Given the heated public debates on Islam in the Netherlands, we expected
that particularly Dutch–Turkish and Dutch–Moroccan migrants – many of whom have
a Muslim background – would feel discriminated against. However, our qualitative
material shows that Dutch–Surinamese respondents also say they experience negative
consequences of the debates about ‘failed integration’, even though these debates do
not specifically focus on them (but rather on Muslims and particularly Moroccan
Muslims). A male Dutch–Surinamese respondent says:
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
10
© 2016 The Author(s)
I am fed up with this country and with Europe. As an ‘allochtone’ [term used in
the Netherlands to refer to a non-native Dutch person] you have to prove
yourself two or three times more, you get underestimated a lot. I don’t want that
anymore. Since Wilders, it has become worse. I ask myself if my children have
a future in this country.
Another Dutch–Surinamese male says: ‘I want to feel like I am a real Dutchman, but
because others don’t see me that way, I can’t either.’ The answers of respondents with
a Surinamese background indicate that, next to the ‘non-Muslim/Muslim’ divide, the
more general division ‘Dutch/non-Dutch’ is at least as important for perceived dis-
crimination. For migrants originating from Surinam, who almost all have only Dutch
citizenship and were born with that nationality, being considered ‘non-Dutch’ is a
sensitive issue. A female Dutch–Surinamese respondent, for instance, reacted emo-
tionally when asked why she chose Dutch instead of Surinamese nationality. She said:
‘I was born Dutch!’
Compared with Dutch–Turkish and Dutch–Moroccan migrants, the Dutch–
Surinamese, because of their perceived (national, linguistic and religious) proximity to
the Dutch appear to have higher expectations of ‘inclusion’ into Dutch society. The fact
that they feel discriminated against most, is consistent with the earlier discussed
‘integration paradox’, which holds that the migrants who are the most integrated feel
least accepted (ten Teije et al. 2010; van Doorn et al. 2013).
Reactive transnationalism: activities
Our central research question concerns the relationship between perceived discrim-
ination on the one hand and transnational activities and transnational identifications on
the other. Table 2 (Model 1) shows the association between perceived discrimination
and various transnational activities (sociocultural, economic, political, and all activities
taken together). As the table shows, there is a significant positive association between
perceived discrimination and participation in transnational activities. This association
between perceived discrimination and transnational engagement is stronger when we
group all three transnational activities together than when we consider them separately.
In fact, there are only modest associations between perceived discrimination and par-
ticipation in both sociocultural and political transnational activities. The association
between perceived discrimination and everyday economic activities with regard to the
origin country is somewhat stronger. All in all, the outcomes support the ‘reactive
transnationalism hypothesis’ (hypothesis 1): first- and second-generation migrants who
experienced discrimination are more inclined to engage in transnational activities.
Table 2 (Model 1) shows significant differences among the three migrant groups
with regard to their engagement in transnational activities. Respondents with a
Surinamese and Moroccan background participate less often in transnational activities
than Dutch–Turkish respondents; this goes for two of the three types of transnational
activities separately and the three types taken together. Only with regard to everyday
economic transnational activities are the differences among the three migrant groups
Table 2: Regression analysis on transnational activities (N=223)
Total transnational activities Sociocultural activities Everyday economic activities Political activities
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
B SE B SE B SE B SE B SE B SE B SE B SE
Constant -0.439*** 0.084 -0.470*** 0.090 -0.557*** 0.118 -0.597*** 0.127 -0.114*** 0.139 -0.129*** 0.149 -0.598*** 0.115 -0.636*** 0.124
Experiences of discrimination -0.132*** 0.038 -0.085*** 0.060 -0.121*** 0.054 -0.063*** 0.085 -0.167*** 0.064 -0.132*** 0.099 -0.113*** 0.053 -0.066*** 0.082
Gender (Male= ref) -0.069*** 0.052 -0.034*** 0.077 -0.109*** 0.073 -0.054*** 0.109 -0.117*** 0.086 -0.071*** 0.127 -0.038*** 0.071 -0.028*** 0.106
Ethnicity (Turkish=ref)
S ur in am e se -0.483*** 0.099 -0.490*** 0.100 -0.812*** 0.140 -0.815*** 0.141 -0.078*** 0.164 -0.091*** 0.165 -0.430*** 0.136 -0.438*** 0.138
Moroccan -0.309*** 0.067 -0.313*** 0.068 -0.507*** 0.094 -0.505*** 0.095 -0.184~ 0.110 -0.181*** 0.111 -0.496*** 0.091 -0.508*** 0.093
Generation (First = ref)
1 .5 ge ne ra t io n -0.413*** 0.067 -0.397*** 0.068 -0.440*** 0.094 -0.427*** 0.096 -0.326*** 0.110 -0.309*** 0.113 -0.468*** 0.092 -0.449*** 0.094
2nd generation -0.465*** 0.066 -0.454*** 0.067 -0.429*** 0.093 -0.422*** 0.095 -0.482*** 0.109 -0.459*** 0.111 -0.517*** 0.091 -0.511*** 0.092
Education (Higher education = ref)
Lower secondary -0.274~ 0.149 -0.299~ 0.154 -0.230*** 0.210 -0.251*** 0.216 -0.147*** 0.247 -0.155*** 0.254 -0.447*** 0.205
-0.501*** 0.211
Higher secondary -0.007*** 0.066 -0.002*** 0.066 -0.066*** 0.092 -0.061*** 0.093 -0.048*** 0.109 -0.041*** 0.109 -0.113*** 0.090 -0.119*** 0.091
Importance of religion (Very important = ref)
No Muslim -0.036*** 0.102 -0.022*** 0.128 -0.110*** 0.144 -0.060*** 0.181 -0.137*** 0.169 -0.148*** 0.212 -0.103*** 0.141 -0.027*** 0.176
Not important -0.006*** 0.096 -0.001*** 0.125 -0.132*** 0.135 -0.178*** 0.176 -0.080*** 0.159 -0.170*** 0.206 -0.090*** 0.132 -0.079*** 0.171
Fairly important -0.096*** 0.090 -0.164*** 0.117 -0.028*** 0.127 -0.035*** 0.165 -0.022*** 0.149 -0.157*** 0.193 -0.223~ 0.124 -0.331*** 0.161
Interaction effects
Discrimination*Gender -0.046*** 0.077 -0.070*** 0.108 -0.071*** 0.126 -0.019*** 0.105
Discrimination*Importance of religion (Very important = ref)
Discrimination*no Muslim -0.077*** 0.096 -0.064*** 0.135 -0.022*** 0.159 -0.166*** 0.132
Discrimination*not important -0.004*** 0.106 -0.066*** 0.148 -0.113*** 0.174 -0.009*** 0.145
Discrimination*fairly important -0.158*** 0.172 -0.003*** 0.241 -0.359*** 0.283 -0.214*** 0.235
N 223 223 223 223 223 223 223 223
Adjusted R
2
0.357 0.351 0.302 0.291 0.131 0.126 0.293 0.287
*** p < 0.001 ** p < 0.010 * p < 0.050 ~ p < 0.100 (two-tailed)
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
12
© 2016 The Author(s)
not significant or more moderate. These findings are consistent with Huijnk and
Dagevos (2012: 95) and van Bochove et al. (2010), who also found that Dutch–Turkish
respondents are generally more transnationally active than those with Moroccan and
Surinamese backgrounds; in our conclusion we will further discuss this issue. Respon-
dents of the 1.5th and second generation (born in the Netherlands, or arrived there
before they were 12 years old) are less engaged in transnational activities than first
generation migrants. The educational level of respondents also has a moderate effect
on their engagement in transnational activities, but only with regard to political
activities and when all transnational activities are taken together. Lower educated
respondents participate less in these transnational activities than higher educated ones;
but one should remind that all respondents have a middle-class economic background
and differences in education level are relatively small. Finally, there are hardly any
differences between different religious categories among our respondents. We distin-
guished Muslims from non-Muslims, and within the first category Muslims for whom
religion is ‘very important’ in their lives and Muslims for whom religion is ‘fairly
important’ or ‘not important’. The only noteworthy difference is that Muslims who say
that religion is ‘fairly important’ for them, are less engaged in transnational political
activities than Muslim respondents who say that religion is ‘very important’ for them.
Model 2 of Table 2 shows the interaction terms with regard to transnational activities
that are necessary to test hypotheses 2 and 3, concerning the role of gender and religion,
respectively. Itzigsohn and Saucedo (2005) found that the association between perceived
discrimination and engagement in transnational activities only goes for male migrants,
not for females. However, we found no significant interaction term (perceived discrimin-
ation*gender). The difference between our finding and that of Itzigsohn and Saucedo
might be related to the fact that our respondents were all relatively successful socio-
economically. Itzigsohn and Saucedo (2005) found that women only invest money in
transnational activities when they have sufficient economic resources, while men do so
regardless of their resources. We may not have found any gender differences in our study
because both men and women had sufficient means to participate in homeland activities.
Neither did we find an interaction effect when looking at the religious engagement of
respondents. We expected that, given heated debates about Muslims and Islam in the
Netherlands, respondents with a Muslim background – particularly those for whom rel-
igion is ‘very important’ – would be more inclined towards transnational activities when
confronted with prejudices and discrimination. However, this is not the case. Hypothesis
3 is also to be rejected as far as engagement in transnational activities is concerned.
Reactive transnationalism: identifications
Table 3 shows the relationship between perceived discrimination and transnational
identifications. The table (Model 1) shows moderate positive effects of perceived
discrimination on transnational identifications. In other words, the more discrimination
respondents experience, the more they are oriented towards the country of origin. This,
again, supports the ‘reactive transnationalism hypothesis’ (hypothesis 1). Another
noteworthy outcome of Table 3 (Model 1) is that Dutch–Moroccan respondents identify
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
13
less with their country of origin than both Dutch–Turkish (reference category) and
Dutch–Surinamese ones. Furthermore, 1.5-generation respondents also identify less
with the origin country than first-generation migrants. Finally, non-Muslims among our
respondents and Muslims for whom religion is ‘not important’ identify less with the
origin country than Muslims for whom religion is ‘very important’.
Table 3: Regression analysis on transnational identification (N=223)
Model 1 Model 2
B SE B SE
Constant -2.763*** 0.154 -2.812*** 0.163
Experiences of discrimination -0.148*** 0.070 -0.041*** 0.108
Gender (Male= ref) -0.062*** 0.095 -0.205*** 0.139
Ethnicity (Turkish=ref)
Surinamese -0.229*** 0.182 -0.220*** 0.181
Moroccan -0.455*** 0.122 -0.423*** 0.122
Generation (First = ref)
1.5 generation -0.281*** 0.122 -0.264*** 0.123
2nd generation -0.175*** 0.121 -0.141*** 0.121
Educational level (Higher education = ref)
Lower secondary -0.187*** 0.273 -0.250*** 0.277
Higher secondary -0.156*** 0.120 -0.146*** 0.119
Importance of religion (Very important = ref)
No Muslim -0.638*** 0.187 -0.474*** 0.231
Not important -0.452*** 0.176 -0.425~ 0.225
Fairly important -0.136*** 0.165 -0.050*** 0.211
Interaction effects
Discrimination*Gender -0.376*** 0.138
Discrimination*Importance of religion
(Very important = ref)
Discrimination*No Muslim -0.200*** 0.173
Discrimination*Not important -0.002*** 0.190
Discrimination*Fairly important -0.077*** 0.309
N 223 223
Adjusted R
2
0.107 0.125
*** p < 0.001 ** p < 0.010 * p < 0.050 ~ p < 0.100 (two-tailed)
Model 2, again, includes the interaction terms with gender and religious engagement
to test hypotheses 2 and 3, respectively. The interaction term with gender results in a
remarkable outcome. Contrary to our expectation based on the research of Itzigsohn
and Saucedo (2005) (reactive transnationalism is found only among men) and different
from our own finding regarding transnational activities (no gender differences), here
we found that reactive transnationalism is stronger for female respondents when it
comes to transnational identifications. In other words, when experiencing discrimin-
ation, female respondents appear to have stronger transnational identifications than
male respondents who experience discrimination. This finding is not easy to explain.
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
14
© 2016 The Author(s)
Additional analysis showed that this gender difference only appears for respondents
with a Turkish or Moroccan (thus mostly Muslim) background and not for those with a
Surinamese (mostly non-Muslim) background. Also, our qualitative material showed
that particularly female respondents who considered themselves Muslims (and were
often physically recognizable as such, since they covered their hair) were reflecting on
their religious identity and their place in Dutch society. A Moroccan–Dutch respondent,
for instance, says ‘I always felt like a “Rotterdammer” [inhabitant of Rotterdam].
Particularly when I came back from Belgium, where my mother lives, and the train
passed the mosque in Rotterdam, it felt like coming home. But that changed a few years
ago. Now we are seen as foreigners.’ There is no significant interaction term of per-
ceived discrimination and religious engagement, and therefore, hypothesis 3 is to be
rejected with regard to transnational identifications. We expected that particularly
Muslims for whom religion is ‘very important’ would have strong transnational orien-
tations when experiencing discrimination, but this appears not to be the case.
To summarize, our outcomes confirm hypothesis 1, which is that perceived
discrimination can result in transnational activities and, to a lesser extent, in transnational
identifications. We have to refute hypothesis 2, which expected this ‘reactive trans-
nationalism hypothesis’ to apply only to men, not to women. Female respondents who
experienced discrimination have even stronger transnational identifications than male
respondents who experienced discrimination. We also have to refute hypothesis 3, which
expected Muslims for whom religion is very important to be more transnationally active
and have stronger transnational identifications when experiencing discrimination.
Conclusion and discussion
This article describes the transnational activities and identifications of three migrant
categories (Dutch–Turkish, Dutch–Moroccan and Dutch–Surinamese first- and second-
generation migrants) in the city of Rotterdam. Typical for our study is that all
respondents have a middle-class socio-economic background. We examined to what
extent migrants’ transnationalism results from perceived discrimination in the host
society. Following Itzigsohn and Saucedo (2002, 2005), we called this ‘reactive trans-
nationalism’. We found that, of the three migrant groups, Dutch–Turkish respondents
are significantly more engaged in transnational activities than Dutch–Moroccan and
Dutch–Surinamese ones. Dutch–Turkish respondents also identify with the origin
country significantly more strongly than Dutch–Moroccan respondents, but not com-
pared with Dutch–Surinamese ones. Dutch–Surinamese respondents, on the other hand,
report perceived discrimination more often, although the difference between the Dutch
Turks is minimal.
Our analyses were not meant to explore possible explanations for the observed
differences in transnational engagement between the three migrant categories in our
study. However, we can mention some factors that may contribute to the stronger trans-
national engagement of Dutch–Turkish respondents compared with Dutch–Moroccan
and Dutch–Surinamese ones. First, for Dutch–Turkish migrants, homeland visits are
easy to organize, for the distance is relatively short and airline tickets relatively
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
15
inexpensive compared with those to Morocco and particularly Surinam (van Bochove
et al. 2010). Furthermore, with Turkey at present a more powerful and relatively
affluent country, for the diaspora it may be a more attractive place in which to keep
close contacts. Finally, migration-related institutions in both the origin and host country
may contribute to the stronger transnational engagement of Dutch–Turkish respon-
dents. For instance, the Diyanet (Directorate for Religious Affairs), which the Turkish
government sponsors, actively supports mosques and cultural activities for the Turkish
diaspora in western Europe. Such institutions may indirectly promote a stronger orien-
tation towards Turkey among immigrants. Prominent Turkish migrant organizations
like Milli Görüs may have the same effect on Turkish migrants and their descendants
in western Europe (Avci 2005). According to Nijenhuis and Zoomer (2012), the Surin-
amese government has shown hardly any interest in its diaspora. Dutch–Surinamese
communities in the Netherlands also lack prominent institutions that promote trans-
nationalism. The same goes for Dutch–Moroccan migrants, despite the Moroccan
government trying to maintain control of emigrants by addressing them as Moroccan
nationals and encouraging remittances (Vancluysen et al. 2009).
Our central research question was whether transnational engagement results from
experienced discrimination in the host country. We found that respondents who experi-
enced more discrimination are more engaged in transnational activities. Respondents
who experienced more discrimination also have stronger transnational identifications.
These outcomes seem to confirm the ‘reactive transnationalism’ hypothesis. However,
lacking longitudinal data, we cannot actually prove a causal relation between the two.
To test the assumption of Itzigsohn and Saucedo (2005) that perceived discrim-
ination only resulted in more transnational engagement for men, and not for women,
we computed interaction terms (perceived discrimination*gender). When testing this
hypothesis for transnational activities, we did not find any gender differences. A
possible reason why our outcomes differ from those of Itzigsohn and Saucedo is that
we only interviewed respondents with a middle-class economic background who are
assumed to have sufficient resources to participate in homeland activities. Itzigsohn and
Saucedo argued that only women with sufficient resources would invest in transnational
activities. When testing it for transnational identifications, we did find gender differ-
ences, but in an opposite direction as assumed in hypothesis 2. Confronted with dis-
crimination, female respondents tend to have stronger transnational identifications than
male respondents who experienced discrimination. However, we found this unexpected
gender difference only for Dutch–Turkish and Dutch–Moroccan respondents, not for
Dutch–Surinamese ones. We argued that women with a Muslim background may react
more strongly to perceived discrimination because their religious identity is often more
visible. Whatever the reason, hypothesis 2, which assumes that the reactive trans-
nationalism hypothesis applies only to men and not to women, is to be rejected.
Religion did not make a difference to the relation between perceived discrimination
and transnational engagement. We assumed that, given the current heated debates in
the Netherlands about Muslims in general and Dutch–Moroccans in particular,
respondents who identify themselves as Muslims and for whom religion is very
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
16
© 2016 The Author(s)
important would react with especially strong transnational engagement when con-
fronted with discrimination. However, this was not the case. The reactive trans-
nationalism mechanism goes for non-religious respondents and for Muslims among our
respondents alike. Hypothesis 3 is also to be rejected. Apparently, it is more the feeling
of being accepted in Dutch society than being a Muslim (despite all the debates about
Islam in the Netherlands) that provokes migrants’ stronger transnational engagement
and transnational identifications. Whether one feels not accepted because of religion or
for any other reason, appears not to make a difference.
In conclusion, we should mention some limitations of this study. First, this research
was not based on a representative sample. All our respondents were middle-class with an
immigrant background. One reason to study this specific target group was that previous
research shows that rather well integrated migrants perceive more discrimination (the
so-called ‘integration paradox’). The consequence may be that our results may not be
valid for all migrants. However, our quantitative outcomes are in line with our qualita-
tive material. Second, our findings lead to some new research questions. It is, for instance,
not quite clear why Dutch–Turkish and Dutch–Moroccan (that is mostly Muslim) females
have stronger transnational identifications than males from the same migrant categories.
More in general, our outcomes call for new research, which in more depth explores the
relationships between discrimination, religion, ethnicity, and transnationalism.
Notes
1. The questionnaire (in Dutch) can be found at: www.eur.nl/fileadmin/ASSETS/fsw/Sociolo
gie/burgerschap/Vragenlijst_ Transnationalisme_en_Stedelijk_Burgerschap.pdf.
2. This was 22,921 euros in 2008, when the interviews were conducted.
3. The other items were ‘I feel at home in the country of origin’; ‘I feel like a real Surinamese/
Turkish/Moroccan (wo)man’; ‘I am attached to the country of origin’; ‘the country of origin
is better than other countries’.
References
Avci, G. (2005) ‘Religion, transnationalism and Turks in Europe’, Turkish Studies, 6 (2),
201–13, doi: 10.1080/14683840500119536.
Branscombe, N. R., M. T. Schmitt and R. D. Harvey (1999) ‘Perceiving pervasive discrimination
among African-Americans: implications for group identification and well-being’, Journal of
Personality and Social Psychology, 77 (1), 135–49, doi: 10.1037/0022-3514.77.1.135.
Carling, J. and K. Hoelscher (2013) ‘The capacity and desire to remit: comparing local and
transnational influence’, Journal of Ethnic and Migration Studies, 39 (6), 939–58, doi:
10.1080/1369183X.2013.765657.
CBS (2014) ‘Jaarrapport integratie 2014’, Den Haag: Centraal Bureau voor de Statistiek,
available at: www.cbs.nl/NR/rdonlyres/4735C2F5-C2C0-49C0-96CB-0010920EE4A4/0/
jaarrapportintegratie-2014pub.pdf.
COS (2011) ‘Bevolking van Rotterdam naar land van nationaliteit, op 1-1-2001 t/m 2011’,
available at: www.rotterdam.nl/COS/standaardtabellen/demografie/D05 Bevolking%20
Rotterdam naar land van nationaliteit, op 1-1-2000-2011.pdf.
Dagevos, J., S. Hoff and A. Soede (2006) ‘Minderheden in de middenklasse’, in Sociaal en
Cultureel Planbureau (ed.) Investeren in vermogen: sociaal en cultureel rapport 2006, Den
Haag: SCP, 117–52, available at: www.scp.nl/dsresource?objectid=20544&type=org&usg.
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
17
Fleischmann F., K. Phalet and O. Klein (2011) ‘Religious identification, perceived discrimination
and politicisation: support for political Islam and political action among the Turkish and
Moroccan second generation in five European cities’, British Journal of Social Psychology,
50 (4), 628–48, doi: 10.1111/j.2044-8309.2011.02072.x.
Green, E. G. T., O. Sarrasin and J. Maggi (2014) ‘Understanding transnational political
involvement among Senegalese migrants: the role of acculturation preferences and perceived
discrimination’, International Journal of Intercultural Relations, 41, 91–101, doi:
10.1016/j.ijintrel.2014.05.001.
Huijnk, W. and J. Dagevos (2012) Dichter bij elkaar: de sociaal-culturele positie van niet-
westerse migranten in Nederland, Den Haag: SCP.
Itzigsohn, J. and S. G. Saucedo (2002) ‘Immigrant incorporation and sociocultural
transnationalism’, International Migration Review, 36 (3), 766–98, doi: 10.1111/j.1747-
7379.2002.tb00104.x.
Itzigsohn, J. and S. G. Saucedo (2005) ‘Incorporation, transnationalism, and gender: immigrant
incorporation and transnational participation as gendered processes’, International Migration
Review, 39 (4), 895–920, doi: 10.1111/j.1747-7379.2005.tb00293.x.
Kasinitz, P., J. H. Mollenkopf, M. C. Waters and J. Holdaway (2008) Inheriting the city: the
children of immigrants come of age, New York: Russell Sage Foundation.
Maliepaard, M. and K. Phalet (2013) ‘Religious expression in the face of discrimination: a study
of Muslim migrants in the Netherlands’, paper presented at the Norface Conference,
London.
Nijenhuis, G. and A. Zoomer (2012) ‘Transnational activities of immigrant organizations in the
Netherlands: do Ghanaian, Moroccan and Surinamese diaspora organizations enhance
development?’, paper prepared for the 2nd meeting of the Transnational Organizations
Network, Princeton, May 2012, available at: www.princeton.edu/cmd/working-papers/
2012TransnationalMeeting/2012-Netherlands.pdf.
Phinney, J. S. (1990) ‘Ethnic identity in adolescents and adults: review of research’,
Psychological Bulletin, 108 (3), 499–514, doi: 10.1037/0033-2909.108.3.499.
Portes, A. and R. G. Rumbaut (2001) Legacies: the story of the immigrant second generation,
Berkeley: University of California Press.
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 (2), 217–37, doi:
10.1080/014198799329468.
Rusinovic, K. (2006) Dynamic entrepreneurship: first and second-generation immigrant
entrepreneurs in the Netherlands, Amsterdam: Amsterdam University Press.
Snel, E., G. Engbersen and A. Leerkes (2006) ‘Transnational involvement and social integration’,
Global Networks, 6 (3), 285–308, doi: 10.1111/j.1471-0374.2006.00145.x.
ten Teije, I., M. Coenders and M. Verkuyten (2010) ‘The paradox of integration: immigrants and
their attitude toward the native population’, Social Psychology, 44 (4), 278–88, doi:
10.1027/1864-9335/a000113.
Tsuda,
T. (2012) ‘Whatever happened to simultaneity? Transnational migration theory and dual
engagement in sending and receiving countries’, Journal of Ethnic and Migration Studies,
38 (4), 631–49, doi: 10.1080/1369183X.2012.659126.
van Bochove, M. (2012) ‘Geographies of belonging: the transnational and local involvement of
economically successful migrants’, Ph.D. dissertation, Erasmus University Rotterdam,
available at: http://repub.eur.nl/pub/37260/Marianne van Bochove Geographies of
Belonging.pdf.
van Bochove, M., K. Rusinovic and G. Engbersen (2010) ‘The multiplicity of citizenship:
transnational and local practices and identifications of middle-class migrants’, Global
Networks, 10 (3), 344–64, doi: 10.1111/j.1471-0374.2010.00292.x.
Vancluysen, K., M. Van Craen and J. Ackaert (2009) ‘Transnational activities and socio-cultural
integration of Moroccan and Turkish descendants in Flemish Belgium’, paper presented at
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
18
© 2016 The Author(s)
the XXVI IUSSP International Population Conference, Marrakech (Morocco), September
2009, available at: http://iussp2009.princeton.edu/papers/90999.
van Doorn, M., P. Scheepers and J. Dagevos (2013) ‘Explaining the integration paradox among
small immigrant groups in the Netherlands’, Journal of International Migration and
Integration, 14 (2), 381–400, doi: 10.1007/s12134-012-0244-6.
Vasta, E. (2007) ‘From ethnic minorities to ethnic majority policy: multiculturalism and the shift
to assimilationism in the Netherlands’, Ethnic and Racial Studies, 30 (5), 713–40, doi:
10.1080/01419870701491770.
Verkuyten, M. (2005) The social psychology of ethnic identity, East Sussex: Psychology Press.
Verkuyten, M. (2010) Identiteit en diversiteit: de tegenstelling voorbij, Amsterdam: Pallas
Publications.
Verkuyten, M. and A. A. Yildiz (2007) ‘National (dis)identification and ethnic and religious
identity: a study among Turkish–Dutch Muslims’, Personality and Social Psychology
Bulletin, 33 (10), 1448–62, doi: 10.1177/0146167207304276.
Voas, D. and F. Fleischmann (2012) ‘Islam moves west: religious change in the first and second
generations’, Annual Review of Sociology, 38, 525–45, doi: 10.1146/annurev-soc-071811-
145455.
Weeks, J. (1990) ‘The value of difference’, in J. Rutherford (ed.) Identity, London: Lawrence &
Wishart, 88–100.
Reactive transnationalism: homeland involvement in the face of discrimination
© 2016 The Author(s)
19
Appendix
Table A1: Means per migrant group of transnational activities (N= 75 per
migrant group, deviant N between brackets)
Range Turkish Moroccan Surinamese
Everyday economic activities
Gives money to family or
friends in origin country
0 =never,
3 =often
1.00 1.19 0.85
Owns house in origin
country
0 =no,
1 =yes
0.15 0.21 0.12
(N=74)
Donates to charities related
to origin country
0 =never,
1 =occasionally,
2 =often
0.46
(N=69)
0.55
(N=58)
0.33
(N=66)
Political activities
Reads newspapers from
origin country
0 =never,
1 =less than once a week,
2 =weekly,
3 =more than once a week,
4 =daily
1.67 0.34
(N=74)
0.67
Performs voluntary work
oriented towards origin
country
0 =no,
1 =yes
0.00 0.07 0.11
Performs social and
political actions oriented
towards origin country
0 =no action,
2 =2 actions
0.27 0.04 0.05
Sociocultural activities
Visits country of origin 0 =no,
1 =yes
1.00 0.95 0.87
Contacts with friends or
family in origin country
0 =never;
4 =once a week
2.69
(N=74)
2.32 2.44
Visits cultural events with
artists from origin country
0 =never,
1 =occasionally,
2 =often
1.11 0.80 0.80
(N=74)
Watches TV from origin
country
0 =never,
1 =less than once a week,
2 =weekly,
3 =more than once a week,
4 =daily
2.85 1.49
(N=74)
0.72
Erik Snel, Margrietha ’t Hart and Marianne van Bochove
20
© 2016 The Author(s)
Table A2: Description of dependent and independent variables (N=225)
Turkish Moroccan Surinamese Total
N % N % N % N %
Gender 75 75 75 225
Male 39 52.0 39 52.0 38 50.7 116 51.6
Female 36 48.0 36 48.0 37 49.3 109 48.4
Age 75 75 75 225
20–30 39 52.0 29 38.7 20 26.7 088 39.1
31–40 17 22.7 29 38.7 28 37.3 074 32.9
41–50 18 24.0 15 20.0 10 13.3 043 19.1
50–65 01 01.3 02 02.7 17 22.7 020 08.9
Generation 74 75 75 224
First generation 20 27.0 28 37.3 24 32.0 072 32.1
1.5 generation 20 26.7 21 28.0 27 36.0 068 30.5
2nd
generation 34 45.3 26 34.7 24 32.0 084 37.4
Level of education 75 75 75 225
No higher education 22 29.3 22 29.3 24 32.0 068 30.2
Higher education 53 70.7 53 70.7 51 68.0 157 69.8
Occupational status 75 75 75 225
Higher 28 37.3 24 32.0 22 29.3 074 32.9
Middle 26 34.7 44 58.7 28 37.3 098 43.6
Other white collar 09 12.0 04 05.3 05 06.7 018 08.0
Other 12 16.0 03 04.0 20 26.7 035 15.6
Religion 75 75 74 224
No religion 10 13.3 02 02.7 08 10.8 020 08.9
Christianity 0 00.0 00 00.0 43 58.1 043 19.2
Islam 64 85.3 73 97.3 5 06.8 142 63.4
Hinduism 00 00.0 00 00.0 16 21.6 016 07.1
Other 01 01.3 00 00.0 02 02.7 003 01.3
Importance of religion 75 75 74 224
No Muslim 11 14.7 02 02.7 69 93.2 082 36.6
Not important 12 16.0 07 09.3
01 01.4 020 09.0
Fairly important 15 20.0 09 12.0 00 00.0 024 10.7
Very important 37 49.3 57 76.0 04 05.6 098 43.7