Ethnic Concentration in the Neighbourhood and Ethnic Minorities’ Social Integration

Abstract

Although ethnic concentration in the neighbourhood is often thought to constrain ethnic minorities’ social ties with majority group members, the results of empirical studies are mixed. However, previous studies have differed in the type of social ties examined—for example, neighbours, friends, marital partners—whereas it is questionable whether ethnic residential concentration constrains weak and strong social ties to the same extent. This study analysed survey data from the Netherlands and found that the greater the ethnic concentration in the neighbourhood, the less often ethnic minorities have social ties with the native Dutch. Ethnic residential concentration is also found to constrain the strength of those social ties in terms of support, advice and frequency of contact. Ethnic residential concentration thus not only constrains the probability of having a social tie with natives in general, but of strong social ties in particular.

Introduction

Ethnic residential concentration, the concentration of ethnic minorities¹ in neighbourhoods, is often seen in both science and policy as a hindrance to ethnic minorities’ social ties with ethnic majority group members² (Bolt et al., 2010; Drever, 2004; Gijsberts and Dagevos, 2007; Mesch, 2002; Smets and Den Uyl, 2008; van der Laan Bouma-Doff, 2007).

Although ethnic concentration is also argued to provide opportunities for social and emotional support and an ‘ethnic career’ (for example, Bolt et al., 1998; Wilson and Portes, 1980), the suggestion that ethnic concentration in the neighbourhood constrains these social ties is alarming. Social ties with the ethnic majority are not only an indication of social integration, but are also found to be a determinant for ethnic minorities’ integration in other respects, such as economic and cultural integration and acceptance (Heath et al., 2008; Pettigrew and Tropp, 2006). Social ties with members of the ethnic majority group can for example improve their proficiency in the majority language (for example, Espinosa and Massey, 1997), help them to learn about the norms and values of the host country (Goldscheider and Goldscheider, 1988) and provide them with information and advice in finding work (Lin, 1999). Social ties between ethnic minorities and the ethnic majority also reduce negative intergroup attitudes (Allport, 1954; Pettigrew and Tropp, 2006). It is therefore important to examine whether and to what extent ethnic concentration in the neighbourhood does indeed lower the probability that ethnic minorities will have social ties with the ethnic majority group.

Empirical studies of the relationship between ethnic residential concentration and ethnic minorities’ social ties with the ethnic majority have, however, not yet produced clear evidence on the existence and extent of a relationship between ethnic residential concentration and these social ties. While some studies have found a relationship between ethnic residential concentration and fewer social ties and less frequent contact with members of the ethnic majority group (for example, Briggs, 2007; Esser, 1986; Gijsberts and Dagevos, 2007; Lievens, 1998; Martinovic et al., 2009; Mesch, 2002; van der Laan Bouma-Doff, 2007; van Tubergen and Maas, 2007; Vervoort et al., 2010), others have not (for example, Drever, 2004; Fong and Isajiw, 2000; Sigelman et al., 1996).

These different outcomes might be the result of differences in the type of social ties studied—neighbours (Esser, 1986), friends (Briggs, 2007; Fong and Isajiw, 2000; Mesch, 2002; Sigelman et al., 1996), marital partners (Lievens, 1998; van Tubergen and Maas, 2007)—and frequency of contact with ethnic majority members in general (Drever, 2004; Gijsberts and Dagevos, 2007; Martinovic et al., 2009; van der Laan-Bouma Doff, 2007). These social ties certainly vary in strength and it may be that ethnic residential concentration constrains weak and strong social ties to differing extents. Ethnic concentration might constrain not only the probability of forming social ties with an ethnic majority member, but also the strength of those ties, impeding the formation of strong social ties with the ethnic majority. Research to date has not compared the effects of ethnic residential concentration on social ties of different strengths or directly studied the effect of ethnic residential concentration on the strength of ethnic minorities’ social ties with the ethnic majority group. Moreover, it is difficult to compare the results of earlier studies, because those studies also differed from each other in several other respects, such as the spatial level of ethnic concentration (for example, municipality, postcode area), the specific ethnic groups, year and country under study.

The main aim of the present study is to build on earlier research by examining ethnic residential concentration not only in relation to the likelihood of having social ties with an ethnic majority member, but also to the strength of those ties. We study these relationships using survey data on the four largest non-Western ethnic minority groups in the Netherlands, which provides information on several indicators of the strength of a social tie (see for example, Marsden and Campbell, 1984). We differentiate between frequency of contact, support and advice received. Social ties are seen as weak when contact is infrequent (less than once a week), when no support is received and when advice is not received from the social tie. In contrast, a social tie is considered strong where contact is frequent (at least once a week), where support is received and where advice is received from the social tie.

To understand the importance of the neighbourhood for ethnic minorities’ further integration, it is important to study whether ethnic residential concentration constrains weak and strong social ties with the ethnic majority to the same extent, since weak and strong social ties are expected to have different significance for ethnic minorities’ further integration. For example, according to Granovetter (1973), a weak social tie is most likely to provide the information needed to find a (better) job, whereas positive and strong social ties mainly affect ethnic minorities’ acceptance (Pettigrew, 1997) and their identification with the host country (Clément et al., 2006).

By also studying ethnic concentration in relation to the strength of social ties, therefore, it is hoped that the present study will add to our understanding of the impact of ethnic concentration in the neighbourhood on ethnic minorities’ further integration.

Theory and Hypotheses

A well-known explanatory model which was originally formulated by Kalmijn to explain interethnic marriage (see Kalmijn, 1998; van Tubergen and Maas, 2007), but which is also applied to weak social ties between ethnic minorities and the ethnic majority (for example, Martinovic et al., 2009), differentiates between preferences, opportunities and third parties.

First, people’s preferences are an important determinant of whom they form a social tie with. Most people prefer social ties with persons who are similar to themselves, a phenomenon called homophily (see for an overview McPherson et al., 2001). People may prefer social ties with others who are similar to themselves as regards various socio-demographic, behavioural and intrapersonal characteristics, but the strongest preference is for persons who are similar in race and ethnicity (McPherson et al., 2001). Consequently, ethnic minorities are likely to prefer social ties with co-ethnics. However, the maxim ‘No mating, without meeting’ applies both for social ties with co-ethnics and with ethnic majority group members (Verbrugge, 1979). The formation of social ties with co-ethnics or the ethnic majority depends on there being opportunities to meet (see also Blau, 1977). Ethnic minorities’ opportunities to form social ties with the ethnic majority also depend on ethnic majority members’ preferences. When members of the ethnic majority are not willing to have social ties with ethnic minorities, the likelihood of such ties being formed is smaller.

Finally, third parties can also play a role. These are other people—for example members of the ethnic community—who can encourage or discourage social ties with other ethnic groups (Kalmijn, 1998). Most assume that members of the ethnic minority community will discourage social ties with the ethnic majority group (for example, Martinovic et al., 2009) since they could pose a threat to their traditional culture and values. Members of the ethnic community might therefore constrain ethnic minorities’ social ties with the ethnic majority by creating demands for conformity and group solidarity and sanction social ties with ethnic majority members (see Galster, 2008; Portes, 1998).

Following the framework of preferences, opportunities and third parties, three reasons can be formulated for expecting that ethnic minorities are less likely to have social ties with the ethnic majority when they live in more ethnically concentrated neighbourhoods. First, the opportunities to meet co-ethnics are greater in such neighbourhoods, whereas the opportunities to meet ethnic majority members are smaller. Secondly, ethnic competition theory states that the ethnic majority feel threatened when the number of ethnic minorities is higher (see Blalock, 1967; Coenders et al., 2004), which makes ethnic majority members in ethnically concentrated neighbourhoods probably less willing to form social ties with ethnic minorities. Thirdly, the influence of third parties, such as members of an ethnic community, is probably greater in ethnically concentrated neighbourhoods than in neighbourhoods with a lower percentage of ethnic minorities (see Galster, 2008; Portes, 1998), which can be expected to result in more pressure from members of the ethnic community to form social ties with co-ethnics rather than with the ethnic majority. Based on these notions, we thus expect that

H1: Ethnic residential concentration decreases the likelihood that ethnic minorities will have social ties with ethnic majority members.

It can, however, be questioned whether ethnic residential concentration decreases the likelihood that ethnic minorities will have weak or strong social ties with the ethnic majority group to the same extent. One argument that follows from the idea of third parties is that ethnic residential concentration not only affects the likelihood that ethnic minorities will have social ties with ethnic majority members in general, but also the tie strength. The idea of third parties suggests that the greater ethnic presence in more ethnically concentrated neighbourhoods gives members of the ethnic community more power to sanction social ties with ethnic majority members. However, previous studies have shown that ethnic minorities’ objections to social ties with ethnic majority members increase with tie strength. For example, whereas contact with colleagues and classmates from the ethnic majority group is often accepted, there is more objection to a marital partner from the ethnic majority (Hagendoorn and Kleinpenning, 1991; Gijsberts and Vervoort, 2007). Members of the ethnic community thus probably mainly sanction strong social ties with ethnic majority members. Ethnic residential concentration would therefore not only constrain the likelihood of having social ties with the ethnic majority group, but also the strength of those ties, mainly by impeding the formation of strong rather than weak ties.

Based on ethnic competition theories (for example, Blalock, 1967; Coenders et al., 2004), it might also be expected that ethnic concentration in the neighbourhood not only constrains the probability of social ties with ethnic majority members, but also the strength of such ties. According to ethnic competition theories, ethnic majority members perceive more ethnic threat in ethnically concentrated neighbourhoods, which probably results in a lower willingness to form social ties with ethnic minorities. However, because of the competition for scarce resources, ethnic majority members who perceive ethnic minorities as a threat might be especially reluctant to form strong social ties, involving providing support and advice, rather than weak social ties with ethnic minority persons. In sum, based on the ethnic competition and third parties mechanisms, it may be expected that ethnic residential concentration also constrains the tie strength of ethnic minorities’ social ties with ethnic majority members

H2a: Ethnic residential concentration decreases the likelihood that ethnic minorities will have strong rather than weak social ties with ethnic majority members.

In contrast, based on the idea of meeting opportunities, no additional effect on social tie strength is expected. Whereas ethnic concentration in the neighbourhood influences the opportunities to meet ethnic majority members and the likelihood of having social ties with them, people’s preferences are probably most influential in whether or not a strong social tie is formed. In line with this reasoning, the study by Mollenhorst et al. (2008) indicated that people’s preferences become relatively more, and meeting opportunities relatively less, important in the formation of stronger social ties. Ethnic residential concentration can thus be expected to constrain the likelihood of having social ties with the ethnic majority, but not to affect the tie strength. This results in the following hypothesis

H2b: Ethnic residential concentration is not related to the likelihood that ethnic minorities will have strong rather than weak social ties with ethnic majority members.

Finally, to gain a further understanding of the importance of the neighbourhood context for ethnic minorities’ social ties with ethnic majority members, we will also study the level of economic disadvantage in the neighbourhood in relation to ethnic minorities’ probability of having social ties with the ethnic majority group and the tie strength. Recent Dutch studies among ethnic minorities (Gijsberts et al., 2010; Vervoort et al., 2010) suggest that, after controlling for the ethnic composition of the neighbourhood, economic disadvantage in the neighbourhood is related to more contact with the native Dutch. Two explanations are suggested. On the one hand, residents of economically disadvantaged neighbourhoods seem more locally oriented in their social networks (Pinkster, 2007), probably resulting in more opportunities for ethnic minorities to meet natives than in more affluent neighbourhoods, where native Dutch residents are probably less locally oriented in their social networks. On the other hand, ethnic minorities in economically disadvantaged neighbourhoods may be more likely to meet native Dutch residents who are more similar to themselves as regards education level and social class than in more affluent neighbourhoods. This might make it easier for ethnic minorities in more economically disadvantaged neighbourhoods to come into contact with natives.

Although ethnic minorities in economically disadvantaged neighbourhoods might have greater opportunities to meet ethnic majority residents, it is, however, uncertain whether these contacts are also more likely to result in strong social ties in terms of support, advice and frequency of contact. Besides testing the proposed hypotheses, the present study will further explore whether economic disadvantage is related to the probability of having social ties with ethnic majority members, and the tie strength.

The aim of the present study is thus to examine whether there is a relationship between ethnic residential concentration and the probability of having social ties with the ethnic majority (H1) and between ethnic residential concentration and tie strength, thereby testing two different hypotheses (H2a and H2b). In addition, the study explores the relationships between economic disadvantage in the neighbourhood and the probability of social ties with members of the ethnic minority and tie strength.

Data, Measures and Data Analysis

Data

The data used to test our hypotheses were drawn from a major survey in the Netherlands, the Living Conditions of Urban Ethnic Minorities survey (LAS2004), which was conducted in the 50 largest Dutch municipalities in 2004/05. This survey included the four largest non-Western ethnic minority groups in the Netherlands: those of Turkish, Moroccan, Surinamese and Antillean origin. A random sample was selected from each ethnic group, stratified for municipality size. We used face-to-face interviews, translated questionnaires and bilingual interviewers, where needed and possible. The final sample consisted of 945 Turks, 912 Moroccans, 759 Surinamese and 807 Antilleans.

Information at the level of the neighbourhood was derived from data from Statistics Netherlands (2004), which was merged with the LAS2004 data. We improved on earlier studies by distinguishing neighbourhoods using neighbourhood codes based on the neighbourhoods as identified by the municipal administrations themselves, following the socio-geographical boundaries and the building style and period of construction. These neighbourhoods are more likely to be perceived as a neighbourhood by their residents than the four-digit postcode areas used in most previous Dutch studies. Moreover, whereas there are around 4000 four-digit postcode areas in the Netherlands, our neighbourhood delineation method identified approximately 10 000 Dutch neighbourhoods. Our sample contained residents of 949 neighbourhoods.

Measures

Dependent variables: tie strength

First, respondents were asked whether they had a native Dutch friend or acquaintance; this resulted in a dummy variable where 0 = ‘no social tie’ and 1 = ‘at least one social tie with a native Dutch person’, which is our dependent variable to test whether ethnic concentration in the neighbourhood constrains the likelihood of having a social tie with a native Dutch person.³

To measure the strength of this social tie, respondents with a native Dutch friend or acquaintance (N = 2508) were invited to keep a native Dutch friend or acquaintance in mind and were asked: how frequently they had contact with this person, ranging from 1 ‘seldom’ to 6 ‘every day’; whether they had received practical support from this native Dutch friend or acquaintance (for example, with household chores, borrowing something, going somewhere together or transporting goods) in the last three months; and, whether they had received advice from this native Dutch friend or acquaintance in the last three months.⁴ The correlations between the frequency of contact, receiving practical support and receiving advice ranged between 0.29 and 0.45. As these correlations were only moderate, we decided to study each indicator of a strong versus weak social tie separately, which also made it possible to test whether ethnic residential concentration affected the probability of having a strong social tie in terms of frequency, in terms of received practical support and in terms of received advice, differently. For practical support, we constructed a dummy variable with 0—a weak social tie, i.e. did not receive practical support; and 1—a strong social tie, i.e. received practical support. A similar scale was constructed for advice received. With regard to the frequency of contact, a social tie was categorised as strong if a respondent had contact at least once a week with their social tie.

Unfortunately, information was only available about the tie strength for one social tie with a native Dutch person, whereas people can have many such ties, both weak and strong. However, it can be expected that respondents only kept a weak social tie with a native Dutch person in mind if they had no strong social ties. Therefore, using the available information we can test whether or not ethnic minorities in more ethnically concentrated neighbourhoods are less likely to have at least one strong social tie compared with having only weak social ties with native Dutch.

Independent variables at neighbourhood level

With regard to ethnic concentration, we had access to population register information on the percentage of non-Western ethnic minorities at neighbourhood level (Statistics Netherlands, 2004), which we merged with our survey data.⁵

We constructed an index for economic disadvantage in the neighbourhood by combining the percentage of low-income residents, the average income and the percentage of residents in receipt of social security benefit in the neighbourhood, again derived from register data (Statistics Netherlands, 2004). These indicators together formed one factor that explained 80.4 per cent of the variance. We standardised the indicators and reversed the distribution of average income in order to create an index in which a high score represented more economic disadvantage in the neighbourhood. Subsequently, we calculated the mean of these standardised indicators. The scale was found to be reliable (Cronbach’s alpha = 0.89).

Independent variables at individual level

To control for ethnic minorities’ preferences with regard to social ties with the native Dutch, two questions were combined’ namely: ‘Would you mind if one of your children had lots of native Dutch friends?’ and ‘Would you mind if one of your children chose a Dutch partner?’. Response categories ranged from 1 (‘would not mind at all’) to 5 (‘would mind a lot’). The mean of these items was included in the analyses. A high score means more opposition to social ties with the native Dutch.

Gender was included as a dichotomous variable with 0 = female and 1 = male. Age was taken into account, measured in years ranging from 15 to 65. For household composition, we differentiated between being part of a couple, a parent in an intact family, a single-parent family and a child living with their parents, with single-person households as the reference group. Education was measured in seven categories, from ‘no education’ to ‘higher education’ and was included as a continuous variable. Respondents who were still at school were included by assigning them to the level of education they were currently attending. We included a dummy for persons who were still at school. For socioeconomic position, non-actives (students, housewives, retired and disabled persons) were chosen as the reference group, and we included dummies in the model for unemployed persons and persons with a low, middle and high status.

Finally, we included some migration-specific background characteristics. We included dummies for ethnic group, with Turks as the reference group. For migration, we included dummies for the first generation of ethnic minorities (second generation as reference group), differentiating between first-generation migrants who had lived less than 10 years, 10—25 years and 25 years or longer in the Netherlands. To measure host language acquisition, respondents were asked whether they ‘yes, often’, ‘yes, sometimes’ or ‘no, never’ had difficulties with speaking Dutch, reading Dutch and/or understanding Dutch. The mean score on these three items formed a reliable scale (Cronbach’s alpha = 0.84). Scores were reversed so that a high score corresponded to better host language acquisition. All independent variables were standardised for the analyses.

Data Analysis

Multilevel logistic analyses were used to test our hypotheses (see for example, Snijders and Bosker, 1999). A multilevel technique was needed to study the role of individual and neighbourhood factors simultaneously. Most importantly, it controlled for dependencies in the data that were the result of respondents living in the same neighbourhood (Snijders and Bosker, 1999). This appeared to be necessary in the present study, because the shares of the variances that can be attributed to differences between neighbourhoods ranged between 5 and 10 per cent (see Table A1 in the Appendix for a more detailed examination of the variance components).

Selection Bias

Finally, it should be noted that we cannot make strong statements about the causal direction of the relationships, because we made use of cross-sectional data, whereas longitudinal or experimental data are needed to study causal relationships. As a consequence, a significant relationship can mean that ethnic concentration does indeed constrain the likelihood that ethnic minorities will form social ties with native Dutch persons, but it is also possible that ethnic minorities without such social ties more often choose to live in an ethnically concentrated neighbourhood—for example, to be close to family and co-ethnics. However, in the Dutch situation, problems of self-selection are probably limited (see also Doff, 2010). Most ethnic minorities in the Netherlands live in social housing and live in the dwelling and neighbourhood that were assigned to them by the local authority. In a representative study of people of Turkish, Moroccan, Surinamese and Antillean origin living in the Netherlands (Vervoort et al., 2010), it was found that the majority of them lived in their neighbourhood out of necessity, as they were unable to find a dwelling somewhere else. Only a few respondents stated that they had moved to their neighbourhood for their social contacts. Moreover, Martinovic and colleagues (2009) showed using longitudinal data that a higher percentage of immigrants in the neighbourhood reduced the development of contacts with the native Dutch over time.⁶

Results

Table 1 shows the percentage of ethnic minorities with at least one social tie with a native Dutch person for different levels of ethnic concentration in the neighbourhood. While in total 73.3 per cent of respondents have at least one such social tie, the percentage is lower in neighbourhoods with a higher ethnic concentration. For example, only 59.7 per cent of respondents living in a neighbourhood with more than 50 per cent ethnic minorities, have a social tie with a native Dutch person.

Table 1.

Percentages of respondents with at least one social tie with a native Dutch person, by ethnic concentration in the neighbourhood (N = 3423)

Percentage of ethnic minorities in the neighbourhood	Percentage of respondents with a social tie with a native Dutch person
0–10	89.8
10–25	79.4
25–50	71.1
50–100	59.7
Total	73.3

As regards the tie strength, Table 2 shows that ethnic minorities in more ethnically concentrated neighbourhoods less often have a strong social tie, measured in terms of frequency of contact, support and advice received. In other words, in a neighbourhood with a higher percentage of ethnic minority residents, ethnic minorities are not only less likely to have a native Dutch friend or acquaintance, but where they do they are less likely to have frequent contact, receive support and receive advice from them.

Table 2.

Percentages of ethnic minorities who have at least one strong social tie with a native Dutch person based on frequency of contact, received practical support or received advice, by ethnic concentration in the neighbourhood (N = 2508)

	Percentage with a strong social tie based on
Percentage of ethnic minorities in the neighbourhood	Frequently contact	Received support	Received advice
0–10	76.7	55.8	64.0
10–25	73.6	51.1	60.5
25–50	68.8	47.6	63.1
50–100	62.7	40.3	53.0
Total	70.4	48.6	60.5

To test whether ethnic residential concentration is still associated with a lower probability of social ties with the native Dutch after controlling for other neighbourhood and individual characteristics, we performed a multilevel logistic analysis (see Table 3). This showed that, after controlling for these characteristics, ethnic concentration in the neighbourhood still reduces the likelihood of ethnic minorities having social ties with the native Dutch, thus supporting hypothesis 1. Moreover, after controlling for individual characteristics, economic disadvantage in the neighbourhood is associated with a higher probability of social ties with the native Dutch.

Table 3.

Multilevel logistic analyses of the probability of having at least one social tie compared with no social tie with a native Dutch person (N = 3423)

Neighbourhood level
Ethnic concentration	−0.533	(0.068)***
Economic disadvantage	0.182	(0.070)**
Individual level
Male (ref. = female)	0.107	(0.050)*
Age	−0.090	(0.071)
Household composition (ref. = single-person household)
With partner	−0.031	(0.066)
With partner and children	−0.202	(0.081)*
Single parent	−0.269	(0.057)***
Child living with parents	−0.062	(0.083)
Education level	0.405	(0.062)***
At school	0.040	(0.079)
Socioeconomic position (ref. = non-active)
Unemployed	0.112	(0.050)*
Low-status job	0.067	(0.053)
Middle-status job	0.126	(0.056)*
High-status job	0.214	(0.073)**
Ethnic group (ref. = Turkish)
Moroccan	0.107	(0.052)*
Surinamese	−0.002	(0.064)
Antillean	0.077	(0.068)
Migration background (ref. = second generation)
First generation, <10 years in NL	−0.281	(0.068)***
First generation, 10–25 years in NL	−0.178	(0.080)*
First generation, >25 years in NL	−0.078	(0.082)
Dutch language proficiency	0.255	(0.054)***
Opposition to social ties with native Dutch	0.398	(0.050)***
Intercept	1.326	(0.057)***

Notes: * p < 0.05; ** p < 0.01; *** p < 0.001.

With regard to the individual characteristics, ethnic minorities with stronger opposition to social ties with the native Dutch are found to be less likely to have a social tie with a native Dutch person. Further, a higher education level, higher-status job and better Dutch language proficiency are all found to be associated with a higher likelihood of having social ties with the native Dutch, which is closely in line with the findings of previous studies (for example, Drever, 2004; Lievens, 1998; Martinovic et al., 2009; Vervoort et al., 2010).

Subsequently, multilevel logistic analyses were executed to test whether ethnic concentration in the neighbourhood is related to the strength of the social ties (see Table 4). The results indicate not only that ethnic concentration in the neighbourhood constrains the likelihood of having social ties with the native Dutch in general, but of having strong social ties in particular; this refutes hypothesis H2b, but is in line with H2a. The percentage of ethnic minorities in the neighbourhood is related to a lower probability of having frequent contact with, receiving support and receiving advice from a social tie with a native Dutch person. However, the effect found is stronger for frequency of contact and receiving support than for receiving advice, for which physical proximity is probably less important.

Table 4.

Multilevel logistic analyses of the probability of having at least one strong social tie compared with only a weak social tie with a native Dutch person based on frequency of contact, received support and received advice (N = 2508)

	Frequency of contact		Received support		Received advice
Neighbourhood level
Ethnic concentration	−0.180	(0.068)***	−0.221	(0.064)***	−0.136	(0.073)~
Economic disadvantage	−0.077	(0.066)	0.053	(0.059)	0.097	(0.066)
Individual level
Male (ref. = female)	0.000	(0.050)	0.048	(0.045)	−0.064	(0.048)
Age	−0.106	(0.075)	−0.225	(0.070)***	−0.148	(0.073)*
Household composition (ref. = single-person household)
With partner	−0.079	(0.068)	−0.117	(0.053)*	−0.127	(0.058)*
With partner and children	−0.183	(0.058)**	−0.091	(0.068)	−0.233	(0.073)**
Single parent	−0.236	(0.075)**	0.007	(0.056)	0.057	(0.063)
Child living with parents	0.036	(0.064)	−0.067	(0.065)	−0.207	(0.071)**
Education level	0.044	(0.078)	0.079	(0.054)	0.249	(0.058)***
At school	0.213	(0.078)**	−0.016	(0.066)	0.029	(0.071)
Socioeconomic position (ref. = non-active)
Unemployed	0.004	(0.054)	−0.064	(0.051)	−0.013	(0.054)
Low-status job	0.112	(0.060)~	−0.010	(0.054)	0.004	(0.057)
Middle-status job	0.100	(0.056)~	0.001	(0.052)	0.084	(0.055)
High-status job	0.057	(0.052)	0.026	(0.047)	−0.003	(0.051)
Ethnic group (ref. = Turkish)
Moroccan	0.179	(0.058)**	0.156	(0.054)**	0.137	(0.057)*
Surinamese	0.079	(0.064)	0.098	(0.058)~	0.105	(0.062)~
Antillean	0.158	(0.065)*	0.074	(0.060)	0.090	(0.063)
Migration background (ref. = second generation)
First generation, <10 years in NL	−0.217	(0.065)**	−0.235	(0.058)***	−0.172	(0.061)**
First generation, 10-25 years in NL	−0.138	(0.073)~	−0.124	(0.063)*	−0.168	(0.068)*
First generation, >25 years in NL	−0.063	(0.075)	−0.087	(0.067)	0.003	(0.072)
Dutch language proficiency	0.081	(0.062)	−0.078	(0.059)	−0.073	(0.062)
Opposition to social ties with native Dutch	−0.112	(0.060)~	−0.242	(0.057)***	−0.168	(0.059)**
Intercept	0.844	(0.053)***	−0.158	(0.048)***	0.391	(0.055)***

Notes: ~p < 0.1; * p < 0.05; ** p < 0.01; *** p < 0.001.

Whereas economic disadvantage in the neighbourhood is found to be associated with the probability of having a social tie with a native Dutch person (see Table 3), no relationship was found between economic disadvantage in the neighbourhood and tie strength (Table 4).

With regard to the individual characteristics, we found that ethnic minorities who are more opposed to social ties with the native Dutch are less likely to have a strong social tie with a native Dutch person. In addition, some interesting relationships emerged between individual background characteristics and tie strength. For example, first-generation immigrants who have lived in the Netherlands for less than 10 years are less likely to have formed a strong social tie with a native Dutch person than second-generation minorities. Moreover, Moroccans appear to be more likely than people of Turkish origin to have at least one strong tie rather than only weak social ties with the native Dutch. Further, single-person households are more likely to receive support and advice from their social tie than persons with a partner, and are more likely to have frequent contact with their social tie than persons who live in a household with children. An explanation for these latter findings might be time constraints and the need for strong social ties; persons with a partner and/or children probably have less time and less need of strong social ties than someone who lives alone. Finally, it is interesting to see that more highly educated minorities do not differ from lower-educated minorities in the frequency of contact or receiving support, but do more often receive advice from their social tie than ethnic minorities with a lower education level.

Additional Analyses

Finally, we performed additional analyses with interactions between ethnic group and ethnic concentration in the neighbourhood. This enabled us to test whether the effect of ethnic concentration is stronger for specific ethnic groups, which might provide a further insight into the mechanisms behind the effects found. For example, previous studies have shown that the native Dutch express more negative attitudes towards Moroccans and Antilleans than towards Turks and Surinamese (for example, Gijsberts and Vervoort, 2007). If the mechanism of ethnic competition theories and a lower willingness by the native Dutch to have contact with ethnic minorities explains (part of) the ethnic concentration effect, this could mean that ethnic concentration in the neighbourhood mainly constrains the probability of forming social ties in general, and strong social ties in particular, for Moroccans and Antilleans more than for Turks and Surinamese.

On the other hand, it is also known that, of these four ethnic groups, Turks are most focused on their own ethnic group, hold the most traditional views on the meaning of family ties and attach most value to retaining their own culture (Dagevos et al., 2007; Dagevos and Gijsberts, 2009). Based on the third parties argument, we could therefore expect the ethnic concentration effect to be stronger for Turks, for whom pressure from members of the ethnic community is probably more likely to play a role.

The results of the additional analyses (see Table A2, Appendix) show that the negative effect of ethnic concentration on the probability of having at least one social tie with a native Dutch person is stronger for Antilleans and Moroccans than for Turks. In other words, for Antilleans and Moroccans in particular, living in a more ethnically concentrated neighbourhood constrains their probability of having a social tie with a native Dutch person. This finding is in line with the expectations following ethnic competition theory, which posits that native Dutch residents in more ethnically concentrated neighbourhoods are less willing to engage in social contact with ethnic minorities, and with Moroccans and Antilleans in particular, thus lowering the latter’s probability of forming a social tie with a native Dutch person. However, the effect of ethnic concentration on the strength of the tie did not differ between the ethnic groups. We thus find no evidence to support the idea that specific ethnic groups are especially less likely to form a strong rather than a weak social tie with a native Dutch person when they live in a more ethnically concentrated neighbourhood.

Conclusions and Discussion

Previous research has not yet established clear evidence for the frequently suggested relationship between ethnic residential concentration and ethnic minorities’ social ties with the ethnic majority group. However, earlier studies often differed in the type of social ties studied, such as acquaintances, friends or a marital partner, whereas it can be questioned whether ethnic residential concentration constrains weak and strong social ties to the same extent. The goal of the present study was therefore not only to study the relationship between ethnic concentration in the neighbourhood and the likelihood of having social ties with the native Dutch, but also ethnic concentration in the neighbourhood in relation to the strength of those social ties. Does ethnic concentration in the neighbourhood constrain not only the likelihood but also the strength of ethnic minorities’ social ties with the native Dutch?

In line with our first hypothesis, our results showed a general negative relationship between ethnic residential concentration and ethnic minorities’ probability of having social ties with the native Dutch. The higher the percentage of ethnic minorities in the neighbourhood, the lower the probability of these social ties. More importantly, our study is the first to show that ethnic residential concentration also constrains the strength of ethnic minorities’ social ties with the native Dutch; ethnic minorities living in a more ethnically concentrated neighbourhood are less likely to have frequent contact, receive support and receive advice from their native Dutch tie. This finding is in line with hypothesis 2a, but refutes hypothesis 2b.

Although it can be reasoned from the premise of meeting opportunities that ethnic residential concentration constrains the probability of having a social tie with a native Dutch person but not its tie strength, our results indicate that ethnic concentration in the neighbourhood does form a constraint for both the probability and the strength of social ties with the native Dutch. Possible explanations for the negative relationship found between ethnic concentration in the neighbourhood and the strength of ethnic minorities’ social ties with the native Dutch can be derived from the idea of third parties and from ethnic competition theories.

The third party mechanism implies that, in more ethnically concentrated neighbourhoods, third parties, such as members of the ethnic community, can influence the probability of social ties between ethnic minorities and the native Dutch, and the tie strength, by discouraging and sanctioning social ties, and especially strong ties, with the native Dutch.

Based on ethnic competition theory, it is posited that, due to higher levels of perceived ethnic threat in ethnically concentrated neighbourhoods, native Dutch residents in these neighbourhoods will be less willing to engage in social contact with ethnic minorities. Owing to the competition for scarce resources, native Dutch might be especially reluctant to form strong rather than weak social ties with ethnic minorities.

Both the mechanisms of third parties and ethnic competition theory thus offer explanations for our findings, but with the data used it is unfortunately not possible to test these mechanisms fully. And although additional analyses provided some circumstantial evidence that the relationship between ethnic concentration in the neighbourhood and the likelihood of having a social tie with a native Dutch person is probably partly due to a lower willingness of native Dutch residents, no such evidence was found for the relationship between ethnic concentration and the strength of ethnic minorities’ social ties with the native Dutch. Further research is therefore needed to look into this further and to study why ethnic minorities in more ethnically concentrated neighbourhoods less often have social ties with the native Dutch in general, and strong social ties in particular. For example, it would be informative to study whether and to what extent members of the ethnic community in such neighbourhoods sanction strong social ties with the native Dutch, and whether and to what extent native Dutch residents in these neighbourhoods are less willing to form (strong) social ties with ethnic minorities.

Finally, taking into account the economic disadvantage in the neighbourhood in relation to ethnic minorities’ social ties with natives showed—in line with previous research (Gijsberts et al., 2010; Vervoort et al., 2010)—that other characteristics being equal, ethnic minorities are more likely to have social ties with natives when they live in a more economically disadvantaged neighbourhood. In addition, our findings indicate that economic disadvantage is not associated with the strength of these ties. Ethnic minorities in more economically disadvantaged neighbourhoods are thus more likely to have social ties with natives, but the strength of these ties does not significantly differ in terms of support, advice and frequency of contact compared with the social ties of ethnic minorities in more affluent neighbourhoods. More research is needed to study the relationship between economic disadvantage and ethnic minorities’ social ties with the native Dutch and the underlying mechanisms in more detail.

All in all, our study has important implications. Since social ties with the native Dutch are thought to be important for ethnic minorities’ economic and cultural integration and acceptance, based on our results, living in an ethnically concentrated neighbourhood does indeed appear to hinder successful integration of ethnic minorities. More specifically, we found that ethnic residential concentration particularly constrains strong social ties between ethnic minorities and natives, which are particularly important for ethnic minorities’ acceptance (Pettigrew, 1997) and identification with the host country (Clément et al., 2006). By contrast, ethnic residential concentration was less strongly related to ethnic minorities’ weak social ties with natives, which are probably more important for ethnic minorities’ economic integration (Granovetter, 1973). Ethnic residential concentration, which is found mainly to constrain strong social ties with the native Dutch, thus probably influences ethnic minorities’ acceptance and cultural integration more than their economic integration.

The negative relationship between ethnic residential concentration and ethnic minorities’ probability of having social ties with natives calls for ways of preventing these negative outcomes. To tackle this, it is necessary to gain a greater understanding of the underlying mechanisms. More research is also needed to examine in more detail the extent to which social ties with natives are beneficial for ethnic minorities’ integration, and which aspects of social ties with the native Dutch are important: the strength of the ties in terms of providing support and advice, or the frequency of contact. And how does this relate to different forms of integration? As long as knowledge about these issues is limited, it remains difficult to predict whether and to what extent ethnic residential concentration impedes the further integration of ethnic minorities. Nevertheless, our results further highlight the importance of not disregarding the neighbourhood context when studying and discussing ethnic minorities’ integration.

Footnotes

Notes

Funding Statement

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Appendix

Table A2.

Additional multilevel logistic regression analyses with interactions between ethnic concentration in the neighbourhood and ethnic group

			Probability of having at least one strong social tie vs a weak social tie (N = 2508) based on
	Probability of having at least one social tie vs no social tie (N = 3423)		Frequency of contact		Received support		Received advice
Intercept	1.338	(0.059)***	0.846	(0.053)***	−0.159	(0.048)***	0.393	(0.055)***
Neighbourhood level
Ethnic concentration	−0.550	(0.070)***	−0.177	(0.068)**	−0.220	(0.064)***	−0.133	0.073~
Economic disadvantage	0.186	(0.071)**	−0.074	(0.066)	0.051	(0.060)	0.096	0.067
Individual level
Ethnic group (ref. = Turkish)
Moroccan	0.131	(0.054)*	0.178	(0.058)**	0.162	(0.056)**	0.138	0.057*
Surinamese	−0.001	(0.068)	0.078	(0.064)	0.099	(0.059)~	0.104	0.062~
Antillean	0.082	(0.068)	0.163	(0.067)*	0.078	(0.062)	0.108	0.066
Interactions
Ethnic concentration * Moroccan	−0.096	(0.053)~	0.059	(0.061)	0.033	(0.060)	0.054	(0.060)
Ethnic concentration * Surinamese	−0.039	(0.054)	−0.010	(0.057)	0.030	(0.054)	0.001	(0.057)
Ethnic concentration * Antillean	−0.123	(0.055)*	0.033	(0.058)	0.029	(0.055)	0.067	(0.058)

Notes: Controlled for gender, age, household composition, educational level, at school, socioeconomic position, migration background, Dutch language proficiency and opposition to social ties with native Dutch. ~p < 0.1; * p < 0.05; ** p < 0.01; *** p < 0.001.

References

Allport

(1954) The Nature of Prejudice. Reading, MA: Addison-Wesley.

Blalock

H. M.

(1967) Toward a Theory of Minority-group Relations. New York: Wiley.

Blau

P. M.

(1977) A macrosociological theory of social structure, The American Journal of Sociology, 83, pp. 26–54.

Bolt

Burgers

van Kempen

(1998) On the social significance of spatial location; spatial segregation and social inclusion, Netherlands Journal of Housing and the Built Environment, 13, pp. 83–95.

Bolt

Özüekren

A. S.

Philips

(2010) Linking integration and residential segregation, Journal of Ethnic and Migration Studies, 36, pp. 169–186.

de Briggs

X. S.

(2007) ‘Some of my best friends are…’: interracial friendships, class, and segregation in America, City and Community, 6, pp. 263–290.

Clément

Singh

S. S.

Gaudet

(2006) Identity and adaptation among minority Indo-Guyanese: influence of generational status, gender, reference group and situation, Group Processes and Intergroup Relations, 9, pp. 289–304.

Coenders

Gijsberts

Scheepers

(2004) Resistance to the presence of immigrants and refugees in 22 countries, in: Gijsberts

Hagendoorn

Scheepers

(Eds) Nationalism and Exclusion of Migrants: Cross-national Comparisons, pp. 97–120. Aldershot: Ashgate.

Dagevos

Gijsberts

(2009) Socio-culturele positie [Sociocultural position], in: Gijsberts

Dagevos

(Eds) Jaarrapport integratie 2009 [Annual report on integration 2009], pp. 226–253. Netherlands Institute for Social Research, The Hague.

10.

Dagevos

Schellingerhout

Vervoort

(2007) Sociaal-culturele integratie en religie [Socio-cultural integration and religion], in: Dagevos

Gijsberts

(Eds) Jaarrapport integratie 2007 [Annual report on integration 2007], pp. 163–191. Netherlands Institute for Social Research, The Hague.

11.

Doff

(2010) Puzzling Neighbourhood Effects: Spatial Selection, Ethnic Concentration and Neighbourhood Impacts. Delft: Delft University Press.

12.

Drever

A. I.

(2004) Separate spaces, separate outcomes? Neighbourhood impact on minorities in Germany, Urban Studies, 41, pp. 1423–1439.

13.

Espinosa

K. E.

Massey

D. S.

(1997) Determinants of English proficiency among Mexican migrants to the United States, International Migration Review, 31, pp. 28–50.

14.

Esser

(1986) Social context and inter-ethnic relations: the case of migrant workers in West-German urban areas, European Sociological Review, 2, pp. 30–51.

15.

Fong

Isajiw

W. W.

(2000) Determinants of friendship choices in multiethnic society, Sociological Forum, 15, pp. 249–271.

16.

Galster

G. C.

(2008) Quantifying the effect of neighbourhood on individuals: challenges, alternative approaches and promising directions, Journal of Applied Social Science Studies, 128, pp. 7–48.

17.

Gijsberts

Dagevos

(2007) The socio-cultural integration of ethnic minorities in the Netherlands: identifying neighbourhood effects on multiple integration outcomes, Housing Studies, 22, pp. 805–831.

18.

Gijsberts

Vervoort

(2007) Wederzijdse beeldvorming [Mutual perceptions], in: Dagevos

Gijsberts

(Eds) Jaarrapport integratie 2007 [Integration report 2007], pp. 282–310. Netherlands Institute for Social Research, The Hague.

19.

Gijsberts

Vervoort

Havekes

Dagevos

(2010) Maakt de buurt verschil? [Does the neighbourhood make a difference?]. Netherlands Institute for Social Research, The Hague.

20.

Goldscheider

F. K.

(1988) Ethnicity, religiosity and leaving home: the structural and cultural bases of traditional family values, Sociological Forum, 3, pp. 525–547.

21.

Granovetter

(1973) The strength of weak ties, American Journal of Sociology, 78, pp. 1360–1380.

22.

Hagendoorn

Kleinpenning

(1991) The contribution of domain-specific stereotypes to ethnic social distance, British Journal of Social Psychology, 30, pp. 63–78.

23.

Heath

A. F.

Rothon

Kilpi

(2008) The second generation in western Europe: education, unemployment, and occupational attainment, Annual Review of Sociology, 34, pp. 211–235.

24.

Kalmijn

(1998) Intermarriage and homogamy: causes, patterns, trends, Annual Review of Sociology, 24, pp. 395–421.

25.

Laan Bouma-Doff

W. van der

(2007) Confined contact: residential segregation and ethnic bridges in the Netherlands, Urban Studies, 44, pp. 997–1017.

26.

Lievens

(1998) Interethnic marriage: bringing in the context through multilevel modelling, European Journal of Population, 14, pp. 117–155.

27.

Lin

(1999) Social networks and status attainment, Annual Review of Sociology, 25, pp. 467–487.

28.

Marsden

Campbell

(1984) Measuring tie strength, Social Forces, 63, pp. 482–501.

29.

Martinovic

Tubergen

F. van

Maas

(2009) Dynamics of interethnic contact: a panel study of immigrants in the Netherlands, European Sociological Review, 25, pp. 303–318.

30.

McPherson

Smith-Lovin

Cook

J. M.

(2001) Birds of a feather: homophily in social networks, Annual Review of Sociology, 27, pp. 415–444.

31.

Mesch

G. S.

(2002) Between spatial and social segregation among immigrants: the case of immigrants from the FSU in Israel, International Migration Review, 3, pp. 912–934.

32.

Mollenhorst

Völker

Flap

(2008) Social contexts and personal relationships: the effect of meeting opportunities on similarity for relationships of different strength, Social Networks, 30, pp. 60–68.

33.

Pettigrew

T. F.

(1997) Generalized intergroup contact effects on prejudice, Personality and Social Psychology Bulletin, 23, pp. 173–185.

34.

Pettigrew

T. F.

Tropp

L. R.

(2006) A meta-analytic test of intergroup contact theory, Journal of Personality and Social Psychology, 90, pp. 751–783.

35.

Pinkster

F. M.

(2007) Localised social networks, socialisation and social mobility in a low-income neighbourhood in the Netherlands, Urban Studies, 44, pp. 2587–2603.

36.

Portes

(1998) Social capital: origins and applications, Annual Review of Sociology, 24, pp. 1–24.

37.

Sigelman

Bledsoe

Welch

Combs

M. W.

(1996) Making contact? Black–White social interaction in an urban setting, American Journal of Sociology, 101, pp. 1306–1332.

38.

Smets

Den Uyl

(2008) The complex role of ethnicity in urban mixing: a study of two deprived neighbourhoods in Amsterdam, Urban Studies, 45, pp. 1439–1460.

39.

Snijders

Bosker

(1999) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modelling. London: Sage.

40.

Statistics Netherlands (2010a) Website (http://www.cbs.nl/nl-NL/menu/themas/dossiers/allochtonen/methoden/begrippen/default.htm?conceptid=37; accessed 20 January 2010).

41.

Statistics Netherlands (2010b) Website (http://www.cbs.nl/nl-NL/menu/themas/dossiers/allochtonen/methoden/begrippen/default.htm?conceptid=1013; accessed 20 January 2010).

42.

van Tubergen

Maas

(2007) Ethnic intermarriage among immigrants in the Netherlands: an analysis of population data, Social Science Research, 36, pp. 1065–1086.

43.

Verbrugge

L. M.

(1979) Multiplexity in adult friendship, Social Forces, 57, pp. 1286–1309.

44.

Vervoort

Flap

Dagevos

(2010) The ethnic composition of the neighbourhood and ethnic minorities’ social contacts: three unresolved issues, European Sociological Review, DOI: 10.1093/esr/jcq02910.1093/esr/jcq029.

45.

Wilson

K. L.

Portes

(1980) Immigrant enclaves: an analysis of the labor market experiences of Cubans in Miami, The American Journal of Sociology, 86, pp. 295–319.