Scholarly article on topic 'Acculturation of Host Individuals: Immigrants and Personal Networks'

Acculturation of Host Individuals: Immigrants and Personal Networks Academic research paper on "Sociology"

0
0
Share paper
OECD Field of science
Keywords
{""}

Academic research paper on topic "Acculturation of Host Individuals: Immigrants and Personal Networks"

Am J Community Psychol (2008) 42:309-327 DOI 10.1007/s10464-008-9209-5

ORIGINAL PAPER

Acculturation of Host Individuáis: Immigrants and Personal Networks

Silvia Domínguez • Isidro Maya-Jariego

Published online: 16 October 2008 © Springer Science+Business Media, LLC 2008

Abstract There has been a vast amount of research on the changes experienced by immigrants, but little is known about the changes experienced by host individuals. This article focuses on the role of host individuals in the networks of relations between immigrant populations and the communities from the dominant culture, as well as the changes experienced by host individuals because of their continuous contact with immigrants. This research applied a network approach to the study of the acculturation of host individuals. Two independent studies were carried out: a systematic analysis of the personal networks of Argentinean (n = 67), Ecuadorian (n = 59), Italian (n = 37) and German (n = 37) residents in Seville and Cadiz (Spain) (Study 1); and an ethnographic study with human service workers for Latin American immigrants in Boston (USA) (Study 2). With two different strategies, the role of host individuals in personal networks of foreigners in the United States and Spain was analyzed. The results show that host individuals tend to have less centrality than compatriots, showing an overall secondary role in the personal networks of immigrants. The lowest average centrality was observed in recent and temporal migrants, whereas the highest corresponded to the individuals with more time of residence in Spain. The personal networks of human service providers in the United States vary in ethnic composition and in their

S. Domínguez (H)

Department of Sociology and Anthropology, Northeastern University, Holmes 525. 350 Huntington Ave, Boston, MA 02115, USA e-mail: s.dominguez@neu.edu

I. Maya-Jariego

Departamento de Psicología Social, Universidad de Sevilla, Calle Camilo Jose; Cela s/n. 41.018, Seville, Spain e-mail: isidromj@us.es

structural properties, and therefore shape different types of integrative bridges for immigrants.

Keywords Acculturation • Immigrants • Host society • Personal networks

Introduction

The acculturation literature has evolved from a unidimensional model to bidimensional models (Berry 1997, 1998). According to the unidimensional model, acculturation can be conceptualized as movements along a continuum ''ranging from the immersion in one's culture of origin to the immersion in the dominant or host culture'' (Cabassa 2003, p. 132). Unidimensional frameworks assume that the acculturating group has no effect on the dominant culture (Cabassa 2003). The unidimensional model is associated with the ''melting pot'' view of immigrant integration, based on assimilation theories. Past generations of immigrants are thought to have become successful by shedding their own ethnic and cultural background and adopting the ethnicity and culture of the host society. This was largely the case with immigrants coming from Northern Europe whose culture closely resembled the Anglo-Saxon Protestant culture of the United States.

Problems with assimilation began to emerge once Irish and Italian immigrants became subjects of discrimination. Although similar to the receiving population in many ways, including skin color and Western European heritage, their Catholic religion and distinctive Mediterranean culture (in the case of Italians) became significant barriers to their assimilation (Aguirre and Turner 2004). As cross-cultural psychology discovered that third generation European immigrants had identity problems and mourned the loss of

their grandparents' culture (McGoldrick et al. 1983), it became clear that assimilation as a theory and value could be problematic even where apparently successful. Thus, assimilation has been displaced by pluralism: immigrant groups could retain aspects of their ethnicity while participating in the mainstream culture (Adams and Strother-Adams 2001). Pluralism is consistent with the bidimen-sional model of acculturation, which allows for the maintenance of the original culture and adherence to the culture of the host society.

In this context acculturation has been defined as a process of mutual change of individuals and groups of different cultures that come into continuous contact (Berry 1986, 1997, 1998; Berry and Sam 1996; Redfield et al. 1936). There has been a vast amount of research on the changes experienced by immigrants (e.g., Birman 1998; Birman et al. 2002, 2005; Cea et al. 2004; Samaniego and Gonzales 1999; Shen and Takeuchi 2001; Vega et al. 1993), but little is known about the changes experienced by host individuals.

This article focuses on the role of host individuals in the networks of relations between immigrant populations and communities from the dominant culture, as well as the changes experienced by host individuals because of their continuous contact with immigrants. First, we analyzed the sociometric role of host individuals in the personal networks of immigrants through a survey with four groups of foreigners living in Spain. Second, we describe the process of acculturation of several human service providers who were working with immigrants, as part of a wider ethnography of Latin-American women living in Boston (USA).

A Network Approach to the Study of Acculturation

Social Network Analysis (SNA) has been recommended as a useful tool to address contextual questions in community science (Luke 2005). It is a broad set of methods, based on relational data, which represent the connections among a set of actors, for the systematic study of social structure. Network methods are useful to analyze the multiple levels of a community, and to describe the relations between a community and its immediate surrounding context. In this sense, it allows researchers to give an account of common but frequently neglected phenomena, namely the existence of overlapping communities or sub-communities, and the experience of multiple belongings (Berger and Neuhaus 1977; Maya-Jariego 2004; Zimmerman 2000).

SNA may be applied to whole or ego-centric networks. A personal network is a special type of social network that is centered on one person, and describes his or her links to other people. The tradition of social support studies has usually focused on the core components of the personal

network, describing the composition and characteristics of about 10 key members. This information has normally been processed as variables that describe the relationships of the respondent. Christopher McCarty (2002) has recommended analyzing a higher number of alteri, the set of individual actors or nodes within the networks, and at the same time, applying the structural analyses that have been traditionally used on whole (sociocentric) network data. The utility of this approach is to avoid limiting the analysis to the strongest ties and to allow the study of the structural properties of the personal network.

This approach is also useful for the analysis of the process of acculturation of immigrants and the host individuals that are in continuous contact with them (McCarty and Molina 2003). Current methods of assessing acculturation are based on psychometric scales that focus on individual adaptation to specific cultural contexts. Such instruments are very useful for understanding how migrants adapt to the new society. A SNA approach, focused on the interactions of immigrants with other individuals (compatriots or not), provides a new way to compare patterns of acculturation (Wutich and McCarty 2003). The interaction between migrants and the individuals of the host society is one of the key elements in the acquisition of new attitudes and behaviors. Measuring differences in the structure, intensity and content of that interaction allows us to interpret the extent to which migrants' attitudes and behaviors change over time. Therefore, acculturation may be analyzed through the changes in the structure, composition and functionality of the personal networks of immigrants. SNA also provides some perspective on members of the host culture most likely to be influenced themselves by immigration, the subject of our second study.

Migration has been characterized as an ecological transition (Bronfenbrenner 1977). The description of the changes in the personal networks of immigrants is a useful strategy for analyzing psychological adaptation to a new country. For instance, we have built a typology of personal networks as a proxy for the psychological adaptation of immigrants (Maya-Jariego 2003a). The consideration of differential network constellations is an efficient measure in that it is at once comprehensive and parsimonious. In several studies with African and Latin-American immigrants in Spain and Indian immigrants in Argentina, we have observed that family reunification and the incorporation of host individuals are two basic strategies to rebuild personal networks in the host society, with clear consequences for the functionality of the support structure (Maya-Jariego 2006).

Acculturation alters the composition of the personal network by increasing its heterogeneity while also affecting the level of structural cohesion, as well-defined groups

of players (e.g., well connected compatriots versus host individuals) appear more frequently. All of these changes lead to reorganization in the distribution of support and leverage functions (Maya-Jariego 2006).

The interaction provokes changes in the behaviors and attitudes of members of the host population as well, which also can be analyzed with the same approach. The observation and analysis of the personal networks of immigrants may be a tool for screening and examining the parts of the host population that are in frequent contact with foreigners. On the other hand, extending the same argument to immigrants, the heterogeneity, structural cohesion, and support functions of host individuals' personal networks is also a tool for describing the acculturation of the dominant culture members. This does not mean we should discard a population survey approach or other research techniques, but it is a first step for analyzing-applying SNA in a mixed method approach to studying the impact of acculturation on the host population.

In this article we report the results of two independent studies: a systematic analysis of the personal networks of Argentinean, Ecuadorian, Italian and German residents in Seville and Cadiz (Spain) (Study 1); and an ethnographic study of human service providers working with Latin American immigrant women in Boston (USA) (Study 2). With both studies we compare personal networks of immigrants and hosts who work with immigrants in two different contexts, through qualitative and quantitative methods. The analysis of host individuals' sociometric position in the personal networks of immigrants (in the Study 1) and the development of a typology describing the personal networks, attitudes and behaviors of host individuals (in the Study 2) provides a complementary network-based description of the acculturation of dominant culture members, examining both sides of the inter-group contact.

Study 1: Bridges Towards the Whole Community: The Role of Host Individuals in the Personal Networks of Foreigners in Spain

Acculturation varies according to the levels and formats of contact between immigrants and host individuals. It is not easy to locate host culture members who are experiencing an active process of acculturation, as a consequence of frequent contact with immigrants. In this article we follow two approaches to identify host individuals experiencing acculturation. In the first study, the personal networks of immigrants are examined as a way to screen Spaniards, that is to say, members of the receiving country. This indirect approach allows the assessment of who are the Spaniards with more active contact with immigrants, and how are they integrated in the personal networks of these immigrants.

The immigrants in the host society become minorities within a minority-majority situation. This inter-group relationship delineates several possibilities for contact. For instance, an immigrant may be well connected inside his or her own group, but may also be in a frontier situation (distributing links with both groups) or inserted into the local networks. The inverse situation is the case for host individuals. We apply two different and complementary approaches, which allow taking into account both perspectives in the inter-group relationship.

The analysis of the personal networks of immigrants in this first study is an indirect approach to researching the acculturation of host individuals. However, getting information from a sample of immigrants made it possible to find different acculturation experiences in host individuals due to our use of a population approach. This approach is probably less affected by social desirability, as is the case when host respondents are reporting on the contacts that they have with minority groups. On the other hand, personal network analysis is a way to analyze the relationship between host individuals in the context of the personal environment of immigrants. Finally, it also provides information on the level of social assimilation of relocated people, providing a context for interpreting the acculturation experiences of host individuals.

In sum, the diversity of patterns of interactions can be described through the specific role that the dominant culture members play for different relocated individuals. Furthermore, this may be a proxy for analyzing the position of the immigrants in the wider social structure. The aim of this first study was to compare the differences in the sociometric properties of compatriots and non-compatriots in the personal networks of immigrants and, second, to identify the various roles of host individuals.

Method

Participants

Participants of the first study were 200 foreigners residing in Spain. They were from Argentina, n = 67; Ecuador, n = 59; Germany, n = 37; and Italy, n = 37. The age range was 16-60 years with a mean of 36.37 (SD = 9.93); with 79 males and 121 females. The average number of years living in Spain was 4.66. Most of them were recent migrants, with a small sub-sample of students who planned to stay for only a short period in Spain.

Procedure and Measures

The Arizona Social Support Interview Scale (ASSIS) (Barrera 1980) was applied to obtain a list and characteristics of support providers. After obtaining a list of support

providers with ASSIS, the respondents were asked to complete the list with ''five other contacts that are also important for you.'' Then a matrix of n x n actors was built, in which the respondent codified each pair from 0, ''no relationship,'' to 2, ''very related.'' The matrix was computed with UCINET 6.0 (Borgatti et al. 2002) and average centrality measures of each network were derived. The information was entered and computed as attributive data with SPSS 13.0. To have a more ample description of the procedure and a clearer distinction between the social support variables and the personal network structural properties, see Maya-Jariego and Holgado (2005).

Centrality is a measure of how connected the node is to other nodes within the network. Degree is the number of direct ties for each node. Closeness for a node is the inverse of the sum of all distances to all other nodes. A single alter is highly close if it is connected by short paths to many other alters. That is to say, the pattern of direct and indirect ties of a specific alter allows him/her to access all the nodes in the network more quickly and efficiently than others. Degree and Closeness measures tend to be strongly correlated. Betweenness is the number of shortest paths between all alters that a node lies upon. For a more detailed treatment of centrality measures and their computation, see Faust and Wasserman (1999). In this study we consider the average centrality of all nodes in the personal network, taking into account the three measures of centrality mentioned above.

The original 12-item version of the SCI by McMillan and Chavis (1986) was applied, with a Likert-type scale ranging from 1, strongly disagree, to 4, strongly agree. The SCI is an adequate, widely used and sound measure of overall Psychological Sense of Community (PSOC), with items that refer to Membership, Influence, Emotional Connection and Needs Fulfillment (Chipuer and Pretty 1999; Obst and White 2004, 2005). The respondents rated their PSOC with three different communities: (a) the Neighborhood where they live in Spain, (b) the former Neighborhood in their home country and (c) the relational community with fellow compatriots in Spain. The Cron-bach's alpha reliability estimates were moderately high for the three scales (respectively, .66, .79 and .81 coefficients). The reliability was also similar for the four immigrant sub-samples (.70 and .83 in Ecuadorians; .65 and .82 in Argentineans; .63 and .77 in Germans; .70 and .86 in Italians).

Figure 1 is an example of the personal network of an Ecuadorian immigrant living in Seville, derived through this procedure. She mentioned 24 alteri, 18 Ecuadorians (white nodes) and 5 Spaniards (black nodes). There are three clearly defined social spaces in the networks. First, there is a group of nine siblings and a Spanish neighbor, Maria; a second group is comprised of friends and

coworkers, both Spaniards and Ecuadorians; and there is a third group with three family members living in Ecuador. Her father and an Ecuadorian friend, María Transito, are key nodes in the network, both in terms of centrality and betweenness; they are well connected with members of the three previous groups, and play the role of bridges among them. The size of the nodes, representing the multiplexity of support providers, shows that the support functions are distributed among all the members of the network.

The graphic represents the relationships among the al-teri mentioned by the respondent. White nodes are compatriots and black nodes are Spaniards (grey nodes refer to other groups). The size of the node is a depiction of the level of multiplexity of each social support provider or node. In this article, Ego (that is to say, each respondent) is excluded in the representation and analysis of the personal networks, following the arguments by McCarty and Wutich (2005).

The black nodes are our object of analysis in this first study. For each case, we computed the percentage of Spaniards in the personal networks (composition), the average centrality of Spaniards in the personal network (centrality) and the multiplexity of support provided by Spaniards (support) (this is direct information of average position of host individuals in the personal networks of respondents).

We followed the prescriptions by Christopher McCarty (2002) to obtain information on personal networks. The implications and possible applications of personal network surveys and methods used in our study have been described elsewhere (Maya-Jariego and Holgado 2005; see also, McCarty and Govindaramanujam 2005).

Results

The respondents mentioned, on average, 17 alteri in their personal networks, with a subset of 10.74 social support providers. Host individuals represent around 45% of the average personal network, but only 28.6% of the social support core. As shown in Table 1, Germans have a significantly broader personal network than the other three groups and mention more host individuals than the Ecuadorians (Scheffe = 3.74, p < .001). At the same time, Germans also have the broadest social support networks, with more Spaniard support providers than the other three groups (Scheffe = 1.91, p < .01, with Argentineans; Scheffe = 2.37, p < .01, with Italians; and Scheffe = 3.16, p < .001, with Ecuadorians).

On the other hand, by comparison, the ethnic composition of Ecuadorians' and Italians' networks is more biased towards the endo-group, that is to say, the group of ethnic origin (family and compatriot relationships). In the case of the Ecuadorians, 61.39% of the alteri and 79.17% of the

Fig. 1 Personal network of an Ecuadorian immigrant (case 44). Black nodes: Spaniards. White nodes: Compatriots. Size of the node: Multiplexity

Table 1 Description of the personal network and the support providers

Variable Ecuadorians (n = 59) Argentines (n = 67) Germans (n = 36) Italians (n = 37) Total F

M SD M SD M SD M SD M SD

Personal network

Size of the personal network 15.10 5.22 16.44 5.10 20.91 5.66 17.24 4.67 17 5.50 9.825 (3,195)*

Number of compatriots 8.89 3.74 8.07 3.65 8.77 3.99 9.02 4.43 8.62 3.88 0.692 (3,195)

Number of host individuals 5.59 4.18 7.58 4.32 9.33 5.26 6.86 3.11 7.17 4.43 6.007 (3,195)*

Social support providers

Size of the social support subset 8.57 4.40 10.24 4.43 14.83 5.80 11.05 4.33 10.74 5.11 13.618 (3,195)*

Number of compatriots 6.54 3.66 6.83 3.39 8.05 3.90 7.78 4.07 7.14 3.72 1.770 (3,195)

Number of host individuals 1.72 2.23 2.97 2.47 4.88 4.37 2.51 2.37 2.86 3.01 9.5 (3,195)*

* p \ .001

support providers are compatriots; for Italians, these figures are, respectively, 56.80% and 75%. Argentineans have 51.56% compatriots in their personal network and 69% in the social support core. Germans have, respectively, 48.45% and 62%.

The average centrality measures were calculated for each of the 199 personal networks with UCINET 6.0 (Borgatti et al. 2002). (We dropped one case because of missing data). Specifically, we computed the average degree, closeness, betweenness and eigenvector centrality of each personal network. These average indexes are a proxy for the structural properties of the personal network itself (and not direct information of the sociometric

position of the respondent). The information was entered and computed as attributive data with SPSS 13.0. The results, comparing the average for Spaniards and Compatriots alteri, are summarized in Table 2.

Host individuals have, on average, a secondary role in the personal networks of immigrants. According to the matched-pairs t-tests, Spaniards have less centrality than compatriots. Specifically, they show significantly lower degree, closeness, betweenness and eigenvector centrality scores. The support functionality is also different, with host society alteri showing less multiplexity than compatriots.

These results are clearly confirmed with Ecuadorian (n = 59) and Argentinean (n = 67) sub-samples.

Table 2 Average centrality measures for host individuals versus. compatriots alteri

Variable Compatriots Host individuals Total t

M SD M SD M SD

Centrality measures

Degree 46.42 23.59 37.29 23.44 44.73 22.44 5.909*

Closeness 50.05 28.57 45.82 27.61 49.42 28.33 5.355*

Betweenness 3.55 3.18 2.06 2.75 2.95 2.10 5.558*

Eigenvector centrality 29.44 19.14 22.70 16.68 28.03 15.20 4.292*

Multiplexity 2.60 0.63 1.83 0.85 2.26 0.54 10.794*

* p < .001

However, there are no significant differences between the structural position of host individuals and compatriots in German (n = 36) and Italian (n = 37) sub-samples. To simplify presentation, only differences are presented in Table 2. Full findings, including those that were non-significant and/or redundant, are available from the authors. Multiplexity is the only indicator showing statistical differences (t(1,35) = 3.443, p < .01, for Germans; and f(1,35) = 6.119, p < .001 for Italians), with lower scores for Spaniards than compatriots.

As both German and Italian groups are small samples we checked the analysis with the Wilcoxon matched pairs non-parametric test, obtaining almost the same results. Only in the case of betweenness in the German sample did we observe slight differences between host individuals and compatriots (z = -2.225, p < .05).

It may be that host individuals play a different role in the personal networks of European residing in Spain, in comparison to Latin-American immigrants. To check this assumption, our next step was to apply a cluster analysis to classify the roles that host individuals play in the personal networks of the participants in the sample.

In consonance with the objectives of the study, we used as criterion variables the average degree, average

closeness, average betweenness, average eigenvector cen-trality and average multiplexity of host individuals in the personal networks of the respondents. First, we applied a hierarchical cluster analysis to observe the structure of the data. Then we proceeded with several K-Mean cluster analyses with a maximum of 10 iterations and a criterion for convergence of .02. The clusters with three, four and five categories produced meaningful and proper classifications of the data, in terms of distribution of the cases and interpretation of the results. Finally, we opted for the 4-category classification because of a clearer interpretation of the patterns of acculturation.

The classification in four categories, presented in Table 3, shows a consistent structure. From Type 1 to Type 4 host individuals show increasing levels of centrality and multiplexity in the personal networks of the respondents. This is consistent with correlations among degree, closeness and eigenvector centrality (with Pearson's r ranging from .655 to .826). In addition, the functions derived from the network—represented in this case with the multiplexity index—show the same pattern.

From Type 1 to Type 4 we observe different sets of personal networks corresponding to the levels of centrality and polyvalence of host individuals for the respondents.

Table 3 Centroids and number of cases of the 4 categories' cluster

Variables Type 1 Type 2 Type 3 Type 4

(n = 8) (n = 65) (n = 87) (n = 34)

Centroids Centroids Centroids Centroids F(3,190)

Host individuals average centrality measures

Degree 17.97 18.02 38.44 75.73 171.370***

Closeness 8.90 15.28 57.69 82.56 467.837***

Betweenness 0.83 1.42 2.81 1.67 4.252**

Eigenvector centrality -24.97 14.44 26.92 38.92 101.808***

Multiplexity 1.32 1.66 1.92 2.09 3.239*

Note: Convergence obtained with 5 iterations. The minimum distance among the initial centroids is 70.341 * p < .05, ** p < .01, *** p < .001

Table 4 ANOVA of the social and demographic characteristics of the 4-category cluster

Variables Type 1 (n = 8) Type 2 (n = 65) Type 3 (n = 87) Type 4 (n = 34)

M SD M SD M SD M SD F

Demographics

Months residing in Spain 21.12 22.52 26.263,4 26.89 69.242 76.75 91.882 99.86 8.911 (3,189)***

Number of family ties in Spain 2.50 5.50 2.07 4.65 3.03 3.78 2.55 3.59 0.661 (3,189)

Age 28.37 9.36 31.21 9.54 34.59 10.16 36.58 9.12 3.420 (3,189)*

Years of academic studies 16.37 4.30 15.90 5.79 16.04 5.60 13.82 3.97 1.555 (3,187)

Network size

Personal network's size 20.504 4.65 18.554 6.16 16.91 4.63 14.381,2 4.93 5.851 (3,190)**

Number of support providers 13.37 6.63 11.724 5.89 10.84 5.10 10.782 4.29 3.412 (3,190)*

Sense of community

Neighborhood in sending country 31.75 7.77 34.35 7.57 34.86 7.41 33.05 7.69 0.772 (3,190)

Neighborhood in Spain 31.25 5.70 29.64 6.15 31.57 6.43 31.32 6.23 1.258 (3,190)

Compatriots 29.00 8.50 28.524 7.56 28.374 8.29 23.022,3 7.04 4.500 (3,190)*

Compatriots' average centrality measures

Degree 16.873,4 11.12 31.203,4 16.93 50.271,2,4 16.03 67.711,2,3 24.13 40.375 (3,190)***

Closeness 8.3134 5.99 19.643,4 12.75 63.581,2,4 13.10 79.231,2,3 13.89 236.611 (3,190)***

Betweenness 0.983 1.25 2.653 2.67 5.201,2,4 3.30 2.083 2.00 17.071 (3,190)***

Eigenvector centrality -12.732,3,4 13.23 23.911,3,4 22.70 35.051,2 11.63 33.341,2 13.00 23.940 (3,190)***

Multiplexity 3.04 0.99 2.57 0.52 2.60 0.66 2.59 0.59 1.364 (3,190)

Superscripts in each cell indicate which pairs of means are significantly different from each other in each row * p < .05, ** p < .01, *** p < .001

The only exception to this pattern is the betweenness measure that shows its highest centroid in Type 3. On the whole, it seems that the categories diverge in the relative importance of Spaniards for immigrants, with Type 3 showing the cases in which Spaniards more clearly play the role of bridges in the personal network.

Finally, the social and demographic characteristics associated with these 4 categories are shown in Tables 4 and 5. First, there are some clear associations revealed in Table 4; specifically, the classification is significantly related to the age, the time residing in Spain, the network size, and the sense of community with compatriots living in Spain.

From Type 1 to Type 4, respondents have spent increasingly more time in Spain, are older and have a lower sense of community with compatriots. On the other hand, they evidence a decreasing size of the personal network and the social support core.

The classification was cross-tabulated with all the qualitative variables in the survey. Only nationality and income were significantly associated with the cluster. Types 1-4 were not related to gender, plans of residence, patterns of remittances, frequency of visits and phone calls to the home country, antecedents of emigration in the family, and marital status.

Types 1 and 2 coincide in some respects with the profile of temporal residents and recent immigrants. Generally speaking, both types tend to have lower income. An important proportion of Type 4 is explicitly planning to stay in Spain.

On the one hand, Cluster 1 represents recent migrants, which are more connected and more identified with other compatriots than with Spaniards. Furthermore, the host culture members have a low centrality in their networks and provide a comparatively low number of types of social support. On the other hand, the respondents classified in Cluster 4 have on average lived in Spain longer and were more likely to express the intention to stay in Spain in the future. They have experienced a longer time of socialization in the new country, and Spaniards play a more important role in their networks, with more centrality, more closeness, and a greater number of types of social support provided. The other two clusters are in between Cluster 1 and Cluster 2, as in a continuum, with the only exception of the betweenness indicator mentioned above.

Discussion

This first study compares the structural position and the support functionality of Spaniards and compatriots and

Table 5 Chi-square of the social and demographic characteristics of the 4 categories' cluster

Variables Type 1 (n = 8) Type 2 (n = 65) Type 3 (n = 87) Type 4 (n = 34)

F (%) Adj. Res. F (%) Adj. Res. F (%) Adj. Res. F (%) Adj. Res. x2

Nationality 19.570 (9 gl)*

Ecuadorians 2 (3.6) -0.2 20 (35.7) 0.4 26 (46.4) 0.3 8 (14.3) -0.8

Argentines 0 (0) -2.1 14 (21.2) -2.6 39 (59.1) 2.9 13 (19.7) 0.6

Italians 3 (8.3) 1.4 13 (36.1) 0.4 12 (33.3) -1.5 8 (22.2) 0.8

Germans 3 (8.3) 1.4 18 (50) 2.3 5 (13.9) -0.6 10 (27.8) -2.3

Plans of residence 19.401 (12 gl)

Come back to the sending country 4(8) 1.6 20 (40) 1.1 22 (44) -0.1 4(8) -2.1

To stay in Spain 2 (1.8) -1.9 32 (28.8) -1.6 52 (46.8) 0.6 25 (22.5) 2.1

To go to another country 2 (16.7) 2.3 4 (33.3) 0.0 6 (50) 0.4 0 (0) -1.6)

Emigration in the family 5.903 (6 gl)

Some migrants before him/her 3 (3.8) -0.2 21 (26.9) -1.5 42 (53.8) 2 12 (15.4) -0.7

Visits to the home country 13.851 (12 gl)

Never since the arrival into Spain 4 (5.6) 0.8 33 (46.5) 2.8 22 (31) -2.8 12 (16.9) -0.2

Less than once per year 1 (2) -0.9 10 (19.6) -2.5 29 (56.9) 2.1 11 (21.6) 0.8

Phone calls to the home country 12.213 (9 gl)

Everyday 3 (5.6) 0.6 25 (46.3) 2.3 19 (35.2) -1.8 7(13) -0.9

Monthly income 23.545 (12 gl)*

Less than 300 euros 2 (14.3) 1.8 7 (50) 1.5 4 (28.6) -1.3 1 (7.1) -1.1

301-600 euros 1 (2) -1 21 (42.9) 2 19 (38.8) -1.1 8 (16.3) -0.4

601-900 euros 5 (11.4) 2.4 14 (31.8) 0 18 (40.9) -0.7 7 (15.9) -0.4

901-1,200 euros 0 (0) -1.3 5 (18.5) -1.6 15 (55.6) 1.1 7 (25.9) 1.1

More than 1,201 euros 0 (0) -1.5 7 (18.9) -1.9 22 (59.5) 1.9 8 (21.6) 0.6

Note: Full findings, with other variables not reported on the table, are available from the authors * p < .05

identifies four different roles of host individuals in the personal networks of immigrants. The results show that host individuals tend to have less centrality than compatriots, showing an overall secondary role (in comparison with compatriots) in the personal networks of immigrants.

Four groups of cases were described, according to the relative importance (or levels of centrality and multiplex-ity) of the host individuals in the personal networks of the respondents. Although the data cannot be interpreted in longitudinal terms, the lowest average centrality is observed in recent and temporal migrants, whereas the highest corresponds to individuals with longer residence in Spain.

However, there is no linear decrease among clusters in the average structural cohesion of the personal network (as could be expected, according to the linear increase in the average centrality of the subset of host individuals). Following these data, we could formulate the hypothesis that the average betweenness of host individuals in the personal networks of immigrants is progressively increasing until the point at which the respondent reaches a certain level of embeddedness in local networks. On the other hand, the

cluster with the highest average betweenness, both among compatriots' and host individuals' nodes, may represent the point at which the acculturation process becomes more active. These interpretations are in line with previous studies (e.g., Maya-Jariego 2006), but they need a longitudinal design (and more analysis) for confirmation.

Through the examination of the different roles that host individuals play in the personal networks of immigrants we are indirectly describing the process of acculturation. The variation of the SCI among the clusters supports this interpretation. However, individual differences may also have a relevant role, which is the focus of Study 2, involving the analysis of the personal networks of host individuals.

Study 2: The Acculturation of Integrative Bridges Who Work with Latin-American Immigrants in USA

''Thus, the mainstream culture, which is highly variegated in any event -by social class and region, among other factors- changes as elements of the

cultures of the newer groups are incorporated into it''

(Alba and Nee 2003, p. 13).

While assimilation used to mean the incorporation of immigrants who lost their culture in the process, the new formulation of assimilation is a dynamic process in which immigrants and majority cultures converge. As social spaces narrow, boundaries along groups diffuse and interchange occurs where culture is converging (Alba and Nee 2003). Human service providers (HSP's) who work with immigrants constitute an interface between converging cultures.

HSP's play an important role in community networks. In the classic studies by Stanley Milgram on acquaintanceship chains, most of the brokers between ethnic communities were those with a professional role (Korte and Milgram 1970; see also, Maya-Jariego 2003b). The role of HSP's is characterized mainly by the fact that they are sometimes between the networks of the minority groups and the wider community. Many of these professionals can be defined as ''integrative bridges'' (IB's); i.e., individuals who catalyze the integration of two dissimilar populations (Domínguez 2005, 2008).

This second study focuses on how HSP's who are members of the dominant culture (European-Americans) are culturally influenced by the immigrants with whom they work. The objective was to understand how their exposure to Latin-American immigrants influences their values, attitudes, language, behavior and interpersonal relationships. The ethnographic data was collected as part of the Welfare, Children and Families Three-City Study and was complemented with a second wave of ethnographic field research. The Three-City Study examined work, welfare, family, money, intimate relationships, and social networks in the lives of low-income families, as well as the institutional resources available to them at the neighborhood level in three U.S. cities, Boston, Chicago, and San Antonio.

Method

Participant-observation and longitudinal ethnographic interviews were conducted during 1999-2002 with women living in public housing in South and East Boston. Field-work in these neighborhoods included assessing access to services, attending meetings and events, and interviewing neighborhood informants. Several HSP's were interviewed and the degree of acculturation to the Latin culture of their clients was based on their abilities to work with immigrants and statements they made about their work. The interviews and observations were documented in field notes and analyzed with QSR N6, and a coding scheme was developed that accounted for the different types of social service

providers and different types of bridges. This modified grounded theory approach incorporates codes that are based on existing literature as well as codes derived inductively (Glaser 1992; Glaser and Strauss 1967). It included codes for experiences and codes for meanings and interpretations of experiences. Integrative bridges and religiously based motivators for work with immigrants emerged early in the analysis. The analysis presented here highlights the acculturation of the host society through its members' contact with immigrants.

The second wave of data collection was done in the summer of 2006 with a return to some respondents to complete missing information and interview several other HSP's with IRB approval and respondents' consent. The respondents were interviewed for 3 h covering the following themes: role and length of work with immigrants; background information and motivation for working with immigrants; how they have changed and the type of changes experienced since they began working with immigrants; and who were the closest people to them and what was their cultural background. A social network grid was completed with the names, types of relationships, age, ethnicity of tie, length of connection, function and who knew who in the list. Because ethnographic methods include participant observation in the study respondents' lives and in their neighborhoods as well as during interviews, the data gathered provided a multifaceted lens that is ideally suited to investigate a typology with both subjective and objective dimensions (Purvin 2007).

The longitudinal ethnographic fieldwork made apparent the acculturation of HSP's who work with Latin-American immigrants, allowing a further exploration through a second round of interviews. The goal of this qualitative study was to add to existing theories and reconstruct them by looking at the ways in which host individuals are influenced by their work and exposure to immigrants (Burawoy 1991). In terms of methodological limitations, data are based on the experiences of 10 HSP's. Nevertheless, the strength of longitudinal ethnographic interviews is that information not intended surfaces naturally. Because respondents were not randomly selected, they are not representative of the city's overall HSP population. Study 2 explores how the society is changing through immigration by focusing on host individuals. These HSP's share many of the demographic characteristics of many other HSP's and therefore, are not anomalous cases.

The participants in this study are ten European-American HSP's, nine women and one man. As Table 6 shows, three (Sherry, Dianne and Sandra) worked as ESL instructors and the others worked in areas of political activism and community organizing (Carol, Susan, Sister Magdalena, Father Murphy, Antonia, Mary, and Ruth). By focusing on their level of acculturation, this study is

Table 6 Characteristics of the participants

Name Age Role Prior experience with immigrants Working with immigrants

Sandra 40 ESL instructor Travel as tourist 11 years

Sherry Mid 50's Teacher/Counselor Grew up in Europe 25 years

Antonia Mid 50's Human service Travel as tourist 15 years

Mary Mid 50's Human service Travel as tourist 15 years

Ruth Mid 40's Human service Travel as tourist 5 years

Sister Magdalena Early 50's Human service 15 years in Bolivia 10 years

Father Murphy Late 30's Human service Two years in Latin America 7 years

Carol 57 Human service integration Lived in Chile 1978-1988 17 years

Susan Late 20's Human service integration Semester in Chile 3 years

Diane 60 ESL instructor Peace corps in Africa 25 years

Note: Names have been changed to maintain confidentiality

concerned with a side of acculturation and immigration that has largely been neglected (Cabassa 2003). The acculturation typology derived from the data is based on the way that the respondents indicated having been changed because of their exposure to Latin-American immigrants.

Results

Towards a Typology of Acculturation of Host Individuals

We found that the level of acculturation depended on the amount of exposure to Latin-American culture experienced by each host member in the sample. Therefore, the typology flows along a continuum of exposure from temporary to immersion. Travelers are those temporarily exposed to Latin-American culture. Frontier Brokers are constantly exposed to Latin-American culture, and Residents are immersed in Latin-American culture. Each type differs along a progressive continuum of changes in values, attitudes, language, behavior and interpersonal relationships. The changes evident in Travelers are cumulative with additional changes being added as evidenced for the Frontier Brokers and finally, the Residents.

Travelers The majority of hosts in this study are travelers. Travelers include two ESL instructors, Sherry and Sandra and three HSP's, Ruth, Antonia and Mary. These individuals maintain the Latin-American immigrants at a distance. They visit the Latin-American culture but always temporarily. Travelers evidence changes in attitude, behavior and, to a certain degree, values but do not manifest changes in language nor interpersonal relationships. International experience is in the background of Sherry and Sandra who attribute changes in attitude and wanting to work with immigrants to that experience. Sandra had been working with immigrants for 11 years. She spent 6 months in England when in college and stated, ''It was an eye-

opening experience for me... to live in another country. The way you think that things are done in this country is the way things are done in other places. is just not true.'' Sherry, who had been teaching ESL for 25 years, also evidenced this change in attitude about cultural differences and immigration. Sherry grew up in Europe, mostly in England but also recalled her United States-born mother being discriminated against and devalued in Germany. Sherry brought up this story when asked about experiences that made her want to work with immigrants.

When I was 10 or so, I went shopping with my mother to the butcher. I was the interpreter for some meat and when we left, two adults said, 'isn't a shame, that little girls' mother is so stupid'. From then on, I refused to tell my mother to come to school. I refused to tell her anything about school because I didn't want her to embarrass me by coming to school. It wasn't until I was twelve that I realized that my mother wasn't stupid. That she was Pre-med in the United States. She just was slow picking up German. But I believed that adults told the truth.

Her story is not unlike so many immigrant children's stories who hear members of the dominant culture speak in demeaning ways about their mothers and fathers or other relatives and this causes them to be ashamed of them and of their culture of origin (Ramirez 1999). This type of shame can cause individuals to internalize oppression about their culture, impede the successful integration of bicultural identities, and have ramifications for partaking in risky behaviors that can reduce life chances in the second generation of immigrants (Portes and Rumbaut 2001). While this shame did affect Sherry's development, it gave her an appreciation for the struggle of immigrants and was part of her motivation for the work an ESL instructor and counselor and her positive attitudes towards immigrants. Sherry sees cultural differences as assets, not weaknesses and this

has allowed her to identify clearly the types of cultural values she most appreciates in the Latin-American culture (Saleebey 1992). Sherry is ''amazed'' at how ''resourceful Latin-Americans are in garnishing social support.'' She attributes this to valuing ''interdependence as opposed to independence'' and the ''collective as opposed to the individual'' in the U.S. Sherry has also become increasingly sensitive to the ''value of legal documentation in accessing services and rights'' and the ''courage'' that it takes to immigrate... particularly women with children.'' The cultural values Sherry wants to adopt ''as soon as someone gives me grandchildren,'' include the way that she sees that Latin-American families ''express love:''

The other thing that I've listened to how moms, all babies are called ''mi amor''. I have never met a baby that's called my it's name. It's always, ''mi amor, mi amor, mi amor'' and now I'm much more aware of children's, how parents, how Latino parents respond to children, the names they call them. I can't wait to be a grandmother, one of these days someone is going to have to give me a grandchild because I've just watched how they treat their children and it's kinda nice.

Sandra credits her work with Latin-American immigrants as having made her more aware of the ''immigrants' struggle.'' ''It's humbling in a way to see what some of these people go through in terms of working two jobs and raising families and still taking the time to come to school and learn English and try to improve their lives here.'' Sandra also credits her work as having made her feel more comfortable with cultural differences. ''I feel like the more exposure people have to immigrants, the more aware you are of cultural differences and understanding differences and maybe. be less threatened by differences.''

Neighborhood historical dynamics account for Mary, Antonia and Ruth's motivations to work with immigrants. Mary and Antonia are service providers in East Boston, a transitional, receiving neighborhood for immigrants that is undergoing a demographic shift from Italian-Americans to Latin-Americans (Domínguez 2005, 2008). Mary had been working at the Women, Infants and Children (WIC) office for close to 10 years and when asked about her motivation for working with Latin-Americans, she explained, ''You know. as an immigrant myself, the grand-daughter of proud Italian parents. I can't discriminate against immigrants. Not only that, we have a duty to help them and I can't feel good with myself if I do not.''

The neighborhood history was fundamental in Antonia's motivation to work with Latin-American immigrants as well. Antonia explained that,

I think that one of the big factors is that we have an island past. East Boston was made up of five Islands

you know? So we have an 'island mentality'.. This means that we take care of our own. I have worked for years in service organizations which are always in a crisis mode because of the low-education status and low-economic status of immigrants arriving all the time.

Ruth is also a traveler with a neighborhood-based consciousness as a resident of South Boston, which is historically an Irish-American neighborhood (O'Connor 1994). This area is well known for having reacted violently against minorities in school and housing integration. Ruth is conscious of the need to help Latin-Americans become integrated into their public housing developments which had become majority Latin-American (Domínguez 2005). Like Antonia and Mary, Ruth is a native of this neighborhood and knows its culture and racial dynamics well. The racial strife caused Ruth to change her attitude about immigrants and begin to work connecting them to services. ''It is very hard here for outsiders so they have to struggle even more and they deserve and need to be included.''

The relationship between travelers and immigrants is that of teacher and student, and it manifests a power differential. This power differential maintains distance between the two and creates boundaries that are formally enforced. The power differential within the ESL instructors and the majority/minority relations based on ethnic identity in East and South Boston is not conducive to the development of substantive friendships with immigrants. In fact, in close-knit communities, networks tend to manifest the ''strength of strong ties'' that lead to a form of ''bounded solidarity'' and ''enforceable trust'' which can restrict and control members of that community's relationships (Lin et al. 1981; Portes 1995; Portes and Sensenbrenner 1993). Nevertheless, Travelers have an increased social sensitivity that changes their attitudes about immigrants, and they exhibit value adoption and behavior change as in the case of Sherry.

Frontier Brokers These individuals are HSP's that stand between two cultures. All the Frontier Brokers made comments related to moral imperatives based on social justice and their experiences living abroad which gave them an understanding of immigrant experiences. In this sample, religion was the basis for Father Murphy and Sister Magdalena, who were actively practicing according to their Catholic belief system, and Carol, who was previously a nun. Religion is known as a carrier of values and motivations behind altruism (Praetorius and Machtmes 2005). These integrative bridges are what Bochner called ''mediating persons, persons who have the ability to act as links between different cultural systems. by introducing,

translating, representing and reconciling the cultures to each other'' (Bochner 1982, p. 29).

In particular, Father Murphy, Sister Magdalena, and Carol's work was influenced by Catholic-based Liberation Theology, which imbued them with a social consciousness about inequality and focused their practice towards social justice. Father Murphy was a pastor in East Boston. Situated across from a public housing development in the middle of a Latin-American immigrant enclave, he had developed a number of social programs associated with his church, including a domestic abuse shelter and an employment center. Father Murphy was highly trusted by the population and was able to carry out a domestic abuse shelter in the context of a close-knit immigrant community. Sister Magdalena was an HSP in a religious-based social services agency in South Boston. She worked hard integrating Latin-Americans who lived in public housing into the services available in the larger neighborhood dominated by Irish-Americans. During her time in Chile, Carol stopped her affiliation with a religious order and came out as a lesbian. Since the 1980s Carol has worked to integrate Latin-American immigrants into the neighborhood of East Boston. All three Frontier Brokers had spent several years in Latin America and manifested changes in values, attitudes, and behavior. They had become fluent in Spanish but had only minimal changes in their social networks. When asked about experiences that led her to work with immigrants, Sr. Magdalena responded,

I think that the years I lived in Bolivia I had a sense of how it felt to be an immigrant. and you know that I grew up in Charlestown, so I know what these folks are like here in South Boston. I used to run a support group for wives of Southie men who were not from South Boston. This just tells you what the implications are for minorities here. I know that South Boston is resource rich and my job is to get Latinos to those resources.

As Frontier Brokers, Father Murphy, Sister Magdalena and Carol had fully incorporated aspects of the Latin-American culture and as such, they were bilingual and bicultural. While bilingualism implies having the ability to use two languages with similar fluidity, bicultural implies being experienced in two cultures which are used in some blended form for the most part while using more of one culture base over another when situations call for it as in code-switching (Zentella 1990). It is widely accepted that in order to work with immigrants, one must be bilingual and bicultural to adequately address their needs. A lack of choice due to provider networks in managed care also limits access to services for Latin-American immigrants (Dominguez and Watkins 2003). A whole field in psychology, social work and psychiatry has risen around the

idea of cultural competence as a necessary step to reduce health disparities existing between majority and minority communities (Benjamin 1993). As HSP's, Father Murphy, Sister Magdalena and Carol were extremely effective in their work with Latin-American immigrants because they were culturally knowledgeable, flexible and class conscious.

Sister Magdalena provided mental health and community building services as the only Spanish-speaking provider working with the Latin-American immigrant families in public housing in South Boston. During our conversations, she discussed many issues that concerned her, demonstrating her sophistication in differentiating structural problems from cultural ones among the Latin-American families she served. She worried about the plight of Latin-American girls, how families used them to help with family duties in ways that did not allow them time to focus on school. Her way of intervening was culturally responsive by not ''tackling this straight with the families but instead I put the information into play at the parenting classes I give.''

Carol verbalized very clearly what she had ''gained'' working with immigrants. She spoke about her work as helping her to ''keep her privilege in check,'' the understanding that ''respect'' was necessary when dealing and interacting with immigrants who she views as needing to be ''treated with dignity.''

Well I think that I'm constantly reminded that it's easy to take life as one knows it for granted. Um it's very. I work with people everyday who don't have health insurance, who maybe have to work 2 jobs. Their kids look like they might be joining gangs, who can't go back to their home countries when a parent dies. Those are all constraints that people have. All of those things that I have and I could easily take those for granted if I weren't reminded on a daily basis that these aren't things that many people have because they need them..

Respect, I think it's very important that people have a place where they can go where they know their experience is gonna be. even if we have to tell them really hard stuff, everything's done with dignity.

Carol also expressed her integrated political beliefs and becoming collective in her thinking when analyzing the consequences of political decisions on the immigrants she serves. Among the bridges concerned with and motivated by social justice, they shared a political consciousness.

Um, you know on a more analytical level. I think the analysis I do around politics has always reflected off what's going to happen to that group of people that I work with. It's not just about what's going to happen

to me or what's going to happen to my kids and grandkids. What's going to happen, you know, decisions that are made in Congress, so I everything that happens, you know, I look through it with that set of eyes. Which, I probably would not do if I didn't work with immigrants.

Frontier Brokers have lived in Latin America and have continuously worked with immigrants for several years. They have achieved increased sensitivity regarding the plight of immigrants as evidenced by their work to help and incorporate immigrants. They are bilingual, and they have incorporated values and behaviors associated with a substantial understanding of Latin-American culture that allows them to be extremely effective in the work they do. All this involvement and exposure facilitates incorporation of immigrants into their own personal networks. Such change would reflect a high level of acculturation to the immigrant culture made by a representative of the dominant culture. Yet, when doing social network inventories, it became clear that their social relationships consist of other individuals with shared traits (religious and/or sexual orientation). Taking a closer look at Carol's social network map (Fig. 2), we can see that she has several long-term relationships with others who share her minority sexual orientation status. Despite working with Latin-American immigrants for years, she only has one immigrant tie, Ana (Black node), who also shares Carol's sexual minority status.

Figure 2 demonstrates Carol's strong ties, defined as ties that offer emotional and instrumental support and those who are seen frequently. The diagram shows that despite Carol's bilingualism and biculturalism and her work with immigrants over 20 years, she has only one immigrant in her network of close ties. Ana is a lesbian from Chile, who is primarily connected to members of Carol's household, Mary, Ishia and Julie, but she also has a strong relationship with Alice, one of Carol's close friends. As such, Ana is integrated into Carol's family life but not substantially to the rest of Carol's Euro-American support network.

Carol also has strong ties that share her interests. Martha, Alice, Fred and Joseph are friends known close to 20 years brought together over a love of books. ''Basically, it's kind of a group of some of us who worked together before and we get together monthly because otherwise we'd just never see each other.'' As Carol explained when talking about her social support network, ''as a lesbian in a relationship, I need to surround myself with others that are supportive.'' As the diagram shows, there are close relationships lasting decades with Kathy and Calvin, and Kim and Anita who are two lesbian couples and Tom and Allen, a gay couple. Carol has gravitated towards others who share the same social identity as sexual minorities. In addition, Carol works long hours and her availability for

socializing is very limited. When asked about the lack of immigrants among her strong ties, Carol explains, ''It is really a matter of time, I don't have enough time.'' As a result, Carol who has a strong and supportive network in place does not have time, to develop more relationships with immigrants despite the fact that she spends most of her day with them. It seems that having a minority status identity needing support from individuals with similar backgrounds can play a role in preventing the incorporation of immigrants into one's network.

These Frontier Brokers reside in two cultures and one subculture. Both Sister Magdalena and Father Murphy have chosen lives of devotion to their god through their participation in institutionalized Catholic organizations, the priesthood and the sisterhood. Their devotion means that their dominant culture is filtered through their religious beliefs and their personal social networks are filled with others with the same devotion and membership in Catholic institutions, the priesthood and the sisterhood. Carol's dominant culture experience is seen through a minority status lens, which is marginalized and opens her up for discrimination. Consequently, these Frontier Brokers do not have room in their social networks. A question for future research would be to see under what conditions Frontier Brokers do incorporate immigrants into their personal networks.

Residents These are HSP's who have immersed themselves in the immigrant culture. Residents at times prepare themselves by living in another country and looking for positions that will undoubtedly bring them closer to that immigrant group. At other times, they get involved with immigrants through work that initially acts as a boundary between the two groups but ends up being breached though a close relationship that develops with a member of that group. Residents end up with increased sensitivity, adoption of values and behaviors, and experience a radical change in their personal networks.

Susan and Diane fit this pattern but at different points. Both HSP's have lived in another country; Diane through the Peace Corps in Western Africa and Susan, during a junior semester at the University of Michigan, lived abroad in Chile. Diane talks about having been influenced and socialized by the activism of the 1960s. Susan's social consciousness was awakened through college courses and her experiences as a volunteer for Habitat for Humanity. Diane's bilingualism and biculturalism resulted from her temporary stay in West Africa and her extensive experience working with Haitian immigrants prior to her current job as an ESL teacher. Diane worked organizing the Haitian community where she learned Creole and achieved a high degree of biculturalism. She sounded very nostalgic when talking about it:

2.1. Acquaintanceship network 2.2. Strong ties network

BLick nodes: Latinos. White nodes: Americans. Grey nodes: other ethnic groups. Size of the node: Mi|]li|:K \::\

Fig. 2 Carol's personal network. a Acquaintanceship network. b Strong ties network. Black nodes: Latinos. White nodes: Americans. Grey nodes: Other ethnic groups. Size of the node: Multiplexity

''You know. I really miss that. I miss that so much.. I used to be so ingrained in that community. I knew the entire families, I knew the dialects, I knew what was happening to their families in Haiti. . my life was rich, culturally and socially speaking. I was always having breakfast, lunch or dinner with them and saw their children grow... I really miss that involvement.''

For Diane, no longer being a resident in another culture is experienced as a significant loss in her life. Her current social network record showed a limited network with people from the dominant culture. She did however have a lingering Haitian immigrant in her personal network that was in an intermittent intimate relationship with her. In this sense, Diane is a resident on leave with the skills to deal effectively with another immigrant population.

Lastly, Susan is the integrative bridge who has achieved the greatest level of current integration of Latin-American culture and now works with immigrant youth organizing them for political action. When asked about what she had adapted from immigrants she differentiates the two cultures by speaking about interdependence and collectivism:

Um. this is a funny question. I think that in terms of community and interdependence. Because I think that in the culture that I grew up it's like very much like you're on your own. Your personal success is like, the most important thing. I think that that's not really my orientation so much. I think when I was in Chile. There was something that struck me about how close the family was. Always with cousins over everyone just seemed like. everyone was around all

the time, and I felt like I was part of this big family, this community. Everyone knew each other. I mean, of course it's not always like that, but it was in that situation.

Yeah I mean, I think its' really like, the way that you think about other people's needs. Maybe even before you own. Your responsibilities to your family or your responsibilities to your friends are like the most important thing. And I think that that came out a lot with my relationship with (Ex boyfriend). Because he would drive his brother, pick his brother up from work. Take his mom shopping every Saturday, take his cousin to get their hair cut like and it was like, so much and I wasn't used to it, and I was like, why are you doing this? Why you have to do that? That's how it is. That's the most important thing. Like ''my mother comes first'' and not in a weird way, but that was definitely a cultural difference in the beginning.

Having relationships with culturally different individuals can be very challenging as was the case early in her relationship with her ex-boyfriend who prioritized his family above others. Nevertheless, Susan's openness allowed her to resolve that difference and now that has become part of her own culture. When looking at Fig. 3, one can see that Susan is well embedded in a network made up of Latin-American immigrants.

As represented in Fig. 3, Susan has four networks of strong ties; household, friendship-based, family and work-based networks. In contrast to Carol (whose case we examined in Fig. 2), Susan's diagram shows that Latin-American immigrants (black nodes in the diagram) have infiltrated her household, work and friendship-based

networks. Susan lives with Isaac, Melvin, José and Andrés who are all roommates. José and Andrés are both immigrants from Peru. Lisa from Mexico, Corina from El Salvador and Marcela from the Dominican Republic are very close friends with Susan who speaks to them daily and sees them weekly. Calvin, a Dominican, is Susan's ex-boyfriend and is now part of her network of friends. Erica and Ana are her remaining friends but they are not as close; ''I see them maybe once a month,'' but they also work with immigrants which seem to be the central theme in Susan's network.

Her work-based network of strong ties includes three immigrants, Antonio and Tricia from Brazil and Roberto from Peru. Consistently with the case of Carol, the remaining tie is European-American and according to Susan, ''has a network of mostly gay people because she is gay and she does not really have friends who are immigrants.'' Susan's family includes Tomas, Jane, Anette and Fred. They are supportive of Susan's work and networks of friends. ''When I went to Chile, they read books about it and when I started working with immigrants, they began to listen to me and take my experience and knowledge seriously.'' Susan's family support of her life style and work has allowed her to immerse fully in the Latin-American culture without feeling as if she was betraying her family's culture. In turn, her family has remained close to Susan.

Carol's and Susan's diagrams are examples of different personal networks among human service providers who work extensively with immigrants. In both cases, it is possible to identify family, work and friends as different

but interdependent social spaces. However, both networks vary in the level of ethnic heterogeneity and structural cohesion. Carol's network is more cohesive (in structural terms) and only has one immigrant among its members, whereas Susan's network is divided in family, work and household spaces (see Fig. 3) and more than half of the members of her network are immigrants. A question worth pursuing is whether residents are only temporary and tied to developmental stages.

It seems that a more active participation in different social spaces (as is the case with a more heterogeneous composition in Susan's case) is reflected in a higher average betweenness. Consequently, immigrants have potential access to a different pool of resources, or geography of opportunity, in each case. Ana, the only migrant in Carol's personal network, has a different context of opportunities to access new resources than the 11 immigrants that exchange their own resources in Susan's personal network.

Discussion

The relationships of these European-American HSP's with immigrants are generally positive. As an aspect of the immigration literature not covered, it demonstrates that members of the dominant culture are also being changed by their exposure to immigrants and those changes may be perceived as beneficial (at least in the case of HSP). This analysis also points to the fact that some service providers work with immigrants because of their neighborhood's historical culture, as in working class and as receivers of

Fig. 3 Susan's personal network. a Acquaintanceship network. b Strong ties network. Black nodes: Latinos. White nodes: Americans. Size of the node: Multiplexity

immigrants. In turn, living in closely-knitted ethnic neighborhoods can limit the HSP's' contact with immigrants from other ethnicities, even when those providers are motivated by social justice.

The bilingual and bicultural HSP's started that process through education and travel abroad but continued accult-urating as workers. As Susan demonstrates, having a supportive family gives her the freedom to go through further acculturation without feeling that she is distancing herself from her family. Lastly, it is significant to see that the need to be surrounded by individuals with similar backgrounds (sexual orientation and/or religion) can limit the incorporation of Latin-American immigrants into personal networks.

Conclusions

The changes experienced by members of the host society as a consequence of contact with immigrant populations have been largely neglected. However, the interaction between members of both groups has an impact on the attitudes, values and behaviors of both collectives. Even if some individuals—immigrants or host society members—do not participate directly in the interactions, they are indirectly exposed to the process of acculturation through their networks of relations. Based on the relative size of the two populations, it is possible to conjecture that the opportunities for contact with members of the other group are greater for immigrants than for host individuals. It seems that the pattern of intergroup interactions delimits the level of exposure of different individuals to the acculturation process.

The two studies described in this article applied a network approach to investigate the other side of acculturation. We presumed that acculturation is expressed in changes in the composition, structure and functionality of the personal networks of immigrants and members of the host society. Two complementary strategies were utilized to get information about those host individuals with high probabilities of having developed strong ties with immigrants. First, the personal networks of foreigners were used as a method for screening host individuals and examining their sociometric role. Second, we also analyzed the acculturation of host individuals within the context of their professional roles working with immigrants.

The personal networks of HSP's vary in ethnic composition and in their structural properties, and therefore shape different types of bridges for immigrants. We have formulated the hypothesis of an inverse relationship between ethnic heterogeneity and the level of structural cohesion in the personal networks of host individuals. The degree of diversity of HSP's personal networks renders a

different array of potential resources for the immigrants connected with them.

We have illustrated how an analysis of immigrants' personal networks may yield relevant information about the acculturative experiences of host individuals. It is useful for screening and locating host individuals experiencing acculturation, and at the same time, it provides a population approach to the diversity of cultural contact situations that both groups face. It seems that this approach is less influenced by social desirability (as in the case of host individuals reporting on their inter-group relationships), but simultaneously may be complementary to the direct study of host culture members. Regardless, this strategy could be refined in the future with a deeper assessment of the relationships between host individuals in the personal networks of immigrants.

We have observed two sides of the same coin. For instance, personal network analysis of immigrants serves to interpret acculturation experiences of host individuals as a function of the level of social assimilation of immigrants with whom they have contact. On the other hand, the direct examination of the acculturative experiences of host individuals illustrates that they play consequently different roles for minority members, whether bridges, exo-group resources, peers, etcetera. The combination of both approaches is a useful strategy to analyze the process of acculturation on both sides and contributes to the development of an interactive model of acculturation. As seen in the second study, the Latin-American immigrants enriched and energized the lives of the representatives of the dominant culture and as such, point out the possibility that immigrants could be a resource to other host communities. Such interventions would constitute an example of fostering diversity as a mechanism to address social problems (Dominguez 2008).

The examination of personal networks reveals (a) the disposition of migrants and host individuals for making new contacts, (b) the acculturative potential of inter-group contacts for both sides, and (c) the potential contribution of host individuals to the social integration of immigrants. For instance, host individuals may have more or less cohesive personal networks and, as a consequence, be more or less prone to incorporating minority members into their pool of social ties. On the other hand, the acculturation of host society members occurs within the context of the process of adaptation of immigrants. Recently arrived foreigners are usually oriented towards other compatriots who are experiencing the same situation and have fewer opportunities to deploy strong relationships with local people. The immigrants who are already inserted in local networks also have more opportunities to establish new links with host individuals, although their acculturative experience was comparatively less intense.

It seems that the level of change overtime corresponds to the level of exposure. In the case of host individuals, we have distinguished between temporary exposure, constant exposure, and immersion. These three levels correspond, respectively to travelers, frontier brokers and residents. Each category represents a different experience of acculturation, but also potentially a different role for immigrants, whether in terms of acculturation of minority individuals or in terms of adaptation and integration. Residents usually have access to the same resources as the immigrants with whom they are in everyday contact, whereas travelers belong to a more dissimilar social niche and can eventually contribute with new and, in some respect, more useful resources for immigrants. In-between, frontier brokers act as a bridge and channel resources between both communities.

What we are illustrating is that the topology of inter-group networks, both at the micro and the macro levels, helps tell the story of the acculturative experience for both immigrants and host individuals. It seems that there is a homophily tendency in the inter-group relations in which cultural diversity is present. The intercultural relations that seem most probable are those between individuals separated by less cultural distance and, in consequence, experience lower acculturative conflict when they establish a link. For instance, a recently arrived migrant might be more likely to deploy a relationship with a ''resident'' than with a ''traveler,'' even though the second could convey more traditional resources and relations from the host society that are needed in the first stage of accommodation in the new country. We have tried to demonstrate that SNA can contribute to understanding how the opportunities of contact evolve over time and within and between groups.

A case in point would be to examine the properties of the personal networks of the host individuals who, in turn, are present in the personal networks of immigrants. On the other hand, weak links between immigrants and host individuals may constitute another type of bridges that has not been analyzed in this article. In the second study, we observed that those who already had cohesive and dense networks may be less likely to include immigrants in their network.

Host population members represent the dominant side and have more control and influence on the definition of the inter-group relationship. Consequently, their attitudes toward the immigrants and toward contact with minority groups play a key role in the acculturation outcomes of both immigrants and host society individuals. ''The other side of acculturation'' is an important side in intercultural relations, though paradoxically most of the literature has put emphasis on the adaptation efforts that international immigrants must carry out to settle in the new country. One of the informants in Study 2 brought up an aspect of

immigrant incorporation that is seldom considered. Carol feels that ''it's good to live in a diverse world'' not only for herself as an American but for immigrants too. ''One of the lessons, one of the hard lessons for immigrants to learn when they arrive to a new place is that they have to kind of look at the society with brand new eyes.'' Immigration and the ensuing diversity provide the host society with an opportunity for change. One of the main tasks of modern social sciences should be to precisely identify the mechanisms for diversity-induced change so that they can be employed as effective tools to address social problems. It remains up to us to understand diversity so that it can be employed as an effective tool in catalyzing positive outcomes.

Acknowledgments This research was supported by the Department of Housing and Urban Development, the Woodrow Wilson Foundation, the Eileen Blackey Fellowship, the American Association of University Women and the Missy Carter Dissertation Fellowship (for the Latina project in USA) and the Consejería de Gobernación de la Junta de Andalucía (for the study of four foreigners groups in Spain). The authors thank the Principal Investigators of the Welfare, Children and Families Three City Study and the array of public and private funding institutions who supported that research. The authors wish to acknowledge the contribution and assistance of Amy Lubitow, Van Tran and the Harvard University Immigration Incorporation Workshop, Daniel Holgado and Raquel Santos (Universidad de Sevilla), and all of the respondents that contributed to both projects.

References

Adams, J. Q., & Strother-Adams, P. (2001). Dealing with diversity. Chicago: Kendall/Hunt Publishing Company.

Aguirre, A. Jr., & Turner, J. H. (2004). American ethnicity: The dynamics and consequences of discrimination. Boston: McGraw Hill.

Alba, R., & Nee, V. (2003). Remaking the American mainstream: Assimilation and contemporary immigration. Cambridge: Harvard University Press.

Barrera, M. (1980). A method for the assessment of social support networks in community survey research. Connections, 3, 8-13.

Benjamin, M. (1993). Child and adolescent service system program minority initiative research monograph. Washington, DC: Georgetown University.

Berger, P., & Neuhaus, R. (1977). To empower people. The role of mediating structures in public policy. Washington, DC: American Enterprise Institute.

Berry, J. W. (1986). The acculturation process and refugee behavior. In C. I. Williams & J. Westemeyer (Eds.), Refugee mental health in resettlement countries. New York: Hemisphere.

Berry, J. W. (1997). Immigrants, acculturation, and adaptation. Applied Psychology: An International Review, 48, 5-33.

Berry, J. W. (1998). Acculturation and health: Theory and research. In S. S. Kasarian & D. R. Evans (Eds.), Cultural clinical psychology: Theory and research and practice (pp. 39-57). New York: Oxford University Press.

Berry, J. W., & Sam, D. L. (1996). Acculturation and adaptation. In J. W. Berry, A. Segal, & C. Kagitcibasi (Eds.), Handbook of cross-cultural psychology: Social behaviors and application (Vol. 3, pp. 291-326). Boston: Allyn & Bacon.

Birman, D. (1998). Biculturalism and perceived competence of Latino immigrant adolescents. American Journal of Community Psychology, 26, 335-354.

Birman, D., Trickett, E., & Buchanan, R. (2005). A tale of two cities: Replication of a study on the acculturation and adaptation of immigrant adolescents from the former Soviet Union in a different community context. American Journal of Community Psychology, 35, 83-101.

Birman, D., Trickett, E., & Vinokurov, A. (2002). Acculturation and adaptation of Soviet Jewish refugee adolescents: Predictors of adjustment across life domains. American Journal of Community Psychology, 30, 585-607.

Bochner, S. (1982). The social psychology of cross-cultural relations. In S. Bochner (Ed.), Cultures in contact: Studies in cross-cultural interaction. Oxford: Pergamon Press.

Borgatti, S. P., Everett, M. G., & Freeman, L. C. (2002). Ucinet for windows: Software for Social Network Analysis. Harvard: Analytic Technologies.

Bronfenbrenner, U. (1977). The ecology of human development. Cambridge: Harvard University Press.

Burawoy, M. (1991). Reconstructing social theories. In M. Burawoy, et al. (Eds.), Ethnography unbound: Power and resistance in the modern metropolis (pp. 8-27). Berkeley: University of California Press.

Cabassa, L. J. (2003). Measuring acculturation: Where we are and where we need to go. Hispanic Journal of Behavioral Sciences, 25, 127-146.

Cea, M. C., Reisen, C. A., Poppen, P. J., Echeverry, J. J., & Bianchi, F. T. (2004). Disclosure of HIV-positive status to Latino gay men's social networks. American Journal of Community Psychology, 33, 107-116.

Chipuer, H., & Pretty, G. (1999). A review of the sense of community index: Current uses, factor structure, reliability, and further development. Journal of Community Psychology, 27, 643-658.

Domínguez, S. (2005). Latina immigrants in public housing: Race relations, social networks, and access to services. Dissertation unpublished, Boston University.

Domínguez, S. (2008). A tale of two projects: Bridging Latin-American immigrant women in public housing. Journal of City and Community, under revise and resubmit.

Dominguez, S., & Watkins, C. (2003). Creating networks for survival and mobility: Social capital among African-American and Latin-American low-income mothers. Social Problems, 50, 111-135.

Faust, K., & Wasserman, S. (1999). Social Network Analysis: Methods and applications. Cambridge: Cambridge University Press.

Glaser, B. G. (1992). Basics of grounded theory analysis: Emergence vs. forcing. Mill Valley: Sociology Press.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. New York: Aldine de Gruyter.

Korte, C., & Milgram, S. (1970). Acquaintance networks between racial groups: Application of the small world method. Journal of Personality and Social Psychology, 15, 101-108.

Lin, N., Ensel, W. M., & Vaughn, J. C. (1981). Social resources and strength of ties: Structural factors in occupational status attainment. American Sociological Review, 46, 393-405.

Luke, D. A. (2005). Getting the big picture in community science: Methods that capture context. American Journal of Community Psychology, 35, 185-200.

Maya-Jariego, I. (2003a, February). A general typology of personal networks of immigrants with less than 10 years living in Spain. Paper presented at XXIII International Social Network Conference, Meíxico.

Maya-Jariego, I. (2003b). Internet, amigos y bacterias: la alargada sombra de Stanley Milgram. Araucaria. Revista Iberoamericana de Filosofía, Política y Humanidades, 5, 3-14.

Maya-Jariego, I. (2004). Sentido de comunidad y potenciación comunitaria. Apuntes de Psicología, 22, 187-211.

Maya-Jariego, I. (2006). Webs of compatriots: Relationship networks among immigrants. In J. L. Perez Pont (Ed.), Geografías del desorden. Migración, alteridad y nueva esfera social (pp. 257276). Valencia: Universidad de Valencia.

Maya-Jariego, I., & Holgado, D. (2005). Lazos fuertes y proveedores multiples de apoyo: comparacion de dos formas de representacion gráfica de las redes personales. Empiria. Revista de metodología de ciencias sociales, 10, 107-127.

McCarty, C. (2002). Structure in personal networks. Journal of Social Structure, 3,(1). http://www.cmu.edu/joss/content/articles/ volume3/McCarty.html.

McCarty, C., & Govindaramanujam, S. (2005). A modified elicitation of personal networks using dynamic visualization. Connections, 26, 61-69.

McCarty, C. & Molina, J. L. (2003, February). Egocentric network research tool. Paper presented at XXIII International Social Network Conference, Meíxico.

McCarty, C., & Wutich, A. (2005). Conceptual and empirical arguments for including or excluding ego from structural analysis of personal networks. Connections, 26, 84-90.

McGoldrick, M., Pearce, J. K., & Giordano, J. (1983). Ethnicity and family therapy. New York: The Guildford Press.

McMillan, D. W., & Chavis, D. M. (1986). Sense of community: A definition and theory. Journal of Community Psychology, 14, 6-23.

Obst, P. L., & White, K. M. (2004). Revisiting the sense of community index: A confirmatory factor analysis. Journal of Community Psychology, 32, 691-705.

Obst, P. L., & White, K. M. (2005). Three-dimensional strength of identification across group memberships: A confirmatory factor analysis. Self and Identity, 4, 69-80.

O'Connor, T. H. (1994). South Boston, My home town: The history of an ethnic neighborhood. Northeastern University Press.

Portes, A. (1995). Economic sociology and the sociology of immigration: A conceptual overview. In A. Portes (Ed.), Economic sociology of immigration: Essays on networks, ethnicity, and entrepreneurship (pp. 1-41). New York: Russell Sage Foundation.

Portes, A., & Rumbaut, R. G. (Eds.). (2001). Legacies: The story of the immigrant second generation. Berkeley: University of California Press.

Portes, A., & Sensenbrenner, J. (1993). Embeddedness and immigration: Notes on the social determinants of economic action. American Journal of Sociology, 98, 1320-1350.

Praetorius, R. T., & Machtmes, K. (2005). Volunteer crisis hotline counselors: An expression of spirituality. Social Work and Christianity, 32, 116-132.

Purvin, D. (2007). At the crossroads and in the crosshairs: Social welfare policy and low-income women's vulnerability to domestic violence. Social Problems, 54, 188-210.

Ramirez, M. (1999). Multicultural psychotherapy: An approach to individual and cultural differences. Boston: Allyn and Bacon.

Redfield, R., Linton, B., & Herkovits, M. J. (1936). Memorandum on the study of acculturation. American Anthropologist, 28, 149-152.

Saleebey, D. (1992). The strengths perspective in social work practice. White Plaines: Longman.

Samaniego, R. Y., & Gonzales, N. A. (1999). Multiple mediators of the effects of acculturation status on delinquency for Mexican American adolescents. American Journal of Community Psychology, 27, 189-210.

Shen, B. J., & Takeuchi, D. T. (2001). A structural model of acculturation and mental health status among Chinese Americans. American Journal of Community Psychology, 29, 387-418.

Vega, W. A., Gil, A. G., Warheit, G. J., Zimmerman, R. S., & Apospori, E. (1993). Acculturation and delinquent behavior among Cuban American adolescents: Toward an empirical model. American Journal of Community Psychology, 21, 113-125.

Wutich, A. & McCarty, C. (2003, February). A personal network method for measuring race and ethnicity. Paper presented at XXIII International Social Network Conference, Mexico.

Zentella, A. C. (1990). Integrating qualitative and quantitative methods in the study of bilingual code switching. Annals of the New York Academy of Sciences, 583, 75-92.

Zimmerman, M. A. (2000). Empowerment theory: Psychological, organizational, and community level of analysis. In E. J. Rappaport & E. Seidman (Eds.), Handbook of community psychology (pp. 43-63). New York: Kluwer Academic.