International Journal of Intercultural Relations 34 (2010) 191–207

Contents lists available at ScienceDirect

International Journal of Intercultural Relations journal homepage: www.elsevier.com/locate/ijintrel

Acculturation, discrimination, and adaptation among second generation immigrant youth in Montreal and Paris John W. Berry a,*, Colette Sabatier b a b

Department of Psychology, Queen’s University, Kingston, Ontario, Canada K7L 3N6 Universite´ Victor Segalen, Bordeaux, France

A R T I C L E I N F O

A B S T R A C T

Article history: Accepted 17 November 2009

Research with immigrant youth has shown that they have a variety of orientations to their acculturation process. These different paths (‘‘acculturation strategies’’) have been described in terms of assimilation, integration, marginalisation and separation (AIMS). This research has also shown that there are variable adaptations or outcomes to acculturation. Of greatest importance is the finding that these variations in adaptation are usually related to the acculturation strategies that youth use. Generally, those who involve themselves in both their heritage culture and that of the national society (by way of integration) have the most positive psychological well-being, and are most adjusted in school and in the community; in contrast, those who are minimally involved with either culture (the marginalisation course), are least well-adapted; and those who are primarily oriented towards one or the other culture (assimilation or separation) generally fall in between these two adaptation poles. The present study examines the generality of this pattern by comparing samples from two societies that have very different policies and practices with respect immigration and acculturation: Canada and France. We measured acculturation attitudes, identity and behaviours along the two basic dimensions of cultural maintenance and social contact, and then cross-tabulated them, creating the four acculturation orientations. In both societies the usual pattern is replicated: youth who prefer integration have higher self-esteem scores than those who are marginalised; youth preferring assimilation and separation fall in between. This pattern was stronger in Canada than in France, a finding that is interpreted in terms of the differential social context faced by youth in the two societies, particularly the differing national policies and youths’ experience of discrimination. ß 2009 Elsevier Ltd. All rights reserved.

Keywords: Acculturation Adaptation Discrimination Immigrant Adolescents National context

1. Introduction Research with immigrant youth (e.g., Berry, Phinney, Sam, & Vedder, 2006) has shown that first and second generation immigrants take a variety of courses through their acculturation process. These different paths or acculturation strategies have been described by terms such as assimilation, integration, marginalisation and separation (AIMS). This research has also shown that there are variable outcomes to acculturation, including both psychological and sociocultural adaptation. Of greatest importance is the finding that these variations in adaptation are usually related to the acculturation strategies that youth use during their acculturation. Generally, those who involve themselves in both their heritage culture and that of the

* Corresponding author. E-mail addresses: [email protected], [email protected] (J.W. Berry). 0147-1767/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijintrel.2009.11.007

192

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

national society (by way of integration) have the most positive psychological well-being, and are most adjusted in school and in the community; in contrast, those who are minimally involved with either culture (the marginalisation course), are least well-adapted; and those who are primarily oriented towards on, or the other, culture (assimilation or separation) generally fall in between these two adaptation poles. In much of this research, the experience of discrimination (or even the perception of discrimination) has a major impact on the ways in which youth acculturate, and on the degree of adaptive success. For example Berry et al. (2006) found that youth who seek integration report the least amount of discrimination, while those who are marginalised have the most; those seeking separation have moderately high discrimination (but less than for the marginalised), and those seeking assimilation have moderately low discrimination (but more than for the integrated youth). With respect to the relationship between discrimination and adaptation, discrimination serves as the single strongest predictor; in a structural equation model (Berry et al., 2006) discrimination has a strong negative impact on both psychological and sociocultural adaptation. These patterns of relationships are present, generally, in a data base that includes immigrant youth in 13 societies of settlement. One question that arises is whether these general findings might vary according to some specific features of the society of settlement. In this 13-country international study, two such features were examined: the degree of actual cultural diversity of a society, and the level of policy support for such diversity. With respect to actual diversity (made up of a number of indicators, such as percentage of immigrants, and ethnolinguistic heterogeneity), greater diversity predicted somewhat greater perceived discrimination, and greater expression of ethnic behaviours; greater diversity also predicted somewhat lower psychological adaptation, but there was no relationship with sociocultural adaptation (Berry et al., 2006, Table 6.3). With respect to diversity policy (indicated by such factors as having a national multicultural policy, and agencies promoting and implementing diversity and equity), the higher the degree of policy support for cultural diversity, the higher the integration and assimilation orientations of immigrant youth, and the higher the level of sociocultural adaptation; however, there was no relationship with discrimination or psychological adaptation (Berry et al., 2006; Table 6.4). The present study examines the generality of this pattern of findings. We retain the same concepts (acculturation strategies, cultural identity, discrimination and adaptation), but operationalise them in different ways. We use different samples (focusing specifically on second generation immigrant youth), and on two cities (Montreal in Canada, and Paris in France) that have clearly defined differences in their policies with respect to acculturation. Such replications and extensions of previous studies can serve to establish the robustness of the generalizations made in earlier research. Samples in Canada and France were selected because they present some important differences that may impact youth acculturation, on their experience of discrimination, and on their adaptation (Noels & Berry, 2006; Sabatier & Boutry, 2006). These societal features include the level of cultural diversity, the national policy concerning cultural diversity, and school and family practices and values. Hence, the comparison may give some indication of the role that national contexts play in the acculturation and adaptation of immigrant youth (Sabatier & Berry, 1994). An initial comparison of immigrant youth acculturation and adaptation in Canada (Toronto) and France (Paris) was extracted from the Berry et al. (2006) study. With respect to acculturation strategies, youth in Toronto were somewhat more oriented toward integration than in Paris (50.4% vs. 45.6%), were less oriented toward separation (11.1% vs. 21.0%) and assimilation (22.1% vs. 26.3%), but were more marginal (16.4% vs. 7.2%) (x2 (3, N = 663) = 23.05, p < .001). With respect to cultural identity, youth in Toronto were higher than in Paris on both their ethnic identity (mean of 4.45 vs. 4.35, t = 2.04, p < .05) and national identity (4.02 vs. 3.29, t = 10.09, p < .001), and on perceived discrimination (2.00 vs. 1.90, t = 2.01, p < .05). Finally, with respect to psychological adaptation, youth in Toronto were less well-adapted than youth in Paris ( .21 vs. 0.02, t = 2.97, p < .01), but on sociocultural adaptation the youth did not differ. With respect to country similarities, both are modern, western democratic and individualistic societies, with a high level of affluence and quality of life. They have a long tradition of receiving immigrants and refugees and their populations are diverse, multilingual, multicultural and multiconfessional. The two research settings (Montreal and its suburbs and Paris area) are two dynamic metropolitan areas that offer a full range of ethnic associations and services with relatively easy access to them. There are social services devoted to helping with their problems, and children can learn their heritage language within the school system. These two areas receive the main proportion of immigrant for their country or Province. Differences between the two countries are situated along three axes: the public policy and public attitudes towards immigration and diversity; the socio-economic conditions of immigrants; and family and school education. France has a moderate score on an index of actual diversity and a moderate level of immigration (Berry et al., 2006). The French political model of immigration and diversity can be characterised as ‘‘assimilationist citizenship’’, where political, socio-economic and civil rights are respected; however, the maintenance of cultural rights is seen as an obstacle to the equality of rights in other domains (see Sabatier & Boutry, 2006). In contrast, Canada has a high score on an index actual diversity and has a high level of immigration. There is also a high degree of policy support for diversity beginning with the adoption of a policy of multiculturalism in 1971 (Noels & Berry, 2006). However, it should be noted that Montreal is in the Province of Quebec, where the Canadian national policy of ‘multiculturalism’ is modified to be a policy of ‘interculturalism’ (Bouchard & Taylor, 2008). The main difference is that in Canada, the policy states that ‘‘while there are two official languages, there are no official cultures’’ (Government of Canada, 1971), in Quebec, priority is given to French language and culture. Thus, the situation facing immigrant youth in Montreal, while largely multicultural, has some elements of assimilation; however, it falls well short of the assimilationist situation in Paris.

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

193

Fig. 1. Acculturation strategies in ethnocultural groups and the larger society.

2. Variables in the study 2.1. Acculturation strategies Up till now, we have referred to different ways of acculturating, using the terms, assimilation, integration, marginalisation and separation (AIMS) for immigrant youth, and multiculturalism (versus assimilation) for the policy orientations of the larger society. To make explicit how we used these terms, Fig. 1 displays them within two parallel acculturation spaces; on the left is that for non-dominant groups (e.g., immigrants and ethnocultural groups), and on the right is that for the larger national society. Early research assumed that acculturating individuals would orient themselves either to one group or the other, in a sense choosing between them; this view was the basis for assessing the acculturation of an individual on a single dimension, ranging from maintaining the heritage culture through to adopting the new culture (e.g., Gordon, 1964). However, an alternative view was proposed by Berry (1974, 1980), who argued that there are two dimensions along which individuals orient themselves. In this view, virtually everyone experiencing acculturation holds attitudes towards two fundamental aspects of acculturation: cultural maintenance and intercultural contact. For each issue, a dimension is shown, with a positive orientation at one end, and a negative one at the other. This two-dimensional conception has been validated in a number of studies (e.g., Arends-To´th & Van de Vijver, 2006; Ryder, Alden, & Paulhus, 2000). Among immigrants, acculturation attitudes are based on the intersection of these two issues: to what extent do immigrants or other non-dominant groups wish to have contact with (or avoid) others outside their group; and to what extent do they wish to maintain (or give up) their cultural attributes? When examined among the population at large (e.g., the dominant or national group), views about how immigrants and other non-dominant ethnocultural groups should acculturate have been termed acculturation expectations (Berry, 2003). For members of the non-dominant ethnocultural group (on the left), the main question is ‘‘How shall we deal with these two issues?’’ while for the larger national society (on the right) the question is ‘‘How should they (e.g., immigrants and ethnocultural groups) deal with them?’’ These two issues define an intercultural contact space (the circles) within which individuals occupy a preferred attitudinal position. Each sector of the circles in Fig. 1 carries a name that has a long-standing usage in acculturation studies. From the point of view of non-dominant ethnocultural groups (on the left of Fig. 1), when individuals do not wish to maintain their cultural heritage and seek daily participation with other cultures in the larger society, the assimilation attitude is defined. In contrast, when ethnocultural group members place a value on holding on to their original culture, and at the same time wish to avoid interaction with others, then the separation alternative is defined. When there is an interest in both maintaining one’s original culture and interacting with other groups, integration is the attitude; here, individuals maintain a degree of cultural integrity, while at the same time they seek to participate as an integral part of the larger society. Finally, when there is little possibility or interest in cultural maintenance (often for reasons of enforced cultural loss), and little interest in having relations with other groups (often for reasons of discrimination) then marginalisation is defined. Acculturation attitudes are presented from the point of view of non-dominant groups, shown on the left side of Fig. 1. This does not take into account attitudes of the larger society, which may play a role in influencing the way in which acculturation takes place (Berry, 1974; Berry, Kalin, & Taylor, 1977; Bourhis, Moı¨se, Perreault, & Sene´cal, 1997). The acceptance of integration along with cultural diversity and equitable participation by the larger society defines the attitude of mutual accommodation now widely called multiculturalism (Berry, 1984). When assimilation is preferred or sought by the dominant group, it can be termed the melting pot. When separation is enforced by the dominant group, it is segregation. When marginalisation is imposed by the dominant group, it is a form of exclusion (Bourhis et al., 1997). Attitudes toward the process of acculturation have been assessed frequently, using various methods (reviewed by Berry, 1997; Berry, Kim, Power, Young, & Bujaki, 1989). The most common procedure is to select a number of domains relevant to intercultural relations (e.g., language use, food preference, values, identity) and then create four statements for the various

194

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

domains, one for each of the four attitude sectors (e.g., Berry et al., 2006; Van de Vijver, Helms-Lornz, & Feltzer, 1999). Another is to create two statements for a particular domain, one for each of the two underlying dimensions and to classify individuals into the four categories by crossing the two dimensions (e.g., Dona` & Berry, 1994; Ryder et al., 2000). We adopt this second approach in this study and chose to cross the dimensions at the median point following the procedure adopted by Dona` and Berry (1994). By creating four acculturation attitude variables, instead of using the two underlying dimensions, we are able to make comparisons with much of the earlier literature that employed the four acculturation strategies. 2.2. Cultural identity Cultural identity refers to the ways in which individuals conceive of themselves in relation to cultural communities; it is a multidimensional and dynamic sense of self as a member of a particular group or groups including affective, cognitive and behavioural aspects (Phinney, 1990, 2003). Just as for acculturation orientations, the cultural identity of immigrants is now understood to have two independent aspects: identity with one’s heritage group (ethnic identity), and identity with the larger society in which a person now lives (national identity). And just as for acculturation attitudes, there is no necessary contradiction between these two identities; individuals can be high or low on the ethnic and national dimensions of identity (Berry et al., 2006; Table 4.1). 2.3. Ethnic behaviour Ethnic behaviour refers to the degree to which individuals keep elements of their culture and behave in accordance to their cultural customs. Among these elements are language retention, ethnic social networks, participation in ethnic associations, attending to media, reading newspapers, and other cultural habits such food (Berry et al., 1989; Knight, Bernal, Garzza, Cota, & Ocampo, 1993; Rosenthal & Cichello, 1986). Ethnic behaviour is a support and a context for identity and acculturation orientations. 2.4. Perceived discrimination Perceived discrimination has been found to be an important predictor of acculturation attitudes, and of how well immigrant youth adapt, both psychologically and socioculturally, to life in their new society (Berry et al., 2006). When adolescents perceive that they are targets of discrimination, they are less likely to orient themselves to the larger society (Berry et al., 2006), and instead to reject the larger society. They are also less likely to exhibit psychological well-being (Gil, Vega, & Dimas, 1994; Greene, Way, & Pahl, 2006; Virta, Sam, & Westin, 2004), and less able to function successfully in school and in the community (Vega, Khoury, Zimmerman, Gil, & Warheit, 1995). 2.5. Adaptation Adaptation to acculturation has become understood as having two components: psychological and sociocultural (Ward, 1996). Psychological adaptation refers to an individual’s personal sense of well-being. In this study, we examine psychological adaptation of immigrant adolescents in terms of four different aspects of self-esteem, and level of stress. Studies have indicated two levels of self-esteem or self-appraisal; one is a general appreciation of self-worth; the other one is domain-related. Immigrant and ethnocultural adolescents may vary in the importance they give to different domains; for example, Asians value more the academic self and Euro-American and African-American value more the physical appearance self than the other groups (Leiderman et al., 1989). While, family, school and peers are the main agencies of adolescent socialization, we assess both the general self-esteem and domain-related self-esteem in three domains (family, social relationships and school). Sociocultural adaptation refers to how well an individual is able to function in their daily lives in school or work, and generally in the community. In this study, we examine two aspects of adolescent sociocultural adaptation. Specifically, we assess adolescent academic performance and adolescent deviance, in terms of a number of activities, such as behaviour problems and the use of drugs. 3. Hypotheses In this study, we examine three main questions: How do immigrant youth acculturate; how well do they adapt; and are there important relationships between, how they acculturate and how well they adapt? These questions are addressed in two national settlement contexts to examine possible differences in how immigrant youth acculturate and how well they adapt in relation to these contextual differences. In addition we examine the role of discrimination in these relationships. Specifically, we hypothesised that: 3.1 The differences in acculturation strategies of youth between the two receiving societies (Canada and France) would be consistent with their policies with respect to acculturation. That is, immigrant youth in Montreal would pursue integration more than in Paris; and those in Paris would pursue assimilation more than those in Montreal.

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

195

3.2 The pattern of relationships between acculturation strategies and adaptation found in previous research with immigrant youth will be replicated using different measurement techniques and scales. That is, those pursuing integration would have more positive adaptation that those pursuing the other three ways of acculturating (assimilation, separation and marginalisation). 3.3 With respects to discrimination: 3.3.1—youth seeking integration and assimilation would experience less discrimination than those seeking separation and marginalisation; 3.3.2—immigrant youth in Montreal would experience less discrimination than those in Paris; 3.3.3—in both national contexts, immigrant youth experiencing more discrimination would have poorer adaptation.

4. Method 4.1. Participants Participants are 718 adolescents (53% are girls) born in the country of settlement (i.e., second generation). The sample was drawn from four second generation immigrant groups living in the Montreal area of Canada (N = 323), Greeks, Haitians, Italians and Vietnamese and five ethnic groups living in the Paris area of France (N = 395) Algerians, Antilleans, Moroccans, Portuguese and Vietnamese (see Table 1). The second generation was chosen because they face the task of making sense of their two cultural worlds, living between the heritage culture of their parents and their ethnocultural community and that of their society of settlement. These particular groups have been selected because either they represented the most important groups (in number) in Montreal and in the Parisian area (which is the case for Greeks, Italians, Algerians, Moroccans and Portuguese), or they are new visible minorities challenging the integration policies (which is the case for Haitians and Vietnamese). Antilleans are French nationals, but those who live in Paris area are geographical immigrants and may face discrimination based on the color of their skin. Adolescents were recruited on an individual basis through several means, mainly through school lists, but also neighbourhood networks. The mean age is 15.5 years (SD = 1.8, min = 11, max = 19). Most of the adolescents live in intact families (with both parents at home 84.9%). The number of eligible adolescents who received the information about the study is not exactly known, but the rate of participation can be estimated at about 66% on the basis of field notes of interviewers. 4.2. Procedure Adolescents filled out a questionnaire in a one-to-one interview at home with a same-ethnic interviewer. Because all adolescents attend Francophone schools, the adolescent questionnaire was presented in French. 4.3. Measures 4.3.1. Acculturation attitudes The construction of the two acculturation attitudes scales followed the procedure of Berry et al. (1989). The scales were developed using a pilot study with 15 second generation adolescents and a few parents from different ethnocultural groups in Montreal in order to identify the most important issues for them during the process of acculturation and adaptation. In addition to the behavioural domains such as food, language, and typical acculturation indices such as marriage and given name, the question of psychological acculturation as noted by Nguyen and von Eye (2002) and by Stevens, Pels, and Vollebergh (2004) appeared as crucial for adolescents especially concerning parent–child relationship. A first questionnaire had 60 questions with two sets of approximately 20 questions concerning each of the ethnic and national cultures; a third set Table 1 Participants in the study. Age

All Montreal Greeks Haitians Italians Vietnamese Paris Algerians Antilleans Moroccans Portuguese Vietnamese

Girls

Boys

N

M

SD

N

%

N

%

718

15.60

1.83

378

53

340

47

88 94 83 58

15.21 15.39 15.52 15.81

1.68 1.75 1.92 1.50

46 45 42 26

52 48 51 45

42 49 41 32

48 52 49 55

89 63 99 94 50

15.65 16.26 15.44 15.81 15.59

2.08 1.62 1.95 1.93 1.59

48 39 55 49 28

54 62 56 52 56

41 24 44 45 22

46 38 44 48 44

196

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

dealt with acculturation to a more global culture. This questionnaire was administered to 30 adolescents. The questions for ethnic acculturation were adapted to make items relevant to each specific ethnocultural group. A set of 45 questions was retained after factor analysis. Most of the questions dealing with a global acculturation orientation were discarded. Factor analysis and Tucker’s Phi calculations indicate that the two factor solution is preferable over a three factor solution, even after item normalization by each cultural group. Internal consistencies (Cronbach’s alpha) for the two samples were good for each cultural group for each dimension, but because the factor analysis revealed different loading across groups we decided to proceed to a series of exploratory and confirmatory factorial analysis in order to obtain the same factorial analysis for all groups with two orthogonal dimensions. This results in two sets of questions (one set for ethnic, and one set for national acculturation attitudes) dealing with various domains that are relevant to adolescents: language, marriage, social networks, values, emotions, parent–child relationship and cultural transmission. As a result, many parallel questions (one for national and one ethnic orientation) that appear to be on a bipolar dimension were discarded after this last step of analysis. By doing this we obtain a true orthogonality and avoid the question of bipolarity as mentioned by Nguyen and von Eye (2002). Two scales were constructed, with eight questions for national acculturation (a = .77, range across ethnic groups: .65–.79); 15 questions for ethnic acculturation (a = .86, range: .63–.91).1 Tucker’s Phi coefficients with the 23 questions for a two factors solution are high for each group, but poor with a three factors solution. Sample items are: - National acculturation attitude: ‘‘I believe that [ethnic] parents should make an effort for their children to develop ties with the Canadian society outside of school’’. - Ethnic acculturation attitude: I find it important that the [ethnic] culture be maintained from generation to generation. The two scales were then crossed at the median point of each scale, following the procedure used by Dona` and Berry (1994). This resulted in each individual being allocated to one of four acculturation attitudes: assimilation, integration, marginalisation or separation. In a discussion of various ways of splitting the two scales, Arends-To´th and Van de Vijver (2006) note that there is a choice between using the median, the mean or the theoretical mid-point, of each scale. There appears to be no ideal way of splitting the scales, since all three make an arbitrary cut near the middle of the scale where (in a normal distribution of scores) most respondents are likely to fall. 4.3.2. Cultural identity We used one question for national and one for ethnic identity (1–5 point). Adolescents evaluate how much ethnicity or national identity is one part of their self-concept: ‘‘To what degree do you feel yourself to be Canadian [or ethnic]?’’ 4.3.3. Ethnic behaviours Several questions were derived from Berry et al. (1989), Knight et al. (1993) and Rosenthal and Cichello (1986). These evaluate the commitment of adolescents in behaviours related to their cultural customs or which can sustain their ethnic identity. The three following dimensions were used to construct an index (1–5 point): proportion of ethnic friends, media and competence in ethnic language (a = .84; range: .60–.86). 4.3.4. Acculturation orientations An overall acculturation score (acculturation orientation) was created to provide an indication of how immigrant youth actually acculturate (including their behaviour and identity), which is more comprehensive than their attitudes or preferences. Factor analyses of the acculturation variables were run for each sample separately, and the weights for the acculturation variables were used to create two orientation scores. Ethnic acculturation orientation score is the combined factor score of ethnic acculturation attitudes, ethnic identity and ethnic behaviour (lowest loading .83 in Montreal, .75 in Paris; explained variance on a single factor, 73% in Montreal, 65% in Paris). The national acculturation orientation score is made up of national acculturation attitudes and national identity (lowest loading .82 in Montreal, .83 in Paris; explained variance on a single factor, 67% in Montreal, 70% in Paris). 4.3.5. Perceived discrimination Twelve questions (derived from Tchoryk-Pelletier, 1989) had Likert response options (1 not at all, to 5 almost always) that tapped two dimensions of discrimination: personal and group discrimination. Six items focus on personal discrimination within the school context and the neighborhood; six other items focus on group discrimination and their future (employment, police and social services, lodging, and efforts). Cronbach’s alphas are for the whole scale .91 (range: .76–.93), for personal discrimination .79 (range: .70–.80), for group discrimination .91 (range: .77–.92). Sample items are: Personal discrimination: ‘‘At school, do you have the feeling you are less accepted than Canadians because you are [ethnic]?’’ Group discrimination: ‘‘Do you have the feeling that your group is discriminated against when it is the time to find a job?’’

1

This questionnaire is available on request to the authors.

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

197

4.3.6. Self-esteem Two scales were used. The Rosenberg (1965) self-esteem scale (with a four point response scale) was used in both samples; it focuses on general self-esteem (a = .77; range: .66–.83). The other scale focuses on specific domains of selfesteem, and is different in each sample. In Paris, Coopersmith’s self-esteem inventory (Coopersmith, 1981) with four dimensions of self-esteem (emotional, familial, social, and school) was used. The scale has five response points (from 1 ‘not typical of me at all’, to 5 ‘completely typical of me’). Alphas are .84 for emotional (range: .81–.87), .80 for familial self-esteem (range: .61–.81), .66 for social (range: .57–.74), and .70 for school (range: .55–.78). In Montreal, the Offer Self-Image Questionnaire (Offer, Ostrov, Howard, & Atknson, 1988) was used. Following the procedure of Laukkanen, Halonen, Aivio, Viinama¨ki, and Lehtonen (2000) and Tremblay, Saucier, and Tremblay (2004) we extracted four dimensions, they are similar as those of the Coopersmith self-esteem inventory. Internal consistency (alpha and factor analysis) is good for all cultural groups. Alphas are .88 for emotional (range: .77–.83), .83 for familial self-esteem (range: .77–.83), .73 for social (range: .68– .74), and .77 for school (range: .75–.81). Both scales have been used successfully in cross-cultural studies with second generation adolescents (Herz & Gullone, 1999; Rosenthal & Cichello, 1986) and in each country. 4.3.7. Academic performance An index was constructed on the basis of several questions about academic performance (grades, peer comparison, academic track and appropriate age for the grade level). There are four levels, ranging from very problematic to good (even excellent) performance. Because of different school systems in the two countries, the classification system was adapted to each sample. 4.3.8. Deviance There were 11 questions (from Dornbusch, Ritter, Chen, & Mont-Reynaud, 1989), ranging from minor behaviour problems (being late at school) to more serious ones (using drugs or having trouble with the police). Responses were made on a Likert response scale (1 never to 5 often); a is .73 (range: .64–.74). 5. Results 5.1. Strategy of analysis The first part of this section presents the descriptive statistics for all the variables in the study and the correlations among intercultural variables. The second presents the distribution of scores for the four acculturation strategies derived from the two acculturation attitude dimensions, and of the two overall acculturation orientations (based on the factor scores of acculturation attitudes, identity and cultural behaviour). In the third part, we examine the relationships between the four acculturation strategies and adaptation variables. In the fourth part we examine relationships between perceived discrimination and the four acculturation strategies. And in the last part, we examine the relationships between perceived discrimination and adaptation variables. Because we are mainly concerned with differences between societies of settlement, the analyses report systematically the country level differences and analyses; however, when it is appropriate and when they illustrate the special situation concerning specific immigrant groups, these are reported, in the text (but they do not appear in tables). 5.2. Descriptive data Means, medians and standard deviations for all variables are presented in Table 2. Multiple analyses of variance reveal that the three ethnic acculturation variables (acculturation attitudes, ethnic identity and ethnic behaviours) are higher in Montreal than in Paris; however, there is no difference on national acculturation attitudes or identity. Perceived discrimination is higher in Paris. Self-esteem is higher in Montreal (for Rosenberg, where the same scale was used in both samples). The means on the other self-esteem measures cannot be directly compared (because somewhat different measures were used, see note b). Correlations among intercultural variables are shown in Table 3. In general, as expected the same-focus variables (ethnic or national) are positively related in both samples: these intercorrelations provided the basis for creating the two acculturation orientation factor scores, as noted above. For example, ethnic acculturation attitudes and ethnic identity are correlated +.55 in Paris and +.65 in Montreal; national acculturation attitudes and national identity are correlated +.39 and +.33 in Paris and Montreal, respectively. In contrast, some other-focussed variables are negatively related in Paris ( .20 for ethnic and national identities, and .21 for national identity and ethnic acculturation attitudes), but they are not related in Montreal. This pattern corresponds to the identity correlations in Berry et al. (2006), where the two identities were negatively correlated in Paris ( .13), but not related in Montreal (+.09). 5.3. Acculturation strategies To examine the first question guiding this study (how immigrant youth acculturate), and to assess the first hypothesis (with respect to sample differences in these acculturation strategies), we employed two measures. The first

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

198

Table 2 Means and standard deviations of acculturation, discrimination and adaptation variables (MANOVA). Montreal

Paris

F

p

M

Md

SD

M

Md

SD

Ethnic acculturation attitudes National acculturation attitudes Ethnic identity National identity Ethnic behaviour

4.12 3.38 4.02 3.33 3.36

4.20 3.38 4.00 3.00 3.44

0.64 0.70 1.05 1.16 0.97

4.00 3.48 3.87 3.32 2.78

4.00 3.50 4.00 3.00 2.70

0.63 0.75 0.95 1.08 0.87

6.02 2.28 4.31 0.25 67.91

0.014 0.131 0.038 0.616 0.000

Perceived discrimination Personal discrimination Group discrimination

1.98 1.67 2.29

1.83 1.33 2.00

0.83 0.74 1.14

2.11 1.75 2.48

1.92 1.50 2.33

0.94 0.85 1.18

4.41 2.50 4.70

0.036 0.115 0.031

Adaptation variables Self-esteem Rosenberg scalea Emotionalb Familialb Socialb Schoolb

3.25 4.61 4.72 4.71 5.16

3.30 4.81 4.89 4.89 5.25

0.47 0.78 0.71 0.75 0.59

3.07 3.67 3.74 3.76 3.49

3.10 3.69 3.83 3.86 3.50

0.48 0.55 0.88 0.64 0.75

27.82

0.000

2.71 1.27

3.00 1.20

0.98 0.29

2.75 1.39

3.00 1.30

0.96 0.37

0.39 18.79

0.532 0.000

Academic performancea Deviance a b

The range of variables is for all 1–5 except range is 1–4. The range of variables is for all 1–5 except differents scales are used for each country, no comparison is possible, range 1–5 in Paris, 1–6 in Montreal.

Table 3 Correlations among identity and acculturation variables. 1

2

3

4

5

6

All 1 2 3 4 5 6 7

Ethnic acculturation attitudesa National acculturation attitudesa Ethnic identity National identity Ethnic behaviour Ethnic orientationb National orientationb

1 0 0.60*** 0.06+ 0.55*** 0.87*** 0.037

0.07+ 0.36*** 0.13*** 0.07+ 0.82***

0.09* 0.45*** 0.82*** 0.09*

0 0.07+ 0.83***

0.76*** 0.07+

0.09*

Montreal 1 2 3 4 5 6 7

Ethnic acculturation attitudesa National acculturation attitudesa Ethnic identity National identity Ethnic behaviour Ethnic orientationb National orientationb

1 0 0.65*** 0.10+ 0.66*** 0.90*** 0.062

0.09 0.33*** 0.06 0.06 0.82***

0.03 0.49*** 0.83*** 0.04

0.19** 0.13* 0.82***

0.84*** 0.08

0.04

Paris 1 2 3 4 5 6 7

Ethnic acculturation attitudesa National acculturation attitudesa Ethnic identity National identity Ethnic behaviour Ethnic orientationb National orientationb

1 0 0.55*** 0.21*** 0.45*** 0.84*** 0.12*

0.05 0.39*** 0.17** 0.08 0.83***

0.20*** 0.41*** 0.82*** 0.15**

0.19*** 0.25*** 0.83***

0.76*** 0.21***

0.20***

a

Ethnic and national acculturation attitudes combine responses to items that were on two independent factors; hence their correlations are zero. b Ethnic and national orientation variables are factor scores that include ethnic and national variables; hence their correlations are inflated by nonindependence. + p < .10. * p < .05. ** p < .01. *** p < .001.

uses the responses on the two acculturation attitude scales (national and ethnic), which were then split at the median to understand their preferences for the four ways of acculturating. The distributions in Paris and Montreal, and among ethnic groups, were calculated (see Table 4); the scattergrams of the bivariate distribution appear in Fig. 2. The distributions in both samples are rather even (x2 (3, N = 718) = 6.27, p = .09), with all four attitudes falling close together

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

199

Table 4 Acculturation strategies categorised by median split of acculturation attitudes (N and %); means on the two underlying attitude dimensions. N

All Assimilation Integration Marginalisation Separation Total Montreal Assimilation Integration Marginalisation Separation Total Paris Assimilation Integration Marginalisation Separation Total

171 201 182 164

%

23.8 28.0 25.3 22.8

718

65 94 80 84

20.1 29.1 24.8 26.0

323

106 107 102 80 395

26.8 27.1 25.8 20.2

National

Ethnic

M

SD

M

SD

4.03 3.98 2.87 2.78

0.39 0.39 0.43 0.52

3.50 4.59 3.55 4.54

0.48 0.26 0.40 0.25

3.44

0.73

4.05

0.63

4.03 3.91 2.84 2.82

0.40 0.34 0.43 0.50

3.47 4.65 3.57 4.54

0.50 0.26 0.39 0.24

3.38

0.70

4.12

0.64

4.03 4.04 2.90 2.74

0.38 0.42 0.43 0.54

3.51 4.53 3.53 4.55

0.47 0.25 0.40 0.26

3.48

0.75

4.00

0.63

in a range between a low of 20.2% for Separation in Paris to a high of 29.1% for integration in Montreal. However, there was wide variation in both countries across ethnocultural groups reflecting the diversity of acculturation within the same society of settlement. In Paris Assimilation was particularly high among Vietnamese and Portuguese, integration high among Algerians, Marginalisation high among Antilleans, and Separation high among Moroccans and Antilleans (x2 (12, N = 395) = 101.25, p < .001). In Montreal, Assimilation was again high among Vietnamese (but almost absent among Greeks), integration was high among Greeks, and Marginalisation and separation were high among Haitians (x2 (9, N = 323) = 81.05, p < .001). The second measure of acculturation strategies was the acculturation orientation factor scores. These are used to create the four acculturation strategies. With this second measure, a similar pattern emerges (x2 (3, N = 706) = 12.63, p < .01), with a rather even distribution (see Table 5). The lowest percentages are for separation in Montreal and for integration and Marginalisation in Paris (all 20.1%), and the highest is for Assimilation in Paris (29.7%). The differences between the two ways of calculating the four acculturation strategies (attitudes vs. the orientation factor scores) are rather small (Kappa coefficient: for Montreal .69, for Paris .49). 5.4. Relationship between acculturation strategies and adaptation variables In this section we examine our third question and second hypothesis (relationships between acculturation strategies and different forms of adaptation). As noted earlier, previous literature (Berry, 1997; Berry et al., 2006) concluded that those pursuing integration had more positive adaptations than those seeking to acculturate in other ways. Least positive were those with marginalisation attitudes, and in between (the order depending on many factors) were those who preferred assimilation or separation. Using the median split method of allocating acculturation attitudes outlined earlier (see Table 6), there is a fairly consistent pattern of adaptation variables having better (more positive adaptation) scores among those preferring integration than among those with the other three ways of acculturating, and they are especially worse among those preferring marginalisation (this is particularly true for school self-esteem in both samples). This pattern reaches significance for all adaptation variables in Montreal (except for deviance), but in Paris, the variation is significant only for social, and for school self-esteem. When the acculturation orientation factor scores are used to generate the four acculturation strategies, the same pattern is found (see Table 7). However, variations in all adaptation variables are now significant in Montreal; and Rosenberg selfesteem is now significant in Paris. Overall, this pattern of relationships between ways of acculturating and adaptation corresponds with the pattern found in much of the earlier research using a different method (i.e., four separate scales) of assessing acculturation strategies. With respect to discrimination we hpothesized: youth seeking integration and assimilation would experience less discrimination than those seeking separation and marginalisation; immigrant youth in Montreal would experience less discrimination than those in Paris; and in both national contexts, immigrant youth experiencing more discrimination would have poorer adaptation.

200

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

Fig. 2. Dispersion diagram of the four classes of acculturation based upon the median.

5.5. Perceived discrimination The mean personal and group discrimination by samples were presented in Table 2. In general, personal discrimination is low (mean of 1.75 in Paris, and 1.71 in Montreal on a five point scale); this difference is not significant. However, group discrimination is higher in Paris (2.48) than in Montreal (2.39, p < .05). Analyses (not in tables) revealed that 72% of adolescents do not feel discriminated against at all by their peers, but 57% think they have to make more effort than their peers because of their origin. Within each sample, there are some variations across domains. For personal discrimination, there are significant differences for discrimination by peers (with greater discrimination perceived in Montreal) and by the police (with greater discrimination perceived in Paris). For group discrimination, there are more sample differences, with adolescents in Paris perceiving more discrimination in the domains of dealing with employment, lodging and social services/ police. The multiple group comparisons (Tukey post hoc test) among ethnic groups (both samples together) of perceived group discrimination indicate three distinct clusters. The groups who feel the most discriminated against are the three visible and culturally distant groups, in decreasing order: Haitians, Moroccans and Algerians (Haitians: M = 3.46, SD = 0.97; Moroccans: M = 3.14, SD = 1.32; Algerians: M = 2.66, SD = 1.23). The groups that feel less discriminated against are immigrants from European countries: in increasing order, Italians and Portuguese and Greeks (Italians: M = 1.61, SD = 0.89; Portuguese:

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

201

Table 5 Acculturation strategies categorised by median split of acculturation orientation factor scores. N

All Assimilation Integration Marginalisation Separation

%

201 173 153 179

Total

National

28.5 24.5 21.7 25.3

706

Montreal Assimilation Integration Marginalisation Separation

86 94 74 65

Total

27.0 29.5 23.2 20.4

319

Paris Assimilation Integration Marginalisation Separation

115 79 79 114

Total

29.7 20.4 20.4 29.5

387

Ethnic

M

SD

M

SD

3.87 3.87 2.91 2.98

0.52 0.56 0.52 0.68

3.60 4.53 3.64 4.47

0.53 0.37 0.51 0.37

3.44

0.73

4.06

0.63

3.87 3.73 2.83 2.89

0.51 0.53 0.44 0.63

3.61 4.65 3.69 4.52

0.56 0.30 0.49 0.29

3.39

0.71

4.12

0.64

3.87 4.03 2.99 3.03

0.52 0.54 0.58 0.70

3.59 4.38 3.60 4.45

0.52 0.39 0.53 0.41

3.48

0.75

4.01

0.62

Table 6 Means of adaptation variables by acculturation strategies based on attitudes. Assimilation M Montreal, N Self-esteem Rosenberg scale Emotional Familial Social School Academic perf. Deviance Paris, N Self-esteem Rosenberg scale Emotional Familial Social School Academic perf. Deviance

Integration SD

65

M

Marginalisation SD

94 ab

M

Separation SD

80

F SD

84

3.21 4.44a 4.55ab 4.70 5.23b

0.41 0.86 0.82 0.84 0.62

3.38 4.86b 4.95c 4.89 5.23b

0.47 0.58 0.53 0.63 0.51

3.12 4.42a 4.50a 4.59 4.96a

0.44 0.87 0.73 0.70 0.60

3.27ab 4.64ab 4.79bc 4.64 5.22b

0.52 0.76 0.70 0.83 0.60

4.55** 6.13*** 7.66*** 2.80* 4.45**

2.97b 1.29

1.00 0.29

2.74ab 1.28

0.82 0.30

2.49a 1.27

1.02 0.25

2.67ab 1.26

1.03 0.33

2.96* 0.21

106

b

M

107

a

102

80

3.06 3.65 3.77 3.73a 3.52b

0.45 0.51 0.84 0.57 0.74

3.16 3.75 3.90 3.98b 3.61b

0.50 0.60 0.83 0.68 0.75

2.99 3.58 3.59 3.61a 3.30a

0.41 0.48 0.82 0.59 0.66

3.04 3.69 3.66 3.71a 3.54b

0.53 0.59 1.06 0.65 0.83

2.17 1.66 2.41 6.58*** 3.22*

2.91 1.37

1.00 0.44

2.63 1.35

0.97 0.29

2.64 1.44

0.99 0.34

2.81 1.39

0.85 0.37

2.07 1.17

Note: Subscripts a, b and c indicate the homogeneity subsets for each variable across acculturation strategies according to Tukey’s post hoc test. * p < .05. ** p < .01. *** p < .001.

M = 1.82, SD = 0.77; Greeks: M = 1.86, SD = 0.67). In between are the two Vietnamese groups and the Antilleans; although visible minorities, they are familiar with the French language and culture. Living in Quebec and in France provides them with a context where cultural distance is reduced compared to the groups that are more highly discriminated against (Vietnamese in Montreal: M = 2.00, SD = 0.85; Vietnamese in Paris: M = 2.08, SD = 0.72; Antilleans: M = 2.46, SD = 1.08). For personal discrimination, the picture is less clear, especially within the school context. All the groups in Montreal feel more discriminated against by peers than the groups in Paris. Haitians, Greeks, Antilleans and Moroccans feel more discriminated against by teachers than other groups. Portuguese and Vietnamese in Paris feel that they are less discriminated against. With respect to discrimination, we proposed three hypotheses. First, in both samples, youth seeking integration and assimilation would experience less discrimination than those seeking separation and marginalisation. Second, we hypothesised that immigrant youth in Montreal would experience less discrimination than those in Paris. Third, we

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

202

Table 7 Means of adaptation variables by acculturation strategies based on acculturation orientation (factor scores). Assimilation M Montreal, N Self-esteem Rosenberg scale Emotional Familial Social School

SD

86

Academic perf. Deviance Paris, N Self-esteem Rosenberg scale Emotional Familial Social School

Integration SD

94

Separation

M

M

SD

74

F SD

65

3.22a 4.53a 4.60a 4.71ab 5.18ab

0.44 0.74 0.70 0.71 0.59

3.46b 4.89b 5.05b 4.94b 5.31b

0.44 0.57 0.49 0.64 0.44

3.06a 4.34a 4.50a 4.53a 5.01a

0.47 0.94 0.77 0.83 0.64

3.22a 4.61ab 4.65a 4.60a 5.14ab

0.47 0.81 0.77 0.81 0.63

11.25*** 7.59*** 11.19*** 4.90** 4.09**

2.88 1.29

0.99 0.29

2.72 1.21

0.85 0.23

2.49 1.27

1.01 0.28

2.69 1.33

1.03 0.37

2.14 2.26+

115

Academic perf. Deviance

M

Marginalisation

79

79

114

3.00 3.60 3.80 3.71ab 3.42

0.47 0.53 0.80 0.62 0.68

3.14 3.77 3.84 3.92b 3.63

0.46 0.61 0.89 0.70 0.78

2.97 3.61 3.63 3.61a 3.35

0.44 0.53 0.81 0.62 0.75

3.14 3.69 3.65 3.79ab 3.56

0.50 0.55 1.00 0.61 0.78

3.24* 1.75 1.28 3.35* 2.45+

2.88 1.37

0.98 0.33

2.78 1.36

0.96 0.45

2.62 1.38

1.02 0.33

2.70 1.43

0.92 0.37

1.27 0.67

Note: Subscripts a and b indicate the homogeneity subsets for each variable across acculturation strategies according to Tukey’s post hoc test. + p < .10. * p < .05. ** p < .01. *** p < .001.

hypothesised and in both national contexts, immigrant youth experiencing more discrimination would have poorer adaptation. 5.5.1. Relationship between acculturation strategies and perceived discrimination For the first of these hypotheses, perceived discrimination means according to acculturation strategies and samples are presented in Table 8 and Fig. 3. The pattern of means differs between Paris and Montreal: in Paris, both personal and group discrimination are highest for integration and separation, and are lowest for Assimilation and Marginalisation; in sharp contrast, in Montreal both discrimination scores are lowest for integration (while remaining high for separation).

Table 8 Personal and group perceived discrimination according to acculturation strategies based on attitudes. Personal

All Assimilation Integration Marginalisation Separation Total Montreal Assimilation Integration Marginalisation Separation Total Paris Assimilation Integration Marginalisation Separation Total

Group

Personal

Group

N

M

SD

M

SD

F

p

F

p

171 199 182 164

1.49a 1.71a 1.67a 1.99b

0.65 0.84 0.69 0.94

2.09a 2.49b 2.19ab 2.81c

0.95 1.28 1.03 1.25

11.63

0.000

13.75

0.000

716

1.71

0.80

2.39

1.17

1.58ab 1.39a 1.81bc 1.90c

0.77 0.46 0.74 0.85

2.16ab 1.96a 2.42bc 2.63c

1.19 0.96 1.2 1.14

9.09

0.000

5.98

0.001

323

1.67

0.74

2.29

1.14

106 105 102 80

1.44a 1.99b 1.56a 2.09 b

0.57 0.99 0.64 1.01

2.04a 2.96b 2.01a 3.00b

0.77 1.34 0.83 1.34

14.74

0.000

24.73

0.000

393

1.75

0.85

2.48

1.18

65 94 80 84

Note: Subscripts a, b and c indicate the homogeneity subsets for personal and group discrimination respectively according to Tukey’s post hoc test within each context.

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

203

Fig. 3. Mean discrimination by acculturation strategies and by country.

Table 9 Correlations among acculturation strategies and total, personal and group discrimination. Montreal Overall

Paris Personal

Group

Overall

Personal

Group

Acculturation attitudes Ethnic National

0.02 0.30***

0.02 0.32***

0 0.23***

0.40*** 0.11*

0.31*** 0.13**

0.41*** 0.08+

Identity Ethnic National

0.03 0.34***

0 0.30***

0.05 0.30***

0.29*** 0.31***

0.23*** 0.25***

0.30*** 0.31***

Acculturation orientation (factor) Ethnic 0.02 National 0.39***

0 0.38***

0.03 0.33***

0.41*** 0.25***

0.34*** 0.23***

0.40*** 0.24***

+

p < .10. p < .05. ** p < .01. *** p < .001. *

These differences become clear when acculturation variables (strategies based on both attitudes and factor scores, and identity) are related to discrimination (Table 9). An ethnic acculturation orientation in Paris is positively correlated with both forms of discrimination (for personal and group), but are unrelated in Montreal. That is, since both integration and separation strategies involve a degree of cultural maintenance, in Paris those seeking to acculturate in these ways seem to run up against a barrier of discrimination. The same holds for cultural identity: having an ethnic identity in Paris is associated with higher perceived discrimination, but they are unrelated in Montreal. 5.5.2. Sample differences in perceived discrimination With respect to the second hypothesis relating to discrimination (see Table 2), overall discrimination was higher in Paris than in Montreal. However, this was the case for only one component (group discrimination), not for personal discrimination. 5.5.3. Relationship between perceived discrimination and adaptation variables With respect to the third hypothesis relating to discrimination, regarding the relationships between perceived discrimination and various forms of adaptation (academic performance, deviance and self-esteem), there is a variable pattern. In Table 10, there are no significant relationships between discrimination and school performance in either sample. In contrast, there are significant relationships between deviance and both kinds of perceived discrimination in both samples, but they are stronger in Paris. Analyses (not in tables) also indicate an ethnic group variation. In Paris, correlations between deviance and personal discrimination vary from .11 for Portuguese to .34 for Vietnamese (p < .05 for all groups except Portuguese), and with group discrimination, ranging from .08 for Vietnamese to .32 for Antilleans (p < .05 for Algerians, Antilleans and Moroccans). In Montreal correlations between deviance and personal discrimination vary from .02 for

204

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

Table 10 Correlations among perceived discrimination and adaptation variables. Montreal

Paris

Personal

Group

Self-esteem Rosenberg self-esteem Emotional self-esteem (anxiety) Familial self-esteem Social self-esteem School self-esteem

Personal

Group

0.13* 0.20*** 0.20*** 0.21*** 0.20***

0.04 0.11* 0.15* 0.12* 0.05

0.04 0.04 0.13* 0.04 0.09

0.10+ 0.02 0.11* 0.08 0.01

Academic performance Deviance

0.06 0.13*

0.02 0.13*

0.04 0.19***

0.02 0.16***

+

p < .10. p < .05. *** p < .001. *

Vietnamese to .18 for Italians (but none reach significance), and with group discrimination, ranging from .13 for Haitians and Vietnamese to .30 for Greeks (p < .05 only for Greeks). There is also a link between perceived discrimination (personal and group) and several indices of self-esteem. Familial self-esteem is negatively correlated with personal and group perceived discrimination at the sample level in Paris and in Montreal. The Rosenberg self-esteem is negatively correlated with personal discrimination in Montreal; however, it is interesting to note that in Paris there is a tendency for a positive correlation with group discrimination. In Montreal, social and emotional self-esteem are negatively correlated with the perceived discrimination (personal and group), while school self-esteem is correlated only with personal perceived discrimination. At the cultural group level, school self-esteem is negatively correlated with personal discrimination but not the group discrimination in the five groups who are not or moderately discriminated against. In addition to this pattern, school selfesteem of Portuguese is correlated with group discrimination. For the three groups who are highly discriminated against (i.e., Algerians, Haitians and Moroccans) and the Vietnamese in Paris, there is no correlation between school self-esteem and perceived discrimination. Notably, there are more correlations between self-esteem and perceived discrimination in the Canadian groups than in French ones. 6. Discussion This study addressed three core questions: how do youth in Montreal and Paris acculturate; how well do they adapt; and are there any relationships between their acculturation attitudes and strategies, and their level of adaptation? These questions were explicitly compared in two national contexts; the two societies were contrasted by their different approaches to immigrant acculturation. One (Canada) is highly culturally diverse and has a form of multiculturalism policy (interculturalism); the other (France) is less diverse, and has an assimilationist policy. In addition, we examined whether the experience of personal and group discrimination differs between these two contexts, and whether such discrimination is related to acculturation and to adaptation. Evidence relating to the first question and hypothesis shows that ethnic acculturation strategies (attitudes, identity and behaviour) are higher in Montreal than in Paris, but that there are no differences in national acculturation (Table 2). This pattern is consistent with the promotion of multiculturalism (and interculturalism in Quebec) in Canada, where the maintenance and expression of ones cultural heritage is encouraged and officially supported. However, despite the promotion of the idea of ‘‘one nation’’ in France (but not in Canada), national acculturation is not greater there. When these two acculturation attitudes are crossed to create the four acculturation strategies (Table 4), the percentage of youth categorised as ‘integration’ in Montreal is slightly higher than in Paris (29.1% vs. 27.1%) for attitudes alone but is considerably higher (29.5% vs. 20.4%) when using the orientation factor scores (composite score). In contrast, youth categorised as ‘assimilation’ are higher in Paris than in Montreal (26.8% vs. 20.1%) for attitudes alone, but there is only a small difference (29.7% vs. 27.0%) using the orientation factor scores. Again, these differences in acculturation strategies are consistent with the different policy orientations between the two countries. With respect to our hypothesis regarding the relationship between discrimination and acculturation strategies (3.3.1) is examined (Tables 8 and 9), in Paris discrimination is highest for those categorised as separation and integration, and lowest for those categorised as assimilation and marginalisation. This appears to show that retaining ones heritage culture in France is a basis for experiencing more discrimination. This is borne out by the correlations in Paris between ethnic acculturation variables (attitudes, identity and factor scores) and discrimination: all coefficients are positive. In contrast, in Montreal, none of the correlation coefficients is positive. That is, being oriented toward one’s ethnic group in Canada does not serve as a basis for higher discrimination, while it appears to be so in France. Again, this difference is consistent with the policy and attitude differences between the two countries: cultural and identity maintenance are considered to be valuable in Canada, whereas

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

205

in France these are in many ways more or less discouraged, the maintenance of culture is seen as an obstacle to equality of chance. For national acculturation, as expected, all correlations with discrimination are negative in both samples. However, these negative coefficients are somewhat larger in Montreal. It appears that being oriented toward the national society is a basis for experiencing less discrimination, especially in Canada. These different relationships in Montreal and Paris show that the finding of Branscombe, Schmitt and Harvey (1999) may be context-dependent. In their study, the experience of discrimination among African-Americans strengthened their ethnic identity and weakened their ties to the national society. This pattern is evident in Paris, but not in Montreal, where discrimination does not enhance an ethnic orientation or identity. With respect to our hypothesis regarding sample differences in perceived discrimination (3.3.2), the overall mean score was higher in Paris than in Montreal; however, this difference was the case only for group discrimination, not for personal discrimination. This difference coincides with reports of public attitudes towards immigrants in Canada (Adams, 2007) and France (CNCDH, 2006). It is interesting to note that in the 13-country study, perceived discrimination was higher in Canada than in France; this is perhaps due both to the lower number of those seeking ‘integration’ and ‘separation’ in France, among whom perceived discrimination is higher and to the measurement of discrimination. In the 13-country study the perceived discrimination focused mainly on personal discrimination while in this study the measure focus both on personal and group discrimination. With respect to the negative relationship between discrimination and orientation towards, and identification with, the national society, the ‘multiculturalism hypothesis’ (see Berry, 2006 for an overview) appears to be supported. This hypothesis asserts that only when individuals are secure in their own identity will they be in a position to accept those who are different from themselves; conversely when people are threatened (for example by pervasive discrimination) they will reciprocate their rejection by the larger society by expressing a negative view towards those who discriminate against them. This pattern is present in both samples. With respect to sample differences in adaptation, these were assessed with two scales: Rosenberg Self-Esteem, and Deviance. For both variables, second generation youth in Montreal are more adapted than youth in Paris. Given the role of discrimination in the lower adaptation reported in previous research (e.g., Berry et al., 2006), we expected that such experience would be related to poorer adaptation in both samples. When discrimination is related to adaptation (Table 10), only some of the correlations reached significance; but when they did, they were in the expected direction (8 were significant in Montreal, and 4 were in Paris). This pattern is consistent with earlier analyses reported by Sabatier and Berry (2008); using multiple regression analyses, and taking out the effects of age and gender, discrimination tended to diminish adaptation more in Montreal than in Paris. With respect to our second main hypothesis (3.2), does it matter how youth acculturate? The general pattern (Tables 6 and 7) shows that there is significant variation across the four acculturation strategies (using both their attitudes and orientation factor scores). This variation is present in Montreal for five of six adaptation variables using attitudes alone, and for all six adaptation variables using factor scores. In Paris, this variation is present for two of the adaptation variables using attitudes (and for three using factor scores). We may conclude that it does indeed matter how youth engage their acculturation. What does this variation show? For all adaptation variables, and in both samples, those youth who are categorised as ‘integration’ have numerically higher adaptation scores, and in nearly almost all cases, those who are categorised as ‘marginalisation’ have lower adaptation scores. Where there are significant differences between specific acculturation strategies, they are present mostly for the contrast between integration and marginalisation. This general pattern corresponds with much of the previous research with adults (reviewed by Berry, 1997) and with youth (Berry et al., 2006). Adaptation scores for the other two acculturation strategies (assimilation and separation) generally fall in between these two end points. Sometimes one has numerically better adaptation scores, sometimes the other does; there are fewer specific significant differences between these two ways of acculturating. Overall, we may conclude that acculturation strategies and adaptation are related in important ways. This is relatively more the case in Montreal than in Paris. With respect to our hypothesis regarding the relationship between discrimination and adaptation (3.3.3), the evidence (Table 10) shows that when there are significant correlations, they generally impacted adaptation negatively. For example in Montreal 9 of 14 correlations were significant (mostly for personal discrimination); and in Paris 4 of 14 correlations were significant (2 each for personal and group discrimination). The finding that discrimination is high for those youth seeking to integrate in Paris (Table 8) signals an important divergence from the general finding in the literature (e.g., Berry et al., 2006); it is also a difference from the finding in this study for Canada (where integration youth generally have the lowest discrimination). What is the situation facing youth in France that makes integration more a risk for discrimination? One clue is that discrimination scores are even higher for youth in Paris who are categorised as ‘separation’. Integration and separation acculturation strategies both involve a high preference for ‘cultural maintenance’ (see Fig. 1). This preference contradicts the general policy orientation in France (favouring ‘assimilation’), and may well draw negative attention from the larger society to both the individual and the ethnocultural group. 7. Conclusion The three questions guiding this study of immigrant youth received answers that are both consistent with and divergent from the literature pertaining to adult immigrants. With respect to the first question, it is clear that immigrant youth hold

206

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

varying views about how they wish to acculturate and about their cultural identities. The two dimensions of interest (cultural maintenance and social participation), and the fourfold conceptualisation of ways of acculturating based upon them, appear to serve as a sensible way for youth to express their preferences. Moreover, the oft-found preference for integration in the adult literature also appears to be supported by the distribution of responses among immigrant youth. Consistent with the view of Arends-To´th and Van de Vijver (2006), we conclude that there can be no doubt that a twodimensional approach is most appropriate for understanding these preferences. Identifying more with one’s ethnic group is not conceptually, nor empirically, opposed to identifying with one’s national society. The unidimensional approach, where individuals must choose between the two poles of acculturation strategies and cultural identities, does not capture the complexity of how youth work out their new lives in their new societies. Second, there is support for the two forms of adaptation: psychological well-being and sociocultural competence are conceptually and empirically distinct among immigrant youth, and they have different predictors. Finally, the pattern of relationships between how youth acculturate and how well they adapt largely replicates the findings with adult immigrants. Those seeking to integrate adapt better than those who are marginalised, with assimilation and separation ways falling in between. However, this set of relationships is stronger and more consistent in Montreal than in Paris. This difference appears to match the differences in public policy and public attitudes between the two societies: it is more difficult to adapt well when one’s ethnicity is being questioned. Finally, we acknowledge the limitations of a cross-sectional approach to the study of the process of acculturation. Ideally, longitudinal research is required to better understand the relationships, and causal links between acculturation strategies and experiences and adaptation. To our knowledge, such a comprehensive longitudinal study has not yet been carried out. To the extant that structural modeling with large data sets is able to approximate causal links among variables, the results from Berry et al. (2006, Fig. 6.2) suggest that acculturation strategies contribute to ethnic contacts, as well as to both psychological and sociocultural adaptations. However, by far the most important contributor to both forms of adaptation was the experience of discrimination. If these interpretations can be generalized to the present study, we may conclude that indeed adaptation is the outcome of acculturation strategies and discrimination.

Acknowledgments This project is supported by a grant to the authors from SSHRC (Social Sciences and Humanities Research Council, Canada) and to the second author by INSERM (Institut national de la sante´ et de la recherche me´dicale, France). References Adams, M. (2007). Unlikely utopia: The surprising triumph of pluralism in Canada. Toronto: Viking. Arends-To´th, J., & Van de Vijver, F. J. R. (2006). Issues in the conceptualisation and assessment of acculturation. In M. Bornstein & L. Cote (Eds.), Acculturation and parent–child relationships (pp. 33–62). Mahwah: Lawrence Erlbaum Associates. Berry, J. W. (1974). Psychological aspects of cultural pluralism. Topics in Culture Learning, 2, 17–22. Berry, J. W. (1980). Acculturation as varieties of adaptation. In P. M. Padilla (Ed.), Acculturation: Theory, models and some findings (pp. 9–25). Boulder, CO: Westview. Berry, J. W. (1984). Multicultural policy in Canada: A social psychological analysis. Canadian Journal of Behavioural Science, 16(4), 353–370. Berry, J. W. (1997). Immigration, acculturation and adaptation. Applied Psychology: An International Review, 46(1), 5–68. Berry, J. W. (2003). Conceptual approaches to acculturation. In K. Chun, P. Balls-Organista, & G. Marin (Eds.), Acculturation: Advances in theory, measurement and application (pp. 17–37). Washington: APA Books. Berry, J. W. (2006). Mutual attitudes among immigrants and ethnocultural groups in Canada. International Journal of Intercultural Relations, 30(6), 719–734. Berry, J. W., Kalin, R., & Taylor, D. (1977). Multiculturalism and ethnic attitudes in Canada. Ottawa: Supply and Services. Berry, J. W., Kim, U., Power, S., Young, M., & Bujaki, M. (1989). Acculturation attitudes in plural societies. Applied Psychology: An International Review, 38(2), 185– 206. Berry, J. W., Phinney, J. S., Sam, D. L., & Vedder, P. (2006). Immigrant youth in cultural transition: Acculturation, identity and adaptation across national contexts. Mahwah, NJ: Lawrence Erlbaum Associates. Bouchard, G., & Taylor, C. (2008). Fonder l’avenir. Le temps de la conciliation. Rapport de la Commission de consultation sur les pratiques d’accommodement relie´es aux diffe´rences culturelles. Building the future: A time for reconciliation]. Montre´al: Commission de consultation sur les pratiques d’accommodement relie´es aux diffe´rences culturelles. Bourhis, R. Y., Moı¨se, L. C., Perreault, S., & Sene´cal, S. (1997). Towards an interactive acculturation model: A social psychological approach. International Journal of Psychology, 32(6), 369–386. Branscombe, N. R., Schmitt, M., & Harvey, R. (1999). Perceiving pervasive discrimination among African Americans: Implications for group identification and wellbeing. Journal of Personality and Social Psychology, 77(1), 135–149. CNCDH. (2006). Rapport de la Commission Nationale Consultative. Report of the national consulting committee]. Paris: La Documentation franc¸aise. Coopersmith, S. (1981). Self-image inventory. Palo Alto: Consulting Psychologist Press. Dona`, G., & Berry, J. W. (1994). Acculturation attitudes and acculturative stress of Central-American refugees. International Journal of Psychology, 29(1), 57–70. Dornbusch, S., Ritter, P., Chen, Z. Y., & Mont-Reynaud, R. (1989). Ethnic differences in family decision-making among adolescents. Paper presented at the meeting of Society for Research in Child Development. Gil, A. G., Vega, W. A., & Dimas, J. M. (1994). Acculturative stress and personal adjustment among Hispanic adolescent boys. Journal of Community Psychology, 22(1), 43–54. Gordon, M. M. (1964). Assimilation in American life. New York: Oxford University Press. Government of Canada. (1971). Multiculturalism policy statement to the house of commons. Ottawa: Hansard. Greene, M. L., Way, N., & Pahl, K. (2006). Trajectories of perceived adult and peer discrimination among Black, Latino, and Asian American adolescents: Patterns and psychological correlates. Developmental Psychology, 42(2), 218–238. Herz, L., & Gullone, E. (1999). The relationship between self-esteem and parenting style: A cross-cultural comparison of Australian and Vietnamese Australian adolescents. Journal of Cross-Cultural Psychology, 30(6), 742–761.

J.W. Berry, C. Sabatier / International Journal of Intercultural Relations 34 (2010) 191–207

207

Knight, G. P., Bernal, M. E., Garzza, C. A., Cota, M. K., & Ocampo, K. A. (1993). Family socialization and the ethnic identity of Mexican–American children. Journal of Cross-Cultural Psychology, 24(1), 99–114. Laukkanen, E., Halonen, P., Aivio, A., Viinama¨ki, H., & Lehtonen, J. (2000). Construct validity of the offer self-image questionnaire in Finnish 13-year-old adolescents: Differences in the self-images of boys and girls. Nordic Journal of Psychiatry, 54(6), 431–435. Leiderman, P. H., Meldman, M. A., & Ritter, P. L. (1989). Parental and peer influences on adolescent self-esteem in a multiethnic high school population. Paper presented at the meetings of the Society for Research in Child Development. Nguyen, H. H., & von Eye, A. (2002). The Acculturation Scale for Vietnamese Adolescents (ASVA): A bidimensional perspective. International Journal of Behavioral Development, 3(3), 202–213. Noels, K. A., & Berry, J. W. (2006). Acculturation in Canada. In D. L. Sam & J. W. Berry (Eds.), The Cambridge handbook of acculturation psychology (pp. 274–293). Cambridge, UK: Cambridge University Press. Offer, D., Ostrov, E., Howard, K. I., & Atknson, R. (1988). The teenage world. Adolescents’ self-image in ten countries. New York: Plenum Medical Book Co. Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psychological Bulletin, 108(3), 499–514. Phinney, J. S. (2003). Ethnic identity and acculturation. In K. Chun, P. Balls-Organista, & G. Marin (Eds.), Acculturation: Advances in theory, measurement and application (pp. 63–81). Washington: APA Books. Rosenberg, G. M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Rosenthal, D. A., & Cichello, A. M. (1986). The meeting of two cultures: Ethnic identity and psychosocial adjustment of Italian–Australian adolescents. International Journal of Psychology, 21(4/5), 487–501. Ryder, A. G., Alden, L. E., & Paulhus, D. L. (2000). Is acculturation unidimensional or bidimensional? A head-to-head comparison in the prediction of personality, self-identity, and adjustment. Journal of Personality and Social Psychology, 79(1), 49–65. Sabatier, C., & Berry, J. W. (1994). Immigration et acculturation. In R. Y. Bourhis J.-P. Leyens (Eds.), Ste´re´otypes, discrimination et relations intergroupes (pp. 261–291). Stereotypes, discrimination and intergroup relationshipsBruxelles: Mardaga. Sabatier, C., & Berry, J. W. (2008). The role of family acculturation, parental style and perceived discrimination in the adaptation of second generation immigrant youth in France and Canada. European Journal of Developmental Psychology, 5, 159–185. Sabatier, C., & Boutry, V. (2006). Acculturation in the Francophone European countries: Context and concepts. In D. L. Sam & J. W. Berry (Eds.), Cambridge handbook of acculturation psychology (pp. 349–367). Cambridge, UK: Cambridge University Press. Stevens, G. W. J. M. , Pels, T. V. M., & Vollebergh, W. A. M. (2004). Patterns of psychological acculturation in adult and adolescent Moroccan immigrants living in the Netherlands. Journal of Cross-Cultural Psychology, 35(6), 689–704. Tchoryk-Pelletier, P. (1989). L’adaptation des minorite´s ethniques. Une e´tude re´alise´e au CEGEP de Saint-Laurent. Adaptation of ethnic minorities. A study in SaintLaurent college]. Montreal, Canada: CEGEP de Saint-Laurent. (Research report). Tremblay, G., Saucier, J.-F., & Tremblay, R. E. (2004). Identity and disruptiveness in boys: Longitudinal perspectives. Child and Adolescent Social Work Journal, 21(4), 387–406. Van de Vijver, F., Helms-Lornz, M., & Feltzer, M. J. (1999). Acculturation and cognitive performance of migrant children in the Netherlands. International Journal of Psychology, 34(3), 149–162. Vega, W. A., Khoury, E. L., Zimmerman, R. S., Gil, A. G., & Warheit, G. J. (1995). Cultural conflicts and problem behaviors of Latino adolescents in home and school environments. Journal of Community Psychology, 23(2), 167–179. Virta, E., Sam, D. L., & Westin, C. (2004). Adolescents with Turkish background in Norway and Sweden: A comparative study of their psychological adaptation. Scandinavian Journal of Psychology, 45(1), 15–25. Ward, C. (1996). Acculturation. In D. Landis & R. Bhagat (Eds.), Handbook of intercultural training (2nd ed., pp. 124–147). Newbury Park, CA: Sage.