Psychological, Sociocultural And Marital Adaptation Of Turk

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2002, 37 (1), 13–26

Psychological, sociocultural, and marital adaptation of Turkish immigrant couples in Canada Bilge Ataca and John W. Berry Queen’s University, Kingston, Ontario, Canada The study examined the acculturation and adaptation of 200 married Turkish immigrants in Toronto, Canada, using selfreport questionnaires. As an extension of research from sojourners to immigrants, and from individuals to married couples, marital adaptation was introduced and three facets of adaptation were differentiated: psychological, sociocultural, and marital. The Ž ndings support the contention that adaptation is multifaceted. Consistent with stress and coping models, psychological adaptation of married couples was associated with the personality variable of hardiness, social support, acculturation attitudes, and discrimination. On the other hand, in line with social learning perspectives, sociocultural adaptation was mostly related to the variables instrumental in acquiring social skills in the new culture, namely, language proŽ ciency and contact with members of the dominant group. Marital adaptation was mostly associated with marital stressors and marital support. The lack of research on gender differences in the differentiation of adaptation was addressed. This differentiation was clearer in men than in women; there were also different variables associated with the facets of men’s and women’s adaptation. The effects of socioeconomic status and gender have also been examined. The Ž ndings made it evident that Turkish immigrants did not acculturate uniformly. Two groups of Turkish immigrants, working class and professionals were clearly distinguished in their acculturation experiences and adaptation. Gender differences were most apparent in the low socioeconomic status group. Women in general were psychologically more vulnerable than men; the group that faced more risk factors were those women of low socioeconomic status. In terms of acculturation attitudes, Turkish immigrant couples strongly endorsed separation; however, those of high socioeconomic status preferred integration and assimilation to a greater, and separation to a lesser extent than those of low socioeconomic status. L’étude a examiné l’acculturation et l’adaptation de 200 immigrants turcs mariés à Toronto, Canada, utilisant des questionnaires d’auto rapport. En ampliŽ ant la recherche des résidents temporels aux immigrants et des célibataires aux couples mariés on a introduit l’adaptation et trois facettes ont été différenciées: psychologique, socioculturelle et maritale. Les résultats conŽ rment l’afŽ rmation que l’adaptation comporte de multiples facettes. L’adaptation psychologique des couples mariés qui est conforme aux modèles de stress et d’affrontement a été associé à la variable de personnalité audace, à l’appui social, aux attitudes relatives à l’acculturation et à la discrimination. Par ailleurs, en parallèle aux perspectives de l’apprentissage social, l’adaptation socioculturelle a été associée en grande partie aux variables instrumentales servant à acquérir des habilités sociales dans la nouvelle culture, notamment la maîtrise de la langue et le contact avec des membres du groupe dominant. L’adaptation maritale a été associée principalement à des éléments de stress et de support dans le couple. Etant donné le manque de différentiation relative au sexe dans la recherche de l’adaptation, cet aspect a été traité. Cette différentiation s’est avérée plus claire chez les hommes que chez les femmes, il y a eu également des variables différentes associées aux facettes d’adaptation des hommes et des femmes. L’in uence du statut socio-économique et du sexe a été également examinée. Les résultats ont mis en évidence le fait que les immigrants turcs ne subissent pas l’acculturation d’une manière uniforme. Deux groupes d’immigrants turcs, des ouvriers et des employés, ont été clairement distingués au point de vue de leur expérience d’acculturation et leur adaptation. Des différences relatives au sexe ont été plus marquées dans le groupe avec le statut social inférieur. Les femmes en général étaient plus vulnérables que les hommes, le groupe qui affrontait le plus de facteurs de risque était celui des femmes de statut socio-économique inférieur. Concernant les attitudes relatives à l’acculturation, les couples d’immigrants turcs ont opté fortement pour une distanciation, cependant les couples avec un statut socio-économique élevé ont préféré l’intégration et l’assimilation et, dans une moindre mesure, la séparation que ceux avec un statut socio-économique inférieur. El estudio examinó la aculturación y adaptación de 200 turcos casados inmigrados a Toronto, Canadá, por medio del uso de cuestionarios de auto informe. De la extensión de la investigación sobre residentes temporales a los inmigrantes, y de los individuos a las parejas de cónyuges, se introdujo la adaptación conyugal y se diferenciaron tres facetas de adaptación: psicológica, sociocultural y marital. Los hallazgos apoyan la aseveración de que la adaptación es multifacética. Consistente con los modelos de estrés y afrontamiento, la adaptación psicológica de las parejas de casados se asoció con la variable de personalidad audacia, con el apoyo social, las actitudes relativas a la aculturación, y la discriminación. Por otra parte, de acuerdo

Requests for reprints should be addressed to Bilge Ataca, who is now at the Department of Psychology, BogÆ aziçi University, Bebek, Istanbul, Turkey 80815 (E-mail: [email protected]). This article is based in part on a dissertation conducted by the Ž rst author under the supervision of the second. We thank Lee Fabrigar for his important contributions to the article.

Ó 2002 International Union of Psychological Science http://www.tandf.co.uk/journals/pp/00207594.htm l

DOI: 10.1080/0020759014300013 5

14

ATACA AND BERRY

con la perspectiva del aprendizaje social, la adaptación sociocultural se relacionó principalmente con las variables instrumentales para la adquisición de habilidades sociales en la nueva cultura, propiamente, dominio del lenguaje y contacto con los miembros del grupo dominante. La adaptación marital se asoció en mayor medida con los estresores conyugales y el apoyo de la pareja. Considerando la carencia de investigación sobre diferencias de género en cuanto a la adaptación, se trató este aspecto. Esta diferenciación se dio con mayor claridad en los hombres que en las mujeres; hubo también diferentes variables asociadas con las facetas de adaptación de hombres y de mujeres. También se examinaron los efectos del nivel socioeconómico y el sexo. Los hallazgos evidenciaron que los inmigrantes turcos no logran la aculturación de una manera uniforme. Dos grupos de inmigrantes turcos, de clase trabajadora y profesionales, se distinguieron entre sí respecto a sus experiencias de aculturación y su adaptación. Las diferencias entre sexos fueron más ostensibles en el grupo de bajo nivel socioeconómico. Las mujeres, en general, resultaron psicológicamente más vulnerables que los hombres; el grupo que encaró más factores de riesgo fue el de las mujeres de bajo nivel socioeconómico. En términos de las actitudes relativas a la aculturación, las parejas turcas inmigrantes abogaron con fuerza por la separación, sin embargo, aquéllos de nivel socioeconómico alto preŽ rieron la integración y la asimilación y, en menor medida, la separación que aquéllos de bajo nivel socioeconómico.

INTRODUCTION Psychological acculturation and its outcome, adaptation , have been the focus of much research on immigration (Berry, 1997). Psychological acculturation (Graves, 1967) refers to the changes in individuals who are experiencing culture changes that result from continuous, Ž rst-hand contact among groups having different cultures. Researchers, drawing from the theoretical frameworks of psychology and applying them to acculturation, have used multiple terms for adaptation in the immigrant, sojourner, and refugee literature. In the immigrant literature, mental health has been the most widely studied adaptive response, following stress and coping models. Immigrants have long been considered a high-risk group for psychologica l disturbance because immigration constitutes major life changes, and stresses associated with such changes can have mental health consequences (Holmes & Rahe, 1967). Although the early literature emphasized the “inevitable” negative consequences of immigration, research in this area has in fact been inconclusive. Findings have been inconsistent, showing higher, equal, or lower rates of mental disorders in immigrants. Hence, such a negative and broad generalization no longer appears to be valid (Berry & Kim, 1988), with social and psychologica l outcomes now known to be highly variable. In the recent literature on psychological adaptation to acculturation, a distinction has been drawn between psychological (emotional/affective) and sociocultural (behavioural) adaptation (Searle & Ward, 1990; Ward, 1996). Psychological adaptation , associated with a stress and coping framework (e.g., Lazarus & Folkman, 1984), refers to psychological wellbeing and satisfaction in a new cultural context. Sociocultural adaptation , based on a social learning perspective (Furnham & Bochner, 1982), refers to acquiring culturally appropriate knowledge and skills, and thereby, to one’s ability to interact with the new culture and to deal with daily problems of living. Individual s experiencing a culture change are socially unskilled in the new cultural setting; some culture learning (the learning of some behaviours from the new culture) as well as culture shedding (the unlearning of aspects of one’s previous repertoire that are no longer appropriate ) (Berry, 1992) are required.

These two facets of adaptation are interrelated; dealing successfully with problems and positive interactions with members of the dominant culture are likely to improve one’s feelings of wellbeing and satisfaction; similarly, it is easier to accomplish tasks and develop positive interpersonal relations if one is feeling well and accepted. Although related, they are conceptually and empirically distinct. Conceptually, they re ect two distinct theoretical approaches to acculturation; empirically, they have different experiential antecedents. Research has shown that psychological adaptation , deŽ ned in terms of wellbeing or mood states (e.g., depression, anxiety, tension, and fatigue), was predicted by personality, life changes, and social support variable s. Searle and Ward (1990) found that extroversion, life events, and satisfaction with host national relations predicted psychological adaptation in Malaysian and Singaporean students in New Zealand. Locus of control, life changes, and personal relationship satisfaction accounted for a substantial portion of variance in psychological wellbeing in student and adult sojourners (Ward & Kennedy, 1992, 1993a). In contrast, assessed in terms of social difŽ culty, sociocultural adaptation was predicted by variables that are related more strongly to cognitive factors and social skills acquisition, such as cultural knowledge, cultural distance, cultural identity, language ability, length of residence in the new culture, and amount of contact with hosts (Ward & Kennedy, 1992, 1993a, b). A major limitation with this empirical distinction is that it has been based on sojourners (e.g., foreign students, international business people, expatriate wives, diplomats), but it has not been demonstrated for immigrants. Although both groups are migrants in voluntary contact, sojourners and immigrants are different in terms of the permanence of their settlement in the new society. Sojourners are short-term visitors who do not intend to stay; however, immigrants intend to have a relatively permanent residence in the new society (Berry & Sam, 1997). Hence, the relative importance of variables in immigrant adaptation may differ from that of sojourners. The present study examines the distinction between the two facets of adaptation for immigrants; in particular, its focus is on a sample of Turkish immigrants in Canada. Another limitation is that acculturation and adaptation have been studied from the point of view of individuals.

ADAPTATION OF TURKISH IMMIGRANTS

However, for the most part, it is not individual s who migrate, but intact family groups. Canadian immigration policy stresses family reuniŽ cation; hence, married couples constitute a large portion of immigrants. Acculturation constitutes a double transition for married immigrants in that both the individua l and the marriage adapt to the new culture. In this context, marital adaptation relates to the mutual accommodation of spouses when each is faced with the new culture and forms of behaviour and different ways of acculturating. Accommodation of spouses to each other is an important issue at a time when ties with the old support networks are lost (Ward, 1996). Marital problems can make life more difŽ cult in the new culture, or conversely, a happy marriage can lead to a successful adaptation . Hence, it is valuable to study immigrant couples’ marital adaptation , which may have an important in uence on their acculturation and positive adaptation . Hence, marital adaptation must be distinguishe d as another facet of the adaptation of married immigrant couples and its relations with the other facets need to be speciŽ ed. This study aims to examine the relationship between marital adaptation and psychologica l and sociocultural adaptation in a systematic fashion by incorporating males and females, and also groups of high and low socioeconomic status. Research has been inattentive to the effect of two important demographic variables, gender and socioeconomic status (SES), in the acculturation literature. Immigrant men and women have different experiences in the new cultural context. On the whole, immigrant women are less likely to speak either of the ofŽ cial languages of Canada than immigrant men. The literature suggests that women immigrants have more difŽ culty in the new cultural context and may be more vulnerable to mental health problems than are men (Beiser et al., 1988; Boyd, 1986). For example, Turkish immigrant women from traditional rural backgrounds with limited education became isolated in the Western, predominantly Christian, industrial urban societies. They had to live without the traditional support of an extended female network, and in a situation in which limited knowledge of language restricted them from participation in the life of the host society (Hearst, 1985). Their marital relationships also underwent some changes. For women who were unemployed and conŽ ned to home, the relationship with the husband took on a more joint conjugal nature due to isolation. The unavailability of the female networks led to isolation, feelings of insecurity, and restriction by the husband. The husband enforced his authority more because it was more difŽ cult to protect his honour in an unfamiliar setting without the buffer of the old female networks (Abadan-Unat , 1985; Keçeli, 1988). In families where women worked and enjoyed greater autonomy and freedom, adjustments had to be made because of the changing sex roles (Abadan-Unat, 1985). In such a context, it is likely that the different experiences of men and women have a differential effect on the facets of their adaptation . A better understanding of these differences is possible if immigrant women and men in the same context and under similar in uences are studied concurrently.

15

Socioeconomic status is also a variable of central importance, especially for immigrants from industrializin g societies where social class variations are very striking. These variations have differential effects on psychologica l variables; failing to differentiate between social class and culture may lead to erroneous attributions of group differences to the latter. Hence, socioeconomic standing must be incorporated in cross-cultural research in order to disentangle its effects from the effects of culture or ethnicity (KagÆ õ tçõ bas¸ õ & Berry, 1989). Socioeconomic status is incorporated in the present study in order to examine its effects on psychologica l acculturation, and thereby, to disentangle its effects from those of the Turkish culture. The present study also aims to extend the work on individuals’ acculturation attitudes to those of married couples. Acculturation strategy is an important concept in the study of adaptation , which has been addressed from the point of individuals until now. In all plural societies, cultural groups and their individual members must deal with the issue of how to acculturate. Four strategies have been proposed (Berry, 1976) based on an individual ’s orientation to two issues: cultural maintenance (to what extent is it important to maintain one’s cultural heritage and identity); and contact and participatio n (to what extent is it important to seek out and participate with other groups in the larger society). When the Ž rst issue is viewed negatively and the second is viewed positively, the Assimilation strategy is deŽ ned. Conversely, when the opposite responses are preferred, the Separation strategy is present. When both issues are viewed positively, the Integration strategy is deŽ ned (i.e., the simultaneous presence of both cultural maintenance, and contact with and participation in the larger society). Finally, opposite to Integration is the strategy of Marginalization, when there is little possibility or interest in cultural maintenance and limited involvement in the larger society. Married couples’ attitudes must be studied because spouses may have different attitudes on how to acculturate and they in uence each other’s attitudes (Kenny, 1996). This may especially be true for spheres where the couple most often act as a unit and make decisions jointly, such as child-rearing, social activity, holiday celebration, etc.

Overview of study The present paper aims to extend Ward’s distinction between psychological and sociocultural adaptation from a sojourner to an immigrant population , as well as from individuals to married couples. There are three objectives: Ž rst, to introduce marital adaptation as the third facet of the overall adaptation of married immigrant couples and to distinguish among psychological , sociocultural, and marital adaptation ; second, to examine the three facets of adaptation separately for males and females; and third, to examine the similarities and differences in acculturation and adaptation in relation to socioeconomic status and gender. The paper is based on a comparative study of Turkish immigrant adaptation to Canada, which also incorporated two sedentary reference groups, namely,

16

ATACA AND BERRY

Turks in Turkey, and Euro-Canadians. However, for the present purposes only the Ž ndings related to the immigrant group will be reported. The focus of the present paper is on married Turkish immigrant couples living in Toronto. Turkish immigrants constitute a small cultural group in Canada. The general pattern of Turkish migration to Canada has been such that it reached its peak before World War I, stopped until after World War II, and peaked during the late 1960s, and again in the early 1980s; it has been increasing rapidly during the last few years (Kaman, 1994). According to the 1996 Census data, 14,430 Turkish immigrants who were born in Turkey live in Canada (Statistics Canada, 1997). The largest Turkish community in Canada, with a populatio n of 5350 Turkish immigrants (2680 females and 2670 males), lives in Toronto (Statistics Canada, 1997). Although small, the size of the Turkish immigrant population has been increasing rapidly and this presents a need to understand how they live through the process of acculturation and adapt to life in Canada. Turkish immigrants are also particularly suitable for the examination of the effect of socioeconomic status on psychologica l acculturation as there are two distinct groups: the upper-middle class professionals and the lower-middle class workers. Most of the literature on Turkish migration has focused on labour migration to Europe (see Abadan-Unat & Kemiksiz, 1986, for a review); research mostly ignored Turkish professionals in general and workers who settled in different continents.

METHOD Participants Two hundred married Turkish immigrants (100 couples) living in Toronto participated in the study. They constituted a “sample of convenience” (Lonner & Berry, 1986) contacted through various Turkish cultural groups, at social events, and by way of an ad in the newsletter of a Turkish association, at a coffee and a grocery shop operated by Turkish immigrants. In order to have a community representative sample, respondents were selected from different age groups and socioeconomic standing, and those with different lengths of residence in Canada. Immigrants from Turkey are a heterogeneous group that includes various ethnic groups such as Turks, Kurds, Armenians, and non-Anatolian Turkish immigrants such as the Turkish Cypriots and Bulgarian, Bosnian, and Azeri Turks. There are also those Turkish immigrants who remigrated to Canada after migrating to Europe. The participants in this study were ethnic Turks emigrating to Canada directly from Turkey at or above age 14. Age 14 was the cut-off for age of arrival to ensure sufŽ cient immersion in the Turkish culture before arriving in Canada. The participants were middle-aged married couples with a wide range of educational and occupational attainment. The mean age was 42.1 years; 39.8 for women and 44.4 for men, ranging from 19 to 73 years. Half the men (50.0%) and 33% of the women had a bachelor’s degree or above.

Eighty-two per cent of the men and 42% of the women were gainfully employed. Of the 62% employed, 22% were professional, 13% had either managerial positions or owned small businesses, 5.5% were white collar, 11% were skilled, and 10.5% were unskilled workers. The participants had been living in Canada for an average of 13.5 years, ranging from less than a year to 40 years. They were married for a mean of 17.2 years and the mean number of children was 1.8.

Procedure A self-administered questionnaire was employed in the study. All measures were originally in English; all except the anxiety measure were translated into Turkish using the method of forward and backward translation with decentration (Brislin, 1986). For the anxiety measure, a standardized Turkish version (Öner & Le Compte, 1985) was used. Participants were visited individually at home and each spouse completed a questionnaire separately in the presence of the researcher. Where possible, the questionnaire was also administered to groups of couples. A group administration of up to four couples per session was held with those who had no concerns about this procedure. The group administration served the purpose of expediency and preventing contamination . This way those who had the potential to be informed about the study ahead of time were exposed to the study concurrently.

Materials The questionnaire included sections on: demographic/ background information, personality, marital variables, acculturation, and adaptation measures. The selection of the speciŽ c predictor variables was based on the strong relations in the literature between the predictor and the particular facet of adaptation . Demographic/background information The demographic variables were: age, sex, length of residence in Canada (years), education, length of marriage, and number of children. Personality Hardiness. The hardiness of one’s personality was measured by a 30-item short form Hardiness Scale (HS; Bartone, Ursano, Wright, & Ingraham, 1989). The items were rated on a 4-point scale ranging from “not at all true” (1) to “completely true” (4), with higher scores indicating more hardiness. Marital variables Marital stressors. Marital stressors were measured by a modiŽ ed version of a marital stressors index (Ilfeld, 1976). It was composed of items re ecting circumstances or con-

ADAPTATION OF TURKISH IMMIGRANTS

ditions of daily marital life that are generally considered problematic or undesirable, including frustration of role expectations, lack of reciprocity between spouses, and a feeling of nonacceptance by one’s spouse. Items concerning children were incorporated for the present study in order to address the central aspect of Turkish marital life. The resulting measure was a 19-item 5-point Likert scale ranging from “strong disagreement” (1) to “strong agreement” (5), with higher scores indicating more marital stressors. Marital support. Support received from the spouse was measured by a 5-item scale developed for the study based on a review of the literature. Respondents rated the quality of their marital relationship and the actual use of marital support on a 5-point Likert scale ranging from “strong disagreement” (1) to “strong agreement” (5), with higher scores indicating more marital support. Items addressed both expressive support, such as “My spouse is very good at understanding my problems,” and instrumental support, including “I can turn to my spouse when I am in need (Ž nancial, informational , or practical).” Acculturation variables Language proŽ ciency. English proŽ ciency was assessed by an 8-item measure in which the respondents were asked to rate their ability to comprehend, speak, read, and write English, sufŽ ciency of their English under certain situations, and the daily amount of English spoken. The responses were scored on a 5-point scale with higher scores indicating more proŽ ciency in English. Social support. Social support was measured by an 11item perceived social support scale developed for the study based on a review of the literature. It included two dimensions of support: perception of availability (e.g., the social network size and social integration as indicated through participation in organizations) and quality of social relations (e.g., presence of solidarity and trust); and two supporting functions of network ties: emotional nurturance (e.g., expressive support) and resource and information assistance (e.g., instrumental support). Respondents indicated whether they had access to someone excluding their spouses (to eliminate the confounding effect of a good or bad marital relationship) on whom they could rely for these functions and rated the perceived closeness of these relations. Cultural distance. The differences perceived between the Turkish and Canadian cultures in terms of various areas (climate, food, educational level, family life, etc.) were ascertained by a 27-item measure developed for the study. The differences were rated on a 5-point scale ranging from “very similar” (1) to “very different” (5), with higher scores indicating greater perceived cultural distance. Contact. Contact with Euro-Canadian s was measured by a 5-item 5-point measure adapted from Kostovcik (1983). Respondents rated the frequency of contact with Euro-Canadians, closeness to a Euro-Canadian friend, and

17

stated how many Euro-Canadian friends they had. Higher scores indicated more contact with Euro-Canadian s. Acculturation attitudes. Eleven attitude domains that are important in the lives of the Turkish immigrants were speciŽ ed: friendship, lifestyle, social activity, food, holiday celebration, language use, decoration, newspaper readership, child-rearing style, children’s values, children’s moving out. Based on Berry’s (1976) acculturation model, four items for each domain were generated, one for each acculturation attitude of integration (e.g., “I prefer to socialize with both Canadians and Turks”), separation (e.g., “I prefer to socialize with Turks more than with Canadians ”), assimilation (e.g., “I prefer to socialize with Canadians more than with Turks”), and marginalization (e.g., “I don’t really care who I socialize with”). Hence, each acculturation attitude scale was comprised of 11 items. Some changes to solve previous measurement problems were undertaken; for example, each statement conveyed one piece of information (i.e., it was not double barrelled) and Ž rst person singular wording rather than third person singular or plural was used. The 44 items were randomly ordered in the scale and the responses were given on a 5point Likert scale ranging from “strong disagreement” (1) to “strong agreement” (5). Higher scores for each acculturation attitude measure indicated higher preference for the particular attitude. Discrimination. Discrimination perceived from EuroCanadians was measured by an adapted version of an 18item Perceived Discrimination Scale (Hocoy, 1993) for use with Chinese-Canadians. Items were scored on a 7-point Likert scale with higher scores indicating higher levels of discrimination perceived by respondents from Euro-Canadian s. The reliability and validity have been well established (Hocoy, 1993) with Cronbach’s alpha ranging from .85 (Dhawan, 1997; Hocoy, 1993) to .88 (Restoule, 1994). The scale has been used in other acculturation studies with Black South Africans (Hocoy, 1997) and various cultural groups in Canada, including Chinese (Hocoy, 1993), Aboriginal (Restoule, 1994), and EastIndian (Dhawan, 1997). Psychological adaptation Four scales measuring depression, stress, anxiety, and satisfaction with life were used to create a composite psychological adaptation score. Depression. The 25-item Generalized Contentment Scale (GCS; Hudson, 1982) was used to measure the degree, severity, or magnitude of nonpsychotic depression, focusing on the affective aspects of clinical depression. The items were rated on a 5-point scale ranging from “rarely or none of the time” (1) to “most or all of the time” (5), with higher scores indicating more depression. Stress. Stress was measured by a 20-item Cawte Stress Scale (Cawte, 1972), which consists of selected items from

18

ATACA AND BERRY

the Cornell Medical Index (CMI; Brodman, Erdman, Lorge, Gershenson, & Wolff, 1952). The Ž rst 10 items assess psychosomatic symptomatology; the second 10 items cover the areas of anxiety, depression, and irritability. Responses are given in a dichotomous “yes” (1) or “no” (0) fashion, with higher scores indicating more stress. The scale has been used in over 30 studies conducted with refugees, immigrants, ethnic groups, Native peoples, and sojourners in Canada (e.g., Berry, Kim, Minde, & Mok, 1987; Dona & Berry, 1994; Kostovcik, 1983). Anxiety. Anxiety was measured by a modiŽ ed version of the 20-item state scale of the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, & Lushene, 1970; standardized Turkish version by Öner & Le Compte, 1985). State anxiety, as a transitory emotional state or condition, addresses how one feels at the moment that he or she is completing the scale. In order to represent the degree of anxiety in the daily lives of immigrants, a broader time frame was addressed by asking respondents to describe their feelings “during the past few weeks.” Responses were rated on a 4-point scale ranging from “not at all” (1) to “very much so” (4), with higher scores indicating more anxiety. The STAI is a widely used anxiety measure that has also been adapted for use in cross-cultural research (Spielberger & Diaz-Guerrero, 1982). Satisfaction with life. Satisfaction with life was measured by a 5-item Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen, & GrifŽ n, 1985). The scale assesses an individual ’s global judgment of his or her quality of life. The SWLS is a widely used life satisfaction scale and has been translated into several languages for crosscultural use (Pavot & Diener, 1993). Responses are given on a 7-point Likert scale ranging from “strong disagreement” (1) to “strong agreement” (7), with higher scores indicating more satisfaction with life. Sociocultural adaptation Sociocultural adaptation was measured by a modiŽ ed version of the Social Situations Questionnaire (SSQ; Furnham & Bochner, 1982) for international students. Respondents were asked to rate the difŽ culty—“never experienced” (0), “very difŽ cult” (1), to “very easy” (7)— they experienced in 23 social situations. Higher scores indicated less difŽ culty with social situations, i.e., better sociocultural adaptation . Marital adaptation Marital adaptation was measured by a modiŽ ed version of the Locke-Wallace Marital Adjustment Test (LWMAT; Locke & Wallace, 1959). Eight items concerning important topics in the Turkish marital life were incorporated. A total of 23 items assessed accommodation of spouses to each other at any given time, with higher scores indicating higher levels of marital adaptation.

RESULTS The unit of analysis was chosen to be the “couple” in order not to violate the assumption of independence in the statistical analyses (Kenny, 1998)1. Descriptive statistics for all scales are shown in Table 1. Socioeconomic status was determined by the mean of each spouse’s education ranging from “no schooling” (1) to “graduate” (8). TABLE 1 Descriptive statistics for variables Scale Mean SD

Variable Socioeconomic status

Possible range Cronbach’s of scores alpha

5.17

2.05

1–8

–

Hardiness

85.03

7.21

30–120

.74

Marital stressors

39.72

9.13

19–95

.85

Marital support

20.72

3.32

5–25

.90

Length of residence

13.50

8.50

1–40

–

Language proŽ ciency

30.80

6.22

8–40

.95

Social support

34.11

8.56

11–55

.94

105.06

9.22

27–135

.87

Contact

11.55

3.76

5–25

.86

Acculturation attitudes Integration Separation Assimilation Marginalization

34.96 39.86 22.70 21.75

6.72 7.53 5.27 5.00

11–55 11–55 11–55 11–55

.84 .89 .83 .78

18–126

.80

Cultural distance

Discrimination Psychological adaptation Depression (R)b Stress (R) Anxiety (R) Satisfaction

51.24 11.48 a

0.00 31.56 3.92 37.66 23.71

2.42 9.73 2.30 7.44 4.74

Sociocultural adaptation 105.27 20.53 Marital adaptation

101.68 12.87

2 23.66–9.69 0–100 0–20 20–80 5–35

.82 .89 .76 .93 .86

0–161

.93

23–123

.95

Psychological adaptation 5 Depression (R) 1 Stress (R) 1 Anxiety (R) + Satisfaction (standardized and summated score). bReversed scale scores. a

1 Various processes lead to nonindependence in couples research (Kenny, 1996). In marital research, if the individual is the unit of analysis and the couple is ignored, the independence assumption is likely to be violated. This will lead to bias in the tests of signiŽ cance and distortion of Type I and Type II errors. However, if the couple is the unit of analysis by using a matched-pairs, repeated-measures analysis, or by creating couple-level scores, then nonindependence does not bias the signiŽ cance testing (Kenny, 1988). The degree of nonindependence between husbands’ and wives’ scores can be estimated by using a Pearson correlation coefŽ cient. However, Kenny and Kashy (1991) recommend that data involving even low levels of interdependence be treated as nonindependent data, and hence, a very liberal test (p 5 .25) be used. These guidelines were followed in the present study. The correlation analyses for the outcome variables revealed nonindependence between spouses: r (96) 5 .14, p 5 .19 for psychological adaptation; r (96) 5 .36, p , .001 for sociocultural adaptation; r (96) 5 .66, p , .001 for marital adaptation. Hence, it was decided to create couple-level scores and treat the couple as the unit of analysis.

ADAPTATION OF TURKISH IMMIGRANTS TABLE 2 Intercorrelations among psychological adaptation measures (N 5 Variables 1. 2. 3. 4.

1

Depression Stress Anxiety Satisfaction ***p ø

– .68*** .76*** 2 .60***

2 – .52*** 2 .35**

3

– 2 .68***

93) 4

–

.001.

In order to form a composite psychological adaptation score, Ž rst, depression, stress, and anxiety scores were re ected to represent less psychological disturbance; second, raw scores on these three measures and the satisfaction measure were standardized and summed. The signiŽ cant correlations among these measures enabled the aggregation of the scores (see Table 2).

Psychological, sociocultural, and marital adaptation of Turkish immigrants Multiple regression and canonical correlation analyses were employed to distinguish among the psychological , sociocultural, and marital adaptation of married Turkish immigrants. Three simultaneous multiple regressions were undertaken separately for couples, males, and females, one for each of the outcome variables of psychological, sociocultural, and marital adaptation (see Table 3). All predictor variable s were included in each of the regression analyses2. At the couple level, the results showed that different variables were signiŽ cant predictors of each dimension of adaptation as expected, which supports the distinctiveness of these constructs. For psychological adaptation , hardiness, length of residence, social support, and marginalization made unique, signiŽ cant contributions to prediction. Marital support, language proŽ ciency, cultural distance, and discrimination had a marginally signiŽ cant tendency to be associated with psychological adaptation . For sociocultural adaptation, contact and language proŽ ciency were unique, signiŽ cant predictors. For marital adaptation, only the marital variables of marital stressors and marital support made signiŽ cant contributions3. 2 This procedure of including all as opposed to only the theoretically relevant variables was employed as a more stringent test of the discriminant validity of psychological, sociocultural, and marital adaptation. It followed the approach of Ward and colleagues (Searle & Ward, 1990; Ward & Kennedy, 1993a, b) in past research. However, contrary to their approach, simultaneous as opposed to stepwise multiple regressions were employed. This was because the use of stepwise regression has been strongly criticized by methodologists (e.g., Cohen & Cohen, 1983; Judd & McClelland, 1989). In stepwise regression, the selection among the independent variables is not guided by theory, but is based on a search through different models based on statistics computed from the particular sample drawn, which increases the likelihood of capitalizing on chance. 3 For exploratory reasons, a separate multiple regression was conducted for marital adaptation that also included length of marriage and number of children as predictors. These two variables did not emerge as signiŽ cant predictors.

19

At the individual level, different predictors were found to make unique, signiŽ cant contributions to the prediction of the psychological adaptation of males and females. Hardiness, marital support, length of residence, social support, cultural distance, and contact were signiŽ cant predictors of the psychological adaptation of males, whereas marital stressors was the only signiŽ cant predictor of the psychological adaptation of females. For sociocultural adaptation , contact made a unique contribution to the prediction of both males and females. Language proŽ ciency, social support, marginalization , and discrimination also emerged as signiŽ cant in the prediction of the sociocultural adaptation of males, and hardiness was another signiŽ cant predictor in the sociocultural adaptation of females. For marital adaptation , marital stressors and support were signiŽ cant predictors of both males and females. For the adaptation of males, socioeconomic status and integration also proved signiŽ cant. Because the three adaptationa l outcome variables correlated among themselves, canonical correlation analyses were employed in order to create independent linear combinations of predictor and outcome variables. For couples, three canonical correlations were found to be signiŽ cant. The Ž rst canonical correlation was .88; the second was .81; and the third was .65. With all three correlations included, F(42.0, 190.6) 5 8.26, p , .001, with the Ž rst canonical correlation removed, F(26.0, 130.0) 5 6.28, p , .001, and with the second removed, F(12.0, 66.0) 5 4.07, p , .001. To enhance the interpretability of the signiŽ cant canonical solutions, loadings were rotated to a varimax criterion (Cliff, 1987). The rotated loadings of the three pairs of canonical variates and the variance accounted for by each pair are presented in Table 4. The Ž rst variate represented a combination of Marital and Psychological Adaptation and was related to marital support, fewer marital stressors, less preference for assimilation and marginalization , lower socioeconomic status, and separation. The second variate indicated an association between Sociocultural Adaptation and contact with Euro-Canadians, language proŽ ciency, integration, length of residence, and socioeconomic status. The third variate indicated an association between Psychological Adaptation and hardiness, less discrimination, less preference for separation, social support, and assimilation. Canonical correlation analysis for males revealed three signiŽ cant correlations of .78, .77, and .64. With all three correlations included, F(42.0, 235.1) 5 6.90, p , .001, with the Ž rst canonical correlation removed, F(26.0, 160.0) 5 6.47, p , .001, and with the second removed, F(12.0, 81.0) 5 4.65, p , .001. For females, only the Ž rst two of the three canonical correlations (.85, .74, and .40) were significant. With all three correlations included, F(42.0, 202.5) 5 5.55, p , .001, with the Ž rst canonical correlation removed, F(26.0, 138.0 ) 5 3.32, p , .001, and with the second removed, F(12.0, 70.0) 5 1.12, p . .10. The rotated loadings of the three pairs of canonical variates for males were similar to those of the couples (see Table 4). The Ž rst variate represented a combination of

20

ATACA AND BERRY

TABLE 3 Standardized regression coefŽcients and variance accounted for by variables predicting the adaptational outcomes of immigrant couples and of males and females Psychological Couple (N = 83)

Variable Socioeconomic status Hardiness Marital stressors Marital support Length of residence Language proŽ ciency Social support Cultural distance Contact Integration Assimilation Separation Marginalization Discrimination R2 a

pø

.06; *p ø

2 2 2 2 2 2 2

.05; **p ø

Sociocultural

Males (N = 97) 2

.14 .27* .23 .26a .28** .25a .25* .17a .13 .06 .15 .23 .23* .20a .64*** 2 2 2 2 2

.01; ***p ø

2

Females (N = 86) 2

.09 .46*** .13 .25* .25* .04 .20* .15a .20* .01 .02 .27 .05 .04 .55*** 2

2 2 2 2

Marital

Couple Males Females (N = 82) (N = 97) (N = 85)

.05 .15 .29* .13 .14 .06 .10 .04 .18 .12 .00 .16 .09 .15 .41*** 2

.12 .09 .06 .14 .04 .28* .00 .05 .41** .05 .06 .10 .10 .15 .65*** 2 2 2

2

.09 .00 .07 .05 .04 .42*** .25** .03 .49*** .02 .00 .19 .19a 2 .27** 2 .57***

Couple (N = 82) 2

.20 .23* .02 .08 .16 .08 .01 .03 .39** .01 .20 .14 .12 .01 .62*** 2 2 2 2 2 2 2 2 2 2 2

Males (N = 96) 2

.05 .17 .33* .43** .01 .05 .08 .02 .18 .09 .03 .02 .11 .04 .69*** 2 2

2 2 2 2 2 2

Females (N = 86) 2

.25* .09 .29* .46*** .09 .03 .06 .11 .00 .21* .02 .23 .02 .05 .59***

.08 .17 .36*** .44*** .12 .09 .05 .10 .10 .04 .04 .02 .06 .14 .60*** 2 2 2

2 2 2 2

.001.

TABLE 4 Canonical correlation analysis for Turkish immigrant couples and for Turkish males and females Couples Varimax rotated canonical loadings Outcome variables Psychological adaptation Sociocultural adaptation Marital adaptation

V1

Predictor variables Socioeconomic status Hardiness Marital stressors Marital support Length of residence Language proŽ ciency Social support Cultural distance Contact Integration Separation Assimilation Marginalization Discrimination Percentage of variance

2

2

2

2

2

2

2

2

2

2

2

.48 .08 .99

V2

.43 .16 .89 .94 .04 .28 .16 .00 .01 .33 .41 .44 .42 .08 22.68

Males

2

2

2

2

2

.09 .96 .06

V3

.49 .30 .09 .06 .49 .73 .29 .02 .84 .51 .27 .29 .18 .31 19.97

V1

2

2

2

2

2

2

2

.88 .27 .04 .31 .70 .21 .02 .18 .07 .44 .09 .33 .32 .53 .36 .23 .61 16.18

2

.41 .02 .98

V2

2 2 2 2 2 2 2 2 2 2

.40 .07 .87 .88 .04 .08 .11 .15 .08 .26 .28 .32 .33 .07 19.15

Females

2

2

2

2

.18 .99 .12

V3

.49 .32 .06 .03 .35 .73 .07 .06 .64 .32 .32 .14 .10 .54 17.54

2

2

2

2

2

.89 .08 .14

V1

.29 .77 .20 .06 .09 .28 .43 .22 .09 .32 .45 .26 .08 .43 14.34

2

2

2

2

2

.51 .93 .15

V2

.64 .74 .04 .25 .54 .65 .36 .24 .83 .59 .50 .53 .03 .38 28.09

2

2

2

2

2

2

2

2

.60 .14 .99 .32 .04 .80 .86 .02 .18 .23 .09 .19 .20 .25 .23 .20 .26 18.93

V 5 Variate. The loadings of the second variate for couples and of the Ž rst variate for males and females have been re ected for ease of interpretation.

Marital and Psychological Adaptation and was related to marital support, fewer marital stressors, and lower socioeconomic status. The second variate represented Sociocultural Adaptation and was related to language proŽ ciency, contact with Euro-Canadians, less discrimination, socioeconomic status, and length of residence, whereas the third variate indicated an association between Psychological Adaptation and hardiness, less preference for separation, social support, and less discrimination .

For females, the rotated loadings of the Ž rst of two pairs of variates represented a combination of Sociocultural and Psychological Adaptation, and was related to contact, hardiness, language proŽ ciency, socioeconomic status, integration, length of residence, assimilation, less preference for separation, less discrimination , and social support. The second variate represented a combination of Marital and Psychological Adaptation, and was related to marital support and fewer marital stressors.

ADAPTATION OF TURKISH IMMIGRANTS

Socioeconomic status and gender similarities and differences in acculturation In order to examine socioeconomic status and gender similarities and differences in the acculturation context, couples were classiŽ ed into high and low socioeconomic status by taking the median split on the socioeconomic status scores. On the scale ranging from 1.5 to 8.0, the median was 5.5. Those couples whose scores were 5.0 and below were classiŽ ed as “low”, and 6.0 and above were classiŽ ed as “high” SES. A 2 (gender) 3 2 (SES) ANOVA was performed on the variables of interest. There were SES and gender differences in English proŽ ciency. The high SES group (M 5 35.14) was more proŽ cient in English than that of low SES (M 5 26.14), F(1, 91) 5 92.76, p , .001; males (M 5 32.87) were more proŽ cient than females (M 5 28.41), F(1, 91) 5 36.28, p , .001. There was also signiŽ cant SES 3 Gender interaction, F(1, 91) 5 6.39, p , .05, indicating that the difference in English proŽ ciency between males (M 5 36.43) and females (M 5 33.84) was less in the high SES group than it was in the low SES group (M = 29.30, M = 22.97 for males and females, respectively). Post hoc analyses using orthogonal contrasts showed that males were more proŽ cient than females in both the high SES group, F(1, 91) 5 6.12, p < .05, and the low SES group, F(1, 91) 5 36.58, p < .001. With respect to contact with Euro-Canadians, the main effects of SES and gender, as well as their interaction, were signiŽ cant. The high SES group (M 5 12.71) was in more contact than the low SES group (M 5 10.41), F(1, 90) 5

Figure 1.

21

9.20, p , .01; males (M 5 12.47) were in more contact with Euro-Canadians than females (M 5 10.65), F(1, 90) 5 13.77, p , .001. There was also a signiŽ cant interaction, F(1, 90) 5 5.52, p , .05. Planned contrasts indicated that in the high SES group males (M 5 13.04) had as much contact with Euro-Canadians as females (M 5 12.38), F(1, 90) 5 0.91, n.s., but in the low SES group males (M 5 11.89) had more contact than females (M 5 8.92), F(1, 90) = 18.39, p , .001. In terms of acculturation attitudes, integration and assimilation had similar patterns of effect (see Figure 1). Males adopted the integration attitude to a greater extent than females, F(1, 92) 5 6.54, p < .05. The low SES group preferred integration less than the high SES group, F(1, 92) 5 51.96, p , .001. The SES 3 Gender interaction was signiŽ cant, F(1, 92) 5 6.47, p , .05. Post hoc analyses using planned contrasts showed that males preferred integration more than females in the low SES group, F(1, 92) 5 13.04, p , .001, but that they did not differ from females in the high SES group, F(1, 92) 5 0.00, n.s. Males also adopted the assimilation attitude to a greater extent than females, F(1, 92) 5 11.60, p 5 .001. Those of low SES also preferred assimilation less than those of high SES, F(1, 92) 5 27.88, p , .001. The SES 3 Gender interaction was marginally signiŽ cant, F(1, 92) 5 3.67, p 5 .058. Males preferred assimilation more than females in the low SES group, F(1, 92) 5 14.11, p , .001, but they did not differ from females in the high SES group, F(1, 92) 5 1.10, n.s. Separation attitude was endorsed more by those of low SES than by those of high SES, F(1, 90) 5 38.26, p , .001; males and females were not different in their attitudes, F(1, 90) 5 0.84, n.s. Males

Acculturation attitudes of Turkish immigrant men and women of high and low socioeconomic status.

22

ATACA AND BERRY

adopted more marginalizatio n attitude than females, F(1, 90) 5 8.23, p , .01, yet the difference was signiŽ cant in the low SES group, F(1, 90) 5 11.72, p , .001, but not in the high SES group, F(1, 90) 5 0.41, n.s. SES and gender effects were not detected on the composite psychological adaptation . However, when the components of psychological adaptation were examined individuall y, some effects were apparent (see Figure 2). Those of low SES were more satisŽ ed with their lives than those of high SES, F(1, 92) 5 4.57, p , .05. Females were marginally more depressed than males, F(1, 88) 5 3.79, p 5 .055. Both the gender and SES main effects on stress were signiŽ cant Females were more stressed than males, F (1, 92) 5 29.70, p , .001, and those of low SES were more stressed than those of high SES, F(1, 92) 5 10.50, p , .01. In terms of anxiety, females were more anxious than males, F(1, 90) 5 4.35, p , .05. With respect to sociocultural adaptation , the main effects of gender and SES, as well as their interaction, were signiŽ cant (see Figure 2). The high SES group was socioculturally better adapted than the low SES group, F(1, 90) 5 24.96, p , .001, and males were better adapted than females, F(1, 90) 5 23.92, p , .001. The signiŽ cant interaction, F(1, 90) 5 5.73, p , .05, indicated that the difference in sociocultural adaptation between males and females was more in the low SES group than in the high SES group. The following post hoc analyses showed that males were better adapted than females in the low SES group, F(1, 90) 5 26.52, p , .001, yet the difference between males and females of high SES did not reach acceptable levels of signiŽ cance, F(1, 90) 5 3.11, p , .09. In terms of marital adaptation, the signiŽ cant main effect of SES indicated that those of low SES had better marital adaptation than those of high SES, F(1, 91) 5 13.44, p , .001.

DISCUSSION The present study supports the contention that adaptation is multifaceted. Although psychological , sociocultural, and marital adaptation of immigrant couples are related to some extent, they are conceptually distinct, and are empirically related to different sources for the most part. The study advances our understanding of the distinction among different facets of adaptation in several ways. First, it extends research from a sojourner to an immigrant population, as well as from individuals to married couples. At the couple level, the Ž ndings on the distinction of psychological and sociocultural adaptation are in broad agreement with past research and further extend the differentiation of facets of adaptation to married immigrant couples. The two facets of adaptation are associated with different variables in accordance with the underlying models. In line with stress and coping models, psychologi cal adaptation of married couples is associated with the personality variable of hardiness, social support, acculturation attitudes, and discrimination . Consistent with social learning approaches to acculturation, sociocultural adaptation is mostly related to the variables instrumental in acquiring social skills in the new culture, namely, language proŽ ciency and contact with members of the dominant group. The utility of this approach lies in the theoretical integration of the acculturation Ž eld as well as its potential for applications. The adaptation of acculturating individual s can be improved by attending to the different sources of the multiple dimensions of adaptation . This is especially important for arriving at theoretically informed and empirically guided applications such as training programmes, including English as a Second Language (ESL) courses, culturally sensitive counselling, counsellors proŽ -

1. Components of Psychological Adaptation

Figure 2.

Psychological, sociocultural, and marital adaptation of Turkish immigrant men and women of high and low socioeconomic status.

ADAPTATION OF TURKISH IMMIGRANTS

cient in various languages, and promotion of an appreciation of pluralism in the public. The second contribution of the present research is the study of marital issues in the lives of immigrant couples. Marital adaptation is introduced as the third facet of the overall adaptation of married immigrants and marital variables are examined in relation to the different facets of adaptation . This is useful in highlighting the role of the spouses in successful adaptation to the new culture. The present Ž ndings support the distinctiveness of marital adaptation from psychological and sociocultural adaptation. Marital adaptation is mostly associated with the variables speciŽ c to marital life, i.e., marital stressors and marital support. Marital variables also display close relations with psychologica l adaptation . This is consistent with studies which examined the relationship between marital variable s and psychological distress among immigrants (Dyal, Rybensky, & Somers, 1988; El Haïli & Lasry, 1997). The third way the present study advances our understanding is by drawing attention to gender differences in the distinction of psychological , sociocultural, and marital adaptation . The distinction is more clear for males than for females, and there are different predictors of the adaptation of males and females. These differences point to the need to consider separate sources for the adaptation of men and women. Social policy and programme developers should recognize that immigrant men and women, especially those with lower socioeconomic status, have different experiences in the accculturation context. The failure to differentiate between the psychologica l and sociocultural adaptation of females points to the possibility that social skills and psychological wellbeing are more closely associated in immigrant women than in immigrant men. The Ž ndings show that most women (76.6%) compared to men (27.7%) in the low SES group are not gainfully employed; these Ž gures correspond to 36.2% for women and 10.6% for men in the high SES group. One of the many functions of work is to enable individuals to interact with others and establish relationships in the larger society (Jahoda, 1981). Not working outside the home, most women of low SES do not get to interact with the larger society and become secluded at home and within their own ethnic circles. This most certainly re ects on their acculturation attitudes. Any idea of a relationship with the larger society is remote; hence, they have the lowest preference for integrationist and assimilationist attitudes of all immigrants. The acquisition of culturally appropriate knowledge and skills, including English language skills, is thereby hindered, leading to poor sociocultural adaptation . Interacting with the new culture and being a part of the society may be more valued and may take more on the part of these women, which may bring greater feelings of satisfaction, competence, and adequacy. Conversely, isolation and not being effective in the new culture may lead to more psychological disturbance. This also explains the association between hardiness and the psychological and sociocultural adaptation in women. Hardiness may be an instrumental trait in immigrant women, with those who are more in control and

23

committed taking the challenge to overcome hardships, to acquire the skills, and to be socially adequate. On the other hand, for men, who are more in contact with the larger society on a daily basis, discrimination is a crucial factor in the sociocultural adaptation . Discrimination from members of the dominant group may push one away from interacting with the society whose members are perceived as not willing to accommodate one’s differences and lessen the efforts to learn social skills, which may lead to more sociocultural difŽ culty. The two economic classes of Turkish immigrants in Canada form two clearly distinct social groups. The two groups do not interact with each other for the most part; they belong to different associations and participate in separate events. The low SES group resembles the Turkish workers in Europe in many ways. They are mostly from rural or low-income urban backgrounds with relatively low levels of education. This group mostly lives in a Turkish neighbourhood in Toronto. In sharp contrast, the high SES group is mostly made up of professionals who live scattered around Toronto. More educated and more proŽ cient in English, this group is in more contact with Euro-Canadian s. Compared to the high SES group, twice as many in the low SES group are not gainfully employed. Similar to their counterparts in Europe, those of low SES practice Islam more and constitute a discernible group with their more traditional and religious apparel. Holding on to their culture and identity, and rejecting relations with the larger society, those of low SES strongly endorse separationist attitudes. Their preference for assimilation and integration attitudes is much less than that of the high SES group. This is very similar to the situation of the Turkish immigrants in Europe. They, too, strongly identify with the Turkish culture and isolate themselves from the larger society (Abadan-Unat, 1985; KagÆ õ tçõ bas¸ õ , 1987; Piontkowski , Florack, Hoelker, & Obdrzalek, 2000). This reluctance to interact with the larger society may be explained by the lower education levels, lack of language skills, and also the experience of discrimination from the dominant group (Berry, Kim, Power, Young, & Bujaki, 1989). Compared to such groups as Chinese-Canadian s (Hocoy, 1993) and East Indian-Canadians (Dhawan, 1997), Turkish immigrants overall experience less discrimination. However, when socioeconomic standing is taken into consideration, those of lower standing perceive as much discrimination from Euro-Canadians as these visible minority groups. They stand out as a group and tend to perceive more discrimination than those Turkish immigrants of higher socioeconomic standing (Ataca, 1998). However, it is difŽ cult to tell the direction of the relationship. It may also be that the more a group separates and is unwilling to participate in the larger society, the more the members of the larger society perceive it as different and are intolerant of its ways (Taylor & Moghaddam , 1994). The low language proŽ ciency and the low levels of contact with the larger society in this lower-educated group indicate less culturally appropriate knowledge and fewer skills to interact with the culture and manage daily life, hence, greater sociocultural difŽ culty. This group has more

24

ATACA AND BERRY

sociocultural adaptation problems than the highereducated group. Research on sojourner adaptation has demonstrated that sociocultural adaptation problems are greater for those who make large, compared to small, transitions (Furnham & Bochner, 1982; Ward & Kennedy, 1993b, 1998). The acquisition of new culture-speciŽ c skills depends on the similarity between the original culture and the new culture: the more dissimilar the two cultures, i.e., the greater the cultural distance, the fewer are the culturally appropriate skills and the more difŽ cult it is to learn new ones. Social class differences in Turkish society are very sharp. The cultural distance between Turks of higher socioeconomic status with urban backgrounds and higher levels of education and those of lower socioeconomic status with rural or low-income urban backgrounds and lower levels of education is as great as the cultural distance between Western European/North American and Turkish society in general (Koçtürk, 1992). This objective distance between the immigrants of lower SES and higher SES is highly apparent in the present study. The higher SES group has more resources such as a higher level of education and greater language proŽ ciency that help manage life in a Western setting more easily. In this respect, this group makes a relatively smaller cross-cultural transition compared to the lower SES group. With more appropriate skills, they have less sociocultural difŽ culty than those of lower SES. Although more stressed and socioculturally less well adapted, this group of lower socioeconomic standing are more satisŽ ed with their life in Canada. This is similar to the feelings of Turkish immigrants in Europe. Despite being segregated and discriminated against, Turkish workers feel satisŽ ed with their immigration experience because they feel they have attained personal goals (Koçtürk, 1992). The immigrants of lower status in the present study point out that their situation has improved in Canada. Although they earn low wages, they make more than they used to and their material standards have improved in terms of housing, schooling of their children, health beneŽ ts, etc. In contrast, immigrants of higher status feel less satisŽ ed with their life in Canada. These relationships can be understood in the context of relative deprivation theory. The objective level of economic standing is disregarded; the lower SES group make comparisons with what their economic condition used to be like in Turkey and feel gratitude, whereas the high SES group make comparisons with their cohorts in Turkey and feel deprived. This points to the loss of status professionals experience in Canada. Credentials and accreditations obtained in Turkey are often not recognized by Canadian authorities; hence, immigrants with credentials suffer loss of status. Those who experience greater loss are also less satisŽ ed with their life in Canada (Aycan & Berry, 1996). The marital relationship can be situated in this broader context of the socioeconomic status of the couples. Women of high SES are on more equal terms with their husbands. They are gainfully employed; hence, they can be more independent and enjoy greater autonomy and freedom than their counterparts in the low SES group. This

may cause con icts between spouses in the high SES group since the husband’s traditional authority is weakened and the more egalitarian roles of the new culture are adopted. However, the traditional roles are the norm, and therefore, are more prevalent in the low SES group. Most women are not employed outside the home; they have the major responsibility for house-related activities and for raising children, whereas men are the breadwinners and dominate over major decision-making. Women are dependent on and submissive to their husbands ; they do not challenge their authority, which prevents tension in the marital relationship. The Ž ndings of the present study has also conŽ rmed the importance of marital problems for women’s mental health. The woman is more dependent and subordinate to her husband in the Turkish culture, hence, has a higher valuation of the relationship (KagÆ õ tçõ bas¸ õ , 1986). This would make her more prone to psychological disturbance resulting from marital discord. It is also the case that primary childcare falls more heavily on the mother in the context of immigration, with fewer skills and less female adult support than in the country of origin. Hence, women experience the stress associated with raising children more than men. Many women respondents in the study complained about having nobody to look after their children. The children also had to grow up alone without getting to know their grandparents, uncles, and aunts. In this socially deprived context, women have to serve the role of the relatives as well as the mother and pass on the traditions and values of the heritage culture. Another possible explanation for the greater psychological difŽ culty of women may be the high rate (58%) of unemployment outside the home among the women in the sample. Literature shows that those who are most at risk are full-time housewives who stay at home to care for their children. Despite the problem of role strain experienced by the employed mother and the greater amount of time spent at home and in childcare compared to her husband, the employed mother has better physical and psychological health than the unemployed mother (Baruch & Barnett, 1986; Gove & Geerken, 1977). The present study focused on married Turkish immigrant couples living in Toronto. However, there is reason to believe that the relationships among variables may be generalizable to most married Turkish immigrants in Canada. On the other hand, some Ž ndings related to the speciŽ c acculturation attitude, amount of contact with the larger society, or degree of perceived discrimination may be speciŽ c to multicultural settings or to English Canada; hence, some caution needs to be exercised in making large generalizations. Social class differences would be expected to in uence psychological variables in the same way across different cultural groups, so results can also be useful for other immigrant groups where class differences are as striking. The present study examined the relations of Turkish immigrants with the dominant society only. In a city like Toronto, where 42% of the population are immigrants, one cannot assume that immigrants come into contact solely with the dominant group. In fact, conversations with the

ADAPTATION OF TURKISH IMMIGRANTS

participants revealed that they were in frequent contact with immigrant groups of different ethnicity. These relations among different ethnic groups are central to the functioning of plural societies and should be examined in future research. One limitation related to measurement is the use of selfreport measures. Among the many biases associated with such measures, social desirability and acquiescence are the two most relevant ones in the present case. Favourable selfpresentation (i.e., social desirability) may be a particular concern because Turkish society is collectivistic ; individuals are likely to present themselves in a socially desirable way (Triandis, 1989). Moreover, in Turkish society, marital and family problems are kept private and there is stigma attached to mental health problems. Desirability is socially construed; what is considered desirable in one social context may not be desirable in another. Social desirabilit y measures have not been cross-culturally validated; hence, such a measure was not used due to the possible cultural bias of the items in these measures. Acquiescence, the tendency to agree with or accept any assertion regardless of its content, is more common among less educated respondents (Krosnick, 1991). The bias may especially have come into effect due to the status difference between the researcher and the respondents of low SES. Finally, the cross-sectional and nonexperimental nature of the study pose some problems regarding the interpretation of the results. Possible cause and effect relationships are suggested, yet it is impossible to make deŽ nitive statements. However, the three-group design employed in the larger study, on which the present paper was based, has allowed us to make stronger claims by way of comparisons. A related point is the speciŽ c Ž nding of the study regarding the distinction among psychological , sociocultural, and marital adaptation. Although the distinction is supported, the interrelationships warrant further attention. These are related facets of adaptation ; hence, it would be desirable for future research to explore the antecedent-consequence distinction between psychologica l and sociocultural adaptation , and also between marital and psychological adaptation . Manuscript received June 1999 Revised manuscript accepted May 2000

REFERENCES Abadan-Unat, N. (1985). Identity crisis of Turkish migrants; First and second generation. In IÇ . Bas¸göz & N. Furniss (Eds.), Turkish workers in Europe (pp, 3–33). Bloomington, IN: Indiana University Turkish Studies. Abadan-Unat, N., & Kemiksiz, N. (1986). Türk dõ s¸ göçü, 1960–1984, yorumlu bibliyografya [Turkish external migration, 1960–1984, annotated bibliography]. Ankara: Ankara University, Faculty of Political Science Publications. Ataca, B. (1998). Psychological, sociocultural, and marital adaptation of Turkish immigrants in Canada. Unpublished doctoral thesis, Queen’s University, Kingston, Ontario, Canada. Aycan, Z., & Berry, J.W. (1996). Impact of employment-related experiences on immigrants’ psychological well-being and

25

adaptation to Canada. Canadian Journal of Behavioural Science, 28, 240–251. Bartone, P., Ursano, R.J., Wright, K.M., & Ingraham, L.H. (1989). The impact of a military air disaster on the health of assistance workers. Journal of Nervous and Mental Disease, 177, 317–328. Baruch, G.K., & Barnett, R.C. (1986). Role quality, multiple role involvement, and psychological well-being in midlife women. Journal of Personality and Social Psychology, 51, 578–585. Beiser, M., Barwick, C., Berry, J.W., daCosta, G., Fantino, A.M., Ganesan, S., Lee, C., Milne, W., Naidoo, J., Prince, R., Tousignant, M., & Vela, E. (1988). After the door has been opened: Mental health issues affecting immigrants and refugees in Canada. A report of the Canadian Task Force on Mental Health Issues affecting Immigrants and Refugees. Ottawa: Ministry of Multiculturalism and Citizenship and Health and Welfare Canada. Berry, J.W. (1976). Human ecology and cognitive style: Comparative studies in cultural and psychological adaptation. New York: Sage/Halsted. Berry, J.W. (1992). Acculturation and adaptation in a new society. International Migration, 30, 69–86. Berry, J.W. (1997). Immigration, acculturation, and adaptation. Applied Psychology: An International Review, 46, 5–68. Berry, J.W., & Kim, U. (1988). Acculturation and mental health. In P.R. Dasen, J.W. Berry, & N. Sartorius (Eds.), Health and cross-cultural psychology: Towards applications (pp. 207–236). Newbury Park, CA: Sage. Berry, J.W., Kim, U., Minde, T., & Mok, D. (1987). Comparative studies of acculturative stress. International Migration Review, 21, 491–511. Berry, J.W., Kim, U., Power, S., Young, M., & Bujaki, M. (1989). Acculturation attitudes in plural societies. Applied Psychology: An International Review, 38, 185–206. Berry, J.W., & Sam, D. (1997). Acculturation and adaptation. In J.W. Berry, M.H. Segall, & Ç. KagÆ õ tçõ bas¸ õ (Eds.), Handbook of cross-cultural psychology, Vol. 3, Social behavior and applications (2nd ed., pp. 291–326). Boston: Allyn & Bacon. Boyd, M. (1986). Immigrant women in Canada. In R.J. Simon & C.B. Brettell (Eds.), International migration: The female experience (pp. 45–61). Totowa, NJ: Rowman & Allanheld. Brislin, R. (1986). The wording and translation of research instruments. In W.J. Lonner & J.W. Berry (Eds.), Field methods for cross-cultural research (pp. 137–164). London: Sage. Brodman, K., Erdmann, A.J., Lorge, E., Gershenson, C.P., & Wolff, H.G. (1952). The Cornell medical index health questionnaire. Journal of Clinical Psychology, 8, 119–124. Cawte, J. (1972). Cruel, poor and brutal nations. Honolulu, HI: University of Hawaii Press. Cliff, N. (1987). Analyzing multivariate data. New York: Harcourt Brace Jovanovich. Cohen, J., & Cohen, P. (1983). Applied multiple regression/ correlation analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum. Dhawan, S. (1997). Caregiving stress and acculturation in East Indian immigrants caring for their elders. Unpublished doctoral thesis, Queen’s University, Kingston, Ontario, Canada. Diener, E., Emmons, R.A., Larsen, R.J., & GrifŽ n, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71–75. Dona, G., & Berry, J.W. (1994). Acculturation attitudes and acculturative stress of Central American refugees. International Journal of Psychology, 29, 57–70. Dyal, J.A., Rybensky, L., & Somers, M. (1988). Marital and acculturative strain among Indo-Canadian and EuroCanadian women. In J.W. Berry & R.C. Annis (Eds.), Ethnic psychology: Research and practice with immigrants, refugees, native peoples, ethnic groups, and sojourners (pp. 80–95). Lisse, The Netherlands: Swets & Zeitlinger. El Haïli, S., & Lasry, J. (1997, June). Roles conjugaux, harmonie maritale et stress psychologique chez des couples immigrants

26

ATACA AND BERRY

Marocains à Montrèal. Paper presented at the annual convention of the Canadian Psychological Association, Toronto, Canada. Furnham, A., & Bochner, S. (1982). Social difŽ culty in a foreign culture: An empirical analysis of culture shock. In S. Bochner (Ed.), Cultures in contact: Studies in cross-cultural interaction (pp. 161–198). Oxford: Pergamon. Gove, W.R., & Geerken, M.R. (1977). The effect of children and employment on the mental health of married men and women. Social Forces, 56, 66–76. Graves, T. (1967). Psychological acculturation in a tri-ethnic community. Southwestern Journal of Anthropology, 23, 337–350. Hearst, S. (1985). Turkish families. In D. Storer (Ed.), Ethnic family values in Australia (pp. 145–172). Sydney: Prentice Hall. Hocoy, D. (1993). Ethnic identity among Chinese in Canada: Its relationship to self-esteem. Unpublished master’s thesis, Queen’s University, Kingston, Ontario, Canada. Hocoy, D. (1997). Apartheid, racism, and black mental health in South Africa and the role of racial identity. Unpublished doctoral thesis, Queen’s University, Kingston, Ontario, Canada. Holmes, T.H., & Rahe, R.H. (1967). The social readjustment rating scale. Journal of Psychosomatic Research, 11, 213–218. Hudson, W.W. (1982). The clinical measurement package: A Ž eld manual. Chicago: Dorsey. Ilfeld, F. (1976). Characteristics of current social stressors. Psychological Reports, 39, 1231–1247. Jahoda, M. (1981). Work, employment, and unemployment: Values, theories and approaches in social research. American Psychologist, 36, 184–191. Judd, C.M., & McClelland, G.H. (1989). Data analysis: A modelcomparison approach. San Diego, CA: Harcourt Brace Jovanovich. KagÆ õ tçõ bas¸ õ , Ç. (1986). Status of women in Turkey: Crosscultural perspectives. International Journal of Middle East Studies, 18, 485–499. KagÆ õ tçõ bas¸ õ , Ç. (1987). Alienation of the outsider: The plight of migrants. International Migration, 25, 195–210. KagÆ õ tçõ bas¸ õ , Ç., & Berry, J.W. (1989). Cross-cultural psychology: Current research and trends. Annual Review of Psychology, 40, 493–531. Karman, D. (1994, May). Kanada göçmen istatistiklerinde Türkiye [Turkey in Canadian immigrant statistics]. Biz Bize, 2, 14–16. Keçeli, B. (1988). The traditional socialisation of the Turkish rural woman and its effects on her migratory settlement process. In R. Akçelik & J. Elley (Eds.), Turkish community in Australia (pp. 203–212). Melbourne: Australian-Turkish Friendship Society Publications No. 3. Kenny, D.A. (1988). The analysis of data from two-person relationships. In S. Duck (Ed.), Handbook of personal relationships: Theory, research, and interventions (pp. 57–77). London: Wiley. Kenny, D.A. (1996). Models of nonindependence in dyadic research. Journal of Social and Personal Relationships, 13, 279–294. Kenny, D.A., & Kashy, D.A. (1991). Analyzing interdependence in dyads. In B.M. Montgomery & S. Duck (Eds.), Studying interpersonal interaction (pp. 275–285). New York: Guilford Press. Koçtürk, T. (1992). A matter of honor: Experiences of Turkish women immigrants. London: Zed Books.

Kostovcik, N. (1983). Psychological adaptation of Malaysian students in Canada. Unpublished honor’s thesis, Queen’s Unviersity, Kingston, Ontario, Canada. Krosnick, J. (1991). Response strategies for coping with the cognitive demands of attitude measures in surveys. Applied Cognitive Psychology, 5, 213–236. Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer. Locke, H.J., & Wallace, K.M. (1959). Short marital-adjustment and prediction tests: Their reliability and validity. Marriage and Family Living, 21, 251–255. Lonner, W.J., & Berry, J.W. (1986) Sampling and surveying. In W.J. Lonner & J.W. Berry (Eds.), Field methods in crosscultural research (pp. 85–110). London: Sage. Öner, N., & Le Compte, A. (1985). (Süreksiz) Durumluk—sürekli kaygõ envanteri elkitabõ [Manual for state-trait anxiety inventory]. Istanbul: BogÆ aziçi University Publications. Pavot, W., & Diener, E. (1993). Review of the satisfaction with life scale. Psychological Assessment, 5, 164–172. Piontkowski, U., Florack, A., Hoelker, P., & Obdrzalek, P. (2000). Predicting acculturation attitudes of dominant and nondominant groups. International Journal of Intercultural Relations, 24, 1–26. Restoule, B. (1994). Cultural identity of Ojibwa youth: The relationship of acculturation, discrimination and multiculturalism to physical and mental health. Unpublished master’s thesis, Queen’s University, Kingston, Ontario, Canada. Searle, W., & Ward, C. (1990). The prediction of psychological and socio-cultural adjustment during cross-cultural transitions. International Journal of Intercultural Relations, 14, 449–464. Spielberger, C.D., & Diaz-Guerrero, R. (1982). Cross-cultural anxiety: An overview. In C.D. Spielberger & R. Diaz-Guerrero (Eds.), Cross-cultural anxiety, Vol. 2 (pp. 3–11). New York: Hemisphere/McGraw-Hill International. Spielberger, C.D., Gorsuch, R.L., & Lushene, R.E. (1970). Manual for State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. Statistics Canada (1997). 96 Census: Nation series edition 1 [CDROM]. Ottawa: Canada Communication Group. Taylor, D.M., & Moghaddam, F.M. (1994). Theories of intergroup relations: International social psychological persectives (2nd ed.). Westport, CT: Praeger. Triandis, H.C. (1989). Self and social behavior in differing cultural contexts. Psychological Review, 96, 269–289. Ward, C. (1996). Acculturation. In D. Landis & R. Bhagat (Eds.), Handbook of intercultural training (pp. 124–147). Thousand Oaks, CA: Sage. Ward, C., & Kennedy, A. (1992). Locus of control, mood disturbance and social difŽ culty during cross-cultural transitions. International Journal of Intercultural Relations, 16, 175–194. Ward, C., & Kennedy, A. (1993a). Psychological and sociocultural adjustment during cross-cultural transitions: A comparison of secondary school students overseas and at home. International Journal of Psychology, 28, 129–147. Ward, C., & Kennedy, A. (1993b). Where’s the culture in crosscultural transition? Comparative studies of sojourner adjustment. Journal of Cross-Cultural Psychology, 24, 221–249. Ward, C., & Kennedy, A. (1998). The measurement of sociocultural adaptation. Paper presented at the meeting of International Academy of Intercultural Research, Fullerton, California.