Int.J.InterculturalRel.Vol.23,No.4,pp.659–677,1999 #1999ElsevierScienceLtd.Allrightsreserved PrintedinGreatBritain 0147-1767/99$20.00+0.00 www.elsevier.com/locate/ijintrel PII:S0147-1767(99)00014-0 THE MEASUREMENT OF SOCIOCULTURAL ADAPTATION COLLEEN WARD* and ANTONY KENNEDY National University of Singapore, Singapore ABSTRACT. This paper examines the construct of sociocultural adaptation and describes the development and refinement of its measurement. Psychometric ana- lyses of the Sociocultural Adaptation Scale (SCAS) are presented based on the compilation of data across a large number of sojourner samples from an emerging program of research. The measurement and patterns of sociocultural adaptation are examined across: (1) 16 cross-sectional samples, (2) 4 longitudinal samples, and (3) 1 paired comparison between sojourning and sedentary samples. Selected cross-sample comparisons are reported, and the relationship between sociocultural andpsychologicaladjustmentacrosssamplesisdiscussed.#1999ElsevierScience Ltd.Allrightsreserved KEY WORDS. sociocultural, adaptation, adjustment, sojourner, cross-cultural .transition. INTRODUCTION Despite four decades of theory and research on ‘‘culture shock,’’ there is still limited consensus as to what actually constitutes sojourner adjustment. The construct has been described, interpreted and measured in varying ways and from numerous perspectives. Adjustive outcomes, for example, have been operationalized and examined in terms of: health-related variables, such as physical symptomatology and medical consultations (Babiker, Cox, & Miller, 1980); perceptual variables, such as cultural awareness (Gannon & Poon, 1997) and perceptual maturity (Yoshikawa, 1988); relational variables, such as feelings of acceptance This research has been supported by grants from the Foundation for Research, Science and Technology, Wellington, New Zealand (No. 8914); University of Canterbury, Department of Psychology; and the National University of Singapore (Nos. 92007 and 95002). *Corresponding author. Department of Social Work and Psychology, National UniversityofSingapore,KentRidge,Singapore117570.;E-mail:[email protected] 659 660 C.WardandA.Kennedy (Brislin, 1981) and quality of relationships with host nationals (Deshpande & Viswesvaran, 1992); intrapersonal or self-concept vari- ables, such as personal development (Gmelch, 1997) and identity con- flict (Leong & Ward, in press); variables linked to task-specific accomplishments, such as job performance (Parker & McEvoy, 1993) and academic achievement (Tanaka, Takai, Kohyama, & Fujihara, 1994); psycho-emotional variables, such as life satisfaction (Yoshida, Sauer, Tidwell, Skager, & Sorenson, 1997) and mood states (Stone Feinstein & Ward, 1990); and behavioral variables, particularly the ac- quisition of culturally appropriate skills (Furnham & Bochner, 1986), including communication e(cid:128)ectiveness (Witte, 1993). Unfortunately, most investigations have been undertaken in piecemeal fashion, making integration and synthesisof research findings di(cid:129)cult, if notimpossible. A smaller number of investigators have embarked upon more sys- tematic programs of research on cross-cultural transition and adap- tation, identifying adjustment domains and, in some cases, constructing predictive models of adjustive outcomes. This research has included data driven, factor analytic, studies by Hammer, Gudykunst and Wiseman (1978), which identified relational, communication and stress management components of intercultural e(cid:128)ectiveness, and work by Black, Mendenhall and Oddou (1991), which inspired research on gen- eral, job and interaction adjustment. It has also included more theory driven contributions by Berry (1997) on acculturative stress and Ward (1996) and associates (Searle & Ward, 1990; Ward & Kennedy, 1996a) on psychological and sociocultural adjustment. PSYCHOLOGICAL AND SOCIOCULTURAL ADAPTATION In an attempt to bring conceptual integration to a fractionated area of research, Ward and colleagues have proposed that cross-cultural ad- aptation may be meaningfully divided into two domains: psychological (emotional/a(cid:128)ective) and sociocultural (behavioral). The former refers to psychological well being or satisfaction; the latter is related to the ability to ‘‘fit in,’’ to acquire culturally appropriate skills and to nego- tiate interactive aspects of the host environment. Accordingly, Ward (1996) has argued that psychological adjustment can best be understood in terms of a stress and coping framework while sociocultural adap- tation is best explained within a social skills or culture learning para- digm. This has been borne out by empirical research which has demon- strated that the two adjustive outcomes, though inter-related, are con- ceptually and empirically distinct. Psychological adjustment, defined in terms of psychological and emotional well-being, is broadly a(cid:128)ected by TheMeasurementofSocioculturalAdaptation 661 personality, life changes, coping styles and social support. As examples, psychological adjustment has been associated with personal flexibility, internal locus of control, relationship satisfaction, approach-oriented coping styles, and use of humor, while psychological di(cid:129)culties in sojourners have been linked to a higher incidence of life changes, loneli- ness, stress, and avoidant coping styles (Berno & Ward, 1998; Searle & Ward, 1990; Stone Feinstein & Ward, 1990; Ward & Kennedy, 1998a; Ward & Rana-Deuba, 1999). Sociocultural adaptation, by contrast, defined in terms of behavioral competence, is more strongly influenced by factors underpinning culture learning and social skills acquisition. Theseinclude lengthofresidenceinthenewculture,culturalknowledge, amount of interaction and identification with host nationals, cultural distance, language fluency and acculturation strategies (Searle & Ward, 1990; Ward &Kennedy,1993a, 1994; Ward & Searle,1991). Research has also revealed that the two adjustive outcomes display di(cid:128)erent patterns of fluctuation over time. The sociocultural adaptation of sojourners predictably follows a learning curve with rapid improve- ment demonstrated over the first few months of cross-cultural transition and then a gradual ‘‘leveling o(cid:128)’’ of newly acquired culture-specific skills. Psychological adjustment is more variable over time although stu- dies have confirmed that the greatest di(cid:129)culties are experienced at the earliest stages of cross-cultural transition (Ward & Kennedy, 1996a, 1996b). Ultimately, the soundness of developing theory and research on psychological and sociocultural adjustment of sojourners rests on the measurement of the adjustive outcomes. In this regard, Ward and col- leagues have relied upon standard psychological assessment techniques for the measurement of psychological adjustment. The Zung Self-rating Depression Scale (Zung, 1965) has been most frequently used because of its widely documented cross-cultural reliability and validity although some studies have additionally incorporated the Profile of Mood States (McNair, Lorr, &Droppleman, 1971), given its inclusionofclassic‘‘cul- ture shock’’ symptoms—tension, depression, anger, fatigue, and con- fusion. The measurement of sociocultural adaptation, by contrast, has been undertaken with an author-devised scale. It is this measurement that is discussedin this paper. The Sociocultural Adaptation Scale (SCAS) was first used by Searle and Ward (1990) in their study of cross-cultural transition and adap- tation of Malaysian and Singaporean students in New Zealand. This paper describes the scale development and documents the usefulness and versatility of the measurement. The psychometric properties of the SCAS are reported along with the analysis of data from 16 cross-sec- tional samples, 4 longitudinal samples, and 1 pair of comparative sojourning and sedentary groups. The paper also reports exploratory 662 C.WardandA.Kennedy research on expanding the SCAS to include cognitive as well as beha- vioral domains. METHOD Materials The development of Sociocultural Adaptation Scale (SCAS) was inspired by Furnham and Bochner’s (1982) 40-item Social Situations Questionnaire (SSQ), based, in turn, on an earlier social skills scale by Trower, Bryant and Argyle (1978) and on interviews with foreign language students in Britain. The first version of the SCAS (Searle & Ward, 1990) contained 16 items. A portion of these items such as ‘‘deal- ing with people staring at you’’ and ‘‘dealing with people of higher sta- tus’’ were taken directly from Furnham and Bochner’s instrument; however, the item content of the SCAS was not confined to social situ- ations. Areas of adaptation such as food and climate were also included in thequestionnaire. Like Furnham and Bochner’s SSQ, the SCAS requires respondents to indicate the amount of di(cid:129)culty experienced in a number of areas by using a five-point scale (no di(cid:129)culty/slight di(cid:129)culty/moderate di(cid:129)culty/ great di(cid:129)culty/extreme di(cid:129)culty). Unlike the SSQ instructions and defi- nitions, however, di(cid:129)culty is not explicitly framed in a(cid:128)ective terms relating to anxiety, discomfort,and embarrassment. The SCAS is a flexible instrument and can be easily modified accord- ing to the characteristics of the sojourning sample. Table 1 lists items that have been used with the various samples. Most versions contain 20–23 items (See Table 2). Some items are specific to sojourner’s desti- nations, e.g., ‘‘eating at food stalls,’’ and others are specific to the ex- periences of student sojourners, e.g., ‘‘expressing ideas in class;’’ however, most items can be used across a diverse range of sojourning groups. The instrument was originally developed as an assessment of intercul- tural competence with emphasis on behavioral domains. The most recent version of the SCAS (with 29 items) has explored the addition of more cognitivedomains (e.g., items35–41). Samples and Procedures Cross-sectional Samples. The bulk of our research with the Sociocultural Adaptation Scale has been undertaken with sojourning samples connected in some way to either New Zealand or Singapore. Sixteen cross-sectional samples are described in Table 2. Sample size TheMeasurementofSocioculturalAdaptation 663 TABLE1 ItemsfromtheSocioculturalAdaptationScalea 1. Makingfriends 2. Usingthetransportsystem 3. Makingyourselfunderstood 4. Gettingusedtothepaceoflife 5. Goingshopping 6. Goingtosocialevents/gatherings/functions 7. Worshippinginyourusualway 8. Talkingaboutyourselfwithothers 9. Understandingjokesandhumor 10. Dealingwithsomeonewhoisunpleasant/cross/aggressive 11. Gettingusedtothelocalfood/findingfoodyouenjoy 12. Followingrulesandregulations 13. Dealingwithpeopleinauthority 14. Dealingwiththebureaucracy 15. Makingyourselfunderstood 16. Adaptingtolocalaccommodation 17. Communicatingwithpeopleofadifferentethnicgroup 18. Relatingtomembersoftheoppositesex 19. Dealingwithunsatisfactoryservice 20. Findingyourwayaround 21. Dealingwiththeclimate 22. Dealingwithpeoplestaringatyou 23. Goingtocoffeeshops/foodstalls/restaurants/fastfoodoutlets 24. Understandingthelocalaccent/language 25. Livingawayfromfamilymembersoverseas/independentlyfromyour parents 26. Adaptingtolocaletiquette 27. Gettingusedtothepopulationdensity 28. Relatingtoolderpeople 29. Dealingwithpeopleofhigherstatus 30. Understandingwhatisrequiredofyouatuniversity 31. Copingwithacademicwork 32. Dealingwithforeignstaffattheuniversity 33. Expressingyourideasinclass 34. Livingwithyourhostfamily 35. Accepting/understandingthelocalpoliticalsystem 36. Understandingthelocals’worldview 37. Takingalocalperspectiveontheculture 38. Understandingthelocalvaluesystem 39. Seeingthingsfromthelocals’pointofview 40. Understandingculturaldifferences 41. Beingabletoseetwosidesofaninterculturalissue aThefirsttenitemsarecommontoallcross-sectionalstudiesandareusedinthecomparative analyses. 664 C.WardandA.Kennedy y e n er N Source WardandSearle(1991)WardandKennedy(1992)WardandKennedy(1993a)WardandKennedy(1993b) WardandKennedy(1993b)WardandKennedy(1993c)WardandKennedy(1994)WardandChang(1997)Kennedy(1998)BernoandWard(1998)WardandInserto(1998)WardandChang(1998)WardandKennedy(1998b)Ward,ChangandLopez-(1999)WardandRana-Deuba(1999)Wardetal.(1999) ngenths) 3 Ra(mo 3^961^1320.5^182^96 0.2^111^3601^472^2285^8.42^603^4201^1802^5221^114 1^106^ s BLE2 onalSample Meanstay(months) 21.8(19.0)25.2(27.6)2.5(2.75)33.4(18.6) 32.6(31.1)90.4(85.2)18.0(11.8)30.7(34.5)5.9(0.7)22.8(16.2)100.3(92.1)35.5(31.3)53.5(89.2)33.0(22.9) 29.7(24.2)6.0 A ti T ec s s- s o 5485 6489840731 40 Cr N 1581714 1512913101010141119 109 w e N n i Sample MultinationalstudentsinNewZealandNewZealandersinSingaporeNewZealandAFSstudentsabroadSingaporeanandMalaysianstudentsZealandMalaysianstudentsinSingaporeBritonsinHongKongNewZealandcivilservantsabroadAmericansinSingaporeSingaporeanstudentsabroadMultinationalstudentsinNewZealandSingaporeansinUnitedStatesHongKongandPRCChineseinSingaporeBritonsinSingaporeFilipinadomesticsinSingapore MultinationalaidworkersinNepalaJapanesestudentsinNewZealand pondentsweresampledatthesametime. s e r 1.2.3.4. 5.6.7.8.9.10.11.12.13.14. 15.16. aAll TheMeasurementofSocioculturalAdaptation 665 ranges from 84 to 191 sojourners (Mdn=119). Mean length of stay var- ied from 2.5 months to 100.3 months(Mdn=30.2). Student samples are identified as such, e.g., Singaporean and Malaysian students in New Zealand and New Zealand AFS students abroad. Adult samples (e.g., Americans in Singapore and Britons in Hong Kong) generally include a cross-section of employed adults and in many cases their accompanying spouses or partners. Most of the adult samples are educationally and occupationally varied; however, in some instances, such as New Zealand civil servants on overseas postings or Filipina domestic helpers in Singapore, the samples are composed pri- marily or exclusively of people engaged in the same type of employ- ment. A number of multicultural samples are also included, such as overseas student samples in New Zealand, and in some instances samples are composed of members of a single cultural/national group who have relocated to a variety of overseas destinations (e.g., Singaporean students abroad). All samples, with the exception of Filipina domestic helpers in Singapore, include both males and females. Thedatapertaining tolengthofresidenceinhostcountry,psychological and sociocultural adaptation have been extracted from larger data sets which examined the prediction of psychological and sociocultural adjustment in sojourners. Longitudinal Samples. Four longitudinal samples are included in this research. The first is composed of 14 (8 males and 6 females) Malaysian and Singaporean secondaryand tertiary students(M age=19.1 years) in New Zealand (Ward & Kennedy,1996a). These students completed a 20 item version of the SCAS within one month of arrival in New Zealand, and again 6 and 12 months later. The second sample included 14 participants in New Zealand’s Volunteer Service Abroad (VSA) program (Ward & Kennedy, 1996b). The VSA program (akin to the American Peace Corps) provides volun- teer workers for various projects in developing countries. Ten men and four women (M age=37.9 years) completed a 23 item version of the SCAS at three points in time: in New Zealand before departure to their overseas assignments, within two months of arrival at their destinations (M=9.7weeks),andapproximatelyoneyearinto theirfieldassignments (M=52.1 weeks). The third sample (a subset of cross-sectional sample # 16 consisted of 35 (17 men and 18 women) Japanese students (M age=18.6 years) in New Zealand (Ward, Okura, Kennedy, & Kojima, 1999). These stu- dents attended an international college specializing in courses for Japanese students who wish to advance to tertiary education in English- speaking countries. The students completed a 20 item version of the SCAS on 4 occasions: the first within 24 h of arriving in New Zealand, 666 C.WardandA.Kennedy h CAS(fullscale)elationwitZSDS ***0.260.25****0.62***0.41 ***0.54**0.23***0.36***0.45***0.31***0.62***0.530.20****0.50***0.27**0.27***0.33 1) S orr 0 c 0 0. ***1,P5 Range 0^321^14.10^300^23 0^240^250^210^280^180^280^170^250^310^272^260^37 0 0. **0.05,P5 Meanscore10itemscale 10.2(5.5)6.2(3.1)10.6(5.5)9.6(5.2) 7.4(4.7)8.4(3.9)7.7(5.8)9.8(5.5)6.8(4.3)9.6(5.8)2.3(3.5)6.8(4.8)10.4(5.5)11.4(5.9)9.7(4.6)11.9(7.0) 5 P amples(* Range 0.1^3.40.05^1.60^3.00^2.3 0^2.30.1^2.50^2.00.04^2.50^1.90.04^3.30^1.30^2.40.1^3.20.4^3.20.1^2.40^3.5 S al n oss-sectio Meanitemscore 1.1(0.5)0.7(0.3)1.1(0.5)1.0(0.5) 0.9(0.4)0.9(0.4)0.8(0.5)1.0(0.5)0.7(0.4)0.9(0.6)0.2(0.3)0.7(0.4)1.0(0.5)1.3(0.5)1.0(0.4)1.1(0.6) 3 Cr LE 16 TAB alein Alpha 0.840.760.850.86 0.840.750.880.860.890.910.910.850.860.850.810.88 c S n atio o.ofemsN 20212020 202122232923232221232320 pt Nit a d A w ural Ne t cul in DescriptiveStatisticsfortheSocio Sample MultinationalstudentsinNewZealandNewZealandersinSingaporeNewZealandAFSstudentsabroadSingaporeanandMalaysianstudentsZealandMalaysianstudentsinSingaporeBritonsinHongKongNewZealandcivilservantsabroadAmericansinSingaporeSingaporeanstudentsabroadMultinationalstudentsinNewZealandSingaporeansinUnitedStatesHongKongandPRCChineseinSingaporeBritonsinSingaporeFilipinadomesticsinSingaporeMultinationalaidworkersinNepalJapanesestudentsinNewZealand 1.2.3.4. 5.6.7.8.9.10.11.12.13.14.15.16. TheMeasurementofSocioculturalAdaptation 667 the second 4 months after arrival, the third 6 months after arrival and the final questionnaire 12 months after the initial arrival and an inter- vening school break. Thefinallongitudinalsample(seecross-sectional sample#9)included 108 (47 male and 61 female) Singaporean students (M age=20.0 years) studying in Australia, New Zealand, People’s Republic of China, the United States and the United Kingdom (Kennedy, 1998). These stu- dents completed a 29 item version of the SCAS three times: 1 month prior to departure (M=25 days), 1 month after arrival (M=4.5 weeks) and 6 months after arrival (M=25.2 weeks) at their overseas desti- nations. Home-based Sample. A total of 142 New Zealand secondary students (49 females, 92 males, 1 unspecified) at home (M age=17.5 years) pro- vided a comparison group for 178 New Zealand AFS students abroad (cross-sectional sample # 3). In this instance 16 common items from the SCAS served as the basis of the comparison of New Zealand students at home and abroad (Ward & Kennedy, 1993a). RESULTS Cross-sectional Comparisons Table 3 reports evidence of scalar reliability. The scale alphas range from 0.75 to 0.91 (M=0.85). Across a diverse selection of sojourning samples the scale has retained good internal consistency. Table 3 also includes evidence of construct validity as indicated by the consistently significant correlations (range=0.20–0.62, M=0.38) between sociocul- tural and psychological adjustment as measured by the Zung Self-rating Depression Scale (Zung,1965). Mean sociocultural adaptation scores based on the common 10 core items in the 16 cross-sectional samples are also presented in Table 3. In addition, the mean item score is presented for all samples. Selected com- parisons by analysis of covariance (controlling for length of residence) are reported below. In all cases comparisons are made based on the maximum number ofcommon SCAS items. The results reveal predictable di(cid:128)erences across sojourning samples and suggest that sociocultural adaptation problems decrease as a func- tion of ethnic and cultural similarity. For example, Chinese sojourners in Singapore experienced less sociocultural di(cid:129)culties (M=8.9) than non-Chinese (British, American and New Zealand) sojourners (M=11.8); F (1, 475)=22.0, P50.0001 (12 item SCAS). In addition, Malaysian and Singaporean students in New Zealand (M=17.5) reported more sociocultural adaptation problems than Malaysian stu- 668 C.WardandA.Kennedy dents in Singapore (M=14.8); F (1, 289)=8.5, P50.01 (17 item SCAS, Ward & Kennedy, 1993b). Data also suggest that fewer adaptation problems are encountered by those with good resources and by those who make transitions to more ‘‘comfortable’’ environments. New Zealand civil servants abroad (M=12.8) reported fewer sociocultural adaptation problems than a younger group of New Zealand AFS students abroad (M=17.1); F (1,258)=4.6, P 5 0.04 (16 item SCAS). Multinationals, i.e., British, New Zealand, American, Hong Kong and PRC Chinese, in Singapore (M=10.9) experienced less sociocultural adaptation problems than mul- tinationals in Nepal (M=13.1); F (1, 579)=9.3, P 5 0.002 (12 item SCAS). While considerable variation exists in sociocultural adjustment across samples, some sojourning groups appear particularly adaptable. Singaporeans in the United States (M=4.0), for example, encountered fewer sociocultural di(cid:129)culties than Americans in Singapore (M=23.4); F (1, 236)=149.0, P50.0001 (23 item SCAS). Similarly, Singaporean students abroad experienced less sociocultural adaptation problems (M=10.6) than New Zealand students overseas (M=16.4); F (1,268)=22.3, P50.0001 (15 item SCAS). Despite these variations, similarities also emerged in the cross-sample comparisons. As might be expected, there was no significant di(cid:128)erence in the sociocultural adaptation of British expatriates in Hong Kong (M=19.5) and Singapore (M=21.3); F51 (21 item SCAS). Nor was a significant di(cid:128)erence found between the 1991 (M=18.6) and 1998 (M=17.7) multinational samples of foreign students in New Zealand; F 51 (17 item SCAS). Longitudinal Analyses Longitudinal data confirmed that sociocultural adaptation problems are greatest during the early stages of transition and that they decrease significantly over time. For the Malaysian and Singaporean students in New Zealand a one way analysis of variance indicated that social di(cid:129)culty was greatest during the first month (M=26.4), that it dropped sharply during the first 6 months (M=17.3) and that it continued in a slightly downward direction 12 months later (M=16.8); F (2, 26)=12.8, P 5 0.0001 (20 item SCAS;Ward & Kennedy, 1996a). For VSA volunteers sociocultural di(cid:129)culties significantly decreased between two (M=27) and 12 (M=24.5) months; t (13)=1.81, P50.05, one tailed (23 item SCAS). Interestingly, the volunteers expected socio- cultural adaptation to be more problematic as the mean predeparture expectations scoreon the SCAS was 32.9 (Ward &Kennedy,1996b).