Introduction This study examined dual group identification in Australian second generation immigrant adolescents. Dual identification, as used in this study refers to self-categorization and self-identification by ethnic minority adolescents `based on the perception and conceptions of themselves as belonging to an ethnic group' as well as mainstream society (Phinney & Rotheram, 1987, p.292) . Increasingly many children and adolescents are growing up in societies where members of the ethnic or racial group with which they may identify, and into which others categorize them, form a minority either in numbers or power and status. There is evidence that minority children are aware, from a very young age, that they are different from others and that the knowledge of their membership of an ethnic group forms a significant aspect of their personal identity and self-definition (Aboud, 1988; Semaj, 1980; Tajfel, 1981; Turner, 1987). While being socialized into their ethnic group, minority youth are simultaneously learning to live in, acquire the values of and identify with mainstream society. Results of preference studies (Cross, 1987; Vaughan, 1987) indicate that ethnic minority parents want their children to develop bicultural values and skills to succeed both in the majority culture and in their own minority culture. Research on cross-cultural training (Triandis, Kurowski, Tecktiel & Chan, 1993) suggests that to improve relationships in culturally diverse situations, one must teach members of each cultural group to extend the attributions they make, so that they consider the attributions other-group members make in addition to the attributions their own group normally make. Such sophisticated social perception is one of the characteristics of bicultural individuals. Turner (1978) claimed that a single source of reference and self-definition tends to restrict the range of behaviour and relationships available to an individual, be they a member of a minority or the majority culture. While there is general support for the notion that dual identification is not only conceptually reasonable but also psychologically desirable, some studies have shown that minority children and adolescents have a complex and problematic identity formation. Further, the development of dual identification in children and adolescents appears to be affected by developmental factors as well as the visibility and status of their specific group within mainstream society (Phinney, 1989; Spencer & Markstrom, 1990). The issue is of relevance to educators in a multicultural as well as increasingly multiracial society such as Australia, whose schools are grappling on a daily basis with the promotion of a positive social identity of its young people as well as their integration into a cohesive society. It is not clear, at present, how second generation immigrant children and adolescents are dealing with the dynamics of dual identification In the Australian context. This study addresses the question of the relationship between minority adolescentsÕ identification with their ethnic group and identification with the majority culture and the factors that affect these identifications. This study aimed to: a) find whether North American (Bat-Chava & Deaux, 1990) findings that majority and minority identifications are independent dimensions will be replicated in Australian ethnic minority adolescents; b) examine whether the two identifications and their relationship to each other vary with age and level of acceptance of their ethnic group by mainstream society; c) gain some qualitative data on how adolescents of different ages negotiate between various aspects of their majority and minority identification. The Relationship between Minority and Majority Identification Bat-Chava & Deaux' (1990) research addressed the question whether majority and minority identifications must be in conflict or whether minority group members can identify both with their minority group to maintain high self-esteem and the majority to achieve success in the mainstream. Their study with Black1 American college students supported their hypothesis that minority and majority identifications were independent dimensions. A number of other studies have shown that dual identification in minority youth is possible and yields favourable results. Studies carried out in the seventies and early eighties of Greek, Latvian, and Jewish Australians (cited by Rosenthal, 1987) and research with deaf adolescents (Weinberg and Sterritt, 1986) found that minorities can live harmoniously within two cultures. Aboud (1992) suggested that adolescents who possess multiple social relations are less likely to be involved in intergroup conflict than those who have only one predominant group affiliation. However, because of their multiple group membership, some minority children may experience inconsistent pressures to conform to the conflicting values of the larger society and their immediate reference group. They may also be unwillingly involved in racial or ethnic conflict instigated by adolescents from other minority as well as majority groups, adults and mainstream institutions that categorize them as lower status outgroups (Aboud, 1992). Phinney and Alipura (1990) found that ethnic identity exploration was higher among ethnic minority than among White American college students. Katz (1981) noted that many American Indian youth swing back and forth between an Anglo-identity and a tribal identity, having to contend with a negative or ambivalent feeling regarding their own ethnic identity. Crocker and Major (1989) investigated the self-protective properties of stigmatized groups, showing that positive evaluation of the attributes that define oneÕs ethnic group and identification with it have a function in protecting the self-esteem of members of groups that are the targets of prejudice or victims of discrimination. Ethier & Deaux (1990) also found that strength of cultural background generally acted as a buffer to perceived threat. However, Steele (1988) and Crocker and Major (1989) suggested that strong minority identification involves the devaluation of the skills and attributes of the majority group and that this, in turn, may be detrimental to success in mainstream society. In Bat-Chava and DeauxÕ(1990) investigation, Black students who highly identified with their minority group perceived more racial tension in their college and had higher levels of stress than those who identified less with their group. Dual Identification and Ethnic Group Status According to Rosenthal (1987) a sense of ethnic identity arises not only from an individual's interaction with his or her ethnic group but in relation to that groupÕs place in the wider society. She referred to a double boundary around minority group members, a boundary from within, related to the socialization and associations within their ethnic community and a boundary from without, resulting from interactions with members of other groups, in particular, the dominant societal group. Bat-Chava and Deaux (1990) also noted that the effect of stigmatization as well as the different cultural heritages of the various groups will affect the ways in which members perceive their group and the majority group. The opportunities afforded and constraints imposed by the majority society, the power, status and prestige accorded an ethnic group and the ensuing degree of social acceptance are important in influencing both oneÕs minority and oneÕs majority identification. The Inquiry on Racist Violence in Australia pointed out that `the more visibly different a person is the more likely he or she is to be subjected to racial discrimination or racist violence' (Human Rights and Equal Opportunity Commission, 1991, p.138). One would expect that adolescents who come from groups that may be ethnically, but not visibly, different from the majority and which are well accepted by the mainstream (OMA, 1989), such as those from the earlier-arrived Mediterranean ethnic groups within Australia, may encounter a weaker boundary from the mainstream to their self-identification as Australian (the term generally applied to the majority population as distinct from the ethnics) and as a result may more readily refer to themselves as both ethnic and Australian. On the other hand, non-acceptance by the majority, a strong boundary from without, may deter identification with the majority group. This may apply to adolescents belonging to groups who are ethnically as well as visibly different, such as those from Asian or Lebanese background, at present the two least-favoured immigrant groups (OMA, 1989) and the ones which, according to the Inquiry, were most likely to be victims of racist intimidation, harassment and violence. A strong boundary from within, imposed by some minority families and ethnic groups, may also hinder dual identification. This applies to ethnic groups who have established institutions that are central to young people, such as school, religion or clubs, so that most of their social interactions take place largely within the group (Rosenthal, 1987). Dual Identification and Developmental Factors Age may also affect the two identifications. Given that identity formations consolidate throughout the period of adolescence, dual identification processes may well undergo developmental changes in this period. As they cognitively mature, ethnic minority adolescents become acutely aware of the evaluations of their group by the majority (Comer, 1988). The data exploring developmental changes through the adolescent years are sparse. One difference that emerged in a study by Rosenthal and Hrynevich (1985) was that younger adolescents were less likely than older adolescents to perceive their group as a culturally cohesive entity. For older adolescents, however, issues such as maintenance and participation in cultural traditions, endogamy, maintenance of ties with other members of oneÕs group and rejection of assimilation were all associated with membership in their ethnic group. This study examined the effect of age on dual identification. Method Participants A sample of 114 Australian adolescents, in age groups of 11-12, 13-14 and 15-16 ((34, 45 and 35, respectively) from a Sydney inner-city primary school and two inner-city high schools, a girlsÕ and a boysÕ school, participated in the study. Close to 40% of the three schoolsÕ student population was of non-English-speaking background (NESB). Participants were selected for the sample if they were born in Australia, but had both parents born in a non-English-speaking country. There were 60 males and 54 females. According to the children's ethnic background, 40% of them were from Mediterranean (Italian, Spanish and Greek), 20% from Asian (Vietnamese, Chinese, Indian and Filipino), 15% from Lebanese and 25% were from a variety of other backgrounds. In one-to-one interviews, participants were asked to respond to the Dual Identification Scale, to elaborate on some of the items on the scale and describe themselves in terms of their ethnicity . Measures Self-Description. Participants were asked which of the following best described their ethnic or national identification: Ethnic (naming their ethnic group based on their parentsÕ birthplace and language spoken at home); Ethnic/Australian; Australian/Ethnic (putting first the one which was more important to them) or Australian. Dual Identification Scale (Bat-Chava & Deaux, 1990). Sixteen items constituting the first two major factors in the original 5-point Likert-type Dual Identification Scale were selected. These two dimensions, respectively, reflected majority group (7 items) and ethnic minority group (9 items) identification. One item in the majority identification scale was dropped in the final analyses because it asked children's preference among their teachers and was quite distinct from the remaining items examining relations with their friends. (Without this item, the subscales are more homogeneous in content, yielding higher reliabilities). Since Bat-Chava and Deaux had developed the scale to measure dual identification in American students, the language of the 15 items was adapted, so it could be understood by Australian adolescents. For example, the item `Almost all the people I have dated are White' was changed to `Almost all the boys/girls (opposite sex) that I really like are Australian' and `I often feel like a chameleon, having to change my colours depending on the race of whom IÕm with' became `I often feel like someone who has to change the way they behave and talk depending on the race/ethnic background of who IÕm with`. Based on extensive piloting of the term, with groups of minority and majority adolescents, Australian 2 was used to refer to the majority Anglo-Australian or in children's words "aussie" culture and people. The Dual Identification Scale includes items measuring the notion that `race does not matter'. Items on the Majority Identification scale refer to friendships or socializing with majority group peers, preference for the majority and distancing oneself from one's ethnic group. Items on the Minority Identification scale refer to friendship and other ties (such as choice of doctor, lawyer, neighbourhood) with oneÕs ethnic group and discomfort with the majority group. Qualitative data. In addition to responses to the Dual Identification Scale, some qualitative data, related to the adolescents bicultural competence was obtained. They were asked to elaborate on their responses to one question on each of the two subscales. Adolescents, who disagreed with item 5 of the Majority Identification Scale `I feel comfortable talking to my Australian friends about my family' and those who agreed with item 8 of the Minority Scale `I often feel like someone who has to change the way they behave and talk depending on the ethnic background of who IÕm with' were asked to `tell and explain more' about their answer. The first item taps into a subject that Australian adolescents often share with their peers, the second relates to the way minority children negotiate between different aspects of their bicultural world, a focus of this study. Results Self-Description Twenty seven per cent of this sample of NESB second generation Australian adolescents identified as hyphenated Australians (e.g. Italian Australian or Australian Italian), placing in equal proportion either Australian or their ethnic group first; 61% identified themselves by sole reference to their ethnic group such as `I am Lebanese' and 11% identified themselves as Australian. Dual Identification Scale The means and standard deviation of students' responses to each of the fifteen items in the two scales are shown in Table 1. As can be seen from the table, adolescents agreed most with majority items 3 and 5, referring to ' having close Australian friends' and 'talking to them about their families' and 'minority items 5 and 6, referring to 'uncomfortable at all-Australian parties' and 'behaving Biculturally'. They agreed least with majority items 4, 6, referring to 'having few same-ethnicity friends' and 'discomfort at all-minority parties' and minority item 1, referring to choosing a seat in the playground 'according to others' ethnicity'. Principal components analyses with the scree test and eigenvalue greater than unity criteria on the majority and minority identification items produced two and three factors respectively (see Table 1). The solutions accounted for 45.7 and 53.1% of the total variance of the two scales. It was found that the orthogonal rotation with the varimax method and the oblique rotation with the oblimin method produced very similar patterns. Consequentially five subscale scores were obtained by summing the respective items in the factor. For Majority Identification, the first factor was labelled as Assimilation Orientation (majority items 7,6,4]), while the second factor was labelled Socializing with the Majority (items 2,3,5). The three factors in the Minority Identification scale were labelled Ethnic Orientation, Discomfort with the Majority and Behaving Biculturally respectively (for items 2,3,1,9; items 5,4,7; and items 6,8) Correlations among the subscales in the same domain of identification (i.e., within majority or minority) were calculated. It was found that the two factors within Majority Scale were not correlated, whereas those within Minority Scale were moderately associated (see Table 2). Children's assimilation orientation (e.g. Almost all the boys/girls that I really like are Australian) was not related to socializing with majority group peers, but their ethnic orientation (e.g. If I were sick I would prefer to go to a well-trained own ethnicity doctor) was related to discomfort in majority settings and difficulty with bicultural behaviours. Table 1 Mean, Standard Deviations, and Factor loadings on Majority and Minority Identification ______________________________________________________________________ Factor loadingsc ______________________________________________________________________ Item contenta Meanb SD123 ______________________________________________________________________ Majority Identification 7. Almost all opposite sex friends Australian 2.32 .91.75 6. Feel uncomfortable at own ethnicity party 2.14 .87.72 4. Have very few own ethnicity friends 2.03 .99.64 2. Feel comfortable in an all Australian class2.53 .96 .77 3. Have many close Australian friends 3.39 1.14.57 5. Feel comfortable talking about family to 3.19 1.10.51 Australian friends Eigenvalue1.561.18 % of variance25.919.7 ______________________________________________________________________ Minority Identification 2. More comfortable with own ethnicity police 2.65 1.03.74 3. If sick prefer own ethnicity doctor 2.74 1.00.74 1.A seat in the playground according to ethnici2.18 .92.64 9. Prefer own ethnicity teacher 2.57 .94.46 5. Uncomfortable at an all Australian party 3.15 1.11.73 4. Prefer to live in a mainly own ethnicity area2.84 1.13.71 7. Prefer own ethnicity lawyer if had problem 2.50 .83.60 6. Behave differently according to context 3.11 1.17.78 8. Find being bicultural difficult 2.71 1.05 .73 Eigenvalue2.421.25 1.11 % of variance 26.813.912.4 ________________________________________________________________________ _ a The direction of wordings of some items have been reversed in the original questionnaire. b Five-point scales for all items, 1 = disagree strongly, 5 = agree strongly. c Only loadings greater than .35 are shown. Correlations between the subscales in the Majority and Minority Identifications were also calculated (see Table 2). It was found that Assimilation Orientation was negatively correlated with Ethnic Orientation, Discomfort with the Majority and Behaving Biculturally, whereas Socializing with the Majority was positively related to Ethnic Orientation. That is, children who sought to identify only with the majority culture (rejecting socializing within one's minority group) identified less with their own group, reported less discomfort in majority settings and less concern with bicultural behaviours. On the other hand, children who did socialize more with majority peers were more likely to have a stronger ethnic identity . Table 2 Correlations among Subscales in Majority and Minority Identification ___________________________________________________________ Subscales 1 2 3 4 ___________________________________________________________ Majority Identification 1. Assimilation Orientation 2. Socializing with Majority.01 Minority Identification 3. Ethnic Orientation-.27** .21* 4. Discomfort with Majority-.44*** .14 .26** 5. Behaving Biculturally-.34*** .01 .28** .22* __________________________________________________________ *p < .05, **p < .01, ***p < .001. Identification and Age The approximately equal number of males and females in the Majority and Minority Identification subscales were compared with t-tests. Results showed that there was no significant difference in all the comparisons, t (112) = .03, .81, .10, .99, and .82 for the five factors in majority and minority identification respectively, all non-significant. Thus, male and female subjects were combined in the following analyses. The number of adolescents in the three age groups were approximately equal (N11-12 = 34, N13-14 = 45, N15-16 = 35) A series of one-way ANOVAs were used to compare the students' responses in the three age groups. As can be seen from Table 3, there was a curvilinear tendency for the middle age group (age 13-14) to be more assimilationist, to feel less uncomfortable with the majority and have less bicultural behaviour concerns (Majority Factor 1 and Minority Factors 2 and 3) as compared with the other two age groups (age 11-12 and 15-16). Furthermore, it was noted that the 15-16 year olds socialized less with the majority and yet identified less with their ethnic group than the younger groups . (Note: these findings need to be treated with caution, because all these differences were not big enough to show that the curvilinear differences were significant). Table 3 ANOVAs of Majority and Minority Identification Subscales by Age Group (in years) ___________________________________________________________________ 11-12 13-14 15-16 ___________________________________________________________________ M M M F-values ___________________________________________________________________ Majority Identification 1. Assimilation Orientation 2.62 3.25 3.16 11.51*** 2. Social with Maj 2.50 2.19 1.80 11.75*** Minority Identification 3. Ethnic Orientation 2.90 2.48 2.24 10.49*** 4. Discomf with Majority 3.08 2.60 2.89 4.59* 5. Behaving Biculturally 3.25 2.64 2.91 4.60* ___________________________________________________________________ *p < .05, **p < .01, ***p < .001. Identification and Self-Described Ethnicity Children's responses were also compared by their country of birth and self-described ethnicity. Separate ANOVAs were employed with parents' country of birth and adolescents' self-described ethnicity being used as the independent variable and the five identification subscales as the dependent variables. There were no significant differences in Assimilationist Orientation, Socializing with the Majority and Discomfort with the Majority among the adolescents from different ethnic background. However, there was a tendency for Asian-Australians to identify more strongly with their ethnic group and have greater difficulties with bicultural behaviours than those from other backgrounds. There was no significant difference in Socializing with the Majority and Ethnic Orientation among adolescents who described themselves as Ethnic, Australian, Ethnic-Australian, or Australian-Ethnic (see Table 4). However, those who described themselves as either ethnic-Australian or Australian-ethnic indicated less discomfort in majority settings and had less difficulty with bicultural behaviours than the other groups. Children describing themselves as Australian were the ones who were most uncomfortable in majority settings and had greatest concerns about bicultural behaviours as compared with the other groups. Table 4 ANOVAs of Majority and Minority Identification Subscales by Self-Described Ethnicity ______________________________________________________________________ Ethnic Aust Aust- Ethnic- F-values Ethnic Aust ______________________________________________________________________ Majority Identif 1. Assimil Orientation 2.93 2.85 3.29 3.39 3.39* 2. Social with Majority 2.12 2.49 1.98 2.25 1.68 Minority Identification 3. Ethnic Orientation 2.54 2.83 2.25 2.5 11.78 4. Disc with Majority 2.92 3.26 2.50 2.41 5.17** 5. Behaving Biculturally 2.99 3.62 2.50 2.35 6.83** ______________________________________________________________________ *p < .05, **p < .01, ***p < .001. Qualitative Responses The qualitative data provides clarification of some of the concerns of this group of adolescents. It focused on their elaboration of two items, Majority item 5 where 38% expressed discomfort with talking to their Australian friends about their families; and Minority item 6, where 43% said that they had no difficulties with being bicultural. Adolescents' comments on each of the 2 questions are found in Appendices A & B. Of the twenty students who elaborated on why they did not feel comfortable talking to my Australian friends about my family only two explained that it was because that was their `personal private life'. The other 18 responses all referred to their perception that their Australian peers would laugh or tease or not understand or call them names because of their differences. Typical answers included: `IÕd get embarrassed, they might tease you. TheyÕll think I am a wog'; `Australians wonÕt understand. they might react in a funny way' and `Our parents teach us different things. My parents more strict. They would tease me'. Forty-four adolescents elaborated on the biculturalism question: I often feel like some one who has to change the way they behave and talk depending on the ethnic background of whom IÕm with. Their responses paint a vivid picture on how they negotiate moving from one cultural world to another in their daily lives. Though their comments indicated a number of patterns, one thing stands out. Their perception was that in their own ethnic communities there was greater emphasis on good manners and more constraints on their behaviour. Some liked the greater freedom they perceived they had in Australian contexts `to muck around' and be `more outgoing'. Others respected and felt more comfortable in their ethnic group as some of the following examples show: "With Lebanese you have to talk serious, with Australians you talk freely" `With Portuguese you behave a polite way, with Australians more rude'; `Behave more mature with Indian people. Talk about cricket. With Australians you talk about subjects and teachers. Muck around more'; `With Aboriginal people you know each other, talk Aboriginal lingo. We handle each other. After fights we make up. With Australians more angry'; `Tongans have different ways-concerts with Tongan dances. I am more myself with Tongans'; `I am more quiet with Australians. With Italians more comfortable'. Four respondents referred only to language difficulties, finding it easier to communicate with same first language peers, with whom they could speak half/half as in the following: 'With Filipinos I talk half Tagalog, half English.' The responses indicate adolescentsÕ awareness of the need for bicultural competencies to function in their different settings. Though two thirds did not think this a difficult task, one third did. It needs to be also noted that three quarters of the group thought that they would not 'feel very comfortable in a class where all the other boys and girls were Australian.' Conclusions The results of this study of dual identification in a sample of Australian-born adolescents from NESB show that the relationship between majority and minority identification is more complex than suggested by Bat-Chava and Deaux' (1990) North American study. The Minority Identification scale yielded three positively related factors, Ethnic Orientation, Discomfort with the Majority and Bicultural Behaviours, indicating coherence of the minority identification dimension. Adolescents who preferred to seek help or services, such as seeing a doctor, asking advice from a policeman or a teacher from their ethnic community, were less comfortable at an all-Australian party and preferred to live in their "ethnic" area . They were also aware of the need for and possible difficulty with bicultural behaviours. The Majority Identification scale, on the other hand, did not represent a coherent dimension in this Australian sample, as it did with Bat-Chava and Deaux' Black American students. The Majority items appeared to tap into two unrelated factors, each of which related to Minority Identification in a different way. The first factor, Assimilation Orientation, which included items such as comfort in a class where all one's peers were majority Australians and being mainly attracted to opposite sex peers from the majority group, was negatively related to all the three Minority Identification factors. That is, adolescents who were oriented towards being "Australian", preferred not to mix primarily within their ethnic community and did not feel that they had to "change the way they behave and talk depending on the race/ethnic background of whom I'm with". The second Majority factor, Socializing with the Majority, however, showed a moderate positive correlation with Ethnic Orientation. In this Australian sample, preference for and greater comfort in mixing primarily with members of one's ethnic community did not preclude socializing with, and being comfortable with mainstream peers. At the same time, preference for Australian friends and being uncomfortable with one's own ethnicity friends did not necessarily mean that one felt comfortable in all-Australian settings or that one had many close Australian friends. It would appear that young people, who turn away from their primary group and look primarily towards the majority group, may potentially face alienation from both and find themselves in no man's land. The findings lend support to the notion that dual identification is psychologically desirable. Adolescents' identification on the two scales did vary according to age. The middle age group, 13-14 year olds, had a stronger assimilationist orientation, indicated less discomfort mixing in mainstream settings and had less difficulty with bicultural behaviours than either the younger or older groups. While seeing themselves as part of mainstream society , they managed to combine the beliefs and behaviours of their ethnic group and that of the majority (as they perceived it). It is the oldest group, 15-16 year olds, who seemed to have the greatest difficulties with their identification. They showed the lowest identification with their ethnic group but were also the group which was least comfortable in socializing with the majority. They felt uncomfortable in an all Anglo-Australian class, did not have many close Anglo-Australian friends and hesitated in talking to them about their families. These results may be explained by ComerÕs (1988) finding that, as ethnic minority adolescents mature cognitively, they become more acutely aware of how the majority evaluates them. At the same time, their weaker ethnic identity means that they do not benefit from the self-protective "buffer" against outside stigma, that, according to Crocker and Major (1989), is provided by a positive evaluation of one's ethnic group. There was a moderate relationship between ethnic background and the two Minority Identification factors. Asian Australian adolescents tended to identify more strongly with their ethnic group and feel more uncomfortable with bicultural behaviours as compared with those from Mediterranean, Lebanese and other backgrounds. This may relate to the fact that Asian Australians are among the most recent wave of immigrants, who, like first generation people in earlier migrations, live primarily within their communities, and/or it may reflect their greater visibility and awareness of barriers from 'without'. Follow-up studies of these Australian-born adolescents as they enter adult life as well as research of the next generation of visibly different young Australians will show whether Australia manages to become a multi-racial society where individuals from diverse backgrounds can freely negotiate their bicultural identification. Adolescents who described themselves only in terms of their majority identification, that is, as 'Australians' had greater difficulties with adaptive bicultural behaviours and paradoxically, were more uncomfortable in all-majority settings as compared with their peers who described themselves in Ethnic-Australian terms. As indicated above, comfort in socializing with majority members appeared problematic for some of the sample, particularly the older group . This may be explained by their elaborations of why they were not comfortable in talking about their families with their Anglo-Australian friends. Most referred to their anxieties and experiences that revealing their `differences' would attract scorn and ridicule from their majority peers or at best lack of understanding. It is as if that part of them that was not readily seen as acceptable by the mainstream culture needed to stay hidden. It would be worth investigating further whether this process is stressful for these young people. The majority of those who elaborated on the bicultural competencies they were developing in negotiating between two cultural worlds stressed the comfort of being able to talk about ingroup things such as cultural activities, soccer scores, what was going in their parentsÕ homelands within their own ethnic community. What stood out was the respect these adolescents had for their parents and ethnic group, in spite of some complaints about the constraints imposed upon them by their traditions and the 'good' behaviour' expected form them at home ("parents more strict"). The greater freedom to be rude and 'muck around', which they perceived as permissible within mainstream culture, did not necessarily engender respect. What was apparent was their stereotypical and rather superficial view of Anglo-Australians and what is and is not acceptable behaviour. Their direct knowledge of mainstream Australian culture came mainly from their interaction with their Anglo-Australian peers within the school context, so that they had little exposure to the responsibilities and independence that is often the other side of the 'freedom' coin. This exploratory study opens up more questions than it provides answers. It has shown that the assumption of the Dual Identification Scale that Majority and Minority identifications are independent dimensions does not apply to these Australian-born adolescents from non-English-speaking backgrounds, who are attempting to straddle the world of the home and the world of mainstream society. Those who showed greater ease in socializing with majority peers were also the ones with a stronger ethnic group orientation. These results suggest that multicultural education policies in Australia, which have emphasized the importance of promoting tolerance for ethnic minority groups to maintain their cultural identity, have met with some success. Yet, both the quantitative and qualitative data in the study indicate that the more mature adolescents were aware that being part of and socializing with mainstream Australian society is not a road as open to them as they may have thought when they were younger. In this context it is worth noting that only 27% of the adolescents in the study described themselves as hyphenated Australians, while 61% identified themselves by sole reference to their ethnic group. Further substantive research is needed in order to identify what are the influences within their social milieu which make dual identification problematic for these young people. This is not merely a question of what the minority group can do to participate in the mainstream as bicultural individuals, which these adolescents are obviously attempting to do, but rather addressing the question to what extent and how the majority group facilitates or hinders this process. In spite of the rhetoric, we are still have the challenge of building a society in which `multiple identities and loyalties can organise and assert themselves with pride' (Lever-Tracy, 1992, p.395). References Aboud, F.E. (1988). Children and Prejudice. New York:Blackwell Aboud, F.E. (1992). Conflict and group relations. In C. Shantz & W.W.Hartup.(Eds). Conflict in child and adolescent development. Cambridge University Press. Bat-Chava, Y. & Deaux,K. (1990, June). Dual group identification, self-esteem and academic performance among minority students. Paper presented at the Second Annual Convention of the American Psychological Society, Dallas,TX. Comer, J.P.(1988). Educating poor minority children. Scientific American, 25 ,9, 42-48. Crocker, J.& Major, B. (1989). Social stigma and self-esteem: The self-protective properties of stigma. Psychological Review, 96, 608-630. Cross, W, E. (1987). A two factor theory of Black identity: Implications for the study of identity development in minority children. In J. Phinney & M.J.Rotheram (1987) ChildrenÕs Ethnic Socialization. Newbury Park, Ca.:Sage. Ethier, K. & Deaux, K. (1990). Hispanics in Ivy: Assessing identity and perceived threat. Sex Roles, 22 (7/8) , 427-440). Human Rights and Equal Opportunity Commission.(1991). Racist Violence: AReportt of national inquiry into racist violence in Australia. Canberra: Australian Government Publishing Service. Katz, P. (1981). Psychotherapy with native adolescents. Canadian Journal of Psychiatry, 26, 455-459. Lever-Tracy,C.(1992). The future of multiculturalism and citizenship. Proceedings of the Second National Outlook Conference, Sydney, 11-13 November, 1992. Bureau of Immigration Research. Office of Multicultural Affairs (OMA). (1989). Issues in Multicultural Australia, 1988. Social Science data Archives for the Office of Multicultural Affairs, Department of Prime Minister and Cabinet. Canberra: Australian National University. Phinney, J.S. (1989). Stages of ethnic identity in minority group adolescents. Journal of Early Adolescence, 9, 34-49. Phinney, J.S., & Allipura, L.L. (1990). Ethnic identity in college students from four ethnic groups. Journal of Adolescence, 13, 171-183. Phinney, J. & Rotheram, M. J. (1987). ChildrenÕs Ethnic Socialization. Newbury Park, Ca.: Sage. Rosenthal, D.& Hrynevich, C. (1985). Ethnicity and ethnic identity: A Comperative study of Greek-, Italian- and Anglo-Australian adolescents. International Journal of Psychology, 20, 723-742. Rosenthal, D. (1987). Ethnic identity development in adolescents. In J.Phinney and M.J. Rotheram (Eds.), ChildrenÕs Ethnic Socialization. Newbury Park, Ca.:Sage. Semaj, M.B. (1980). The development of racial evaluation and preference. A cognitive approach. Journal of Black Psychology, 6, 59-79 Spencer, M.B . & Markstrom-Adams, C. (1990). Identity processes among racial and ethnic minority children in America. Child Development, 6, 290-310. Steele, C.M. (1988). The Psychology of self-affirmation. Sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology, 21, (pp.261-302). New York: Academic Press. Tajfel, H. (1981). Human groups and social categories.: Studies in Social Psychology. Cambridge: Cambridge University Press. Triandis, H.C., Kurowski, L.L., Tecktiel, A., & Chan, D.K.S. (1993). Extracting the emics of diversity. International Journal of Intercultural Relations, 17, 217-234. Turner, J.C. (1978). Social comparison, similarity and ingroup favouritism. In H. Tajfel (Ed.), Differentiation between social groups. London: Academic Press. Turner, J.C. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Basil Blackwell. Vaugham, G. 1987). A social psychological model of ethnic identity development. In J. Phinney & M.J.Rotheram (Eds.), ChildrenÕs Ethnic Socialization. Newbury Park, Ca.:Sage. Weinberg, N . & Sterritt, M.(1986). Disability and Identity: A study of identity patterns in adolescents with hearing impairments. Rehabilitation Psychology, 31, 95-102. Footnotes 1. Though many people like the term African-American, not all African American agree and most of the earlier literature refers to Black- Americans (see Triandis et al., 1993). The term that is used in this study is that of the authors of specific research cited. 2. The term 'Australian' is used in this study when reporting the results, since it was invariably used by the participants in this study (as it is commonly used by most 'Australians') to refer to the majority group, Anglo-Australians. In other parts of the paper we refer to 'Anglo-Australians'. APPENDIX A ELABORATION of "disagree" ANSWERS TO MAJORITY SCALE ITEM 5 I feel comfortable talking to my Australian friends about my family. BOYS' SCHOOL: YEAR 8 (13-14 years) I'd get embarrassed, they might tease you.They'll think I am a wog. You can't talk Lebanese matters what is happening in Lebanon) with Australians. Because the way Lebanese parents treat their children is different, Australians might laugh. Australians won't understand. They might react in a funny way. None of their business. Australian would think my parents are different. If they know I am different religion they'd call me names. BOYS' SCHOOL: YEAR 10 (15-16 years) About war. The might think I am bad. They think my Dad was Vietcong. I feel close to other Chinese. Australians mightn't care. My personal private life. Our parents teach us different things. My parents more strict. They would tease me. Aussies. They don't understand our culture. GIRLS' SCHOOL : YEAR 8 (13-14 years) Australians may not like your culture, they might tease me. Don't talk about my family life to anyone. Don't understand our culture.They'd think it's weird GIRLS' SCHOOL: YEAR 10 (15-16 years) They are different-wouldn't understand and help me. My Australian friends have different values.They wouldn't understand. They may think my parents are more strict.They may criticize. Australians don't know how it is.Spanish parents are different. They are allowed to do much more than I do. They would feel sorry for me APPENDIX B ELABORATION OF "Agree" ANSWERS TO MINORITY SCALE Item 6 I often feel like someone who has to change the way they behave and talk depending on the ethnic background of whom I'm with BOYS' SCHOOL : YEAR 8 (13-14 years) With Lebanese people you can joke around; you know their parents.With Australians you can't. Different religion. I'm Orthodox. Some Australians have different words. With Lebanese you have to talk serious, with Australians you talk freely. With Yugoslavs if I don't behave they start gossiping, with Australians it doesn't matter. Portuguese play different games,different T.V. shows. With Portuguese I speak freely. With Australians not so. With Portuguese you behave a polite way, with Australians more rude. Hong Kong people dress differently, more neat, not in jeans.Use Chinese slang. With Greeks I talk half Greek, half Australian. With Australians I need to speak Australian. More comfortable with Lebanese, more private. Behave more mature with Indian people.Talk about cricket.With Australians you talk about subjects and teachers. Muck around more. BOYS' SCHOOL: YEAR 10 (15-16 years) With Philipinos I talk half Tagalog,half English. With Chinese friends I speak Chinese and English. With Australians only English. With Indians I talk about Indian music,cultural staff. With Australians sport. They wouldn't understand me. Here I learn about Jesus, in Hindi you have many gods. With Greeks I talk half Greek, half Australian. With Australian I need to speak Australian. Talk about different things. To Portuguese soccer, to Australians cricket. With Mauritian friends I care, with Australians I care less. With Spanish older people behave like friends. Some Australians I know only for business. With Australian friends lazy. Lebanese more enthusiastic. With Australian friends, you can do anything.With Vietnamese you obey. GIRLS' SCHOOL: Year 8 (13-14 years) More polite with Greeks, you don't swear. More rude with Australians. With Tongan people you speak Tongan, cautious; with Australians more outgoing. Closer to Portuguese friends. With Portuguese talk about own culture, different things. With my non-English -speaking friends I talk about soccer; Australians not interested. Tongans have different ways-concerts with Tongan dances. I am more myself with Tongans I talk different things with Aboriginals.With Whites you are more quiet. With Aboriginal people you know each other; talk Aboriginal lingo.We handle each other. After fights we make up.With Australians more angry. We are brought up differently.More comfortable with each other. With Italians I have to behave good; with Australians more free. GIRLS' SCHOOL : YEAR 10 (15-16 years) The way you speak with Greek friends ,you are more comfortable. They are more similar. More open with Tongans,more respect. With Australians I don't respect as much. I don't want my Australian friends to see how I look (all properly dressed for Church). With Australians I am more myself. With Lebanese I feel more restricted. With Italians I am more open, a bit wild. With Australians more quiet. With Australians you speak different things. With Columbians I speak about Columbian things-soccer games,going to Columbia.I don't go for Australian boys,not so educated. With Spanish people you can use Spanish expressions, talk about Spain, things in common. With Australians you can't do it. I am more quiet with Australians. With Italians more comfortable. With Greek friends you joke the Greek way. They'll understand more.