Intergenerational transmission of ethnic identity and life satisfaction of Roma minority adolescents and their parents

Journal of Adolescence 45 (2015) 296e306

Contents lists available at ScienceDirect

Journal of Adolescence journal homepage: www.elsevier.com/locate/jado

Intergenerational transmission of ethnic identity and life satisfaction of Roma minority adolescents and their parents Radosveta Dimitrova*, Laura Ferrer-Wreder, Kari Trost Stockholm University, Sweden

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 11 November 2015

This study investigates intergeneration transmission of ethnic identity as a resource for life satisfaction of Roma adolescents and their parents. Historically, Roma represent the largest ethnic minority in Europe. They have been exposed to severe discrimination, social exclusion, and poverty. Therefore, identifying resources for their life satisfaction is theoretically and practically important. The present study included 1093 participants, of which there were 171 Roma adolescents (age: M ¼ 14.96 years, SD ¼ 1.85), 155 mothers (age: M ¼ 36.16 years, SD ¼ 5.77) and 123 fathers (age: M ¼ 39.68 years, SD ¼ 6.06). Further, a comparison group of 248 mainstream adolescents with their mothers (n ¼ 221) and fathers (n ¼ 175) was also included in the study. Adolescents and their parents provided data on ethnic identity (MEIM; Phinney, 1992) and life satisfaction (SWLS; Diener, Emmons, Larsen, & Griffin, 1985). Results indicated that Roma youth were lower on endorsement of ethnic identity and average on life satisfaction compared to their mainstream peers. A structural equation model showed that ethnic identity was a positive predictor of life satisfaction for both adolescents and their Roma parents. Furthermore, parents' ethnic identity was a predictor of adolescent life satisfaction. We concluded that for Roma youth and their parents, ethnic identity represents a salient source for life satisfaction and an intergenerational continuity of identity and life satisfaction exists. © 2015 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

Keywords: Ethnic identity MEIM Roma adolescents Life satisfaction Bulgaria

Intergenerational transmission is only one dimension of a wide variety and range of activities and exchanges that can occur between different generations. Intergenerational transmission is defined as the process through which earlier generations psychologically influence attitudes and behaviors of the next generation, and is usually examined in two or more generations of participants (van Ijzendoorn, 1992; Serbin & Karp, 2003). This study investigates intergenerational transmission of ethnic identity among a sample of Roma adolescents and their parents, and its relation to their life satisfaction compared to a sample of Bulgarian mainstream youth with their mothers and fathers. Roma are the largest indigenous minority group in Europe, characterized by marked social marginalization and in most cases unique continuous life challenges (European Union Agency for Fundamental Rights, 2010; Gerganov, Varbanova, & Kyuchukov, 2005). Although multiple generational ties exist in the Roma population, intergenerational transmission has not been studied in the Roma community to the extent one may expect. The main focus on this study is on Roma families with adolescent children living in Bulgaria as this is a marginalized social group. Yet, scholars have acknowledged the importance of understanding intersections between

* Corresponding author. E-mail addresses: [email protected], [email protected] (R. Dimitrova). http://dx.doi.org/10.1016/j.adolescence.2015.10.014 0140-1971/© 2015 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

297

minority and mainstream or dominant groups as a way to work against the reification of mainstream groups as the lone representation of normativity (Yuval-Davis, 2006). Other studies with ethnic minority samples in several instances also utilized mainstream groups which informed our approach to the present study (e.g., Crocetti, Fermani, Pojaghi, & Meeus, € r, Fleischmann, & Phalet, 2011; Schwartz, Zamboanga, Weisskirch, & Rodriguez, 2009). This study acknowl2011; Güngo edges this argumentation and prior empirical precedent by taking into account a group of mainstream youth with Bulgarian ethnic heritage in comparison to their peers with Roma ethnic background. In Bulgaria 89% of Roma students have a primary school education, 10% have a secondary education as a final education level in adulthood (Ringold, 2000). Also, in Bulgaria official policies on improvement of their conditions are scarce or suffer from a lack of follow through (Amnesty International, 2007; Civil Society Monitoring Report Bulgaria, 2013). Marginalization of the Roma population remains a concern. In order to promote inclusion and decrease economic strain of marginalized groups, the World Bank and the Open Society Institute have taken steps to promote education, housing, employment and physical health programs for Roma (Ringold, Orenstein, & Wilkens, 2005). Although these programs may be beneficial for economic and social welfare inclusion, there remain significant gaps in understanding individual and intergenerational resources that may support life satisfaction among Roma, from the standpoint of normative developmental processes such as identity development. This study is novel in that it promotes understanding of Roma as an important ethnic minority in Europe with unique socio-cultural features and a long history of marginalization. We address relevant research in this group by adding important methodological and conceptual contributions to the literature. We know little if anything about Roma compared to other ethnic minorities and the sparsely available research has rarely investigated positive aspects in Bulgarian Roma families, such as life satisfaction as a part of positive development. We also add important findings to what is known about the joint role mothers' and fathers' ethnic identity may play in their adolescent child's ethnic identity and life satisfaction. This is a little researched area for Roma families in Bulgaria and we also have the strength of having a comparison group of families who are in the Bulgarian mainstream society. Therefore, we investigated intergenerational transmission of ethnic identity, and attempted to document the ways in which ethnic identity may function as a resource for well-being of Roma adolescents and their parents. Ethnic identity and life satisfaction Ethnic identity concerns the process of identification with an ethnic or cultural group accompanied by a sense of belonging, relatedness, and commitment to that group (Phinney, 1989; Phinney & Ong, 2007). Ethnic identity development is not static, but part of an ongoing process that involves combinations of the presence and absence of exploration and commitment (i.e., identity diffusion to achievement; Erikson, 1968; Marcia, 1980). An important distinction in the use of the term ethnic identity is warranted. The term is often used in the sense of an ethnic minority/immigrant identity based on cultural heritage. However, in this study, we referred to ethnic identity in terms of identification with the dominant culture (Bulgaria) for mainstream youth and identification with the ethnic culture (Roma) in the Roma minority group. Such operational definitions and measurement of ethnic identity have been used in the literature (e.g., Barany, 2001; Prieto-Flores, 2009). In large scale of multi-ethnic youth, findings showed that achieving a stable, positive identity can be challenging (Crocetti et al., 2011). It has also consistently been reported that an optimally developed ethnic identity, which is characterized by strong sense of attachment or belonging to one's ethnic group, identifying and engaging with ethnic practices and recognizing positive virtues about own ethnic group, is positively related to life satisfaction and psychosocial adjustment in various ethnic minority groups (e.g., Rivas-Drake et al., 2014; Schwartz et al., 2009; Smith & Silva, 2011). The strong relationship between ethnic identity and life satisfaction has also been documented in Roma youth in Eastern Europe (Dimitrova, Chasiotis, Bender, & van de Vijver, 2014). Relatedly, studies have shown that older adolescents had a clearer and more firmly defined sense of ethnic identity relative to younger adolescents (e.g., French, Seidman, Allen, & Aber, 2006). Moreover, gender differences may exist in that identifying with one's ethnic group is more important for adolescent girls than boys (Yip & Fuligni, 2002). With regards to Roma youth, little is known about ethnic identity of Roma. Of the existing scarce literature, the findings are inconclusive and mixed. Most are based on cross-sectional studies using specifically developed quantitative instruments to measure ethnic identity. For example, a tendency among Roma to identify with the national mainstream culture has been observed in Central and Eastern Europe (e.g., Barany, 2001; Prieto-Flores, 2009), as well as among Roma living in Western Europe (e.g., Marushiakova & Popov, 1993). It has also been found that Roma youth and their mothers have reported, on average, moderate levels of endorsement of both Bulgarian national and Roma ethnic identity (Dimitrova et al., 2014). Based on such findings, no recommendations can be given to Roma youth in Bulgaria since more documentation of how these youth are identifying is needed. A goal of the present study is to pursue more information about youth identity development within the Roma population. The purpose of this study was to explore the relations between optimally developed ethnic identity (i.e., strong sense of attachment, belonging and engaging with ethnic practices of the Roma group) and life satisfaction for Roma youth and their parents in comparison to a group of Bulgarian mainstream youth with their mothers and fathers (i.e., their ethnic identity as a sense of attachment and belonging to the Bulgarian group). Although complex and sometimes disparate findings exist for Roma families, prior work on ethnic identity has shown it often to be strength related to well-being among ethnic minority populations (e.g., Smith & Silva, 2011). We acknowledge the diversity in prior work by addressing the question of whether

298

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

Roma ethnic identity is important to life satisfaction in Roma families. Because there is not an extensive amount of prior research on this matter, and the available evidence shows mixed findings, we aimed at providing a more descriptive contribution on ethnic heritage identity and its relations to well-being among Roma and thereby extend knowledge regarding what is known about this ethnic group. Intergenerational transmission of ethnic identity Intergenerational transmission is one of the major mechanisms of cultural continuity and a basis for culture mainte€r et al., 2011; Nauck, 2001; Scho € npflug, 2001). nance or adoption among immigrant and ethnic minority groups (Güngo Intergenerational transmission has been found to be salient in ethnic minority groups and to contribute to a strong collective ~ a-Taylor, Zeiders, & Updegraff, 2013). For example, intergeneracultural continuity (Farver, Narang, & Bhadha, 2002; Uman tional transmission processes have been shown to have importance for intercultural contact and ethnic identity of second generation adolescents in five different groups of migrant families: Italian, Greek, and Turkish work migrants, German repatriates from Russia, and Russian Jewish immigrants in Israel (Nauck, 2001). Results from 400 parentechild dyads in each group confirmed that intergenerational transmission has a significant relation to acculturation processes in migrant families; especially strong was the transmission of ethnic identification between parents and children. Additional work has provided evidence on intergeneration transmission using cross-sectional design with mixed-methods qualitative interview and quantitative survey data. Research conducted with Asian Indian mothers and fathers shows that ethnic identity maintenance was enhanced by engaging in cultural celebrations, keeping strong family ties, and a rejection of Western values (Inman, Howard, Beaumont, & Walker, 2007). In a cross-cultural comparative study of Turkish and Moroccan €npflug (2001) found that across ethnic groups and parentechild dyads in Germany and the Netherlands, Phalet and Scho national contexts, parents with strong collectivistic values of family relatedness tended to have more collectivistic children. Verkuyten, Thijs, and Stevens (2012) examined the relation between religious, ethnic, and national identity among MoroccanDutch Muslim adolescents and their parents in the Netherlands. Compared to their parents, adolescents showed stronger national identification and weaker religious and ethnic group identification. Furthermore, there was evidence for strong intergenerational transmission in parent's religious group identification and their children's religious, ethnic, and national identifications. Finally, a recent cross-sectional study of 194 Roma adolescents and their mothers investigated the relation between ethnic, familial, and religious identities as salient components of collective identity and their relation to life satisfaction (Dimitrova et al., 2014). A structural equation model (SEM) showed evidence to support the intergenerational continuity of collective identity in that mothers' and their adolescent children's collective identity was positively related to one another. Further, results indicated that collective identity (which included ethnic identity in terms of Roma and Bulgarian identity) was positively associated with greater life satisfaction in both adolescents and mothers. Yet, a limitation of this study is that it included only mother-adolescent dyads from only one ethnic population. With a new sample, however, we do build upon prior work by studying ethnic identity as a single construct rather than as one part of collective identity and as a resource for life satisfaction. The present study, also addresses gaps in the research by testing intergenerational continuities of ethnic identity and life satisfaction in a large sample of youth, mothers, and fathers with Bulgarian and Roma ethnic background. This is particularly important as understanding ethnic identity experiences of both parents and their importance to children's outcomes may facilitate actions to improve well-being for the family as a whole and for Roma at individual, family and societal levels. Further, the role of fathers and the ways in which both paternal figures may work together when raising their children is a relevant family process in ethnic minority context (e.g., Chuang & Tamis-LeMonda, 2013). Family systems theory (e.g., Grusec & Hastings, 2014) and empirical research which indicates the importance of fathers to child adjustment and well-being (Feinberg, 2003) also provides support for the value of including fathers in the present study. Further, much of the available research relevant to the present study has focused only on mothers or families in the United States, whereas much less is known about both mothers and fathers in ethnically diverse samples from Europe and whether parental (including fathers') ethnic identity has similar consequences for child well-being. In summary, previous literature suggests that a) Roma youth may show low endorsement of ethnic identity (identification with their own heritage Roma culture) compared to their mainstream peers' ethnic identity (identification with their own mainstream culture); b) age and gender related differences may appear in the ethnic identity of youth from minority and mainstream ethnic groups; c) ethnic identity is likely to be a salient factor for the life satisfaction of ethnic minority (Roma) groups; d)the process of cultural transmission from one generation to the next is expected to be a significant factor for ethnic identity of Roma. Context, hypotheses, and research questions Study participants were youth of Roma minority and Bulgarian mainstream origin in Bulgaria, a post-communist country in Eastern Europe where many Roma suffer poor life satisfaction, segregation and lack of official policies for improvement of their conditions (Amnesty International, 2007). In fact, the latest monitoring report in Bulgaria concluded that despite the large number of strategic documents and programs that have been adopted, the political will is lacking to substantively improve the situation of Roma (Civil Society Monitoring Report Bulgaria, 2013). Roma estimates in the country range between 325,343 and 800,000 people out of a national population of nearly 7 million (National Statistics Institute, 2011).

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

299

We investigated the intergenerational transmission of ethnic identity and its importance to well-being among Bulgarian families, those in the Roma minority and those in the national mainstream by taking into account three theoretical perspectives. First, family systems theory emphasizes the role of family factors, such as parentechild relationships in influencing developmental outcomes (Grusec & Hastings, 2014). As the primary unit of analysis is the family, development is conceptualized in terms of interactions between people, rather than any one individual's behavior. Second, the actorepartner interdependence model (APIM, Cook & Kenny, 2005) integrates a conceptual view of interdependence in dyadic relationships. Central components are the actor effects (how much individuals' current behavior is predicted by their own past behavior) and the partner effects (how much one person is influenced by another person). In terms of parentechild relationships, the model portrays interaction effects among actors and partners as partial determinants of child outcomes. Third, in line with interactionist perspective, child outcomes result through the crossing of parent and child characteristics (Sameroff & Chandler, 1975). The interactionist model also assumes that developmental outcomes depend on the extent to which parent and child characteristics match or fit together (Lerner, 1993). We build on these theoretical perspectives (e.g., family systems theory, the actorepartner interdependence and interactionist models) in postulating that parents and the family are key agents of socialization for children and adolescents and that the intergenerational continuity of ethnic identity and life satisfaction is likely to exist in many families. We investigated ethnic identity and life satisfaction in Roma adolescents and their parents by advancing two hypotheses and two research questions. First, we expected group differences in average levels of ethnic identity and life satisfaction such that Roma adolescents compared to their mainstream peers would have a) weaker Roma ethnic identification relative to a stronger mainstream ethnic identification in mainstream Bulgarian youth (Hypothesis 1a) and b) lower levels of life satisfaction (Hypothesis 1b). Second, we explored age and gender related differences in line with prior work (French et al., 2006; Phinney, 1989), suggesting that older compared to younger adolescents (Research Question 1), and girls compared to boys (Research Question 2) would report more pronounced ethnic identity. Lastly, in accordance with previous findings (Dimitrova et al., 2014) and the theoretical frameworks guiding this study (Cook & Kenny, 2005; Grusec & Hastings, 2014; Sameroff & Chandler, 1975), it was posited that parents' ethnic identifications would have positive associations with their children's identifications. It was also hypothesized that ethnic identity would have strong and consistent relations with life satisfaction, for parents and their children, and that parents' and children's life satisfaction would be positively related to one another (Hypothesis 2, see Fig. 1). In particular, we tested a model of associations between ethnic identity and life satisfaction among Roma and mainstream parents and youth. The SEM model used to test hypothesis 2 was based on similar path models testing intergenerational transmission in ethnic minority groups (Dimitrova et al., 2014; Verkuyten et al., 2012). However, these previous models used only mother reports and referred to only one ethnic group, a gap this study addressed by testing a multi-group path model of youth, mothers and fathers with Bulgarian and Roma ethnic backgrounds. Overall, results of these prior studies (Dimitrova et al., 2014; Verkuyten et al., 2012) confirmed that there was a significant intergenerational continuity of ethnic identity and well-being between mothers and their children in diverse ethnic minority contexts like Moroccan-Dutch Muslim youth and Roma-Bulgarian youth. Therefore, we expect the same pattern of findings to emerge in our samples of Roma parents and their adolescent offspring in Bulgaria. Method Participants and procedure In total, there were 1093 participants, of which 419 were adolescents (age: M ¼ 15.16 years, SD ¼ 1.90). In the Roma adolescent group, there were 171 adolescents (M ¼ 14.96 years, SD ¼ 1.85). In the Roma maternal group, there were 155 Ethnic identity mother

Life satisfaction mother Life satisfaction adolescent

Ethnic identity adolescent

Life satisfaction father

Ethnic identity father

Fig. 1. Theoretical intergenerational transmission model of ethnic identity and life satisfaction.

300

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

mothers with a mean age of 36.16 (SD ¼ 5.77). In the Roma paternal group, there were 123 fathers with a mean age of 39.68 (SD ¼ 6.06). In addition to the Roma participants, there were 248 Bulgarian mainstream youth (M ¼ 15.30 years, SD ¼ 1.93) and their parents as reference groups (see Table 1). We included a comparison group of Bulgarian mainstreamers for several reasons such as this group's relevance to the study questions (i.e., expected ethnic group differences in ethnic identity and life satisfaction). Criterion sampling was used (Palys, 2008) by including youth who were aged 14 to 18 and reported Roma and Bulgarian background. From all participating schools students that met the criterion were invited to fill out the surveys. Usually students are grouped in classes based on their age so all selected classes had students that fit our criterion. Participants were invited to fill out the survey on a voluntary basis in the classroom and all of them agreed to do so. With this criterion in mind, all youth were sampled in schools as the most appropriate sampling strategy in the local context and in line with the nature of our research objectives. For the purposes of this cross-sectional study design, approximately 400 families were approached via four public schools to acquire their participation. The response rate was very high among mothers (90%), fathers (70%) and youth (98%). Missing data for parents across measures ranged from 11% (mothers) and 27% (fathers) due to work load and impossibility to fill in the survey. Data missing from adolescents were due to the participant not being at school at the time of administration or delayed delivery of the questionnaires to the research team. Missing data at student level were 2%. The adolescent groups in this study differed in terms of family socioeconomic status (SES) as measured by occupation and education of both parents, with Roma youth having a lower SES than their mainstream counterparts, c2(2, N ¼ 278) ¼ 137.18, p < .001. The adolescent participant group did not differ in terms of age or gender. For further analyses, we dichotomized age in middle (11e14 years old) and late (15e18 years old) adolescent groups (see Verkuyten et al., 2012). Participants for this study were recruited through four public high schools in Southern Bulgaria with a high concentration of Roma inhabitants. Prior to data collection, local school authorities were informed about the purpose and method of the study to acquire their consent and participation. The study was part of a larger project on Roma ethnic minority groups in Bulgaria and was approved by the Bulgarian National Agency for Child Protection and the Swedish Regional Ethics Committee. All field workers received training in the administration of the survey and a weekly supervision was conducted. Materials were distributed to all parents containing information about the study and parental consent forms. Active consent was used in this study. Upon receiving the signed parental consent forms, youth filled in the survey in their classroom during the school day. Parents filled in questionnaires in their homes. Both parents were assisted with the questionnaire if requested. All participants (adolescents and both parents) were offered a small gift for their participation. On average, the parent and youth questionnaires took 45 min to complete. Measures All participants completed a short questionnaire assessing their ethnicity, gender, and age. Data on family SES was reported by participating youth, mothers and fathers. SES was computed by creating a composite score of all participants' scores on occupation (unskilled, semi-professional, and professional job) and education (primary, secondary, and university degree) levels of both parents. This score was coded in three levels by adopting SES criteria as score resulting in three levels of low, middle, and high SES (Oakes & Ross, 2003). The same measures, with slight wording changes, were applied for assessing ethnic identity and life satisfaction among adolescents and their parents. All measures were translated from English into Bulgarian by four bilingual speakers while adhering to the standard guidelines to ensure linguistic equivalence (Van de Vijver & Leung, 1997). Also, two separate focus groups of teachers and students were involved in assessing the appropriateness of the translated measures. The questionnaires were presented only in Bulgarian, because all Roma pupils acquire literacy skills exclusively in Bulgarian. All measures showed very good internal consistency across cultural groups and respondents (see Table 1). Table 1 Sample characteristics for Roma and mainstream adolescents and parents. Roma ethnic minority

Age Mean (SD) Gender, % Female Male SES, % Low Middle High Ethnic identity, M (SD) Life satisfaction, M (SD)

Bulgarian mainstream

Youth n ¼ 171

Mothers n ¼ 155

Fathers n ¼ 123

Youth n ¼ 248

Mothers n ¼ 221

Fathers n ¼ 175

14.96 (1.85)

36.16 (5.77)

39.68 (6.06)

15.30 (1.93)

40.03 (7.51)

43.54 (6.77)

48 52

e e

e e

53 47

e e

e e

3.42 (.73) a ¼ .89 3.68 (1.12) a ¼ .80

46 49 5 3.73 (.85) a ¼ .91 3.45 (1.29) a ¼ .85

3.70 (.62) a ¼ .87 4.39 (1.02) a ¼ .79

3 28 69 3.77 (.76) a ¼ .91 4.33 (1.17) a ¼ .88

3.61 (.77) a ¼ .92 4.31 (1.10) a ¼ .86

3.41 (.83) a ¼ .91 3.75 (1.34) a ¼ .88

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

301

Ethnic identity Ethnic identity was measured by the Multi-Ethnic Identity Measure (MEIM; Phinney, 1992). The measure consists of 12 items rated on a 4-point Likert scale ranging from 1 (Strongly disagree) to 4 (Strongly agree). Sample items include “I am happy that I am a member of the group I belong to”, “I have spent time trying to find out more about my ethnic group, such as its history, traditions, and customs”, and “I participate in cultural practices of my own group, such as special food, music, or customs.” A mean score ranging from 1 to 4 is obtained for an ethnic identity score. Instructions were given to the Roma participants in this study to fill in the measure with regards to the identification with their own Roma heritage culture. The factor structure of the MEIM has been shown to comprise two factors: a developmental-cognitive component (ethnic identity search) and an affective component (affirmation, belonging and commitment) (Phinney, 1992). For the purposes of this study, we used a one factor approach indexing global ethnic identity based on previously adopted similar approach (Bracey, Bamaca, & Umana-Taylor, 2004). The scale is a widely used measure of ethnic identity across variety of contexts and cultures and has consistently shown good reliability, typically with alphas above .80 across a wide range of ethnic groups and ages (Phinney, 1992). Recently, the measurement invariance of the MEIM has been tested in Bulgarian, Dutch, and Greek adolescents, showing good psychometric properties for the groups under investigation (Mastrotheodoros, Dimitrova, Motti-Stefanidi, Abubakar, & van de Schoot, 2012). Life satisfaction The Satisfaction with Life Scale (SWLS) was applied to measure global life satisfaction (Diener, Emmons, Larsen, & Griffin, 1985). The scale consists of five items rated on a 7-point scale (1 ¼ strongly disagree to 7 ¼ strongly agree). Sample items include “In most ways, my life is close to my ideal”, “I am satisfied with life”, and “If I could live my life over, I would change almost nothing”. The SWLS is a widely used and reliable measure of life satisfaction and there has been tremendous amount of research on its validity across different cultures (e.g., WesterneEastern European, American, and African). Further support for the validity of the scale is provided by strong evidence of its discriminant validity and high internal and testeretest reliability (Diener et al., 1985). A recent study, confirmed measurement invariance of the SWLS across three immigrant groups, namely, immigrants from the Former Soviet Union (FSU), Turkish-Bulgarians, and Turkish-Germans (Ponizovsky, Dimitrova, Schachner, & van de Schoot, 2012). Results Preliminary analyses explored outlier, normality and missing data patterns across samples. Participants' leverage values were used to identify univariate outliers, and these analyses showed that there were fourteen outlier cases on the scale scores across samples. The substantive analyses were conducted with and without these outlier cases, and showed overall consistent results. Thus, the analyses described here include the outlier cases. Kurtosis and skewness values at the scale level for all informants did not exceed an absolute value of two (Raykov & Marcoulides, 2008). List-wise deletion at scale level indicated 2% (youth), 11% (mothers) and 27% (fathers) missing data. Missing values analysis (i.e., Little's MCAR test) showed nonsignificant results, suggesting that these data were missing completely at random. Further, we tested hypotheses of the study in two main steps. First, we investigated the possibility of group differences in ethnic identity and life satisfaction among youth by performing a MANCOVA with group (2 levels: Roma and mainstream), age and gender as independent variables and ethnic identity and life satisfaction as dependent variables, and SES as covariate. Second, we tested associations between mothers', fathers' and youths' ethnic identity and life satisfaction by employing a path model by means of Structural Equation Modeling in AMOS (Arbuckle, 2009). We tested our path model in two steps. First, we estimated two separate models for each group by including all parent and youth variables (Fig. 1). Second, we tested this model simultaneously across Roma and mainstream groups. Fit indices adopted to interpret the model fit were the c2 test, the root mean square error of approximation (RMSEA; recommended value  .08) and the comparative fit index (CFI; recommended value  .90) (Browne & Cudeck, 1993; Marsh, Hau, & Grayson, 2005). Group, age and gender differences in identity and life satisfaction Table 1 shows the mean scores of ethnic identity and life satisfaction across groups. Hypotheses 1a and 1b concerned ethnic group, age and gender related mean level differences in ethnic identity and life satisfaction. The multivariate tests for group Wilks' lambda ¼ .96, F(1, 412) ¼ 8.15, p < .001, h2 ¼ .039 and age were significant, Wilks' lambda ¼ .97, F(1, 412) ¼ 5.65, p < .001, h2 ¼ .027. In line with study predictions, namely hypotheses 1a and 1b, Roma adolescents, compared to their mainstream peers, showed a significantly weaker ethnic identity (F(1, 412) ¼ 7.12, p < .001, h2 ¼ .017) and lower levels of life satisfaction, F(1, 412) ¼ 15.61, p < .001, h2 ¼ .034. Concerning age and gender related differences, middle adolescents reported greater life satisfaction than late adolescents, F (1, 412) ¼ 11.29, p < .001, h2 ¼ .027. No significant gender differences emerged for adolescents with different ethnic backgrounds regarding their ethnic identity or life satisfaction. Thus, research questions 1 and 2 were not supported. However, there was an unexpected finding in that there was a significant gender by ethnic group interaction for life satisfaction showing that Roma adolescent girls reported significantly lower life satisfaction relative to Roma adolescent boys and mainstream adolescents, F(1, 412) ¼ 4.28, p < .001, h2 ¼ .011.

302

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

Table 2 Fit indices of ethnic identity and life satisfaction path model for of Roma and Bulgarian youth and parents. Model

c2 (df)

Configural invariance Structural weights Structural covariances Structural residuals

20.29 30.08* 36.32* 43.92*

(12) (19) (22) (27)

Dc2 (Ddf)

RMSEA

CFI

e 9.79 (7) 6.24 (3) 7.6 (5)

.041 .037 .039 .039

.988 .984 .980 .976

Note: most restrictive model with a good fit is printed in italics. D ¼ change in parameter; *p < .05.

Links between parental and youth ethnic identity and life satisfaction In a further step, we tested the second study hypothesis concerning the intergenerational transmission of ethnic identity and life satisfaction. We examined the hypothesized relationships between mothers' and fathers' ethnic identity and adolescents' ethnic identity and life satisfaction via SEM. We also considered associations between variables of ethnic identity and life satisfaction among parents. First, we conducted two separate models for each ethnic group. The two models showed poor fit for samples with a Roma (c2(4, N ¼ 171) ¼ 38.46, p < .001, RMSEA ¼ .225 and CFI ¼ .907) and a mainstream background, c2(4, N ¼ 248) ¼ 21.83, p < .001, RMSEA ¼ .134 and CFI ¼ .947. Therefore, we proceeded with the second step in testing our model simultaneously across groups. This multi-group path model investigated associations between ethnic identity and life satisfaction (SWLS) scores simultaneously among Roma and mainstream families. Direct relations between ethnic identity and life satisfaction of mothers and fathers and those of their children were allowed. The fit of this model was adequate, c2(27, N ¼ 419) ¼ 43.93, p < .021, RMSEA ¼ .039 and CFI ¼ .976. As can be seen from Table 2, all models showed good fit and we therefore opted for the most restrictive model represented by the structural residuals. The standardized path coefficients for the structural residuals model are reported in Fig. 2. There were significant direct paths from ethnic identity of mothers and fathers to the same identity in youth; these paths refer to direct-specific influences of parental ethnic identity on child ethnic identity. For both Roma and mainstream groups, ethnic identity was positively linked to life satisfaction in adolescents, mothers and fathers (Fig. 2). Yet, the link between ethnic identity and life satisfaction was much stronger among mothers than fathers. Moreover, in the group of mothers there was a direct, positive association between mothers' life satisfaction and their children's life satisfaction. This link was not significant in the fathers' group, suggesting that fathers' life satisfaction was unrelated to adolescents' life satisfaction. It can be concluded that in both generations, ethnic identity was positively linked to life satisfaction and that Roma and mainstream adolescents and parents with a stronger ethnic identity tended to report more life satisfaction. An important feature of these results involves intergenerational continuity across generations and ethnic groups. We found that mothers' and fathers' ethnic identity showed a strong, positive association with adolescent ethnic identity and the same holds true for life satisfaction. In fact, results showed a positive association between mothers' and fathers' ethnic identity and adolescent life satisfaction. It can be concluded that parental ethnic identity and life satisfaction are important for ethnic identity and life satisfaction of adolescents. Adolescent life satisfaction was indeed predicted by mothers' rather than by fathers' life satisfaction.

Ethnic identity mother

.18***

.23*** Life satisfaction mother .61***

.28***

.72***

Life satisfaction adolescent

.30***

Ethnic identity adolescent

.04 .13*** Ethnic identity father

Life satisfaction father

.22***

Fig. 2. Path model of ethnic identity and life satisfaction of adolescents and parents across groups. Note. The parameters represent standardized coefficients for the structural residuals model. ***p < .001.

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

303

Discussion In a multicultural world, the formation of ethnic identity is clearly an area of pressing concern, particularly in light of efforts to better support the likelihood for equal opportunities throughout life amongst diverse youth (e.g., Ferrer-Wreder, Montgomery, Lorente, & Habibi, 2014; Motti-Stefanidi, Berry, Chryssochoou, Sam, & Phinney, 2012). An optimal ethnic identity is the product of effort and a supportive ecology e individuals engage in identity work (e.g., talking to others, considering and trying out different life options), and can be supported in that work by important others (also called identity agents see Schachter & Ventura, 2008), in order to determine what their ethnic group affiliation(s) mean to them, in their own lives e and coming away from this process with a personally meaningful sense of belonging to a particular ethnic or other culture-based identity. Importantly, aspects of culture can change what is normative in terms of the quality and quantity of identity exploration and commitment in different cultures (Phinney & Baldelomar, 2011), and this is why is it is critical to document normative identity development processes in culturally diverse families. In this study, hypotheses were first posed in terms of exploring potential group differences on main study constructs. It was predicted, and study results from the tested MANCOVA supported the expectation that Roma adolescents would be lower than their mainstream adolescent counterparts in ethnic identity (i.e., hypothesis 1a). The ethnic identity in question for the Roma youth would be a Roma identity and for the mainstream adolescents it was a Bulgarian identity. Few prior studies provide a specific precedent for this result (see Dimitrova, Chasiotis, Bender, & van de Vijver, 2013 for a related study but this time in terms of a mainstream Bulgarian identity for both Roma and mainstream Bulgarian adolescents), and thus this finding contributes to an emerging evidence base on this question. Although research has generally found weaker or inconsistent links between positive characteristics like self-esteem and life satisfaction and minority youths, the present study clearly illustrates positive correlates to well-being for Roma youth. In line with Phinney's earlier work (1992), the present study did find positive relationship with ethnic identity and life satisfaction within the Roma community. Little research exists on Roma youth and their parents regarding positive youth development. This study illustrates that not taking a positive mainstream orientation can be linked with life satisfaction. However, Roma youths ethnic identity was a predictor of life satisfaction which conflicts with other findings linking ethnic minority adjustment with adaptation to mainstream youth. Marginalization theory (Berry, 1997) can provide an explanatory framework for the group differences on ethnic identity. It can be argued that this result is consistent with aspects of marginalization theory, in that a socially contested identity, such as a Roma identity in a Bulgarian context, may be associated with less development, as evidenced by low to moderate levels of ethnic identity development, relative to a Bulgarian identity for mainstream adolescents. This connection to marginalization theory is however, tentative in that this study is cross-sectional and captures only a glimpse of identity development at one point in time, and what is normative in terms of ethnic identity work for Roma adolescents living in Bulgaria is not yet well established through longitudinal studies, and thus an accurate sense of what might be low, moderate, or high levels of ethnic identity work would require additional research. As alluded to, the other aspect of marginalization theory important to consider would be, in this case, Roma adolescents' endorsement of a mainstream Bulgarian identity, which was not measured in the present study, and would provide a more complete analysis picture also for study hypothesis two which was tested in the SEM model. Also, in connection to the broader notion of understanding marginalization among adolescents with a Roma background, the present findings illustrate that many Roma youth are not thriving in terms of life satisfaction. For instance, it was hypothesized and the study results from the tested MANCOVA confirmed that the average life satisfaction of Roma adolescents in this sample was lower relative to mainstream adolescents (i.e., study hypothesis 1b). Lower life satisfaction among Roma relative to mainstream Bulgarian adolescents is consistent with a prior cross-sectional study of adolescents (e.g., Dimitrova et al., 2013). Other research questions related to age and gender-related differences in ethnic identity indicated no such differences as was evident. Thus, this study makes an important contribution to the research literature by examining potential age and gender differences in Roma ethnic identity, in particular, and finding evidence for similarities rather than differences. Future research should be conducted in order to explain the potential of age-related differences when using longitudinal design across age cohorts. One unexpected finding was found between Roma girls and the other adolescents in the present study. Roma girls fared worst on life satisfaction when compared to all other adolescents in the study including mainstream girls, mainstream boys and Roma boys. This finding for girls highlights the importance of studying intersectionality in the study of social identity, including ethnic identity (Kiang, 2014; Phinney & Baldelomar, 2011). Future studies with this population should be mindful to examine potential subgroup perspectives. This finding also provides empirical support for studying and understanding the diversity within Roma youth. It is also important to mention that the lack of research on Roma youth also extends onto intervention implementation. See Ferrer-Wreder et al. (2014) for a review of the efficacy of several ethnic identity oriented interventions. Since none of the reviewed interventions have been systematically implemented or tested with European Roma youth or their families, careful formative research and cultural adaptation of existing evidence-based strategies could be beneficial for Roma. Other unique results in the study illustrate the importance of the intergenerational transmission of ethnic identity as a resource rather than a hindrance for youth positive development. With the use of structural equation models, it was shown to be a relevant aspect for all study participants. Independent of being mainstream Bulgarian or Roma, parents' ethnic identity was positively related to one another, showing some concordance among parents. This is in line with our predictions and theoretical frameworks guiding this study (Cook & Kenny, 2005; Grusec & Hastings, 2014; Sameroff & Chandler, 1975) that

304

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

parents are a major source of socialization during childhood and adolescence as the main agents of ethnic-related issues to their children. For mothers, fathers, and adolescents, a more well-developed ethnic identity was associated with greater life satisfaction. The present results are in line with a prior cross-sectional study with Roma adolescents and their mothers living in Bulgaria (Dimitrova et al., 2014). Because this is an emerging field of inquiry, future longitudinal studies extending beyond the parentechild relationship including siblings and extended family members could provide greater insight into the meaning of our findings. Such studies would also assist in understanding which cultural-based identities maybe of most benefit for Roma adolescents living in and outside Bulgaria. Finally, another unique aspect of the study lifts up the importance of intergenerational transmission of ethnic identity. Although varying, both mothers and fathers' ethnic identity was positively associated with their adolescent child's ethnic identity and this was the case in both Roma and mainstream families. This finding is consistent with a relevant prior study which also showed intergenerational transmission of collective identity (made up of Bulgarian, Roma, familial and religious identity) among Roma adolescents and their mothers (Dimitrova et al., 2014). Therefore, parents act as active and reflective agents for the identity formation of their children by interacting with their offspring and mediating social influences on identity (Schachter & Ventura, 2008). Some nuances in the present study results also point to the particular importance of mothers. Interestingly, mothers' life satisfaction was positively associated with their adolescent child's life satisfaction however, this relation was not observed in fathers and their adolescent child, with a non-significant association found between fathers' and adolescents' life satisfaction. This finding generalized across both study groups (Roma and mainstream). To gain better insight into why fathers' life satisfaction was not related to their child's life satisfaction it would be important to have had additional study measurement, such as documentation of the living situation and quality of parentechild interactions in participating families. Although this study did document that intergenerational transmission of ethnic identity took place in the studied families, it was limited in that did not have the type of measurement approach to document how this type of transmission might have taken place. What specific activities and exchanges promote identity work around a Roma or mainstream Bulgarian identity? Do the nature and intensity of these activities and exchanges vary by ethnic/national groups and what role might other key social identities play in these activities (i.e., identity as it concerns other culture related identity such as social and economic status, religiousness, urban or rural identities, see Cohen, 2009). Despite the noted study limitations regarding limited measurement, study results clearly highlight the importance of parents and family as a context for ethnic identity development and that such identity work can pay off in terms of greater life satisfaction, although it is important to note that it cannot be assumed that the directionality of intergenerational transmission is one way from parents to children, and is likely to be reciprocal in nature. Study limitations, strengths, and conclusions This study is valuable because it involved an understudied ethnic minority group that experiences clear marginalization (e.g., Civil Society Monitoring Report Bulgaria, 2013). Efforts are needed to better support Roma families living in Bulgaria and in other European nations. This study also breaks new ground in that it is one of the first to investigate the intergenerational transmission of ethnic identity with Roma mothers, fathers and their adolescent children. Although this study is important on a number of grounds, study limitations include the cross-sectional nature of the design, which compromises the ability to infer directionality or causality. Moreover, all measurement was self-reported. Indeed, a more diversified measurement approach (i.e., qualitative and quantitative and the addition of observational ecology-oriented assessment) could have provided insight into the study hypotheses and research questions. Additional investigations may expand measurement of ethnic identity by including both Roma and Bulgarian identity for the Roma participants as to compare bicultural identity assets of this minority group. Although the study response rate was acceptable (ranging from 70% to 98% across informant groups), missing survey responses from fathers was not optimal at 27%. It is also important to note that the term Roma was used to describe the ethnic minority group. However, it is recognized that the present study finding do not generalize to all families with a Roma heritage presently living in Bulgaria. By necessity, for example, the sample selection was limited to Roma families with an adolescent who was attending school. Thus, the present study sample is a selective one since secondary education Roma student attendance is estimated to be at only 10% (Ringold, 2000). The study sample is also drawn from the Southern regions of Bulgaria further making this sample selective from a regional standpoint. While limited, this study does contribute to a newly emerging evidence base on this important European minority group by highlighting the importance of the intergenerational transmission of ethnic identity in Bulgarian families, those with a Roma and mainstream family heritage. Consistent with the growing evidence base on ethnic identity interventions (e.g., Ferrer-Wreder et al., 2014), ethnic identity work is an important developmental process to enhance systematically through planned interventions, as well as through the identification of existing ecological assets, such as those within the family. Acknowledgments The authors would like to acknowledge the support by a COFAS FORTE (Swedish Research Council for Health, Working Life and Welfare) Marie Curie Grant (Forte Projekt 2013-2669) to the first author. We are also extremely grateful to Dr. Eva Jecheva and the

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

305

National Agency for Child Protection, all schools and teachers and in particular Neli Filipova, Ivanina Noncheva, Radka Kostandinova, Albena Damianova, Stoyka Jekova, Svetla Atanasova, Neli Kiuchukova, Ani Angelova, Lilia Stoyanova, Tianka Kardjilova, Zvetan Terziev, Elena Ianeva, Venizslav Chobanov, Kamelia Mateva, and Petko Petkov for their help in carrying out the study.

References Amnesty International. (2007). Europe: discrimination against Roma. Retrieved from http://www.amnesty.org/en/library/info/EUR01/012/2007. Arbuckle, J. (2009). Amos 19. Crawfordville, FL: AMOS Development Corporation. Barany, Z. (2001). The East European gypsies: Regime change, marginality and ethnopolitics. Cambridge, United Kingdom: Cambridge University Press. Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology, 46, 5e34. http://dx.doi.org/10.1111/j.1464-0597.1997.tb01087.x. Bracey, J. R., Bamaca, M. Y., & Umana-Taylor, A. J. (2004). Examining ethnic identity and self-esteem among biracial and monoracial individuals. Journal of Youth and Adolescence, 33, 123e132. http://dx.doi.org/10.1023/B:JOYO.0000013424.93635.68. Browne, M. W., & Cudeck, R. (1993). Alternative ways in assessing model fit. In K. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136e162). Newbury Park, CA: Sage Publications. Chuang, S. S., & Tamis-LeMonda, C. S. (2013). Gender roles in immigrant families. New York: Springer ScienceþBusiness Media. Civil Society Monitoring Report Bulgaria. (2013). Implementation of the national Roma integration strategy and decade action plan in 2012. Retrieved from http://www.romadecade.org/cms/upload/file/9270_file4_bg_civil-society-monitoring-report_en.pdf. Cohen, A. B. (2009). Many forms of culture. American Psychologist, 64, 194e204. http://dx.doi.org/10.1037/a0015308. Cook, W. L., & Kenny, D. A. (2005). The actorepartner interdependence model: a model of bidirectional effects in developmental studies. International Journal of Behavioral Development, 29, 101e109. http://dx.doi.org/10.1080/01650250444000405. Crocetti, E., Fermani, A., Pojaghi, B., & Meeus, W. (2011). Identity formation in adolescents from Italian, mixed, and migrant families. Child & Youth Care Forum, 40, 7e23. http://dx.doi.org/10.1007/s10566-010-9112-8. Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71e75. http://dx.doi.org/10. 1207/s15327752jpa4901_13. Dimitrova, R., Chasiotis, A., Bender, M., & van de Vijver, F. J. R. (2013). Collective identity and well-being of Roma minority adolescents in Bulgaria. International Journal of Psychology, 48, 502e513. http://dx.doi.org/10.1080/00207594.2012.682064. Dimitrova, R., Chasiotis, A., Bender, M., & van de Vijver, F. J. R. (2014). Collective identity and well-being of Bulgarian Roma adolescents and their mothers. Journal of Youth and Adolescence, 43, 375e386. http://dx.doi.org/10.1007/s10964-013-0043-1. Erikson, E. (1968). Identity: Youth and crisis. New York, NY: Norton. European Union Agency for Fundamental Rights. (2010). Addressing the Roma issue in the EU. Retrieved from http://fra.europa.eu/fraWebsite/roma/roma_ en.htm. Farver, J. A., Narang, S. K., & Bhadha, B. R. (2002). East meets West: ethnic identity, acculturation, and conflict in Asian Indian families. Journal of Family Psychology, 16, 338e350. http://dx.doi.org/10.1037//0893-3200.16.3.338. Feinberg, M. E. (2003). The internal structure and ecological context of coparenting: a framework for research and intervention. Parenting: Science and Practice, 3, 95e131. http://dx.doi.org/10.1207/S15327922PAR0302_01. Ferrer-Wreder, L., Montgomery, M. J., Lorente, C. C., & Habibi, M. (2014). Identity interventions for adolescents: promoting optimal identity. In T. Gullotta, & M. Bloom (Eds.), The encyclopedia of primary prevention and health promotion: Adolescent volume (pp. 1278e1287). New York: Springer. http://dx.doi.org/ 10.1007/978-1-4614-5999-6_183. French, S. E., Seidman, E., Allen, L., & Aber, J. L. (2006). The development of ethnic identity during adolescence. Developmental Psychology, 42, 1e10. http://dx. doi.org/10.1037/0012-1649.42.1.1. Gerganov, E., Varbanova, S., & Kyuchukov, H. (2005). School adaptation of Roma children. Intercultural Education, 16, 495e551. http://dx.doi.org/10.1080/ 14675980500378607. Grusec, J. E., & Hastings, P. D. (2014). Handbook of socialization: Theory and research. New York & London: Guilford Press. €r, D., Fleischmann, F., & Phalet, K. (2011). Religious identification, beliefs, and practices among Turkish Belgian and Moroccan Belgian Muslims: Güngo intergenerational continuity and acculturative change. Journal of Cross-Cultural Psychology, 42, 1356e1374. http://dx.doi.org/10.1177/0022022111412342. van Ijzendoorn, M. H. (1992). Intergenerational transmission of parenting: a review of studies in nonclinical populations. Developmental Review, 12, 76e99. http://dx.doi.org/10.1016/0273-2297(92)90004-L. Inman, A. G., Howard, E. E., Beaumont, R. L., & Walker, J. A. (2007). Cultural transmission: influence of contextual factors in Asian Indian immigrant parents' experiences. Journal of Counseling Psychology, 54, 93e100. http://dx.doi.org/10.1037/0022-0167.54.1.93. Kiang, L. (2014). Ethnicity and ethnic identity in context: commentary on Gjerde. Development, 57(4), 213e221. Lerner, J. V. (1993). The influence of child temperamental characteristics on parent behaviors. In T. Luster, & L. Okagaki (Eds.), Parenting: An ecological perspective (pp. 101e120). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Marcia, J. E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 159e187). New York, NY: Wiley. Marsh, H. W., Hau, K.-T., & Grayson, D. (2005). Goodness of fit evaluation in structural equation modeling. In A. Maydeu-Olivares, & J. McCardle (Eds.), Contemporary psychometrics: A festschrift to Roderick P. McDonald (pp. 275e340). Mahwah, NJ: Erlbaum. Marushiakova, E., & Popov, V. (1993). Tsiganite v Bulgariya [The gypsies of Bulgaria]. Sofia, Bulgaria: Club'90. Mastrotheodoros, S., Dimitrova, R., Motti-Stefanidi, F., Abubakar, A., & van de Schoot, R. (2012). Multigroup ethnic identity measure (MEIM) in Bulgarian, Greek and Dutch adolescent samples. European Journal of Developmental Psychology, 1e8. http://dx.doi.org/10.1080/17405629.2012.688099. Motti-Stefanidi, F., Berry, J., Chryssochoou, X., Sam, D. L., & Phinney, J. (2012). Positive immigrant youth adaptation in context: developmental, acculturation, and socialepsychological perspectives. In A. Masten, D. Hernandez, & K. Liebkind (Eds.), Realizing the potential of immigrant youth (pp. 117e158). New York, NY: Cambridge University Press. National Statistics Institute (NSI). (2011). Census 2011. Sofia, Bulgaria: NSI Press. Nauck, B. (2001). Intercultural contact and intergenerational transmission in immigrant families. Journal of Cross-Cultural Psychology, 32, 159e173. http://dx. doi.org/10.1177/0022022101032002004. Oakes, J. M., & Ross, P. H. (2003). The measurement of SES in health research: current practice and steps toward a new approach. Social Science & Medicine, 56, 769e784. http://dx.doi.org/10.1016/S0277-9536(02)00073-4. Palys, T. (2008). Purposive sampling. In L. Given (Ed.), The SAGE encyclopedia of qualitative research methods (pp. 698e699). Thousand Oaks, CA: SAGE Publications, Inc. €npflug, U. (2001). Intergenerational transmission of collectivism and achievement values in two acculturation contexts: the case of Turkish Phalet, K., & Scho families in Germany and Turkish and Moroccan families in the Netherlands. Journal of Cross-Cultural Psychology, 32, 186e200. http://dx.doi.org/10.1177/ 0022022101032002006. Phinney, J. (1992). The multigroup ethnic identity measure: a new scale for use with adolescents and young adults from diverse groups. Journal of Adolescent Research, 7, 156e176. http://dx.doi.org/10.1177/074355489272003. Phinney, J. S. (1989). Stages of ethnic identity development in minority group adolescents. Journal of Early Adolescence, 9, 34e49. http://dx.doi.org/10.1177/ 0272431689091004. Phinney, J. S., & Baldelomar, O. A. (2011). Identity development in multiple cultural contexts. Bridging cultural and developmental approaches to psychology: New syntheses in theory, research, and policy (pp. 161e186). New York, NY: Oxford University Press.

306

R. Dimitrova et al. / Journal of Adolescence 45 (2015) 296e306

Phinney, J. S., & Ong, A. D. (2007). Conceptualization and measurement of ethnic identity: current status and future directions. Journal of Counseling Psychology, 54, 271e281. http://dx.doi.org/10.1037/0022-0167.54.3.271. Ponizovsky, Y., Dimitrova, R., Schachner, M. K., & van de Schoot, R. (2012). The satisfaction with life scale: measurement invariance across immigrant groups. European Journal of Developmental Psychology, 1, 526e532. http://dx.doi.org/10.1080/17405629.2012.707778.  (2009). Does the canonical theory of assimilation explain the Roma case? Some evidence from Central and Eastern Europe. Ethnic and Racial Prieto-Flores, O. Studies, 32, 1387e1405. http://dx.doi.org/10.1080/01419870903006988. Raykov, T., & Marcoulides, G. A. (2008). An introduction to applied multivariate analyses. NY: Routledge. Ringold, D. (2000). Roma and the transition in Central and Eastern Europe. Washington: World Bank. Ringold, D., Orenstein, M. A., & Wilkens, E. (2005). Roma in an expanding Europe: Breaking the poverty cycle. Washington, D.C.: World Bank. ~ a-Taylor, A. J., Markstrom, C., French, S., Schwartz, S. J., et al., , Ethnic and Racial Identity Study Group. (2014). Feeling good, Rivas-Drake, D., Syed, M., Uman happy, and proud: a meta-analysis of positive ethnic-racial affect and adjustment. Child Development, 85, 77e102. http://dx.doi.org/10.1111/cdev.12175. Sameroff, A. J., & Chandler, M. (1975). Reproductive risk and the continuum of caretaking casualty. In F. D. Horowitz (Ed.), Review of child development research (pp. 187e244). Chicago: University of Chicago Press. Schachter, E. P., & Ventura, J. J. (2008). Identity agents: parents as active and reflective participants in their children's identity formation. Journal of Research on Adolescence, 18, 449e476. http://dx.doi.org/10.1111/j.1532-7795.2008.00567.x. € npflug, U. (2001). Intergenerational transmission of values: the role of transmission belts. Journal of Cross-Cultural Psychology, 32, 174e185. http://dx. Scho doi.org/10.1177/0022022101032002005. Schwartz, S. J., Zamboanga, B. L., Weisskirch, R. S., & Rodriguez, L. (2009). The relationships of personal and ethnic identity exploration to indices of adaptive and maladaptive psychosocial functioning. International Journal of Behavioral Development, 33, 131e144. http://dx.doi.org/10.1177/0165025408098018. Serbin, L. A., & Karp, J. (2003). Intergenerational studies of parenting and the transfer of risk from parent to child. Current Directions in Psychological Science, 12, 138e142. http://dx.doi.org/10.1111/1467-8721.01249. Smith, T. B., & Silva, L. (2011). Ethnic identity and personal well-being of people of color: a meta-analysis. Journal of Counseling Psychology, 58, 42e60. http:// dx.doi.org/10.1037/a0021528. ~ a-Taylor, A. J., Zeiders, K. H., & Updegraff, K. A. (2013). Family ethnic socialization and ethnic identity: a family-driven, youth-driven, or reciprocal Uman process? Journal of Family Psychology, 1, 137e146. http://dx.doi.org/10.1037/a0031105. Van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis for crosscultural research. Thousand Oaks, CA: SAGE. Verkuyten, M., Thijs, J., & Stevens, G. (2012). Multiple identities and religious transmission: a study among Moroccan-Dutch Muslim adolescents and their parents. Child Development, 83, 1577e1590. http://dx.doi.org/10.1111/j.1467-8624.2012.01794.x. Yip, T., & Fuligni, A. J. (2002). Daily variation in ethnic identity, ethnic behaviors, and psychological wellbeing among American adolescents of Chinese descent. Child Development, 73, 1557e1572. http://dx.doi.org/10.1111/1467-8624.00490. Yuval-Davis, N. (2006). Intersectionality and feminist politics. European Journal of Women's Studies, 13, 193e209. http://dx.doi.org/10.1177/ 1350506806065752.