Bilingualism, Integration in schools and effects on Social identity : A qualitative study on how language shape social identity of secondgeneration Afghani children in a Swedish context

(1)

Linköping University | Master’s thesis, 15 credits| Master’s programme in Child Studies

Autumn 2018|D52942396

Bilingualism, Integration in schools and effects on Social identity

A qualitative study on how language shape social identity of second-generation Afghani children in a Swedish context

Maroof Hewad

Supervisor: Asta Cekaite Examiner: Mats Andren

Linköping University SE-581 83 Linköping, Sweden +46 013 28 10 00, www.liu.se

(2)

Lastly, I want to thank Linköping University for such an amazing, interesting and content rich master program titled as Master Program in Child Studies. The entire study period have been a personal development that made me reflect upon so much in the field of child studies. Thank you TEMA BARN team 

(4)

ABSTRACT

Immigration is a common phenomenon in contemporary society and results in challenges with maintaining cultural identity while integrating with a new culture. This is particularly important in the family unit, where first-generation immigrant communities may have different levels of experience and cultural values compared to their offspring (Second-generation immigrants). Integration in second-(Second-generation immigrants is often related to bilingualism and the concept of a dual social identity, embracing elements of host and native culture. The aim of this study was to explore the perspective of parents on the role of bilingualism on integration and identity in their children. A sample of five families, originally from Afghanistan and now living in Sweden, was purposively selected to complete semi-structured interviews exploring bilingualism, social identity and integration. The findings of these interviews revealed three key themes relating to the adoption of Swedish identity by children, the loss of Afghani identity, and the role of the school and society in facilitating these identities.

Overall, the interview data suggested that children living in Sweden identified as Swedish and engaged in cultural and social activities consistent with integration. Bilingualism was an important aspect of holding a largely Swedish identity, although parents valued dual identity of children. Often, immersion in Swedish culture and limited opportunities to socialise as an Afghani cultural unit were viewed as threats to maintaining or developing a clear Afghani identity. The role of schools, parents and society in general is considered important in these processes, particularly with regards to supporting cultural and religious values of the family unit, while promoting successful integration. Bilingualism was universally viewed as a positive tool that could facilitate integration and dual identity, where suitable opportunities arose.

Keywords: Bilingualism, Integration, Identity, children, Parents, social identity, first – generation, second- generation

(5)

INTRODUCTION

Immigration is an important topic in education, health care, social care and politics in contemporary society (Young, 2016:111). Although immigration is hardly a new feature of the human population, shifts in immigration patterns have occurred throughout history (Chin, 2017:77). Immigration may be driven by a multitude of factors, including social and political changes in one’s home nation, a desire for prosperity or improved life opportunities, and additional personal factors that contribute to decision-making (Kalter et al., 2018:20). In recent years, Europe has seen a growing number of immigrants arriving from Central Asian and African communities, where war, social unrest and persecution have prompted an exodus in populations (Geddes and Scholten, 2016:66).

In Sweden, there has been a steady influx of immigrant families since the 1990s (Bygren and Szulkin, 2010:1305). It is estimated that close to 100,000 individuals migrate to Sweden every year, leading to a net migration of over 50,000 individuals per annum (Bygren and Szulkin, 2010:1305). The present figures of immigration exceed those seen in previous peak periods, such as following the Yugoslavian war in the 1990s, reflecting contemporary wars and civil unrest, driving a significant proportion of asylum seekers as part of the immigrant cohort. Indeed, population data from 2012 (Statistics Sweden, 2012:1) suggested that 6,525 asylum seekers were among a total of 13,248 immigrants from Afghanistan (aged 20–64 years) during the year. The Afghan immigrant population represents a relatively small proportion of total refugees during this time period but reflects an increasing level of cultural diversity within Sweden over time, particularly as immigration levels appear to be consistent over the past few years (Statistics Sweden, 2012:1).

The importance of understanding the epidemiology of immigrant populations, first- and second-generation, in Sweden lies in the potential challenges facing these populations. Immigrant communities may be at risk of poorer socioeconomic outcomes than the host population and may have specific needs in terms of education, health care and social care (Hooijer and Picot, 2015:1880). This is particularly true of asylum seekers fleeing war or civil unrest, as seen in a proportion of Afghan immigrants. However, from a broader perspective, the capacity of immigrant populations to engage with a new culture and to thrive within that

(6)

culture reflects how individuals within that population can overcome challenges associated with relocation due to asylum seeking or other causes of immigration.

One of the challenges facing any immigrant population is that of integration into the host society (Alba and Foner, 2015:34). Integration is a complex phenomenon and can be defined according to the perceptions of the individual, perceptions of the host society, physical actions (i.e. active citizenship) and language use, all of which serve as markers or proxies to integration (Alba and Foner, 2015:34). On a fundamental level, integration implies the adoption of practices and sociocultural norms associated with the host society, which may be distinct to those engrained in the culture of immigrant communities (Heckmann and Schnapper, 2016:21). Therefore, the impact of integration on identity is considered an important topic in contemporary research (Alba and Foner, 2015:34).

Social identity is a term used to describe the extent to which one belongs to a specific social group, which informs self-concept and emotional attachment to that group (Padilla and Perez, 2003:37). When an individual has dual social identity, he or she has a sense of belonging to two distinct social groups. Social identity is considered an important proxy for the process of social integration, particularly among the immigrant community, as this encapsulates how an immigrant identifies with the host community/society (McNamara, 1997562). However, it is also important to consider integration from the perspective of the host society, as integration cannot successfully occur unless immigrants are accepted as members of the host community (Bhugra and Becker, 2005:18). Successful integration is therefore a complex process that embodies interactions between host and immigrant societies, as well as cognitive states and perceptions of both parties.

The positive effects associated with integration into a host society have been widely noted in the literature (e.g. Koopmans, 2010:12). These include positive effects on well-being, psychological and physical health, and opportunities to flourish among the immigrant community (Masten et al., 2012:22). Furthermore, integration within a society can promote social activities, active citizenship and economic prosperity (DeVoretz, 2008:680; Body-Gendrot and Martiniello, 2016:67). These outcomes reflect the importance of social identity in promoting social cohesion in wider society. Furthermore, social integration may have an impact on future generations, facilitating deeper connections with the host society among second-generation immigrants (Koopmans, 2010:20).

(7)

One of the earliest and most important interactions a second-generation immigrant has with the host society is within formal education (Portes and Rumbaut, 2001:56). In contrast to home life, school exposes children to different social and cultural practices. This is evident in the predominance of host/native students in the school, the overt and covert cultural practices underlying the education process, as well as the use of language in schools (Heath et.al. 2008:212). Indeed, many second-generation immigrants will learn their native language and speak this at home with their parents or wider family members, while the school environment requires that the child adopts the host language to achieve successful education outcomes, socialisation and integration (Portes and Rumbaut, 2001:99). Hence, bilingualism (or multilingualism) may be viewed as an important aspect of early social integration among school-aged children.

Indeed, bilingualism in schools has been recognised as an important factor in combining the language, culture and identity of children (Shin et.al. 2017:77). Therefore, bilingualism within schools may be an important marker of how second-generation immigrants adapt to new social and cultural norms, embodying identities consistent with their native or host nations. Bilingualism not only reflects how second-generation immigrants adopt the practices and norms of their new nation/culture, but reflects how parental values and cultural norms are maintained across generations.

Therefore, children and parents within immigrant families are likely to have contrasting experiences of identity and integration, with bilingualism in children serving as a key observable manifestation of this duality. How bilingualism is viewed by first-generation immigrants as a part of the process of their children integrating into society and forging new cultural identities is an interesting research topic that is under-explored and has the potential to delineate the complexity of generational dynamics within the process of immigration.

RESEARCH PROBLEM AND QUESTIONS

The research problem for this study was based on a number of factors relating to the rise in immigration in Sweden and the importance of social identity and integration for these immigrant communities and populations. As immigrants integrate within Swedish society,

(8)

there is an important need to understand how this process may influence social identity and a sense of belonging to one group or several groups.

The importance of this research can be considered from the perspective of school-aged children, parents, the wider migrant community and educational staff within Sweden. Although the effect of bilingualism on social identity and integration has been considered for other languages in Sweden (e.g. Finnish, Romani), there is no clear consideration of how languages from distant regions may influence these areas. This includes immigrants from Afghanistan, who form an increasingly large group in Sweden. Appreciating how children of immigrants from Afghanistan integrate into schools and forge a social identity may be important in shaping how schools incorporate bilingualism into schools, to maximise participation and inclusion. Furthermore, it is important to consider how parents react to and perceive this integration process, as part of a wider cultural understanding of the process.

The aim of this study was to explore the perspectives of immigrant parents of school-age children who speak a language other than Swedish or English in the home setting, with a focus on how language shapes social identity in this group. Specifically, the research question was: “How do Afghani immigrant parents perceive the influence of bilingualism on the social identify of second-generation Swedish children?”

This overarching research question was also supplemented with the following questions, which served to analyse more specific aspects of the topic:

 How do immigrant Afghani parents perceive the influence of bilingualism on

integration into Swedish schools?

 How do immigrant Afghani parents perceive the influence of integration into Swedish

schools on cultural identity?

 How do Afghani parents perceive that their child is affected by having a dual social

identity?

The following literature review considers the wider justification for this research, including the specific features of Swedish society and culture that may impact on integration and social

(9)

identity. Furthermore, consideration is given to the theoretical bases of social identity and integration, grounding potential research findings within clear frameworks.

LITERATURE REVIEW

Overview

This section aims to provide an overview of the key publications to date in the area of bilingualism and social identity among immigrant communities, with a specific focus on the Swedish context. This section will begin with an overview how social identity is defined in immigrant communities, with consideration of the theoretical background of this construct.

Social identity and immigrant communities

Identity is a nuanced concept that may be understood from multiple perspectives, depending on the position of the researcher and the discipline in which identity is examined (McNamara, 1997:562). While broadly reflective of the ways in which people conceive of themselves and are defined or characterised by others, conceptions of identity are inconsistent and contradictory across the fields of sociology, psychology, philosophy and linguistics (Hogg, 2006:114). To overcome this barrier to understanding identity, the present study adopted the social identity framework developed by Tajfel and Turner (1979:56).

According to Tajfel, (1981:44), social identity can be defined as the aspect of self-concept attributable to the value and recognition associated with membership of a social group, or social groups. Three main components of social identity have been recognised within this framework: a cognitive component (self-categorisation), an emotional component (affective commitment) and an evaluative component (group self-esteem) (Bergami and Bagozzi, 2000:555). Social identity is therefore applicable to all individuals, as every person belongs to at least one social group at some point in their lives. However, social identity is not constant, as membership in these groups can change over time as well as the meaning attributed to membership in these groups, and therefore individual-level social identity changes in response to social

(10)

constructions and contexts (Reicher, 2004:925). However, it is recognised that there is a greater level of constancy to social identity in relation to ethnicity, race, religion and nationality, as these collective social definitions are often an intrinsic part of an individual’s life course.

Multiple factors may influence the development of a social identity within a given context. These factors include aspects of the groups with which individuals identify, opportunities to engage with these groups, and sociocultural norms or pressures to identify with certain groups (Brown and Capozza, 2016:78). Among immigrant populations, social identity is closely linked to the idea of a cultural identity, whereby individuals are defined (or define themselves) according to the nation or culture with which they identify (Brown and Capozza, 2016:79). Culture in this context will be considered in the broader sense of belonging to an ethnic or national group in the present paper, as this definition correlates closely with the experiences of immigrants who move between ethnic groups and nations. When immigrants interact with host nations, there are many factors that can influence how cultural identity is shaped over time. Initially, there is a process whereby immigrant populations tend to seek out other populations or communities within the host nation, allowing for social groups to form and cultural identity to be maintained (Verkuyten, 2018:88). This is particularly true in older immigrants, including adults within a family unit, who seek stability and have a clear cultural identity based on their experiences (Peters, 2017:55). Over time, these immigrants interact with the host community and society in an increasingly diverse number of ways, leading to exposure to factors that may influence social identity. The degree to which the host nation culture contrasts with the native cultural values of the immigrant population may serve as a barrier to integrating with the host nation (Andreouli, 2013:167).

An important point to consider is that first- and second-generation immigrants have different experiences of identity formation. First-generation immigrants are largely influenced by their nation of birth and their formative years are spent within that nation, leading to a clear cultural and social identity (Haller and Landolt, 2005:1188). This sense of identity may persist even following migration to another country based on the persistence of cultural identity once established. However, for second-generation immigrants, exposure to the host nation often occurs during key development or formative years (i.e. childhood and adolescence), which may shape social identity significantly (Haller and Landolt, 2005:1184). Therefore, second-generation immigrants are more likely to adopt a dual social or cultural identity, consistent with influences of their parents’ nation and culture (and corresponding family values) and their

(11)

exposure to the host nation. This dual identity is an important factor in characterising the potential for integration and recognises the effects of not only host nation interactions as a driver of cultural identity, but also the influence of parents and family members.

The following section highlights the importance of parents (first-generation immigrants) on the social identity of children within a host nation. Parental influences and experiences may influence social identity among children, serving as both negative or positive factors in how a child interprets the host culture and native culture.

Considering the parent-child dynamic in social identity among immigrants

As well as factors relating to the host nation, one must consider the opposing forces relating to persistence and maintenance of native (i.e. birth nation) cultural identity. These influences may be derived from social groups within the host nation, contact with family and friends in the birth nation, and family members who have travelled to the host nation. In particular, hierarchical systems within families, including parent-child dyads, may lead to specific influences on cultural identity among second-generation immigrants (Koh et.al. 2009:602).

An interesting perspective on the development of identity among immigrants can be derived from ecological theory, as developed by Bronfenbrenner (1977:155). The ecological theory of human development suggests that an individual is heavily influenced by his or her environment, including interactions between micro- and macro-systems (Torres et.al. 2009:578). These systems may be characterised in multiple ways and include society-level influences and influences relevant to the individual, family and peer group (Darling, 2007:204). It has been noted that parental influences are heavily associated with the adoption of cultural identities among second-generation immigrant family members (Phinney et.al. 2001:138). This includes the degree to which children of immigrants learn the language of the host nation (second language), yielding bilingualism or multilingualism.

Specifically, it is recognised that parental influences can determine the degree to which second-generation immigrants develop bilingualism, based on prioritisation of first language learning and acceptance of second language learning (DeHouwer, 2007:414). First-generation immigrants can influence the cultural identity of their children through a variety of overt and subtle cues, based on emotional factors, social contacts, behaviours and praise or satisfaction

(12)

(King and Fogle, 2013:177). This includes parents using only first languages when at home, as well as explicit pressures to maintain cultural heritage and traditions through the use of language.

Although it may be perceived that parental or family influences would promote maintenance (vitality) of culture through first language learning and maintenance of cultural values and traditions consistent with the nation in which they were born, this is not always the case (Daglar et.al., 2011:266). Indeed, on the contrary, parents may promote second language learning among their offspring in a direct attempt to promote integration into the host nation (Barkhuizen, 2006:66). Drivers underlying this push for integration of the second-generation may be diverse, depriving from the practicality of maintaining a first language (e.g. communicating with wider family members or social groups in native countries), fear of isolation if first language learning is promoted at the expense of second language learning, and wider attitudes towards integration into the host nation as either desirable or undesirable for their children (Karpava, 2018:110).

Bilingualism in offspring can be challenging for first-generation immigrants, as it requires a significant amount of effort to rear a child according to bilingual strategies (Curdt-Christiansen and Lanza, 2018:125). Furthermore, proficiency in both first and second languages needs to be monitored in children to ensure adequate sociolinguistic development and the potential for integration within society (Shin, 2017:156). The impact of bilingualism on the potential for second-generation immigrants to shape their social or cultural identity in the host nation is a potentially important consequence of this integration process.

The following section of this literature review considers the importance of bilingualism as a potential proxy for social identity in the wider literature, providing a refined focus for the exploration of integration in this context. This section will consider contemporary literature linking bilingualism to changes in identity among second-generation immigrants, while highlighting gaps in the literature.

(13)

Bilingualism as a driver of cultural identity

Although there is a dearth of studies exploring the influence of bilingualism on social identity among Afghanistan immigrants in Sweden, data from other populations suggest that there may be a link between these two factors. This includes immigrant populations in Sweden, as well as immigrants in other nations in Europe. For instance, Hochman and Davidov (2014:344) analysed the link between proficiency in German language among immigrants in Germany and the affects on national identity, hypothesising that these constructs are inter-related. The authors found that language proficiency in German among first-generation immigrants led to an increase in national identification as ‘German’, but that national identity had no reciprocal influence on language proficiency. This study therefore highlights how second language learning may influence identity on a nationalistic level among immigrant populations. However, the findings are limited to first-generation immigrants and adult populations, which may poorly reflect the link between language and identity among second-generation children in the school setting.

Gogonas and Michail (2015:198) explored the use of languages in a cohort of Albanian immigrants (n=200) living in various parts of Greece, including first and second-generation immigrants. Interestingly, the authors noted that bilingualism was a fluid concept and the use of languages varied according to setting and attitudes towards integration. Greek (second language) was principally spoken in settings outside of the home, while Albanian (first language) was spoken most often at home and among family members, particularly by the second-generation immigrant cohort.

The importance of the study by Gogonas and Michail (2015:198) lies in the realisation that both first and second-generation Albanian immigrants in Greece recognise challenges in maintaining language-related aspects of social identity (ethnolinguistic vitality), with individuals adapting language use based on the degree to which they wish to integrate into Greek society. This finding highlights the importance of language as a tool for integration, while potentially highlighting a shift in social identity corresponding with changes in language use over time.

With respect to the study by Gogonas and Michail (2015:198) it is important to note that the cultures of Greece and Albania, while distinct, are not completely contrasting and therefore

(14)

there may be more pronounced effects of social identity loss or change when differences between languages or cultures are more pronounced, as is the case with the cultures of Afghanistan and Sweden. Indeed, this is the same challenge as noted in much of the published literature in the role of bilingualism in defining aspects of social identity in Sweden, which focuses on Finnish language speakers; cultures of Finland and Sweden, and corresponding social identity, may be more closely related than those of Afghanistan and Sweden.

The final section of this review considers how integration of immigrant communities occurs and is theorised, with a focus on Swedish schools. This includes an overview of integration as a policy objective in Sweden and consideration of the diverse factors influencing integration, with consideration of the influences of bilingualism and identity. Finally, this section will isolate gaps in the knowledge base relating to the effect of bilingualism in school-aged immigrants in Sweden and the impact this may have on integration and identity.

Integration, identity and bilingualism in Swedish schools

When considering the importance of social identity in immigrant communities in Sweden, it is necessary to appreciate how integration occurs within the school setting, as social identity among second-generation immigrants is fundamentally linked to school experiences. Sweden has long been viewed as an exemplar of multiculturalism among world nations, based on positive multicultural immigrant integration policies and resistance to political pressures to respond to perceived multicultural crises in other European nations (Borevi, 2013:140). The basis of this reputation rests not only in wider social policy and liberalism within the values of social integration models, but also in the promotion of integration in schools (Borevi, 2013:144).

Integration has been recognised as a priority in Swedish immigration policy. Specifically, the civic orientation handbook ‘About Sweden’ notes that integration reflects a sense of belonging where an individual lives and that everyone should “feel they are a part of Swedish society” (City of Gothenburg and the County Administrative Board of Västra Götaland, 2018:13). Specifically, integration within the context of this policy can be viewed as the meeting and interaction of different groups in society. This interpretation of integration is consistent with that noted in the wider literature (e.g. Favell, 2016:11; Kraal and Vertovec, 2017:34), although

(15)

there is less clarity regarding the measurable components of integration in Swedish policy documents.

Evidence for the success of integration in Sweden can be found in a variety of publications (e.g. Borevi, 2013:149; Kalter et.al. 2018:22). However, specific analyses of the experiences, integration and identity of immigrant populations require further examination within the context of literature already presented in this review. It has been noted that immigrant youth in Sweden negotiate identity, literacy and language in concert (Davila, 2017:3), highlighting the connection between these areas. Furthermore, studies have suggested that promotion of bilingualism within Swedish schools may facilitate (and reflect) the formation and maintenance of multicultural identities among children (Bagga-Gupta, 2017:102; Gynne and Bagga-Gupta, 2013:479). Several studies will be considered, reflecting the contemporary evidence base on this topic to show how bilingualism may be viewed as a key aspect of social identity and integration among second-generation immigrants in Sweden.

Ethnographic research conducted in Scandinavia suggests that the identities of children and youth may be constructed, at least partly, based on the spoken languages of the individual, as well as some additional factors. For instance, Henning-Lindblom and Liebkind (2007:161) evaluated identity and language competence in Swedish and Finish among youth living in Finland, and found that perceived competence in the Finnish language did not reduce identification of students with Swedish-speaking Finns. However, multiple cultural or ethnolinguistic identities developed in individuals who were competent in Swedish but had a large network of linguistic contacts who were predominantly Finnish. These findings suggest that ethnolinguistic identities became more nuanced, rather than restrictive, based on language competence and social contact/integration with host populations. Therefore, dual identity and diversity in identification with host language speakers may occur in immigrants within Scandinavia.

Furthermore, teaching of second languages in schools may also influence the social identity of children in Sweden (Bagga-Gupta, 2017:102; Gynne and Bagga-Gupta, 2013:479). In Sweden, the designated national language is Swedish, which is taught as a first language to all Swedish school children. However, English is also taught as a second language in schools due to the widespread use of English throughout the world. Therefore, the school experience of Swedish children is typically defined by learning two languages simultaneously, although not all schools

(16)

routinely teach English. For immigrant children with parents who speak their mother tongue at home, bilingualism or multilingualism is common. It has been argued that schools and wider social institutions need to recognise the importance of bilingualism or multilingualism within school education environments, recognising language and identity as key components of learning and conceptualisation of learning processes (Bagga-Gupta, 2017:102). Furthermore, an analysis of Swedish bilingual school settings (Gynne and Bagga-Gupta, 2013:479) found that multilingualism influences social positioning through everyday social interactions and practices. Indeed, the interactions of different languages and the languages used in instructional contexts in schools can play a role in defining meaning and conceptualisation of learning.

Taken together, these studies suggest that bilingualism is an important aspect of social identity and can play an equally important role in defining the educational environment, including the ways in which students learn and approach learning. It should be noted that data on second language teaching and bilingualism in Swedish schools tend not to focus on immigrant populations associated with refugee or asylum seeker status, including Afghan immigrants. The Afghan immigrant experience has been described in a limited number of studies, although social and cultural identities have generally not been explored in detail.

Furthermore, there are no contemporary studies focusing on Afghan immigrants living in Sweden, based on a search of the literature. This suggests that these experiences have not been adequately defined or explored and that factors contributing to social identity formation in this group are unknown. As Afghan immigrant numbers in Sweden have increased over time, and appear to be continuing to increase, there is an emerging need to understand how these communities integrate into Swedish society. Furthermore, there is a need to appreciate how second-generation Afghan immigrants develop their social identity based on multiple influences, including parental values and exposure to the school system in Sweden.

Summary and unanswered questions

This literature review highlights the definition and relationship between important concepts in understanding social identity, integration and bilingualism in immigrant populations. Two key theoretical perspectives are considered as drivers of social and cultural identity formation among immigrant populations, reflecting the wider effects of belonging to a specific group or

(17)

groups (social identity theory) and being influenced by micro- and macro-level environmental factors, including parental attitudes and guidance (ecological theory). Although these theories are distinct in terms of how they characterise the drivers behind development of an individual over time, within the context of immigrant identity formation they can be combined to highlight the importance of parental drivers and perceptions of second-generation immigrant identity formation.

The literature to date suggests that second-generation immigrants encounter two distinct sources of identity within the host nation: the family or parental unit (home life) and the school environment. Many differences are apparent between these environments and their influence on the individual, including the use of first or second languages. Bilingualism can be considered a direct consequence of adopting a dual cultural identity within a host nation and reflects the balance between integration in the host society and ethnolinguistic vitality or the persistence of native cultural identity. For many second-generation immigrants, the process of integration is a part of life and a necessary adaptation to promote integration, which yields friendships, social connections and opportunities to thrive. However, first-generations are uniquely placed to observe how bilingualism may be linked to cultural identity formation in their offspring, suggesting that these perspectives may provide valuable insights into how bilingualism in particular influences social identity.

Despite evidence to date supporting these assertions, there is a paucity of published research in this area and remaining unanswered questions. Although there are data available in the context of the influence of bilingualism on social identity in Swedish schools, no studies to date have focused on the Afghan immigrant population in Sweden. Furthermore, data on perceived influences of bilingualism on social identity in second-generation immigrants is particularly limited in the wider literature. These knowledge gaps limit how social identity is understood within this context and limits the potential for schools to promote and facilitate bilingualism for integration in Sweden.

(18)

METHODS

Overview

This section provides an account of the methodological approach and practical measures used to meet the aim of this study. Theoretical considerations are noted when deciding on the research strategy and are used to guide the development of data collection and analytical approaches. Details are provided to ensure transparency in the research process, while permitting replication of the study by another researcher. Finally, ethical issues are noted, with an account of how these were minimised in the study.

Theoretical assumptions and approach

Central to the research process is the need to understand how knowledge can be understood and conceptualised to meet the aims and objectives of a study (Hughes and Sharrock, 2016:11). Within the social sciences, two major research philosophies are recognised, with divergent implications on knowledge generation and understanding: positivism and interpretivism (Walliman, 2017:33).A interpretivism approach is considered relevant for this study because interpretivismn provides a more nuanced approach to understanding experiences and phenomena within social sciences and contexts.

Five key approaches are recognised in the qualitative literature: case studies, grounded theory, phenomenology, ethnography and narrative approaches (Creswell and Poth, 2017:12). Although these approaches all have their individual merits and nuances, the ethnographic approach was considered of greatest relevance to the present research topic.

Ethnographic methods have been advocated when adopting a theoretical perspective on the interactions of societies and cultures to produce certain phenomena (Silverman, 2016:66). Indeed, the basis of ethnography is that cultural meanings, interpretations and practices can only be understood where the researcher can observe or impassively collect and analyse data relative to the culture of the participants. Ethnography can be considered an appropriate approach in this instance, as the aim of the study was to understand how culture, beliefs and

(19)

values specific to Afghan culture influence interpretations of phenomena and behaviours in children (specifically, bilingualism, integration in schools and effects on identity).

The influence of selecting the ethnographic approach can be seen in the characteristics of the population selected for the study, the data collection method used, and the lens through which data were analysed, as discussed in the following sections.

Population and sample

One of the key issues when considering the perspectives and experiences of immigrant populations is the need to focus on either first- or second-generation data. As discussed in the background literature, there is an important difference between these generations in terms of the development of identities when interacting with host nations. The focus of this study was specific to the experiences of second-generation immigrants (children) within the Swedish school system. However, for several reasons the perspectives of parents of these children were sought and therefore the key participant group for this study comprised second-generation Afghan parents of children attending Swedish schools.

The reasons for selecting parents rather than children were two-fold. Firstly, it was considered more feasible to access parents within the research process, as additional ethical criteria and permissions would be needed to directly collect data from children (Tisdall et.al. 2008:88). Secondly, the perspectives of parents on how cultural/social identity is formed in children provides a useful perspective on the key issues of the research question, particularly as these processes may be intrinsic to the development of children, yielding little insight into their influence.

The participants were recruited through the Afghanistan Associations local community in southern (skåne) Sweden, as a convenient local area for the author to access the population. Specifically, Afghan parents living with their school-aged children will be selected purposively within this population. Purposive sampling was preferred, as random sampling and other methods would be less likely to identify a participant group with characteristics specific to the aims of the study. Indeed, purposive sampling allowed for individual meeting key criteria to

(20)

be identified and evaluated within the study, maximising the relevance of the findings to the research aim and providing in-depth data on the topic (Etikan et.al. 2016:3).

Eligibility criteria included the need for participant parents to be born in Afghanistan (and to identify with that national group) and for children to have been born in Sweden and attending Swedish schools. Putative participants were ineligible if they had specific communication difficulties (precluding optimal data collection), had children born in Afghanistan now attending Swedish schools, or had children who experienced significant challenges relating to socialisation or integration, including disability, learning challenges and chronic illness that led to missing school. No age restrictions were place on the parents or child participating in the study and it was not necessary for either to be fluent in English or Swedish, as the researcher was able to communicate in their native tongue.

A sufficient number of participants was required to ensure the potential for adequate data collection and data saturation (Sandelowski, 1995:180). Although there are varying standards applied to the definition of data saturation, it has been suggested that combining the number of participants and the depth of analysis achieved during data collection can yield sufficiently robust data that may be revisited during a second data collection process, if necessary (O’Reilly and Parker, 2013:193). Therefore, sample size was largely determined by the willingness of parents to engage in the study and the practical constraints of performing data collection (see below) in the participants. A total of five families (couple parents in joined interview), of which four pair/couple parents and one single parent were included in the final study, reflecting a number of participants commonly used in this form of data collection (Bryman, 2015:201). These Families/parents were directly approached through relevant contact details available from the Afghanistan Association community’s website1 and invited to participate in the study. The Data was transcribed from the participants’ language (Afghani language) to English and due to participants’ wish a translated version of the interview question (Afghani version- Appendix A) were provided to the participants.

(21)

Data collection

Numerous strategies may be used to guide the methodological approach of qualitative research, depending on the intended aims of the project and the anticipated outcomes (Ritchie et al., 2013:125). Qualitative methods include case studies, observational methods in populations and cross-sectional collection of data using surveys, questionnaires, focus groups and interviews. These methodological approaches to data collection are closely related to the qualitative approach adopted (i.e. ethnography, narrative, phenomenological, grounded theory or case study) and therefore selection of the methodology is contingent upon the basic assumptions of the qualitative approach.

The ethnographic method is commonly associated with observational data sets, whereby the researcher impartially observes a culture or group over time, noting behaviours and interactions within the group to develop an in-depth understanding of the culture (Reeves et.al. 2008:1020). However, this may not be a feasible approach to appreciating how cultural values lead to interpretation of phenomena, may be time-consuming and may also be exploitative if not performed carefully (Taylor.et al., 2015:80). Therefore, other data collection strategies have been considered compatible with the ethnographic approach.

Interviews are considered one of the most diverse data collection strategies in practice, within the fields of both qualitative and quantitative analysis (Sutton and Austin, 2015:227). Interviews have a number of advantages over other data collection methods used in qualitative research. Therefore, it allows for a for a focused, in-depth method of data collection from a small sample of participants, while permitting the researcher to actively engage with participants to develop data further and explore specific issues as they emerge.

Semi-structured interviews permit a combination of these approaches and greater flexibility in the research process by pre-specifying key topics of discussion, while permitting the researcher and participant to guide the depth of discussion in key areas or to discuss associated areas of interest to the topic (Galletta, 2013:33). Therefore, semi-structured interviews were used in this study to provide maximal opportunities to explore key topics and the specific experiences of participants simultaneously.

(22)

Each interview was conducted with the participant on a one-to-one (couple parent as one unit) basis at a time and place agreed upon in advance. Following recruitment and contact with participants, the purpose of the interview was explained and the arrangements were made by telephone or email, depending on the communication preferences of the individual. The interview process was influenced by the recommendations of Bryman (2015:201,469), based on the conduct of the interviews, content of the interviews, and the subsequent analytical processes employed.

The interview setting was agreed upon by the researcher and participant, including the home of the participant, where preferred. Key criteria of the setting included the need for a quiet and private space, where interruptions would be minimised and the participant was free to speak about any topic they wished. Each interview lasted approximately 90 minutes to allow for sufficient data collection and exploration of the key issues defined in the research guide/schedule. The interview schedule and questions used to guide the interview process are included in Appendix A.

All interviews were recorded using a tape recorder (with back-up recordings made using a voice recorder loaned from the Linköping University) to permit later transcription and analysis. No notes were taken during the interview, to ensure that the participant was not distracted and that the interviewer could maintain a focus on the responses, permitting more effective engagement with the participant. Following completion of the interview, the participant was thanked for their time and participation. Opportunities to ask questions, clarify the intended outcomes of the research and to withdraw from the study were also provided.

Data analysis

Numerous approaches may be employed to analyse interview transcript data, depending on theoretical assumptions of the researcher and the intended outcomes of the analysis (Grbich, 2012:61). From a theoretical perspective, this study is grounded in an interpretivist philosophy, specifically relating to ethnographic interpretation of the data set. Reeves and Colleagues (2008:1020) note that analytical approaches in ethnography generally focus on an inductive process of thematic analysis. This process implies that data from sources or transcripts are

(23)

analysed and themes generated as more data are included within the study (Rezaei, 2016:10). As this analytical strategy may be easily applied to data from semi-structured interviews (Smith and Firth, 2011:55), a suitable thematic analysis framework was selected to facilitate data analysis.

The framework of Braun and Clarke (2006:86) was specifically selected, as this is widely used in the literature and is considered a structured approach suitable for use by novice researchers (Willig, 2013:71). The framework comprises six key stages of elements, all of which were followed precisely to yield the themes for discussion. These stages progress through a within-text analysis of the data, whereby individual transcripts or data elements are analysed to a between-texts analysis, which involves comparison of texts or data sets.

The first stage involves a complete reading of all transcripts to become familiar with the data set (Braun and Clarke, 2006:86). Each transcript was read several times to encourage familiarisation with the content prior to making any notes or proceeding with the analysis. The second stage involved the initial generation of codes, which were defined as interesting features of the data set (Braun and Clarke, 2013:55). More specifically, codes were words, phrases or sentences that formed an interesting account of the topic under exploration. These codes were isolated for all interview transcripts, with specific markers/highlights made on transcript data, with reference to the line of text in the document. Importantly, each transcript was independently analysed in this manner to generate a unique set of codes relevant to the content of the transcript, rather than being influenced by codes from other transcripts.

Once coding was completed for all manuscripts these codes were compared across the entire data set and used to generate themes (Braun and Clarke, 2006:88). These themes represented shared patterns of thought or meaning between the codes and were developed based on the emerging meanings from across the entire data set, with respect to the aims and objectives of the study. The data were then reviewed to form a thematic map, whereby themes were refined and developed, leading to clear, named themes that represented the key findings of the data set (Braun and Clarke, 2006:89). Once the themes were developed, the transcript data was read once again to ensure clarity in the meaning of the themes, and the relevance of the themes to the data set; this check ensured that the research did not distort or misinterpret data during the analytical process. The final stage of the thematic analysis process involved composing a

(24)

narrative discussion of each theme, based on quotations from within the transcripts and supporting interpretation/narration.

Although this framework provides a robust basis for interpretation of the data set relative to the aims and objectives of the study, further consideration of the quality of the research process is provided in the following section.

Quality in the research process

One of the challenges with research is the need to ensure the reliability and validity of the data collection and analysis approaches (Noble and Smith, 2015:34). These terms are typically not applied to qualitative data, which is less interested in analysing the precision with which findings reflect the data (validity) or statistical markers of consistency in data collection procedures (reliability) and more interested in how the findings and research process can be considered relevant, truthful, consistent and applicable (Roberts et.al., 2006:1).

It has been proposed that standards of quality in qualitative research should reflect four main points: truth-value, consistency, neutrality (confirmability) and applicability (Noble and Smith, 2015:34). Truth-value implies that researchers recognise how their experiences and viewpoints may influence methodological bias to ensure that participant perspectives are truly presented, without bias (Onwuegbuzie and Leech, 2017:238). This requires reflexivity in the research process and clarity in the presentation of data, linking actual transcribed data to interpretation of that data (Alvesson and Skoldberg, 2017:200). Therefore, clarity in the use of interpretative frameworks and the presentation of data were ensured in the present study, along with reflexivity in data presentation.

Consistency and neutrality can be equated to reliability of the data set and describe the trustworthiness of the research method and how researchers may bias interpretation of participant views during data collection (Noble and Smith, 2015:35). Transparency in the research process, data collection stages, data analysis stages and consistency in the study aims and objectives in relation to data outputs were used to promote consistency and neutrality. Finally, the applicability of the findings to other groups or settings was also considered, as the small sample size from a specific region may not reflect experiences of the entire Afghan

(25)

diaspora in Sweden (Ryan-Nicholls and Will, 2009:3). Therefore, a detailed, rich account of participants and the local setting was provided to allow for critical interpretation of applicability of the findings outside of the present research context.

Ethical considerations

Any research processes involving human participants should be scrutinised with regards to the ethical standards of the treatment of participants and the processes employed by researchers (Miller et.al. 2012:19). Within different research contexts there are various ethical priorities that need to be promoted, although many ethical principles are consistent across settings and disciplines. In the context of the present study, there was the need to consider the ethical factors relating to beneficence (potential for good), the avoidance of harm (non-maleficence), justice, autonomy and vulnerability of the study participants and their families (Beauchamp and Childress, 2011:144).

The aim of the study was to provide an insight into the experiences and interpretations of phenomena of an immigrant population living in Sweden. While it was not expected that the participants would directly benefit from the study, beneficence was considered relevant in that a deeper knowledge of bilingualism and identity in the Swedish school context may yield improvements in the knowledge base for future research, and the development of hypotheses that may be tested in future studies. The risk of harm to individuals within the study was minimal, based on a lack of exposure to physical or psychological harm. Emotional or psychological factors were appreciated during discussion of immigration issues and identity, but the topic of conversation was not considered sensitive or likely to cause distress to participants. If distress or discomfort was evident or noted by the participants, the interview process would have been terminated and further advice sought from a supervisor.

Justice is an important issue in research involving vulnerable populations, including immigrants. All participants were treated equally in the research process and non-discriminatory practice emphasised the need to maintain sensitivity and objectivity during data collection and interpretation (Denzin and Lincoln, 2011:33). Finally, all participants had their autonomy preserved by ensuring that written consent was provided prior to data collection.

(26)

Consent forms were completed by all prospective participants (Appendix B) to provide informed, written consent for participation in the study. Furthermore, consent was based on a detailed account of the study purpose, processes and intended outcomes, without coercion or incentives to participate. Participants could ask questions or withdraw from the study at any point, further emphasising autonomy of the participants (Denzin and Lincoln, 2012:33).

All collected data were made anonymous by removing participant information (e.g. name, address, contact details) and assigning a pseudonym/number to the transcript. Data were stored on secured tape-recorder and only accessed by the researcher. Data are only kept for as long as necessary to complete analysis and in accordance with local institution data storage policies.

ANALYSIS

Overview

A total of five families, four couple parents and one single parent, were interviewed to form the primary data set for subsequent thematic analysis. The researcher interviewed the four parents in pairs as couple/pair parents and the fifth parent was interviewed alone as single parent upon participant’s own wish. Three main themes were derived from the responses and are presented according to the standard approach advised in the Braun and Clarke (2006:77) framework. This includes an analytical overview of the data set, accompanied by quotations to reinforce the interpretation of each theme. Where relevant, sub-themes are also presented to structure the discussion of the content.

THEME ONE: The Role of the Swedish language in Integration and Acceptance

The first theme highlights the general findings that parents perceive their children as having a strong Swedish identity. Membership in Swedish society was viewed as an important aspect of well-being and socialisation. This form of membership tended to be defined in a number of discrete ways, based on specific behaviours and observations of parents. This included the formation of close friendship groups between their children and native Swedish children, acceptance of Swedish culture (clothing choices, music and food preferences) and in adopting

(27)

Swedish traditions, including religious and cultural traditions. Therefore, integration was viewed as a process that affected multiple aspects of the life of the child, reflecting a wide level of engagement with multiple aspects of Swedish culture that contrast with Afghani culture.

To some extent, integration with Swedish society was viewed as a complete process, whereby children appeared to identify with Swedish society to such an extent that this defined their social identity. Parents often reported that their children essentially identified as Swedish and often did not recognise a key difference in cultural values with other Swedish children (natives). However, some cultural differences were evident, often due to exposure to parents and family traditions, as well as in terms of physical appearance compared to other Swedish children. However, Swedish identity was seen as a vital manifestation of how the child interacted with society, while this identity was viewed as essential to a sense of belonging and was associated with feelings of pride and social wellbeing.

“They feel proud and count themselves as part of the collection…They are as of them. They are they. They are Swedish.” Family A

“His speaking Swedish is something he is proud of and he is happy.” Family B

A key theme that emerged from the interview transcripts was the recognition that language was strongly associated with integration into the Swedish community. The ability to speak Swedish was viewed as an integral marker of social integration and was a clear facilitator of social engagement with classmates, friends and wider society. The use of Swedish language was dominant among children, spoken far more often than Afghani languages (even at home), and was felt to represent a way for children to understand the world and their place in it, particularly with reference to Swedish society. Specifically, a sense of security was associated with the use of Swedish language, which often manifested as parental perceptions of confidence and socialisation of children when using Swedish languages:

“Their security is the culture and language they speak in here” Family D “His security comes with the language he is comfortable.” Family B

(28)

“…language is security…” Family E

“My children are more social I think because of their bilingualism and that they are bilingual.” Family C

Therefore, there was a clear significance attached to bilingualism as a component of integration and general wellbeing of the child. Indeed, bilingualism was viewed as advantageous over knowing a single language, as parents felt that this allowed children to draw on multiple vocabularies and enhanced communication. Furthermore, bilingualism was viewed as having an important link with identity, allowing an individual to have dual identity, one associated with each language. Furthermore, this bilingualism was seen as a means of navigating the complexity of have Afghani roots while being largely embedded within Swedish society.

“Bilingualism has really played a vital role in all aspects of this integration” Family C

“…knowing one language makes you one person and knowing two languages makes you be two persons” Family A

“His being bilingual has given him this tendency to feel secure and safe in his identification.” Family B

One of the ways in which bilingualism promoted integration into Swedish culture was through face-to-face contact with other students, although it was recognised that social media communication was increasingly common and formed a central part of adapting to Swedish culture and the use of Swedish language. Activities on social media were generally conducted in Swedish by the children of parents in this study, emphasising the importance of language in facilitating communication. Furthermore, social media was viewed as an additional element of identity, allowing children to forge an emotional attachment with Swedish society and culture. This may also be applied to the use of social messaging applications and platforms, which could be used to connect to Afghanistan-based relatives, adding the potential to promote dual identity.

(29)

“…children are emotionally attached to the Swedish community. Because it’s a trend in the Swedish community that youngsters and children often have own social media pages and contacts.” Family D

Overall, parents generally held positive views of bilingualism, as well as recognising the benefits of bilingualism to their children. Bilingualism was viewed as a means of integrating with society and in forging personal and social identity and parents generally noted that they were in favour of promoting bilingualism as a key aspect of their child’s development and well-being. This finding suggests that first- and second-generation immigrant family members are broadly accepting of bilingualism and actively engage in learning the host language.

“I am proud of my children because they are bilingual. I will promote their bilingualism in all manners.” Family C

“We as parents have no problem with our children participating in school activities. That is a positive thing.” Family E

However, it was evident that bilingualism alone was not the only factor accounting for perceived integration of children into society, as many factors were noted where children expressed a preference for Swedish culture, activities or social norms. Indeed, defining social identity and integration was often based on parents’ perceptions of child behaviours, in addition to the use of language, demonstrating complexity in how these concepts were understood.

”They are in all manners Swedish. They have a Swedish taste of music. They have Swedish friends a lot. They eat mostly Swedish food. They talk Swedish more than they talk Afghani. If that identifies them as being, Swedish so I guess this is it.” Family E In summary, this theme emphasised the degree to which second-generation children embedded into Swedish culture and how integration was associated with a clear sense of belonging to Swedish society. Bilingualism was an important aspect of forming this social identity and while parents were aware of the importance of this mechanism, it was appreciated that children were not always aware of that mechanisms due to the normality of living in Sweden and relative

(30)

lack of exposure to other cultures. The issue of maintaining an Afghani identity in this context is explored in the following theme.

THEME TWO: Loss of Afghani Social Identity and Challenges in maintaining two

Identities

This theme highlights the relative challenges of engaging in Afghani and Swedish culture and social identity for second-generation immigrants, from the perspective of parents. Parents tended to find that their children were more heavily embedded in Swedish society and culture than Afghani culture, reflecting how often children socialised with Swedish children compared to Afghani children, as well as reflecting the wider use of Swedish or Afghani languages on a daily basis.

A key finding was that younger children in the family were more likely to integrate with Swedish society, while also having a lower sense of Afghani identity. These children were perceived as almost fully integrated into Swedish society, which was partly helped by younger engagement with the language and a lack of experience in Afghani languages. The lack of pre-existing identification with Afghani cultural heritage facilitated engagement with Swedish culture, which was also linked to bilingualism. However, young children were also recognised as having a lack of cultural or social identity due to poor understanding of these concepts and a lack of true appreciation of the differences between cultures and cultural backgrounds:

“He has no idea of really what Afghani culture is in all senses. This may come after he gets a bit older.” Family B

“My children are not that big enough to really identify themselves.” Family C

“Because you have no other culture really that you could rely on, but you are programmed in this new culture in all senses. Bilingualism have been a vital aspect of this.” Family A

(31)

In contrast, older children were seen as having a greater level of conflict between Afghani and Swedish social identity. This manifested as hiding elements of Afghani culture, including visual displays of culture (e.g. clothing) and selecting between cultures during engagement with others. Indeed, children were generally seen to have a preference for music, clothing and other values consistent with Swedish culture, rather than Afghani culture. Rejection of Afghani culture in this way was not viewed as overtly negative, but rather a reflection of how children adapted to the Swedish environment and cultural values of their friendship groups.

“I have seen my oldest (10 years) have always masked her Afghan social identity and put on Swedish identity …” Family A

“My children didn’t feel comfortable with having own clothes but instead chose to have Swedish clothes.” Family A

Association with Afghani heritage among children was also viewed as distinctly negative by some parents, with children feeling shame about being from this culture or presenting this culture to other children. These negative feelings were considered a potential driver for wanting to abandon Afghani identity and to fully immerse in Swedish culture. This was concerning to some parents and was considered an obstacle to promoting wider engagement with the Afghani culture and traditions. Furthermore, parents were concerned that abandoning Afghani culture was linked to abandoning a Muslim culture that was held in high esteem and was considered a fundamental part of the child’s identity. Indeed, religious identity was often valued more highly than Afghani identity, complicating an appreciation of how identity arises and persists as a consequence of culture.

“My children sometimes feel shame… not proud to say that they are from Afghanistan and that they speak Afghani language.” Family A

“…it is better to refer to yourself as Afghan… I want my children to, in any case, be a good Muslim and obey the religion.” Family C

“They know what Ramadan is (fast month). They don’t fast and they will never do as far as I know.” Family E

(32)

Specifically, identification with the Afghani culture could be viewed as a means of distancing oneself from Swedish culture, which children were not keen to do. It was also considered that Swedish names and culture were perceived as more desirable or with a higher level of social status that Afghani names, which would lead to children confronting or denying aspects of their heritage. Furthermore, life in Sweden was viewed as generally positive, particularly when contrasting with Afghanistan during holidays or visits, where differences in facilities and infrastructure influenced how children viewed the fundamental divide between identities. A reluctance to discuss these elements with Swedish classmates, out of shame or embarrassment, may lead to a rejection of Afghani cultural values

“I have also witnessed that my children saying ‘our last names are not good because it not matching the Swedish community’” Family A

“My children, they sometimes talk very positive about being Swedish… talks of hygiene that in Sweden it’s more facility. There is electricity.” Family C

“…my children know Afghanistan is a poor country and living standards are weak and not good.” Family C

Parents generally acknowledged and appreciated the difficulties inherent in children growing up in Sweden and integrating into Swedish society while having parents and family from Afghanistan. To some parents, it was inevitable that Afghani social identity needs to be sacrificed as part of the integration process into a new culture or is an inevitable consequence of high levels if exposure to Swedish culture versus Afghani culture.

“We as parents are reminders for them” Family A

“…in my child’s case, that has not been so active as his attachment to Afghan community been not that much that is why his attachment been with Swedish society.” Family B

Bilingualism, Integration in schools and effects on Social identity : A qualitative study on how language shape social identity of secondgeneration Afghani children in a Swedish context

Bilingualism, Integration in schools and effects on Social identity

Table of Contents

Acknowledgements