سرمایه اجتماعی و فاصله فرهنگی به عنوان پیش بینی کننده های ترک تحصیل زود هنگام: مفاهیمی برای اقدام جامعه برای مهاجران داخلی ترکیه

This paper examines the influence of rural to urban migration on early school dropout from compulsory education through effects of social capital drawn from community, in a sample of Turkish youth. The loss of community-based sources of social capital may be the underlying reason for significantly lower levels of school attainment of immigrant youth in comparison to native youth in culturally diverse societies. In the present study, with a sample of 764 adolescents, we show that rural to urban migration at school age (5–15 years) increases a child's odds of dropping out from compulsory education about 103% compared to when the child is not migrated, above and beyond other significant structural risk factors like child labor, having an illiterate mother and no stable house income. The effect of migration on dropping out is significant for adolescents who migrated to the metropolitan Istanbul, but not to other less developed cities, which are similar to the immigrants’ home cities in terms of socio-economic and cultural milieu; 94.5% of migrants to Istanbul but 74.5% of migrants to other cities had dropped out. Positive student–teacher relationship as a form of school social capital increased the odds of staying in school by 65% for migrants to Istanbul. We suggest that social capital factors are critical in the educational attainment and acculturation of migrant children. Interventions should target immigrant children, their families and the communities they live in.

The purpose of this paper is to examine the influence of rural to urban migration on school dropout through effects of social capital drawn from the community. “Community” is increasingly regarded as an important source of social capital for migrants helping them to have access to social capital. Indeed social capital and community are frequently seen as synonymous (Ryan, Sales, Tilki, & Siera, 2008). Migration disrupts the socially structured relationships in communities. Migrant populations develop strategies or processes which have been studied as “acculturation” (Berry, 1980), “assimilation” (Gordon, 1964), “socio-cultural and psychological adaptation” (Ward and Kennedy, 1994 and Ward and Kennedy, 1999), or “adjustment” (Richardson, 1979), depending on the discipline and theoretical paradigm. These concepts vary, but the social psychological process itself is rather similar: it is the reconstruction and renegotiation of the socially structured relationships and identity following transition to a different culture (Colic-Peisker & Walker, 2003). Social capital, though not consistently defined, is linked to the “relations among persons” on whom an individual can draw support (Coleman, 1988). Social capital theory is key to the understanding of effects of migration on a wide range of social outcomes, one being acculturation. The crucial point is that socially structured relations between individuals in social groups (e.g. families, schools, neighborhoods) are sources of social capital that increase the capabilities of individuals in the process of acculturation (Hagan, MacMillan, & Wheaton, 1996). Acculturation strategies occur as a process and are set in a broader socio-cultural context (Ward, 2008). To help us better understand the socio-cultural context of migrants’ acculturation environment, resources and provision of social capital is crucial. Coleman's (1988) analysis of school dropout revealed that the frequency of residential mobility has the strongest overall effect on school dropout. Compared to non-migrants, migrants are significantly less likely to complete high school and more likely to have lower levels of educational attainment and occupational status (Hagan et al., 1996). These results are consistent with recent studies of the effects of migration on educational outcomes and supports Coleman's argument that “for families that moved often, the social relations that constitute social capital are broken at each move” (1988, p. 113). Migration leads to a loss of community-based sources of social capital. Family's capacity to provide compensatory social capital in the form of parental support and involvement may not compensate this loss, since the family itself suffers from the ill-effects of migration. The social capital explanation for the negative association between migration and school performance is that migration damages and sometimes completely severes acculturation and social ties that “inhere in the family relations and in community and that are useful for cognitive or social development of a child” (Coleman, 1990). Along these lines this paper aims to look at the effects of community both at the level of family, neighborhood and school as a provider of much needed social capital in the disrupted environment of migrant children. Rather than employing a generalized notion of social capital and assuming equivalence with community, it may be more helpful to distinguish between the different types and levels of social capital and resources that communities provide. Provision of social capital at the level of school and family leads to different educational outcomes. Students’ performance in school is enhanced by strong social connections both within and between families. For example, when parents have a close relationship with the child, they can monitor the child's school progress closely, to reinforce expectations, and to provide guidance for school-related matters. Similarly students and parents who are connected with the school and community receive additional information about the performance of the child. Norms regarding school performance are stronger and more consistently enforced in communities with strong ties. Social ties with family and community can be consequential for students’ performance in schools (Coleman, 1988). In addition to the parent–child connection, children have ties to teachers, peers, and community members. Students may also be affected by parents’ loss of connection to the school and community. In short, migration affects the maintenance of all of these community connections. 1.1. Rural to urban migration in Turkey People migrating both across national borders and within a country share common problems such as participating in the labor-market and accessing the healthcare, social and educational systems in the new communities and societies they move to. Globally, researchers from different disciplines are striving to understand the short-term effects of migration and long-term integration prospects of immigrants to the social and economic contexts in the host countries (Valverde & Vila, 2003). Education of the young migrants is one of the most important challenges that require large scale policy development and implementation. All children have the right to education (Article 28, UN Convention on the Rights of the Child, 1989). However, research has shown that, in many countries with diverse populations, immigrants on average have lower levels of school attainment than the natives (Glick & White, 2004). For example, by 2003, in the U.S.A., 13.4% of the natives (aged 25 or more) but 33% of immigrants who had immigrated within the past 10 years had not completed high school (Glick & White, 2003). Similarly, in Sweden, 92% of the native pupils, but 64% of the newly migrated pupils received final grades from compulsory education that qualified them to continue at the upper secondary level (Böhlmark, 2003). Migration results in processes such as loss of access to traditional means of livelihood, inability to benefit from citizenship rights, and inability to benefit from the right to education and from educational opportunities, together with housing problems, poverty, child labor. All these problems result in social exclusion (Yukseker, 2007). Rural to urban migration in Turkey has always been one of the crucial social and political issues. Turkish society has experienced rural to urban shifts dramatically starting in the second half the 20th century (Kagitcibasi, Cemalcilar, & Baydar, 2009). Industrialization in large cities and the introduction of agricultural machinery in rural areas brought about an internal migration from rural to urban areas and from Eastern Turkey to Western Turkey (Aksel, Gun, Irmak, & Cengelci, 2007). According to the 2000 population census, nearly 28% of the population was born in a different province than they now reside in.1 This ratio goes up to 62% for Istanbul, a major province that has drawn migrants for years. Starting with 1950s and well into the 1970s rural to urban migration was the dominant pattern in Turkey. In the later periods rural to urban migration slowed down and urban to urban migration became the dominant pattern. The unstable political climate of the 1980s brought a different pattern of migration, namely the internal displacement. Internal displacement took place in Eastern and Southeastern Turkey between 1984 and 1999 due to the armed conflict between Turkish military and the Kurdish separatist movement in the region. Internally displaced persons (IDPs), who originally had resided in villages were resettled in Eastern and Southeastern city centers, and a significant proportion gradually migrated to larger cities in Turkey. The net migration rate out of Southern and Eastern Turkey reached to about 50% of the population in that region, with a majority migrating to Istanbul (1,159,703 people in 2000, comprising about 1/9th of the city's population). Recent urban to urban migration and the displacement caused the loss of “traditional” life styles based on agriculture and made migrants dependent on the market economy for their most basic needs (Yukseker, 2007). Leaving traditional life styles and social networks and moving to cities have had profound cultural and social effects on people. When migrants moved to new environments they lost their traditional social networks as their most significant support system. Work became informal and irregular for men; inability of men to find work placed a burden on women and children. Recent wave of migration and specifically displacement has also had a profound effect on the education of children. Families stopped sending their children to school to minimize their expenses, or they had to choose regarding which one would attend school. Boys are often the ones who are sent to school (Yukseker, 2007). Together with economic and social changes, migration to a different, more affluent, more educated socio-cultural environment brings important changes in individuals’ learning environments, cultural values and development (Greenfield, 2009). Different qualities, skills, and social relations become adaptive in the host culture and this shift triggers immense social and psychological changes. For example, in an earlier study Parker, Kleiner, and Needelman (1969 cited in Furnham & Bochner, 1986) had shown that, people from rural districts moving to a large town in the U. S. were more likely to suffer from mental illnesses than those migrating to other rural towns. Two recent studies with Turkish migrants also revealed similar adaptation difficulties individuals suffer from even when they migrate within their native country. In their 1996 study, Phalet and Hagendoorn examined Turkish youngsters’ adjustment to acculturative transitions in two acculturation contexts: migration from rural to urban settings in Turkey and international migration from Turkey to Belgium. Their findings revealed that youth that migrated to urban cities in Turkey suffered from higher internal adjustment problems (e.g. distress) and lowered achievement values than those who migrated to Belgium. Those migrated to Belgium, on the contrary suffered from more external adjustment problems (e.g. problem solving). The authors suggest that reliance on in-group ties protected Turkish immigrant youth in Belgium against stressful cross-cultural encounters. In another study, Aksel et al. (2007) compared the experiences of two groups of adolescents migrating from a village in Eastern Turkey; the first moved to a nearby town (with similar socio-cultural characteristics to their native) and the second moved to a large, metropolitan city in Western Turkey. Their findings showed that those who moved to the large city suffered from lower self-esteem and lower levels of satisfaction with life compared to those who moved to the nearby small cities. In the present study, after investigating the role of “migration” on youngsters’ school status in general, we seek to further examine the experiences of those who move from their rural towns to nearby towns or small underdeveloped cities and those who move to the metropolitan Istanbul. Istanbul is quite different from the migrants’ native cities, for example in terms of job opportunities and population. It represents a much different culture, with more individualistic values. Our assumption in this paper is that migration to a nearby small town and migration to metropolitan Istanbul demand different adaptation strategies from migrants, for these locations portray significantly dissimilar cultural milieu in terms of resources and provision of social capital. Hence, we predict that migrating to a distant place (both geographically and culturally) even within the same country will have more detrimental effects on students’ education than migrating to new places with familiar culture. 1.2. Education system in Turkey Turkey has one of the youngest populations in the world; age cohort 5–19 years consists of 26.9% of the population (State Statistics Institute, 2007). Formal educational system consists of primary school, middle school, high school and tertiary levels of schooling. Education is free in state schools. Based on a law passed in 1997 primary schools provide 5 years of compulsory schooling followed by 3 years of compulsory middle school. At 1998, the mean school attainment for the adults was 5.3 years. According to the 2007 statistics by State Statistics Institute, net education enrollment is 24% for pre-school, 90.3% for primary school and 56.51% for secondary school in Turkey. The lack of educational infrastructure, regional disparities due to ethnic tensions, migration and uneven economic development have prevented the system to create a fully functioning 8-year primary education structure (Rankin and Aytac, 2006 and Smits and Hosgor, 2006). Moreover, the gender disparities in education are especially noteworthy. By 2008, the net primary school enrollment ratio for girls was 87.93%, whereas this ratio was 92.25% for boys. Adult literacy rates showed a similar tend, 79.6% for females and 95.3% for males. Like in many other countries, with recent campaigns and policies, major progress in enrollment rates is observed over the last decade, especially for girls. However, as also suggested in the Education for All (EFA) Report, access to schooling does not necessarily mean retaining these new entrants within the education system until they get a degree (UNESCO, 2008a and UNESCO, 2008b). Turkey is a particularly interesting case, as school dropout is a problem experienced significantly in the early years of the primary school education unlike the dropout problems in other developed and developing countries, where dropping out from school emerges as a problem during high school years. In a recent study, the average grade of dropout is found to be 4.4 years in Turkey; close to 14% of enrolled students leave the school system before completing the compulsory 8-year program (Goksen, Cemalcilar, & Gurlesel, 2006). Consequently it is not surprising that, in the latest EFA Report, Turkey was mentioned as “one of the 33 countries out of 86 which were not expected to reach EFA's goal for achieving universal primary school enrollment by 2015” (UNESCO, 2008a and UNESCO, 2008b, p. 180). The definition of dropout varies across different research. Some of the commonly cited definitions are; failure to complete compulsory primary schooling and not attending further education or training; failure to gain qualifications or school leaving certificates; failure to gain qualifications required for participation in further education; and failure to gain qualifications required for access to a wide range of labor market opportunities to sustain life chances (Kritikos & Ching, 2005). In this paper we use the first definition and categorize those adolescents who have failed to complete compulsory primary schooling by the time they are 14 years old which is the legal age limit for primary schooling as dropouts. 1.3. Social capital and school dropout Just as dropout is a problematic term, so is social capital. Morrow (1999) defines social capital as an elusive concept and suggests that the use of the term is inherently problematic and needs to be empirically grounded before it can be used in social policy formulations. Despite the disagreements in defining social capital, as Portes (1998) has claimed, there is consensus among researchers that social capital stands for the ability to secure benefits by virtue of membership in social networks. The notion of social capital as it relates to children was developed by Coleman (1990). Coleman defines social capital by its function, as resources derived from people's social ties. Coleman believed that increasing social capital in the family (by investment of time and effort in shared activities or helping house work) and the school (by strengthening social relationships between teachers and parents, and between students and teachers) would increase students’ academic achievement Coleman, 1988 and Coleman, 1990. Scholars have also suggested that these two forms of social capital interact to promote school achievement and to deter children from dropping out of school (e.g. Crosnoe, 2004 and Parcel and Dufur, 2001). Especially in situations where family social capital is not adequate, school social capital may be functional in providing students with the resources they are lacking. Experiencing favorable conditions both at home and at school may further enhance students’ school performance (Parcel & Dufur, 2001). Recent conceptualizations of social capital take a broader approach and incorporate various other sources, like the school and the neighborhood as extra-familial forms of social capital. (Duncan et al., 1994 and Furstenberg and Hughes, 1995). In a study reviewing the conceptualization of social capital as applied to the well-being of youth, Morrow (1999) has criticized existing research for putting too much emphasis on family-effects as the resources of social capital and suggested that, school-related relationships and contexts should also be incorporated in the theory and study of social capital in addition to understanding the effects of families. In addition to their role as organizing the social environment, schools provide a primary site for social interaction (Crosnoe, 2004). The bonds children form with their teachers constitute an important aspect of the school social capital. Both ethnographic studies and survey based studies denote that students who leave school before graduation often cite a lack of social and academic support (Croninger and Lee, 2001 and Rumberger, 1987). Additionally, for at-risk groups (poor, migrant students) perceived positive teacher expectations mediated between sociological status risk factors and alienation from school. As Croninger and Lee (2001) suggested, a supportive and stimulating community can create powerful incentives to stay in school when faced with social and academic problems, especially for disadvantaged groups such as migrants. Research has also shown that extracurricular activity participation significantly reduces a student's likelihood of dropping out of school (Beekhoven & Dekkers, 2005). As such, social capital provides an attractive conceptual link between academic performance and immediate social/community contexts, most notably, the household, school and neighborhood of migrant populations. In this paper, we focus on first generation migrant youth who had migrated from their original towns in Turkey to either nearby small cities or metropolitan Istanbul. We study internal migration as a predictor of dropping out of compulsory education, in relation to other structural factors such as gender, poverty, parental education and related family characteristics (Rumberger, 1987, Rumberger, 2004 and Wagenaar, 1987) and then investigate family- and school-related factors that can be modified through practice, interventions and system changes, or policy reforms to increase the likelihood of students’ completion of basic compulsory education. Particularly, we study the effects of social ties/relationships above and beyond general social risk factors and investigate whether students’ access to social capital from school-related interactions (or school-based social resources) reduces the risk of dropping out from compulsory schooling. We propose three main hypotheses: 1. Rural to urban migration is a significant predictor of dropping out of compulsory education for those who migrate at school age. 2. The detrimental effects of migration will be stronger for those who migrate to a large city that is culturally more distant to the immigrants’ native towns, compared to those who migrate to nearby cities with more similar cultures. 3. Having positive social capital decreases the risk of dropping out of school even when other social risk factors are present.

Our findings reveal that migration at school age increases a child's odds of dropping out of compulsory education about 103% compared to the case when the child did not migrate. For males when the structural risk factors (child labor, illiterate mother, and unstable employment of the father) are present, migration further increases the likelihood of dropping out more than females. There is a significant effect of migration on dropout for the students who migrated to Istanbul. This effect is not observed in the group which migrated to the other cities which are less developed. A closer examination indicates that, migration and the instability of fathers’ employment were significant predictors of dropping out for only those who migrated to Istanbul. Migration at school age, but not before, increases at-risk groups’ odds of dropping out of primary school about 750% only when families migrate to Istanbul. Among the school social capital factors only positive student–teacher relationships have a significant effect on school status of Istanbul migrants. For migrants of the other project cities, both positive teacher-student relationships and parents’ involvement in their child's school work increase the odds of staying in school. Our findings suggest that the impact of structural risk factors such as child labor, illiterate mother and unstable employment status of fathers are very significant on the dropout rates of migrant. However these factors are resistant to change and difficult to ameliorate even with sound policies. On the other hand, our study further shows that social capital factors are indeed critical in the educational attainment of migrant children. Factors such as teacher-student relationships and parental involvement in the child's school work contribute to migrant child's staying in school. These factors, luckily, are more open to modification than social structural risk factors. Among the most revealing of our findings is the differential dropout rate of migrant children in Istanbul and other less developed cities. Istanbul is a huge metropolitan city with a population of almost 11 million. Only in 2007, there were 374.868 registered migrants to Istanbul. Within this group, 57.767 were aged between 5 and 14 years (State Statistics Institute, 2007). Istanbul is a city of global economic relationships, flow of activities and diverse ethnic and economic groups residing in close proximity to each other. As such it severely lacks the elements in facilitating community formation and permanent settlement for migrant populations. As they arrive in the metropolitan city they encounter the formidable task of acculturation and re-building their disconnected ties and identities. Migrants’ ability to mobilize social capital and successfully engage in community ties is significantly hindered in Istanbul. Newly arrived migrants are not able to derive social capital and fulfill their practical and socio-cultural needs from within their own communities. Many forms of social disengagement including school dropout are inevitable consequences of migration to this metropolitan city. Even within a single country different migration patterns lead to distinct adaptation strategies which are displayed through resources and provision of social capital. Our findings indicate at least four distinct environments (cultures) of the migrant populations. The first group is the non-migrant natives of the less economically developed cities with established social ties. The second group is the migrants who moved to the nearby towns which are economically less developed, yet similar in terms of the structured relationships. The third group is the non-migrant Istanbul residents with low income and low education, however with more established resources of adaptation. The final group is the migrants who moved to Istanbul recently with very limited economic means and education. This group experiences serious social exclusion in an environment with minimum opportunities to fulfill their physical and socio-psychological needs from within their own communities. We believe migration to these distinct environments (cultures) reflects different patterns of acculturation. Furthermore, the detrimental effect of migration was more profound when individuals experienced this transition at the time they were about to start school. Migration at earlier ages did not have the same strong effect on dropping out from school. We also believe that this finding suggests that with time, when migrant families become more established in their new societies, that is when they adapt, their children may find the necessary support systems to continue their schooling. As opposed to migrant children in Istanbul, children in other project cities have a slightly different experience of migration. Migration to these cities takes place within shorter distances from their home cities or villages where socio-economic and cultural milieu is not radically different. Integrating into a community requires fewer challenges both in financial and cultural terms. More homogeneous (and familiar) structure of the other cities makes access to resources of social support and integration a much less painful process for migrant families. Therefore, our finding indicating a difference in dropout rates of migrant children in Istanbul and other cities is telling, albeit indirectly, about the effect of community in provision of social capital and educational attainment. Community plays a crucial symbolic role in generating people's sense of belonging. As well as helping us to build a sense of self, communities also enable us to navigate our way around the demands and contingencies of everyday living. In an era where the pace of migration is expected to increase in the foreseeable future, providing children with services aimed at their education, health and general well-being must be a policy priority (UNESCO, 2008a and UNESCO, 2008b). In this general context, early and sustained intervention is critical because the decision to leave school is not an instantaneous decision, but rather a process that occurs over time. Research shows that leaving school early is the outcome of a long disengagement from school and community (Christenson, Sinclair, Lehr, & Godber, 2001). The problem of school dropout cannot be understood in isolation from contextual factors. Schools are embedded primarily in localized community settings (Bronfenbrenner, 1979). Schools serve as a broad context for acculturation not only for academic development, but for cultural learning too. Hence, early dropout reflects a complex interaction among student, family, school and community variables (South, Baumer, & Lutz, 2003). Research has shown that the relative effects of family background and school on educational attainment and achievement vary depending on the socio-economic condition of the societies (Buchmann & Hannum, 2001). Even though research in more industrialized societies like the U.S.A., Britain and Japan has shown family background characteristics to be more important than school factors in determining children's educational outcomes (Coleman, 1990 and Peaker, 1971), research in less industrialized, developing countries has shown the reverse effect. For example, in an earlier study exploring pupil achievement in Africa, Asia, Latin America and the Middle East, Heyneman and Loxley (1983) showed that in low income countries, the effect of school and teacher was greater than the effect of family characteristics on primary school students’ academic achievement. As discussed above, especially for residentially mobile groups and immigrants, in addition to the family and the school, the community children live in also comes into the picture as a factor influencing their academic outcomes. However, as Buchmann and Hannum (2001) have mentioned in their review of education and stratification in developed countries, there is a lack of research on how community factors, operating independently or in conjunction with schools and families, shape educational outcomes in less industrialized contexts. School and family policies and practices are critical for the positive development of youth. In a recent evaluation of a school-based intervention program designed to enhance social capital in vulnerable families with limited human capital through a school-based family support program, Terrion (2006) showed that, when families are given opportunities to develop social capital by being more connected to their communities and more involved in the school, their children showed less behavioral problems and developed more positive attitudes toward education and more adaptive behavior in the school. Improved attitude toward education is considered part of this intervention. Raising the educational aspirations of children in these communities will improve academic performance, because young people tend to adopt the educational attitudes and behaviors of their families and peers (South et al., 2003). Investments in school-based family support programs improving educational attitudes and behaviors are likely to impact the disadvantaged migrant communities. Raising the educational aspirations of children cannot be realized without the contribution of the family. Adult education programs at community levels would serve this purpose. They would increase the educational expectations and aspirations of families and establish a norm about the value of education. They would also help the development of interpersonal bonds in communities which is central to the construction of social connectedness and thus social capital. In most migrant communities and neighborhoods in Turkey there are relatively few community centers or any common meeting locations where residents come together, share their experiences or join certain training programs. Schools, in this sense, are alternative spaces compensating for the lack of community centers. Designing training programs for adults, holding seminars and workshops, organizing social activities located in neighborhood schools would not only bring isolated individuals of the migrants group together but at the same time would build a bridge between the residents and the school environment. All these suggestions implicitly note that there is not one single migrant culture. There are distinct migrant cultures with their specific problems, demands and adaptation strategies. Policy suggestions need to recognize this diversity and develop sound and culturally sensitive action plans. Finally, community interventions are needed that will accurately identify students who are at risk of dropping out and that will recognize the systems that place those youth at risk. The challenge lies in targeting students and families who are in need of intervention based on efficient and accurate predictors that incorporate the influence of school environment and other contextual factors such as the neighborhood the child resides in. Crucial questions here are for whom, and under what conditions, specific outcomes can be achieved. Further research should be conducted to determine which interventions are most effective with specific populations and in specific contexts.