An overview of the risk factors that contribute to child labor in Turkey: Implications for research,policy,and practice

Abstract

Child labor is a serious issue in Turkey. This article provides a review of the current literature on risk factors associated with child labor in Turkey. Emphasizing their multilayered nature, the article examines risk factors contributing to child labor in Turkey by clustering them under individual, family, and structural factors. Recommendations for future research, policy, and practice are also discussed.

Keywords

child labor implications review risk factors Turkey

Turkey is a developing country that struggles with child labor (Göncü et al., 2009). In a recent report (International Labor Organization [ILO], 2009a), Turkey was listed as the third worst of 16 countries in terms of weekly work hours for children. In 2006, 22 percent of Turkey’s population (approximately 16 million) consisted of children between the ages 6 and 17. Six percent of these children engaged in some form of economic activity (Turkish Statistical Institute [TUIK], 2007). Forty percent of the working children engaged in agricultural work and 60 percent in non-agricultural work (TUIK, 2007). Only 53 percent of working children were paid for their labor, whereas 44 percent were unpaid and worked in family businesses such as farms or textile workshops (TUIK, 2007). Rates of child employment in Turkey have decreased significantly. While in 1994, the rate of child employment for children up to age 17 years was 15 percent, this rate was 6 percent in 2006 (TUIK, 2007). Yet, the most recent poverty rate (17%) reported by TUIK (2009), and economic recessions and migration to urban areas still make low-income households vulnerable to child labor (Bakırcı, 2002).

Since the first nationwide child labor survey in 1994, several studies have examined the determinants of child labor in Turkey (Dayıoğlu, 2006; Değirmencioğlu et al., 2008; Karatay, 2000). However, more research is needed to understand the complex, multilayered nature of child labor and to identify the gaps in this literature in Turkey. This article presents risk factors contributing to child labor by clustering them under structural, family, and individual factors. This approach not only allows for a contextualization of child labor in Turkey and an understanding of its multilayered nature, but also provides opportunities to outline new policies, interventions, and directions for future research.

Literature review

Parents’ decision to send their children to work is influenced by a variety of structural, family, and individual factors.

Structural factors

Structural factors represent macro-level obstacles that affect and limit people’s lives and opportunities (Barnes, 2008). In Turkey, structural factors including economic development, migration and urban poverty, societal norms, legal context, and geographical area (urban/rural) contribute to child labor.

Economic development

Until recently, Turkey’s economy has undergone a major financial crisis almost every five years (Degğirmenciogğlu et al.,2008). Both in 1994 and 1999, the economy shrank by 6.1 percent (Dayıoğlu, 2005) and the purchasing power of Turkish citizens was reduced by persistently high rates of inflation (Değirmencioğlu et al., 2008). Turkey has made significant attempts to improve its socio-economic infrastructure. According to the Human Development Index, Turkey ranked 84th among 177 countries in terms of development in 2007/2008, and its economy was among the world’s 20 largest economies in 2006, with GDP per capita reaching US$5482 (ILO, 2009b). Despite these improvements, in 2008, the rate of unemployment was still 11 percent, and 34 percent of individuals lived below the poverty level (TUIK, 2009). Since economic recessions and unemployment are likely to hit low-income families more severely, child labor often becomes a viable alternative to increase family income (Bakırcı, 2002; Değirmencioğlu et al., 2008).

Migration and urban poverty

Migration from rural to urban areas is closely related to increasing urban poverty and hence, to child labor (Yıldız, 2007). Beginning in the 1950s, many rural families migrated to cities due to a shift from an agricultural to an industry-driven economy (Değirmencioğlu et al., 2008). These families experienced significant levels of poverty in the city, and most children joined the labor force to supplement family income (Acar, 2010).

Urban poverty has been on the rise since the 1990s due to internal migration (Şen, 2005). Unlike earlier migration waves, the migration in the 1990s mainly involved Kurdish citizens who were forced out of their villages in eastern and southeastern Turkey due to the armed conflict between the Kurdistan Workers’ Party and government forces (Değirmencioğlu et al., 2008). As Kurdish families who moved to cities lacked the necessary education and skills to find employment, they quickly became the new urban underclass and resorted to child labor as a survival strategy (Altuntaş, 2003).

The majority of working children in urban areas are from Kurdish families who migrated from eastern and southeastern Turkey during the last 15 to 20 years (Yıldız, 2007). Karatay (2000) found that the 70 percent of his 905 participants who worked on the street came from families who migrated from those areas of Turkey. Similarly, almost all of the street-working children in Altuntaş’s (2003) study had migrated from southeastern Turkey due to forced migration.

Societal norms

How a society defines childhood affects child labor practices (Hendrickson, 2009). Until recently, parents in Turkey prioritized children’s economic value (e.g. their contribution to household economy and household chores) (Kağıtçıbaşı, 1998). However, the values that Turkish society attached to children have changed in the last decades (Kağıtçıbaşı and Ataca, 2005). Compared to their data from 1975, Kağıtçıbaşı and Ataca noted that in 2005, families emphasized the psychological benefits of having children (e.g. joy, fun, companionship) more than children’s economic value. Nonetheless, rural and less affluent urban groups still attached greater value to children’s actual contribution to the family economy compared to better-off urban groups.

Culturally prescribed gender roles discourage women from working outside of the home (Şen, 2005). According to patriarchal gender roles, women take care of the house and produce offspring, whereas men provide their families with income and security (Gündüz-Hoşgör and Smits, 2008). This belief system obliging women to stay at home significantly decreases their employment rates (Gündüz-Hoşgör and Smits, 2008), and places children in traditional low-income families at a disadvantage as they are more likely to be mobilized into the workforce.

Legal context

Child labor is defined as ‘work that deprives children of their childhood, their potential and their dignity, and that is harmful to physical and mental development’ (ILO, n.d., para. 3). Turkey was among the first six countries to join ILO’s International Programme on the Elimination of Child Labor in 1992 (ILO, 2009b). In 1995, Turkey also ratified the 1989 UN Convention on the Rights of the Child (Akşit et al., 2001), followed by the ILO Minimum Age Convention in 1998, thereby increasing the minimum age of employment to 15 years (ILO, 2009b). In 2001 the ILO Worst Forms of Child Labor Convention was also ratified to prevent and eliminate all forms of child labor for children under the age of 18 (ILO, 2009b).

Although the legal adult age in Turkey is 18, under Turkish law the definition of ‘child’ or ‘minor’ varies in employment-related regulations. The minimum age for signing employment contracts is age 18, but children younger than 18 can do so with parental permission (Bakırcı, 2002). Conversely, the Turkish Labor Act forbids employing children under age 15, but allows children between ages 13 to 14 to perform light work without specifying what it entails (Bakırcı, 2002). Determining which age range defines child labor is critical for understanding its scope and for designing effective policies and interventions.

The Apprenticeship Law regulates the practice of apprenticeship in Turkey. It recognizes apprentices as ‘students’ not ‘workers’ (Altıntaş, 2006). According to this law, apprentices should be over age 14 and under age 19, attend a vocational education center one day a week, and work five days a week (Altıntaş, 2006). While apprentices are considered students, they work long hours for small amounts of money (Altıntaş, 2006).

Another relevant initiative was the extension of compulsory education from five to eight years (Dayıoğlu, 2005), a legislation requiring children to stay in school until age 15. Despite opposition from small informal establishments that used child labor, this new law was enforced in the 1997–8 academic year (Dayıoğlu, 2005). While some research (Dayıoğlu, 2005; ILO, 2009b; Tor, 2010) indicated that the extension of compulsory education helped to curb child labor, Akkoyunlu-Wigley and Wigley (2008) argued it did not affect child labor rates.

Geographical area (rural/urban)

Children’s work takes place in both rural and urban areas of Turkey. According to TUIK (2009), 47 percent of working children between ages 6 to 17 worked in urban areas; 53 percent worked in rural areas. In cities, children work as street vendors polishing shoes or selling water (Altuntaş, 2006). They also work as apprentices or blue-collar workers in small enterprises, as employees in restaurants or coffeehouses, or as unpaid workers in family establishments (Akın, 2009). In rural areas, children assist their families in agricultural work or animal husbandry (Ertürk and Dayıoğlu, 2004), or work as shepherds, seasonal workers, or farm laborers for others (Göncü et al., 2009).

Until recently, child markets operated in northwestern villages where poor parents rented their children to better-off families for agricultural work (Değirmencioğlu et al., 2008) or as shepherds (Ertürk and Dayıoğlu, 2004). Fathers took their children, invariably sons, to the market for potential employers (Değirmencioğlu et al., 2008). If the child was found fit, he was sent to work for and stay with the hiring family for an extended period of time. Child markets in northwestern Turkey were eliminated after they came under scrutiny, but it is not known whether this practice has been completely abolished (Değirmencioğlu et al., 2008).

Family factors

Family characteristics including household income, household size, parental education, and parental beliefs contribute to child labor in Turkey.

Household income

In Turkey, child labor occurs frequently in low-income families (Göncü et al.,2009). As in other developing countries (Blunch and Verner, 2000; Emerson and Souza, 2008; Ray, 2000), family income is inversely related to child labor (Acar, 2010; Dayıoğlu and Assaad, 2003; Yılmaz, 2004).

In families where children worked, the child’s income constituted 13 percent of the household income. In about one-fifth of these households, more than one child worked, increasing the contribution to 25 percent (Dayıoğlu and Assaad, 2003). Dayıoğlu and Assaad showed that while only 14 percent of households fell below the poverty line (roughly $2000 monthly for a family of four in 2010) when children’s earnings were included, this percentage increased to 26 percent when children’s income was excluded from the household income.

Mothers’ wages did not contribute significantly to child employment, possibly due to the small percentage of working women (16.5%) (Dayıoğlu and Assaad, 2003). Studies estimate that at least 90 percent of mothers of working children are unemployed (Acar, 2010; Karatay, 2000). Fathers’ wages, however, correlates negatively with child labor (Dayıoğlu and Assaad, 2003). Dayıoğlu and Assaad found that a 10 percent increase in adult male hourly wages resulted in a 2 percent decline in male children’s and 0.5 percent decline in female children’s employment probability, supporting the link between child labor and household income level.

Similar to other studies (Acar, 2010; Karatay, 2000), Yılmaz (2004) found that 75 percent of fathers held a job in her ethnographic study of street-vendor children from an inner-city migrant neighborhood of Istanbul. Yet, these were mostly temporary and low-paying jobs such as construction or street work. Almost no mothers worked, except a few who worked at home for low-paying textile industry. Families acknowledged that their children’s financial contribution was crucial for survival.

Household size

In 2008, the poverty rate for families with three or four members was 8.5 percent compared to 38.2 percent for families with seven or more members, which underscores the greater risk for poverty for larger households (TUIK, 2009). Studies consistently show negative correlations between household size and child labor (Acar, 2010; Karatay, 2000). Karatay’s study (2000) on street-working children in Istanbul showed that 72 percent of the participants had three to seven siblings, 24 percent had more than seven siblings, and only 4 percent had one to two siblings. Similarly, Akşit et al. (2001) found that the average household size for children who worked on the street in three large cities in Turkey was larger (7.8) than the overall average household size (5.8) in Turkey in 1998 (Hacettepe Institution of Population Studies, 1999). Acar (2010) noted that socio-demographic differences might exist among children working in different industries. Street-working children lived in larger households (seven siblings) compared to those in the service sector (four siblings).

Parental education

Research suggests that parents of working children have less education (Acaroğlu, 2010; Karatay, 2000; Yılmaz, 2004). In their study of children working on the street, Akşit et al. (2001) found that almost 70 percent of mothers and 32 percent of fathers were illiterate compared to 6 percent of males and 22 percent of females in the general population (World Resources Institute, 2003). Karatay (2000) also reported that 72 percent of mothers and 14 percent of fathers were illiterate in his sample of working children in Istanbul.

As in other developing countries (Blunch and Verner, 2000; Das and Mukherjee, 2007; Ferguson, 2005), Dayıoğlu and Assaad (2003) found that both maternal and paternal schooling had an important although different impact on child labor. Father’s schooling had an indirect impact through increased paternal wages whereas mother’s schooling had an independent impact with an additional year of schooling reducing male and female participation in child labor by 0.7 percent and 0.5 percent respectively.

While Dayıoğlu and Assaad do not provide a rationale for this finding, it is possible that low maternal education correlates with child labor because mothers do not have decent paying jobs due to their lack of education. Thus, children work to support their family. Mothers with less education may also be more open to child labor because they may place less value on education.

Parental beliefs

How parents conceptualize childhood and child labor affects their decision-making processes and childrearing practices. In rural Turkey, it is believed that child labor cannot be harmful because children who work become mature. As children in rural households start working in the fields or as shepherds as early as age 6, parents likely perceive child labor in other contexts as normal (Yılmaz, 2008).

Patriarchal gender roles also affect parents’ perceptions and actions about child labor. In a study of children from three metropolitan cities who worked on the streets, Akşit et al. (2001) showed that both parents and children believed that it was ‘shameful’ for girls to work on the streets, particularly after age 12. According to Akşit et al., the social norm of keeping girls at home provided protection from the hazards of street work.

Another critical issue is parents’ beliefs about education. According to Göncü et al. (2009), middle-class parents expect children to spend after-school time on school work, whereas lower-income children are expected to devote extra time to household chores or to the family business. Like urban families, rural families with a sustainable income prioritize children’s education over work (Ertürk and Dayıoğlu, 2004). Yet, Ertürk and Dayıoğlu (2004) argue that rural middle-class families that rely on labor-intensive farming or animal husbandry prioritize household work and agricultural activities over children’s participation in school. Children’s gender also influences parental beliefs about education. Especially for girls, for whom patriarchal gender roles dictate to be good wives and mothers, education is not prioritized (Ertürk and Dayıoğlu, 2004; Göncü et al., 2009).

However, Akşit et al.’s study (2001) suggests that educational background and culture do not always influence parents’ beliefs. Despite low levels of education, 64 percent of mothers in Diyarbakır and 66 percent of mothers in Istanbul valued education, and believed that education would increase their children’s chances for a better future. Yet, children in these families also had high dropout rates due to school expenses (Akşit et al., 2001) or to poverty (Yılmaz, 2004, 2008).

Individual factors

Children’s age, gender, and child birth order also determine which children are more likely to work.

Age

As in other developing countries (Blunch and Verner, 2000; Fafchamps and Wahba, 2006; Ray, 2002), older children are more likely to work (Dayıoğlu, 2006; Dayıoğlu and Assaad, 2003). Karatay (2000) found that, although children as young as four years old worked on the street, the majority of working children were between ages 10 and 15, with street work decreasing steadily after age 15. Karatay suggested that children older than 15 might lose their ‘attractiveness’ in eliciting empathy from potential customers on the street and might turn to other jobs, such as work in construction or textile industry. Yılmaz (2008) suggested a similar trajectory. In her study, children between the ages of 6 and 13 were mostly engaged in street-vending activities such as selling Kleenex or shoe shining. After age 12, they worked in textile sweatshops. Thus, their disappearance from the street does not imply that older children stop working.

Children working on the street may be younger than children who work in the service sector (Acar, 2010). In Acar’s study, 42 percent of children working on the street started working at age eight or younger. Most children in the service sector started working between ages 13 to 15. The mean age of children working on the street was 12.3 whereas the mean age of those in the service sector was 17, suggesting that the service industry might be the next step for these children. However, it is also likely that these two groups have distinct trajectories since the study showed difference in other characteristics such as parental employment and parental education.

Gender

Girls in Turkey are less likely to work compared to boys. According to TUIK (2009), boys constituted 66 percent of working children between the ages of 6 and 17. Gender moderates how other factors affect child labor. Dayıoğlu (2006) found that boys were more likely to work compared to girls when the father’s employment prospects were poor. Also, while boys were more likely to work as they got older, this was not the case for girls.

Gender affects the type of work children engage in. Patriarchal gender definitions encourage Turkish girls to stay at home, and not act in ways (e.g. interact with males outside the family) that will shame the family (Göncü et al., 2009). In most cases, these expectations create a strict gender division of labor and limit girls either to the home and domestic labor (Ertürk and Dayıoğlu, 2004) or to work as wage earners or unpaid family workers in agricultural settings (Dayıoğlu and Assaad, 2003). For girls who work in an industry, their workplace is selected by their father and is usually owned by a relative or family friend (Göncü et al., 2009). Industries are also ‘gendered’. Girls usually work in textile and carpet industries, and boys mostly work in auto repair, construction, and restaurants (Ertürk and Dayıoğlu, 2004).

Gender also affects the number of hours children work. Dayıoğlu and Assaad (2003) found that females worked about 6 hours less per week than males. Also, girls worked an average of 8.9 months whereas boys work an average of 10.1 months.

Child birth order

With few exceptions (Karatay, 2000), the relationship between birth order and child labor has been understudied. In Karatay’s study with 905 children, 18 percent of street-working children were first-born and 20 percent were third-born children. As the mean birth order of working children was 3.6. Karatay suggested that parents might consider other factors such as age and/or school success in making decisions regarding child work.

In other developing countries, child birth order plays an important role in child labor (Emerson and Souza, 2008; Fafchamps and Wahba, 2006). The oldest male children are more likely to participate in market work, whereas the oldest female children are likely to undertake subsistence work in order to support their family and their younger siblings. A pilot field study in Turkey found that child birth order was more important than gender in deciding which child to send to work (Sensoy Bahar, 2011). Oldest children work regardless of their gender in order to support the family and/or to be able to send younger children to school. This collective coping strategy may give all children in the family an opportunity to receive some education.

Implications

Given that there are various determinants at multiple levels contributing to child labor, a holistic approach should be adopted in research, policies and interventions.

Research implications

At the family level, the role of parental education needs to be better understood. If parents are not convinced of the long-term financial benefits of education, they will be less likely to prioritize it (Buchmann, 2000). Whether low parental education correlates with child labor because it affects parental beliefs about education and/or because it limits parental job opportunities has implications for intervention.

Individual factors identified as critical for child labor in other developing countries (e.g. child birth order, school success) should be further explored in the Turkish context. The relationship between gender and child birth order in child labor warrants special attention in Turkey given the critical role of patriarchal gender norms. Studies working with large datasets and more case-based qualitative studies are both needed to further explore the role of child birth order and its interaction with gender. Elucidating this could help identity which siblings are at higher risk for child labor and be important in tailoring preventive strategies and interventions.

Another important individual factor to be studied is children’s academic success. Orazem and Gunnarsson (2004) suggest that parents prefer keeping older children in school if they are successful, and sending younger children who are not academically as promising to work. For low-income families in Turkey, school success may be an important factor in determining which child is sent to work.

Qualitative and quantitative studies that take an ecological perspective (Bronfebrenner, 1994) to examine the interaction between the structural, family, and individual levels are needed. While quantitative studies can provide a systematic understanding of patterns of interactions across the structural, family, and individual levels, qualitative studies can offer insights to how families negotiate different variables across these levels. An ecological perspective is needed both to better understand the dynamics that contribute to child labor and to envision and implement comprehensive policies and interventions to curb this problem. Research should also focus on protective factors that minimize the occurrence of child labor in high-risk populations. Protective factors, once identified, can be integrated into preventive and intervention efforts targeting this population. Existing interventions should be evaluated using pre and posttest designs to assess their effectiveness.

Policy and practice implications

Interventions and policies should be closely informed by research and target multiple levels. As poverty is a major determinant of child labor, government efforts to eradicate poverty should continue. As a means to increase household income, vocational trainings can be offered to parents who migrate from rural areas to cities and do not have the necessary skills to hold jobs in an urban environment. Microcredit programs can also be considered as potential resources. Some researchers caution against microcredit programs with this population because family business may increase the risk of child labor (Dayıoğlu, 2006). Thus, official agreements can be signed between microcredit providers and users not to use child labor. Social workers can also lobby to create developmentally appropriate paid summer internships for older children where they can learn skills such as computer use. Local governments can partner with middle and high schools in low-income communities to create paid internship opportunities for children with regular school attendance. The availability of such internships may also encourage families to send their children to school.

The majority of working children engage in agricultural work in rural areas and in small establishments in urban areas, which are less likely to be scrutinized by the government. Thus, social workers should advocate for a wider monitoring of child labor.

At the family level, interventions can focus on informing parents on the importance of education, consequences of child labor, and the link between education and social mobility. Similarly, more work is needed with communities where child labor is widespread in order to change attitudes about child labor. More community centers and free health clinics should be established in low-income neighborhoods in order to provide families with resources and services. Efforts should also focus on raising public awareness about child labor and discouraging consumption of products that have relied on child labor. Under close monitoring, employers can be encouraged to mark their products so as to indicate that children were not involved in its production.

Conclusion

Despite significant efforts to understand the complexity of child labor and to eliminate this practice, Turkey needs to systematically expand its research and intervention efforts in this area. Child labor in Turkey cannot be viewed in a social vacuum divorced from contributing political and socio-economic conditions such as urban poverty, migration, and traditional gender relations in the family and society. A more contextual, multilayered approach to child labor will allow for more effective interventions at micro-, meso-, and macro-levels.

Footnotes

Acknowledgements

The author wishes to express her gratitude to Teresa Ostler, Professor of Social Work at the University of Illinois at Urbana-Champaign; and Robin L. Jarrett, Professor of African American Studies and Human and Community Development at the University of Illinois at Urbana-Champaign for their valuable guidance, which contributed to this article.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author biography

Ozge Sensoy Bahar, MSW, is a PhD student at the School of Social Work, University of Illinois at Urbana Champaign, USA.

References

Acar

. (2010) ‘Children Working in the Streets and in the Service Sector in Turkey: A Comparative Study’, International Journal of Human Sciences 7(1): 1009–19.

Acaroğlu

. (2010) ‘An Empirical Approach about Some Important Features of Child Labor in Turkey’, Business and Economic Horizons 3(3): 135–46.

Akın

. (2009) ‘Working Conditions of the Child Worker in Turkish Labor Law’, Employee Responsibilities and Rights Journal 21: 53–67.

Akkoyunlu-Wigley

Wigley

(2008) ‘Basic Education and Capability Development in Turkey’, in Nohl

Akkoyunlu-Wigley

Wigley

(eds) Education in Turkey, pp. 271–98. New York and Münster: Waxmann Publishing.

Akşit

Karancı

Gündüz-Hoşgör

(2001) ‘Working Street Children in Three Metropolitan Cities: A Rapid Assessment’, available online at: http://www.oit.org/public/libdoc/ilo/2001/101B09_343_engl.pdf.

Altıntaş

. (2006) ‘Child Labour in Turkey from a Global Perspective: A Case Study of Working Children in Istanbul’. Unpublished Thesis.

Altuntaş

. (2003) Mendile, Simite, Boyaya, Çöpe….Ankara Sokaklarında Çalışan Çocuklar. İstanbul: Iletisim.

Altuntaş

. (2006) ‘Displaced Persons through the Lens of Social, Economic, and Class Dynamics’, TESEV International Symposium on ‘Internal Displacement in Turkey and Abroad: International Principles, Experiences and Policy Proposals, Istanbul, Turkey, 4–5 December.

Bakırcı

. (2002) ‘Child Labour and Legislation in Turkey’, The International Journal of Children’s Rights 10: 55–72.

10.

Barnes

. (2008) ‘A Case Study of the Working Poor Single Mother Experience: An Analysis of the Structure versus Agency Discourse’, Journal of Poverty 12(2): 175–200.

11.

Blunch

Verner

(2000) ‘Revisiting the Link between Poverty and Child Labor: The Ghanaian Experience’, World Bank Policy Research Working Paper No. 2488, World Bank, Washington, DC.

12.

Bronfenbrenner

. (1994) ‘Ecological Models of Human Development’, in Husen

Postlethwaite

T.N.

(eds) International Encyclopedia of Education, 2nd edn, pp. 1643–7. New York: Elsevier Science.

13.

Buchmann

. (2000) ‘Family Structure, Parental Perceptions, and Child Labor in Kenya: What Factors Determine Who Is Enrolled in School?’, Social Forces 78: 1349–78.

14.

Das

Mukherjee

(2007) ‘Role of Women in Schooling and Child Labor Decision: The Case of Urban Boys in India’, Social Indicators Research 82(3): 463–86.

15.

Dayıoğlu

. (2005) ‘Patterns of Change in Child Labour and Schooling in Turkey: The Impact of Compulsory Schooling’, Oxford Development Studies 33: 195–210.

16.

Dayıoğlu

. (2006) ‘The Impact of Household Income on Child Labor in Urban Turkey’, Journal of Development Studies 42(6): 939–56.

17.

Dayıoğlu

Assaad

(2003) ‘The Determinants of Child Labor in Urban Turkey’, Economic Research Forum Working Paper Series No. 0302, Cairo, Egypt.

18.

Değirmencioğlu

Acar

H.B.

(2008) ‘Extreme Forms of Child Labour in Turkey’, Children & Society 22: 191–200.

19.

Emerson

P.M.

Souza

A.P.

(2008) ‘Birth Order, Child Labor, and School Attendance in Brazil’, World Development 36(9): 1647–64.

20.

Ertürk

Dayıoğlu

(2004) ‘Gender, Education and Child Labour in Turkey’, available online at: http://www.ilo.org/ipecinfo/product/viewProduct.do?productId=346.

21.

Fafchamps

Wahba

(2006) ‘Child Labor, Urban Proximity, and Household Composition’, Journal of Development Economics 79(2): 374–97.

22.

Ferguson

. (2005) ‘Child Labor and Social Capital in the Mezzosystem: Family- and Community-based Risk and Protective Factors for Street-working Children in Mexico’, Journal of Social Work Research and Evaluation 6(1): 101–20.

23.

Göncü

Ozer

Ahioglu

(2009) ‘Childhood in Turkey: Social Class and Gender Differences in Schooling, Labor, and Play’, in Fleer

Hedegaard

Tudge

(eds) Childhood Studies and The Impact of Globalization: Policies and Practices at Global and Local Levels, pp. 67–85. New York: Routledge.

24.

Gündüz-Hoşgör

Smits

(2008) ‘Variation in Labor Market Participation of Married Women in Turkey’, Women’s Studies International Forum 31(2): 104–17.

25.

Hacettepe Institute of Population Studies (1999) Demographic and Health Survey 1998. Ankara, Turkey: Hacettepe Institute.

26.

Hendrickson

M.D.

(2009) ‘Childhood and Child Labor in the Brazilian Amazon’, paper presented at the University of Pittsburgh Latin American Studies Association International Congress, Rio De Janeiro, Brazil.

27.

International Labor Office (n.d.) ‘About Child Labor’, available online at: http://www.ilo.org/ipec/facts/lang--en/.

28.

International Labor Office (2009a) ‘Give Girls a Chance: Tackling Child Labour, a Key to the Future’, available online at: http://www.ilo.org/ipecinfo/product/viewProduct.do?productId=10290.

29.

International Labor Office (2009b) ‘Ending Child Labour: A Comprehensive Review of Turkish Experience’, available online at: http://www.ilo.org/public/english/region/eurpro/ankara/areas/childlabourinturkiyepdf/clreviewturkishexperience.pdf.

30.

Kağıtçıbaşı

. (1998) ‘The Value of Children: A Key Issue to Gender Issues’, International Child Health 9(1): 15–24.

31.

Kağıtçıbaşı

Ataca

(2005) ‘Value of Children and Family Change: A Three-Decade Portrait from Turkey’, Applied Psychology 54(3): 317–37.

32.

Karatay

. (2000) ‘Istanbul’un Sokaklari ve Calisan Cocuklar’, in Sayita

S.U.

Sirin

M.R.

(eds) Istanbul Cocuk Kurultayi Arastirmalari Kitabi, pp. 453–541. Istanbul: Istanbul Cocuklari Vakfi.

33.

Orazem

P.F.

Gunnarsson

(2004) ‘Child Labor, School Attendance and Performance: A Review’. Department of Economics Working Paper Series No. 04001, Iowa State University.

34.

Ray

. (2002) ‘The Determinants of Child Labour and Child Schooling in Ghana’, Journal of African Economies 11: 561–90.

35.

Sensoy Bahar

. (2011) ‘“Education is important but…”: A pilot study on low-income Kurdish migrant mothers’ attitudes and decisions about child labor and education’, Journal of Global Social Work Practice 4(2). Available online at http://www.globalsocialwork.org/vol4no2/Bahar.html

36.

Şen

. (2005) ‘Poverty Alleviation, Conflict, and Power in Poor Displaced Households: A Study of the Views of Women in Diyarbakir’, New Perspectives on Turkey 32: 113–35.

37.

Tor

. (2010) ‘Status of Child Labor in Turkey’, Journal of World of Turks 2(2): 25–42.

38.

Turkish Statistical Institute (2007) ‘Cocuk Isgucu Arastirmasi, 2006’, T.C. Basbakanlik Turkiye Istatistik Kurumu Haber Bulteni 61: 1–3.

39.

Turkish Statistical Institute (2009) ‘2008 Yoksullluk Arastirma Sonuclari’, T.C. Basbakanlik Turkiye Istatistik Kurumu Haber Bulteni 205: 1–2.

40.

World Resources Institute (2003) ‘Population, Health, and Human Being, Turkey’, available online at: http://earthtrends.wri.org/pdf_library/country_profiles/pop_cou_792.pdf.

41.

Yıldız

. (2007) ‘Calisan Cocuklar: “Sorun” mu? “Cozum” mu?’, Gaziantep Universitesi Sosyal Bilimler Dergisi 6(2): 57–66.

42.

Yılmaz

. (2004) ‘Street-Vendor Children in Istanbul: The Visible Facet of Urban Poverty’, in Naumovic

Jovanovic

(eds) Childhood in South East Europe: Historical Perspectives on Growing up in the 19th and 20^th Century, pp. 125–47. New Brunswick, NJ: Transaction Publishers.

43.

Yılmaz

. (2008) ‘Entrapped in Multidimensional Exclusion: The Perpetuation of Poverty among Conflict-Induced Migrants in an Istanbul Neighborhood’, New Perspectives on Turkey 38: 205–34.