key: cord-0995243-0w02f9q2
authors: Beydoun, Zahraa; Abdulrahim, Sawsan; Sakr, George
title: Integration of Palestinian Refugee Children from Syria in UNRWA Schools in Lebanon
date: 2021-01-04
journal: J Int Migr Integr
DOI: 10.1007/s12134-020-00793-y
sha: 1a92ef91b885d2d827a6c2b79b07fe3e638d6144
doc_id: 995243
cord_uid: 0w02f9q2

Following the Syrian conflict that began in 2011, Lebanon received more than one million refugees including 44,000 Palestinian refugees from Syria (PRS). PRS children were integrated into existing schools run by the United Nations Relief and Works Agency (UNRWA). Despite efforts by UNRWA to integrate the newly displaced into its services in Lebanon, only 58% of 6–18-year-old PRS children were enrolled in school in 2014. Informed by ecological systems theory, we examined the role of parental characteristics in determining school enrollment among PRS children following displacement into Lebanon. Utilizing data from the 2014 UNRWA Vulnerability Assessment (N = 12,378 6–18-year-old children), we specified crude and adjusted logistic regression models to predict child school non-enrollment including a set of variables on head of family characteristics (gender, age, education, and presence/absence of chronic disease) and post-displacement household characteristics (crowding, wealth, camp residence, region, and type of dwelling). The results show that, adjusting for household characteristics, a child living in a family whose head has secondary education or higher is more likely to be enrolled in school compared to one living in a family headed by someone with less than secondary education. Parental education remains the strongest predictor of child school enrollment despite displacement-related household disadvantage. To break the cycle of intergenerational educational disadvantage, it is critical for UNRWA to proactively design school retention programs for PRS children living in families whose head had limited access to education. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12134-020-00793-y.

There are currently 79.5 million forcibly displaced people worldwide of whom 26 million are refugees who have crossed an international border (UNHCR 2020). Refugees are persons who have fled their homes and who are unable to or unwilling to return due to a well-founded fear of persecution. An estimated 40% of forcibly displaced persons are children below age 18 whose education is disrupted and who experience limited access to schools in host countries as well as a number of barriers related to language of instruction and discrimination. Education is a basic human right that is protected in various international legal frameworks, including the Convention on the Rights of the Child (OHCHR 1995) . States are responsible to guarantee the right to education to all children, including refugee children, on their territory (UNESCO 2017) . Further, the 1951 Convention on the Status of Refugees states that refugee children should be accorded the same right to education as nationals (UNHCR 2010) . The Sustainable Development Goal 4 (SDG4) on education cannot be achieved without ensuring the right of refugee and internally displaced children to attend school.

The benefits of education for children in displacement are numerous. Schools provide safe spaces and help children maintain a sense of normalcy following a traumatic experience. Since 2012, UNHCR has been promoting the model of integrating refugee children within national education systems and moving away from the model of building separate, refugee-only schools inside or near refugee settlements (Dryden-Peterson 2016 ). Yet, integrating refugee children into the educational system of host countries presents with major challenges related to logistics, resources, and educational policies (Culbertson and Constant 2015) . These challenges are particularly pressing in low-and middle-income countries (LMICs) of first asylum that host the majority of refugees worldwide (86%) and that have stretched resources and weak public education infrastructures.

Since the beginning of the Syrian crisis in 2011, the influx of large numbers of Syrian refugees into neighboring countries has created an education crisis, placing a heavy demand on this sector in Lebanon, Jordan, and Turkey (Culbertson and Constant 2015) . In Lebanon in particular, a country where the public education system has historically been weak and under-funded (Nahas 2011) , the integration of refugee children into schools has struggled to keep pace. In 2015, Lebanon, a country of 4.5 million, hosted around 1.1 million Syrians registered with the United Nations High Commissioner for Refugees (UNHCR), the majority of whom (75%) are women and children, and 38% are school aged (5-17 years old) (UNHCR 2017). In May 2014, the Lebanese Ministry of Education and Higher Education (MEHE) has implemented the RACE strategy (Reaching All Children with Education) to enroll Syrian refugee children in Lebanese public schools in an afternoon shift. Due to a gap between refugee education policies and the implementation of these policies on the ground in Lebanon (Buckner, Spencer, Cha 2017) , around 50% of school age Syrian children were not enrolled in an educational institution during the 2017-2018 academic year (UNHCR 2018).

In addition to Syrian refugees, Lebanon also received 44,000 Palestinian refugees from Syria (PRS) between 2012 and 2014. These are the descendants of Palestinians who were displaced into Syria in 1948 as a result of the establishment of the state of Israel. The influx of PRS into Lebanon increased in 2013 following the destruction of one of the largest refugee camps around Damascus; around half of this population moved into existing Palestinian refugee camps in Lebanon (Abdulrahim and Harb 2015) . Like other Palestinian refugees, PRS receive education and health services from the United Nations Relief and Works Agency (UNRWA), not from UNHCR. As such, PRS children were integrated into existing UNRWA schools, which are accessible to all Palestinians, following displacement.

UNRWA and Access to Education for Palestinian Refugees UNRWA was established by the United Nations (UN) General Assembly in 1949 and commenced its operations in 1950. Unlike UNHCR which is mandated to provide all refugees with protection and durable solutions, UNRWA primarily provides health and education services to Palestinian refugees in five fields of operation: West Bank, Gaza, Syria, Jordan, and Lebanon (Bocco 2009 ). Within a couple of decades after its establishment, UNRWA increased secondary school graduation among Palestinian refugees and achieved near gender parity in educational enrollment (Rosenfeld 2009 ). This progress, however, eroded over the years due to a number of factors, mainly dwindling financial resources, as the organization relies primarily on voluntary contributions of donor countries, and exclusionary host country policies (Bocco 2009; Roberts 2010) . In Lebanon, for example, policies that exclude Palestinian refugees from the labor market have created disillusionment with education as a pathway to social mobility (Abdulrahim and Khawaja 2011) .

UNRWA operates 65 elementary, preparatory, and secondary schools that educate Palestinian refugee children in Lebanon (PRL). While 97% of PRL elementary students are enrolled in school, only 84% of preparatory and 61.2% of secondary students are enrolled (Chaaban et al. 2016) . Following the Syrian humanitarian crisis and the displacement of PRS into Lebanon, the agency integrated PRS children into its existing schools (UNRWA 2015) . Shortly after their arrival, UNRWA implemented an "Education in Emergencies" strategy designed to integrate PRS children into existing schools and provide them with psychosocial support to mitigate the impact of war trauma and create a sense of normalcy (UNRWA 2015) . The strategy was implemented through three progressive phases whereby a child was first provided with one week of psychosocial activities, followed by three weeks of learning skills, followed by a period of transition before integration in school (UNRWA 2015) . Yet, despite UNRWA's efforts, a vulnerability assessment conducted in 2014 revealed that, overall, only 58% of school aged children were enrolled in school and that enrollment drops precipitously after age 12 for both boys and girls (Abdulrahim and Harb 2015) .

Ecological models, which focus on how individuals interact with their physical and social environment, have been widely used as theoretical frameworks in sociology, behavioral sciences, education, and public health (Darling 2007; Sallis, Owen, Fisher 2015) . One of the earliest ecological systems, theories was advanced by developmental psychologist Bronfenbenner (1979) who proposed that research on child development should examine the interaction between the child and his or her environment. The environment is conceived of as a set of nested structures that surround a developing child and exert their influence at multiple levels: the micro-, meso-, exo-, and macro-system levels. Bronfenbrenner (1979 Bronfenbrenner ( , 1986 defined the micro-system as the pattern of relations the child experiences in a setting that has unique physical and material features. The meso-system, on the other hand, comprises the interrelations between the child's micro-system and other settings in which the child interacts directly, such as the home and school. Finally, the exo-system refers to interactions in settings in which the child participates indirectly while the macro-system describes the ideological structure of the society in which the child lives.

According to ecological systems theory, the family and school are critical settings in which child development takes place at the meso-level and influence his or her educational outcomes. Bronfenbrenner (1986) suggested that examining school experiences and family-school interactions are prerequisites for better-informed research on child development. Family characteristics, including size, housing conditions, and parental educational level, have been defined by Johnson (1994) as components of the "meso-risk" that can affect a child's educational outcomes. Parental education in particular exerts a strong influence on children's educational outcomes as has been demonstrated in numerous studies in LMICs and high-income countries on the intergenerational transmission of educational inequalities (Connelly and Zheng 2003; De Broucker and Underwood 1998; Heineck and Riphahn 2009; Hertz, Jayasundera, Piriano, Selcuk, Smith, Verashchagina 2007; Huisman and Smits 2009 ). Mothers' and fathers' educational variables capture the availability of human capital in the family that can protect a child and lessen his or her risk of educational disruptions even under unpredictable negative circumstances that affect an entire population (e.g., war and displacement).

The literature highlights multiple pathways through which parental education influences children's educational mobility. The two main pathways are parental engagement and availability of tangible resources. Educated parents have a higher regard for education, value it as an outcome for their children, and engage more with their children in education-promoting activities (Black and Devereux 2011) . Moreover, educated parents are more likely to have tangible resources needed to militate against their children dropping out of school in order to work or provide for the family (Breen and Goldthorpe 1997; Huisman and Smits 2009 ). However, the positive association between parental education and children's school integration and performance may not always hold among population groups disadvantaged by income, race, or citizenship categories. For example, studies have shown that the correlation between parental education and child educational attainment is stronger in wealthy compared to poor families (Huang 2013) and is weak among rural and migrant families (Golley and Kong 2013) .

There is general acceptance that educational disruptions are an inevitable consequence of war and displacement. Nonetheless, it is worth exploring whether factors such as the availability of schools in the first country of asylum or parental and household characteristics can buffer against the risk of children's school non-enrollment. According to ecological systems theory, these factors operate at what Bronfenbenner (1979) defined as the meso-system that surrounds a child and influences his or her development. In the present study, we examined the predictors of school nonenrollment among PRS refugee children in UNRWA schools 1 year after they became displaced in Lebanon in 2014. Specifically, we tested whether the socio-demographic characteristics of the parents as a meso-system determinant, mitigate against school non-enrollment following displacement (a macro-system characterized by sudden loss of resources and extreme uncertainty). A study on PRS in Lebanon in which the school context can be held constant, as UNRWA schools are universally accessible to all Palestinian children, provides an opportunity to test whether pre-displacement parental socioeconomic position buffer against the deleterious impact of displacement on child school enrollment.

In this study, we utilized data from a vulnerability assessment conducted by UNRWA on all PRS families who arrived in Lebanon between 2012 and 2014. The data were collected during the summer of 2014 on 12,735 families (44,227 individuals) through utilizing a questionnaire adapted from the VASyR (Vulnerability Assessment of Syrians) questionnaire developed by UNHCR and the World Food Program. The questionnaire gathered information on housing conditions, education, health, income and expenditures, household assets, and food security. The analyses presented in this paper focus on the educational outcomes of 6-18-year-old children (N = 12,378). As many families did not have a child in the 6-18 age range, the total number of families included in the analyses is 5972.

Our dependent variable is school non-enrollment, deduced from a question in the questionnaire on whether the child was enrolled or not enrolled in school at the time of the assessment. We analyzed two sets of independent variables-characteristics of the head of the family and household characteristics. The head of the family characteristics are gender (woman/man), age (18-29, 30-39, 40-49, > 50) , presence of a chronic illness (yes/no), and educational level (dichotomized as < secondary versus ≥ secondary). Household characteristics include region of residence in Lebanon (Beirut, Beqaa, Saida, Tripoli, Tyre), camp residence (living inside vs outside a Palestinian refugee camp in Lebanon), and crowding index (calculated based on the minimum standard space required for healthy living which is 3.5 m 2 per individual). Other household characteristics include type of residence in 3 categories (independent dwelling; informal dwelling such as room in an apartment, unfinished shelter, or warehouse/garage; and tent, barracks, or hut). A wealth index was constructed using 20 items on household facilities (e.g., type of toilet and access to sufficient water) and assets (e.g., refrigerator, winter clothes, and blankets). Principal components analysis was used to create wealth quintiles (highest, high, medium, low, lowest). Information on the construction of the wealth quintiles is presented in the supplementary document.

Descriptive analyses were first carried out to explore the predictors of school nonenrollment by child sex and age, and to describe the head of the family and household characteristics. Following, we conducted multivariable analyses to test the hypothesis that head of family characteristics remain strong determinants of child school nonenrollment controlling for household characteristics. We specified crude and adjusted logistic regression models to predict child school non-enrollment. Model 1 included four variables that capture the characteristics of the head of the family (gender, age, education, and chronic illness), adjusted for sex and age of child. Model 2 (full model) included five household characteristics-region of residence, camp/non-camp residence, type of residence, wealth, and crowding, adjusted for head of family characteristics included in model 1 and sex and age of child. Crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. All statistical analyses were conducted using SPSS version 22. Table 1 presents the sex and age distribution of all 6-18-year-old children in the study (N = 12,378), socio-demographic characteristics of the head of the family, and household characteristics. The sample has an almost equal proportion of male (48.5%) and female (51.5%) PRS children. Slightly more than half of the children (51.9%) are 6-11 years old, 20.4% are 12-14 years old, and 27.6% are 15-18 years old. More than three quarters of PRS children live in families headed by a man (76.9%). Further, 44.2% of family heads are 40-49 years old, the majority (75.1%) have less than secondary education, and 34.2% have a chronic illness. As to household characteristics, most PRS children live in crowded households (83.8%) and almost equal numbers live inside (50.6%) and outside (49.4%) camps. Moreover, only 14.5% live in Beirut while the majority live outside the capital (Saida 33.5%, Tyre 18.9%, Beqaa 16.7%, and Tripoli 16.4%). Finally, the majority of children live in an independent dwelling (79.5%); 16.3% live in an informal dwelling; and only 4.2% live in a tent, barracks, or hut.

In 2014, 58% of 6-18-year-old PRS children were enrolled in school. Only half of 6year-old PRS children were enrolled in school with no difference between boys (50.7%) and girls (51.5%); see Fig. 1a and b. Between age 7 and 12, school enrollment for both girls and boy exceeded 70%. Enrollment, however, dropped precipitously after age 12 and showed differentiation by sex favoring girls; at age 17, 26.7% of girls and only 14.5% of boys were enrolled in school.

Crude logistic regression results (Table 2) show that boys were at a higher risk of school non-enrollment compared to girls (OR 1.21, 95% CI 1.13-1.30) and, as expected, the odds of non-enrollment were almost eight times higher for 15-18-yearold children compared to the 6-11 age group (OR 7.71, 95% CI 7.01-8.48). As to characteristics of the head of the family, crude results show that children who lived in families headed by women have slightly higher odds of school non-enrollment (OR 1.18, 95% CI 1.08-1.29). Moreover, children in families headed by someone who is younger than 29 or older than 50 years of age, who has less than secondary education, and who has a chronic illness exhibited higher odds of school non-enrollment. All household characteristics predicted child school non-enrollment in crude models. As expected, poverty (low and lowest wealth quintiles) and living in poor quality housing (informal dwelling and tent/barracks/hut) predicted higher odds of school nonenrollment. Living outside camps was predictive of higher odds of school nonenrollment (OR 1.32, 95% CI 1.23-1.42) and living outside Beirut was protective against school non-enrollment (all ORs for the four regions were below 1 and statistically significant). These findings are perhaps due to the fact that UNRWA schools in the capital Beirut are crowded and located outside camps due to limited space. In the four other regions, UNRWA schools are located inside or on the outskirts of camps and are therefore more accessible to children who live inside camps. Oddly, living in a crowded household did not have an effect on school non-enrollment.

In model 1, we examined the effect of the head of the family characteristics, adjusting for child sex and age only, on school non-enrollment. Children living in family headed by a woman or someone with a chronic illness lost statistical significance in the adjusted model. In model 2, we examined the effect of the characteristics of the head of the family on school non-enrollment, after adjusting for both household characteristics that reflect post-displacement socioeconomic conditions and child sex and age. Multivariable logistic regression results showed that living in a family headed by someone with less than secondary education remained one of the strongest predictors of school nonenrollment (OR 1.71, 95% CI 1.55-1.89). Moreover, the odds of school nonenrollment in the adjusted model 2 remained higher for children in the low and lowest wealth quintiles (OR 1.44, 95% CI 1.27-1.65 and OR 1.98, 95% CI 1.72-2.28, respectively); those living outside camps (OR 1.48, 95% CI 1.35-1.62); and those living in a tent, barracks, or hut (OR 1.89, 95% CI 1.52-2.35).

War and forced displacement disrupt the lives of whole populations, particularly children who are the most vulnerable social group. Children's access to education in the first country of asylum is determined by factors that operate at multiple levels, host country policies, the availability of schools, and the mesolevel factors that surround a child including the human capital of their parents. Our results showed that whereas household characteristics that reflect poor family economic conditions in the host country are important determinants of child educational disruptions due to displacement, the characteristics of the head of the family remained strongly predictive of a refugee child's access to schooling. Children who lived in a family headed by a woman or whose family head had less than secondary education were at higher risk of school non-enrollment, irrespective of such household characteristics as camp residence, housing quality, or wealth. Even under conditions of mass displacement, when families lose their property, livelihood, and social protections, parental education continues to play a critical role in transmitting advantage (or disadvantage) to children and to protect them from school non-enrollment following forced displacement.

These findings are consistent with ecological systems theory, which advanced that parental influence is a main meso-system determinant of child development. Johnson (1994) extended the definition of meso-risk as an outcome of the social and economic characteristics of the family's housing conditions and parental education level. The available literature on the educational experiences of refugee children has primarily focused on macro-level factors that determine their integration in resettlement countries in North America and Europe. For example, at a macro-level, resettled children in the United States (US) whose refugee parents come from high-income countries were more likely to be enrolled in school compared to those whose parents come from wardevastated and low-income countries (Hooper, Zong, Capps, and Fix, 2016) . Studies that have examined determinants of child school enrollment in countries of first asylum also focused primarily on macro-level factors related to policies and educational systems (Dryden-Peterson 2016).

Our study findings highlight the importance of examining the meso-level factors that facilitate refugee children's access to schooling, particularly parental education which in many cases is the only capital refugees carry with them when they flee war and violence. Despite the presence of UNRWA schools in Lebanon which proactively sought to integrate PRS children after their arrival, parental education, a measure of the human capital a family carried with them post-displacement, continued to play a role in determining a child's risk of non-enrollment. PRS parents with secondary education or higher were able to mobilize resources even in uncertain circumstances to ensure their child's integration in school. Although PRS parents are expected to be familiar with the UNRWA system as their children attended its schools in Syria before displacement, those with less than secondary education may have been less able to navigate the system in Lebanon. Given the role of education in enhancing human capabilities and preparing individuals to better navigate their environment in uncertain situations, educationally disadvantaged PRS parents may have found it challenging to adapt to the UNRWA education system in Lebanon.

School non-enrolment can set in motion a cascade of events that put PRS children on a negative social trajectory, particularly in the context of limited work opportunities for Palestinian refugees in Lebanon coupled with the country's high cost of living compared to Syria. Boys may turn to work in the informal economy, where they experience exploitation and oftentimes exposure to harmful substances, while girls may be pushed to marry early in pursuit of economic security. Our findings on the association between parental education and the risk of school non-enrollment among PRS children in Lebanon highlight the importance of instituting targeted programs to proactively integrate children who are most at risk in UNRWA schools. Some strategies that have been successfully implemented in poverty settings, although not among refugees, are to provide cash to poor families on condition that they keep their child enrolled in school (Schultz 2004; De Janvry et al. 2006) . However, implementing a conditional cash strategy would be challenging for UNRWA as the organization is currently facing unprecedented budget cuts that threaten its ability to carry on with its mandate and to continue to provide education to Palestinian refugees. In August 2018, the US, the largest contributor to UNRWA's core program budget, announced that it will withdraw all of its funding (Beaumont and Holmes 2018) . Despite the dramatic turn of events, UNRWA schools opened their doors in September 2018 and 2019 to Palestinian students including PRL and PRS students in Lebanon (Anderson and Zaatari, 2018 ). Yet, the future of these children, who do not have access to Lebanese public schools, remains unclear particularly in light of the COVID-19 global pandemic and the severe economic downturn in Lebanon.

Although the present study contributes knowledge on the determinants of school enrollment of PRS refugee children, it suffers from a couple of methodological limitations mainly due to its reliance on secondary data gathered by UNRWA to assess vulnerability in general. First, our analysis is limited by the crude nature of the education outcome measure available-whether a child was enrolled in school or not at the time of the assessment. This measure does not capture displacement-related school dropout, meaning whether a child was enrolled in school in Syria but dropped out in Lebanon as a consequence of displacement. While this may not influence our results for young children whose school enrollment is high, out of school children that are older than 12 years of age are a mix of those who were not enrolled in school before displacement and those who were but dropped out after displacement. Moreover, the vulnerability assessment data we analyzed did not capture school achievement, an intermediate outcome that is oftentimes examined in studies on children's risk of school dropout. A second main limitation is the study's focus on a specific refugee group and a UN agency that already had established schools in the host country, which is not the situation of most refugees around the world. Although the context of PRS in Lebanon differs from that of Syrian or other refugee groups, given the presence of UNRWA, the findings on the link between head of family education and the risk of school nonenrollment is potentially generalizable to other displacement settings. These limitations notwithstanding, the present study adds to the literature that has focused on the protective role of parental education on their children's school enrollment and achievement in LMICs and highlights important implications for the local context. Following displacement and sudden loss of livelihood, the amount of human capital parents carry with them may be the most critical meso-level determinant of their children's adaptation and social mobility in the host country.

Profiling the vulnerability of Palestine refugees from Syria living in Lebanon

The cost of being Palestinian in Lebanon

UNRWA schools open but future far from certain

US confirms end to funding for UN Palestinian refugees

Recent developments in intergenerational mobility

Explaining educational differentials: towards a formal rational action theory

The ecology of human development experiments by nature and design

Ecology of the family as a context for human development: research perspectives

UNRWA and the Palestinian refugees: a history within history

Between policy and practice: the education of Syrian refugees in Lebanon

Survey on the Socioeconomic Status of Palestine Refugees in Lebanon

Determinants of school enrollment and completion of 10 to 18 year olds in China

Education of Syrian refugee children: managing the crisis in Turkey

Ecological systems theory: the person in the center of the circles

Refugee education in countries of first asylum: beaking open the black box of pre-resettlement experiences. Theory and Research in Education

Intergenerational education mobility: an international comparison with a focus on postsecondary education

Can conditional cash transfer programs serve as safety nets in keeping children at school and from working when exposed to shocks

Inequality in intergenerational mobility of education in China

Intergenerational transmission of educational attainment in Germanythe last five decades

The inheritance of educational inequality: international comparisons and fifty-year trends

Young children of refugees in the United States: integration successes and challenges

Intergenerational transmission of educational attainment: the role of household assets

Effects of household-and district-level factors on primary school enrollment in 30 developing countries

An ecological framework for conceptualizing educational risk

Financing and political economy of higher education: the case of Lebanon

Office of the High Commissioner for Human Rights (OHCHR)

Palestinians in Lebanon: Refugees living in long-term displacement

From emergency relief assistance to human development and back: UNRWA and the Palestinian refugees

Ecological models of health behavior

United Nations High Commissioner for Refugees (UNHCR). (2010). Convention and protocol relating to the status of refugees

United Nations High Commissioner for Refugees (UNHCR)

United Nations High Commissioner for Refugees (UNHCR). (2020). Figures at a glance

The United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) (2015)

The online version contains supplementary material available at https://doi.org/ 10.1007/s12134-020-00793-y.Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.