No Lost Generation: Supporting the School Participation of Displaced Syrian Children in Lebanon

Abstract

This study documents the impact of a cash transfer programme – known as the No Lost Generation Programme (NLG) and locally as Min Ila (‘from to’) – on the school participation of displaced Syrian children in Lebanon. An initiative of the government of Lebanon, the United Nations Children’s Fund (UNICEF), and the World Food Programme (WFP), the programme provided cash for the benefit of children enrolled in afternoon shifts at public primary schools. It was designed to cover the cost of commuting to school and to compensate households for income forgone because children were attending school instead of working. Commuting costs and forgone income are two critical barriers to child school participation. The analysis relies on a geographical regression discontinuity design to identify the impact halfway through the first year of programme operation, the 2016/2017 school year. The analysis finds substantive impacts on school attendance among enrolled children, which increased by 0.5 days to 0.7 days per week, an improvement of about 20 per cent relative to the control group. School enrolment among Syrian children rose rapidly across all Lebanon’s governorates during the period of the evaluation, resulting in supply-side capacity constraints that appear to have dampened positive enrolment impacts.

Acknowledgements

The authors thank UNICEF Lebanon for the opportunity to carry out this study and for financial and technical support. They would like also to recognise the many individuals and organisations without which it would not have been possible to complete this study. They thank the Programme Management Unit of the Reaching All Children with Education Project in Lebanon’s Ministry of Education and Higher Education (MEHE), including Sonia Khoury, Bane Khalife, Georges Ghassan, and Maroun Hobeika, the UNICEF Middle East and North Africa Regional Office, the UNICEF Office of Research - Innocenti, the United Nations Refugee Agency (UNHCR), the World Food Programme (WFP) (Jordi Renart and Soha Moussa), and Statistics Lebanon for technical or financial support at various stages of the project.

Special thanks go to Violet Speek-Warnery, Georges Haddad, Sharlene Ramkissoon, Juan Santander, Maxime Bazin, Abed Alrahman Faour, Louisa Lippi, and Georges Fares at UNICEF Lebanon for their technical support during the research and during the design and implementation of the fieldwork. The authors thank Rabih Haber, Hanane Lahoud, and Elie Joukhadar of Statistics Lebanon and Mohammed Elmeski of the American Institutes for Research for their support during the implementation of the fieldwork. They gratefully acknowledge feedback on the initial evaluation design by members of the UNICEF Office of Research Review Group, including particularly Sudhanshu Handa and Amber Peterman. They also thank Gustavo Angeles, Amber Peterman, and Dan Sherman for detailed comments on an earlier draft of this report.

The acknowledgements would be incomplete without mentioning the team of able research assistants. Kevin Kamto of the American Institutes for Research provided invaluable data support, and the project manager, Mariela Goett, helped coordinate and support all activities. The authors also acknowledge the input of the team of enumerators and supervisors at Statistics Lebanon whose dedication during data collection ensured that the data collected were of high quality.

The patience shown by Syrian refugee households, community leaders, and community members during interviews is also gratefully acknowledged. It is hoped that the insights derived from the information they provided will translate into valuable support for their communities.

The data used in this study are currently not in the public domain. The authors will make any information about data release available online. The findings, interpretations and conclusions expressed in this paper are those of the authors and do not necessarily reflect the policies or views of UNICEF.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Puri et al. (2017) note that, although rigorous impact evaluation methods can be helpful in learning about the effectiveness of interventions following humanitarian emergencies, such evaluations are more difficult to implement in the aftermath of humanitarian crises, where ethical concerns sometimes prevent the use of control or comparison groups. For this reason, only a few studies have examined the impact of cash transfers in humanitarian contexts in a rigorous manner.

2. See Education (database), United Nations Educational, Scientific, and Cultural Organisation, Institute for Statistics, Montreal, http://data.uis.unesco.org/; EdStats: Education Statistics (database), World Bank, Washington, DC, http://datatopics.worldbank.org/education/.

3. When the present study was designed, the value of the monthly WFP payment was US$27; the value of the monthly Multipurpose Cash Assistance Programme benefit was US$174, and the value of the winter subsidy was US$100–US$147.

4. US$20 is the average expenditure per child on the Caritas-UNICEF bus services.

5. Information on the reason for non-attendance is sought during the field visit, and household members (children) may be referred for complementary services offered by the government, UNICEF, or other agencies that address non–income-related constraints, such as the need for psychosocial support or difficulty keeping up with the Lebanese curriculum. In this way, the cash transfer programme can contribute to an integrated package of support. This process had not yet been launched at the time of the follow-up data collection reported in this study.

6. If the list provided by UNHCR contained fewer than 90 households, the evaluation team included additional neighbouring cadastres as necessary in an attempt to obtain a list of at least 90 households in the vicinity of each school.

7. To find and reach the threshold of 20 households per cadastre needed for the study interviews, the survey teams typically required lists of more than 50 households per cadastre.

8. The target sample was the 20 households per cadastre, for a total of 1,480 households. However, 40 households either did not meet the eligibility criterion of the presence of a child 5–14 years of age or could not be interviewed.

9. During the 2015/2016 school year, Caritas and UNICEF ran a school bus programme that provided transport to school for children living far from a school. The programme was stopped in the governorates of Akkar and Mount Lebanon during the 2016/2017 school year, but continued in the rest of the country. However, because the bus service is available only to children living 2.5 kilometres or more from the nearest school, the presence or absence of the bus service is not expected to confound the results presented in this study.

10. Information on school enrolment is missing on 1.5 per cent of the children in the sample.

11. Per capita income is calculated by summing household income from wages, self-employment, cash from friends, cash from support agencies, food aid, remittances, and the sale of assets over the previous 30 days and then dividing by household size.

12. School enrolment has an inverted U-shaped relationship with age. School enrolment peaks roughly at the age of 10 (not displayed).

13. New afternoon shifts were opened in response to the increased demand in other parts of the pilot governorates; however, this change was not captured by the study because the sample was selected from those households living in the vicinity of afternoon shift schools that already existed.

14. Baseline statistics on attendance were not available during the week prior to the interview because the baseline was conducted during the summer break. Thus, changes in attendance among these groups cannot be measured. It is only possible to determine how these groups compared with each other at follow-up.