JUNE 2021 COX’S BAZAR PANEL SURVEY: RAPID FOLLOW-UP ROUND 2 STATUS OF EDUCATION AMONG SCHOOL-AGED CHILDREN IN COX’S BAZAR This is the third in a series of briefs to disseminate findings from high-frequency rapid follow-ups on the CBPS sample, by the Poverty and Equity GP of the World Bank. CBPS EDUCATION BRIEF 2 EDUCATI ON COVI D-19 CHIL D RE N This brief summarizes findings from rapid welfare tracking surveys in Cox’s Bazar. Two rounds of tracking surveys were implemented via phone interviews in 2020 to monitor the impacts of the COVID-19 crisis on labor markets, wages, and household coping strategies. The first round was conducted during the COVID-related lockdowns in April-May 2020. A second round was conducted from October-December 2020 (roughly 6 months after the government-imposed lockdowns), and also included a short module on education. The brief recaps key findings from the 2019 baseline on education, followed by updated findings on the status of school- aged children from the 2nd round of the rapid follow-up surveys in light of COVID-19 induced school closures and disruptions. These rapid phone surveys are built on the Cox’s Bazar Panel Survey (CBPS), which is a multi-topic survey that focused on socio-economic outcomes and access to services. The baseline CBPS survey, implemented in March-August 2019, was designed to be representative of the recently displaced Rohingya population (displaced after August 2017) in Cox’s Bazar and the host community. Within the host community, the survey was further stratified into high ex- posure (HE, within 3 hours walking distance of a Rohingya camp) and low exposure (LE, more than 3 hours walking distance from a Rohingya camp) areas within the district. The overall sample size of the CBPS baseline was 5020 households, split roughly equally across Rohingya camps and host communities, and within the latter, equally among HE and LE areas. In this second tracking survey, 3,438 households originally surveyed in the baseline were covered. KEY MESSAGES: 1. Children in high exposure areas demonstrate gains in education access since 2019 despite COVID-19 closures, possibly due to the heightened focus of humanitarian efforts on alternative access to learning. At baseline in 2019, HE areas had higher dropout rates particularly from secondary school, and a higher share of out-of-school children, compared to LE areas. By late 2020, high exposure host households continued to report higher rates of school dropouts but also reported higher shares of school-aged children remaining engaged in educational activities despite school closures. 2. 3. Beyond COVID closures, financial constraints remained a key Despite COVID-induced closures in educational programs in factor behind dropouts, in addition to gender-specific con- the Rohingya camps, there has been an increase in education- straints. Half of the school aged children in the host communi- al engagement among teenage children. This may reflect in- ty reported not studying due to COVID closures in 2020, with low creased efforts by UNICEF in ensuring basic grade competen- exposure children reporting higher inactivity than high exposure cies in preparation for the GoB decision to expand the scope counterparts. Financial constraints remain the most pervasive of education for Rohingya children to secondary equivalent non-COVID related issue for all school-aged children, potentially levels under the Myanmar curriculum. While rates of current made worse by COVID-induced labor market shocks. Households, education engagement among children 5-12 years old have seen particularly in HE regions, reporting a lower number of income declines (possibly due to COVID-19 closures), the 13-16 age group sources compared to baseline were significantly more likely to re- reported more than double the rates of educational engagement port financial constraints and the need to start working as reasons compared to 2019. for children not to study anymore. CBPS EDUCATION BRIEF 3 IMPLICATIONS FOR POLICY AND PROGRAMMING: 1. Faced with overwhelming concerns that school closures due to COVID would have wide-rang- ing and long-term implications on youth literacy and education continuity, the humanitarian education sector strengthened its focus on maintaining and expanding access to alternate learning modes, including through home-based caregiver-led education and distance learn- ing. This response to a tangible nation-wide bottleneck was relatively more successful in HE areas, given the higher presence of humanitarian efforts. 2. This success story suggests that it may be worthwhile to invest in alternative learning modes for school-age children at risk of dropping out of school, by bringing schooling closer to home, rather than relying solely on physical attendance in classrooms. Analysis of reasons behind school drop-outs suggest that such strategies may help teenage boys at risk of dropping out flexibly manage school and work responsibilities; and help teenage girls continue their edu- cation in the face of increasing mobility constraints. 3. Bringing education closer to the home may also help the HE host community take advantage of opportunities for increased labor market earnings from tertiary education. Residents of Ukhia and Teknaf face relatively long travel times to the nearest secondary schools. Expand- ing education and learning access could help keep secondary-school age children in school, and increase the pool of young, educated adults who can gain from the increased supply of relatively well-paying jobs in and around the humanitarian effort. CBPS EDUCATION BRIEF 4 ABOUT THE COX’S BAZAR PANEL SURVEY AND HIGH FREQUENCY ROUNDS. Bangladesh’s local economy started experienc- TIMELINES ing impacts of the COVID-19 crisis in early to mid- March 2020, with the first case being reported Baseline Mar-Aug 2019 on 7 March. A full countrywide lockdown was in place from 26 March-28 May 2020. The first round Round 1 Apr-May 2020 of the CBPS high-frequency tracking surveys was conducted within the government lockdowns Round 2 Oct-Dec 2020 (between April-May 2020) and focused on cap- turing key trends in the labor market related to the economic contraction. of school-aged children in studies, reasons for not studying and households’ intention to send Findings from the second round, conducted ap- children back to school post-closures. Findings proximately 6 months following the lockdowns are presented using the survey rounds as repeat- (October-December 2020) are summarized in ed cross-sections, and confirmed through panel this brief. Round 2 surveyed 1,092 households in analysis. high exposure upazilas (primarily Ukhia-Teknaf), 1,088 households in low exposure upazilas The education update initiated in Round 2 is (Cox’s Bazar Sadar, Ramu, Pekua, Chakaria) and part of a broader, mixed methods study on as- 1,662 households in Rohingya camps. Education pirations and well-being which aims to under- data was collected as part of the household ros- stand parental aspirations and ambitions for ter in the baseline survey and added as a mod- their children, and the capacity of the Rohingya ule in the Round 2 rapid follow-up. Education and Bangladeshi hosts in Cox’s Bazar to achieve updates were collected on current engagement those aspirations. HOST COMMUNITIES IN 2019: LOW LITERACY ADULT POPULATION WITH GENDER GAPS; AMONG SCHOOL-AGED CHILDREN, HIGH PRIMARY ENROLMENT RATES BUT SIGNIFICANT SECONDARY SCHOOL DROPOUT. According to the CBPS baseline conducted in school were high in 2019 (Figure 2). There were no 2019, only 60 percent of host adults in Cox’s Ba- visible gender gaps, except in low-exposure host zar can read; one-third of the adult population communities, where girls reported higher enroll- have no schooling at all, and an additional 25 ment rates. However, secondary enrolment rates percent have reached primary school comple- were lower than the national average: 54 percent tion or less. There is also a gender gap in educa- for boys and 63 percent for girls, compared to na- tional attainment between adults: 37 percent of tional rates of 63 percent and 66 percent respec- adult women have no schooling, compared to tively (Figure 2). This was mirrored in the much 29 percent of men. Adults in high exposure areas higher rates of dropouts in secondary schools, reported lower educational attainment than low for both girls and boys, especially in high expo- exposure adults, with larger male-female gaps in sure regions (Figure 1). On average, more than a attainment in Ukhia and TeknafAmong children third of secondary school age children drop out of school-going age, enrollment rates in primary before completing this level. CBPS EDUCATION BRIEF 5 REASONS FOR NON-ATTENDANCE AND DROPOUTS AMONG SCHOOL-AGE CHILDREN VARY SHARPLY BY GENDER. Figure 1: Baseline (2019) enrolment rates in Figure 2: Baseline (2019) dropout rates in host communities host communities 97% 94% 94% 99% 91% 91% 68% 63% 54% 51% 51% 55% 42% 44% 38% 37% 33% 29% 5% 4% 5% 6% 2% 0% Male Female Male Female Male Female Male Female Male Female Male Female Hosts High Exposure Low Exposure Hosts High Exposure Low Exposure Primary School enrollment (ages 7-12) Primary School Drop-out Rates Secondary School Drop-out Rates Secondary School enrollment (ages 13-18) Within the host community, boys and girls are Figure 3: Age distribution by gender for baseline (2019) school attendance generally equally likely to be attending school until the age of 12. Gender gaps in attendance 100% start to appear in this cohort of children of sec- 80% ondary school age, with girls more likely to at- 60% tend school than boys. By age 18, this pattern is 40% somewhat reversed (Figure 3). 20% 0% 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Educational costs and the lack of sufficient funds Male Female for education were most cited reasons for host girls not attending school (36 percent), followed by marriage (18 percent) and family/social restric- tions (16 percent). Out-of-school girls from the Figure 4: Reasons for not going to school host community were likely leave school for mar- during baseline (2019) riage between the ages of 17 and 20 (28 percent). 0% 20% 40% 60% 80% 100% 5 to 8 The primary reason cited for not attending 9 to 12 amongst host community school-aged boys Boys 13 to 16 were financial costs of education (41 percent) Age range 17 to 20 followed by reporting completing education or 5 to 8 not wanting to study (22 percent), and the need 9 to 12 Girls to work (14 percent). 13 to 16 17 to 20 Do not want/Completed Too old/too young Other Marriage Work/Family chores Family Restrictions No money/too expensive CBPS EDUCATION BRIEF 6 DESPITE HAVING A HIGHER SHARE OF SCHOOL-AGE CHILDREN OUT OF SCHOOL AT BASELINE AND IN ROUND 2, CHILDREN IN HE AREAS WHO WERE IN SCHOOL, WERE MORE LIKELY TO REMAIN ENGAGED IN STUDIES AMIDST SCHOOL CLOSURES IN 2020, COMPARED WITH THEIR PEERS IN LE AREAS. Figure 5: Panel transitions between baseline school aged children in the host community are and R2 education status,out of all school also found to correlate significantly with educa- All hosts tion level and current employment status of the household head, suggesting that such house- LE holds may place greater value on continuing HE their child(ren)’s education through alternative 0% 20% 40% 60% 80% 100% channels such as online learning and private tui- Studyinig in baseline and R2 Studying in R2 but not in Baseline tions, and have the means to do so. Studying in Baseline but not in R2 Not studying in both rounds Panel transitions between the baseline and R2 uncover similar topline trends: despite a larger Between October-December 2020 (Round 2 of proportion of school age children out of school in the tracking surveys, R2), two out of three host both rounds in HE areas, both retention in school households reported that school-aged children during COVID, and new entrance to school during (5-18) were actively studying, with the propor- COVID are higher in HE areas. At baseline and R2, tions in HE being larger than LE and with no 13 percent of school age children in HE areas were notable difference between boys and girls. The out of school compared with 8 percent in LE areas. higher rates of student engagement could be However, the share of children currently studying attributable to the government’s nationwide in R2 is higher in HE areas, at 75 percent, com- efforts in ensuring remote-learning platforms pared with 67 percent in LE areas. This is partially during COVID-19 closures, with high exposure explained by two positive shifts in HE areas: (a) a regions receiving added attention from the edu- higher share of children studying in R2 relative to cation cluster in Ukhia-Teknaf as part of the Ro- baseline; and (b) and a lower share of children in hingya response. Educational continuity among school in baseline but not in school in R2. FINANCIAL CONSTRAINTS CONTINUE TO BE A KEY FACTOR BEHIND DROPOUTS BEYOND COVID CLOSURE, WITH ADDITIONAL CONSTRAINTS FOR GIRLS. Among children not studying during R2, 50 per- children. For girls, the most common reasons cent were reported as being temporarily out of reported was included marriage and other so- school due to COVID-related school closures. cial constrains, whereas for boys, financial con- 99 percent of their parents reported that the straints paired with needing to join the labor children would return to school when they re- force as most common reasons for not studying. opened. However, COVID-19 closures seem to Panel analysis confirm these trends showing that have impacted education more strongly in LE ar- COVID-related school closures were the main eas than in HE areas close to camps. Indeed, the reason for households reporting children drop- share of LE school aged children reporting being ping out of school between the baseline in 2019 out of school and not studying due to COVID-re- and R2 in late 2020. For those out of school in lated school closures is twice as high as in HE both rounds, structural barriers such as social re- areas (63 vs 32 percent). Among those who were strictions, marriage, need to work and financial out of school due to non-COVID-related reasons, constraints remain important constraints. financial constraints were the key reason for all CBPS EDUCATION BRIEF 7 HOW DO GAINS IN EDUCATION ACCESS AND ATTAINMENT AFFECT BROADER WELL-BEING IN COX’S BAZAR? Analysis on the returns to education on the la- ly higher returns if they are wage workers. For bor market indicate that, a priori, statistically self-employed individuals, earning levels do not significant returns to education for hosts emerge seem to be impacted significantly by education, only when individuals have obtained some level but instead depend on their sector of activity. of tertiary education. However, there are some differences in returns across high and low expo- While tertiary education attainment clearly sure areas. For HE hosts, returns to tertiary ed- generates returns in employment in the form of ucation are significantly positive only when the monthly salaried, formal jobs (all host individu- individual is a wage worker in the service sector. als with tertiary education who were employed This may be related to the recent emergence of had these types of jobs), only 3.65 percent of NGO jobs in Ukhia-Teknaf related to the Rohing- hosts had attained this level of education at ya response and efforts to support the neighbor- baseline. The paucity of educated individuals is ing host community. More generally, individuals also evident in the fact that more than four-fifths with tertiary education only receive significant- of monthly salaried jobs are held by individuals with higher secondary education or less. ROHINGYA COMMUNITIES1: VERY LOW LITERACY ADULT POPULATION WITH LARGE GENDER GAPS; SCHOOL-AGED CHILDREN IN CAMPS ONLY HAD LIMITED ACCESS TO INFORMAL LEARNING CENTERS IN CAMPS AT BASELINE. In 2019, 62 percent of Rohingya adults reported host communities. Among school-aged children never having attended school, and an additional in camps, despite reliance on informal learning 22 percent who attended did not complete pri- centers, access to some form of educational at- mary school. Only 23 percent of Rohingya adults tainment was already better than their previous reported being able to read, with considerably generations with 58 percent attending schools. larger gender gaps than what was observed in Table 1: Schooling Attainment for adults (18+), by stratum and gender Gender No Some Complete Some Complete Religious Schooling Primary Primary Secondary Secondary Education Male 51% 24% 2% 12% 2% 4% Female 71% 21% 1% 2% 0% 5% 1 As per the regulations set by the Government of Bangladesh, there is no formal curriculum operational inside camps. Education is provided to Rohingya under the Learning Competency Framework and Approach (LCFA), based on which the “Guidelines for Informal Education Programming” (GIEP) was drafted. Levels I and II i.e., the equivalent levels up to grade 2 in a formal school system are currently approved, with higher levels approval and operationalization in process. CBPS EDUCATION BRIEF 8 Figure 6: Educational attainment by gender Figure 7: Age distribution by gender for of school aged children in camps during the baseline (2019) school attendance 2019 baseline 100% 84% 90% 79% 80% 70% 60% 46% 50% 45% 40% 31% 29% 30% 20% 9% 15% 6% 6% 10% 0% 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Primary Secondary Primary Secondary Have never School School School School Attended Male Female enrollment enrollment Drop-out Drop-out School (ages 7-12) (ages 13- Rates Rates (ages 7-18) 18) Male Female At baseline in 2019, Rohingya children in camps had no access to a formal education system with Rohingya children do not have access to a formal standardized curriculum and instruction with education system with standardized curriculum and more than 70 percent of the children going to instruction. The education cluster led by UNICEF, in learning centers (LC’s) run by various non-gov- collaboration with the Government of Bangladesh ernment organizations (NGOs). Consequentially, (GoB) built the Learning Competency Framework while primary enrolment rates for the communi- and Approach (LCFA), which is currently followed ty were on the higher end (84 percent for boys, 79 across informal Learning Centers (LC’s) in camps. percent for girls) with effectively no gender gaps, Based on the LCFA, the GoB drafted the “Guidelines the status of secondary education attainment was for Informal Education Programming” (GIEP) pol- equally abysmal. Enrollment rates are substan- icy, a 2-year program designed to facilitate attain- tially lower, and the gender gap worse (Figure 6)2. ment of basic competencies across age groups 5-14 years. Under the GIEP, Levels I and II are equivalent There was also clear evidence of divergence in to pre-primary up to grade 2 teaching English and school attendance amongst the Rohingya chil- Burmese literacy, numeracy, life skills; Levels III and dren by gender after the age of 10. Only 45 per- IV are equivalent to grades 3-8, with the addition cent of Rohingya girls aged 12 were attending of science as a subject. Apart from learning centers school, compared to 73 percent of boys of the under the LCFA, adolescents aged 15 to 18 years re- same age (Figure 7). The gap increases with age ceived literacy, numeracy, life-skills and vocational as by the age 15, a negligible proportion of Ro- skills training in centers such as Multi-purpose Child hingya girls are enrolled in any type of schooling and Adolescent Centre’s run by UNICEF. (5 percent), compared to more than a third of boys (33 percent). These rates of enrolment coincide with updated findings according to the 2020 Multi Sectoral Needs Assessment (MSNA), 2 which states that 76 per cent of girls and 80 per cent of boys aged 6 to 14 were attending NGO-run LCs. CBPS EDUCATION BRIEF 9 DESPITE LOCKDOWNS AND SCHOOL CLOSURES, RATES OF REPORTED The Government of Bangladesh’s decision to ex- pand access to education for the Rohingya under the ENGAGEMENT IN EDUCATION Myanmar curriculum is expected to benefit post-pri- CLEARLY INDICATE THE EXPANSIONS mary students significantly. On 29 January 2020, IN ACCESS. the Government of Bangladesh eased regulations of education, granting approval for the Myanmar Cur- Despite the protracted operational contractions riculum Pilot (MCP) in the camps. This pilot plans to on account of COVID-19, half of the school-aged initially target Rohingya students from grades 6-9, an children (54 percent) reported studying during older cohort of children/adolescents who currently Round 2 (Oct-Dec 2020), indicative of the efforts by have lower access than younger groups. The cur- the education cluster in attempting to ensure edu- riculum is being prepared to include Burmese, En- cation continuity through a range of door-to-door glish, mathematics, science and social studies. Later service delivery and distance-learning methods. phases are expected to expand access across grades and subjects covered. While rates of current education engagement among previously well-established groups aged The education sector in Cox’s Bazar had planned to 5-12 years old have seen expected declines introduce the Myanmar curriculum in camps starting (plausibly due to COVID-19 closures), the 13-16 in April 2020, initially targeting 10,000 Rohingya stu- age group on the contrary reported more than dents in grades six to nine. COVID restrictions start- double the rates of educational engagement ing in end-March however classified education un- compared to 2019. There may be two ways to der non-essential operations, putting a halt to said explain this, apart from the obvious factor of improvements in education access inside camps. children from lower age groups in baseline con- tinuing to study. 50% (i) Ongoing preparations for expanding educa- tion access under the Myanmar curriculum may be responsible for increased appetite and partic- 39% 40% ipation among these age-groups. 36% (ii) home-based caregiver led education modali- ties in camps introduced in response to Learning 21% Center (LC) closures may have catalyzed higher participation among female adolescents, whom parents are reluctant to send to co-educational 9% LCs for socio-cultural reasons3. This hypothesis is further strengthened by the fact that increase 1% 2% in participation in this age-group was driven by girls (+28 percentage points; boys: +16 percent- 5 to 8 years 9 to 12 years 13 to 16 18 to 21 years years age points). Baseline (2019) Round 2 (2020) 3 Source: Sreeparna Banerjee, “Children Left Behind: Challenges in Providing Education to the Rohingya Children in Bangladesh,” ORF Issue Brief No. 465, May 2021, Observer Research Foundation.