#7 POLICY LESSONS ON IMPROVING EDUCATION OUTCOMES GENDER INNOVATION LAB FEDERATION EVIDENCE SERIES GENDER INNOVATION LAB FEDERATION The Gender Innovation Lab (GIL) Federation is a World Bank community of practice coordinated by the Gender Group that brings together the Bank’s five regional GILs: Africa (AFR), East Asia and Pacific (EAP), Latin America and the Caribbean (LAC), Middle East and North Africa (MNA), and South Asia (SAR). Together, they are conducting impact evaluations of development interventions to generate evidence and lessons on how to close gender gaps in human capital, earnings, productivity, assets, voice and agency. With over 188 impact evaluations in 66 countries completed to date, the GIL Federation is building the evidence base for governments, development organizations, and the private sector to increase uptake of effective policies that address the underlying causes of gender inequality. Significant progress has been made in closing school, they usually marry at an older age and reduce gender gaps in primary and secondary enrollment their fertility. rates worldwide. However, girls still have lower In the Philippines, the EAP GIL evaluated the long-term expected years of schooling than boys in some regions, impacts of a national conditional cash transfer program particularly in Sub-Saharan Africa, and boys have worse that was targeted at mothers of girls ages 12-14.5 The educational outcomes than girls in other countries, most transfers were randomly allocated at the barangay level notably in Latin America and the Caribbean.1 Barriers to (administrative unit). The study shows that girls whose the continuation of schooling for girls are linked to child mothers received the conditional cash transfer program marriage, early pregnancies, sexual harassment, and for 1.5 years experienced reductions in fertility and social norms around girls’ education.2 The COVID-19 delayed marriage until their twenties, in comparison to pandemic has also impacted schooling of both girls and girls whose mothers lived in control areas not selected boys. The transition to remote learning hurt girls who for the transfers. Although there were no long-term often have fewer technical skills and less access to the impacts on labor market outcomes, the evidence internet than boys.3 In other cases, boys had higher suggests that conditional cash transfers targeted during economic opportunities than girls and were more likely the transition to adolescence can delay adolescent to drop out from school in response to the economic childbirth and marriage and help keep girls in school stress generated by the pandemic.4 The GIL longer. Federation is generating rigorous evidence around the world to understand what works, and what does The Africa GIL supported a randomized controlled trial not, in narrowing gender gaps in education. This note (RCT) of a three-year scholarship program targeting presents evidence on three key findings. Nigerian middle school girls as part of the Sahel Women Empowerment and Demographic Dividend (SWEDD) project. The study finds that the program improved girls’ educational outcomes (reducing their likelihood of FINDING 1. CASH TRANSFERS ARE EFFECTIVE dropping out of school), postponed marriage, and AT DELAYING MARRIAGE AND KEEPING increased their life satisfaction.6 Girls in villages GIRLS IN SCHOOL assigned to receive the program were 53 percent less likely to drop out of school and 49 percent less likely to Financial resources labeled for girls’ education, in the be married than girls in villages not assigned to receive form of conditional cash transfers or scholarships, can the program. improve schooling outcomes. When girls stay longer in FINDING 2. LIFE SKILLS TRAINING WITH mindsets, socio-emotional skills, and other behavioral MENTORS AND PEERS CAN IMPROVE GIRL’S factors.9 The results of the impact evaluation show that these mindsets, skills, and behaviors can be taught. EDUCATIONAL ATTAINMENT The Africa GIL conducted an RCT to evaluate the Some 2,100 lower-secondary schools were randomly Sisters of Success (SOS) program in Liberia, which assigned into three groups: one receiving a six-session aimed to support girls’ transition into adolescence and curriculum teaching growth mindset and self- adulthood by delivering life skills training through management, one receiving the six-session curriculum mentors.7 Girls ages 12–15 in the treatment group were paired with tools and activities for daily classroom use, matched with a volunteer woman mentor from their and a control group that did not receive any new community who delivered a life skills curriculum focused curriculum.10 Both interventions shifted some of the on building social and emotional skills. The control socio-emotional skills and study behaviors taught in the group was not offered any intervention. training. Impacts were stronger for students from lower socio-economic households, with lower baseline Girls in the treatment group saw improvements in their grades, and in lower performing schools. The more interpersonal relationships with peers and parents and intensive intervention with tools and activities for daily increases in their educational attainment (both in use led to some additional improvements for students; primary school completion and secondary school however, they came at a higher cost and burden for the enrollment) four years after the program. These impacts schools. Although the impact evaluation shows that in- were already observed the first year after the program school training programs can boost targeted skills and and persisted more than four years after the end of the behaviors that the literature stresses for academic program. They were detected in both survey data and success, neither intervention led to improved exam administrative records. These effects were concentrated scores or aspirations for higher levels of education in among the younger girls (age 12–13 at baseline), the weeks after the program. suggesting that supporting girls as they transition into adolescence may yield greater impacts. No effects were observed on sexual activity or pregnancy. FINDING 3. PHONE-BASED OUTREACH CAN IMPROVE ADOLESCENT’S MOTIVATION AND The LAC GIL supported an RCT in Brazil to measure the ASPIRATIONS AND ENCOURAGE THEM TO effects of peer-led life skills training focused on reproductive health and goal setting.8 High school STAY IN SCHOOL students served as peer educators to disseminate The economic and educational disruptions generated knowledge. Different selection criteria to choose these by the COVID-19 pandemic have increased dropout peer educators were randomized across schools using rates around the world. Low-cost interventions, such as social network data elicited at baseline. In one-third of phone-based outreach, have proved to be effective at treatment schools, peer educators were selected mitigating those effects. according to their network centrality. In another third of treated schools, the most popular students, according In Bangladesh, the SAR GIL conducted an RCT to to their peers’ ranking, served as educators. In the estimate the effects of a phone-based outreach remaining schools, students chosen by the school program designed to foster a growth mindset among administration disseminated information. Overall, the adolescents to overcome the negative impacts of the intervention significantly increased contraceptive use, pandemic on their motivation.11 Adolescents in the study reduced teenage pregnancy, and increased intended were enrolled in grades 7 and 8 across 109 schools post-secondary school enrollment. Because different prior to the onset of the pandemic. Students from 73 types of peer educators reached different students, schools were randomly assigned to participate in the effects differed by treatment arm. Notably, the treatment program, while students in the other 36 schools served had smaller impacts when peer educators were as a control group. The evaluation finds that the phone selected by the school rather than by one of the outreach was associated with a 9 percent increase in network-based methods. adolescent motivation and aspirations across all genders. The program mitigated the reduction in Baseline data from an impact evaluation conducted by university aspirations for girls, who suffered more from the EAP GIL in Indonesia show that 8th grade boys have, this than boys during the pandemic.12 on average, lower grades and school attendance than girls, and these gender differences are linked to their Another low-cost alternative is the use of text messages. educational outcomes. The study pooled different data The LAC GIL conducted a pilot study in Ecuador where sources to construct a dataset with 32,000 observations sexual and reproductive health education was of children ages 6 to 18.14 The study used a two-stage implemented via text messages.13 In line with Finding 2, least squares strategy exploiting exogenous changes in the program used peers to send the information. It was tariffs on female-intensive sectors to estimate the effects successful at reducing teenage pregnancy and of maternal employment on child health outcomes. The increasing adolescent girls’ self-reported educational findings indicate that maternal employment can have aspirations as well as school continuation. positive long-term impacts on children's development outcomes. Mothers’ employment significantly increased their children’s years of schooling and enrollment in FINDING 4. EMPOWERING WOMEN CAN school and positively affected health outcomes, such as IMPROVE CHILDREN’S EDUCATION height for age and hemoglobin levels. A study by the EAP GIL in Indonesia shows that maternal employment can lead to improved children’s FOR MORE INFORMATION, PLEASE CONTACT Diego Ubfal dubfal@worldbank.org 1818 H St NW Washington, DC 20433 USA https://www.worldbank.org/en/topic/gender ACKNOWLEDGMENTS This brief is a product of collaboration between the World Bank Gender Group and the Gender Innovation Labs. It was prepared by Daniel Halim, Diego Ubfal, and Rigzom Wangchuk with key inputs from Diana Arango, Elizaveta Perova, and Rachael Pierotti. It was copy-edited by Leslie Ashby. Other contributors include Lourdes Rodriguez Chamussy, Maria Emilia Cucagna, Isis Gaddis, Markus Goldstein, Jacobus Joost De Hoop, Forest Brach Jarvis, Hillary C. Johnson, Lili Mottaghi, Michael B. O'Sullivan, Laura B. Rawlings, Javier Romero, Jayati Sethi, and Emcet Tas. The World Bank GILs and the GIL Federation are supported by the Umbrella Facility for Gender Equality (UFGE), a multi-donor trust fund administered by the World Bank and supported with generous contributions from Australia, Canada, Denmark, Finland, Germany, Iceland, Ireland, Latvia, the Netherlands, Norway, Spain, Sweden, Switzerland, United Kingdom, United States, the Bill and Melinda Gates Foundation, and the Wellspring Philanthropic Fund. ENDNOTES 1 Fotso, Arlette Simo and Daniel Halim. 2019. Human capital for all: are we there yet? 5 things to know about gender equality and human capital. Blog, World Bank. 2 Baird, Sarah, Sarah Bunker, Shwetlena Sabarwal, and Jeffifer Seager. 2020. Investing in girls’ education in the time of COVID-19. Blog, World Bank. 3 Sosale, Shobhana, Sarah Bunker, Laura McDonald, and Eliana Rubiano-Matulevich. 2021. The opportunity to build back better from COVID-19: Fostering gender parity in education and skills development for a larger workforce. Blog, World Bank. 4 Geven, Koen, Sheena Fazili, Sheena, Ayesha Tahir, and Tazeen Fasih. 2022. SMS Girl Data Insights. SAR Gender Innovation Lab Policy Brief, World Bank. 5 Dervisevic, Ervin; Elizaveta Perova, and Abhilasha Sahay. 2021. Long-Term Impacts of a Short Exposure to CCTs in Adolescence: Evidence from the Philippines. World Bank Policy Research Working Paper 9617. 6 Giacobino, Hélène, Elise Huillery, Bastien Michel, and Mathilde Sage. 2022. Schoolgirls Not Brides: Secondary Education as a Shield Against Child Marriage. Working Papers DT/2022/01, DIAL. 7 Africa GIL. 2016. Sisters of Success: Measuring the Impact of Mentoring and Girls Groups in Supporting Girls' Transition into Adolescence and Adulthood, in Liberia: Full Length Baseline Report. World Bank. 8 Baumgartner, Eric, Emily Breza, Eliana La Ferrara, Victor Orozco, and Pedro Rosa Días. 2021. The nerds, the cool and the central. Peer education and teen pregnancy in Brazil. Preliminary Results, June 6, 2021. 9 EAP GIL. 2018. Why Do Indonesian Adolescent Boys Have Poorer Schooling Outcomes than Girls? EAPGIL Policy Brief Issue 5, World Bank. 10 Johnson, Hillary, Daniel Pinzón Hernandez, Kali Trzesniewski, Taufik Indrakesuma, Renos Vakis, Elizaveta Perova, Noël Muller, Samantha De Martino, and Diego Catalán Molina. 2020. Can Teaching Growth Mindset and Self- Management At School Shift Student Outcomes and Teacher Mindsets? Evidence From a Randomized Controlled Trial in Indonesia. World Bank. 11 Baird, Sarah, Jennifer Seager, Shwetlena Sabarwal, Silvia Guglielmi, and Maheen Sultan. 2020. Adolescence in the Time of COVID-19: Evidence from Bangladesh. Policy Brief, The World Bank. 12 Seager, Jennifer, T.M. Asaduzzaman, Sarah Baird, Shwetlena Sabarwal, and Salauddin Tauseef. 2022. Gender, Growth Mindset, and Covid-19: A Cluster Randomized Controlled Trial in Bangladesh. Review of Economic Analysis, 14(2), 183-219. 13 Cuevas, Facundo, Marta Favara, and Megan Rounseville. 2015. Quito Text Me Maybe: Piloting a new intervention for teenage pregnancy prevention. Technical Report. The World Bank. 14 Dervisevic, Ervin, Maria Lo Bue, and Elizaveta Perova. 2021. Maternal employment and children’s outcomes. Evidence from Indonesia. WIDER Working Paper 2021/186.