Delaying Child Marriage through Community-Based Skills-Development Programs for Girls: Results from a Randomized Controlled Study in Rural Bangladesh

"BALIKA results show that programs that educate girls, build their skills for modern livelihoods and engage their communities can reduce the likelihood of child marriage by one-third and produce better health, educational and social outcomes for girls."

This report describes the rationale behind, implementation of, and results from the Bangladeshi Association for Life Skills, Income, and Knowledge for Adolescents (BALIKA) project (see Related Summaries below for a full description). BALIKA is a randomised controlled trial to evaluate approaches to prevent child marriage and improve life opportunities for girls. The opening sections of the report explore the extent of the culturally entrenched practice of early marriage in Bangladesh, where two out of three girls are married before the legal age of 18. The objective of the programme is to bring about change in girls' lives and their status in the community so that they are perceived as assets rather than as burdens and liabilities to their families and to society. A consortium comprising Population Services and Training Center (PSTC), mPower Social Enterprise (mPower), Center for International Development Issues Nijmegen (CIDIN), and the Population Council partnered on the project.

More than 9,000 in- and out-of-school girls aged 12-18 in 72 communities in three districts of Bangladesh that are considered "hotspots" for early marriage (Khulna, Satkhira, and Narail) participated. Communities were assigned to one of three arms in which girls received either (i) education support through tutoring in math and English; (ii) training on gender rights and negotiation, critical thinking, and decision making; or (iii) livelihoods training in entrepreneurship, mobile phone servicing, photography, and basic first aid. All of the girls met weekly with locally selected young female mentors and peers in safe, girl-only locations, called BALIKA centres, which helped girls develop friendships, receive training on new technologies, borrow books, and acquire the skills they need to navigate the transition from girlhood to adulthood. Girls would use these skills within their communities, helping to build their confidence, demonstrate their achievements, and elevate their profiles. Another 24 communities served as the control arm of this study: no services were provided.

To measure the impact of each intervention strategy in relation to the others and to the control group, a baseline survey was conducted before the project was implemented, and an endline survey was conducted after the project had been in place for 18 months. Specifically, the research team selected 96 potential centres and randomized them into 4 categories. A population enumeration was conducted in the catchment area of centres, defined as anyone living within a 2-kilometer (1.2-mile) walking distance of the centre. The baseline survey was designed to cover an average of 120 adolescents in each of the 96 communities. At baseline, 11,609 respondents were successfully interviewed between March and August 2013. The endline survey began after programme implementation was completed and was conducted over a period of four months between August and November 2015, and an attempt was made to interview all adolescents interviewed at baseline. A total of 9,982 interviews were successfully completed, yielding a re-contact rate of 86%. To assess the impact of the BALIKA programme, the researchers performed a difference-in-differences (DiD) analysis adjusting for three key sociodemographic characteristics: age, religion, and wealth quintile.

In short, the baseline survey confirmed a high prevalence of child marriage in the study area, and more than two-thirds of girls married by age 19. Even within communities where child marriage is common, the practice is more common among the economically poor. Marriage is seen as a parental decision, and seeking the young person's consent is not common. Only 20% of girls completed a secondary or higher level of education; 52% of girls who had dropped out reported marriage as the reason for school discontinuation. Adolescent girls in the study area had few opportunities to earn money - only 10% reported ever working for pay and 7% said they were currently working to earn money. Those who were married and had lower educational attainment in general held more gender-inequitable values.

In terms of BALIKA impact, the results show that girls in BALIKA communities were nearly one-third less likely to be married as children than girls living in communities not reached by the BALIKA project (0.69-0.77 relative odds adjusted for age, religion and family wealth status). By the end of the study, girls who were not married at baseline were one-fourth less likely to be married (0.76–0.78 relative odds adjusted for age, religion, and family wealth status).

• In BALIKA communities where girls received educational support, girls were 31% less likely to be married as children at endline than girls in the control communities.
• In communities where girls received lifeskills training on gender rights and negotiation, critical thinking, and decision making, girls were 31% less likely to be married as children at endline than girls in the control communities.
• In communities where girls received livelihoods training in entrepreneurship, mobile phone servicing, photography, and basic first aid, girls were 23% less likely to be married as children at endline than girls in the control communities.
• Compared to girls outside BALIKA communities, the study found that girls participating in the project were: 18% more likely to be attending school; 20% more likely to have improved mathematical skills if they received education support and gender-rights awareness training; and one-third more likely to be earning an income if they received gender-rights awareness or livelihoods skills training.

In addition to child-marriage indicators, Figure 4 of the paper summarises the results reported in Section 3 on the impact of the interventions on schooling, work, gender-awareness, and sexual and reproductive health and rights indicators. On a number of indicators, such as knowledge (and confidence) that girls can say no to marriage, rejection of norms reinforcing gender-based violence, knowledge of sexually transmitted infection (STI) and HIV transmission, and behaviours seeking reproductive health services, improvements were significantly greater in the villages that participated in the BALIKA programme.

These results are from an intent-to-treat analysis, in which the impact of each intervention strategy on child marriage is measured among all girls who live in the community, not just those girls who participated in the BALIKA programme.

Quotes throughout the report offer quotations from focus group discussions (FGDs) and key informant interviews that also demonstrate qualitative impact. For example, one BALIKA participant from Pankhali, Dacope, said: "I learned from BALIKA that I can say 'no' to a marriage proposal. I learned that if a marriage proposal comes and I am too young to marry, I am able to express my opinion to convince my parents. If I couldn't convince them, then I would seek out someone in the family who would understand me or else I would consult with my friends."

Building on these results, the BALIKA project makes recommendations on the best approaches to delay child marriage in Bangladesh for policymakers, donors, and programme implementers, which include:

• Design programmes based on context analysis: The drivers of child marriage vary greatly around the world. In Bangladesh, the dominance of arranged marriages, dowry, concerns about the reputation and safety of daughters, and a strong sense of duty among parents to marry their daughters drive decisions about early marriage.
• Reach girls at an early age and while they are in school: Age-differentiated patterns of programme participation and decline in child marriage as measured in odds ratios suggest that interventions are more successful in delaying marriage among 16- and 17-year-olds and for education interventions for the under age 16 marriages.
• Provide girls with the skills they need: All girls received basic life skills training to help develop critical thinking and personal and interpersonal competencies needed to lead fuller, healthier lives.
• Create girl-centred platforms: Safe spaces were established for girls to come together and forge a common identity helped improve self-confidence, health, and well-being.
• Engage the community: When the community is engaged in conversations around girls' vulnerabilities and approaches to ensure their health and well-being, interventions are more successful.
• Use technology: Girls received access to digital learning material and training on new technologies - skills that help traverse the digital divide. These skills helped build confidence, elevate their profiles within the community, and enable girls to communicate with the world around them.