While enrollment in secondary school has increased significantly in the past 20 years, net enrollment rates remain below 70% in low- and middle-income countries (LMICs) (UNESCO Institute for Statistics, 2021). Factors such as engagement in paid work and early marriage draw adolescents away from schooling during these critical years of development. In Bangladesh, where a reverse gender-gap in enrollment has emerged in recent years at both the primary and early secondary levels, gender inequities remain in the intrahousehold allocation of educational resources, with boys receiving greater investment in the quality of schooling (Xu, Sonchoy, and Fujii, 2022). Furthermore, at the secondary level, dropout rates among girls outpace boys at 42% compared to 33% (GIRL Center Research Brief, 2020; BANBEIS 2018). Furthermore, recent school closures due to the covid-19 pandemic threaten to undo progress in enrollment and widen disparities in investment, especially for girls and those from disadvantaged backgrounds. Adolescents in LMICs are disproportionately affected due to longer durations of school closures and lower access to distance learning modalities (World Bank, 2020; Baird et al., 202), and adolescents in Bangladesh are among the most negatively affected globally, with complete school closures that lasted 63 weeks.
We are piloting evidence-based approaches among adolescents in Bangladesh to promote re-enrollment, increase secondary school completion, and promote gender equitable behavior. This research evaluates, through a school-based cluster Randomized Controlled Trial (cRCT) with 109 schools, two interventions that were virtually-delivered during covid-19-related school closures: (1) a gender-neutral Growth Mindset (GM) programming around malleable intelligence and (2) Girl Rising (GR) programming that focuses on gender norms around girls’ education that is layered on top of the GM programming. We layer GR on GM in order to test whether programming that speaks particularly to the constrains faced by adolescent girls has additional positive impacts on adolescent outcomes.
Intervention
There are two main interventions:
1. Eight weeks of group-based Growth Mindset Programming (GM) delivered via phone: Growth mindset is rooted in Carol Dweck’s pioneering research on malleable intelligence; when children (and adults) have a growth mindset, they believe that their ability depends on effort and are not constrained by fixed intelligence – the belief that talents are innate (Dweck, 2008). Adolescents were organized into groups of 3 students, according to school of attendance, gender, and grade and engaged in an eight-week intervention between April 5 and June 3, 2021. In week one, students participate in a group call where a facilitator reads an essay on malleable intelligence followed by discussion. In week two, students write a letter to a friend explaining malleable intelligence, and there is a second group call to discuss the contents of the letter. In weeks three to seven, students receive true/false statements around growth mindset concepts via SMS, with a final wrap up group call occurring in week eight. In weeks five and seven, there are additional refresher group calls to review concepts. Adolescents received a certificate of completion at the end of the intervention.
2. Eight weeks of group-based Girl Rising programming (GR) delivered via phone, in combination with videos and a companion story book: Girl rising is a gender sensitization program that uses powerful storytelling tools to build adolescents' voice and agency and underscore the importance of education. The objective of the intervention is to create awareness of gender-based discrimination, change dominant gendered perceptions, promote gender-equitable attitudes, and provide tools to participants to translate attitude change and greater aspirations into behavior change (Vyas 2020). Adolescents were organized into groups of 3 students, according to school of attendance, gender, and grade and engaged in an eight-week intervention between September 27 and November 20, 2021. Adolescents are provided a physical story book with six stories of six girls from six countries, accompanied by six companion videos, shedding light on a variety of gender-based issues, such as early marriage. The first week of the intervention introduced the programming structure and stories. Weeks two through seven comprise of adolescents reading and watching an assigned story, followed by a group phone call of 20-25 minutes with a facilitator to discuss the story. In the final week of the intervention, adolescents write their own stories and engage in a final group call.
Experimental design
The study sample comprises 2,220 adolescents across 109 schools.
1. Control (No intervention): 744 students (attending grades 7 and 8) in 36 school clusters. The target population comes from school registration lists in February and March 2020.
2. Treatment 1 (Growth Mindset Programming (GM)): 733 students (attending grades 7 and 8) in 36 school clusters. The target population comes from school registration lists in February and March 2020. These students participate in the 8-week Growth Mindset programming.
3. Treatment 2 (Growth Mindset and Girl Rising Programming (GM+GR)): 743 students (attending grades 7 and 8) in 37 school clusters. The target population comes from school registration lists in February and March 2020. These students participate in the 8-week Growth Mindset programming. They are additionally engaged in 8 weeks of Girl Rising programming.
Design details
Randomization method
Cluster randomization in the office using Stata 16. Randomization was stratified by urban/rural and government/MPO cells.
Medium term (2022)
Education and learning measured by: (1) a binary indicator for grade 10 completion (highest grade completed >=10)*; (2) an index comprised of (i) a binary indicator for currently enrolled in school; (ii) proportion of time spent at school or studying in a typical day (hours in a typical day/24); and (iii) learning measured by total scores on reading and math tests.
*Grade 10 completion unable to be measured in 2022 as adolescents were expected to be currently enrolled in grades 9 and 10 at that time.
Socioemotional Skills measured by an index comprised of: (i) a grit scale (generated from seven items following Alan, Boneva, and Ertac [2019]); (ii) a malleability scale (generated from six items following Alan, Boneva, and Ertac [2019]; and (iii) a growth mindset scale (generated from four items from the World Bank STEP survey [2014]).
Gender Attitudes measured an index comprised of: (i) a scale of 4 education gender attitudes; (ii) an IAT score of gender bias around education and domestic work; and (iii) a scale of gender equitable responses to a vignette around early marriage, street harassment, gender roles, and mobility.
All composite indexes are constructed by standardizing each component to the mean and standard deviation in the control group and taking the unweighted mean of the standardized components.
Long term (2025)
Education and learning measured by: (1) a binary indicator for grade 10 completion (highest grade completed >=10); (2) an index comprised of (i) a binary indicator for currently enrolled in school; (ii) proportion of time spent at school or studying in a typical day (hours in a typical day/24); and (iii) learning measured by total scores on reading and math tests; (3) an index comprised of items listed in (1) and (2).
Marriage before the age of 18 measured with a binary indicator (among females).
Engaged in paid work, schooling, or training measured with a binary indicator.
Gender attitudes measured by a gender stereotypical roles scale comprised of eight items.
All composite indexes are constructed by standardizing each component to the mean and standard deviation in the control group and taking the unweighted mean of the standardized components.
Secondary outcomes
Medium term (2022)
Aspirations measured by an index comprised of: (i) a binary indicator for aspire to university education; (ii) a binary indicator for aspire to professional career; (iii) a binary indicator for set an education-related goal in the next week; (iv) a binary indicator for set an education-related goal in the next year; and (v) a binary indicator for adolescent has a role model outside the household.
Socioemotional skills measured by an index comprised of: (i) confidence and curiosity scale of eight items developed by Jukes (2018); and (ii) degree of control the adolescent feels over their lives (scale 1-10).
Economic Empowerment measured by: (1) an index comprised of (i) proportion of time spent in domestic work or caring for others; (ii) proportion of time spent in paid work; (2) a binary indicator for being married; (3) a binary indicator for being engaged in paid work; (4) an index comprised of (i) a binary indicator for having money the adolescent controls; and (ii) a binary indicator that the adolescent has savings.
Psychosocial well-being measured by a binary indicator of minimal depression (PHQ-9<=4).
Social inclusion measured by a binary indicator that the adolescent has a trusted friend.
Gender attitudes measured by an index comprised of (i) a stereotypical roles scale comprised of eight items; (ii) a gender consciousness scale comprised of four items; and (iii) a gender malleability beliefs scale comprised of two items.
All composite indexes are constructed by standardizing each component to the mean and standard deviation in the control group and taking the unweighted mean of the standardized components.
Long term (2025)
Socioemotional Skills measured by an index comprised of: (i) a grit scale (generated from seven items following Alan, Boneva, and Ertac [2019]); (ii) a malleability scale (generated from six items following Alan, Boneva, and Ertac [2019]; and (iii) a growth mindset scale (generated from four items from the World Bank STEP survey [2014]).
Education Gender Attitudes measured by an index comprised of: (i) a scale of 5 education gender attitudes; (ii) an IAT score of gender bias around education and domestic work; and (iii) a scale of gender equitable responses to a vignette around early marriage, street harassment, gender roles, and schooling.
Gender attitudes measured by an index comprised of (i) a gender consciousness scale comprised of four items; and (ii) a gender malleability beliefs scale comprised of two items.
Economic Empowerment measured by: an index comprised of (i) a binary indicator for being engaged in paid work, school, or training; (ii) a binary indicator for having money the adolescent controls; (iii) a binary indicator that the adolescent has savings; (iv) a binary indicator for aspiring to a professional career; and (v) an index of time use control.
Mental health measured by (i) a binary indicator of minimal depression (PHQ-9<=4); (ii) a binary indicator of minimal anxiety; and (iii) degree of control the adolescent feels over their lives (scale 1-10).
Social inclusion measured by: (i) a binary indicator that the adolescent has a trusted friend; (ii) a binary indicator that the adolescent can express opinion to friends; (iii) a binary indicator that the adolescent can express opinion to adults; and (iv) a binary indicator that the adolescent has a role model.
All composite indexes are constructed by standardizing each component to the mean and standard deviation in the control group and taking the unweighted mean of the standardized components.
Cited by
No citation data available for this registration yet.