90360 enGender Impact: The World Bank’s Gender Impact Evaluation Database Women’s Empowerment in Action: Evidence from a Randomized Control Trial in Africa Update from 2012 paper: Empowering Adolescent Girls: Evidence from a Randomized Control Trial in Uganda Oriana Bandiera, Niklas Buehren, Robin Burgess, Markus Goldstein, Selim Gulesci, Author(s) Imran Rasul and Munshi Sulaiman Contact mgoldstein@worldbank.org Country Uganda Gender Based Violence, Voice and Agency, Economic Opportunities and Access to Organizing Theme Assets Status Project is ongoing, 2 year evaluation is complete Intervention Category Training, Cash Transfer Sector Social Development Women in developing countries are disempowered relative to their contemporaries in developed countries. High youth unemployment and early marriage and childbearing interact to limit human capital investment and enforce dependence on men. In this paper, we evaluate an attempt to jump-start adolescent women’s empowerment in the world’s second youngest country: Uganda. In this two- pronged intervention, adolescent girls are simultaneously provided vocational training and information on sex, reproduction, and marriage. Relative to adolescents in control communities, after two years the intervention raises the likelihood that Abstract girls engage in income generating activities by 72% (mainly driven by increased participation in self-employment) and raises their monthly consumption expenditures by 41%. Teen pregnancy falls by 26% and early entry into marriage/cohabitation falls by 58%. Strikingly, the share of girls reporting sex against their will drops from 14% to almost half that level and preferred ages of marriage and childbearing both move forward. The findings indicate that women’s economic and social empowerment can be jump-started through the combined provision of vocational and life skills and is not necessarily held back by insurmountable constraints arising from binding social norms. Gender Connection Gender Focused Intervention Girls vocational knowledge or skills, labor force participation, psychological agency, Gender Outcomes reproductive health Clustered Randomized Control Trial - clustered at village level. (difference in IE Design difference analysis) The authors evaluate the Empowerment and Livelihood for Adolescent girls program operated by the Ugandan NGO, BRAC Uganda. The program initiated "Adolescent Development Clubs" that aimed to provide vocational skill training Intervention and life skills to adolescent girls aged 14-20. The club is led by a female mentor. The clubs provide courses that provide income generating skills with focus geared towards micro-enterprises. Life skills courses cover topics related to sexual reproductive health, menstruation, pregnancy, family planning, rape, conflict Last updated: 2 September 2014 1 enGender Impact: The World Bank’s Gender Impact Evaluation Database resolution, and leadership. Participation in the club and in each specific course is voluntary. The club also hosts recreational activities such as plays, games and opportunities for socializing. Intervention Period The majority of clubs were established between June and September 2008 BRAC Uganda chose 10 branch offices to receive the program, 5 from rural regions and 5 from Semi-Urban regions. In each branch 15 communities were identified; 10 of which randomly received the treatment. Overall, there were 150 communities in the sample, 100 in the treatment group and 50 in the control group. From each of Sample population the communities, 40 girls were randomly selected to be surveyed. There were 5,966 girls surveyed: 3,964 resided in treatment communities while 2,002 girls lived in control communities. 82% of the girls were tracked for the follow up survey. The baseline characteristics between the girls who were tracked and who atritted are not significantly different. For the Intention to Treat Effect, the comparison condition is the control communities that did not have clubs. For the Average Treatment Effect, the Comparison comparison condition is girls who did not join the clubs. It is important to note that conditions almost 5% of girls in control communities did join clubs in treatment communities. Additionally, approximately 21% of girls in treatment communities actually participated club activities. Randomization occurred at the community level. Average Treatment Effects are analyzed at the individual level for girls who actually participated in the clubs. Unit of analysis Intention to treat effects are analyzed at the village level for girls who joined and did not join the clubs. The baseline survey was administered between March and June 2008, the follow up Evaluation Period survey was administered between March and June 2010 Relative to adolescents in control communities, after two years the intervention raises the likelihood that girls engage in income generating activities by 72% (mainly driven by increased participation in self-employment) and raises their monthly Results consumption expenditures by 41%. Teen pregnancy falls by 26% and early entry into marriage/cohabitation falls by 58%. Strikingly, the share of girls reporting sex against their will drops from 14% to almost half that level and preferred ages of marriage and childbearing both move forward. Many of the outcome measures are self reported; it is possible the girls are just saying what they think the interviewers want to hear. However, girls report Primary study increased condom use and not other contraception that is discussed in the limitations intervention. This result corresponds to the availability of condoms and lack of availability of other contraception in local areas. Bank Netherlands, MasterCard, Nike, The Gender Action Plan, Improving Funding Source Institutions for Pro-Poor Growth at DFID Bandiera, O., Buehren, N., Burgess, R., Goldstein, M., Gulesci, S., Rasul, I., Sulaiman, M. (2014). Women’s Empowerment in Action: Evidence from a Reference(s) Randomized Control Trial in Africa. International Growth Centre. Working Paper, February Link to Studies http://www.theigc.org/sites/default/files/Bandiera%20et%20al%202014.pdf Microdata Last updated: 2 September 2014 2