Impact evaluation of the Women for Women International Programme in reducing intimate partner violence and strengthening livelihoods amongst women in Afghanistan
The Women for Women International (WfWI) intervention is a year-long combined economic and social empowerment intervention for marginalised women survivors of conflict. The primary objective of the intervention is to support women to achieve four key outcomes: women earn and save money, women improve their health and well-being, women influence decisions in their homes and communities, women connect to networks for support. The intervention consists of several components such as:
i) Informational Training: focused on critical information about savings, health, rights and networking.
ii) Skill building: these are hands on sessions on numeracy, basic business skills and vocational skills.
iii) Resource provision: This component complements the information training and skill building and provides participants with financial support, referrals and opportunities for savings
The program is being evaluated through individually randomised control trial amongst women in Afghanistan. The primary outcome for the Randomised Control Trial was reduction in women's experience of intimate partner violence and improvement in women's mental health. Secondary outcomes include improved earnings and savings, improved women relationship with spouse and parent-inlaw, and increased involvement in decision making. The study population are unemployed women aged between 18 and 45, earning less than US$1.25 a day from selected districts in Kabul and Nangarhar. For practical reasons, a total of 5 districts were selected (3 in Kabul and 2 in Nangarhar). A total of 6 villages were recruited into the study. Women were individually recruited from the selected villages and randomly allocated to the 2 study arms after screening for eligibility. Randomisation of women to the study arms was done at village level. The study protocol provides further details on the study:- rationale, randomization, sample size and data collection ( <https://doi.org/10.1186/s12889-018-5029-1>).