Thriving Together: Creating Social Support Groups for the Reintegration and Empowerment of Formerly Abducted Women in Northern Uganda

A total of about 200 women, of which 80-100 previously abducted and 100-120 never abducted, will take part of the study. We aim to scale up the study and involve more women in the future, conditional on the success of this initial investigation. In collaboration with local partners, our program will unfold in three phases. Phase 1 will consist of 10 half-day workshops in which the study participants, in groups of 16-20 women, will engage in a set of incentivized behavioral games and an extensive survey. The workshop will generate important baseline data that will be compared to endline data, which we will collect in 2021, six months after the completion of Phase1, during Phase 3 of the project. Phase 2 will consist of 12 by-monthly meetings, taking place between (tentatively) September and April, in which half of the study participants (30 formerly abductees and 45 never abducted women) will be invited to set, discuss and advise peers on achievable goals aimed at bettering their mutual conditions. These meetings will happen in groups of 5 women, under the guidance of a meeting facilitator. Phase 3 will follow the same methodology as Phase 1, i.e., it will consist of 9 workshops in which all the 150 women will participate in incentivized behavioral games and a survey.
August 2022 note: We secured more funding from The World Bank, allowing us to expand the sample size to 600 women, of which 300 formerly abducted and add mental health counseling to the treatments. When we pre-registered the pilot (before the baseline survey) we were planning to implement women support groups only - given the small sample size. We have now added: 1) group mental health counseling/therapy (150 women, i.e., 30 groups of 5 women) and 2) individual mental health counseling (150 women). This in addition to the support group treatment (150 women, i.e., 30 groups of 5 women). More funding will allow us to also collect hair samples just before the RCT and at endline (provided local cultural conditions permit).

2021-04-15

2021-11-01

Behavioral outcomes: grit, risk, prosociality, competitiveness, cognitive skills
Intervention outcomes: goals achieved, social capital, wellbeing, economic outcomes, hopes and aspitation (education, income, life)
Psychological wellbeing: depression, stress
Social support: instrumental, informational, and emotional social capital

Grit = from an experiment in which women are solving puzzles of different level of difficulty and asked (before final stage) whether they want to practice more.
Risk= from experimental game with unitary lottery protocol
Prosociality= from a series of dictator games towards 4 categories of recipients: woman formerly abducted, woman never abducted, random woman from same village, random man from same village
Competitiveness= from experimental game in which subjects are offered the option to be paid piece-rate (low amount) or to enter competitive tournament (higher amount) in a ball-tossing game
Cognitive skills= from standard Raven's tests
Intervention outcomes: goals achieved, social capital, wellbeing, economic outcomes, hopes and aspirations (education, income, life)
Psychological wellbeing: Edinburgh Scale

Stress reactions

Stress reactions: from 16 questions in the survey to be compiled in: fight, flight, tend, befriend categories.
Stress from hair samples (if the cultural conditions in the field will allow us to sample).

Phase 1. At baseline and endline (not covered by this proposal) we will conduct a survey and a series of incentivized lab in the field experiments during 10 half-day workshops, involving groups of 20 women each. The survey will serve to record women’s socio-economic background, demographics and, importantly, measures of experienced trauma and psychological well-being, such as stress, anxiety and depression (using internationally validates instruments), and several measures of social support from different characters in a woman’s network. The lab in the field experiments will serve to measure women’s attitudes, preferences and behaviors that we hypothesize are impacted by traumatic experiences and may be an obstacle/resource to individual life improvements. We will implement incentivized games to measure: 1) Cognitive Ability; 2) Grit; 3) Competitiveness; 4) Social shyness; 5) Preferences for giving to in-group versus out-group members; 6) Risk preferences. Furthermore, we will collect drawings from the women’s children to test for a potential intergenerational transmission of trauma. The initial workshops will provide us with baseline data for the second phase of the project.
Phase 2. We will then proceed with piloting a program ---we are not seeking PEP support at this time--- that utilizes peer-support and goal-setting to potentially impact in a positive manner the participants’ psychological well-being, social support, and, ultimately, their economic conditions. The program relies on bimonthly meetings for small groups of 6 women for the duration of 6 months organized around the setting of short-term goals aimed at bettering the women’s personal situations. This approach tests the principle that individuals could break cycles of poverty through light behavioral nudges and by harnessing the formation and/or the strengthening of local social capital.

Not available

At the end of all initial workshops and six-months group intervention, a lottery will be help by randomization via computer by field manager to randomly assign which participants will continue to the group meetings stage of the experiment.

Individual level and group level (5-6 women in each).

No

600

This is a behavioral study planned for a total of roughly 600 women.

N/A

Power calculation. Due to natural constraints and logistical necessities (i.e., the limiting factor availability of formerly abducted women) we will be now able to enroll in the project 300 FA women (treatment group) and 300 NFA women (control group). Based on parameters from our previous studies on individual choice competitiveness and prosociality experiments among victimized and not victimized subjects in Bosnia and Sierra Leone, with 300 subjects the minimum detectable effect size to reach the significance level of 5%, power of 80% (with a control mean of about 0.35, std1=0.48, std2=0.50) will be about 0.195. We will now have 600, so power will increase.