Today’s subject matter is depressing. But, it is also worth continuing to read because, for many, the numbers will be eye-opening, and because with an improved focus on the types of programs targeted to communities suffering from these problems, we may be able to substantially improve the welfare of children and women.
In a study just published in Social Science and Medicine – Population Health, we evaluated the impact of Girl Empower – an intervention that aimed to equip adolescent girls with the skills to make healthy, strategic life choices and to stay safe from sexual abuse using a cluster-randomized controlled trial with three arms in Nimba County, Liberia. Among the target population of 13-14-year-old females, one in five had had their sexual debut at baseline and approximately 30% of the first sexual encounters among this sexually active group had been nonconsensual. Over one-third (37.3%) reported ever experiencing sexual violence: 7.8% had been physically forced to have sex, 8.4% had been non-physically pressured (coerced/persuaded) to have sex, 24.7% had had someone unsuccessfully attempt to have sex with them, and 28.9% had been touched in a sexual way.
There are a number of recent studies that examine the effectiveness of interventions to reduce sexual violence against women and children – gender-based and intimate partner. Such interventions include safety net programs, cash or in-kind transfers combined with behavior change communication (BCC) interventions, and girls’ clubs. See Buller et al. (2016), Peterman et al. (2017), Yount et al. (2017) for reviews and systematic review of reviews; see Bandiera et al. (2020) on girls’ clubs; Buchmann et al. (2018) for empowerment programs and incentives to delay marriage; and Roy et al. (2019) for BCC (for child nutrition) and cash/food transfers. [Also see these helpful slides by Hidrobo on cash transfers and IPV. You can find links to all the studies cited here in our paper that is linked in the previous paragraph.]
Our study differed slightly from some of these studies above. First, the target population was females in early adolescence (13-14 years old with mean age 13.7 at baseline). The mentoring program, designed and run by the International Rescue Committee (IRC), consisted of (i) Girl Empower life skills curriculum, facilitated by local female mentors weekly for approximately 40 weeks; (ii) Caregiver discussion groups, facilitated by IRC staff; (iii) Individual savings start-up for the girls; and (iv) Capacity building for local health and psychosocial service providers. Second, we wanted to assess the marginal gains from making participation in Girl Empower conditional on a small transfer to the caregiver of the girls. At $1.25 per session attended (or a maximum of $40 for perfect attendance to 32 sessions), the transfer amounts were less than 10% of the cost of the program per person. Our hypothesis was that these transfers may enhance program effectiveness at a small cost, by boosting program participation among early adolescents. We did not have a 2x2 factorial design: instead we had a control group, GE, and GE + cash – with 28 communities randomly allocated to each arm.
[My co-author Marie-France Guimond (IRC) wrote a very useful piece that summarizes the program and its impacts, accompanied with some stories of the participants, photos, and a video. Here, I talk a little bit more about some details and mechanisms.]
The first interesting finding was that attendance in both GE and GE+ were quite high – around 28 out of 32 possible sessions. Based on the previous experience of our collaborators from the IRC and Population Council running similar interventions elsewhere (as well as the attendance rates in ELA clubs in Uganda reported by Bandiera et al. 2020), we had anticipated much lower attendance rates in the GE arm, leaving room for plenty of improvement in GE+. We do not know whether it was the diligent work of IRC Liberia in rolling out the intervention or high demand in Nimba County, Liberia (a post-conflict/post-Ebola setting) that was responsible for the 88% participation rate. Whatever it was, this was not differential between GE and GE+, meaning that any differences in impacts between these two variants of the mentoring program cannot be due to increased participation in the combined intervention.
Figure 2 in the paper summarizes the impacts on seven pre-specified domains at the 24-month follow-up, i.e. more than one year after the mentoring interventions and the small cash transfers ended (Tables 1-7 then unpack each of these domains, so that the readers can see the impacts on sub-components. All indices are constructed such that positive effects are desirable outcomes):
At 24 months, the standardized effects of both GE and GE+, compared to control, on sexual violence, schooling, psychosocial wellbeing, and protective factors were small (β, ≤ 0.11 standard deviations [SD]) and not statistically significant at the 95% level of confidence. However, we found positive effects on indices of Gender Attitudes (GE: β, 0.21 SD, p<0.05; GE+: β, 0.23 SD, p<0.05), Life Skills (GE: β, 0.22 SD, p<0.05; GE+: β, 0.29 SD, p<0.01), and Sexual and Reproductive Health, or SRH (GE: β, 0.24 SD, p<0.01; GE+: β, 0.37 SD, p<0.01; F-test for GE=GE+: p=0.075).
Both GE and GE+ caused small improvements in the participants’ attitudes towards gender equity and towards IPV. Similarly, both GE variants caused small to moderate improvements in knowledge of HIV/AIDS and condom effectiveness, as well as financial literacy. However, GE+ was 50% more effective in reducing child marriage and number of sexual partners, delaying sexual debut, and increasing safe sex. The point estimate of the intention-to-treat effect on the SRH index in the GE+ arm was 0.37 SD, which is a substantial effect for these types of interventions. As the small amount of conditional cash offered to caregivers did not increase participation in the GE sessions as we had initially hoped, what caused the differential effect between the two variants on child marriage, teen fertility, and risky sexual behavior?
A potential mechanism is a pure income effect. Previous studies have shown that positive income shocks can reduce early marriages and teen pregnancies in both developing and developed countries. Study participants in both GE and GE+ received the US$2/month savings payments, for a modest sum of US$16 during the Girl Empower program, which were provided unconditionally, i.e. not tied to program participation. While the original study design envisioned the participation incentives to be small (US$0.50 per session attended) in order to minimize the possibility of a direct income effect on the outcomes of interest, the actual incentive payment was US$1.25 per session attended – due to a miscommunication between the study team and the IRC Liberia staff. The implied transfer to the caregiver of a regular program participant of approximately US$6/month constitutes more than 10% of per capita consumption in Liberia,41 and likely more in the study area, because Nimba is a poorer than average rural county in Liberia. As such, it represents a non-negligible increase to household income during the 8-month program, which could have directly affected child marriage and sexual behavior among GE+ beneficiaries.
However, the 24-month follow-up data was collected approximately one year after the last cash transfer payment, at which pointy one might have expected the effects of cash to have dissipated. For example, Baird et al. (2019) finds that such short-term reductions in marriage and fertility from a two-year cash transfer program disappeared very quickly upon the cessation of support. Hence, the evidence here also raises the possibility of an indirect income effect: perhaps, the additional income provided the space for the GE+ participants to better internalize the lessons from the mentoring program and to reinforce their newly obtained knowledge and skills towards sustained behavior change. All participating caregivers were engaged in sessions on the effects of sexual abuse and early marriage on adolescent girls, their families, and the community at large. While the caregivers in the GE+ groups were free to use the cash transfers as they saw fit, they were encouraged to use these funds in ways that would promote the adolescent girls’ safety and well-being. In this scenario, a positive interaction effect between the mentoring program and cash transfers would produce a larger effect than either mentoring or cash transfers alone. Sadly, since we did not include a study arm providing cash transfers alone (due to a variety of considerations and constraints at the study design phase more than five years ago), we cannot formally test this hypothesis.
Future trials may experiment with comparing Girl Empower with cost-equivalent cash transfers, in addition to Girl Empower plus cash as we have done here. It is also important, however, to identify how programs like Girl Empower can be modified or enhanced to reduce the sexual abuse of adolescent girls, the rates of which are unacceptably high, especially in conflict and post-conflict settings like the one studied here.