Please find below projects funded by IPA's Intimate Partner Violence (IPV) Initiative.
 

Addressing Intimate Partner Violence: Evaluating Interventions with Male Perpetrators vs Female Victims, with Regents of the Utrecht University in the Netherlands
Researchers: Sonia Bhalotra, José Raimundo Carvalho, Pietro Biroli, Karlijn Morsink, Joseph Vecci
Country: Brazil
Partners: Fundación Capital, Departamento para la Prosperidad Social
Type of Research: Randomized Evaluation

In Brazil, 1,000 couples that include male IPV perpetrators will be randomly selected into one of four groups: (1) male-perpetrators mandated by the court to attend 12 weekly group meetings (“Gender Reflexive Groups”), (2) an empowerment and mindfulness training for female partners of male perpetrators, (3) both male and female programs, or (4) neither program. The study will look at how these training programs—separately and concurrently—influence further IPV, reported IPV recidivism, and mental health of the participants.

Business Training, Soft Skills, Women’s Economic Empowerment and IPV among Female Microentrepreneurs, with Grupo de Análisis para el Desarrollo in Peru
Researchers: Suzanne Duryea, Verónica Frisancho, Martin Valdivia
Country: Peru
Partner: Grupo de Análisis para el Desarrollo in Peru
Type of Research: Randomized Evaluation

In Peru, researchers will randomly select 1,913 women clients from Financiera Confianza’s group lending program into four alternative groups: (1) hard skills training on running a business, (2) soft skills training focused on entering the labor market, (3) hard and soft skills training, and (4) neither program. This study aims to provide insight into how hard and soft skills acquisition impact business outcomes and, subsequently, IPV.

Can Soft Skills Training Improve Enterprise and Employment Growth in Tanzania? A Randomized Evaluation of the STRYDE 2.0 Program in Tanzania
Researchers: Margherita Calderone, Nathan Fiala, Lemayon Melyoki, Annekathrin Schoofs, Rachel Steinacher
Country: Tanzania
Partner: TechnoServe
Type of Research: Randomized Evaluation

Numerous governments in low- and middle-income countries, such as Brazil and Mexico, have adopted conditional cash transfer (CCT) programs as a social safety net, but most recipients of these transfers have little or no experience with formal financial products. To address this financial capability gap for recipients of Colombia’s government CCT program, Fundación Capital designed LISTA, a program founded on the notion of “freeing financial education” from the classroom via the use of tablet-based apps. Researchers collaborated with Fundación Capital and the Government of Colombia to conduct a randomized evaluation of LISTA to study its impact on financial knowledge and behavior. LISTA had significant positive impacts on participants’ financial knowledge, attitudes, practices, and performance. Many of these impacts persisted after two years, while some waned over time. The long-term impact on participation in the formal financial system was limited, suggesting that financial education alone may not be sufficient to increase formal financial inclusion.

Cash Transfers and Role Models: Evaluating the Impact of Economic and Psychological Empowerment on Intimate Partner Violence in Kenya
Researchers: Mahreen Mahmud, Kate Orkin, Emma Riley
Country: Kenya
Partner: GiveDirectly
Study Type: Randomized Evaluation

Women across developing countries report limited direct control over economic resources and experience high rates of physical and sexual intimate partner violence. In rural areas of Siaya county in western Kenya, researchers are evaluating the impact of alleviating economic or psychological constraints individually or in combination on women’s control of household resources and experience of intimate partner violence.

Rates of intimate partner violence are high—61% of women in the study control group reported that their partner had committed at least one act of physical violence in the last 12 months. Nearly 50% of men and women aged 15-49 agreed that a husband is justified in hitting or beating his wife for at least one of the five specified reasons (United Nations, 2015). The high rates of violence and attitudes that violence is acceptable point to unfavourable social norms for women. However, women do own some share in household assets—an average of 21% of the value of non-land household assets, primarily poultry and small durables, in the study control group—and there are no major restrictions on their movement outside the household. To learn more, read IPA's project summary.

Effects of Incentives for Seasonal Migration on Intimate Partner Violence in Bangladesh
Researchers: Ahmed Mushfiq Mobarak, Alejandra Ramos
Country: Bangladesh
Type of Research: Randomized Evaluation

This paper studies the effects of seasonal migration incentives on intimate partner violence. There are at least three reasons why migration may have consequences for violence among the ultra-poor. First, migration induces a positive income shock that lessens poverty-stressors and, with it, violence. Second, the extra-income associated with the migration of the male spouse increases his bargaining power and lessens woman's say in the household decisions. Third, when the man migrates, the woman spends less time with her potential perpetrator and faces a lower risk of victimization. To address the net effect of migration on violence, the researchers collected data on intimate partner violence among the female respondents of the No Lean Season intervention 2017 program-year endline survey. The results indicate that 77% of these women have experienced violence at the hands of an intimate partner during their lifetime, and almost one in every two has been physically or sexually abused by her partner in the last six months. The research team's analysis suggests that migration reduces violence. As such, women of households receiving interest-free migration loans are less likely to experience physical or sexual violence. Violence, however, is higher in villages that did not receive the migration incentive but are surrounded by villages that did receive it. The researchers hypothesize this increase is explained by the fact that, when the male spouse is less likely to migrate, the woman spends more time with the perpetrator of violence. The evidence provided in this paper is in line with exposure reduction theory and suggests that seasonal migration has the potential to improve well-being by providing women with periods of reduced violence throughout the year.

Engaging Women in Mobile Money Markets in Bangladesh
Researchers: Lore Vandewalle, Xavier Giné, Jessica Goldberg
Country: Bangladesh
Partner: bKash
Type of Research: Randomized Evaluation

Women are underrepresented as providers and as users of mobile money services in Bangladesh: there are almost no female agents and as of 2017, the gender gap in account usage stood at 29%. Vandewalle, Giné and Goldberg will conduct an RCT to evaluate two interventions that interfere with the supply and demand of mobile financial services for women. The project has three specific objectives. First, the researchers will estimate the differential economic impact of hiring male and female mobile banking agents on workers and on business development. Second, the project allows studying the effect of experiencing female employment on attitudes towards female labour force participation. Finally, they will test whether limited access to female agents forms a gender-specific constraint to being a mobile money customer, to account usage and to taking up loans.

At the supply side, the researchers collaborate with bKash—a major provider of mobile money—to increase the number of women working as mobile money agents. To do so, businesses will receive a subsidy to hire an employee to manage the bKash transactions. The stores will be randomly allocated to one of three groups: control, male employee, or female employee. On the demand side, they will collaborate with BRAC to study the digitisation of microfinance loan repayments by female clients. To do so, BRAC will identify 90 branches that will be randomly allocated to one of three arms: control, voluntary digitisation or mandatory digitisation. Microfinance groups covered by branches in the voluntary digitisation arm will decide whether to reimburse loans in cash or whether they prefer to repay their loans using mobile money. The groups in the control arm will continue reimbursing in cash, and those in the mandatory arm will be required to repay their loans using mobile money. The cross randomisation of these two interventions lets us test whether female agents increase access for female customers.

Improving the effectiveness of financial services targeting women and female micro-entrepreneurs in particular is a key policy question. This reseach can contribute to policymaking in this area in several ways. First, the digitisation of loan reimbursements will naturally increase the pool of women who have mobile money accounts and who gain experience using them. This will positively influence account usage, which the literature shows is economically empowering. This may also impact loan usage by women and influence female led businesses in particular. Second, encouraging the wider employment of women as mobile money agents may create opportunities for other women—and not only for BRAC clients—to use mobile money. Third, the direct employment of women in this project may change attitudes and reduce barriers to female labour force participation.

Escaping Poverty - The Effect of Cognitive Behavioral Therapy on Intimate Partner Violence Study in Ghana
Researchers: Nathan Barker, Gharad Bryan, Dean Karlan, Angela Ofori-Atta, Christopher Udry, Julia Vaillant
Country: Ghana
Partner: Heifer International
Type of Research: Randomized Evaluation

The Graduation Approach, a model for holistic livelihoods programs, has been proven to have lasting impacts on poor families’ income, assets, food security, and mental health, but these programs can be expensive to implement. Innovations for Poverty Action is working with researchers in Ghana to evaluate the impact of different variations of the Graduation program—without coaching visits, with cognitive behavioral therapy, without additional services, and a cash transfer only—on household income, consumption, assets, psychological health, and other measures of well-being.

Evaluation of a Holistic Entrepreneurship and Leadership Development Program at the Secondary Level in Uganda: Effects on Intimate Partner Violence (Educate!), Uganda [Round 1]
Researchers: Paul Gertler, Laura Chioda
Country: Uganda
Partner: Educate!
Type of Research: Randomized Evaluation

Education systems in Africa face many challenges in equipping students with the skills needed to be successful in adulthood. While secondary schools are largely fee-based and require substantial investments for poorer families, graduates face few formal employment opportunities and often lack the entrepreneurial skills required to start or operate their own small businesses. Teaching students the skills required to be successful in life or to compete in the formal labor market has the potential to reduce youth unemployment, drive economic growth, and reduce poverty. But whether such skills can be taught is an open question and a policy priority. Researchers (Dana Carney (UC Berkeley), Laura Chioda (World Bank), David Contreras (UC Berkeley), and Paul Gertler (UC Berkeley)) partnered with IPA and Educate! to conduct a randomized evaluation of the Educate! Experience program, a leadership and social entrepreneurship skill development program for youth in their last two years of secondary education. The program was successfully implemented during the 2012-2013 school years. A follow-up survey was conducted in 2017 to analyze the impact of the program on students’ skills, economic outcomes, educational attainment, community involvement, and gender empowerment outcomes in Uganda. The study also investigated whether the program impacted gender social norms and intimate partner violence (IPV)-related outcomes. Four years after the intervention, researchers found that the Educate! Experience program had strong and meaningful impacts on Educate! graduates’ intra- and inter-personal soft skills such as creativity, grit, ability to manage stress, persuasion, and self-efficacy. Educate! graduates appear to focus more on long-term goals and report being more in control of aspects of their lives, as well as more empowered to implement actions towards achieving their plans.

Extending a Large-Scale Conditional Cash Transfer Evaluation to Include Impacts on Intimate Partner Violence, Dominican Republic
Researchers: Paul Gertler, Seth Garz
Country: Dominican Republic
Partner: Gabinete de Coordinación de Políticas Sociales
Study Type: Randomized Evaluation

In countries with conditional cash transfer programs, volunteers or modestly-paid community workers often play a significant role in supporting beneficiaries in complying with the program requirements upon which their cash transfers are conditioned. Despite the important role that these community workers play, there is little research on their impact on the well-being of beneficiaries or on how best to recruit and select community workers in order to maximize productivity and beneficiary outcomes. In partnership with the Government of the Dominican Republic, researchers are evaluating the impact of conditional cash transfer program community workers on beneficiary wellbeing. Researchers will also distinguish between the impacts of cash transfers alone and community worker visits, compare the impacts of paid versus volunteer community workers, and evaluate the impact of different community worker recruitment techniques.

The Impact of a Gender Targeted Graduation Model and Norm-Transforming Intervention on Intimate Partner Violence in Malawi
Researchers: Michael King, Tara Bedi, Julia Vaillant, Markus Goldstein, Alejandra Ramos
Country: Malawi
Partner: Concern Worldwide

In Malawi, Concern Worldwide is implementing a ‘graduation model’ intervention with 3,300 households in 200 villages. This involves a cash transfer for the extreme poor complemented with skills training, access to savings accounts, and an asset transfer. The research study will estimate the impact of (1) the program delivered to women, (2) the program delivered to men, and (3) the program combined when delivered with a joint couples’ empowerment training.

Impacts of Skilling and Employment Opportunities on Female Rural-to-Urban Migrant Workers and their Families: A Randomized Controlled Trial, with Good Business Lab in India
Researchers: Achyuta Adhvaryu, Anant Nyshadham, Priyanka Sarda, Huayu Xu
Country: India
Type of Research: Randomized Evaluation

Labor-intensive manufacturing is growing rapidly in developing countries. Yet significant wage gaps exist both across geographic and gender boundaries: the urban-rural wage gap is as high as 45% in some areas of India. Industries that specifically carry disproportionate amounts of female employees, such as garment production, could provide a way to enhance successful migration and provide important job skills, and thus begin to narrow gender and wage gaps in labor participation. This project will analyze and address methods to alleviate the barriers to successful rural-to-urban migration for women.

Partnered with one of the world’s largest garment manufacturers in India, researchers will run a randomized controlled trial that aims to facilitate Indian women’s rural-urban migration and assess impacts on the wellbeing of workers and their families. Ten vocational training centers will be established at randomized locations across the Indian states of Karnataka and Andhra Pradesh, covering women living in more than 1,200 villages. The centers will provide vocational skills and a guaranteed employment match in urban Bangalore following training. Extensive panel data will be collected on women from nearly 3,000 households, and impacts on women’s experiences of increased bargaining power, well-being, and annual income will be assessed.

Measurement Error in Self-Reported Data and Solutions: Experimental Evidence for Intimate Partner Violence in Peru
Researchers: Jorge Agüero, Úrsula Aldana, Erica Field, Verónica Frisancho, Javier Romero
Country: Peru
Type of Research: Randomized Evaluation

The prevalence of intimate partner violence (IPV) is generally measured with direct questions, such as the widely used Domestic Violence module of the Demographic and Health Surveys (DHS). Since IPV prevalence is a sensitive topic, these data may suffer from non-random measurement error, raising concerns about the validity of descriptive studies, impact evaluations, and policy choices. This study tests new methods that provide greater degrees of privacy when trying to measure IPV prevalence but without increasing the cost of collecting data. Researchers randomized a sample of over 8,000 women in rural and peri-urban Peru into three groups. Participants in two of the groups will respond to IPV-related questions that provide varying levels of anonymity compared to DHS-type questions. These surveys provide either partial anonymity, relying on secret reporting through a ballot, or full anonymity through an indirect questioning technique The third group—a comparison group—will respond to the face-to-face standard DHS questionnaire on IPV. Researchers find that reported prevalence rates increase with the level of privacy, particularly for the most severe violence episodes. Even though rates are higher when measured this way, the estimates are less precise when compared to those obtained from the ballot. These strategies may extend to other sensitive topics and behaviors in health and beyond.

The Roles of Contraceptive Prices, Perceived Mortality Risk, and Social Norms in Fertility Choices in Burkina Faso
Researchers: Pascaline Dupas, Seema Jayachandran, Adriana Lleras-Muney, Idrissa Ouili, Pauline Rossi
Country: Burkina Faso
Partners: United States Agency for International Development (USAID), Development Media International, Global Innovation Fund, Maxmind
Type of Research: Randomized Evaluation

Mass media reaches a large and growing share of the population in developing countries, but can it be used to tackle poverty and change behaviors? IPA partnered with researchers and Development Media International to evaluate the impact and cost-effectiveness of an intensive, two-and-a-half-year mass media radio campaign in Burkina Faso that promoted family planning and modern contraception use.  The campaign led to a 6-percentage point increase in the modern contraceptive prevalence rate, largely by reducing misconceptions about potential side effects. The impact on contraception translated into a 10 percent reduction in births and an increase in women’s self-assessed health and well-being. Researchers estimate that the nationwide campaign scale-up led to 225,000 additional women using modern contraception, at a cost of US$7.7 per additional user.

Skills to Prevent Intimate Partner Violence in Uganda: Better Partners and Better Choices (Educate!), Uganda [Round 2]
Researchers: Paul Gertler, Laura Chioda
Country: Uganda
Partner: Educate!

Building on previous findings from the first round research, researchers will conduct a seven-year follow-up of the Educate! youth entrepreneurship and mentoring program. This long-term follow-up seeks to understand to what extent reductions in IPV are explained by partner selection. In addition, the study will assess whether the program lead to long-term reductions in IPV among female participants and the spouses of male participants.

Tablet-based Financial Education and Intimate Partner Violence in Colombia
Researcher: Matthew David Bird
Country: Colombia
Partners: Fundación Capital, Departamento para la Prosperidad Social
Type of Research: Randomized Evaluation

Numerous governments in low- and middle-income countries, such as Brazil and Mexico, have adopted conditional cash transfer (CCT) programs as a social safety net, but most recipients of these transfers have little or no experience with formal financial products. To address this financial capability gap for recipients of Colombia’s government CCT program, Fundación Capital designed LISTA, a program founded on the notion of “freeing financial education” from the classroom via the use of tablet-based apps. Researchers collaborated with Fundación Capital and the Government of Colombia to conduct a randomized evaluation of LISTA to study its impact on financial knowledge and behavior. LISTA had significant positive impacts on participants’ financial knowledge, attitudes, practices, and performance. Many of these impacts persisted after two years, while some waned over time. The long-term impact on participation in the formal financial system was limited, suggesting that financial education alone may not be sufficient to increase formal financial inclusion.

Understanding the Impacts of Cash Transfers and Reducing IPV Measurement Error: Experimental Evidence of Audio Computer-Assisted Self-Interviewing (ACASI) in Liberia and Malawi
Researchers: Shilpa Aggarwal, Jenny Aker, Dahyeon Jeong, Naresh Kumar, David Sungho Park, Jonathan Robinson, Alan Spearot
Countries: Liberia, Malawi
Type of Research: Randomized Evaluation

In a cross-country study in Liberia and Malawi, researchers will interview 4,900 women across three randomized evaluations using either audio computer-assisted self-interviewing (ACASI) or “standard” face-to-face interviews. Researchers will assess if different methods of administering the same IPV survey changes women’s willingness to disclose whether they were victims of violence.