Education is not only a human right, but also a powerful tool for women's empowerment and a strategic development investment. There is a clear multiplier effect to educating girls; women who are educated are healthier, participate more in the formal labor market, earn more income, have fewer children, and provide better healthcare and education to their children compared to women with little or no education (Klugman et al. Citation2014). The benefits of education thus transmit across generations as well as to communities at large. Where girls have greater educational and economic opportunities, they are more likely to pursue those opportunities than to have children in their teenage years. Yet a host of structural, social, and financial barriers prevent girls’ enrollment and completion of both primary and secondary schools.
Over the past two decades, uneven progress has been made toward gender equality in global education goals. The most recent UNESCO data show that of 161 countries, 60 percent have achieved gender parity in enrollment at the primary school level, compared to only 38 percent of countries at the secondary level. Major gender imbalances persist, especially in low-income countries, just 20 percent of which have reached gender parity at the primary level, and only 10 percent at the secondary level.Footnote1 This is a major global challenge. At current rates of progress, UNESCO projections show that it will take on average 63 years longer for the poorest girls in low-income countries to achieve parity in lower secondary school completion compared to the richest boys (UNESCO and UNGEI Citation2014). Across sub-Saharan Africa, for example, the richest boys are expected to reach this target by 2021, while the poorest girls will not reach it until 2086. Even among the children in the richest households in this region, there is a gender gap of at least 8 years.
Millions of girls around the world are barred from entering and completing their education due to social norms about gender roles that limit women and girls’ agency—the choices they are allowed to make about their own lives. What are these norms? Simply put, norms are social expectations and prescriptions reflected in the informal rules, beliefs, and attitudes of a society. For example, analysis of Gallup World Poll data across 11 countries in Europe and Central Asia shows that on average one quarter of men and an equal share of women believe that it is morally wrong for a woman to delay getting married to further her education or to start a career. The range is from less than 10 percent in Latvia to over 50 percent Azerbaijan (Klugman et al. Citation2014). Such norms are often also reflected in formal structures of society, including discriminatory laws. As documented by Women Business and the Law, 128 countries around the world have at least one law that is discriminatory against women and girls, 56 have more than five, and 28 have more than 10 (Klugman et al. Citation2014).
Such adverse norms are reinforced by sanctions that can be positive or negative, imposed by people belonging to the same reference group or by the state (cited in Klugman et al. Citation2014; Mackie and Lejeune Citation2009; Muñoz Boudet, Petesch, and Turk Citation2013). These norms vary across regions and countries, but in the context of education, adverse norms can include attitudes and practices that devalue girls’ education and advancement, sexism in the curricula, violence in and around schools, early childbearing, and early marriage (Gennari et al. Citation2015).
Across all developing countries, one out of three girls is married before her 18th birthday (UNICEF Citation2013). These averages mask considerable variation across regions and countries. Among the 111 countries with data, the prevalence of child marriage ranges from 2 percent in Algeria and Libya to as high as 74 and 75 percent in Niger and Bangladesh, respectively (Parsons and McCleary-Sills Citation2014a; Raj and Boehmer Citation2013). Regional rates of child marriage are highest in South Asia and sub-Saharan Africa, and relatively low in Latin America. While the prevalence of child marriage in India is not the highest recorded, the sheer size of its population means that the country accounts for one-third of the world's child brides, the highest number of any country in the world (Klugman et al. Citation2014; see also Nguyen and Wodon Citation2015a on global trends in child marriage).
This picture persists despite the fact that an increasing number of governments now deem child marriage illegal. Such laws are often not effective, either because they are not enforced due to countervailing norms, or because there are widespread exceptions. Customary or religious law in some countries makes exceptions for the minimum age of marriage, allowing a girl to marry before age 18 with her parents’ permission (Vogelstein Citation2013). Beyond being a violation of girls’ human rights, child marriage is one of the most pervasive and earliest forms of gender-based violence (Coomaraswamy Citation1999; UNICEF Citation2013). Child marriage has a range of negative repercussions. Recent analysis shows strong correlations with the risk of intimate partner violence (Hanmer and Klugman Citationforthcoming). It has also been shown to limit women's health, decision-making, labor force participation, and other options for exercising their agency (Hindin and Fatusi Citation2009; see also Parsons et al. Citation2015 for a review). This includes their ability to enroll in and complete school, which is a fundamental predictor of their later development and achievement.
This articleFootnote2 examines the link between girls’ education and child marriage, including the important social and structural barriers that prevent girls’ school completion and the protective role of education on girls’ voice and agency. It presents findings from two important sets of World Bank Group (WBG) studies: (1) new analyses of Demographic and Health Survey (DHS) data from more than 50 countries (Klugman et al. Citation2014), and (2) a synthesis of findings from 27 recent impact evaluations (IEs) (Parsons and McCleary-Sills Citation2014a, Citation2014b).
Girls’ Agency
Thirty years ago, Sen (Citation1985, 205) defined agency as what a person is free to do and achieve in pursuit of whatever goals or values he or she regards as important. Agency invokes an ability to overcome barriers, to question or confront situations of oppression and deprivation and, as individuals or together with others, to have influence and be heard in society. Agency has intrinsic value; it is important in its own right regardless of whether or not its exercise leads to increased well-being. As articulated by Kabeer (Citation2008), women's agency leads to empowerment when its exercise questions, challenges, or changes regressive norms and institutions that perpetuate the subordination of women.
The impacts of education on women's agency are potentially far reaching, as noted above. An example from rural Bangladesh illustrates: when girls were asked how education had made their lives different from their mothers’, they replied that it had “helped them find a voice,” allowed them “to have a say,” to “speak and be listened to” (World Bank Citation2008, 40). Child marriage reflects lack of agency in several ways. First, the individual's basic right to “free and full consent” to marry at “full age,” as enshrined in the Universal Declaration of Human Rights, is denied. Second, child marriage is directly linked to girls’ increased risk of other forms of gender-based violence, including physical and emotional abuse by husbands and other family members, as well as marital rape and sexual coercion (Klugman et al. Citation2014; UNICEF Citation2008). Third, and the focus of this section, there is persuasive evidence linking child marriage and educational deficits.
Lack of education is both a risk factor and an outcome of child marriage (Klugman et al. Citation2014). Evidence from Bangladesh and sub-Saharan Africa suggests that women who married early are over five percentage points less likely to be literate and over eight percentage points less likely to have any secondary education (Field and Ambrus Citation2008; Nguyen and Wodon Citation2015b). Every year of early marriage significantly reduces the probability of girls completing secondary school (Nguyen and Wodon Citation2015b). Although it is difficult to disentangle causation, the associations are clear and consistent. Other research has underscored that education is an effective protective mechanism against early marriage. Across 18 of the 20 countries with the highest prevalence of child marriage, girls with no education were up to six times more likely to marry are than girls with a secondary education (ICRW Citation2006). The longer a girl stays in school, the less likely she is to be married before the age of 18 and have children during her teenage years.
The evidence discussed in this section draws on our analyses, as presented in the book Voice and Agency: Empowering Women and Girls for Shared Prosperity (Klugman et al., Citation2014), which explored how women and girls experience multiple deprivations of agency, including limited control over household resources, child marriage, and attitudes that expose women to increased risk of violence (Klugman et al. Citation2014). Using the latest available DHS data from 54 countries covering 78 percent of the developing world's population, we found that 4 in 5 women experience at least one of these deprivations of agency (Figure 1). More than 1 in 2 women had been married before age 18, and 1 in 4 of these women also condoned intimate partner violence and lacked control over household resources.
Published online:
23 October 2015Figure 1. Overlapping Agency Deprivations, 54 Developing Countries. Source: Klugman et al. (Citation2014, Figure 1.3).
Figure 1. Overlapping Agency Deprivations, 54 Developing Countries. Source: Klugman et al. (Citation2014, Figure 1.3).
It was not possible to establish causation using the available DHS data, but there were some interesting and consistent associations. We found that girls living in poor households are twice as likely to marry before the age of 18 compared with girls in wealthier households, as are rural girls compared with those from urban areas. This is consistent with other studies finding that the girls most likely to marry early are those with the least education and lowest economic status (Raj Citation2010). In terms of the implications of early marriage, our analysis found that women who married after the age of 18 were more likely to feel able to refuse sex with their partners compared to those who married early, and that this likelihood increases with each year that marriage is delayed beyond 18. Being married as a child is also associated with a 20 percent higher probability of experiencing intimate partner violence.
A central finding of the patterns of voice and agency across the 54 developing countries was that education, especially at the secondary level and beyond, significantly reduces women's and girls’ risk of experiencing agency deprivations. This is shown in Figure 2. For example, 90 percent of women with no more than a primary education experience at least one of the listed constraints, compared with 65 percent of women with at least secondary education. Nearly 20 percent of women with only a primary education experience all three deprivations, compared to five percent with secondary schooling and higher. We also see that almost one in five rural women with a primary education experience all three deprivations compared with 1 in 50 urban women with a higher education.
Published online:
23 October 2015Figure 2. Share of Women with Agency Deprivations. Source: Klugman et al. (Citation2014, Figure 1.5).
Figure 2. Share of Women with Agency Deprivations. Source: Klugman et al. (Citation2014, Figure 1.5).
Further analysis of these data explored the domains of sexual and reproductive health and rights, household decision-making and freedom of movement (Hanmer and Klugman Citation2015). Highlights include a series of negative findings around marriage: being married—either monogamous or polygamous marriage—reduces sexual autonomy, and being married before age 18 further reduces sexual autonomy.Footnote3 Women married early are also more likely to have their movement restricted by their husbands. These analyses also confirmed that education plays a positive role, and the possibility of threshold effects. Controlling for other factors, the analysis reveals that the completion of secondary education and above is associated with doubling or tripling of measures of sexual autonomy (Hanmer and Klugman Citation2015). The husband's education is also associated with a higher likelihood of his partner having sexual autonomy, albeit to a lesser extent. Having primary or secondary education does not reduce the probability of mobility restrictions, but women with higher education are less likely to be restricted. Education is also associated with a higher probability that women can make decisions about large purchases, though the effect is less marked.
Girls’ Agency and Gender Equality in Education
Progress in primary and secondary school enrollment has accelerated over the past two decades toward the Dakar Framework for Action goals established in 2000, and the Millennium Development Goals. For example, globally in 2000 there were only 81 girls for every 100 boys completing primary school. In 2010, this ratio was 93 girls for every 100 boys (UNESCO Citation2015). Two-thirds of all countries have reached gender parity in primary enrollment and almost half have done so at the secondary level (UNESCO Citation2015).
Across the globe, 58 million children and 63 million adolescents are not in school, and girls still represent more than half of out-of-school children—31 million primary school age girls and 34 million lower secondary school age girls are not enrolled in school (UNESCO Institute for Statistics and UNICEF 2015). An estimated 43 percent of all out-of-school children will never go to school, but the gender disparities are clear—48 percent of girls compared to 37 percent of boys in this group will never enroll. Gender disparities in secondary school enrollment remain widest in sub-Saharan Africa and South and West Asia (84 and 93 girls per every 100 boys enrolled), while the gap actually favors girls in Latin America and the Caribbean (107 girls per 100 boys enrolled). While the gender gap in enrollment is narrowing, other important gender gaps remain. Even when enrolled, girls continue to lag substantially behind boys in secondary completion rates. In Africa and South Asia for example, boys remain 1.55 times more likely to complete secondary education than girls (UNESCO Citation2015). As underlined above, there are major gaps between children in the richest and worst off households, with the latter lagging significantly, and poorest girls being farthest behind (UNESCO Citation2014).
So what works to attract and keep girls in school? Here we review the evidence. Over the years, many countries have experimented with multi-sectoral approaches to overcome the persistent infrastructure and cultural barriers. McCleary-Sills and Benveniste (Citation2015) highlight a number of promising strategies, including the following:
Providing scholarships or cash transfers to girls. Bangladesh pioneered this solution decades ago to make schooling more affordable for girls and indirectly reduce child marriage. Today, in Bangladesh as many or more girls than boys attend primary and secondary schools, and female literacy has soared from 65 percent in 1999 to 83 percent in 2012.
Hiring more female teachers. The lowest proportion of female teachers is in countries with wide gender disparity in enrollment. Shifting this balance can be a way to attract girls to school and, in turn, encourage greater number of educated young women to choose careers as teachers. Yemen has made important investments in this area, especially by training and hiring hundreds of female teachers to work in rural areas who can be positive role models for girls (World Bank Citation2013).
Reducing distance to schools, especially in areas where safety is an issue for girls, by building more schools or improving public transportation, as illustrated by the example of Afghanistan below.
Building separate toilet blocks for adolescent boys and girls in schools. By building school latrines, India significantly increased girls’ enrollment, particularly for puberty-age girls.
Carrying out gender-sensitivity training for teachers. In Africa, the Forum for African Women Educationalists has trained over 6,600 teachers since 2005 in developing and using gender-equitable learning materials and preventing sexual harassment in schools.
Other important measures also have protective elements. These range from ensuring adequate and safely located latrines and sanitation facilities for girls, well-lit and visible access routes to and from school, pairing female and male teachers and teaching assistants in the classroom, and practicing restorative discipline that avoids physical, psychological, and verbal violence. Many girls’ schools in Afghanistan and Pakistan, for example, have made it a priority to protect students from physical harm, even from extreme threats of being attacked with acid (Reyes, Kelcey, and Dias Varela Citation2013). By erecting school boundary walls, providing community supervision, and rallying the support of religious community leaders, schools and communities working together have been able to make protecting students from gender-related violence a priority (Glad Citation2009).
These strategies echo recent studies conducted by the WBG, which find that tackling girls’ access to quality education requires multi-faceted approaches. Recent analyses of 30 WBG-supported IEs indicate that strategies that combine structural interventions with individual- and family-level financial incentives show the greatest promise for improving education outcomes and leveling the playing field for girls (Parsons and McCleary-Sills Citation2014b). Impacts of programs on education vary depending on the outcome measured, as well as whom the intervention targets. Across the three education outcomes included in the analysis—enrollment, attendance, and retention/dropout—conditional cash transfer (CCT) programs targeted at girls only, rather than at their families, had the most consistently significant impacts.
The success of both conditional and unconditional cash transfer (UCT) programs suggests that measures to overcome the financial constraints facing families can be effective. Transfer payments in CCT programs went to the girl's parents (Alam, Baez, and Del Carpio Citation2011; Chaudhury and Parajuli Citation2010; Filmer and Schady Citation2008; Hassan Citation2010), to the girl directly (Heath and Mobarak Citation2012; Khandker, Pitt, and Fuwa Citation2003), or were split between the two (Baird et al. Citation2010, Citation2011). All of these programs had a positive impact on their intended education outcomes, regardless of the direct recipient of the cash transfer.
Among those interventions studied, programs that achieved greater improved education outcomes for girls than boys included programming components in addition to cash transfers. For example, in Pakistan, a public–private partnership provided newly established schools with additional space for girls and boys to attend at no cost, which worked to increase access to education, and resulted in higher enrollment (Barrera-Osorio et al. Citation2011). Providing uniforms and supplies in addition to school vouchers or cash transfers increased attendance and retention, as did provision of meals and medical care at school. Investing in the education system—training teachers, improving school infrastructure and materials, and increasing accountability of schools to the Ministry of Education—also improved rates of attendance and retention, especially among girls. This suggests that sustained impact on education—from higher rates of enrollment to increased attendance and lower dropout rates—requires multi-level programs. Interventions that invest in school infrastructure and personnel in addition to financial incentives appear to be a good route to boosting secondary school completion.
An IE in Bangladesh (Heath and Mobarak Citation2012) usefully highlights the importance of measuring the impact of education initiatives in the broader labor market and the demand for labor. The outcome measured was school attendance in areas where garment factories were present. Introduction of a garment factory in a community resulted in increased school enrollment for girls ages 5–16, while enrollment for girls ages 17 and 18 decreased. This may indicate that younger girls began enrolling in school to prepare themselves for jobs, while older girls left school to work in the factories. These findings highlight the need to consider labor supply and demand in the context of analyzing school enrollment and completion.
Girls’ Agency
Educational attainment and child marriage are inextricably linked. Promoting education for girls can prevent child marriage, while preventing child marriage can promote secondary school completion. Recognizing the importance of these links, programmatic efforts to reduce child marriage have expanded around the world, with programs increasingly seeking to address education, poverty, and lack of economic opportunities as drivers of child marriage.
Successfully addressing child marriage requires solutions that are tailored to local context and successfully engage community and religious leaders in efforts to chip away at the norms that perpetuate the practice. Lemmon and ElHarake (Citation2014) point out, even as education and economic opportunities become more widespread, countries where religious and traditional justifications for child marriage are deeply entrenched will not see an end to child marriage without shifts in social norms.
Research shows no single religion is particularly associated with high rates of child marriage (ICRW Citation2007). Our global analysis of child marriage prevalence across 111 countries shows that prevalence of child marriage varies greatly and that no pattern emerges by the countries’ predominant religion (Klugman et al. Citation2014). For example, the proportion of women married before age 18 is 75 percent in Niger and 18 percent in Indonesia, both Muslim-majority countries. In Central Africa Republic and Mexico, both Christian-majority countries, the prevalence is 68 percent and 23 percent, respectively. The prevalence of child marriage in India, a Hindu-majority country, is 47 percent, while neighboring Sri Lanka, with a majority of its population identifying as Buddhist, has a prevalence of only 12 percent. There are nonetheless clear regional dimensions. Of the 20 national “hot spots” for child marriage around the world, 18 are in sub-Saharan Africa and South Asia, but only two are in Latin America and the Caribbean (ICRW Citation2007).
At the same time, since families and parents often look to culture and religion to justify child marriage, religious traditional leaders can be particularly effective advocates for ending child marriage and leading a collective shift in social norms (Lemmon and ElHarake Citation2014). In Senegal for example, community based awareness campaigns, supported by the non-profit organization Tostan, enlisted the support of religious leaders to successfully reduce the accepted practice of female genital cutting (FGM) as well as child and forced marriage. Tostan's approach has led to a change in social norms, encouraging 5,423 villages in Senegal to move away from FGM and child and forced marriage (Klugman et al. Citation2014). Another example is the “Safe Age of Marriage” intervention in Yemen, which attributes its success in reducing child marriage to a participatory process of securing endorsement of religious leaders and community stakeholders to increase age of marriage (Pathfinder International Citation2010).
The challenges of child marriage have been long recognized by developing country governments, with various prevention initiatives emerging in the past two decades. It is instructive to learn from evaluations of child marriage prevention efforts, which have addressed the complexity of girls’ social lives and the gender norms that proscribe their roles and expectations for their achievement. In 1994, the Haryana State Government in northern India launched Apni Beti Apna Dhan (ABAD, or “Our Daughter, Our Wealth”), a program that gave poor families an incentive to keep their daughters in school and unmarried until age 18 and sought to improve family and community perceptions about the value of girls. The evidence suggests that the initiative was successful in important respects. An IE found that girls in the treatment group were significantly more likely to be in school and unmarried at age 18 than girls in the control group, although the impact on attitudinal changes about the value of girls are less clear (Nanda, Datta, and Das Citation2014).
Another example comes from rural Amhara region of Ethiopia, where 1 in 5 girls is married before her 15th birthday and nearly half (44 percent) are married by age 17. In 2004, the regional government partnered with the Population Council to implement a two-year project (Berhane Hewan [“Light for Eve”]) intended to delay marriage and keep girls in school. Through close collaboration and engagement with community and religious leaders, the project provided families with cash conditioned on their daughters’ remaining unmarried and in school for the duration of the program. Other elements included social mobilization of girls led by female mentors, provision of school supplies, livelihood training for out-of-school girls, and “community conversations” on early marriage and reproductive health. Two years later, girls (ages 10–14) enrolled in the program were 90 percent less likely to be married compared with girls in the control group. Participants were also three times more likely to be in school compared with girls in the control group (Erulkar and Muthengi Citation2009).
The findings from these evaluations resonate with the results from seven WBG-supported IEs for which age at marriage was an outcome (full details presented in Parsons and McCleary-Sills Citation2014b). A core set of intervention components emerge as consistently most effective: CCTs, teaching life and vocational skills, providing mentors, increasing access to free education, engaging in health awareness conversations with girls and their communities, and addressing the underlying economic drivers of child marriage (i.e. poverty). The programs evaluated were generally designed with the aim of boosting girls’ educational attainment, with increasing age at marriage as a secondary goal. Three of the evaluations examined CCT programs that were “cash for attendance” interventions (Baez et al. 2010; Baird, McIntosh, and Özler Citation2011; Baird et al. Citation2010), with one of those IEs comparing the conditional versus unconditional arms of an intervention (Baird, McIntosh, and Özler Citation2011). Such programs offered financial incentives, conditional on girls’ attendance in school. In the Zomba program in Malawi, girls received US$1-5 per month, in addition to school fees, if their monthly school attendance rate was above 80 percent. The UCT arm offered identical cash transfers but without a school attendance requirement (Baird, McIntosh, and Özler Citation2011). The program achieved substantial and statistically significant delays in marriage and childbearing among girls receiving the unconditional transfer, compared to those in the control group. A second program in Bangladesh (Khandker, Pitt, and Fuwa Citation2003) included a subsidy for school uniforms, as well as support for school curriculum reform, and infrastructure improvements. The results of the latter program showed that the stipend program increased girls’ secondary education substantially, while it had no discernible effect on the schooling of boys.
Another study evaluated a school-based reproductive health curriculum intervention in Kenya, focusing on the arm of the intervention that reduced the cost of education with the aim of keeping children in school longer (Duflo et al. Citation2006). The authors found that the program had little impact on students’ knowledge, attitudes, and behavior, or on the incidence of teen childbearing. However, reducing the cost of education by paying for school uniforms reduced dropout rates, teen marriage, and childbearing. Finally, an evaluation that tried to capture the effects of structural factors in labor market supply and demand and their impact on child marriage in Bangladesh (Heath and Mobarak Citation2012) found, as noted above, that the intervention increased enrollment for younger girls but not older girls.
It is clear from the foregoing synthesis that multiple strategies are being tested to improve girls’ education and delay marriage. Around half of the interventions among the WBG programs that were evaluated were multi-pronged, with cash transfers accompanied by additional educational and structural components. One clear result is that cash transfers appear to be very promising, presumably by helping to overcome the underlying poverty and economic drivers of school dropout and early marriage. In all the CCT programs, significant decreases occurred in early marriage, measured by the proportion of girls in the treatment group married by age 18 at the end of the intervention compared with girls in the control group. Further, it is important to note that unconditional transfers can be effective. In the Zomba program in Malawi (Baird, McIntosh, and Özler Citation2011), there were substantial and statistically significant delays in marriage and childbearing among girls in the UCT compared to those in the control group. Indeed, the observed impact among girls in the CCT arm was smaller and not significant. The largest impact among girls in the UCT arm was among those who had dropped out of school after the program began, which suggests that the decision to delay marriage is influenced not only by increased educational attainment, but also by the economic needs and stability of a girl and her family.
Our survey of the evidence also suggests that “extended” CCT programs that incorporated at least some of the following components into programming are effective: assigning mentors for girls, providing recreational activities and vocational skills, improving education facilities, and providing subsidies for uniforms. However it is difficult to judge the relative importance of each of these individual components, and which combination works best. Moreover, what is most effective very likely varies across local contexts. From the evident success of the programs in Ethiopia and India as well as results from a range of innovative programs, it is clear that the drivers of child marriage—poverty, limited opportunities for educational attainment and vocational training, and the low perceived value of girls in society (perceptions that are linked with various cultural and religious traditions)—all need to be addressed to reduce the prevalence of child marriage.
Conclusions
Leveling the playing field for girls in education has proved to be a global challenge, despite the progress of recent decades. This is a reflection of the persistent gendered barriers that prevent many girls around the world from attending and completing school, including child marriage. Families make decisions about their daughters’ marriages within the context of social norms, financial constraints, and economic opportunities. Changing these norms is the key to ending child marriage and achieving gender equality in education. This lasting change can be achieved through positive engagement of influential community members, including religious leaders. Programs and strategies that increase girls’ access to education—through financial incentives to attend school, expanding economic opportunities, and more gender-equitable schools, have been shown to result in delaying marriage beyond the girl's 18th birthday.
Our review of the evidence suggests that successful strategies have sought to address multiple social and economic drivers of child marriage, at the individual and family levels, although more longitudinal data analysis and continued IEs of programs are necessary to determine the long-term effects. Finally, though not addressed in detail here, preventing child marriage requires strengthened legal and policy frameworks, to ensure increased awareness and greater enforcement of existing laws. As long as child marriage continues unchecked, we will not reach global development goals of universal education and gender equality in education.
Related Research Data
Notes
1 Low-income countries that have not achieved gender parity in primary school (in order from lowest to highest): Angola (0.64), South Sudan (0.66), Afghanistan (0.72), Central African Republic, Chad, Yemen, Eritrea, Comoros, Guinea, Niger, Cote d'Ivoire, Pakistan, Cameroon, Democratic Republic of the Congo, Mali, Benin, Sudan, Djibouti, Cape Verde, Comoros, Dominican Republic, Lebanon, Mozambique, Swaziland, Papua New Guinea, Togo, Antigua and Barbuda, Nigeria, Liberia, Algeria, Ghana.
2 The quantitative data analyses reported in this manuscript were supported by funding from the Nordic Trust Fund and the Umbrella Facility for Gender Equality (UFGE), both of the World Bank Group. The UFGE also provided funding for the enGENDER IMPACT analyses included here. Jeni Klugman would like to thank the Women and Public Policy Program at the Kennedy School, Harvard University, for support. Analysis reported here of the Gallup World Poll was conducted by Emma Samman (Overseas Development Institute) for the Voice & Agency book cited throughout this manuscript (Klugman et al. Citation2014).
3 We use the term sexual autonomy to refer to the results for both the ability to ask for the use of a condom and to refuse sex.