INTRODUCTION

National and international communities are increasingly recognizing child marriage

as a serious problem, both as a violation of girls’ human rights and as a hindrance
to key development outcomes. As more program, policy, donor and advocacy constituencies
pledge commitment, resources and action to address this problem, it becomes important
to examine past efforts and how well they have worked. Finding model solutions to address
child marriage has been a challenge because, while there has been increasing investment
in programs during the last decade, many are not well-documented, and even fewer
are well-evaluated.

In this brief, we summarize a systematic review of child marriage prevention programs

that have documented evaluations. Based on this synthesis of evaluated programs,
we offer an analysis of the broader implications for viable solutions to child marriage.
Our findings show that child marriage prevention programs have indeed expanded in
number and scope during the last decade; almost two dozen have documented some
type of an evaluation. The largest number of evaluated programs is in South Asia,
especially in Bangladesh and India. Programs in a broader range of African and Middle
Eastern countries, including Ethiopia and Egypt, are also adding to the evidence base.
On balance, the results from this composite of evaluations lean toward positive findings,
indicating that a set of strategies focusing on girls’ empowerment, community mobilization,
enhanced schooling, economic incentives and policy changes have improved knowledge,
attitudes, and behavior related to child marriage prevention. The strongest, most consistent
results are shown in a subset of programs fostering information, skills, and networks for
girls in combination with community mobilization. While many child marriage prevention
programs are only beginning to explore possibilities of going to scale, there are encouraging
signs that large-scale structural efforts aimed at other goals, such as education, health,
and poverty reduction, are beginning to make a connection with child marriage prevention.
A smaller, but growing set of such programs is providing tentative but promising evaluation
results, laying the foundation for building new partnerships and leveraging scarce resources.
THE PROBLEM
Based on the United Nations (UN) Convention on the Rights of the Child, child marriage
refers to marriage under age 18 (UN 2000). Marriage before the age of 18 is a fundamental
human rights violation. Child marriage disproportionately affects young girls, who are much
more likely to be married as children than young boys (Mathur et al 2003; UNICEF 2005;
Save the Children 2004). The latest international estimates indicate that worldwide,
more than 60 million women aged 20–24 were married before they reached the age of 18
(UNICEF 2007).
INTERNATIONAL CENTER FOR RESEARCH ON WOMEN
3

The extent of child marriage varies substantially between countries; the highest rates are
found in West Africa, followed by South Asia, North Africa/Middle East and Latin America
(Clifton & Frost 2011). However, as Figure 1 shows, countries with the highest percentage
of females 20 to 24 who report having been married before 18 are not the same countries
in which the largest number of girls are vulnerable to this practice. When considering the
number of girls aged 10-19 across various countries, more girls are at risk of child marriage
in India than in most other countries combined. In fact, given population size and significant
rates of child marriage in Afghanistan, Bangladesh, India, and Nepal, about one-half of the
girls who are affected by child marriage live in South Asia (UNICEF 2007).

Early marriage leads to

early childbearing, with significantly higher maternal mortality and morbidity rates as
well as higher infant mortality rates (Mensch 2005; UNICEF 2005; Save the Children
2004; Bott & Jejeebhoy 2003). Moreover, child marriage has negative effects on girls’
education. Girls with low levels of schooling are more likely to be married early, and
child marriage virtually puts an end to a girl’s education (Mathur et al 2003; Mensch 2005;
Jejeebhoy 1995). A child bride’s lack of education and peers limits her support systems,
and without skills, mobility, and connections, she is constrained in her ability to overcome
poverty for herself, her children, or her family. Young girls married to older men with more
sexual experience are also at greater risk of HIV infection (Clark 2004), and child brides
are at heightened risk of violence in the home (ICRW 2005; Santhya et al 2010).
The lack of education, health, physical safety, and autonomy deprives girls of their
basic human rights, and it also acts as a brake to social and economic development.
National and international indicators on maternal health, education, food security,
poverty eradication, HIV/AIDS, and gender equality are all negatively linked with high
child marriage rates. In fact, child marriage undermines the achievement of each of
the eight Millennium Development Goals and global targets to reduce poverty worldwide
(Hervish & Feldman-Jacobs 2011; UNICEF 2006; IPPF 2007).
SOLUTIONS: WHAT DO WE KNOW?
Efforts to address child marriage in many parts of the world date back to the 1920s.
For example, the first legislative attempt to end child marriage in India was through
the passage of the “Sarda Act” in 1929 (Mukherjee 2006). More recently, legal reform
began to gain ground in the 1970s and 1980s, as countries such as Bangladesh, India
and Indonesia established or raised the legal minimum age of marriage to 18 for girls.
During the same period, human rights activists and the United Nations launched efforts
to address harmful traditional practices affecting women. However, programmatic
interventions to eradicate this practice have only gained momentum since the 1990s,
coinciding with the attention to adolescent reproductive health at the Cairo International
Conference on Population and Development in 1994 and the assertion of women’s and
girls’ human rights at the UN International Conference on Women in Beijing in 1995
(UNICEF 2001).
INTERNATIONAL CENTER FOR RESEARCH ON WOMEN
5
Several organizations have examined the causes and consequences of child marriage
in some detail (Bott & Jejeebhoy 2003; Mathur et al 2003; UNICEF 2001), and many have
highlighted promising programmatic approaches to prevent child marriage (e.g., USAID
2009; Hervish & Feldman-Jacobs 2011). However, comprehensive reviews to take stock
of existing programs have been more limited. The International Center for Research on
Women (ICRW) undertook one such effort in 2007, and identified 66 child marriage-related
programs in 30 countries (Jain & Kurz 2007). ICRW conducted a follow-up review that
focused solely on India and identified 58 program and policy efforts targeting child
marriage in the country (Mukherjee et al 2008).
These reviews have uncovered important insights on the scope and range of interventions
to address this problem. Many programs recognize the multitude of factors driving the
persistence of child marriage. These interventions have tried comprehensive or
integrated approaches that engage communities, families and policymakers, while
attempting to impart to girls skills, opportunities and empowerment. However, prevention
efforts were not always focused in the countries with the highest rates of child marriage
(Jain & Kurz 2007). And many efforts lacked scale and were not integrated into larger
government initiatives or private sector drivers of economic and social change to be
sustainable in the long run (Mukherjee et al 2008).
Most importantly, existing reviews of initiatives to prevent child marriage indicate
that few of these have been evaluated (Jain & Kurz 2007; Hervish & Feldman-Jacobs
2011; Mukherjee et al 2008). For example, only 10 percent of the programs identified
in the 2007 ICRW scan had been evaluated. Still fewer were evaluated using rigorous
methodologies or included information about the evaluation process. Thus, while we
know something about what has been attempted to prevent child marriage, we know
little about how successful these efforts have been.
To address this gap, this brief focuses on and systematically analyzes the small number
of evaluated programs with the goal of preventing child marriage. This analysis provides
guidance not only on what has worked, but what can be done to strengthen current and
future efforts to prevent child marriage.
Solutions to End Child Marriage WHAT THE EVIDENCE SHOWS
6
WHAT WORKS: A SYSTEMATIC REVIEW
In 2009, the World Health Organization (WHO) commissioned ICRW to undertake
a systematic review that would use the established WHO methodology to identify,
review and appraise research evidence relevant to the prevention of child marriage.
In collaboration with WHO, ICRW conducted an extensive search of international,
regional and WHO databases to identify program interventions and policy strategies
that had documented measurement of change in child marriage-related behaviors
and/or attitudes. We identified additional programs by conducting a general online
search, examining websites of organizations known for their involvement with child
marriage prevention and emailing relevant staff, and reviewing a wide range of program
scans and other documents in the published and grey literature. In total, we identified
more than 150 potentially relevant efforts to prevent child marriage. However, only 23
of these documented an attempt to measure change in child marriage-related behaviors,
knowledge, or attitudes among relevant stakeholders. These programs were implemented
between 1973 and 2009, with several of the programs continuing through the present,
and evaluations were published between 1991 and 2010.
Our search suggests a significant increase in the number of interventions targeting child
marriage during the last decade. The majority of programs work directly with girls, offering
them opportunities to obtain skills and education. Most efforts engage with families and
community members and attempt to change underlying social norms that perpetuate the
practice of child marriage. Many interventions also offer economic incentives to parents
to promote education and healthy behaviors as well as prevent child marriage. As increased
resources are dedicated to preventing child marriage, interest in learning the impact of these
efforts has also increased. Although 23 is not a large number, this is the first critical mass
of evaluated programs. Of course, the "positive bias" in the documentation and publication
of evaluations may shape the conclusions we draw about them; unsuccessful programs
are less likely to be documented and published.
It is important to note that a few promising approaches—especially at scale—are underway,
but do not as yet have completed evaluations. An important example is the Haryana state
government’s Apni Beti Apna Dhan program in India, which provides to girls and their
families cash incentives that are conditional on the daughters remaining unmarried until
age 18 (Sinha & Young 2009). The first cohort of girls enrolled in the program will turn
18 in 2012, and ICRW is initiating an evaluation of the program. Similarly, the Population
Council is investigating the effectiveness and cost of different strategies to delay marriage
in four sub-Saharan African countries, including Ethiopia, Kenya, Tanzania and Uganda,
building on its successful Berhane Hewan program in Ethiopia (USAID 2011). Pathfinder’s
PRACHAR program in Bihar, India, has also built on positive results to scale up and
evaluate activities in subsequent phases (Rahman & Daniel 2010).
INTERNATIONAL CENTER FOR RESEARCH ON WOMEN
7
Other large-scale programs may be poised to undertake evaluations on child marriage
in the near future. For example, programs targeting girls’ safety and reduction of violence
in schools, such as the Safe Schools program sponsored by the United States Agency for
International Development (USAID) in Ghana and Malawi, have child marriage prevention
as one objective, but data to evaluate this outcome is currently not collected (USAID 2008).
Similarly, governments and multi-lateral agencies undertaking large-scale initiatives to
eradicate poverty, expand economic opportunity, promote women’s empowerment and
promote girls’ education have the potential to prevent child marriage because they are,
in effect, addressing some of its drivers. However, these broader initiatives have not always
been developed with child marriage in mind, and therefore do not collect relevant data.

EVALUATED PROGRAMS: AN OVERVIEW

We begin with an overview of the 23 evaluated programs. Figure 2 presents when these
programs were initiated and when evaluations were published. The figure illustrates that
much of the activity on child prevention is recent: more than half (13) of the programs
were initiated in the last decade. Seven programs were initiated in the 1990s, coinciding
with the two major UN conferences on reproductive health and women’s rights. Only
three programs had documented evidence earlier than the 1990s. The figure also helps
explain why so little evidence has been available to date on the success of programs
addressing child marriage: 15 out of 23 evaluations were published since 2006.
Figure 3 illustrates the targeted outreach for these programs. Not surprisingly given the
nascent stage of child marriage interventions, the vast majority of programs have had
limited outreach and have yet to go to scale. In fact, nine of 23 programs aimed to reach
fewer than 5,000 beneficiaries and almost three-quarters or 14 programs reached fewer
than 15,000 beneficiaries. Only six programs reached more than 60,000 beneficiaries.
Figure 4 shows that only five out of 23 programs had preventing child marriage as a
primary objective. A significant proportion of the programs (7 out of 23) included child
marriage-related outcomes only as an incidental objective. These programs were often
educational, infrastructure or cash incentive interventions implemented and/or supported by
national or local governments; bilateral or multi-lateral donors such as the World Bank, USAID,
or the Swedish International Development Authority (SIDA); or large non-governmental
organizations (NGOs) such as Building Resources Across Communities (BRAC)1
in Bangladesh. As such, these interventions also had the greatest scope for
expansive outreach.
1 BRAC was formerly known as the Bangladesh Rural Advancement Committee

Educating and mobilizing parents and community members

Parental and community engagement is the second most frequently used strategy,
employed by 13 of 23 programs. As mentioned, this strategy usually accompanies other
approaches described here, especially strategy 1. The primary motivation behind this
strategy to “create an enabling environment” is the understanding that the decision to
marry girls early is generally in the hands of family and community elders, and that the
resulting stigma and sanctions for failing to meet social expectations are administered
by the broader community. Girls rarely have the power to decide on their own if, when or
whom to marry. Thus, parental education and community mobilization attempt to change
social norms and forge a more supportive, less punitive environment for girls and families
who are willing and ready to change the custom of early marriage. Enlisting parents
and community members helps to mitigate possible unintended consequences of girls’
participation in programs, and also reinforces a program’s messages and activities.
Programs employing this strategy include the following range of interventions:
• One-on-one meetings with parents, community and religious leaders to gain support.
PROGRAM EXAMPLES: Ethiopia Early Marriage Evaluation Study (EMES) (Ethiopia),
Kishori Abhijan (Bangladesh)
• Group and community education sessions on the consequences of and alternatives
to child marriage.
PROGRAM EXAMPLES: Ishraq (Egypt), PRACHAR (India)
• Parental and adult committees and forums as guides to life skills and sexual and
reproductive health curricula.
PROGRAM EXAMPLES: Adolescent Participatory Project (Nepal), Maharashtra Life
Skills Program (India)
• Information, education, communication (IEC) campaigns—using various
platforms—to convey messages about child marriage, schooling, rights,
reproductive health and other topics.
PROGRAM EXAMPLES: Integrated Action on Poverty and Early Marriage (IAPE)
(Yemen), Gender Quality Action Learning Program (GQAL) (Bangladesh)
• Public announcements and pledges by influential leaders, family heads,
and community members.
PROGRAM EXAMPLES: Tostan (Senegal)
This strategy is generally implemented as an accompaniment to others, and it is difficult
to assess the extent to which community education and mobilization efforts contribute
to program failure or success, because most evaluations are not designed to isolate the
impact of this component. At the same time, most program implementers argue that it
may well be impossible to implement programs aimed at such significant social change
without actively engaging community members.
2..

The power of community mobilization can perhaps be best assessed by examining the
few programs that have taken on community mobilization as their core activity. On the
whole, such evaluations demonstrate success, although their designs do not conform
to the highest level of methodological rigor, which in itself presents a significant challenge
for broad-based community interventions. In this category is the Integrated Action on
Poverty and Early Marriage program in Yemen, which sponsored a campaign to raise
awareness of parents, grandparents and youth about the consequences of early marriage.
The advocacy efforts were instrumental in the introduction of a minimum legal age
of marriage in Yemen. The Tostan approach implemented in Senegal does not focus
on legal change, and is instead built around informal community education and
awareness-raising that facilitates community mobilization, sometimes in the form of
public declarations against harmful practices, such as female genital cutting and early
marriage. Tostan boasts a large number of converted communities through documented
pledges but finds less conclusive results on actual declines in child marriage.
Enhancing the accessibility and quality of formal schooling for girls
Research shows that girls’ education is strongly associated with delayed marriage.
Girls with secondary schooling are up to six times less likely to marry as children when
compared to girls who have little or no education (UNICEF 2007). The causality of this
relationship is debated, although more recent research suggests that being out of school
puts girls at risk of marriage rather than marriage being a reason for girls being pulled
out of school (Lloyd 2006). Schooling is protective against marriage for at least two
reasons. Normatively, simply being in school helps a girl to be seen as a child, and
thus not marriageable. Other than home, schools can be seen as a “safe space” for girls.
This seems to be largely the case, even with continued parental concerns about violence
or sexual harassment in many settings. Thus, as it becomes a socially acceptable
alternative, school attendance helps to shift norms about early marriage. Additionally,
the experience and content of schooling help girls to develop social networks and acquire
skills and information, all of which contribute to their ability to better communicate and
negotiate their interests. For parents and society, better quality and secondary education
may make the returns to human capital investment in girls more obvious and justifiable.
For all these reasons, formal schooling should have emerged as one of the strongest
program interventions to prevent child marriage, especially given the existing momentum
toward universal schooling and increased emphasis on secondary education. Past reviews,
however, have found this strategy to be under-utilized as a route to child marriage
prevention (Jain & Kurz 2007; Mukherjee 2008). In this review, we find encouraging
signs of an increasing number of programs employing formal schooling as a means