Theses Doctoral

“Child marriage” declines as social change? The influence of global priorities, social determinants and norms in changing adolescent marriages in southcentral Uganda, 1999-2018

Spindler, Esther J.

Over the last 20 years, adolescent health researchers, practitioners and advocates have zeroed-in on the global problem of ‘child marriage.’ Defined as a formal or informal marital union before 18 years, child marriage affects both boys and girls, but disproportionally affects girls. Globally, child marriage is noticeably prevalent but on a downward trend, with the proportion of 20-24 year old women marrying before 18 years decreasing from 25% to 19%, from 2008 to 2020 (UNICEF, 2018; 2022). Extensive research has shown the adverse consequences of marrying during adolescence, ranging from increased risk of maternal mortality and birth complications, intimate partner violence (IPV), adverse mental health and intergenerational poverty outcomes (Burgess et al., 2022; Clark, 2004; Nour, 2009; Otoo-Oyortey & Pobi, 2003; UNICEF, 2018). From a rights perspective, child marriage is considered a violation of girls’ and boys’ ‘right’ to fully consent into marriage before reaching age of majority, internationally recognized as 18 years of age (Bruce, 2003; Nour, 2009). As such, child marriage is recognized as a human rights violation under several international treaties, including the Convention on the Rights of the Child (CRC).

The term ‘child marriage’ is commonly used to convey the human rights violations that early marital practices have for under-age girls and boys. While the term ‘child marriage’ has mobilized consensus and solidarity toward the issue, this terminology also homogenizes the issue of marriage as a problem affecting the ‘girl child’ with little to no agency in the marriage decision-making process. More specific to Uganda, this ‘child marriage’ terminology can be problematic where marriage more commonly occurs during middle to late adolescence (15-19 years) and when adolescents may exert varying degrees of agency and consent in the marital decision-making process. Except for Chapter 1 which explores ‘child marriage’ global and national movements, I intentionally use the terminology ‘adolescent marriage’ (as marriage before age 18), rather than ‘child marriage,’ throughout this dissertation.

Despite the global push to ‘end child marriage’ over the last decade, there is limited research about how broader social and structural factors may be driving declines in adolescent marriage (Muthengi et al., 2021; Plesons et al., 2021). In particular, we have a limited understanding about how global efforts, social processes and norms might work together to drive marriage declines among adolescents. Through a mix of policy, quantitative and qualitative methods, this dissertation examines the policy, structural and social mechanisms that have contributed to declining adolescent marriage among adolescent girls in the context of southcentral Uganda.

Chapter 1 begins with a broader contextual lens, examining the political evolution of the global ‘child marriage’ movement, and how the ‘problem’ of child marriage was then taken-up by government and civil society actors in Uganda. This chapter is informed by 20 key informant interviews with Ugandan and global stakeholders working on child marriage and a desk review of over 130 documents gathered across four years. This chapter highlights how the global ‘child marriage’ movement marked a political shift in adolescent girl funding, repackaging the issue of early marriage as an issue of ‘child protection.' The focus on child protection, rather than adolescent sexuality, was instrumental in mobilizing attention from liberal and conservative funders in the Global North and policy-makers in the Global South. In the priority country of Uganda, multiple factors influenced the national policy uptake of child marriage, including: 1. Regional campaigns that created consensus among Eastern and Southern African country leadership to address child marriage; 2. The availability of national data that showed the reach and severity of child marriage within Uganda; 3. The cultural and political appeal of child marriage as an issue of ‘child rights’, rather than one of ‘sexuality,’ and; 4. A network of government leaders, academics, international non-governmental organizations (INGOs) and civil society organizations (CSOs) who coalesced behind the issue in Uganda.

Chapter 2 focuses-in on the southcentral region of Uganda, leveraging close to 20 years of quantitative data to understand how social and structural factors are affecting adolescent marriage declines in the region. Using data from 13 surveys (1999-2018) of the Rakai Community Cohort Study (RCCS), I couple decomposition and causal inference methods to assess how social determinants and adolescent pregnancies have contributed to adolescent marriage declines among 15 to 17 year old girls. I find that both marriages and pregnancies among adolescent girls substantially declined over the last 20 years, from 24% to 6%, and 28% to 8%, respectively, between 1999 and 2018, as a result of educational and economic improvements. Among all social determinants, girls’ secondary schooling was more closely associated with lower risk of marriage and pregnancy (aOR marriage = 0.09; 95%CI=0.07, 0.12; aOR pregnancy = 0.14; 95% CI=0.11; 0.19). In the causal mediation analyses, lower pregnancy rates partially explained the positive effect of higher secondary schooling on lower risk of adolescent marriage. Decomposition analyses showed that the declines in adolescent marriage between 1999 to 2018 were primarily attributed to pregnancy declines, and to a lesser extent, improvements in education and SES. These findings reemphasize the sizeable role of education in preventing adolescent marriages, in line with Uganda’s national educational investments such as universal primary education (UPE). Yet, these findings also underline the importance of adolescent pregnancy prevention to delay age at marriage.

In the same region of southcentral Uganda, Chapter 3 uses secondary ethnographic data to more deeply explore the social mechanisms and norms that have contributed to changes in adolescent marriages. I qualitatively explore how the region’s social and economic changes have affected social norms about adolescent sex, courtship, and marriage in Rakai, Uganda. This analysis is informed by 16 focus group discussions and 15 key informant interviews conducted in 2018 with younger and older women and men, ranging from 16 to 77 years old. In comparing generational perspectives, I identify a ‘normative transition’, in which new structures are transforming courtship and marriage processes for young people. First, the HIV epidemic significantly weakened family structures, and in the process, courtship and marriage guidance previously provided by families and elders; second, the loss of land ownership in between generations has made marriage preparations more difficult for young people; and third, new social spaces outside the family home – including discos, mobile phones and schools - have expanded young people’s romantic geographies prior to marriage. These changes have reduced the importance of the family institution in the marital decision-making process, while increasing young women’s and men’s autonomy in engaging in premarital sex, choosing their partners, and delaying marriage. Although these changes have delayed age at marriage beyond adolescence, this transition has introduced unanticipated challenges for young people as they enter adulthood, including lack of overall parental, familial and elder guidance in their relationship and marriage formation processes.

Taken together, these findings highlight the complexity of adolescent marriage changes and prevention efforts at the global, Ugandan, and southcentral region of Uganda. First, global and national ‘child marriage’ movements played a significant role in the uptake of child marriage as an issue of ‘child protection’, rather than one about ‘sexuality’ in Uganda. Yet looking at the context of southcentral Uganda, adolescent pregnancies and adolescent marriages declines appear to be closely linked, highlighting the importance of conceptualizing adolescent marriage as not just a child protection issue, but one of adolescent sex and sexuality. Lastly, I find that broader structural and social changes in Rakai have substantially changed adolescent norms around sex, courtship, and marriage, delaying age at marriage in between generations. However, young people are encountering new challenges as they enter adulthood and romantic relationships in the absence of pre-existing elder and familial systems and networks. Additional research should focus on understanding the unintended consequences of catalyzing norm change and delaying age at marriage, including how these changes might affect familial and community relationships and kinships.

Twenty years into the global push to end ‘child marriage’, this dissertation research provides new insights into the complex structural, social and sexuality drivers of adolescent marriage changes in Uganda. Despite the substantial progress in adolescent marriage declines, this research points to key gaps that will need to be addressed to improve adolescent SRH rights and needs in Uganda, the East African region, and beyond. Of particular importance is the need to center adolescent sexuality within current child marriage efforts, as well as focusing on the broader social changes affecting adolescent relationship formation, rather than exclusively focusing on age at marriage as a marker of social change.

Geographic Areas


  • thumnail for Spindler_cumc.columbia_0054E_10084.pdf Spindler_cumc.columbia_0054E_10084.pdf application/pdf 2.45 MB Download File

More About This Work

Academic Units
Population and Family Health
Thesis Advisors
Santelli, John S.
Dr.P.H., Mailman School of Public Health, Columbia University
Published Here
September 14, 2022