The work performed as part of the project and the main achievements can be divided into four parts.
First, the project examined disparities in the age at first union and first birth, and the role of education in shaping them, among women within LMICs. Taking a novel approach and using quantile regression analysis, this study explored whether additional schooling shifts relatively early union formation and childbearing (i.e. lower parts of the distributions of the age at first union and birth) similarly or differently than it shifts other parts of these distributions. It was shown that the association between an additional year at school and the age at first union and birth was weaker in the lower than the upper parts of the distributions. Thus, education has a relatively weaker effect on the reduction of early first unions and births and plays an unequalizing role in shaping family formation within countries. These findings are key to understanding persistently high levels of early marriages and pregnancies, despite educational expansion.
The second part of the project explored gender differences in family formation, combining data from countries at different stages of development and gender inequality levels. First, patterns of differences in childlessness between women and men were examined according to the countries’ levels of gender inequality. The key findings suggested that in settings with lower levels of gender inequality, thus higher levels of women’s empowerment, the prevalence of childlessness is higher among both women and men, but among men in particular. The results highlighted the interplay between individuals’ union formation trajectories and their level of education in shaping these associations. Second, the patterns of gender disparities in union formation were examined across countries with different levels of development. The results uncovered that in countries with high levels of development, a significant proportion of unions involve cross-national couples that differ starkly in their characteristics between women and men, and exhibit asymmetries according to partners' country of birth, age, and educational level. Overall, the results shed new light on the links between family formation and contextual factors and highlighted that to gain a fuller understanding of these processes, it is critical to consider differences in the behaviors of women and men.
Third, the collaborations developed over the course of the fellowship allowed the study of the causes and consequences of two important dimensions of family formation: unintended pregnancies and fertility levels. Given that many pregnancies among women in LMICs happen not only early but are also unintended, the association between birth intentions and adverse outcomes, namely infant mortality, was examined. The use of fixed-effects models showed that compared to wanted births, infant mortality was higher after an unwanted or mistimed birth. The risk of infant mortality after an unwanted/mistimed pregnancy was particularly high in nations with low human development. To further cast light on the importance of taking contextual factors into account when examining family formation, the final collaborative study focused on researching the impacts on fertility levels of disasters. Evidence synthesis methods were employed to show that, overall, fertility levels were negatively affected by disasters. With growing risks of adverse events (e.g. pandemics), this study delivered insights into an important factor affecting family formation globally.
Fourth, the project permitted skills development. The fellow attended courses, organized seminars at the host institution and sessions at international conferences, supervised Master and PhD students, expanded their research network, and developed new collaborations that resulted in high-quality publications. Overall, the objectives of the action were satisfactorily achieved.
