Fertility Decision-Making in Brazil: Impact of High School Full Time on the Prevention of Early Pregnancy among Young Women in Amazon

This study aimed to analyse, using open-source data from Brazil, the impact of high school full time on the odds of early childbearing in Amazonian municipalities. The final dataset comprised 359 observations spanning 62 municipalities in Amazon, covering the period from 2010 to 2015. The methodology employed was a Generalised Linear Mixed Model (GLMM). Overall, municipalities with enrolments in high school full time are negatively correlated with the log-odds of the early pregnancy rate among girls between 15 to 19 years old. The odds ratio calculated is lower than one (0.95), meaning that when the municipality has enrolments in high school full time, the odds of early childbearing are likely to decrease in 4,41%. This relationship is particularly significant in low-to-medium income municipalities: the calculated odds ratio is lower than one in both models (between 0.92 and 0.93), and there is a 7.88% and 7.01% decrease in the odds of early pregnancy for respectively low-income and medium-income municipalities when the dummy for enrolments in high school full time is equal to one. However, these findings are merely suggestive and somewhat inconclusive. Challenges in modelling and methodology, alongside significant difficulties in data collection, undermine the robustness of this study, necessitating cautious interpretation of its results. Future research could benefit from employing alternative methodologies, such as causal inference models or qualitative methods, to achieve more robust findings. Additionally, the collection of primary data on social norms and religion is recommended, as these factors could significantly enhance the understanding of the determinants of early fertility within the Brazilian context.