Commonly used methods for modelling early life adversity (e.g., sum-scores, latent class or trajectory ap-proaches, single-adversity approaches, and factor-analytical approaches) have not been able to capture the complex nature of early life adversity. We propose network analysis as an alternative way of modelling early life adversity (ELA). Our aim was to construct a network of fourteen adverse events (AEs) that occurred before the age of 16 in the TRacking Adolescents Individual Lives Survey (TRAILS, N =1029). To show how network analysis can provide insight into why AEs are associated, we compared findings from the resulting network model to findings from tetrachoric correlation analyses. The resulting network of ELA comprised direct re-lationships between AEs and more complex, indirect relationships. A total of fifteen edges emerged in the network of AEs (out of 91 possible edges). The correlation coefficients suggested that many AEs were associated. The network model of ELA indicated, however, that several associations were attributable to interactions with other AEs. For example, the zero-order correlation between parental addiction and familial conflicts (0.24) could be explained by interactions with parental divorce. Our application of network analysis shows that using network analysis for modelling the ELA construct allows capturing the constructs' complex nature. Future studies should focus on gaining more insight into the most optimal model estimation and selection procedures, as well as sample size requirements. Network analysis provides researchers with a valuable tool that allows them as well as policy- makers and professionals to gain insight into potential mechanisms through which adversities are associated with each other, and conjunctively, with life course outcomes of interest