From Childhood Delinquency to Adulthood Aggression: Using Ecological Momentary Assessment to Examine How Developmental Trajectories Forecast Emerging-Adult Aggression in Daily Life

A substantial body of research has examined the link between childhood delinquency and later antisocial behaviour in adulthood. One of the most influential frameworks explaining these developmental trajectories is the dual taxonomy model (Moffitt, 2017). This model distinguishes between adolescent-limited antisocial behaviour, which is confined to adolescence, and life-course persistent antisocial behaviour, which typically begins with delinquent behaviour earlier in childhood and continues into adulthood (Moffitt, 2017). The dissertation aims to interpret the data patterns within the framework of existing theories such as the dual taxonomy model; however, some patterns may extend beyond what current theories can fully explain. Furthermore, to date, Ecological Momentary Assessment (EMA) has not been employed to investigate how early delinquency translates into day-to-day aggression in adulthood, which could reveal rich insights into how aggression manifests in daily contexts following different developmental trajectories of delinquency. Understanding these dynamics may illuminate how antisocial tendencies persist into emerging adulthood and help predict which individuals are at greater risk of heightened aggression later in life based on their childhood and adolescent delinquency patterns. Thus, this dissertation seeks to address this gap by examining how trajectories of delinquent behaviour from childhood and adolescence predict momentary aggression in emerging adulthood (ages 21-26), using data from the z-proso Zurich Cohort Study. The dependent variable of aggression, captured via EMA, will be measured across multiple domains, including provocation, emotionally driven aggression, and substance-related aggression, with childhood delinquency trajectories as the independent variable. The analysis will employ growth mixture modelling to identify trajectories of delinquency over time and examine their associations with EMA outcomes, controlling for key covariates such as gender, socioeconomic status (for z-proso data this will be parental occupational prestige). Gender is particularly important, as research indicates that life-course persistent offending is less common among females than males (Moffitt, 2001, 2004; Ribeaud & Eisner, 2010; Archer, 2004). Country of origin will also be reported in the descriptive statistics. Additionally, if scope and time permits, the dissertation will also examine trajectories of aggressive behaviour from childhood and adolescence and how they predict momentary aggression in emerging adulthood (ages 21-26). This will offer further insights as aggression was measured earlier (from age 7 onwards) in the z-proso data than delinquency (from age 11 onwards). Using the z-proso dataset offers several advantages; its large, prospective longitudinal design establishes a clear time order, supporting stronger plausible causal pathways compared to retrospective studies. This is particularly important as there are still few longitudinal studies tracking antisocial behaviour over the lifespan, with most evidence based on retrospective accounts of offending (Moffitt, 2017; Huesmann, Dubow, & Boxer, 2009). The sample is also ethnically diverse, including major minority groups, which allows for more generalisable findings and addresses another limitation from previous studies which relied largely on WEIRD populations (Ribeaud & Eisner, 2010; Moffitt, 2017).