British Criminology Conference

Activity: Participating in or organising an event typesParticipation in conference


'Unravelling Youth Crime Patterns using Trajectory Modelling' Abstract: The existence of the age-crime curve is something of a ubiquitous phenomenon within life-course criminology. However, exploring data at the aggregate level conceals important differences in the underlying individual offending patterns. Longitudinal offending data allows us to study individual trajectories of offending and to better understand the various complex pathways into and out of offending. Recent advances in the development of statistical approaches to studying offending trajectories have made it possible to examine naturally occurring patterns of behaviour amongst large groups of people and such methods are perfectly adapted to longitudinal data. The most widely used of these are semi-parametric group based modelling (SGM) or latent growth mixture modelling (LGMM), both of which classify offenders into distinctive trajectory groups on the basis of observed offending histories, thus modelling sample heterogeneity in offending trajectories over time. Using six sweeps of data from the Edinburgh Study of Youth Transitions and Crime, this paper will use SGM to explore the complexities of individual offending trajectories amongst a cohort of over 4000 young people in contemporary Scotland and will expose the frailties of relying on aggregate data to understand youth offending behaviour.
PeriodJul 2006
Event typeConference
LocationGlasgow, United KingdomShow on map


  • age-crime curve
  • trajectory modelling
  • life-course criminology
  • youth offending
  • longitudinal studies