ORI GIN AL PA PER Déja ̀ vu All Over Again: Investigating Temporal Continuity of Adolescent Victimization

Abstract An important theoretical issue in the study of criminal victimization focuses on state dependence or the extent to which prior victimization causally affects subsequent victimization. Framing this issue are two conceptual arguments that posit divergent pre-dictions regarding the impact of past victimization. While ‘‘victim labeling’ ’ suggests that the experience of prior victimization increases victimization in the future, ‘‘victim ratio-nality’ ’ suggests that past victimization decreases subsequent victimization. To date there has been little longitudinal research that explicitly evaluates these competing viewpoints. In response, the current study uses panel data from a sample of adolescents to assess the merit of these opposing theoretical frameworks. Against this theoretical backdrop, it is shown that prevailing strategies for the analysis of panel data—random-effects and fixed-effects—produce erroneous results that add confusion to the debate regarding the nature of state dependence processes in criminal victimization. Drawing from advances in econo-metrics, two estimators that can overcome the problems that plague more traditional panel data models are used to provide a test of the competing victimization perspectives. Results from those analyses suggest support for the thesis that past victimization increases sub-sequent victimization, but the magnitude of this positive relationship is more modest than is indicated by analytic models that have been utilized in prior longitudinal victimization research