Introduction: the unusualness and contribution of life span longitudinal studies of aggressive and criminal behavior

Prospective longitudinal studies of violent and criminal behavior have been reviewed by Farrington [2007, Appendix] and Farrington and Welsh [2007, Appendix to Chapter 2]. These authors focused on longitudinal surveys of hundreds of people in community samples, spanning at least 5 years, including repeated personal contacts with the participants, and publishing findings on violent or criminal behavior. They found that only 35 major surveys of this type had ever been conducted. Of these, only 10 studies have thus far followed up participants into their 40s. In organizing this special section, the collaborative aim of the four lead authors of the included articles (Lars Bergman, David Farrington, Rowell Huesmann, and Lea Pulkkinen) was to bring together results from the most noteworthy of these surveys. We excluded the Danish birth cohort study of Brennan et al. [2000] because it was largely based on police records, as opposed to personal contacts with participants. We excluded the Young Lawbreakers as Adults Study of Lang et al. [2002] because of its case–control design, comparing delinquent and control males. We wanted to focus on surveys of community samples. We excluded the Cambridge–Somerville Youth Study of McCord [1991] because unfortunately Joan McCord had passed away. We excluded the UK National Survey of Health and Development of Wadsworth [1991] because that only collected criminal record data up to age 21. Regarding the Kauai longitudinal study [Werner and Smith, 2001], we invited Emmy Werner to submit an article, but unfortunately she was unable to do so because of illness. Regarding the Montreal longitudinal study of adolescents [Le Blanc and Frechette, 1989], we invited Marc Le Blanc to submit an article, but unfortunately he declined because of his retirement. This special section therefore presents results obtained in the four life span longitudinal studies directed by the four lead authors. Long-term surveys are needed to investigate the development of aggression and criminal behavior, risk and protective factors at different ages, and the effects of life events on the course of development [Farrington, 2003]. Because of recent interest in categories of offenders [Moffitt, 2006], we agreed that we would all identify adolescence-limited and persistent offending or aggression in our surveys, but we would add a new category of adult-onset offending if data were available. Previous studies have been focused on the antecedents of offending, but we agreed that we would investigate the later life outcomes of these different categories of offenders, compared with nonoffenders. We also agreed that we would investigate predictors of these different categories at different ages. We hope that this special section contributes to knowledge in at least two distinct ways. First, based on four of the largest and most long-lasting longitudinal surveys of aggressive and criminal behavior, this section presents the most extensive information available about the development of adolescencelimited, late-onset, and persistent offenders from childhood to adulthood. Second, because it is very unusual for several different researchers to carry out parallel analyses, this section establishes the extent to which findings about these different categories might be replicable in four different countries. We believe that it is very important to investigate the extent to which results are replicable in different contexts.