Childhood broken homes and adult violence: An analysis of moderators and mediators Delphine Theobald a , David P. Farrington b , Alex R. Piquero c, ⁎ a King's College London, Institute of Psychiatry, London, United Kingdom b Cambridge University, Institute of Criminology, Cambridge, United Kingdom c University of Texas Dallas, Program in Criminology, 800 W. Campbell Rd., GR31, Richardson, TX, 75080-3021, United States abstract article info Available online 18 January 2013 Purpose: Children who experience a family breakdown due to the separation or divorce of their parents may be at an increased risk of violent offending especially if they have experienced a disadvantaged upbringing that included low family income, marital disharmony, or parental criminality. Methods: Using data from the Cambridge Study in Delinquent Development, we examine the effect of experienc- ing a broken home on subsequent violence. We further examine possible mediators and moderators of this effect. Results: We find that self-reported violence and hyperactivity act as possible mediators on the effect of a broken home on later violent convictions. We also find evidence for moderation of this effect by harsh discipline, nervousness, low family income, and having a young mother. Conclusions: We discuss the findings in the context of future research and possible policy recommendations. © 2012 Elsevier Ltd. All rights reserved. Introduction Traditionally, a common theme in many criminological theories is the important role of the family in socializing children (Hirschi, 1995; Farrington, 2010). When families do not socialize their children or when families break apart, the negative effects that may emerge will involve both the adults and the children, especially if there is con- tinued conflict between them (Thornberry et al., 1999). This break- down can be regarded as a process with problems that emerge years before an actual separation or divorce—often beginning with one or both partners disengaging emotionally from the relationship (Amato, 2000). This disengagement can also result in conflict over emotional matters and financial issues (Booth & Amato, 1991, 2001). From the parents' point of view, the breakdown may act as a trigger for increases in their own antisocial and offending behavior (Farrington & West, 1995; Horney et al., 1995; Theobald & Farrington, 2012). Conflict between the parents can also have delete- rious effects on the children where they can experience inconsistent parenting, reduced affection and warmth, and reduced supervision— all of which have been found to be predictive of later delinquency (Farrington, 1992; Rebellon, 2002). 1 This paper examines the effect of experiencing family breakdown due to separation/divorce (up to the age of 14) by the males in the Cambridge Study in Delinquent Development (CSDD) and the effect this life event has on their subsequent violent offending. Such an in- vestigation is relevant to extant research because of the general lack of longitudinal investigations of long-term effects of broken homes on adult functioning—especially with respect to violent criminal behavior—and may help us to understand the mechanisms involved. Before we present the results of our investigation, we first briefly re- view prior research on the effect of family disruption on later delin- quency and offending. Effects of marital breakdown on children's offending With increases in family instability over the last 40 years in the UK and other westernized nations, the effect of family disruption on chil- dren has become one of the primary areas of research in criminology and other disciplines such as developmental and family psychology (Amato, 2001; Wells & Rankin, 1991). This link between broken homes and delinquency has been established over several decades using data from some important longitudinal studies (Glueck & Glueck, 1950; West & Farrington, 1973; Fergusson et al., 1986; Kolvin et al., 1988; Mednick et al., 1990; Farrington, 1992; Coughlin & Vuchinich, 1996; Juby & Farrington, 2001). For example, in the Thousand Family birth co- hort study of Newcastle boys, Kolvin et al. (1988) found that the percent- age of boys who experienced family disruption up to the age of 5 years who were convicted up to age 32 was doubled (53%, compared with 28% of the remainder). Farrington (1992) found that disrupted families were as strong a predictor of delinquency as other major risk factors (e.g., hyperactivity, low family income, large family size, low school at- tainment, poor parental supervision and poor child rearing). Journal of Criminal Justice 41 (2013) 44–52 ⁎ Corresponding author. E-mail address: apiquero@utdallas.edu (A.R. Piquero). 0047-2352/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jcrimjus.2012.12.003 Contents lists available at SciVerse ScienceDirect Journal of Criminal Justice