Nature of violence by girls
Research has shown that in general, the nature, severity, frequency, and victim characteristics of violent offenses committed by girls are significantly different from those committed by boys. Violence by girls is more often reactive and within social relations en less often instrumental compared to violence by boys. Motives for offenses committed by girls were more often seen in the social sphere or within relations (revenge, jealousy and gossip) than in boys. In a recent U.S. study of the explanations for violent offenses by girls as seen by probation officers, the three most frequently cited explanations for girls were: 1) emotional outburst; 2) relational violence; and 3) history of abuse (Fusco, 2011). For boys, these three explanations were not once mentioned. The three most frequently cited explanations for violent offenses for boys were: 1) ego driven; 2) peer pressure; and 3) survival. The most common victims of violence by girls are brothers / sisters and peers.
Research, however, also demonstrated that there is a subgroup of girls/young women who seem to show more 'masculine' forms of violence (e.g., see a report by MacKenzie & Johnson, 2003) . In this subgroup of females, instrumental aggression, hostility, committing robberies and criminal gang membership is more prevalent.
Risk and protective factors for violence in girls
Research has demonstrated that - although many risk factors are valid for both boys and girls, there are several risk factors that have stronger predictive validity in girls, for example, abuse and childhood traumas, problems with social interacton and family and running away from home. Hawkins and colleagues (2009) found that family connectedness and religiosity provided significant protection for girls, but not for boys. Positive social relationships were found to have a stronger protective effect for adolescent girls compared to boys. Currently, a youth version of the SAPROF suited for both boys and girls is being developed; the SAPROF - Youth Version.
Risk assessment in girls
Research has demonstrated that general risk assessment tools have less strong predictive power in girls than in boys and that gender-responsive tools are needed. Despite the many important advances in the field of violence risk assessment in the past thirty years and the fact that many risk assessment tools have become available for populations of different ages and for different types of violence, virtually no ‘specific’ tools have been developed for the assessment of risk for antisocial or violent behavior in adolescent female offenders. One exception is the Early Assessment Risk List for Girls (EARL-21G; Levene et al., 2001) for girls between 6 and 12 years old. Other than the risk factors valid for both boys and girls, this instrument contains two items specific to girls; Caregiver-daughter interaction and Sexual development. Positive results have been found in terms of reliability, predictive validity and clinical applicability of the EARL-21G (Augimeri, Enebrink, Walsh, & Jiang, 2010).