Adolescent Family Violence: Findings from a Group-Based Analysis


Current understanding of adolescent family violence (AFV) is fragmented, with research describing offenders and offending at the aggregate level or exploring different forms of AFV in isolation. The current study aimed to describe and compare AFV offender groups drawn from the same population. A sample of 2717 adolescent offenders (12–17 years old) who were reported to Victoria Police for family violence offending in the 2014 calendar year were sorted into five mutually exclusive categories based on the relationship with their primary victim/survivor—mother, father, younger sibling, older sibling and other family members. The groups were then compared across a number of dimensions, including the characteristics of the young person and ‘type’ of violence and abuse perpetrated, as well as prior offending patterms. The analysis identified key differences between offenders based on their relationship with their primary victim/survivor. Mothers emerged as a high-risk cohort because the violence they were experiencing was often frequent and had been ongoing for extended periods of time. They were also the most likely cohort to report being afraid of the offender. Adolescents who were primarily violent towards other family members (eg grandparents) also emerged as key group for future examination due to their use of violence in various contexts, and other offending. Group-based analyses of AFV offenders can provide valuable insights into the differences and similarities underlying this offender population. Future studies should aim to build on this research.

Family violence research conducted since the 1970s has consistently identified children and adolescents as victim/survivors of family violence, which includes witnessing intimate partner violence between parents or carers (Nancarrow et al. 2012). However, it is increasingly recognised that children and adolescents are also perpetrators of violence, capable of both committing significant acts of abuse against family members (Purcell et al. 2014) and causing serious harm (Edenborough et al. 2008; Kiselica and Morrill-Richards 2007; Lopes et al. 2019).

Adolescent family violence is an umbrella term that has been used to describe violence and abuse perpetrated by children and young people (typically below 18 years of age) towards their family members (Moulds et al. 2019). This may include parents/carers (child-to-parent violence), siblings (sibling-to-sibling violence), familial and non-familial carers (child-to-carer violence) and grandparents (elder abuse).

Although the prevalence of adolescent family violence (AFV) is currently unknown (Coogan 2011; Moulds et al. 2016; Walsh and Krienert 2009), reviews of the literature have yielded estimates which suggest that 7–13% of families residing in the community experience AFV (Contreras and Cano 2014; Routt and Anderson 2011). Meanwhile, recent Australian estimates from Victoria indicate that approximately 10 % of all family violence incidents responded to by police involved an adolescent offender (State of Victoria 2016).

Despite accounting for a significant minority of family violence incidents, very little research has explored AFV (Edenborough et al. 2008; Moulds et al. 2016). This is perhaps primarily because of cultural attitudes towards children where we tend to view them as ‘innocents’. As a result, we are quick to blame any negative behaviours displayed by young people on their parents, particularly the mother (Desir and Karatekin 2018; Edenborough et al. 2008; Gallagher 2004), or alternatively, subsume their violence under the more palatable term ‘challenging behaviours’ (Downey 1997).

The absence of research on this issue, as well as our general reticence in naming it for what it is—violence and abuse—means that understanding of this cohort of offenders is limited. This in turn has impacted attempts to identify and support affected families, rehabilitate and treat offenders and also prevent the onset and escalation of these harmful behaviours (Desir and Karatekin 2018; Edenborough et al. 2008). Certainly, within the literature there is a consensus that AFV is a distinct form of family violence that should be theoretically and practically set apart from adult-perpetrated intimate partner and family violence, child abuse and other juvenile violent behaviours (Coogan 2011; Edenborough et al. 2008; Gallagher 2004; Harbin and Madden 1979; Holt 2016; Kiselica and Morrill-Richards 2007). For example, many of the traditional criminal justice responses that may be implemented when adult-perpetrated intimate partner or family violence is detected, such as protection orders and removing the offender from the home, may be inappropriate for juveniles who are often dependent on their victim/survivors for accommodation and financial support. Further, much of the research that has looked at adolescent family violence has examined it from a family systems perspective. This is primarily driven by research findings which suggest that the presence of other forms of abuse within the household (eg intimate partner violence between parents/carers; Brezina 1999), as well as dysfunctional attachment between parents and children and parent mental and physical health issues often co-occur with AFV (Boxall et al. 2020; Downey 1997). This in turn suggests that this form of violence may require a holistic intervention involving the entire family, whereas those tailored to adult family violence offences and juvenile offending typically focus on individual responsibility and accountability (Boxall et al. 2020; Moulds et al. 2019).

Because it is a fledging area of inquiry, it is understandable that current AFV-focussed research has been preoccupied with answering the overarching question of what does AFV ‘look like?’ Particularly, studies have focused on describing the characteristics of young people who perpetrate reported and unreported AFV (eg Freeman 2018), the nature of the violence and abuse (eg Evans and Warren-Sohlberg 1988; Walker and Woerner 2018) and the characteristics of victims/survivors who experience AFV (Desir and Karatekin 2018). In attempting to answer these questions, researchers have either treated AFV populations as homogenous and described them at the aggregate level (Freeman 2018), or have focused on one type of AFV in isolation, particularly child-to-parent violence (Contreras and Cano 2014; Edenborough et al. 2008; Evans and Warren-Sohlberg 1988; Holt 2016; Moulds et al. 2019; Routt and Anderson 2011; Walsh and Krienert 2007). Research that compares different ‘groups’ within AFV populations—beyond gender (see for example Freeman 2018)—are relatively rare (for notable exceptions see Moulds et al. 2019; Nowakowski-Sims 2019).

Group-based analyses—and the related development of typologies—are very common within criminology. Certainly the literature abounds with examples of researchers using various techniques to identify cohorts within the same population that are differentiable from others based on factors like motivation and purpose of the offending behaviours, dimensions of their criminal career (ie when they start and stop), the nature of their offending and characteristics of the offenders themselves (see for example Moffitt 1993). Typologies have been developed for a range of crime types, the most relevant to AFV being intimate partner violence between adults (Holtzworth-Munroe and Stuart 1994; Johnson 2010), serious juvenile criminal offenders (Cohen and Short 1958; Hillege et al. 2017) and juvenile sex offenders (Fox and DeLisi 2018).

The persistent appeal of typologies is that they simultaneously complicate and simplify our understanding of complex criminal behaviours. On the one hand, they help us to ‘see’ the heterogeneity underlying criminal behaviour—not every offender, even those perpetrating similar crimes, are motivated by the same factors, will follow the same patterns of offending, be deterred by the same factors, or pose the same level of threat or risk to victim/survivors (Gibbons 1975; Weber and Bouman 2017). However, typologies or similar classification systems also bring a sense of order to criminal populations, and in turn provide some clues for how we should respond to certain offender groups (Fox and DeLisi 2018). Certainly, IPV researchers have argued that typologies may be useful for identifying individuals who are highest risk for reoffending and perpetrating severe harm—and in turn victims/survivors who are highest risk for being impacted by violence (Kelly and Johnson 2008; Wangmann 2011), matching offenders with appropriate treatment (Capaldi and Kim 2007; Saunders 1996; Weber and Bouman 2017), and determining the most appropriate processes for resolving disputes within abusive relationships in the context of legal separation (Boxall and Payne 2017) as well as the outcomes of these processes (eg child custody arrangements; Altobelli 2009). The extent to which typologies have been able to live up to this promise—and relatedly, have been translated into practice—are questions for another time (Boxall et al. 2015b; Weber and Bouman 2017).

This is not to suggest that we should plunge headfirst into an attempt to identify a typology of young people who perpetrate violence against their family members. Considering the state of the AFV research field, such an analyses would probably be premature. Instead, what is needed is more exploratory research which looks at the differences (and similarities) between young people detected for AFV within the same population, using easily identified criteria, to see what patterns emerge. For practical and conceptual reasons, the relationship between the victim/survivor and young person is perhaps one of the more obvious places to start. On the practical side of things—something which criminologists often forget is a necessary consideration when hoping to translate typologies into practice (Boxall et al. 2015b)—the relationship between the victim/survivor and young person is easily identified by the different organisations and agencies who are first or secondary responders to this form of family violence. In particular, this information can be gleaned by the police who are often the first point of contact between families experiencing AFV and the criminal justice system.

Second, differentiating between young people detected for AFV on the basis of their relationship with the victim/survivor is supported by the research evidence that is available so far. Looking across the growing number of disparate studies of AFV, particularly as it relates to specific forms of AFV like child-to-parent violence (CPV), sibling-to-sibling violence (SSV) and child-to-carer violence (CCV), there is evidence that they may differ on a number of key dimensions, including prevalence, the gender of perpetrators and the motivations underlying the violence. For example, the occurrence of CPV has been attributed to a variety of factors, including family dynamics, the nature and warmth of the relationship between the young person and their parents and communication styles (and deficits) (Brezina 1999; Contreras and Cano 2014). Further, a number of studies have suggested that young people may use violence as a means of stopping the abuse that they themselves are experiencing at the hands of their parents (Brezina 1999; Evans and Warren-Sohlberg 1988). In this way, CPV may have an instrumental purpose (Brezina 1999).

In comparison, causes attributed to SSV include power dynamics between siblings, including ‘sibling rivalry’ and competition for family resources (Kiselica and Morrill-Richards 2007; Noland et al. 2004; Tippett and Wolke 2015), and experiences of abuse from a parent or carer (see for example Noland et al. 2004) and peer-to-peer bullying (Tippett and Wolke 2015). Finally, the small number of studies that have looked at the presence of AFV within kinship carer arrangements has identified that disruptions to attachment to primary carers, lack of communication between carers and children, as well as trauma of being removed from—or given up—by their parents (Bullock and Thomas 2007; Day and Bazemore 2011; Kosberg and MacNeil 2003), may contribute to the abuse of carers.

The above review suggests that there are some key differences and similarities between distinct groups of AFV offenders. However, because these studies have described groups of AFV offenders in isolation, rather than making comparisons between groups drawn from the same population, it is unclear whether these observed variations are attributable to differences in causes and antecedent risk factors associated with the onset of violence towards different family members, or inconsistent sampling methodologies, data sources and definitions of AFV. To build our understanding of AFV offenders, offending and reoffending, this study aimed to explore and describe trends that emerge from an analysis of a group of young people recruited from the same population. Further, exploring variations in offender traits, offending trends and trajectories within the same population is useful for the purpose of risk assessment, particularly the identification of ‘high-risk’ cohorts of offenders and victims/survivors (e.g., those who are likely to be revictimized), and developing appropriately targeted interventions for AFV offenders and victims/survivors.


This study involved the analysis of a sample of young people who were apprehended by Victoria Police for family violence offences in the 2014 calendar year. Adolescents were between the ages of 12 and 17 at time of the reference incident, defined here as the first family violence incident in 2014 for which they were reported to Victoria Police.

The authors were provided with an extract of historical offence data for an initial sample of 3453 young people reported for AFV offences during the 2014 calendar year. Young people were removed from the sample if the relationship between the perpetrator and victim/survivor was not clear at time of the reference incident or the young person was not 12–17 years old in 2014. The final sample consisted of 2717 AFV perpetrators. Two thirds of offenders were male (N = 1800, 66%) with the remaining being female (N = 908, 34%; information was missing for nine young people). The mean age of offenders was 15 years at time of reference incident. These findings are consistent with previous research involving the examination of police data for AFV offenders (Condry and Miles 2013; Coogan 2011).

Data used for the current analysis was drawn from the Law Enforcement Assistance Program (LEAP) which is maintained by Victoria Police. The LEAP is an electronic database that includes information about all offences reported to or detected by Victoria Police, regardless of outcome (eg arrest or charge). The dataset is fully relational, in that all offences are assigned a unique ID, and individual offenders and victim/survivors are assigned one ID that can be linked across offences.

In addition to the minimum information recorded for all offences recorded in the LEAP (eg date and time of offence and charge), officers responding to family violence are also required to complete an L17 form which asks for detailed information about the characteristics of the the incident, including the characteristics of the offender, victim/survivor (referred to in LEAP as the affected family member), the nature of the violent and abusive behaviour (eg physical assault, financial abuse) and the history of violence between the victim/survivor and the young person. At time of data extraction, L17 forms were completed in hardcopy but then uploaded into the LEAP and linked to the original offence data.

Historical offence data was extracted for all identified young people, for both family violence and non-family violence incidents. This included all family violence incidents involving the initial offender sample that were reported to police prior to their reference incident, as well as their prior charge data (criminal offences that resulted in the young person being charged with an offence). Charge data included offences that were committed outside of family contexts, such as property, justice and drug offences. Charge data was prioritised over apprehension data as it ‘filters’ out offences that were deemed insufficient to be charged by police, generally highlighting the more ‘severe’ offences of adolescents.

Data was extracted by Victoria Police data custodians on request by the first author. This research was approved by the Australian Institute of Criminology’s Human Research Ethics Committee.


For consistency with Victorian legislation (Family Violence Protection Act 2008 (Vic)), family violence was defined to include both physical (eg sexual assault, pushing, shoving, hitting), non-physical family violence offences (eg stalking and intimidation, verbal abuse, financial abuse, emotional/psychological abuse) and breaches of Family Violence Intervention Orders (ie protection orders). Family members include both biological and non-biological relatives, including fathers, mothers, grandparents, step-parents, siblings and carers. It does not include current or former intimate partners.

The relationship between the young person and their primary victim/survivor was used to sort young people into five mutually exclusive groups (see Table 1). For simplicity, the primary victim/survivor for each young person was identified based on the number of occasions they were identified as the affected family member in AFV incidents involving the offender. Specifically, the victim/survivor involved in the largest number of incidents was coded as the primary victim/survivor. This measure gives a more appropriate indication of who the adolescent predominantly targeted, rather than who the victim/survivor was at the time of the reference incident. This approach is consistent with methods used by other researchers to identify the target of AFV (Contreras and Cano 2014).

Table 1 Offender groups

The relationship between the victim/survivor and offender was determined based on information provided in the L17 forms. However, it is important to note that although the L17 form identifies whether the victim/survivor was a family member, they provide very little detail about the specific nature of the relationship. The notable exception to this is parent/child relationships.

For the purpose of this study, the following definitions were used to differentiate between familial relationships:

  • mother – includes biological and adoptive, identified as parent/child relationship, and that the victim/survivor was female;

  • father - includes biological and adoptive, identified as parent/child relationship, and that the victim/survivor was male;

  • older sibling – includes biological and adoptive, identified as a family member, was under the age of 21 years at time of the reference incident, and was younger than the offender at time of the reference incident;

  • younger sibling – includes biological and adoptive, identified as a family member, was under the age of 21 years at time of the reference incident, and was younger than the offender at time of the reference incident;

  • other family member – includes foster family members, grandparents, aunts and uncles, identified as a family member who did not meet the criteria for the other groups.

As shown in Table 1, the most common primary target of AFV was mothers (N = 1595, 59%), with three quarters of primary victims/survivors being female (N = 2002, 76%). Fathers were the primary victim in 15% of cases. Younger siblings were the primary target of the violence in 10% of cases (N = 268), while older siblings were the primary targets in 6 % of cases (N = 151). Finally, other family members such as grandparents and foster carers were the primary target of the violence in 11% of situations (N = 292).

Data Analysis

Once extracted, the data was analysed by the authors using STATA software. Critically, because this study was exploratory in nature and underpinned by grounded theory principles, the data was allowed to ‘speak for itself’ – this means the analysis was not guided by specific hypotheses or theoretical frameworks (Charmaz and Belgrave 2012). Instead, themes and findings were emergent from the data. For simplicity, data was analysed using bivariate methods (chi-square tests of significance).

Differences between the groups of young people identified in Table 1 were examined across a small number of factors: the gender and age of adolescents, nature of the abusive behaviour (type and frequency), impact of the violence (victim/survivor fear), family violence histories, and other offending histories.


Gender of Offenders

Consistent with the previous literature, a larger proportion of males than females were apprehended by Victoria Police for AFV offending, and this held across all offender-victim/survivor relationships; mothers (62% vs 38%), fathers (72% vs 28%), younger siblings (85% vs 15%), older siblings (73% vs 27%) and other family members (65% vs 35%). However, although mothers were the most common primary targets of AFV for both male and female offenders, female offenders were statistically more likely than males to target their mothers (67% vs 54%, χ2(4) = 63.69, p < 0.001). In comparison, males were more likely to be primarily violent towards their fathers (16% vs 13%) and younger siblings (13% vs 5%) than their female counterparts. However, males and females were as likely as eachother to be violent towards other family members (11% vs 11%) and older siblings (6% vs 4%).

Age of Offenders

Adolescents who were primarily violent towards their younger (N = 74, 28%) or older siblings (N = 35, 23%) were themselves younger (12–13 years) at time of the reference incident. Meanwhile, adolescents who were violent towards their fathers (N = 191, 47%) or other family members (N = 143, 49%) were likely to be older (16–17 years). These differences were statistically significant (χ2(8) = 23.4, p = 0.003).

Type and Frequency of Abuse

The nature of the abusive behaviours used by young people against their primary victim – both at time of reference incident and historically - differed across the five groups (see Table 2). Adolescents who targeted their siblings were more likely to be physically violent towards them when compared with the other family members (χ2(4) = 11.5, p = 0.02). For example, 26% of adolescents primarily targeting their younger (N = 70) or older (N = 39) siblings had been physically violent towards them on a least one occasion, compared with one in five young people who were physically violent towards their mothers (N = 306, 19%) or fathers (N = 73, 18%). Further, adolescents targeting younger siblings were more likely to be reported for sexual violence on at least one occasion, relative to other groups (N = 139, 52%; Fischer’s exact p < 0.001).

Table 2 Reported family violence offending types, by primary victim/survivor group (%)

Young people who were primarily violent towards their mothers were statistically more likely to use non-physical means to intimidate and control them than other groups. In particular, adolescents were more likely to damage their property (N = 295, 19%; χ2(4) = 55.8, p < 0.001) and be emotionally abusive towards them (N = 1342, 84%; χ2(4) = 276.2, p < 0.001). Adolescents who were violent towards other family members were more likely to be reported for threatening violence (N = 30, 10%, χ2(4) = 18.0, p = 0.001).

The majority of young people had only been violent towards the primary victim/survivor on one occasion—the reference incident (N = 2068, 76%). Twelve percent (N = 322) had been reported for violence and abuse towards their primary victim on two occasions, or three or more times (N = 327, 12%).

Adolescents who primarily targeted their mothers were statistically more likely to have been violent towards them on three or more occasions than other groups (N = 267, 17%; χ2(8) = 129.56, p < 0.001). In comparison, adolescents who were violent towards family members other than their mother (eg their father) were more likely to have only been violent towards them once. What this indicates is that once the abusive behaviour commences, mothers are more likely than fathers, siblings and other family members to be subjected to ongoing and repeated reported violence (see Table 3).

Table 3 Family violence history, by primary victim/survivor group (%)

Impact on the Victim/Survivor—Feelings of Fear

As part of the L17 form, police are asked to assess the extent to which the victim/survivor was afraid of the adolescent after the incident. In the majority of cases, primary victim/survivors were not reported as being fearful (N = 2078, 77%). However, this was not consistent across primary victim/survivor groups. Again, mothers were more likely to report being fearful or very fearful (N = 413, 26%) when compared with the other groups, while fathers (N = 55, 13%) were less likely to be fearful or very fearful (χ2(4) = 30.1, p < 0.001).

History of Family Violence Offending

The family violence offending histories of adolescents were analysed across a small number of dimensions:

  • age at time of first reported family violence offence;

  • number of family violence victims/survivors (including the primary victim); and

  • and number of prior family violence offences and charges (see Table 3).

Overall, the majority of adolescents were 14 years or older when they were first came to the attention of police for family violence offending (N = 1909; 70%). One in four (N = 708, 26%) were 12–13 and a small number were 7–11 years old (N = 100, 4%).

Although only accounting for a small proportion of the sample, as shown in Table 3 it appears that young people primarily targeting their mothers were more likely to be reported for their abusive behaviours at a younger age than the other groups (χ2(9) = 50.6, p < 0.001). On the other hand, adolescents primarily targeting their fathers (N = 156, 38%) or other family members (N = 108, 37%) were older at time of their first reported violent offence against their primary victim.

The majority of adolescents had only been detected for violence towards one family member—the primary victim/survivor (N = 2280, 84%; see Table 3). Approximately one in ten had been violent towards two family members (N = 333, 12%) and 4 % had three or more detected victim/survivors (N = 104). Young people who were primarily targeting other family members were more likely to have been violent towards multiple family members (3 or more, N = 23, 8%) than the other groups. In comparison, adolescents primarily violent towards their mothers were more likely to have only been reported for violent towards her (N = 1362, 85%; χ2(8) = 20.9, p = 0.007).

Other Offending

Two in five adolescents (N = 1065 39%) had previously been charged with any non-family violence related offence. More specifically:

  • one in three (N = 965, 36%) had been charged with a property offence;

  • one in 20 (N = 171, 6%) had been charged with a drug offence; and

  • one in seven (N = 342, 13%) had been charged with a justice-related offences (eg breach of orders).

Notably, a significant proportion of the sample also had histories of violence towards non-family members. Although only a small number had been charged with an intimate partner violence offence (against a current or former partner; N = 58, 2%), one in five (N = 510, 19%) had been charged with a violent offence against someone who was not a current or former intimate partner or a family member, such as a stranger, acquaintance, teacher or friend.

Young people who were primarily violent towards other family members were more likely to have histories of non-family violence-related offending. Not only were they more likely to have been charged for any non-family violence related offence than other groups (N = 143, 49%; χ2(4) = 15.8, p = 0.003), they were more likely to have been charged with a property offence (N = 128, 44%; χ2(4) = 19.7, p = 0.001) and justice offence (N = 54, 18%; χ2(4) = 19.4, p = 0.001). Further, as shown in Fig. 1, this group were also more likely to be violent in non-familial contexts (N = 84, 29%; χ2(4) = 26.8, p < 0.001) and towards intimate partners (N = 18, 29%; χ2(4) = 10.3, p = 0.036). In comparison, young people targeting their mothers or fathers were less likely to be violent towards non-family members or intimate partners.

Fig. 1

Charges for non-family violence related offences, by offender group (%). ote: Percentage totals do not equal 100 due to non-mutually exclusive categories. a: Limited to violent offences that were perpetrated against non-intimates or family members (eg strangers, acquaintances, teachers, friends). * p < 0.05 ** p < 0.01 *** p < 0.001. Source: Victoria Police 2014–2015 [computer file]

In addition to being more likely to be charged with violence towards non-family members, adolescents targeting other family members were also more likely to have been charged on multiple occasions (see Fig. 1). For example, a quarter (N = 73, 25%) of this group had been charged four or more times with other violent offences, compared with 17% (N = 26) of those who were primarily violent towards older siblings (χ2(8) = 35.7, p < 0.001). On average, young people targeting other family members had been charged with 5.1 other violent offences (range = 0–97). The next highest was young people targeting older siblings, with an average of 3.8 violent offences (range = 0–93). Similar results were found in relation to property offending, with adolescents who had been primarily violent towards other family members more likely to have been charged for four or more property offences (N = 63, 22%; χ2(12) = 31.4, p = 0.002). There was no relationship between the offender groups for the number of drug charges (χ2(12) = 11.4, p = 0.498).


The aim of the current study was to expand our currently limited understanding of AFV by exploring the differences and similarities between groups of AFV offenders drawn from the same population. Differentiating between young people by their relationship to their primary victim/survivor resulted in a number of valuable insights. In particular, there was evidence that certain groups—of offenders and victim/survivors—may be higher-risk than others for experiencing negative outcomes as result of the violence and abuse perpetrated against them.

Unsurprisingly, the first high-risk group is mothers. The overall picture that emerged from the study is that violence targeted at mothers is ongoing, frequent, starts earlier and is likely to make women fearful for their safety. This is certainly consistent with previous research which highlights that not only are mothers the most common target of AFV, they are also one of the most vulnerable groups for experiencing harm as a result (Fitzgibbon et al. 2018). The cumulative impact of sustained patterns of violence can be significant, including loss of employment, breakdown in relationships with the young person and other family members and deterioration in the general health and wellbeing of victims/survivors (Edenborough et al. 2008).

It is also worth noting that young people who were targeting their mothers were unlikely to be violent to other family members or in other contexts, particularly outside the home. This may further contribute to mothers’ vulnerability because it closes off potential avenues for help-seeking. If the adolescent were violent towards other people, their likelihood of being detected and reported for their behaviour would potentially increase, as would the network of people who are aware of the abuse and could provide assistance. When the violence is targeted solely at mothers and occurs behind closed doors, women may become isolated. The desire to keep the violence hidden can be heightened for mothers who are concerned about being blamed for the abuse, want to protect their children from negative outcomes associated with their contact with the criminal justice system, and maintain the façade of a happy home (Edenborough et al. 2008; Fitzgibbon et al. 2018).

The second group of high-risk victims/survivors are younger siblings. Although they were unlikely to be abused on more than one occasion, they were statistically more likely to experience physical and sexual violence. In particular, half of younger siblings being primarily targeted by their siblings were experiencing some form of sexual violence. The long and short-term impacts of physical and sexual violence range significantly, although it can involve mental illness, injury and even death (Relva et al. 2013). The vulnerability of younger siblings to experiencing physical harm is heightened by the physical disparities between them and adolescents, which means they are less able to protect themselves and escape (Coogan 2011; Eriksen and Jensen 2006). This could also explain why younger siblings are targeted by adolescents—their relative physical strength means that they are low-risk targets.

Another potential reason underpinning adolescent’s use of physical and sexual violence against their younger siblings is that their abusive actions are difficult to detect and so there is less risk involved. Detection is difficult not only because the behaviour itself may be hidden, but also because physical violence behaviour may be mislabelled as sibling conflict and rivalry which is a relatively typical occurrence in many families, while sexual violence and abuse may be viewed as sexual ‘experimentation’ or ‘exploring’ (Relva et al. 2013). Targets of the violence may even have difficulties identifying their experiences as abusive (McDonald and Martinez 2016).

The final group that will be discussed here are young people targeting other family members. These young people emerged as a group of particular interest because of their frequent use of violence against multiple people within familial and non-familial contexts, and their engagement in other offending. These young people were also more likely to be reported for multiple incidents of violence against victims/survivors. Although we did not examine the relationship between young people and their primary targets within this group, it is likely that in many situations the victim/survivor was the young person’s grandparent who was providing care in the context of formal or informal kinship carer arrangements. Because of their age, elderly carers are vulnerable to experiencing a range of negative harms associated with AFV, including significant injury. Considering the increasing pressure on statutory child protection systems to prioritise kinship carer arrangements over non-familial and institutional arrangements, there is an obvious need to explore the abuse of elderly carers in the context of kinship arrangements in more depth.

That young people who were targeting other family members were also involved in other forms of criminal behaviour – both violent and non-violent – indicates that family violence may be part of the overall criminal careers of these young people. This is certainly consistent with other research from the family and domestic violence space, which has found evidence of some offenders being ‘generalists’, with extensive and varied offending careers of which family violence comprises only one part (Boxall et al. 2015a; Coghlan and Millsteed 2017; Holtzworth-Munroe and Stuart 1994; Moulds et al. 2019).

Although it was beyond the scope of this study to explore the role of family violence in the offending trajectories of adolescents, this would be a valuable area of future research. In particular, understanding the role of family violence in the offending trajectories of young people could provide some insight into the causes of the violence, as well as potential treatment options. For example, that some groups of young people are involved in both AFV and other forms of criminal offending indicates that the risk factors for AFV and general offending may overlap in some cohorts, which in turn which could be indicative of a generally antisocial or violent personality, or low self-control and emotional dysregulation. If true, the treatment needs of this cohort of offender are likely be very different from those of a young person who is ‘only’ violent within familial contexts, particularly mothers. That these young people appear to be capable of controlling their behaviours in non-familial contexts would suggest that their abusive behaviours are driven by a desire for control and dominance, or a need to ‘punish’ their family members.

The perceived ‘targeting’ of mothers could be in part influenced by their views of the role of their mothers in their lives, for example that their mothers are primarily responsible for their happiness, wellbeing and providing them with whatever support they require. Although the lens of dominance and control has primarily been used to understand male-perpetrated intimate partner and family violence (Stark 2009), it may be helpful for explaining the occurrence of AFV perpetrated against mothers, by both daughters and sons (Downey 1997). In situations where mothers are perceived to have ‘failed’ or diverged from acceptable behaviours as defined by young people, abuse and violence may become a means by which they attempt to change these behaviours or punish them.

Taken together, what this suggests is that identifying the primary target of the violence, the gender of the primary target, as well as the occurrence of others forms of offending (violent and non-violent) could be important for identifying the underlying causes for the violence and abuse perpetrated by young people, as well as possible treatment options. For example, young people who are violent towards other family members and are involved in other offending may benefit from broad-based intervention programs that aim to mitigate the underlying causes of the offending behaviour (eg prior trauma and emotional dysregulation), as well as co-occurring situational factors that may exacerbate these behaviours (eg alcohol and drug use). However, for young people who are targeting their mothers and family members in isolation, treatment may focus on their relationship with family members, attachment to parents, and providing them with support to ‘unlearn’ negative attitudes or views of women and the role of mothers in their lives and societally. Future research should focus on delving into these findings in more detail, by exploring this new proposed schema within different populations of young people who have been detected for AFV offending, as well as using this schema to predict future reoffending and treatment outcomes.


This study represents one of the first attempts to differentiate between groups of AFV offenders using easily identifiable criteria that are operationable by first and second responder agencies like the police. Further, due to improvement in data collection by Victoria Police (and Australian police agencies in general) regarding incidents of domestic and family violence over the past 10 years, it was possible to extract data for a large enough sample size to facilitate bivariate analyses of the differences between these groups.

This said, there are several limitations associated with police recorded data, the most important being that it only captures incidents that are reported to authorities. Although there is very little research which has looked at factors associated with reporting of AFV to the police, some insights may be gleaned from the intimate partner violence research. In particular, this research has shown that incidents involving physical violence where the victim/survivor felt they were more likely to be seriously harmed as a result (eg the offender was intoxicated or they had a weapon) were more likely to be reported to the police, or where there had been prior abuse and violence (Voce and Boxall 2018). If these findings are consistent when applied to the AFV space, it could be that only more serious or ‘severe’ incidents of violence and abuse perpetrated by young people are being reported to the police.

Second, the information collected is a by-product of mandatory police recording methods, where research is not the primary objective (Beare 2012). Finally, police data is subject to a level of discretion by the officer recording the information. For example, the current dataset included information about the victim/survivor’s level of fear and offender’s likelihood of reoffending, where a level of discretion and judgment must be exercised by the attending officer (Beare 2012).


One of the main benefits of groups-based or typological analyses is that it demonstrates that within crime categories, offences and offenders are not a ‘unitary phenomenon’ (Johnson 2010: 3). Certainly the current study has demonstrated that AFV offenders’ do not all look the same, and do not follow the same patterns of behaviour. However, the relationship between adolescents and primary victim/survivors was only selected as a starting point for sorting AFV offenders into groups. Other options for identifying offender groups include their psychological profile and the presence of specific cognitive traits (eg lack of empathy) and the primary form of violence used against family members. However, the findings from this study suggest that there may be utility in exploring a schema that focuses on the presence of non-familial offending (ie distinguishing between generalists and specialists), and the gender of the target of the violence.

Regardless of whether the current approach is the most appropriate means for differentiating between offenders, it does demonstrate the potential viability of such approaches in understanding AFV. Certainly, this analysis has identified the potential of using typologies for identifying who is most at risk of experiencing harm as a result of AFV, and so should be prioritised for engagement in support services aimed at addressing the behaviour.


  1. Altobelli, T. (2009). Family violence and parenting: future directions in practice. Australian Journal of Family Law, 23(3), 194.

    Google Scholar 

  2. Beare, M. (2012). Data sources: Police data. In M. Beare (Ed.), Encyclopaedia of transnational crime and justice. SAGE Publications: US.

    Google Scholar 

  3. Boxall, H., & Payne, J. (2017). Another tool in their arsenal? The potential of domestic violence typologies to inform family law alternative dispute resolution processes. Family Law Review, 7(1), 20–29.

    Google Scholar 

  4. Boxall, H., Payne, J., & Rosevear, L. (2015a). Prior offending among family violence perpetrators: A Tasmanian sample. Trends & Issues in Crime and Criminal Justice, no. 493, Canberra: Australian Institute of Criminology

  5. Boxall, H., Rosevear, L., & Payne, J. (2015b). Domestic violence typologies: What value to practice? Trends and Issues in Crime and Criminal Justice, no. 494, Canberra: Australian Institute of Criminology.

  6. Boxall, H., Morgan, A., Voce, I., & Coughlan, M. (2020). Responding to adolescent family violence: Findings from an impact evaluation. Trends & issues in crime and criminal justice no. 601. Canberra: Australian Institute of Criminology.

  7. Brezina, T. (1999). Teenage violence toward parents as an adaptation to family strain evidence from a national survey of male adolescents. Youth & Society, 30, 416–444.

    Article  Google Scholar 

  8. Bullock, K., & Thomas, R. L. (2007). The vulnerability for elder abuse among a sample of custodial grandfathers: an exploratory study. Journal of Elder Abuse & Neglect, 19, 133–150.

    Article  Google Scholar 

  9. Capaldi, D. M., & Kim, H. K. (2007). Typological approaches to violence in couples: a critique and alternative conceptual approach. Clinical Psychology Review, 27(3), 253–265.

    Article  PubMed  Google Scholar 

  10. Charmaz, K., & Belgrave, L. (2012). Qualitative interviewing and grounded theory analysis. SAGE Handb. Interview Res. Complex. Craft 2, 347–365.

  11. Coghlan, S., & Millsteed, M., (2017). Identifying the differences between generalist and specialist family violence perpetrators: Risk factors and perpetrator characteristics. In Brief, no. 8 Melbourne: Crime statistics agency

  12. Cohen, A. K., & Short Jr., J. F. (1958). Research in delinquent Subcultures1. Journal of Social Issues, 14, 20–37.

    Article  Google Scholar 

  13. Condry, R., & Miles, C. (2013). Adolescent to parent violence: framing and mapping a hidden problem. Criminology & Criminal Justice, 14(3), 257–275.

    Article  Google Scholar 

  14. Contreras, L., & Cano, C. (2014). Family profile of young offenders who abuse their parents: a comparison with general offenders and non-offenders. Journal of Family Violence, 29, 901–910.

    Article  Google Scholar 

  15. Coogan, D. (2011). Child-to-parent violence: challenging perspectives on family violence. Child Care in Practice, 17, 347–358.

    Article  Google Scholar 

  16. Day, S. E., & Bazemore, G. (2011). Two generations at risk: child welfare, institutional boundaries, and family violence in grandparent homes. Child Welfare, 90, 99.

    PubMed  Google Scholar 

  17. Desir, M. P., & Karatekin, C. (2018). Parental reactions to parent-and sibling-directed aggression within a domestic violence context. Clinical Child Psychology and Psychiatry, 23, 457–470.

    Article  PubMed  Google Scholar 

  18. Downey, L. (1997). Adolescent violence: a systemic and feminist perspective. Australian and New Zealand Journal of Family Therapy, 18, 70–79.

    Article  Google Scholar 

  19. Edenborough, M., Jackson, D., Mannix, J., & Wilkes, L. M. (2008). Living in the red zone: the experience of child-to-mother violence. Child & Family Social Work, 13, 464–473.

    Article  Google Scholar 

  20. Eriksen, S., & Jensen, V. (2006). All in the family? Family environment factors in sibling violence. Journal of Family Violence, 21, 497–507.

  21. Evans, E. D., & Warren-Sohlberg, L. (1988). A pattern analysis of adolescent abusive behavior toward parents. Journal of Adolescent Research, 3, 201–216.

    Article  Google Scholar 

  22. Fitzgibbon, K., Elliott, K., & Maher, J. (2018). Investigating adolescent family violence in Victoria: Understanding experiences and practitioner perspectives. Monash: Monash University.

    Google Scholar 

  23. Fox, B., & DeLisi, M. (2018). From criminological heterogeneity to coherent classes: developing a typology of juvenile sex offenders. Youth Violence and Juvenile Justice, 16(3), 299–318.

    Article  Google Scholar 

  24. Freeman, K. (2018). Domestic and family violence by juvenile offenders: offender, victim/survivor and incident characteristics. Crime and Justice Statistics Brief, 136. Retrieved from

  25. Gallagher, E. (2004). Parents victimised by their children. Australian and New Zealand Journal of Family Therapy, 25, 1–12.

    Article  Google Scholar 

  26. Gibbons, D. C. (1975). Offender typologies: two decades later. British Journal of Criminology, 15(2), 140–156.

    Article  Google Scholar 

  27. Harbin, H., & Madden, D. (1979). Battered parents: a new syndrome. The American Journal of Psychiatry, 136, 1288–1291.

    CAS  Article  PubMed  Google Scholar 

  28. Hillege, S. L., Brand, E. F. J. M., Mulder, E. A., Vermeiren, R. R. J. M., & van Domburgh, L. (2017). Serious juvenile offenders: classification into subgroups based on static and dynamic charateristics. Child and Adolescent Psychiatry and Mental Health, 11, 67.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Holt, A. (2016). Adolescent-to-parent abuse as a form of “domestic violence”, a conceptual review. Trauma, Violence, & Abuse, 17, 490–499.

    Article  Google Scholar 

  30. Holtzworth-Munroe, A., & Stuart, G. (1994). Typologies of male batterers: three subtypes and the differences among them. Psychological Bulletin, 116, 476–497.

    CAS  Article  PubMed  Google Scholar 

  31. Johnson, M. (2010). A typology of domestic violence: Intimate terrorism, violent resistance, and situational couple violence. Boston: Northeastern University Press.

    Google Scholar 

  32. Kelly, J. B., & Johnson, M. P. (2008). Differentiation among types of intimate partner violence: research update and implications for interventions. Family Court Review, 46(3), 476–499.

    Article  Google Scholar 

  33. Kiselica, M. S., & Morrill-Richards, M. (2007). Sibling maltreatment: the forgotten abuse. Journal of Counseling & Development, 85, 148–160.

    Article  Google Scholar 

  34. Kosberg, J. I., & MacNeil, G. (2003). The elder abuse of custodial grandparents: a hidden phenomenon. Journal of Elder Abuse & Neglect, 15, 33–53.

    Article  Google Scholar 

  35. Lopes, P. P., Relva, I. C., & Fernandes, O. M. (2019). Psychopathology and sibling violence in a sample of Portuguese adolescents. Journal of Child & Adolescent Trauma, 12, 1–11.

    Article  Google Scholar 

  36. McDonald, C., & Martinez, K. (2016). Parental and others’ responses to physical sibling violence: a descriptive analysis of victim/survivors’ retrospective accounts. Journal of Family Violence, 31, 401–410.

    Article  Google Scholar 

  37. Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: a developmental taxonomy. Psychological Review, 100, 674–701.

    CAS  Article  PubMed  Google Scholar 

  38. Moulds, L., Day, A., Mildred, H., Miller, P., & Casey, S. (2016). Adolescent violence towards parents – the known and unknowns. Australian and New Zealand Journal of Family Therapy, 37, 547–557.

    Article  Google Scholar 

  39. Moulds, L., Mayshak, R., Mildred, H., Day, A., & Miller, P. (2019). Adolescent family violence towards parents: a case of specialisation? Youth Justice, 19(3), 206–221.

    Article  Google Scholar 

  40. Nancarrow, H., Hanley, C., & Viljoen, R. (2012). Domestic and family violence: Awareness, attitudes and experiences (preliminary findings). Rockhampton: Central Queensland University.

    Google Scholar 

  41. Noland, V. J., Liller, K. D., McDermott, R. J., Coulter, M. L., & Seraphine, A. E. (2004). Is adolescent sibling violence a precursor to college dating violence? American Journal of Health Behavior, 28, S13–S23.

    Article  PubMed  Google Scholar 

  42. Nowakowski-Sims, E. (2019). An exploratory study of childhood adversity and delinquency among youth in the context of child-to-parent and sibling-to-sibling violence. Journal of Family Social Work, 22, 126–145.

    Article  Google Scholar 

  43. Purcell, R., Baksheev, G. N., & Mullen, P. E. (2014). A descriptive study of juvenile family violence: data from intervention order applications in a children’s court. International Journal of Law and Psychiatry, 37, 558–563.

    Article  PubMed  Google Scholar 

  44. Relva, I., Fernandes, O., & Mota, C. (2013). An exploration of sibling violence predictors. Journal of Aggression, Conflict and Peace Research, 5(1), 47–61.

    Article  Google Scholar 

  45. Routt, G., & Anderson, L. (2011). Adolescent violence towards parents. Journal of Aggression, Maltreatment & Trauma, 20(1), 1–19.

    Article  Google Scholar 

  46. Saunders, D. (1996). Feminist-cognitive-behavioral and process-psychodynamic treatments for men who batter: interaction of abuser traits and treatment models. Violence and Victims, 11, 393–414.

    CAS  Article  PubMed  Google Scholar 

  47. Stark, E. (2009). Coercive control: The entrapment of women in personal life. Oxford University Press.

  48. State of Victoria (2016). Royal Commission into family violence: Summary and recommendations. Melbourne: Victorian Government.

  49. Tippett, N., & Wolke, D. (2015). Aggression between siblings: associations with the home environment and peer bullying. Aggressive Behavior, 41, 14–24.

    Article  PubMed  Google Scholar 

  50. Voce, I., & Boxall, H. (2018). Who reports domestic violence to police? A review of the evidence. Trends & issues in crime and criminal justice, no. 559. Canberra: Australian Institute of Criminology.

  51. Walker, S., & Woerner, J. (2018). Adolescent sibling violence in Victoria. In fact no. 8. Melbourne: Crime statistics agency

  52. Walsh, J., & Krienert, J. (2007). Child–parent violence: an empirical analysis of offender, victim, and event characteristics in a national sample of reported incidents. Journal of Family Violence, 22, 563–574.

  53. Walsh, J., & Krienert, J. (2009). A decade of child-initiated family violence: comparative analysis of child-parent violence and parricide examining offender, victim/survivor, and event characteristics in a national sample of reported incidents, 1995-2005. Journal of Interpersonal Violence, 24, 1450–1477.

    Article  PubMed  Google Scholar 

  54. Wangmann, J. (2011). Different types of intimate partner violence-an exploration of the literature. Australian Domestic & Family Violence Clearinghouse.

  55. Weber, T., & Bouman, Y. H. A. (2017). Intimate partner violence: variations in perpetrators and treatment allocation. Journal of Interpersonal Violence, 35(3–4), 1012–1034.

    Article  PubMed  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Hayley Boxall.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Boxall, H., Sabol, B. Adolescent Family Violence: Findings from a Group-Based Analysis. J Fam Viol (2021).

Download citation


  • Family violence
  • Adolescent
  • Youth
  • Sibling abuse
  • Carer abuse