An analysis of current Australian program initiatives for children exposed to domestic violence.

It is difficult to estimate precisely the number of children exposed to domestic violence. In a review of Victorians domestic violence legislation it was revealed that children under five years were present in: 65 per cent of domestic disputes involving the threat or use of a gun; in 79 per cent of disputes involving a weapon (usually a knife); and in almost two-thirds of disputes where property was damaged (Wearing 1992). While in a more recent survey of Australian youth, one quarter of young people sampled reported having 'witnessed' (1) an incident of physical domestic violence against their mother or step-mother (Indermaur 2001) (2). Such research highlights the size of the 'living with domestic violence' problem. Yet it is only quite recently that attention has begun to be paid to this issue.

Most domestic violence research and practice has focused primarily on the incidence of violence against women, with relatively little attention being paid to the plight of children who have been exposed to the violence (Smith, O'Connor & Berthelsen 1996; Sullivan, Bybee & Allen 2002). Similarly, the child protection and family support systems have, until recently, tended to overlook children who have been exposed to domestic violence in the mistaken belief that 'children are untouched by the chaos happening around them in the family home' and a belief that the absence of physical harm meant that no real harm had occurred (Blanchard 1993:31). Thus, children who live with domestic violence have been called the 'silent', 'forgotten', or 'invisible' victims of domestic violence (Osofsky 1998; Edleson 1999).

Throughout the 1990s however, there was increasing evidence of both the widespread nature of domestic violence and the serious impact that being exposed to violence may have on children and young people. Studies in the United States, Canada, the United Kingdom and Australia have produced substantial evidence to suggest that many children who have been exposed to domestic violence develop a variety of social and mental health problems as a result of their experiences (Bookless-Pratz & Mertin 1990; Mathias, Mertin & Murray 1995; Kolbo, Blakely & Engleman 1996; Henning, Leitenberg, Coffey, Bennett & Jankowski 1997; Osofsky 1998; Laumakis, Margolin & John 1998).

Further, while the majority of children who have been exposed to domestic violence do not participate in further family violence, research has identified an association between growing up in a violent family and subsequent involvement in violent adult relationships as either an offender or victim (for example, Edleson & Tolman 1992; Avakame, 1998; Markowitz 2001). Although the extent of the intergenerational transmission of violence attributed to exposure to domestic violence is not known, the best estimates put the extent of the intergenerational transmission of child maltreatment, as a whole, at 30-37 per cent (Kaufman & Zigler 1987; Ryan, Davies & Oates 1977; all cited in Tomison & Poole 2000).

In recognition of the size and impact of the problem, children who have been exposed to domestic violence have recently become the target of research and therapeutic interventions across both the domestic violence and child protection/family support sectors. However, while knowledge about the impact of exposure to domestic violence has been increasing, and programs specifically aimed at supporting the children of battered women have begun to be documented, there is currently a paucity of information available about the nature and effectiveness of such interventions and their outcomes for children (Blanchard 1993; Carter, Weithorn & Behrman 1999, Graham-Bermann 2000).

In order to begin to address the knowledge 'gap' about interventions for children who have been exposed to domestic violence in Australia, data collected from a national audit of child abuse prevention programs (Tomison & Poole 2000) was analysed. The intention is to use the Audit findings as a means of describing this burgeoning area of service delivery. It should be noted however, that the Audit does not provide a detailed analysis of programs as it is based on summaries or overviews of programs collected from the service providers. Thus, rather than a detailed analysis of program operations, the aims is to provide a description of the types of programs in operation in this new area of service provision and to place the identified programs within a broader context of the existing literature.

Australian Audit of Child Abuse Prevention Programs

In 1999, the National Child Protection Clearinghouse with the support of the Australian Council of Children and Parenting and State and Territory child protection departments, undertook an audit of child abuse prevention programs operating in Australia. The aim was to:

   identify trends and any gaps in service provision, to identify
   programs from which service providers can learn, and thus avoid
   'reinventing the wheel' with regard to program development
   (Tomison & Poole 2000: 1).

The target group for the Audit were researchers and practitioners in the government and non-government sectors from across the nation, who were directly involved in child abuse prevention, and/or those assumed to have an interest in child maltreatment and related areas.

Using the information resources of the Clearinghouse, in conjunction with assistance provided by a number of government and non-government agencies, a national master list of agencies thought to be running child abuse prevention programs was created. The sample was further extended by snowball sampling in order to develop as comprehensive a sample of prevention programs as possible.

Program information was collected by way of a questionnaire developed by the Clearinghouse, where respondents were asked to provide a brief description of their program(s) and information about the sections of the population and forms of maltreatment that were being targeted. They were also asked to describe the program in terms of a set of key program descriptors (e.g. home visitation; parent education; alternative care), to provide information about the theoretical or practical basis of the program and describe any program evaluations that had been undertaken (Tomison & Poole 2000).

Criteria for inclusion

The criteria for inclusion in the Audit were first, that any material submitted had to have an explicit child abuse prevention focus. Using the public health model of classification (Tomison and Poole 2000), programs could be classified as operating at the primary ('whole of community' or universal), secondary (targeting at risk sections of the population) or tertiary (preventing the recurrence of maltreatment) level.

Second, the material submitted had to describe a program of activity. Material that did not meet these criteria was kept for use as ancillary information.

Classification of the programs

Once accepted, the programs were categorised using the following classification based on previous Clearinghouse audits (Tomison & Poole 2000). The classifications were developed from typologies commonly used in the literature and the observation and assessment of current trends in prevention activity (Tomison & Poole 2000):

Community Education programs: media campaigns, information materials and training programs for professionals or community groups;

Personal Safety or Protective Behaviours programs: school-based programs that educate children in order to help them identify and protect themselves from possible abuse and/or unsafe situations;

Family Support programs: providing support for 'at risk' families (caregivers and children), and those where maltreatment has already occurred;

Child Focused programs: non-school-based programs focused predominantly on meeting the emotional, behavioural and cognitive needs of children or young people;

Child and Family Centres: multiservice community centres that take a holistic approach to preventing child maltreatment and promoting healthy communities by providing support to children and families on a number of levels; and,

Offender Programs: focused primarily on preventing the recurrence of sexually or physically abusive behaviour, or the development of offending behaviour.

The resultant sample of programs was developed into a Child Abuse Prevention Programs database, which has been made available to the wider community online on the Clearinghouse website (http://www.aifs.org.au/na2.htm).

Overview of the Audit sample

As of December 2000, 1244 separate entries had been included on the Child Abuse Prevention Programs database. However, agency entries often contained information on more than one program that was being run, (for example, agencies could run community education and personal safety programs separately or in combination). For audit purposes (as per previous audits), each of the various programs was classified as a separate program. Under this approach, the 1244 agency entries created a total of 1814 individual programs.

This study

In order to develop a more detailed picture of current initiatives designed to address the needs of children exposed to domestic violence (and their parents), a search was conducted of the Child Abuse Prevention Programs database with the aim of identifying programs that focused specifically on these children. Once these programs were identified, they were classified and analysed on a number of dimensions: program type, prevention type, geographic location, type of evaluation and also by sector. As noted above, only broad overview information was provided about each program and it is not therefore possible to provide a detailed analysis of the program content.

For the purposes of this analysis, services designed primarily to support (and house) women fleeing violent relationships, but where there was also a specific focus on supporting children exposed to domestic violence, were classified as programs from the domestic violence sector. Similarly, there are a range of programs in the child protection and family support fields that have been designed to protect children from harm and to support families to achieve a reasonable standard of health and wellbeing. Operating through a range of community agencies a subset of programs are designed specifically to offer support to children exposed to domestic violence. These programs were classified as operating within the child protection sector.

Of the 1244 agency entries included in the Audit, 36 (3%) organisations were identified as operating 65 individual programs where there was a focus on providing support to children who had been exposed to domestic violence. Clearly programs for children who have been exposed to domestic violence make up a very small proportion of the prevention programs collected during the Audit. Just under half (48%) of the sample of programs were being run by the domestic violence sector, with the remaining programs (52%) operated by the child protection sector.

Geographical distribution 

In the National Audit the geographical distribution of the sample of programs approximated the national population distribution (Australian Bureau of Statistics 2000), with the majority of programs operating across the Eastern seaboard and around the various State and Territory capital cities. The lesser populated States and Territories (Australian Capital Territory, Northern Territory and Tasmania) were over-represented in the Audit, possibly due to smaller and more cohesive professional networks facilitating the distribution of audit material (Tomison & Poole 2000). A similar State-Territory distribution was observed for the sub-sample of programs addressing children who had been exposed to domestic violence. However, while there was a proportional representation of rural and remote programs located across the nation for the Audit as a whole, the 'exposed to domestic violence' programs were over-represented in rural areas.

Program classification

Given that relatively few programs were identified in the Audit as having a focus on providing assistance to children who had been exposed to domestic violence, caution is needed when interpreting the findings of this analysis. From Table 1 it is apparent that over half of the identified programs were family support (34%), or child focused programs (24%). Personal safety and community education programs accounted for 20% and 17% of the programs respectively. The programs were operating in almost equal proportions within both the child protection and domestic violence sectors. The only difference appeared to be a slightly greater emphasis on personal safety and community education programs in the domestic violence sector, with the opposite trend evident for the child protection sector. Child and Family Centres and offender programs were not well represented in either sector.

Type of maltreatment

From service provider reports about the forms of violence being addressed by their programs it was apparent that almost half (48%) of the programs had taken an holistic approach to violence prevention and were designed to prevent all forms of child maltreatment (i.e. physical, sexual, emotional abuse and neglect) as well as domestic violence (see Table 2). A further eight programs were reported to have the aim of addressing all forms of violence or abuse, with the exception of sexual abuse. Fourteen programs were reported to see their primary role as addressing domestic violence (including children's exposure to domestic violence); not surprisingly these were operating in the domestic violence sector.

Level of prevention

As noted above, programs were classified by Clearinghouse staff using the information provided by the programs as targeting primary, secondary or tertiary prevention, under the public health model, and/or as operating as a health promotion program (Tomison & Poole 2000). While classifying programs by prevention level is useful for research and administrative purposes, this classification has been criticised on the grounds that it makes somewhat artificial distinctions between programs that are not reflected in the realities of case management practice (Tomison & Poole 2000).

For example, it has been found that many programs address issues across more than one level of prevention (Tomison & Poole 2000). Thus, programs may be designed for families 'at risk' of harming their child (secondary) and for families where the aim is the prevention of further harm (tertiary level). This point is illustrated in Table 2, where two-thirds of the identified programs were reported to contain elements that impacted across more than one level of prevention; the majority classified as having a secondary and tertiary prevention focus.

This finding is not surprising for a number of reasons. First, most programs run by the domestic violence sector operate in women's refuges where the women have been victims of violence and have fled to safety with their children. Thus, the children had either been directly exposed to violence (tertiary level) or had been (and sometimes were reported to still be) at risk of either witnessing violence or being assaulted (secondary prevention).

Similarly, the majority of programs run by the child protection sector were also classified at the secondary or tertiary levels. Although child protection and family support service providers were more likely to be dealing with 'intact' families of children and their caregivers. A greater number of primary-level prevention programs was attributed to the higher number of community education programs being run in the child protection sector.

Key themes and underlying tenets

Given the recency of the development of programs for children who have been exposed to domestic violence, one useful means of further describing the nature of current interventions and trends in program design is to assess the key themes, concepts or philosophies underlying program design. Overall, seven key tenets or themes were identified (in at least two entries) across the two sectors (through analysis of the 36 agency entries) (see Table 3).

These tenets are consistent with those identified in other studies, such as the Commonwealth Government's Partnerships Against Domestic Violence 'Models of Service Report' (1999) which list the general goals of services providing assistance to children who have been exposed to domestic violence. Further, the overriding aim of most of the domestic violence sector programs appears to be to teach children personal safety skills and to deal with their emotional, behavioural and cognitive needs (therapeutic support). Such tenets have been identified as forming the basis for most interventions or programs for children exposed to domestic violence (Sullivan et al. 2002).

Personal safety

Personal safety programs are the major school-based primary prevention initiatives currently in operation. Depending on the specific program, these programs have the aim of educating school-age children to protect themselves from sexual abuse, child maltreatment as a whole and/or other potentially unsafe situations (Tomison & Poole 2000). For all such programs however, the objective is to educate children, not to make them solely responsible for their own safety (Hummer 1993). The most common and consistent positive outcomes of such programs have been children's heightened knowledge of abuse, increased child disclosures of abuse (Plummer 1993; Reppucci & Haugaard 1993; Michaelson 1997) and, in some programs, an improvement in parents' awareness and ability to protect their children from harm (Hummer 1993).

Overall, it should be noted that the available evidence provides strongest support for an ability to prevent the recurrence of maltreatment (via an increased rate of children's disclosures of abuse) (Hummer 1993; Reppucci & Haugaard 1993; Michaelson 1997); that is, a clear role in tertiary prevention. There is little evidence that the programs can prevent the initiation of abuse, particularly intrafamilial abuse.

In light of the location of most of the programs in this study (women's refuges) it is understandable that children in such situations would be taught personal safety procedures as:

   'Protective behaviours training can be useful in working with
   children who have lived with domestic violence by encouraging
   them to develop a safety plan, and identifying people with whom
   they feel safe' (PADV 1999:1.18).

An example of a program run in a women's refuge with a personal safety element was the 'Child Support-Breaking Free' Program run by Lismore Women's and Children's Refuge in New South Wales:

   'The program aims to identify children's needs and help them
   cope with the experience of living with domestic violence. Child
   Support Workers attempt to build the children's self esteem
   through one-to-one activities, play and protective behaviours
   education' (Child Abuse Prevention Programs database).

Therapeutic support

A number of programs were designed to meet children's emotional and therapeutic needs, and address any ongoing trauma associated with exposure to domestic violence. It was not possible, given the limited evaluations carried out on the programs (see Evidence-based practice) and the available literature, to clearly determine the effectiveness of these programs at meeting children's therapeutic needs. The 'Children's Domestic Violence Group' program run by the Albany Women's Refuge in Western Australia, provides an example of programs that had adopted a therapeutic support focus:

   'This program aims to assist children ages 8-10 who have
   experienced domestic violence. The objectives of the program
   include: to provide a safe place where these children can be
   heard and assisted to heal from the effects of family violence;
   to work at a preventable level to break the cycle of violence
   and to reduce isolation, blame and guilt which may of these
   children carry in connection to the violence. It also aims to
   give the children the right to privacy and forum where they can
   express themselves safely' (Child Abuse Prevention Programs
   database).

Conflict resolution and anger management

With regard to the child protection sector, the most common tenet mentioned was teaching children anger management and conflict resolution skills, which is linked to the aim of 'ending the cycle of violence', by teaching children alternatives to violent behaviour and thus, to reduce the risk of intergenerational transmission of violence.

'Children whose mothers have been victims of domestic violence have often been reported as ... show[ing] signs of anxiety and low self esteem' (Parkinson & Humphreys 1998:150). Not surprisingly, raising children's self esteem was a key element of child protection and family support-type programs. The 'Program for Children from Families where Violence has occurred' run by the Oakrise, Child & Adolescent Mental Health Service in Tasmania aimed to improve self-esteem and teach personal safety skills.

   'This is a ten week program for children aged 9-12 years, at the
   same time the children's group is operating, SURVIVORS will be
   holding a group to support women from relationships where
   violence has occurred. The purpose of the children's group is to:
   empower children to break the secret of family violence so they
   learn they are not the only ones with this experience; to teach
   children that abuse is not ok; to teach children safe and
   nonviolent ways of relating; and to improve children's self
   esteem' (Child Abuse Prevention Programs database).

Educating adults about the impacts of exposure to domestic violence

Another key tenet that was the basis of most of the community education and the (few) offender programs that were in operation, was educating adults about the impact of living with domestic violence on children's development. Many programs that target the perpetrators of domestic violence aim to assist the offender to understand how their violent actions affects others in the family, particularly their children. It has been found that violent fathers either deny or are unaware of the impact of their violence on their children (Mathews 1995). Therefore offender programs aim to focus on 'children's experiences, feelings and behaviours resulting from witnessing violence at home' (Matthews 1995:117). An example of such a program is, the 'Stopping Violence Program' run by the Department for Correctional Service in South Australia:

'This a 12 week group work program for men who have abused their female partner or ex-partners. They are encouraged to take responsibility for the violence; to recognise and stop all forms of violence and understand the impact on children witnessing domestic violence' (Child Abuse Prevention Program Database).

Evidence-based practice

Using the results of this and previous audits of prevention programs, Tomison and Poole (2000) identified a general trend of service providers becoming increasingly aware of the importance of conducting evaluations of their work. This was evident in the growing number of agencies attempting to evaluate their work and the increasing recognition by service providers of the importance of 'adopting a degree of rigour in program evaluation research' (Tomison & Poole 2000:101).

The majority of programs (approximately 60%) addressing children who have been exposed to domestic violence were being evaluated using simple, non-experimental client attendance and satisfaction measures. One quarter of the identified programs had attempted to incorporate pre- and post- test comparisons of the program's impact on participants. However, most programs had stopped short of conducting more rigorous forms of evaluation where control over internal and external variables was exercised, a control or comparison group was employed and follow up assessments were undertaken.

Given that agencies often face some large barriers that can inhibit the conduct of empirical evaluations, this finding is not unexpected. The likelihood that a quality evaluation will be undertaken is reduced by a number of factors, including:

* service providers' fear that external evaluations may produce negative findings and therefore threaten future funding;

* their perception that evaluation diverts meagre resources away from service provision;

* a lack of staff evaluation expertise;

* ethical concerns surrounding issues such as the assignment of children to a control rather than a treatment group; and,

* the long time frame required to undertake a thorough evaluation when most programs only run for a short time (Tomison 2000a).

The evaluation of programs addressing the needs of children exposed to domestic violence

Previous studies have criticised the current standard of evaluations for programs aimed at addressing children's exposure to domestic violence (Tomison 2000a; Graham-Bermann 2000; Sullivan et al. 2002). Although, 34 of the 36 agencies operating programs for children exposed to domestic violence had attempted to evaluate their program(s), the results reflected the wider Audit findings. The level of evaluation activity was similar across both the domestic violence and child protection sectors, and was comprised generally of non-experimental, simple client satisfaction and attendance measures, with few programs appearing to have undergone rigorous evaluation.

It has been suggested that more attention has been placed on the effective evaluation of programs aimed at the deterrence and treatment of the perpetrators of violence than on those programs designed for children exposed to domestic violence (Graham-Bermann 2000). This can possibly be explained by the fact that while programs for offenders have been in existence for some time, the emergence of programs for children has occurred relatively recently. It would be expected that as the number of programs for children who have been exposed to domestic violence grow, so will the quantity and quality of evaluations.

Discussion

This study is one of the first in Australia to identify the types of programs being delivered to children who have been exposed to domestic violence, in what is a new and developing, area of service provision. Although the sample size of programs being studied was too small to allow for substantial conclusions to be drawn, the study does highlight some of the key issues in this area.

In order to ensure children who have been exposed to domestic violence receive the best service response possible, it is necessary to address a number of key issues. These include: the need to increase the availability and funding of services; the development of improved cooperation across the child protection and domestic violence sectors; and the enhancement of the quantity and the quality of research in the field, in particular program evaluation studies.

Resourcing

Clearly, many areas of public welfare would benefit from additional funding. The growing evidence of the numbers of children negatively affected by exposure to domestic violence, combined with the current paucity of programs, as illustrated by the low number of agencies identified running such programs, would appear to provide a good argument for some prioritisation of funding for this area.

As Blanchard (1993) suggests, the majority of Australian counselling and support services that children who have been exposed to domestic violence are often referred to, are attempting to deal with a range of problems with limited resources, and as a result the waiting lists for services are often long. Despite the comparatively good representation of rural programs in this (unrepresentative) sample, accessibility to services is generally hampered in rural and remote areas.

Another Australian study supports this contention. (Smith et al. 1996:8) reported that:

   'many mothers indicated to the interviewers that they were still
   seeking counselling for their children because services were
   limited in this area. As services appear to be poorly funded or
   non-existent, it is clearly an important role for social work to
   support and advocate for services of this nature.'

In addition, the 2001 Victorian Government Review into Victim Services received several submissions suggesting that greater funding needs to be put into services for children who have been exposed to domestic violence. The Domestic Violence and Incest Resource Centre (DVIRC) noted that the situation is 'grave with little free counselling available, [and] children are often left to cope alone'. The long waiting lists for children to receive assistance (e.g. Brenda House Refuge 2001; DVIRC 2001) and the negative consequences of the wait for services were also noted.

   'Children may wait a long time for counselling and they then
   may have to strip away the defences they had built up to heal
   from what they had seen which is really revictimising an already
   traumatised child' (DVIRC 2001:2).

In order to assist children who have been exposed to domestic violence, a primary objective should be to develop and expand programs and services to meet these children's needs. Before it is possible to determine which types of programs should be funded however, greater information is required about the effectiveness and outcomes obtained by the various types of programs. From this study it was evident that even at this early stage in service provision, a range of approaches and program types are operating to address the issue of children's exposure to domestic violence. Family violence is a complex social problem. It is generally acknowledged that no one intervention or program type is able to fully prevent its occurrence. What is required is 'a comprehensive, multi-faceted approach to prevention' (Tomison 2000a:18b), but one based on better evidence of the relative effectiveness of the various service models.

The Australian Government's Partnerships Against Domestic Violence Strategy (PADV 1999) has gone some way towards meeting this need by funding a small number of services and by documenting the methods of program delivery used. Although this Strategy has also produced a number of 'best practice' guidelines for this area of intervention (Gevers 1999), these standards are based on the results of consultation with providers rather than rigorous evaluations of program outcomes for children. While these guidelines are useful in offering some idea of current practice in this area, it could be suggested that they need to be further tested by evaluation.

Holistic violence prevention

It has been suggested (e.g. Stanley & Goddard 1993; Parkinson & Humphreys 1998; Tomison 2000b) that a holistic approach to dealing with domestic violence and child abuse is a necessary step in order to achieve the effective prevention of family violence in its entirety. Historically however, there has been substantial fragmentation of research, policy and practice in the family violence field, with 'domestic violence and child protection [having] been seen as separate issues, with a sharp demarcation line being drawn between them' (Parkinson & Humphreys 1998:147). Partly as a result of this fragmentation, until recently the needs of children exposed to domestic violence have been overlooked by both the child protection system and domestic violence services.

Yet this analysis provides encouraging evidence of the growing recognition of domestic violence as a child protection issue (and vice versa) (Parkinson & Humphreys 1998). As mentioned, the majority of agencies in the Audit (81%) were attempting to address both exposure to domestic violence and at least one form of child abuse. A small number of agencies (7) went further. Rather than merely incorporating children's issues into existing adult programs they had developed a separate child-focused program that ran concurrently with a program designed for the parents. The 'Surviving Together' Women and Children's Group run by the Amaroo Neighbourhood Centre in Victoria provided a good example of this trend:

   '"Surviving Together" is a structured group for women and their
   children who are, or who have been affected by, seeing abuse in
   the home. The group explores ways of dealing with the situation
   as well as strategies for the future. Separate sessions are held
   for mothers and their children between the ages of 7-11 years,
   at the same time and in the same venue. The women's group is
   facilitated by a family counsellor and the children's group by a
   children's worker. The mother's and children's sessions deal with
   the same issues, such as anger, but in different ways. At the end
   of the group women and children come together to discuss what
   they have done, promoting open communication. The course lasts for
   eight to ten weeks' (Child Abuse Prevention Programs
   database).

While the findings indicated that many of domestic violence services were setting up child focused programs, run by workers who specialise in working with children (N=8) there was little evidence of interagency or intersectoral collaboration per se. Interagency contact, particularly where dedicated child focused staff are not available within an agency, provides an opportunity to obtain expert knowledge of children's behaviour, needs and effective therapeutic responses. Conversely, child protection and other family support agencies dealing with cases of domestic violence can benefit from access to expertise with regard to the adult victims/survivors of domestic violence. In order for each sector to better meet the needs of both children and adult caregivers, there is a need to further enhance communication and cooperation between domestic violence agencies and specialised children's services. Those working in the field should realise that:

   'the convergence of their approaches in targeting common risk
   and protective factors means that the results of their programs
   are likely to overlap ... Categorical approaches to prevention
   that focus on single domains of functioning should be expanded
   to more comprehensive programs with multiple goals' (Durlak
   1998, as cited in Tomison and Poole 2000:103)

This is a necessary prerequisite if both the mothers and children are to receive the specific assistance they need.

Future research directions

Despite increasing attempts to address the problem of children being exposed to violence, as the results of this study show, few quality evaluations have been conducted. It is therefore difficult to determine the most effective approaches to supporting and protecting children and preventing the various forms of family violence (Graham-Bermann 2000).

Clearly, if effective interventions are to be more widely employed, and scarce resources better spent, there is a strong need to produce rigorous evaluations of existing programs aiming to assist children exposed to domestic violence. Given the many difficulties agencies face in carrying out evaluations, there is a need to find ways of encouraging and enhancing evaluation research.

One relatively easy step would be to encourage agencies to keep detailed records about their programs as a 'valuable, achievable, first step in evaluating service delivery that may provide a richness of data able to be collected relatively easily' (Tomison & Poole 2000:108). [For a more detailed discussion of the sorts of information that could be collected, see Tomison & Poole 2000.]

It is also important that professional supports are provided to agencies wishing to carry out evaluations. Some universities already have small departmental evaluation units. While the National Child Protection Clearinghouse operates a Program Evaluation Consultancy Service, offering assistance to service providers who want to evaluate child abuse prevention programs they are running or plan to run. Services include assistance in developing evaluation design, evaluation workshops, site visits and access to program evaluation literature.

Examining program impact

At present few studies have attempted to determine how different children react to different types of programs in order to establish those children that are most likely to be assisted by particular program models. Specifically, it is contended that research should be carried out to study the impact of different types of programs on children from different cultural groups, and on the types programs of programs best suited for early intervention and postvention (treatment) roles (Graham-Bermann 2000).

Finally, researchers investigating the risk factors that may heighten children's vulnerability to various social ills, such as child abuse, have consistently identified some children who are able to achieve positive outcomes in the face of adversity--children who are 'resilient' despite facing stressful, high risk situations (Kirby & Fraser 1997). Resilience appears to be determined by the presence of risk factors in combination or interaction with the positive forces (protective factors) that contribute to adaptive outcomes (Garmezy 1985, 1993).

Thus, the nature and extent of the negative impacts of children's exposure to domestic violence may be moderated or mediated by a variety of protective factors including: age of the child; children's coping strategies and individual resiliency; gender; the nature, severity and frequency of the violence; whether the pattern of violence has ceased; attendant environmental factors, such as the mother's ability to parent, and the availability of legal and social protection (Suderman & Jaffe 1997; Edleson 1999).

Research is needed to determine more clearly the factors that lessen the impact of exposure to domestic violence. As Edleson states:

   'the current literature offers only glimpses of children's
   resilience and the factors in their environment that lessen
   or heighten the impact of the violent events swirling around
   them. It is these protective factors-about which we know
   little-that may lead us to design more effective interventions
   to minimize the impact of violence on children' (Edleson
   1999: 865).

The role of schools

Despite a lack of outcome evaluations demonstrating their effectiveness, schools (and secondary schools in particular) are perceived to offer a prime opportunity to intervene early to prevent the development of violent behaviour in intimate relationships (Reiss & Roth 1993; Webster 1993; Suderman & Jaffe 1997). They have often been identified as an important venue for the social development of children, with many engaged in the prevention of a variety of social ills, combined with the promotion of pro-social behaviour in children (i.e. health promotion or life skills education) (Tomison & Wise 1999).

It has also been found that school-based programs 'can reach so many children and youths, reduce misunderstandings about domestic violence, give safety information and planning and offer positive alternatives for conflict resolution and relationship development'(Carter, Weithorn & Behrman 1999:14)

A study which interviewed children who had been exposed to domestic violence found that 'schools were mentioned by all children interviewed.. as being a potentially important source of support following a family crisis' (Blanchard 1993:35).

However the present study found that only one of the 37 agencies identified as running a program for children exposed to domestic violence, was based in the education system. It has been suggested by other writers that it would be beneficial if the more general social development programs currently being run in schools could be extended to include more assistance for children who have been exposed to domestic violence (Blanchard 1993), in part, in order to prevent these children falling into the cycle of violence and becoming perpetrators or victims of domestic violence as adults.

As a precursor to schools playing a greater role in supporting children who have been exposed to domestic violence, it would be beneficial to ensure that all teachers are well informed and trained to be aware of the issues surrounding domestic violence. The prevalent nature of domestic violence in Australia means that every Australian teacher is likely to confront this issue having '... three to five children in each classroom who are contending with violence at home' (Blanchard 1993:35).

It may be unrealistic to expect such programs in isolation, to be enough to alter complex socially derived patterns of behaviour. Well-designed curricula could, however, be useful components of a wider strategy to prevent family violence (Webster 1993). Further, schools have a real opportunity to intervene and support those children who have been affected by their exposure to domestic violence.

In conclusion

The issues for children exposed to violence are serious, with more and more studies showing that a substantial proportion of these children suffer serious and long lasting impacts (Kolbo, Blakely & Engleman 1996, Henning et al 1997, Blumental, Neeman & Murphy 1998, Edleson 1999, Pepler, Catallo & Moore 2000). Unfortunately, the development of research and interventions in this area has been relatively slow. It has only been in recent years that children have been considered as 'victims' of domestic violence. This delay is due, in part, to both the domestic violence services and child protection or family support services failing to acknowledge the harm caused to children who have been exposed to domestic violence and subsequently not addressing the needs of these victims.

As this analysis indicates however, there has been a (relatively recent), positive shift by both sectors to recognise exposure to domestic violence as a key form of child abuse, and to set up programs that attempt to address the issues being faced by child victims/survivors. It seems clear that substantially more research is required to gain a better understanding about this form of abuse and to determine which interventions are most effective in preventing harm and meeting existing survivors' needs. As Morley and Mullender note:

   ... in a field where our ignorance certainly outweighs our
   knowledge, child care practitioners and policy makers--as well
   as children themselves and their mothers--can only benefit
   from well conducted studies which cast light on children's
   experiences and needs (1995:39).

Table 1. Programs targeting children exposed to domestic
violence--by type and sector

                                             Sector

Program type                    Domestic     Child      Total
                                Violence   Protection

Community Education                 5           8         13
Personal Safety or Protective       6           4         10
  Behaviours
Family Support                     11          11         22
Child-Focused                       9           8         17
Child and Family Centres            0           1          1
Offender                            0           2          2
Total                              31          34         65

Table 2. Classification of individual programs by type and sector

                                                      Sector

Maltreatment                             Domestic     Child      Total
                                         Violence   Protection

Physical, emotional abuse and neglect        0           1          1
Physical, emotional, sexual abuse
  & exposure to domestic violence            0           3          3
Physical, emotional abuse, neglect
  and exposure to domestic violence          1           7          8
Emotional abuse and exposure to
  domestic violence                          1           0          1
Physical, sexual abuse and exposure
  to domestic violence                       0           2          2
Physical, emotional abuse and exposure
  to domestic violence                       3           2          5
Exposure to domestic violence               11           3         14
All forms of child maltreatment and
  exposure to domestic violence             15          16         31
Total                                       31          34         65

Table 3. The programs classified by level of prevention and sector

                                               Sector

                                  Domestic     Child
Prevention Level                  Violence   Protection   Total

Primary                               0           5         5
Primary and Secondary                 4           1         5
Secondary                             0           2         2
Secondary and Tertiary               18          12        30
Tertiary                              6          10        16
Primary, Secondary and Tertiary       2           4         6
Health Promotion and Tertiary         1           0         1
Total                                31          34        65

Table 4. Key themes and tenets by sector

                                                  Sector

Themes and Tenets *                  Domestic     Child      Total
                                     Violence   Protection

Teaching children personal safety
  skills                                 9          3          12
Conflict resolution skills/ anger
  management                             3          6           9
Ending the cycle of violence             2          0           2
Reducing children's sense of
  isolation, blame, guilt and
  responsibility                         4          2           6
Emotional & therapeutic support
  (including the enhancement of
  self-esteem)                          10          6          16
Enhancing the mother-child bond          3          0           3
Educating adults as to the impact
  of exposure to domestic violence
  on child development                   2          5           7

* themes not mutually exclusive

Endnotes

(1) 'Witnessed domestic violence' can encompass a range of children's experiences that go beyond merely seeing or hearing violence. Such as being hit or threatened while in her/his mother's arms; being used as a hostage, or as a means of ensuring a mother's return to the home; being used as a physical weapon against a parent; being forced to watch or participate in assaults; being interrogated or involved in spying on their mother; and attempts to break down the parent-child bond (Tomison 2000a). As a result, feminist researchers have criticised the use of the term on the grounds that 'witnessing' implies merely observing violence and therefore fails to reflect the realities of living in a violent family (Irwin & Wilkinson 1997). Thus, the terms 'exposed to domestic violence' or 'living with domestic violence' will be used throughout this article.

(2) There is no universal definition of 'domestic violence'. In this paper the most common Australian definition is used, where domestic violence refers to the violence occurring between persons who are, or who have been, in an intimate relationship. It is generally acknowledged that domestic violence includes physical and sexual violence, threats and intimidation, emotional and social abuse and economic deprivation (Laing 2000).

References

Australian Bureau of Statistics. (2000) Australian Demographic Trends, Cat. No. 3102.0, Australian Bureau of Statistics, Canberra.

Avakame, E. (1998) 'Intergenerational Transmission of Violence, Self-Control, and Conjugal Violence: A comparative Analysis of Physical Violence and Psychological Aggression', Violence and Victims, 13 (3), 301-315.

Blanchard, A. (1993) 'Violence in Families: The Effect on Children', Family Matters, Australian Institute of Family Studies, (34), 31-36.

Bookless-Pratz, C. & Merton, E (1990) 'The behavioural and social functioning of children exposed to domestic violence : a pilot study', Children Australia, 15 (3), 4-7.

Brenda House Inc. (May, 2001) Submission to the Victorian Government Review of Services to Victims of Crime. (http://www.justice.vic.gov.au)

Carter, Weithorn, and Behrman (1999) 'Domestic Violence and Children: Analysis and Recommendations', The Future of Children, 9 (3), 4-20.

Domestic Violence and Incest Resource Centre (June, 2001) Submission to the Victorian Government Review of Services for Victims of Crime (http://www.justice.vic.gov.au)

Edleson, J. (1999) 'Children's Witnessing of Adult Domestic Violence', Journal of Interpersonal Violence, 14 (8), 839-870.

Edleson, J. & Tolman, R. (1992) Interventions for men who batter: An ecological approach, Sage Publications, Newbury Park.

Garmezy, N. (1985), Stress-resistant children: the search for protective factors, in J.E. Stevenson (ed.), Recent Research in Developmental Psychology, Pergamon Press, Oxford.

Garmezy; N. (1993), 'Children in poverty: resilience despite risk', Psychiatry, vol. 56, pp. 127-136.

Gevers, L. (1999) Kids and DV: Practice Standards for Children and Young People Who Have Lived With Domestic Violence, Queensland Department of Families, Australia

Graham-Bermann, S. (2000) 'Evaluating Interventions for Children Exposed to Family Violence', Journal of Aggression, Maltreatment & Trauma, 4 (1), 191-215.

Henning, K., Leitenberg, H., Coffey, P., Bennett, T. & Jankowski, M. (1997) 'Long-term psychological adjustment to witnessing interpersonal physical conflict during childhood', Child Abuse and Neglect, (21) 6, 501-512.

Indermaur, D. (2001) Young Australians and Domestic Violence Trends and Issues, no. 195, Australian Institute of Criminology.

Irwin, J. & Wilkinson, M. (1997), 'Women, children and domestic violence', Women Against Violence, (3) pp. 15-22.

Kirby, L. & Fraser, M. (1997), 'Risk and resilience in childhood', in Fraser, M. (ed.), Risk and Resiliency in Childhood: An Ecological Perspective, NASW Press, Washington, DC.

Kolbo, J., Blakely, E. and Engleman, D. (1996) 'Children Who Witness Domestic Violence: A Review of Empirical Literature', Journal of Interpersonal Violence, 11 (2), 281-293.

Laing, L. (2000) 'Progress, trends and challenges in Australian responses to domestic violence', Issues Paper, Australian Domestic Violence Clearinghouse, Canberra, No. 1.

Laumakis, M., Margolin, G. & John, R. (1998) 'The Emotional, Cognitive and Coping Reponses of Preadolescent Children to Different Dimensions of Marital Conflict', in G. Holden, R. Geffner & E. Jouriles (eds), Children exposed to Marital Violence, American Psychological Association, USA.

Markowitz, F. (2001) 'Attitudes and family violence: linking intergenerational and cultural theories,' Journal of Family Violence, (16) 2, 205-218.

Mathews, D. (1995) 'Parenting Groups for Men Who Batter', in E. Peled, P. Jaffe & J. Edleson (eds), Ending the Cycle of Violence: Community Responses to Children of Battered Women, Sage Publications, USA

Mathias, J., Mertin, P. & Murray, A. (1995) 'The Psychological Functioning of Children from Backgrounds of Domestic Violence', Australian Psychologist, 30 (1), 47-56.

Michaelson, R.C. (1997) 'School-based child sexual abuse prevention: What works? Results of the three year Child Sexual Abuse Prevention Program (CSAPP) evaluation'. Paper presented at the Sixth Australasian Conference on Child Abuse and Neglect, October 1997, Adelaide.

Morley, R. & Mullender, A. (1995) 'Domestic Violence and Children: What do we know from research' in Morley, R and Mullender, A. (eds), Children Living with Domestic Violence: London, Whiting and Birch.

Osofsky, J. (1998) 'Children as invisible Victims of Domestic and Community Violence', in G. Holden, R. Geffner & E. Jouriles (eds), Children exposed to Marital Violence, American Psychological Association, USA.

Partnerships Against Domestic Violence (1999) Models of Service: for Working with Children and Young People who have lived with Domestic Violence, Families, Youth and Community Care, Queensland.

Parkinson, P. & Humphreys, C. (1998) 'Children who witness domestic violence-the implications for child protection', Child and Family Law Quarterly, 10 (2), 147-159.

Pepler, D., Catallo, R. and Moore, T. (2000) 'Consider the Children: Research Informing Interventions for Children Exposed to Domestic Violence', Journal of Aggression, Maltreatment & Trauma, 3 (2), 1-7.

Plummer, C.A. (1993), 'Prevention is appropriate, prevention is successful', in R.J. Gelles and D.R. Loseke (eds), Current Controversies on Family Violence, Sage Publications, Newbury Park, California.

Reiss, A. & Roth, J. (eds), (1993), Understanding and Preventing Violence: Report of the National Research Council Panel on the Understanding and Control of Violent Behavior, National Academy Press, Washington, DC.

Reppucci, N.D. & Haugaard, J.J. (1993), 'Problems with child sexual abuse prevention programs', in R.J. Gelles & D. R. Loseke (eds), Current Controversies on Family Violence, Sage, Newbury Park, CA.

Smith, J., O'Connor, I. & Berthelsen, D. (1996) 'The effects of witnessing domestic violence on young children's adjustment', Australian Social Work, 99 (4), 3-10.

Stanley, J. & Goddard, C. (1993) 'The association between child abuse and other family violence', Australian Social Work, 46 (2), 3-8.

Suderman, M. & Jaffe, P. (1997), 'Children and youth who witness violence: New directions in intervention and prevention', in D.A. Wolfe, R.J. McMahon & R.DeV. Peters (eds), Child Abuse: New directions in prevention and treatment across the lifespan, Banff International Behavioral Science series, Sage, Thousand Oaks.

Sullivan, C.M., Bybee, D.I. & Allen, N.E. (2002) 'Findings from a community-based program for battered women and their children', Journal of Interpersonal Violence, (17), 9, 915-936.

Tomison, A.M. (2000a) 'Evaluating Child Abuse Prevention Programs', Issues in Child Abuse Prevention, National Child Protection Clearinghouse, (12).

Tomison, A.M. (2000b) 'Exploring Family Violence: Links between child maltreatment and domestic violence', Issues in Child Abuse Prevention, National Child Protection Clearinghouse, (13).

Tomison, A.M. and Poole, L. (2000) Preventing child abuse and neglect: findings from an Australian audit of prevention programs, Melbourne, Vic: Australian Institute of Family Studies.

Tomison, A.M. and Wise, S. (1999) 'Community-based approaches in preventing child maltreatment', Issues in Child Abuse Prevention, National Child Protection Clearinghouse, (11).

Wearing, R. (1992) Monitoring the impact of Crime (Family Violence) Act 1987, funded by the Family Law Foundation, Victoria.

Webster, D. (1993) 'The unconvincing case for school based conflict resolution for adolescents', Health Affairs, vol. 12, pp. 126-141.

Katie Kovacs is a Project Officer at the National child Protection Clearinghouse, Australian Institute of Family Studies

Dr. Adam Tomison is an (Acting) Principal Research Fellow at the Australian Institute of Family Studies and the Research Advisor for the National Child Protection Clearinghouse

Correspondence should be sent to: Katie Kovacs, Australian Institute of Family Studies, 300 Queen St, Melbourne, Vic 3000, [e-mail]