Catherine Humphreys Audrey Mullender research in practice Children and Domestic Violence Children and domestic violence: a research overview of the impact on children Catherine Humphreys and Audrey Mullender research in practice w w w . r i p . o r g . u k 2 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN about this pamphlet One of the main ambitions of research in practice is to give local authorities and voluntary organisations better access to reliable research that has been distilled with a particular readership in mind. This series of occasional pamphlets covers key practice areas, identified by practitioners, and key research strategy issues, identified by planners and policy makers. We believe our methods chime well with the developing national agenda for building more effective, more consistent services for children on an increasingly robust platform of research evidence. This pamphlet looks at the much neglected question of the impact of domestic violence on children. Too often, usually out of sympathy for the plight of women at home, childcare workers have turned away from domestic violence and, as a result, women and children alike have suffered continuing abuse and distress. Yet some of the research on intervention and what can be done to promote better outcomes is heartening and there are some surprising findings, including the views of the children themselves. Part of the intention in this series is to stimulate discussion about how research can be used for the benefit of children and their families. The partnership between research in practice and its member agencies supports this process in a number of ways – through meetings, email discussion groups and the ‘learning zone’ areas of our website (www.rip.org.uk). Members are encouraged to make use of the Discussion Groups we have set up for this and also the other pamphlets in the series. Pamphlets are provided free to our members. Others may purchase copies through our website or our Dartington office. The series is an example of a wider range of activities that together support the adoption and implementation of research findings. More information about research in practice can be found on the back of this pamphlet. CHILDREN AND DOMESTI C VIOLENCE contents 5 background 6 defining domestic violence incidence 8 the experiences of children living with domestic violence witnessing abuse links with physical abuse severity of violence other risks abuse during pregnancy links with child sexual abuse flow-on effects 13 the impact on children of living with domestic violence impact on behavioural and emotional well-being impact on cognitive abilities the different effects of age the different effects of ‘race’ and ethnicity the ‘cycle of abuse’ children’s resilience 18 overlap with post-separation violence and child contact continuing to witness or be abused recovery from the effects of violence 19 children’s views about living with domestic violence 21 children as social actors 21 worrying attitudes among young people more generally 22 what helps? raising young people’s awareness direct work with children who have lived with domestic violence work in refuges groupwork one-to-one work 26 conclusion 28 references 3 4 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN CHILDREN AND DOMESTI C VIOLENCE 5 background ‘No one thinks enough of the kids - thinks what effect it has on them. It doesn’t just affect the mother - it’s the kids... Because they’re the ones that have got to see it, and hear it.’ (17 year old girl)1 Strategies for denying or minimising domestic violence are still evident in current social work with children and families. A documentary analysis from the case files of two children and families’ teams showed the following patterns: · failure by professionals to report to child protection conferences known incidents of domestic violence so causing the issue to become lost · inappropriately naming violence (usually by the man towards the woman) as ‘marital conflict’ or ‘arguments’ so that the source of the danger was not identified and its impact was obscured · shifting the focus of assessment to other issues such as mental health or alcohol abuse which, while also present, were not the cause of the danger and indeed, sometimes resulted from it · naming the mother’s behaviour as an equally, or a more serious problem than the man’s violence even when the file documented significant evidence to the contrary · identifying the man as the ‘cornerstone of the family’ rather than challenging his violence towards his partner; and only naming the most extreme forms of physical violence (the ‘atrocity story’) while ignoring other aspects of a pattern of emotional, physical and sexual abuse (Humphreys, 2000). While the study also produced evidence of more sensitive practice, the dominant pattern of minimisation and avoidance can be singled out as a key problem that social work must overcome if practice is to become more effective in this area. This is not a new finding, but one consistently replicated in other studies (Maynard, 1985; Milner, 1993; Farmer and Owen, 1995; Brandon and Lewis, 1996), thus suggesting an issue which is long overdue for attention. This review of the research will address the following issues: defining domestic violence and the gendered pattern of abuse; the experiences of children living with domestic violence and the overlap with other forms of abuse; the impact on children of living with domestic violence; the problems associated with post-separation violence; and policy implications and practice interventions. Sources from UK-based work are used wherever available and are supplemented with research from the USA, Canada and Australia. 1 Except where otherwise stated, children’s words are cited from interview or questionnaires in the authors’ own research studies. 6 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN defining domestic violence ‘He says he loves my Mum but he lies. He tells Mum to do everything at home. He never gave Mum any money. He hit my Mum, I saw it. I tried to look happy but I wasn’t inside. He never played with me - I felt lonely... My Dad hit my sister with a plate and she started bleeding on her head. She was red everywhere... I am scared when I have to see my Dad sometimes, that he will hurt me and shoot me. He said lots of times that he would do that to all of us.’ (6 year old boy, talking to Higgins, 1994) These words of a young Asian boy, resident in a Women’s Aid refuge raise a number of issues which are central to defining what we mean by domestic violence: · a range of abusive behaviours · the effects on children and the overlap with other forms of child abuse · the issue of power and control · escalation over time and the fears and reality of post-separation violence. While terms and definitions will include as many issues as they exclude (see Mullender and Morley, 1994; and Humphreys et al., forthcoming), the definition used for this review includes the issues mentioned above, as well as: · the social collusion in a gendered pattern of violence, with women more likely to be abused · diversity in patterns of abuse · the wide-ranging impact of domestic violence. ‘Domestic Violence typically involves a pattern of physical, sexual and emotional abuse and intimidation which escalates in frequency and severity over time. It can be understood as the misuse of power and exercise of control (Pence and Paymar,1996) by one partner over the other in an intimate relationship, usually by a man over a woman, occasionally by a woman over a man (though without the same pattern of societal collusion) and also occurring amongst same sex couples. It has profound consequences in the lives of individual, families and communities’ (Mullender and Humphreys, 1998) Although women may be violent to men, and there is certainly abuse in some same-sex relationships, historically, statistically and globally the predominant pattern is one of men’s violence towards women (see summaries in Dobash and Dobash, 1992 and Mullender, 1996). At times, research has appeared to challenge this pattern suggesting that women are equally violent towards their partners (Straus et al. 1980; Mirrlees-Black, 1999). A major flaw in many such studies is their use of self-completion questionnaires often based on the Conflict Tactics Scale (CTS). This research tool has been heavily criticized for its emphasis on physical acts taken out of context (Yllo, 1988; Dobash and Dobash, 1992). Hence, there is no discrimination between acts of self-defence or attack, CHILDREN AND DOMESTI C VIOLENCE 7 nor in the level of impact of the violence. Mirrlees-Black explains though, that the initial finding in her research which showed similar rates of men and women being assaulted, does not mean that men are equally victimised. Her closer analysis of the findings indicated that: ‘Men were less upset by their experience, considerably less frightened, less often injured, and less likely to seek medical help.’ (Mirrlees-Black, 1999) When attention is focused upon severe, damaging violence the dominant pattern of concern remains violence towards women by men. Some groups of women face further barriers to help. A study by WAITS (Women Acting in Today’s Society, 1995) of the takeup of domestic violence services by Asian, African Caribbean and Arab women found that approximately 50% of women who had experienced domestic violence waited five years before they sought help, 38% for between five and ten years, and several women waited more than twenty years. Language barriers and often unjustified fears about the type of accommodation they would find in refuges acted as major constraints for black women leaving abusive relationships (Rai and Thiara, 1997). Such obstacles effectively increase the power and control of the abuser in these situations. Greater hurdles must also be overcome by disabled women seeking help. They may be dependent upon their abuser also to be their carer; community care packages and adaptations to the home may not be easily transferable and may have taken years to put in place; only one-quarter of refuges are adapted for some women with disabilities such as wheelchair users. Again, the woman’s difficulties in accessing appropriate help, combined with the losses incurred if she were to separate and the ease with which she can be isolated by her abuser, increase his power and control in the relationship, compounding the impact of abuse. North American research in this area found in a survey of 245 women with disabilities that 40% had experienced abuse (primarily by spouses or ex-spouses) and 12% had been raped. More than half of the women had not sought help, mostly due to fear and dependency (Riddington, 1989). incidence ‘There have been lots of occasions w[h]ere I have sat there and watch[ed] my dad hurt my mum. I’ve seen him strangle her, punch her and generally hurt her.’ (16 year old girl) Physical attacks are what we think of first in considering domestic violence and are also what the law is largely designed to deal with2 . Many children in Britain regularly see or hear such assaults taking place. Research shows consistently that as many as one in three women will experience domestic violence (McGibbon et al., 1988; Mooney, 1994; Dominy 2 The Protection from Harassment Act, 1997 now recognises the impact of threatening behaviour and stalking and is therefore a step forward in acknowledging the wider parameters of abuse. 8 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN and Radford, 1996; Statistics, Canada, 1994). This is exemplified in the random household survey in a North London borough where one in three women admitted having experienced, at some point in their lives, violence in an intimate relationship worse than being grabbed, pushed or shaken, with similar rates across all social and ethnic groupings (Mooney, 1994). When women are asked about events in the year immediately preceding the interview, figures are once more remarkably consistent. Between one in eight and one in ten women have been victims during that period (Mooney, 1994; Stanko et al.1998). These figures clearly indicate that domestic violence is endemic and that it is not just a problem of the poor or of any specific community. Home Office Crime Statistics (year on year) indicate that something approaching two women a week in England and Wales die at the hands of a man with whom they are in, or have been in, an intimate relationship. This accounts for almost half the women who are killed in this country. Strangulation is the most common method used, which makes ‘opportunity’ very easy to find. Violence does not cease when the relationship does. The danger of a woman dying may actually increase upon separation; a third of women killed by male partners are already living apart (Edwards, 1989) and the threat may intensify during attempts to end the relationship (Wilson and Daly, 1992). A third of police calls to incidents of domestic violence come from separated women who are being harassed by ex-partners (Kelly, 1999). The message for practitioners is that when a woman says her abuser has threatened to kill her, or a child has heard this said and discloses it, they must be believed. The only safe way to work with any case of domestic violence is to offer levels of confidentiality and support that assume that this could be, or could become, a ‘life or death’ situation. the experiences of children living with domestic violence ‘Whenever he saw us arguing, he used to be crying his eyes out all the time. He was ‘Mummy, Mummy’. And coming to me all the time.’ (mother of a 2 year old boy) witnessing abuse The majority of children whose mother is being abused are aware of it, often more so than their parents realise. In the NCH Action for Children (Abrahams,1994) research, 87% of the 108 mothers in the study believed that their children had witnessed or overheard violence. Almost all had seen their mothers upset or crying. This echoes the findings of earlier studies. Jaffe et al. (1990), in Canada, found that children knew of incidents their parents thought had been hidden from them. CHILDREN AND DOMESTI C VIOLENCE 9 Increased attention is being given to the effects on children of living with domestic violence. A central concern in this review is to promote an holistic approach, whereby the links between different forms of abuse are made. In particular, the unhelpful divide between violence and abuse towards women is not separated from children’s experiences of this abuse and other forms of child abuse. links with physical abuse ‘I have seen him hit all the members of the family at least once.’ (10 year old boy) There are a number of ways in which children may be directly physically abused or injured when living with domestic violence. They may be accidentally injured because they become caught in the violence which is usually directed at their mothers; they may be the subject of separate incidents of physical abuse by the same man who is also abusive of their mothers; or they may also suffer greater levels of physical punishment or abuse from their over-stressed mothers. incidence Estimates of the overlap between men’s violence towards women and physical abuse of children, gauged largely through North American research, lie in the range of 30 to 66 per cent (summary in Edleson, 1999). In recent years, a similarly wide range has emerged here. At least a third of children on the Child Protection Register in one London Borough were found to have mothers who were known to be being abused (London Borough of Hackney, 1993). An overview of 1,888 child protection referrals across several local authorities found that in 27% of cases domestic violence was an issue in the family (Gibbons, et al.1995). A cluster of factors which significantly affects the prevalence rate of identified domestic violence in cases of known child abuse include: who is looking for it; how knowledgeable they are about domestic violence; and whether active questioning occurs. The separation in both policy and practice of child abuse from woman abuse has meant that the level and effects of domestic violence are consistently under-estimated (Maynard, 1985; Brandon and Lewis, 1996). In two further studies, also drawn from populations of children known to have been abused, the number of cases that came to light increased dramatically once questions about domestic violence were asked. The use of a simple domestic violence monitoring form introduced into the practice of professionals working with children who had been abused increased the numbers of cases in which domestic violence emerged as an issue from one-third to two-thirds of the families assessed at an NSPCC centre (Hester and Pearson 1998). Similarly Farmer and Owen (1995) found that, by the end of their research, they knew of twice as many mothers who were the subject of domestic violence than were known to the participants at the original case conference. They found that 59% of mothers whose children were on the child protection register were 10 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN themselves also the subject of violence; moreover, these were also often the cases with the worst outcomes in the sense that the children were least safe. These studies suggest that active screening for domestic violence is an important practice development. Yet only 40% of social services departments ask these questions at some point in the child protection process, and in only 14% of cases are the results systematically recorded (Humphreys et al., forthcoming). An important consideration is that active screening should only occur alongside active training to ensure that more sensitive, safer intervention occurs. The Department of Healthcommissioned training pack (Barnardos, 1997) and reader (Hester et al., 2000) contribute to increasing standards in this area. However, a recent audit showed that, while 70% of social service departments had attended dissemination workshops on the training pack, to date, only 45% had taken the training forward (Department of Health, 1999). severity of violence At its most extreme, physical abuse will result in death. A study of 30 child death inquiries (James,M. 1994) found a strong association between domestic violence and fatal child abuse. The deaths of Kimberley Carlile, Sukina Hammond and Toni Dales all involved violence towards the child’s mother by her male partner, who then went on to kill the child. The invisibility of domestic violence on social work agendas meant that in the reports of individual enquiries from the 1980s this issue was not recognised until much more recently (O’Hara, 1994). Such a pattern highlights the problems for social workers of confronting dangerous men (Milner, 1993; O’Hagan and Dillenburger, 1995). Further concerns are raised in a USA study by Ross (1996) involving an analysis of questionnaires from 3,363 parents. This study found that there was an almost 100% correlation between the most chronic violent abuse of women by men and those men’s physical abuse of children. The same high correlation did not hold for those women in the study who were defined as most violent towards their male partners. other risks ‘Me and my sister would jump on his back (to stop him hurting Mum), but he would just hit you off… Then you would go flying because he was so big.’ (child cited in Abrahams, 1994) Certain other risks increase the physical abuse of children living with domestic violence. Sometimes they become caught in the cross-fire and are injured in the same incident in which their mother is assaulted. They may be injured when they have tried to intervene to protect their mothers (and vice versa, their mother may be assaulted when she has intervened to protect her children). An early study by Pagelow (1982) in the United States, for example, found that approximately half the children who had both witnessed the domestic violence and been physically assaulted had been injured during a domestic violence incident. CHILDREN AND DOMESTI C VIOLENCE 11 A number of such studies also make the point that, while the patterns of physical abuse of the child are greater from the man abusing the child’s mother, there are heightened risks from their over-stressed mother as well (Straus et al., 1980; O’Keefe, 1995). The link between domestic violence and child abuse is such that, where one is present, questions should always be asked about the other as a means of creating safer, more sensitive assessments and intervention strategies. abuse during pregnancy ‘I remember he kicked my mum in the stomach when she was pregnant.’ (13 year old girl) A serious concern is now also arising in relation to attacks on women by their partners during pregnancy. An overview article of research in this area (Mezey and Bewley 1997) highlights dangers to the woman and to the unborn child. Domestic violence may commence or escalate during pregnancy (Hillard, 1985; Bohn, 1990; Stewart and Cecutti, 1993). The danger does not end with the birth. Indeed, the postpartum is the period of greatest risk of moderate to severe violence (Gielen at al, 1994). Two per cent of women surveyed in GP waiting rooms in Hackney reported a miscarriage they believed to have been caused by a violent partner (Stanko et al. 1998). The attacks on women during pregnancy raise the possibility of a strong link between children’s physical disabilities and domestic violence. Little research has been done in this area. Hague et al. (1996) found that a quarter of refuges in a national survey had a disabled child resident. Such information points to an area urgently needing further exploration. Practically, it also raises issues for professionals about the need for screening for domestic violence in anti-natal and health visiting services and for particular attention to the exploration of the needs of women presenting with physical injuries and/or post-natal depression. links with child sexual abuse ‘She had seen her father pull out a knife on me, so I guess that must have given her a message about how unstable he could be. She would know that she couldn’t tell [anyone].’ (a mother reporting on the link between her own abuse and her child’s sexual abuse in Humphreys, 1991) Although there is less research evidence available, an overlap between domestic violence and child sexual abuse is nevertheless emerging (Goddard and Hiller, 1993). Research from within the statutory child care system has shown a link between domestic violence and elevated levels of child sexual abuse. A case file analysis of 250 children in substitute care who had been sexually abused or sexually abusing found that 39% came from families where there was domestic violence (primarily violence towards the child’s mother). This rate rose to 55% in a more 12 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN detailed follow-up of 40 children (Farmer and Pollock, 1998). The latter figure reflects the incidence of domestic violence in a study by Hester and Pearson (1998) who found that over half the children (n = 111) who had been sexually abused and were attending an NSPCC centre had been living with domestic violence. Smaller qualitative studies of women whose children have been sexually abused similarly reveal a high proportion who had suffered domestic violence (Hooper, 1992; Forman, 1995). These studies highlight the problems that arise if woman abuse is separated from child abuse. Frequently, in cases of child sexual abuse, positive outcomes for children are highly dependent upon the level of support they receive from the non-offending parent (usually mother). Tackling the violence and abuse which the mother has experienced, and the necessity of acknowledging her fear and the issues of safety for her are essential elements in investigation, assessment and on-going work with the family. Similarly, acknowledging the context of violence in which sexual abuse occurred can be important in work with some sexually abused children and their siblings. flow-on effects Other flow-on effects from children’s experiences of domestic violence involve the disruption of their community, family and friendship networks, moving schools, and often losing their home and possessions (Mathias, et al., 1995). Black children and their mothers may find it particularly traumatic to leave communities in which they have found protection from racism and positive support for their cultural and religious life (Mama, 1996; Bhatti-Sinclair , 1994). For Traveller women and their children, the only way to escape from a violent abuser may be to give up their cultural way of life as it is particularly difficult to disappear within the Traveller networks (Van Cleemput et al., undated). For disabled children the disruption of particular schooling and care packages may be extremely difficult to replace. These networks of support and resources which respond to their particular needs may have taken years to put in place and put an added restraint on women leaving an abusive situation. Separation also frequently involves financial hardship which falls disproportionately on women when they leave any relationship (Land, 1999). Violence adds further complications, as women in these situations often believe that fighting for their share of the house or possessions is too risky (Abrahams, 1994). When taken together, the flow-on effects of domestic violence are considerable and often a key restraint to women leaving abusive situations. ‘Staying for the sake of the children’, is not just about children having a father, but about the very real emotional and financial hardship created if and when women decide that it is not safe to stay in their own home. These points highlight the need for practical support and information on welfare and legal rights in social work intervention in situations of CHILDREN AND DOMESTI C VIOLENCE 13 domestic violence. Moreover, actively creating links for women from ethnic minorities with sympathetic women’s organisations will be essential in developing practice in this area. the impact on children of living with domestic violence ‘(My sister) does stupid stuff, like she lights a match, puts it out and burns her hand with it. And she’s got scars on her arms.’ (child quoted in Abrahams 1994) Children cannot help but be affected by their experiences of abuse and violence. Substantial literature now exists which addresses the effects on children. Edleson (1999) identifies 84 studies in this area to date. Fortunately, overview papers and chapters have been written which draw together the key research findings (James, M. 1994; Peled and Davis, 1995, Hester et al. 2000). Children suffer the effects of domestic violence in a number of different ways. They include: a) the impact on their behaviour and emotional wellbeing and b) the effects on their cognitive abilities and attitudes. General findings from research in these areas will be discussed before exploring the particular influences of age and ethnicity. impact on behavioural and emotional well-being Generally, children witnessing domestic violence have significantly more frequent behavioural and emotional problems than children who are not in these abusive environments. A Canadian study of 102 children in shelters (refuges) shows that more than a quarter of the sample (20% girls and 34% boys) had scores on tests which put them in the clinical range of problems for social competence and behaviour problems. This was 2.5 times the rate of a comparison group of children from nonviolent families (Wolfe et al. 1988). Children who have been both physically abused as well as witnessing the violence tend to show the highest levels of behavioural and emotional disturbance (Hughes (1988; Sternberg et al. 1993; O’Keefe, 1994). Individual children react in very different ways to the violence they are witnessing or experiencing. Some children are reported to be more aggressive and anti-social - ‘externalised behaviours’ (Hughes, 1988; O’Keefe, 1995; Maker et al. 1998), others have high rates of depression, anxiety and trauma symptoms – ‘internalised behaviours’ (Sternberg et al. 1993; Mathias et al. 1995; McCloskey et al. 1995). The findings from these studies that compared control groups of children in non-abusive environments are echoed in qualitative studies such as the research on childworkers in refuges which reported children who showed aggression (including tantrums, destructive behaviour, hurting other children) and withdrawn and non-communicative behaviour (Hague et al. 1996). 14 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN impact on cognitive abilities ‘It got so bad I had to go to special classes….I found it so hard to concentrate ….They always thought I was slow.’ (child quoted by Abrahams, 1994) Children’s cognitive abilities are also often affected by their experiences of living with domestic violence. One study found that children who had recently left violent situations and were currently residing in refuges, showed significantly lower levels of competence on a number of parameters, including school performance, than children from a comparison group (Wolfe et al. 1986) , with younger children being particularly affected (Rossman, 1998). It has been suggested that this may be due to the distracting impact of what is going on at home, as well as disrupted schooling from moving to refuges and new addresses if the family is being pursued (Jaffe et al. 1990). An example is provided by a USA study (Montminy-Danna, 1997) which compared 37 children known to be living with violence (but not in a refuge and not thought to have been directly abused themselves) with other children not living with violence. It reported below average school performance in the children from violent homes, a doubled rate of absence from school, behavioural problems observed in class, and 38% of the sample (more than double the expected rate) receiving some form of special educational service. However, the evidence of the effects of domestic violence on children’s academic or cognitive abilities is conflicting. Mathias et al. (1995) found no significant difference between children witnessing violence and other children. Practitioners report that school may be a safe and happier place for children than a difficult home life. Of immediate practical concern is getting children into school when they have to leave home or move refuge; local autonomy of schools has made this more difficult (Hague et al.1996). the different effects of age ‘At first it was, I wouldn’t leave my mam. Wouldn’t leave her anywhere. I was round her all the time. And then , when I was about 14, I used to just stay out all the time. Stayed out as much as I [could]. Anywhere I could, just to get out of the house.’ (17 year old girl) The precise way in which an individual child reacts depends on age, personality, circumstances and so on. Clearly, like the young woman quoted above, reactions may also change over time. Children need to make sense of their experiences and developmental abilities are crucial in determining how this might occur. Infants are only able to indicate their distress through their health or development. Babies under one who witness domestic violence have been characterised with poor health, poor sleeping habits and excessive screaming (Jaffe et al., 1990). They are, of course, entirely dependent on others for their care, yet, as we saw above, the postpartum period is a CHILDREN AND DOMESTI C VIOLENCE 15 particularly high risk period for moderate to severe violence, thus making it difficult for some women to offer the quality of care they would wish. Children of pre-school age tend to blame themselves for adult anger (Jaffe et al., 1990) and to register the highest levels of behavioural disturbance when living with violence (Davis and Carlson, 1987; Hughes, 1988). School-age children often continue to demonstrate behavioural problems. There is less research evidence available about young people: perhaps because teenagers are under-represented in refuges, the source of many research samples (Hester et al. 2000). the different effects of ‘race’ and ethnicity ‘There were problems about our neighbours saying things that were not always nice and not liking my colour.’ (12 year old Asian girl cited in Humphreys, forthcoming) Black authors have been referring for some time to factors which may compound the problems for black children, including the difficulty in seeking help (Mama, 1996) , threats of abduction, language and interpreting problems, and the loss of a community which may protect children from racism(Imam, 1994). More recently, a multi-ethnic study seeking children’s own views has found that black children fear bad advice and unsympathetic treatment from white organisations, a fear which reduces the sources of help open to them (Mullender et al. 2000). Asian children worry about becoming isolated from their communities if they accept such help, a concern which is borne out by the finding, in the same study, that those who had experienced such disruption were less settled and secure than others who had been able to remain within their family and community support structures. The role of the family is significant in other ways, too. Mullender et al. (2000) found that proportionately more black children than white had gone to family members as a first choice of help, but that Asian children sometimes also spoke of the collusion of their father’s extended family with the abuse of their mother, including by female relatives such as a grandmother and an aunt. The practical implications of the effects on children of domestic violence raise a number of issues, including: the need for individualised and age-appropriate assessments; acknowledgement of the trauma which may be associated with witnessing domestic violence; the need for particular vigilance where children are both witnessing and experiencing other direct forms of abuse; and attention to schooling and the need to assess and respond to the educational needs of some children who have been living with domestic violence. the ‘cycle of abuse’ An area which is a continued focus of discussion, and where the issue of the impact of domestic violence on children’s lives is given persistent attention lies in the so called ‘cycle of violence’. This concept proposes 16 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN variations on the notion that ‘violence breeds violence’: either that children growing up with domestic violence are more likely themselves to become abusive adults (Widom, 1989) or that boys will become adult abusers and girls, women who are victimised. Learning theory is used to explain how the pattern is replicated (Straus et al., 1980). The concept and much of the research in this area has been subjected to detailed criticism (Stark and Flitcraft, 1985; Morley and Mullender, 1994). Morley and Mullender (1994) draw attention to a number of issues. First, although many studies suggest a transgenerational pattern, this cannot be an automatic assumption because no study shows 100% of adult perpetrators or victims of domestic violence having had experiences of violence in their childhood, and many report fewer than 50% with this correlation. This means that, in these latter studies, the majority of victims or perpetrators in fact come from non-violent backgrounds. Second, there are many methodological problems with the research in this area. For example, samples are taken from very particular populations of domestic violence offenders, criminals or women in refuges, thus skewing the data or ignoring the number of adults who have grown up living with domestic violence and who go on to live without violence in their adult lives. There are often no control groups of adults who are not living with violence. Studies are often looking for a correlation or an association between particular sets of factors in childhood related to quite specific adult behaviours. This is very different from stating that one causes the other. Numerous intervening variables may also be at play. There are also problems with the definitions employed, both of violence in adults which is often based on the much criticized Conflict Tactics Scale (see Yllo 1988 for a detailed critique of this research tool), and of experiences of childhood abuse where reference may be made to something as vague as ‘disturbed or violent childhoods’ - a phrase which could cover a substantial proportion of any population. Third, if there is any weight in the inter-generational transmission theory at all, it is in relation to men not women. The percentage figures for women as either victims or abusers are consistently lower than 50%. This is exemplified in studies such as that of Pagelow (1981), then replicated by Telch and Lindquist (1984) and O’Leary and Curley (1986), who found no linkage between women as domestic violence perpetrators or victims and those women’s experiences of childhood violence. However, some studies suggest a more persistent pattern for some boys and their subsequent violence towards women in adult relationships. A more detailed exploration of the issue of adult violence may show that ‘the double whammy’ of severe physical abuse as a child and witnessing domestic violence may be a particular risk factor for adult male violence (Spaccarelli et al., 1994), though it needs to be re-iterated that this is not a simplistically causal relationship. Many young boys with these experiences will go on to grow into non-abusive adults. Human beings are not simplistically determined by their upbringing - they can CHILDREN AND DOMESTI C VIOLENCE 17 make choices and avoid repeating negative experiences from the past. Many young people (and their parents) are frightened that living with violence marks them out as inevitably bound for a violent future themselves. It is important to dispel such myths, while also providing positive opportunities for parents to learn to discipline in non-physical ways and for young people to learn at home and at school that violence is wrong and is not appropriate as a means of conflict resolution. children’s resilience ‘(My oldest daughter) had something there and she was a real extrovert… She had this sort of stable network… That helped her through them years…but (the violence) had an opposite effect on (her younger sister)… She didn’t have a lot of friends, and when she does have friends they don’t last long.’ (mother quoted in Abrahams, 1994) Just as it is important to avoid making ‘cycle of violence’ assumptions about children’s futures, so there are no certainties about the way in which any individual child will react to living with domestic violence. While an increased likelihood of problems for children living with domestic violence has been demonstrated (see above), by no means all children are negatively affected. Jaffe et al. (1990), for example, took care to draw attention to the children in their Canadian research who showed few negative symptoms and higher levels of competence than comparison groups. A study by Hughes and Luke (1998) of 58 women and their eldest children living in a women’s refuge found that 60% of the children were either not distressed or only very mildly distressed. It is important to remember these findings when making child protection assessments and also when considering what family support can be offered. Each situation needs to be assessed individually to explore safety strategies with both the women and the children, the effects of the violence in all areas of the child’s life, and the networks which may support the child including at school and in the extended family. Moore and Pepler (1998), in a study of 113 children of mothers in a women’s refuge and a control group of 100 children in two-parent, non-violent families, found that the mothers’ behaviour and mental health played a key role in their children’s adjustment. The conclusion to be drawn is that it is not simply living with violence that automatically causes problems for children, but the kind of life the child lives when viewed in the round. The most recent research (McGee, forthcoming; Mullender et al., forthcoming) also reminds us that children take their own actions and come to their own conclusions about violence and about their family context. Their resilience and coping strategies, and the use they are able to make of supports in their families and communities, also help to explain why some children come through the experience relatively unscathed. This complex interplay of risk and protective factors reinforces the need to shift from blaming women for their ‘failure to protect’, to exploring strengths, the potential to create 18 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN places of safety and support for survivors and challenges to domestic violence offenders. overlap with post-separation violence and child contact ‘I definitely do not want to see him. I would like to see my mum’s brothers and sisters and my Nan [but] I hate his guts. I never want to see him again… our dad found out w[h]ere we had moved to - he threw a brick through the window. And he hit mum in the street and swore... I felt scared and worried. He said he was going to kill our mum.’ (7 year old boy) It is becoming increasingly clear that violence does not cease on separation and that a particular danger for both women and children is the child contact arrangements which arise from separation and divorce. The incidence of post-separation violence (see p8) has clear implications for children. Risks to children arise in a number of areas: continuing either to witness violence or to be the subject of direct abuse; continued flow-on effects of violence; and the interruption of their recovery from violence. continuing to witness or be abused ‘I didn’t mind them phoning him now and again or writing to him, but that was all… On the fourth week he rang and he really upset C., slagging me off, calling me all sorts of names. C. started crying and I told her to put the phone down. She said ‘he is still a nasty, horrible person and I will never speak to him again’…M. (daughter) was the witness when he tried to kill me, and she has never forgotten that, as young as she is – it has stuck in her mind.’(from Humphreys, 2000, forthcoming) The UK research on child contact and domestic violence has pointed to consistent problems that arise in negotiating safe contact arrangements. A comparative, in-depth study undertaken in England and Denmark by Hester and Radford (1996) found that a majority of women interviewed had initially wanted their children to have contact with their father but that, for the majority, the contact arrangements had proved dangerous. Only 7 of the 53 mothers interviewed in England were eventually able to organise contact in such a way that there was no threat to their own safety or their children’s well-being. In the remainder, the man continued to behave in an abusive or intimidatory fashion. Most of the women were re-assaulted after separation, and all of this continuing abuse was linked in some way to child contact (Hester and Radford, 1996). Surveys of refuges (Womens Aid, 1997) and work by AMICA (Aid for Mothers Involved in Contact Action)have also produced worrying results of both women and children being abused via contact arrangements (Radford et al., 1999). While these studies are drawn from particular populations or self-selected samples, which means that their findings cannot be reliably generalised, they do give cause for grave concern. CHILDREN AND DOMESTI C VIOLENCE 19 Certainly, the evidence is sufficient to identify handover as a ‘hot spot’ for repeat violence to women and also a more generalised continuing danger of direct and indirect abuse of children. recovery from the effects of violence Several studies point to the fact that children have the ability to recover from the effects of violence once they are in a safer, more stable environment (Church, 1984; Wolfe et al, 1986; Johnson, 1995; Mertin, 1995). Time really can be a healer in these circumstances. Children who have been most recently exposed to violence tend to show the most marked problems, while those who have moved on and are no longer living with violence may be much less disturbed, both emotionally and behaviourally (Wolfe et al., 1986). The issues of child contact where domestic violence has occurred are some of the most contentious for practitioners. The development of safe child contact is a problem yet to be effectively addressed. Social workers need to be more aware that separation is not ‘a panacea’ for children’s safety where domestic violence and child abuse have been a problem, given that within current private law child contact is denied under only the most extreme circumstances. Assessment and planning will need to tackle these issues much more directly to improve future practice. This includes making recommendations about child contact within child protection conferences, providing resources for high vigilance supervision or, if contact is considered unsafe, ensuring that such recommendations inform court welfare reports and the courts. children’s views about living with domestic violence ‘Grown ups think they should hide it and shouldn’t tell us but we want to know, we want to be involved and we want our mums to talk with us about what they are going to do, we could help make decisions.’ (group interview in Mullender et al., 2000) Whereas earlier research in this field had focused on interviewing adults (mothers and relevant professionals) or conducting psychological tests to ascertain the impact of domestic violence on children, the most recent studies in the UK have asked children and young people themselves about their own experiences and views (McGee, forthcoming; Mullender et al., 2000). From this work, it emerges clearly that children and young people can talk about what they have lived through, and that they are willing to do so provided that the subject is approached sensitively and appropriately. Furthermore, they reveal themselves as active participants in situations of domestic violence and not as passive victims (see below). This means that adults’ attempts to protect children or to offer them help need to build on children’s own understandings and coping strategies if they are to be effective. Mothers who avoid talking to their children in order to protect them 20 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN from the horror of what is going on are likely to be under-estimating both what the children already know and also the extent to which they would appreciate being able to discuss it. Some children even recognise that their mothers are trying to avoid upsetting them (McGee, forthcoming). Only 17% of the mothers in the NCH Action for Children study (Abrahams,1994), and a third in that by Mullender et al. (2000), had talked to their children about the violence, yet all the children in the in-depth phase of the latter study demonstrated awareness of what had been happening. In McGee’s (forthcoming) study, 16 out of the 54 children interviewed had themselves initiated conversations with their mothers about the violence while it was still going on. Communication became easier for many women and children once they were out of the violence, and explanations also became important at that time, although some children avoided re-opening painful memories or discussing current worries, for example about contact with their fathers, as they felt their mothers had enough to cope with in building a new life. Where children could talk to their mothers, this could be very helpful in dealing with unresolved feelings and in making sense of all that had happened. Children do not always or only want to talk to their mothers. Almost all the 54 children interviewed in the Mullender et al. (2000) study reported that it would be useful to talk to someone else they could trust, on their own terms. This included family, friends and neighbours, and sometimes teachers, counsellors or social workers. Grandparents and siblings were important sources of support to many children in McGee’s (forthcoming) study, brothers and sisters mainly for comfort and grandparents sometimes also as people who could intervene or provide a safe place. However, McGee found a minority of families that had been locked into silence over long periods of time. For the children she interviewed, friends were typically a first choice as a source of emotional support, but they were carefully selected as able to keep a secret for reasons of shame, fear and family loyalty. Adults were still considered necessary for practical help. Children want to be listened to, to be taken seriously, to be told what is going on and to be involved in decisions (Mullender et al., 2000). Currently, professionals across the inter-agency range come across to children as lacking sensitivity; children often do not feel they are noticed, believed or offered support in their own right (Mullender et al. 2000). Similarly, McGee (forthcoming) found that children were deterred from disclosing events at home by not expecting to be believed - particularly if the violent man, as a plausible adult, were to deny their account - and they were also frightened of the probable repercussions for themselves and/or their mothers. Children were acutely aware of social stigma and frequently sensed that domestic violence should not be talked about, even when no one at home had actually told them not to do so. CHILDREN AND DOMESTI C VIOLENCE 21 children as social actors Children do not just experience, understand and give meaning to their social circumstances, they participate in them as social actors. Mullender et al. (2000) found a wide range of coping strategies where domestic violence was concerned, which included children keeping themselves and their brothers and sisters away from the danger, staying around or physically intervening to protect their mothers, summoning help, and offering their mothers emotional support. McGee (forthcoming) found children using a similar set of practical coping strategies and also psychological mechanisms such as blocking out the violence, monitoring it closely as a way of feeling some control over the situation, or talking it through, even if only with a toy or a pet. In both Mullender et al.’s and McGee’s work there are hair-raising accounts of children climbing out of bedroom windows, either to reach a sibling or to go and get help. While safety planning as a form of intervention with children needs to build on children’s natural ways of coping, practitioners must also be careful to teach children not to put themselves in greater danger while trying to help others. Overall, safety planning can help children feel less powerless in what is inevitably a terrifying and distressing situation (McGee, forthcoming, and see section on groupwork below). worrying attitudes among young people more generally ‘Some women just need a slap to the jaw and put into the bedroom to calm down.’ (young male focus group member cited in Burton and Kitzinger, 1998) Studies of general populations of children and young people reveal a worrying degree of tolerance of violence against women, suggesting that primary prevention work in schools and elsewhere is an important focus for change. A large survey, involving 2,039 14-21 year olds in Scotland and the North West of England, revealed that almost half the young men and a third of the young women could envisage circumstances in which they thought it would be acceptable for a man to hit his female partner (Burton and Kitzinger, 1998). One in eight young men, for example, considered ‘nagging’ a justification for violence. In addition, as many as one in five young men were tolerant of the idea of a man forcing his wife to have sex. Similar concerns are raised by a recently completed project (Mullender et al., 2000), funded in the Economic and Social Research Council ‘Children 5-16 Programme’, this time asking 8 to 16-year olds (using two, age-related questionnaires) what they thought about domestic violence. A large sample of 1,395 children was obtained through primary and secondary schools in three contrasting areas of England. At all ages, boys were less clear than girls as to who was at fault in incidents where a 22 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN man was violent towards a woman, and more likely to excuse the perpetrator. Boys’ attitudes began to diverge from those of girls at around the age of 13. By 15 to 16 years of age, 78% of girls as against only 56% of boys, saw the man as solely responsible for his own violence in one particular scenario described in the questionnaire. More boys than girls at all ages believed that some women deserve to be hit and over threequarters of 11-12 year old boys thought that women got hit if they made men angry. Thus we have two sets of findings which show particular cause for concern about the attitudes held by boys and young men. what helps? raising young people’s awareness ‘It was very interesting and I learned a lot from it. It gave me ideas which I could tell other people so something could be done in our school.’ (Young Persons’ Zero Tolerance Conference Report, 1996) Young people in research contexts have said that they would like to learn more about domestic violence (Burton and Kitzinger 1998; Mullender et al., 2000). These studies showed that discussion in focus groups was valued and the idea of an educational campaign based around participative discussion encompassing peer and media pressure, parent and teacher attitudes, was welcomed. Educational materials are already available, though they have yet to be evaluated. A teaching pack entitled STOP: Schools Take On Preventing Domestic Violence (London Borough of Islington, 1995), is perhaps the best known. Guidance for Schools, produced in Leeds (Leeds Education and Leeds Inter-Agency Project, not dated) is aimed at raising the awareness of educators. More extensive developments in Ontario, Canada, link work on violence against women into broader learning about mutual respect, including between adults and children, and peer-based conflict resolution skills. In the UK, a comprehensive focus on bullying could potentially be extended into a wider anti-violence curriculum, including also anti-racism, provided that the gendered content on violence against women did not become submerged. Practice experience in Canada has also shown that a holistic approach - involving parents, local media, community arts groups, school governors, and so on in a range of events, and developing relevant teaching materials right across the curriculum - can be enjoyable as well as fruitful (Mullender, 1994b). Any work with pupils needs to be preceded by adequate preparation for teaching staff. Mothers and children in Mullender et al.’s (2000) study reported a low level of awareness in schools about the problems caused by living with domestic violence, with little sympathy for the impact on the ability to learn or on children’s behaviour. It is also essential to have CHILDREN AND DOMESTI C VIOLENCE 23 active links in place with relevant women’s groups (such as Women’s Aid) and with child protection/child welfare organisations, since the work may well lead to disclosures from children currently living with direct and/or indirect abuse. All work in schools needs to emphasise that domestic violence is widespread (each child concerned is likely to think their family is the only one experiencing it) and that violence against women is a crime and no longer socially tolerated. Awareness-raising work in youth settings is another exciting possibility for changing attitudes. A successful youth strategy was, for example, built into a wider Zero Tolerance campaign in Fife (Reid-Howie Associates, 1996). The strategy’s use of peer education theatre was evaluated very positively by participants. Using take-up as another measure of effectiveness, the theatre production, the youth conference and a promotional rock concert, between them, reached thousands of young people in the Fife region. This youth campaign provided further backing for work in schools. A majority (88%) of those responding after watching the play felt that school was the best place to explore the issues. Evaluation forms distributed at the youth conference once again drew a 95% view that Zero Tolerance should form part of the school curriculum. direct work with children who have lived with domestic violence Direct work with children who have lived with, or who are still living with, domestic violence can take place in refuges, in specially established children’s groups, and individually, through counselling or other oneto-one work. It tends to focus broadly on issues of recovery and safety and, where evaluated, has typically emerged very positively. work in refuges We’re all getting away from violence and abuse… That’s why we are all here. It’s good.’ (11 year old girl) A telephone survey of childwork in refuges (Hague et al., 1996) found that 80% of Women’s Aid-affiliated groups in England and 55% of unaffiliated groups had designated children’s workers. Cost was the main obstacle for others. Despite chronic under-resourcing, refuge childwork emerged from the research as a major resource for children. At the time of the survey, 164 refuge groups were housing 2271 children, an average of 14 children each at any one time and an estimated 145 different children each over the year. The children were concentrated at the lower end of the age range, with 85% under 11. Disability awareness and inclusivity for black and minority ethnic children were widely on the agenda, but there can be problems in identifying appropriate services from other agencies. Mainstream work with children revealed by the study covered: play work and provision of play facilities; working with and through mothers to help children; direct work with children through one-to-one work, 24 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN groupwork, workshops and children’s meetings; liaison and advocacy with other agencies; and the development of no violence and child protection policies. Although there were some negatives for children in the lack of space, quiet and privacy, and in restrictions on telling friends where they lived and on bringing them home, refuges lived up to their name both in helping children to feel safe and in providing a time for mother and children to be together in safety, and sometimes to rebuild their relationship (Hague et al., 1996). Children and young people particularly appreciated the ‘no violence’ rule in refuges, both because it made them feel safe and because they consider it illogical to flee violence against women only to inflict it on children. In this respect, too, refuges are an under-recognised national resource, having put considerable effort into working with women to develop non-violent forms of discipline. groupwork ‘Make them feel good about themselves and make them feel it wasn’t their fault.’ (12 year old in Mullender et al., 2000) Refuges excepted, most work aimed at intervening to help children recover from and move on from their experiences of living with domestic violence has been in the form of groupwork, both in the UK and overseas (Mullender, 1994a). It is furthest advanced in Canada and the USA but, increasingly, interesting projects are developing here (Humphreys et al., forthcoming). In London, Ontario, in Canada, there is over a decade’s inter-agency collaboration in this work and every child who needs it can be referred by any of the child welfare, child protection or child health agencies in the city to an appropriate group (Mullender, 1994a). As a result, a rolling programme of groups for 4 to 16-year olds admitted 185 children into 25 groups in 1996 (Loosley et al., 1997). A full-time co-ordinator acts as a single access point and co-runs many of the groups, with all the key agencies involved also contributing a proportion of staff time as groupworkers. As well as obtaining positive feedback from 92% of children and 87% of parents/carers surveyed, evaluation of the groupwork programme has demonstrated effective outcomes (Loosley et al., 1997). The groups can help children feel less responsible for the violence. Learning in the groups includes how to keep safe from both direct and indirect abuse, and how to seek help safely, as well as about non-violent conflict resolution and about abuse and non-violence more generally (Marshall et al., 1995; Loosley et al., 1997). Children’s groups in the USA have also been positively evaluated, both by children who have attended and by their mothers (Peled and Edleson, 1992; 1995). The same study showed that children’s groups are effective in helping children break out of the isolation of keeping a ‘family secret’. CHILDREN AND DOMESTI C VIOLENCE 25 Some individual differences in outcomes, according to the child’s personality, family circumstances, and personal and family history of abuse, would perhaps indicate a need for careful group selection. In particular, children who disclose direct abuse are considered by practitioners as better placed in a group dealing specifically with that issue so that all group members, in both sorts of group, can experience the inclusiveness of having lived through similar experiences. one-to-one work ‘I think there should be a hell of a lot more done, to talk to them... It’s all right [saying]… ‘Everything’s going to be all right’... I think some kids should have counselling... You shouldn’t have that sort of strain on a kid.’ (17 year old girl) An alternative to groupwork is to offer children an individual opportunity to talk about their experiences. This may also be a better option where a child is currently too distressed or disturbed to settle into a group, or where continuing danger and the likelihood of further unplanned moves make commitment to group attendance unrealistic. Most of the children interviewed in the NCH Action for Children (Abrahams,1994) study said they thought it would have been useful to have had someone to talk to, to make sense of the violence and feel less isolated. Few formal psychological or counselling services in the UK are offered by professionals with a knowledge of domestic violence. This is worrying because practitioners who work with children and young people in any specialist setting - be it because they are misusing drugs, committing offences, not attending school, or the like - will be coming across youngsters whose problems may have been triggered by living with violence at home. Typically, there are no routine procedures in place in mainstream children’s services to monitor or assess this issue or its impact (Humphreys et al., forthcoming). Children who have actually received counselling after living with domestic violence have found it useful and recommend it to other children (Mullender et al., 2000). Mothers also consider that children’s counselling services are essential (op. cit., and NCH Action for Children, 1994). Research which mapped services for children across the statutory and voluntary sectors in the UK found that there were few specialist services to cater for the needs of children and their experiences of domestic violence outside Women’s Aid. Only six projects (1.3% of projects surveyed) in four children’s charities surveyed (Barnardos, Children’s Society, NCH Action for Children and NSPCC) provided dedicated domestic violence projects. However, a further 73% (n = 332) gave attention to the needs arising from domestic violence in more general projects (Humphreys et al., 2000). Smaller charities, such as Refuge and Family Services Units, also provide specialist counselling and groups for women and children living with domestic violence. 26 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN However, it is not only in direct service provision that the development of more sensitive practice occurs. A growing body of work explores the context for more responsive and effective intervention where there is domestic violence (Home Office, 2000; Hague et al., 1996). In both adult and children and families’ teams, social work practice which places responsibility on the domestic violence offender and which strengthens the support and safety planning for women and children is being developed. This work is particularly challenging as it requires working jointly across services that have traditionally been separated. Directions for practice lie in the following areas: joint assessments which may involve a person from an adult’s team specialised in disabled people, mental health, alcohol abuse or probation alongside a worker from a children and families’ team; ‘family budgets’ which share responsibility for resources across teams; close supervision of cases where domestic violence is an issue to ensure worker safety and guard against minimisation and collusion with offenders; systematic screening and monitoring for domestic violence in the context of training and increased sensitivity to the impact of domestic violence; the development of safety planning in partnership with the survivors of domestic violence (usually women and children); the development of perpetrator programmes which adhere to the guidelines established by the National Practitioner Network (1998) ; multi-agency working at both an organisational and individual case level; and the evaluation of new initiatives and practice developments (Humphreys and Mullender, forthcoming). conclusion There is a great deal of research on children and domestic violence, an increasing amount of it ‘home grown’ in the UK and benefiting from the broader awareness in British research that children are not silent victims but social actors who have their own perceptions and understandings of events. A deterministic belief in a ‘cycle of violence’ has, in the past, made professionals worry about children who are living with domestic violence as a ‘bad bet’ when they become the partners and parents of the future. The message from research, however, is that the experience of living with domestic violence, though distressing and not without its adverse effects for many children, can also give some children and young people (at least in the short to medium term) an insight into adult relationships which other young people do not share. Surveys of general populations of young people reveal a degree of tolerance towards violence against women, especially amongst boys and young men, which suggests that work in schools is needed as a preventative measure. For those who have lived with domestic violence, individual and groupwork that focuses on practical safety planning and emotional recovery can be of enormous value but is not yet CHILDREN AND DOMESTI C VIOLENCE 27 comprehensively available. Children’s workers in refuges are a national resource that could be built on with relatively modest additional funding and the major national children’s charities are also developing important projects. Inter-agency work through domestic violence forums, Area Child Protection Committees and the like is essential in order to ensure well targeted and effective efforts, but equally important is partnership with women and children so that we never forget whose issue this is and who are the experts on what domestic violence actually means at the personal level. 28 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN references Abrahams, C. (1994) Hidden Victims: children and domestic violence, London, NCH Action for Children. Barnados, NSPCC and University of Bristol (1998) Making an Impact: children and domestic violence, London: Barnardos in association with Department of Health. Bhatti-Sinclair, K. (1994) ‘Asian women and violence from male partners’ in Lupton, C. and Gillespie, T. (eds) Working with Violence, Basingstoke, Macmillan. Bohn, D. (1990) ‘Domestic violence and pregnancy: implications for practice’ Journal of Nurse Midwifery 35, 19-21. Brandon, M. and Lewis, A. (1996) ‘Significant harm and children’s experiences of domestic violence’, Child and Family Social Work, 1, 3342. Burton, S. and Kitzinger, J. with Kelly, L. and Regan, L. (1998) Young People’s Attitudes towards Violence, Sex and Relationships. A Survey and Focus Group Study, Edinburgh, The Zero Tolerance Charitable Trust. Research Report 002. Church, J. (1984) Violence against Wives: its causes and effects Christchurch, New Zealand. John Church (Publisher) Davis, L. and Carlson, B. (1987) ‘Observation of spouse abuse: what happens to the children?’, Journal of Interpersonal Violence 2, 278-291. Department of Health (1999) The Family Law Act 1996 Domestic Violence Survey Report on Local Authority Implementation of Part IV, London, Department of Health. Dominy, N. and Radford, L. (1996) Domestic Violence Intervention in Surrey: towards an effective inter-agency strategy response, London, Roehampton Institute. Dobash, R. and Dobash, R. (1992) Women, Violence and Social Change, London, Routledge. Edleson, J. (1999) ‘Children’s witnessing of adult domestic violence’, Journal of Interpersonal Violence, 14, 839-870. Edwards, S. (1989) Policing Domestic Violence London, Sage. Farmer, E. and Owen, M. (1995) Child Protection Practice: Private Risks and Public Remedies, London, HMSO. Farmer, E. and Pollock, S. (1998) Substitute care for sexually abused and abusing children, Chichester, John Wiley and Sons. Forman, J. (1995) Is There a Correlation Between Child Sexual Abuse and Domestic Violence? An exploratory study of the links between child sexual abuse and domestic violence in a sample of intrafamilial child sexual abuse cases, Glasgow, Women’s Support Project. CHILDREN AND DOMESTI C VIOLENCE 29 Gibbons, J., Conroy, S. and Bell, C. (1995) Operating the Child Protection System: a study of child protection practice in English local authorities, London, HMSO. Gielen, A., O’Campo, P., Faden, R., Kass, N. and Xue, X. (1994) ‘Interpersonal conflict and physical violence during the childbearing year’, Social Science and Medicine, 39, 781-787. Goddard, C. and Hiller, P. (1993) ‘Child sexual abuse: assault in a violent context’, Australian Journal of Social Issues, 28, 20-33. Hague, G., Kelly, L., Malos, E. and Mullender, A. with Debbonaire, T. (1996) Children, Domestic Violence and Refuges: a study of needs and responses, Bristol Women’s Aid Federation (England). Hester, M., Pearson, C. and Harwin, N. (2000) Making an Impact: a reader, London, Jessica Kingsley. Hester, M. and Pearson, C. (1998) From Periphery to Centre: domestic violence in work with abused children,Bristol, Policy Press. Hester, M. and Radford, L. (1996) Domestic Violence and Child Contact Arrangements in England and Denmark, Bristol, Policy Press. Higgins, G. (1994) ‘Hammersmith Women’s Aid childhood development project,’ in A. Mullender and R. Morley (eds) Children Living with Domestic Violence, London, Whiting and Birch. Hillard, P. (1985) ‘Physical abuse in pregnancy’, Journal of Obstetrics and Gynaecology, 66, 185-190. Hooper, C.-A. (1992) Mothers Surviving Child Sexual Abuse, London, Routledge. Hughes, H. (1988) Psychological and behavioral correlates of family violence in child witnesses and victims’, American Journal of Orthopsychiatry, 58, 77-90. Hughes, M. and Luke, D. (1998) ‘Heterogeneity in adjustment among children of battered women’ in Holden, G., Geffner, R. and Jouriles, E. (eds) Children Exposed to Marital Violence, Washington DC., American Psychological Association. Humphreys, C. (1991) Child Sexual Assault Disclosure: mothers in crisis, PhD dissertation held at the University of N.S.W., Sydney, Australia. Humphreys, C. (2000) Social Work, Domestic Violence and Child Protection: challenging practice, Bristol, Policy Press. Humphreys, C., Hester, M., Hague, G., Mullender, A. Abrahams, H. and Lowe, P. (2000 forthcoming) From Good Intentions to Good Practice: mapping services working with families where there is domestic violence, Bristol, Policy Press. Humphreys, C. and Mullender, A. (forthcoming) ‘Working with Domestic Violence’, in Horwath, J. and Shardlow, S. (eds) Making Links: assessment and roles across social work specialisms, Lyme Regis, Russell House Publishing. 30 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN Imam, U. (1994) ‘Asian children and domestic violence’ in Mullender, A. and Morley, R. (eds) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. Jaffe, P., Wolfe, D. and Wilson, S. (1990) Children of Battered Women, Newbury Park, CA, Sage. James, M. (1994) Domestic Violence as a Form of Child Abuse: identification and prevention, Canberra, National Child Protection Clearing House. Johnston, J. (1995) ‘Domestic violence and parent-child relationships in families disputing custody’, Australian Journal of Family Law (Special Issue) 12-25 Kelly, L. (1999) Domestic Violence Matters: an evaluation of a development project, Home Office Study 188, London, The Stationery Office. Land, H. (1994) ‘The demise of the male bread winner - in practice but not in theory: a challenge for social security systems’, in Baldwin, S. and Falkingham, J. (eds) Social Security and Social Change: new challenges to the Beveridge model, Hemel Hempstead, Harvester Wheatsheaf. Leeds Education and Leeds Inter-Agency Project, Women and Violence (not dated) Guidance for Schools: violence against women and children by known men, Leeds City Council, Department of Education. Loosley, S., Bentley, L., Lehmann, P., Marshall, L., Rabenstein, S. and Sudermann, M. (1997) Group Treatment for Children Who Witness Woman Abuse: a manual for practitioners, London, Ontario, The Community Group Treatment Program. (Available from: The Children’s Aid Society of London and Middlesex, P.O. Box 7010, Depot 1, London, Ontario, Canada). London Borough of Hackney (1993) The Links Between Domestic Violence and Child Abuse: developing services, London, Hackney Council Press and Publicity Team. London Borough of Islington (1995) STOP: Schools Take On Preventing Domestic Violence, London: London Borough of Islington, Women’s Equality Unit. Maker, A., Kemmelmeier, M. and Peterson, C. (1998) ‘Long term psychological consequences in women of witnessing parental physical conflict and experiencing abuse in childhood,’ Journal of Interpersonal Violence, 13, 574-589. Mama, A. (1996) The Hidden Struggle: statutory and voluntary sector responses to violence against Black women in the home, London, Whiting and Birch. Marshall, L., Miller, N., Miller-Hewitt, S., Sudermann, M., Watson, L. (1995) Evaluation of Groups for Children who have Witnessed Violence, London, Ontario, Centre for Research on Violence Against Women and Children. CHILDREN AND DOMESTI C VIOLENCE 31 Mathias, J., Mertin, P., and Murray, A. (1995) ‘The psychological functioning of children from backgrounds of domestic violence’, Australian Psychologist, 30, 47-56. Maynard, M. (1985) ‘The response of social workers to domestic violence’ in J. Pahl (ed) Private Violence and Public Policy, London, Routledge. McClosky, L., Figueredo, A. and Koss, P. (1995) ‘The effects of systemic family violence on children’s mental health’, Child Development, 66, 1239-1261. Mertin, P. (1995) ‘A follow-up study of children from domestic violence’ in Australian Journal of Family Law, 9, 76-85. Mezey, G. and Bewley, S. (1997) ‘Domestic violence and pregnancy’, British Journal of Obstetrics and Gynaecology, 104, 528-531. McGibbon, A., and Kelly, L. (1989) Abuse in the Home: advice and information, London Borough of Hammersmith and Fulham. McGee, C. (forthcoming) Childhood Experiences of Domestic Violence, London, Jessica Kingsley. Milner, J. (1993) ‘A disappearing act: the differing career paths of fathers and mothers in child protection investigations’, Critical Social Policy, 13, 48-61. Mirrlees-Black, C. (1999) British Crime Survey 1996: domestic violence findings, London, Home Office. Montminy-Danna, M. (1997) A comparative study of school performance of children who witness family violence and their non-reporting agemates, MSW, unpublished. Mooney, J. (1994) The Hidden Figure: domestic violence in North London, London, Islington Police and Crime Prevention Unit. Moore, T. and Pepler, D. (1998) ‘Correlates of adjustment in children at risk’, in Holden, G., Geffner, R. and Jouriles, E. (eds) Children Exposed to Marital Violence, Washington DC, American Psychological Association. Morley, R. and Mullender, A. (1994) ‘What do we know from research?’ in Mullender, A. and Morley, R. (eds) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. Mullender, A. (1994a) ‘Groups for child witnesses of woman abuse: learning from North America’, in Mullender, A. and Morley, R. (eds) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. Mullender, A. (1994b) ‘School-based work: education for prevention’, in Mullender, A. and Morley, R. (eds) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. 32 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN Mullender, A. (1996) Re-thinking Domestic Violence: the social work and probation response, London, Routledge. Mullender, A. and Morley, R. (1994) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. Mullender, A. and Humphreys, C. (1998) Domestic Violence and Child Abuse: policy and practice issues for local authorities and other agencies, London, Local Government Association. Mullender, A., Kelly, L., Hague, G., Malos, E. and Imam, U. (2000) Children’s Needs, Coping Strategies and Understandings of Woman Abuse. End of award report submitted to the ESRC Award No: L129 251037, Coventry, University of Warwick, Department of Social Policy and Social Work. Mullender, A., Hague, G., Imam, U., Kelly, L. and Malos, E. (forthcoming) ‘It Makes Me Sad’: children’s perspectives on domestic violence, London, Sage. National Practitioners’ Network (1994, with revisions in 1997 and 1998) Statement of Principles and Guidelines for Good Practice for Intervention Programmes Working with Men towards Ending their Violence and Abuse to Women Partners: The National Practitioners’ Network Perspective, held at DVIP, London, on behalf of NPN. O’Hagan, K. and Dillenburger, K. (1995) The Abuse of Women Within Childcare Work, Open University Press. O’Hara, M. (1994) ‘Child deaths in the context of domestic violence: implications for professional practice’, in Mullender, A. and Morley, R. (eds) Children Living with Domestic Violence: putting men’s abuse of women on the child care agenda, London, Whiting and Birch. O’Keefe, M. (1995) ‘Predictors of child abuse in maritally violent families’, Journal of Interpersonal Violence, 10, 3-25. O’Keefe, M. (1994) ‘Linking marital violence, mother-child/father child aggression, and child behavior problems’, Journal of Family Violence, 9, 63-78. O’Leary, K. and Curley, A. (1986) ‘Assertion and family violence: correlates of spouse abuse’, Journal of Marital and Family Therapy, 12, 281-289. Pagelow, M. (1982) ‘Children in violent families: direct and indirect victims’, in Hill, S. and Barnes, B. (eds) Young Children and their Families, Lexington, Lexington Books. Peled, E. and Edleson, J. L. (1992) ‘Multiple perspectives on group work with children of battered women’, Violence and Victims, 7, 327346. Peled, E. and Edleson, J. (1995) ‘Process and outcome in small groups for children of battered women’, Ending the Cycle of Violence: Community Responses to Children of Battered Women, Thousand Oaks, CA, Sage. CHILDREN AND DOMESTI C VIOLENCE 33 Peled, E. and Davis, D. (1995) Groupwork with children of battered women: a practitioner’s manual, Thousand Oaks, CA, Sage. Pence, E. and Paymar, M. (1996) Education Groups for Men Who Batter: the Duluth model, (2nd edition), New York, Springer. Radcliffe, L., Sayer, S. and AMICA (1999) Unreasonable Fears? Child Contact in the Context of Domestic Violence, Bristol, Women’s Aid Federation of England. Rai, D. and Thiara, R. (1997) Re-defining Spaces: the needs of Black women and children in refuge support services and Black workers in Women’s Aid, Bristol, Women’s Aid. Reid-Howie Associates (1996) The Zero Tolerance Campaign in Fife: post campaign research, September-November 1996, Edinburgh, Reid-Howie Associates. Riddington, J. (1989) Beating the ‘Odds’: violence and women with disabilities (Position Paper 2), Vancouver, DisAbled Women’s Network, Canada. Ross, S. (1996) ‘Risk of physical abuse to children of spouse abusing parents’, Child Abuse and Neglect, 20, (7) 589-598. Rossman, B. (1998) ‘Descartes’ error and post-traumatic stress disorder: cognition and emotion in children who are exposed to parental violence’, in Holden, W., Geffner, R. and Jouriles, E. (eds) Children Exposed to Marital Violence, Washington DC, American Psychological Association. Spaccarelli, S., Sandler, I. and Roosa, M. (1994) ‘History of spouse violence against other: correlated risks and unique effects in child mental health’, Journal of Family Violence, 9, 79-98. Stanko, E., Crisp, D., Hale, C. and Lucraft, H. (1998) Counting the Costs: estimating the impact of domestic violence in the London Borough of Hackney, London, Crime Concern. Stark, E. and Flitcraft, A. (1985) ‘Women-battering, child abuse and social heredity, what is the relationship?, in Johnson, N. (ed) Marital Violence, London, Routledge and Kegan Paul. Statistics Canada (1994) Family Violence in Canada, Ottawa, Canadian Centre for Legal Statistics. Sternberg, K., Lamb, M., Greenbaum, C., Dawud. S., Cortes, R., Krispin, O., and Lorey. F. (1993) ‘Effects of domestic violence on children’s behavior problems and depression’, Developmental Psychology, 29, 44-52. Stewart, D. and Cecutti, A. (1993) ‘Physical abuse in pregnancy’, Canadian Medical Association Journal, 149, 1257-1263. Straus, M., Gelles, R. and Steinmetz, S. (1980) Behind Closed Doors: violence in the American family, California, Sage. 34 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN Taylor-Browne, J. (ed) (forthcoming) Reducing Domestic Violence: what works? London, Home Office. Telch, C. and Lindquist, C. (1984) ‘Violent versus non-violent couples: a comparison of patterns’, Psychotherapy, 21, 242-248. Van Cleemput, P., Storey, M. and Traveller Women Living in Sheffield (undated) ‘Supporting Gypsy and Traveller Women and Children Living with Domestic Violence’, Community Health, Sheffield (leaflet). Weinehall, K. (1997) ‘To grow up in the vicinity of violence: young people’s stories about domestic violence’, Akademiska avhandlingar vid Pedagogiska institutionen, Umea universitet, 45, 328-342. Women Acting in Today’s Society (1995) Freedom From Abuse: domestic violence in Asian, African and Arab communities, Birmingham, WAITS. Yllo, K. ‘Political and methodological debates in wife abuse research’ in Yllo, K. and Bograd, M. (eds) Feminist Perspectives on Wife Abuse, Newbury Park, Sage Publications. Widom, C. (1989) The Intergenerational Transmission of Violence, New York, Guggenheim Foundation. Wilson, M. and Daly, M. (1992) Homicide, New York, Aldine de Gruyter. Wolfe, D., Zak, L., Wilson, S. and Jaffe, P. (1988) ‘Child witnesses to violence between parents: critical issues in behavioural and social adjustment’, Journal of Abnormal Child Psychology, 14, 95-104. Women’s Aid (1997) Women’s Aid Federation Briefing Paper on Child Contact and Domestic Violence, Bristol, Women’s Aid Federation. Young Persons’ Zero Tolerance Conference Report (1996) Voices of the Future, Auchterderran, Fife: Zero Tolerance Campaign. CHILDREN AND DOMESTI C VIOLENCE 35 about the authors Audrey Mullender is Professor of Social Work at the University of Warwick. She served for four years, until 1999, as Editor of The British Journal of Social Work and is the author of over 100 publications in the social work field, including ten books. These reflect her three main research interests: children living with domestic violence, birth relatives of adult adopted people and groupwork as a means of empowering the users of social work services. Catherine Humphreys is a senior lecturer in the Department of Social Policy and Social Work at the University of Warwick. She worked for 15 years as a social work practitioner in child care and mental health in Australia before coming to England in 1994. Her research, writing and practice interests have been in the areas of child sexual abuse and domestic violence. She has been an active member of her local women’s aid group for the past six years. 36 A RESEARCH REVIEW OF THE IMPACT ON CHILDREN