Jeffrey L. Edleson
Revised April 1999

Children who witness violence between adults in their homes have become more visible in the spotlight of public attention. The purpose of this document is to further an understanding of the current literature on the effects of witnessing adult domestic violence on the social and physical development of children. Out of 84 studies reporting on children's witnessing of domestic violence originally identified, 31 studies met criteria of rigorous research (see Edleson, 1999), with 18 of them comparing children who witnessed adult domestic violence to other groups of children, 12 others using multiple regression procedures to compare subjects along a continuum of violence exposure or by demographic characteristics, and one study applying qualitative research methods. The findings of these 31 studies can be divided into three major themes: (1) the childhood problems associated with witnessing domestic violence; (2) the moderating factors present in a child's life that appear to increase or decrease these problems; and (3) an evaluation of the research methods used in the studies reviewed.

Children's Problems Associated with Witnessing Violence

Reviewed studies report a series of childhood problems statistically associated with a child's witnessing domestic violence. These problems can be grouped into the three main categories presented in more detail below: (1) behavioral and emotional; (2) cognitive functioning and attitudes; and (3) longer-term.

Behavioral and emotional problems
The area in which there is probably the greatest amount of information on problems associated with witnessing violence is in the area of children's behavioral and emotional functioning.Generally, studies using the Child Behavior Checklist (CBCL; Achenbach & Edelbrock, 1983) and similar measures have found child witnesses of domestic violence to exhibit more aggressive and antisocial (often called "externalized" behaviors) as well as fearful and inhibited behaviors ("internalized" behaviors), and to show lower social competence than other children. Children who witnessed violence were also found to show more anxiety, self-esteem, depression, anger, and temperament problems than children who did not witness violence at home. Children from homes where their mothers were being abused have shown less skill in understanding how others feel and examining situations from others' perspectives when compared to children from non-violent households. Peer relationships, autonomy, self-control, and overall competence were also reported significantly lower among boys who had experienced serious physical violence and been exposed to the use of weapons between adults living in their homes.

Overall, these studies indicate a consistent finding that child witnesses of domestic violence exhibit a host of behavioral and emotional problems. A few studies have reported finding no differences on some of these measures but these same studies found significant differences on other measures.

Another aspect of the effects on children is their own use of violence. Social learning theory would suggest that children who witness violence may also learn to use it. Several researchers have attempted to look at this link between exposure to violence and subsequent use of it. Some support for this hypothesis has been found. For example, Singer et al. (1998) studied 2,245 children and teenagers and found that recent exposure to violence in the home was a significant factor in predicting a child's violent behavior.

Cognitive functioning and attitudes
A number of studies have measured the association between cognitive development problems and witnessing domestic violence. While academic abilities were not found to differ between witnesses and other children (Mathias et al., 1995), another study found increased violence exposure associated with lower cognitive functioning (Rossman, 1998). One of the most direct consequences of witnessing violence may be the attitudes a child develops concerning the use of violence and conflict resolution. Jaffe, Wilson and Wolfe (1986) suggest that children's exposure to adult domestic violence may generate attitudes justifying their own use of violence. Spaccarelli, Coatsworth and Bowden's (1995) findings support this association by showing that adolescent boys incarcerated for violent crimes who had been exposed to family violence believed more than others that "acting aggressively enhances one's reputation or self-image" (p. 173). Believing that aggression would enhance their self-image significantly predicted violent offending. Boys and girls appear to differ in what they learn from these experiences. Carlson (1991) found that boys who witnessed domestic abuse were significantly more likely to approve of violence than were girls who had also witnessed it.

Longer-term problems
Most studies reviewed above have examined child problems associated with recent witnessing of domestic violence. A number of studies have mentioned much longer-term problems reported retrospectively by adults or indicated in archival records. For example, Silvern et al.'s (1995) study of 550 undergraduate students found that witnessing violence as a child was associated with adult reports of depression, trauma-related symptoms and low self-esteem among women and trauma-related symptoms alone among men. Witnessing violence appeared to be in dependent of the variance accounted for by the existence of parental alcohol abuse and divorce. In the same vein, Henning et al. (1996) found that among 123 adult women who had witnessed domestic violence as a child greater distress and lower social adjustment existed when compared to 494 non-witnesses. These findings persisted even after accounting for the effects of witnessing parental verbal conflict, being abused as a child, and level of reported parental caring.

Factors Influencing the Degree of Problems Associated with Witnessing Violence

Several factors appear to moderate the degree to which a child is affected by witnessing violence. As will be seen below, a number of these factors also seem to interact with each other creating unique outcomes for different children.

Abused and witnessing children
Hughes, Parkinson and Vargo (1989) have suggested that both witnessing abuse and also being abused is a "double whammy" for children. Their study compared children who were both abused and had witnessed violence to children who had only witnessed violence and to others who had been exposed to neither type of violence. They found that children who were both abused and witnesses exhibited the most problem behaviors, the witness-only group showed moderate problem symptoms and the comparison group the least. This same pattern appears in series of other studies. Children seem to agree. In one study they indicated that the experience of being abused or both abused and a witness is more negative than witnessing adult domestic violence alone (McClosky, Figueredo & Koss, 1995).

The combination of being abused and witnessing violence appears to be associated with more serious problems for children than witnessing violence alone. Silvern, et al. (1995) found, however, that after accounting for the effects of being abused, adult reports of their childhood witnessing of interparental violence still accounted for a significant degree of their problems as children. Silvern and her colleagues caution that witnessing domestic violence may result in traumatic effects on children that are distinct from the effects of child abuse.

Child characteristics
Some findings point to different factors for boys and girls that are associated with witnessing violence. In general, boys have been shown to exhibit more frequent problems and ones that are categorized as external, such as hostility and aggression, while girls generally show evidence of more internalized problems, such as depression and somatic complaints (Carlson, 1991; Stagg, Wills & Howell, 1989). There are also findings that dissent from this general trend by showing that girls, especially as they get older, also exhibit more aggressive behaviors (for example, Spaccarelli, et al., 1994).

Children of different ages also appear to exhibit differing responses associated with witnessing violence. Children in preschool were reported by mothers to exhibit more problems than other age groups (Hughes, 1988).

Few studies have found differences based on race and ethnicity. O'Keefe's (1994) study of white, Latino, and African-American families of battered women found that all the children were viewed by their mothers as having serious emotional and behavioral problems. The only difference found between the groups was on social competence; African-American mothers rated their children more competent when compared to other mothers' ratings of their own children.

Time since violent event
The longer the period of time since exposure to a violent event the fewer effects a child experiences. For example, Wolfe, Zak, Wilson and Jaffe (1986) found more social problems among children residing in shelters than among children who had at one time in the past been resident in a shelter. The effect of the immediate turmoil may temporarily escalate child problems as observed in a shelter setting.

Parent-child relationship factors
A number of authors have discussed a child's relationship to adult males in the home as a key factor. Peled (1996) suggests that children's relationships with their battering fathers were confusing, with children expressing both affection for their fathers and resentment, pain and disappointment over his violent behavior.

Children's relationships to their mothers have also been identified as a key factor in how children are affected by witnessing domestic violence. Some have conjectured that a mother's mental health would negatively affect a child's experience of violence but the data are conflicting. Wolfe, Jaffe, Wilson and Zak (1985) found that maternal stress statistically accounted for a large amount of child behavior problems. Another study of child witnesses of violence, however, found that mothers' mental health did not affect a child's response to violence in the home (McClosky et al., 1995).

Family support and children's perceptions of their parental relationships have also been identified as key parent-child variables. For example, Durant et al. (1994) found home environments to be important among the 225 urban black adolescents they studied. Adolescents exposed to community and domestic violence appeared to cope better if they lived in more stable and socially connected households.

Research Methods Used to Study Child Witnessing

Interpreting this literature raises several problems based on the research methodologies applied. These include problems with definitions, samples, sources of information, measures, and research designs. Each is reviewed below. While together these flaws raise serious questions about this body of literature, these problems should not cause us to dismiss findings that are consistently replicated across different studies using different methods and samples.

Definitions
A significant problem in this body of literature is that many researchers have failed to differentiate abused children from those who are not themselves abused but who witness family violence. For example, Kolbo (1996) notes that of the 60 child witnesses he studied at a non-shelter domestic violence program all but two were also targets of violence. Some authors do not even identify the degree to which the children studied are both abused and witnessing violence. Rather, they sometimes present their data as representative of children who only witness violence. As Silvern et al. (1995) have stated, "the relationship between reported partner and child abuse should warn that research could be flawed if it is assumed that shelter samples of children have been exposed solely to partner abuse" (p. 195).

Samples
Another issue in this literature is that most studies draw on samples of children and their mothers who are located in shelters for battered women. While this research generates very important information for shelter-based programs, residing in shelters may be a very stressful point in a child's life and not representative of his or her mental health in the long run. Not only have shelter-resident children most likely witnessed a violent event but they have also been removed from the familiar surroundings of their homes, neighborhoods and often their schools.

Sources of reports
Who reports the child's problems in a study may also skew the information we receive. Almost all of the studies reported above relied on mothers' reports of their children's problems. O'Brien, John, Margolin and Erel (1994) have shown that many parents report their children are unaware of violence between the adults when the children, in fact, report awareness of it. Studies that rely on the reports of only parents to define witnessing may incorrectly classify significant numbers of children as non-witnesses. Studies have also shown that in reports of other forms of maltreatment there are discrepancies between child, parent, clinician and agency ratings of problems. Sternberg, Lamb and Dawud-Noursi (1998) have found that child witnesses of violence and their parents differ significantly on the problems they report to researchers.

Measures
The over-reliance on a single reporter is a theme that is carried through to the measures used in these studies. The reason "internalized" or "externalized" behavior problems are so frequently mentioned in this literature is a direct result of the repeated use of the Child Behavior Checklist as mentioned earlier. Very few investigators have ventured beyond the use of this measure of a few others such as the Trauma Symptoms Checklist and there is not currently a standardized measure developed that addresses the unique problems experienced by children who witness violence at home. Such measures should include an assessment of a child's perceived safety. Other variables not yet measured include disruption in child's social support network among extended family members, school personnel and friends, the safety and effect of visitation arrangements, and the effect of changed economic factors on the child's development.

Design
A final weakness in this area of study is that most studies are correlational. As Holtzworth-Munro, Smutzler and Sandin (1997) point out, these studies only show associations between being a witness and some other variable such as a behavior problem. We generally speak of the effects of witnessing violence on children's development. In reality, however, these studies reveal only an association between the variables without predicting that one variable caused the other to occur or vice versa. Many people make the assumption that finding an association is the same as finding that a particular event such as witnessing violence caused a child's problems.

Implications

The studies reviewed for this document provide strong evidence that children who witness domestic violence at home also exhibit a variety of behavioral, emotional, cognitive and longer-term developmental problems. Each child will experience adult domestic violence in unique ways depending on a variety of factors that include direct physical abuse of the child, his or her gender and age, the time since exposure to violence, and his or her relationship with adults in the home. Significant percentages of children in the studies reviewed showed no negative developmental problems despite witnessing repeated violence. We must be careful to not assume that witnessing violence automatically leads to negative outcomes for children.

These data are primarily based on samples of children living in shelters for battered women. This has been used as a criticism of these studies on the grounds that shelter residence is a time of crisis and not representative of a child's on-going life. These data do, however, provide shelters with a much better understanding of the problems many of their resident children may be experiencing. And despite the limitations of some individual studies cited, the number and variety of studies so far reported provide a strong basis for accepting the overall findings.

There is a danger that these data may lead some child protection agencies to more frequently define child witnessing of violence as a form of child abuse or neglect. It is not uncommon to see battered women charged with "failure to protect" their children from a batterer. Many child protection agencies continue to hold battered mothers solely responsible for their children's safety. These actions are often based on the belief that separating from a batterer will always be the safest path for the battered woman and her child.

Yet these actions on the part of the child protection system ignore the reality that the majority of assaults and murders of battered women occur after they have been separated or divorced from their perpetrator. Such actions also ignore the reality that battered mothers often make decisions about their relationships with male partners based on their judgments of what will be best for their children.

The responsibility for creating a dangerous environment should be laid squarely on the shoulders of the adult who is using violent behavior, whether or not that adult is the legal guardian of the child. Responsibility and blame should not be placed on adult survivors in the home. Holding the violent abuser responsible for ending the use of violence is the path that leads to safety for these children and their abused mothers.

It is likely that the outcomes of additional studies on this topic will be reported in the immediate future. The responses to existing and future studies should be to identify ways to provide safety to both children and any abused adults who also reside in their homes.

Author of this document:
Jeffrey L. Edleson, Ph.D.
Director, Minnesota Center Against Violence & Abuse
Professor, School of Social Work
University of Minnesota
jedleson@umn.edu

April 1997 (Revised April 1999)

Recent Reviews of the Literature

Edleson, J. L. (1999). Children's witnessing of adult domestic violence. Journal of Interpersonal Violence, 14(8), (Forthcoming August 1999).

Holtzworth-Munroe, A., Smutzler, N., & Sandin, B. (1997). A brief review of the research on husband violence. Part II: The psychological effects of husband violence on battered women and their children. Aggression and Violent Behavior, 2, 179-213.

Margolin, G. (1998). Effects of witnessing violence on children. In P. K. Trickett, & C. J. Schellenbach (Eds.), Violence against children in the family and the community (pp. 57-101). Washington, DC: American Psychological Association.

Additional References Cited

Achenbach, T. M., & Edelbrock, C. (1983). Manual for the child behavior checklist and revised child behavior profile. Burlington, VT: University of Vermont Department of Psychiatry.

Carlson, B. E. (1991). Outcomes of physical abuse and observation of marital violence among adolescents in placement. Journal of Interpersonal Violence, 6, 526-534.

DuRant, R. H., Cadenhead, C., Pendergrast, R. A., Slavens, G., & Linder, C. W. (1994). Factors associated with t he use of violence among urban Black adolescents. American Journal of Public Health, 84, 612-617.

Henning, K., Leitenberg, H., Coffey, P., Turner, T., & Bennett, R. T. (1996). Long-term psychological and social impact of witnessing physical conflict between parents. Journal of Interpersonal Violence, 11, 35-51.

Hughes, H. M. (1988). Psychological and behavioral correlates of family violence in child witness and victims. American Journal of Orthopsychiatry, 58, 77-90.

Hughes, H. M., Parkinson, D., & Vargo, M. (1989). Witnessing spouse abuse and experiencing physical abuse: A "double whammy"? Journal of Family Violence, 4, 197-209.

Jaffe, P., Wilson, S., & Wolfe, D. E. (1986). Promoting changes in attitudes and understanding of conflict resolution among child witnesses of family violence. Canadian Journal of Behavioral Science, 18, 356-366.

Kolbo, J. R. (1996). Risk and resilience among children exposed to family violence. Violence & Victims, 11, 113-128.

Mathias, J. L., Mertin, P., & Murray, A. (1995). The psychological functioning of children from backgrounds of domestic violence. Australian Psychologist, 30, 47-56.

McClosky, L. A., Figueredo, A. J., & Koss, M. P. (1995). The effects of systemic family violence on children's mental health. Child Development, 66, 1239-1261.

O'Brien, M., John, R. S., Margolin, G., & Erel, O. (1994). Reliability and diagnostic efficacy of parents' reports regarding children's exposure to marital aggression. Violence and Victims, 9, 45-62.

O'Keefe, M. (1994). Linking marital violence, mother-child/father-child aggression, and child behavior problems. Journal of Family Violence, 9, 63-78.

Peled, E. (1996). Supporting the parenting of battering men: Issues and dilemmas. Manuscript submitted for publication.

Rossman, B. B. (1998). Descartes's error and posttraumatic stress disorder: Cognition and emotion in children who are exposed to parental violence. In G. W. Holden, R. Geffner, & E. N. Jouriles (Eds.), Children exposed to marital violence (pp. 223-256). Washington, DC: American Psychological Association.

Silvern, L., Karyl, J., Waelde, L., Hodges, W. F., Starek, J., Heidt, E., & Min, K. (1995). Retrospective reports of parental partner abuse: Relationships to depression, trauma symptoms and self-esteem among college students. Journal of Family Violence, 10, 177-202.

Singer, M. I., Miller, D. B., Guo, S., Slovak, K., & Frierson, T. (1998). The mental health consequences of children's exposure to violence. Cleveland, OH: Cayahoga County Community Mental Health Research Institute, Mandel School of Applied Social Sciences, Case Western Reserve University.

Song, L., Singer, M., & Anglin, T. (1998). Violence exposure and emotional trauma as contributors to adolescents' violent behaviors. Archives of Pediatric and Adolescent Medicine, 152, 531-536.

Spaccarelli, S., Coatsworth, J. D., & Bowden, B. S. (1995). Exposure to serious family violence among incarcerated boys: Its association with violent offending and potential mediating variables. Violence and Victims, 10, 163-182.

Stagg, V., Wills, G. D., & Howell, M. (1989). Psychopathology in early childhood witnesses of family violence. Topics in Early Childhood Special Education, 9, 73-87.

Sternberg, K. J., Lamb, M. E., & Dawud-Noursi, S. (1998). Using multiple informants to understand domestic violence and its effects. In G. W. Holden, R. Geffner, & E. N. Jouriles (Eds.), Children exposed to marital violence (pp. 121-156). Washington, DC: American Psychological Association.

Wolfe, D. A., Jaffe, P., Wilson, S. K., & Zak, L. (1985). Children of battered women: The relation of child behavior to family violence and maternal stress. Journal of Consulting and Clinical Psychology, 53, 657-665.

Wolfe, D. A., Zak, L., Wilson, S., & Jaffe, P. (1986). Child witnesses to violence between parents: Critical issues in behavioral and social adjustment. Journal of Abnormal Child Psychology, 14, 95-104.

Bibliography on "Children Who Witness Violence: Research & Intervention" by J. L. Edleson.

Bibliography from the Project to Address Violence through Education (PAVE) at the University of Minnesota on "Young children and violence" .

Mothers & Children: Understanding the Links Between Woman Battering and Child Abuse is a briefing paper by Jeffrey L. Edleson for a recent strategic planning meeting on the Violence Against Women Act.

In the Best Interest of Women and Children: A Call for Collaboration Between Child Welfare and Domestic Violence Constituencies is a briefing paper by Susan Schechter and Jeffrey L. Edleson prepared for a Wingspread Conference of a similar title.

Child Witness to Domestic Violence is a brief paper written by Kathryn Conroy, DSW, on the effect on children of witnessing their mothers being battered.

Children and Family Violence: The Unnoticed Victims is a May 1994 report by Gabrielle M. Maxwell of New Zealand's Office of the Commissioner for Children.

An art gallery from the Domestic Abuse Project in Minneapolis of 13 images drawn by children who have witnessed violence.