Domestic Violence Exposure Effects on Children: Qual & Quant Studies

Introduction: Domestic Violence as a Public Health Concern

Domestic violence is an ongoing experience of physical, psychological, and sexual abuse in the home, and is often a method used by one adult to establish control and power over another person. Exposure by children to marital aggression is now a recognized public health concern. The investigation of the effects of such exposure on the functioning of a child is a significant societal concern. The two studies reviewed below — one qualitative, one quantitative — each address different dimensions of this problem, together offering a broader understanding of how domestic violence affects children's psychological and behavioral development.

Qualitative Study: Children of Incarcerated Mothers

DeHart and Alshuler (2009) examined incarcerated women's accounts of their children's exposure to violence prior to the mother's incarceration. The researchers noted that previous research had explored the consequences of children having their mothers incarcerated — concluding that incarceration alone has few direct consequences for children's health and behavior — but that there is a dearth of research on these children's experiences prior to the mother's incarceration. This gap is a central factor in determining appropriate interventions for children who are subsequently living with the other parent, in extended kinship care, or in community placements. The study was designed to provide a foundation for studying violence exposure by examining the mothers' own accounts, allowing researchers to identify the types of exposure children experienced and to better determine implications for social work practice and policy.

Participants were randomly sampled from a maximum-security state correctional facility. The sample of 60 women included 52% African Americans and 48% Caucasians, with an age range of 18 to 70 (median age of 31 years). The women were incarcerated for a variety of offenses ranging from murder to forgery, burglary, and grand larceny, and were serving sentences ranging from 15 months to life imprisonment. The median time already served at the completion of the study was just under four years.

Open-ended interviews were conducted regarding each participant's family and relationship history, history of victimization, and lifetime delinquency and interactions with the justice system. Participants were not directly questioned about their children's violence exposure; however, this information emerged naturally during the interviews. Handwritten notes were transcribed and analyzed using a grounded theory methodology and ATLAS/ti qualitative software.

The qualitative study examined the effects of exposure to maternal partner violence (independent variable) on children's psychological well-being and behavior (dependent variables). These variables were broadly defined, as the study was intended to provide a foundation from which future research could better define methods and measures for further investigation.

According to the researchers, 90% of the sample had children living with them prior to incarceration, and 75% of those women mentioned the impact of abuse on their children — meaning approximately 42 out of 56 children had been exposed. The authors provide rich descriptions drawn from interview transcripts. Psychological and emotional effects of witnessing violence were categorized into fear and worry about violence occurring, with reactions ranging from withdrawal to severe emotional responses. Behavioral effects included passivity (hiding), attempting to intervene and stop the abuse, acting out, running away, substance abuse, and committing delinquent acts. In many cases, the children were also directly abused rather than merely witnessing violence; these instances are discussed briefly in the study.

This study reinforces a broader theme in the literature: dysfunctional family interactions affect children, and children are not immune to the effects of abusive situations even when they are not directly abused themselves.

Quantitative Study: Dual Exposure and Adolescent Outcomes

The potential for inaccurate or biased recall is very high in this study. Furthermore, there is no clear distinction between the effects of merely witnessing parental abuse, being directly abused, or experiencing both. The researchers could also have specified the documented types of effects associated with children witnessing violence — drawing on existing literature — and mapped their findings onto those established categories, rather than offering broadly generalized conclusions.

Reference: De Hart, D. D., & Alshuler, S. J. (2009). Violence exposure among children of incarcerated mothers. Behavioral Science, 26(5), 467–479.

Moylan et al. (2010) posit what they call a dual exposure effect — or "double whammy" — wherein children exposed to both witnessing domestic violence and direct child abuse fare worse with respect to later outcomes than do children exposed to only one form of violence. Having reviewed previous research on this dual exposure effect and found mixed results, the researchers conducted a study examining several adolescent outcomes that have demonstrated empirical links to childhood adversity. They assessed a range of internalizing and externalizing behaviors and hypothesized: (1) that exposure to violence of any type in childhood will increase a child's risk for detrimental outcomes, and (2) that children exposed to both abuse and witnessing domestic violence will demonstrate increased risk compared to those exposed to either form alone. The researchers were also interested in whether gender moderated the relationship between early exposure and adolescent outcomes.

Data came from the Lehigh Longitudinal Study, a study of children and families begun in the 1970s to examine the consequences of child maltreatment. Participants were recruited from multiple settings in Pennsylvania, including Head Start programs, child welfare abuse and protective service programs, day care programs, and private nursery schools. A total of 416 participants were assessed through adolescence: 229 (55.0%) were male; 81.5% (n = 339) were Caucasian; 11.7% (n = 49) were biracial; 5.0% (n = 21) were African American; 1.4% (n = 6) were American Indian or Alaska Native; and 0.2% (n = 1) was Pacific Islander.

At the time of this research, three waves of data had been collected (preschool, school-age, and adolescence), with a fourth wave underway as the original child participants had reached adulthood. Data from the preschool and school-age assessments were collected through interviews with parents. Data from the adolescent assessment came from face-to-face interviews and individually administered questionnaires completed by both parents and adolescents.

Violence exposure (dichotomous) and domestic violence exposure (dichotomous) served as independent variables, with gender entered as a moderator and covariate. Abuse exposure was determined via official records, mothers' reports, and adolescents' reports, and required confirmation across all three sources. Dependent variables included internalizing and externalizing behaviors as measured by a standardized youth questionnaire, a measure of depression, and a measure of delinquency.

Standardized measures were used for the dependent variables. The Achenbach Youth Self-Report Inventory (YSR) was completed by adolescent participants, with subscales from both the internalizing and externalizing composite scales scored and used in analyses. The Beck Depression Inventory (BDI) and a Delinquent Acts Questionnaire were also used as dependent measures.

Regression models indicated that children exposed to any form of violence in childhood had higher levels of both internalizing and externalizing behaviors as adolescents compared to those who had not been exposed. Children who had both witnessed domestic violence and been direct victims of abuse were at higher risk for the full range of internalizing and externalizing problems in adolescence than children exposed to only one type. Overall, females scored higher on measures of internalizing behaviors and males on externalizing behaviors, and gender did moderate the effects of earlier exposure on adolescent outcomes.