Abstract
The main purpose of this study was to examine the association of family polyvictimization in childhood with the victimization or perpetration of spousal abuse and the perpetration of child abuse in adulthood. While associations between maltreatment in childhood and subsequent perpetration or victimization in adulthood are well documented, their association with polyvictimization in childhood (i.e., experiencing multiple types of victimization) has received less attention. This research aims to empirically investigate 1) whether early experiences of family polyvictimization are predictive of subsequent experience or spousal abuse or perpetration of child abuse and 2) whether there are gender differences in those associations. Through conducting Chi-square analysis and logistic regression analysis with South Korea’s National Domestic Violence Survey 2019, the study found significant empirical evidence that polyvictimization is predictive of perpetration or experience of spousal abuse and perpetration of child abuse. Particularly, adults who experienced polyvictimization in childhood were more likely to perpetrate child abuse regardless of the type of abuse (p < .001). Polyvictimization in childhood was also significantly associated with perpetration and victimization of spousal abuse (p < .001). In regard to gender differences, this study found that males were more likely to be polyvictimized by family in childhood (15.11%) than females (10.23%), and polyvictimization was found to increase the likelihood of females being revictimized in adulthood. More attention should be paid to victimization by multiple types of violence within the family and its influence on intergenerational transmission of violence.
Introduction
Polyvictimization, an exposure to multiple and varied traumatic violence (Bidarra et al., 2016; Chan et al., 2019; Finkelhor et al., 2007a), has significant long-term effects on children’s developmental outcomes (Finkelhor et al., 2007b; Rapsy, et al., 2019), compared to single victimization (Devries et al., 2017). While prior studies have extensively examined the specific and individual contributions of specific types of victimization (e.g. sexual, physical, and emotional abuse; neglect; bullying; and witnessing domestic violence) on various mental and behavioral health outcomes (Chan et al., 2017; deLara, 2019; Salokangas et al., 2019; Yoon et al., 2018), recent research has increasingly documented children and youth who have experienced multiple types of victimization over the course of their life (Finkelhor et al., 2007b, 2011). Particularly, youth who experience one type of abuse from their parents or main caregiver is more likely to suffer other types of abuse in their family (O’Leary, et al., 2000; Wolfe, 2018) because family members share common risk factors of violence (Finkelhor et al., 2007a). Compared to repeated experience of the same type of abuse, polyvictimization is associated with considerably more severe developmental outcomes (Finkelhor et al., 2007b) including physical and mental health, as well as other psychopathological and psychosomatic symptoms (Chan et al., 2017; Hesketh et al., 2010). In addition, a significant overlap between child abuse and neglect in a family has been found (Gracia et al., 2020), in that approximately 25–40% of children experienced more than one form of abuse (Clément & Bounchard, 2005; Finkelhor et al., 2007b).
Added on to the abuse and neglect, spousal abuse has been increasingly recognized as a distinct form of child abuse and neglect, having an array of negative impact on children’s developmental outcomes (Kimber et al., 2018). While not a direct form of victimization, studies suggest that many children witness spousal abuse between their parents or caregivers (Borrego et al., 2008). Witnessing parent’s spousal abuse between caregivers is a trauma that may have devastating effects on all domains of a child development (Howell et al., 2016). For example, children who witnessed spousal abuse in their family showed immediate negative outcomes such as aggressive behavior to peers and dating partners (McCloskey & Lichter, 2003), post-traumatic distress disorder (PTSD), and depression (Moretti et al., 2006). Long-term adverse outcomes were also reported, including problems in relationships with intimate partners (Rosenbaum & Leisring, 2003) or in substance abuse in adulthood (Caetano et al., 2003). Children who witnessed spousal abuse are known to be at a greater risk of being abused by parents in other forms (Chan, 2011; Devries et al., 2017). Given the detrimental impact of merely witnessing spousal abuse on developmental outcomes (Wood & Sommers, 2011), it is worth regarding it as a type of child maltreatment and examining its cumulative impact on later outcomes.
Research on revictimization (Kimerling et al., 2007) has consistently indicated a link between childhood maltreatment and subsequent perpetration or victimization in adolescence or adulthood. For example, prospective studies have found that being abused or neglected in childhood increases the likelihood of perpetrating child maltreatment later as a parent (Thornberry & Henry, 2013). In particular, those who reported being a victim of physical or psychological abuse in adolescence and spousal abuse in adulthood also reported to have been victims of physical or sexual abuse in childhood (Messing et al., 2012) and witnessed parental violence as children (Franzese et al., 2017). Also, over 60% of survivors of child sexual abuse were sexually revictimized as adults (Classen et al., 2005). Likewise, prior research has documented that experiencing or witnessing parental violence as a child is one of the most consistent predictors of subsequent spousal abuse in adulthood (Aakvaag et al., 2017; Dugal et al., 2018; Godbout et al., 2009). However, these lines of research has paid less attention to polyvictimization, mostly focusing on a single type of child abuse or neglect, often child sexual abuse. The evidence for an association between polyvictimization in childhood and later perpetration is both scant and mixed.
Potential gender differences in the current associations can be considered. For instance, females are usually more likely to suffer sexual abuse (Jung et al., 2019; Martin et al., 2004), while males are more likely to suffer physical abuse (Thompson, et al., 2004). However, gender differences in the link between abuse and neglect in childhood and spousal abuse in adulthood have received less attention, leading to inconclusive results (Jung et al., 2019). While Lackey (2003) found that experiencing child abuse and neglect was a significant cause of spousal abuse in later life only for men, an indirect effect of child abuse and neglect on spousal abuse was found for female (Herrenkohl et al., 2004).
The Current Study
While the co-occurrence of child abuse, neglect, and spousal abuse in a family and its effects on child development have been well documented, the scope of previous studies has been limited in several different ways. First, previous works failed to capture the intersectional characteristics of multiple abuse victimization (Finkelhor et al., 2007a; Wolfe, 2018), including witnessing spousal abuse, and its predictive capacity into their conceptualization of polyvictimization and later experience or perpetration of child abuse. Second, the evidence for an association between polyvictimization in childhood and later perpetration is scant and mixed. Third, previous research has not fully examined polyvictimization and its effects in Asian countries. Fourth, previous studies failed to capture gender differences in the link between polyvictimization in childhood and both the experience and perpetration of spousal abuse in adulthood.
The current study builds on prior work and examines the associations between polyvictimization in childhood and subsequent perpetration of child abuse in the South Korean population. The central research questions are (a) Do early experiences of family polyvictimization have an impact on subsequent experience or perpetration of spousal abuse and child abuse? (b) Are there gender differences in the link between childhood polyvictimization, spousal abuse and child abuse in adulthood? We hypothesized that exposure to more types of family polyvictimization would increase the likelihood of perpetrating child abuse and neglect as well as perpetration and experience of spousal abuse, beyond exposure to individual type of child abuse. In this study, polyvictimization was limited to the experiences gained from families to examine the intergenerational transmission of violence within a family context. Identifying co-occurrences of various types of child abuse and neglect within a family may reveal the cumulative burden of this victimization and provide implications for further prevention and intervention in child abuse and neglect.
Methods
Data and Sampling
Data was taken from the Domestic Violence Survey in 2019, which was conducted by the South Korean Ministry of Gender Equality and Family (MOGEF). According to the Act on the Prevention of Domestic Violence and Protection, etc. of Victims in 1997, MOGEF has conducted a triennial survey of domestic violence in Korean society. The survey specifically focuses on the perpetration and experience of spousal abuse and child abuse (MOGEF, 2019). For the purpose of this study, we used the 2019 survey, which is the most recent survey about domestic violence. The survey data was collected by door-to-door interviews from August 2019 to November 2019, and respondents aged 19 or older were sampled through multi-stage probability proportion sampling and systematic sampling based on 2017 Korea National Population and House Census conducted by Statistics Korea (MOGEF, 2019). Trained interviewers visited 28,310 sampled households and among them, 9060 households were successfully responded to the interview. This survey is aware of the possibility that respondents are reluctant to reveal their experiences of victimization. For this reason, the safety protocol to protect respondents was strictly enforced by keeping survey participation and responses confidential to other family members. The success rates of this survey were 48.8% and it was significantly increased than the response rates in 2016 (44.1%) and 2013(34.2%) (MOGEF, 2019). A total number of 9060 households were sampled and one adult of each household was interviewed by a professionally trained interviewer. Considering the higher risks of spousal abuse victimization of women, the gender ratio was adjusted 2:1, which means women were twice more sampled than men (MOGEF, 2019). Among 9060 subjects in that survey, this study only used the data from 5688 married individuals because the sample of 2019 DVS included other types of households including single households. Due to the purpose of this study to explore the association between polyvictimization in childhood and violence between partners (spousal abuse), this study only utilized the information of married (or cohabitating) couples.
Variables
Polyvictimization in Childhood
The Domestic Violence Survey asked respondents about five types of negative experiences in the family when they were 18 years old or younger: emotional abuse (i.e., blaming and insulting), minor physical abuse (i.e., corporal punishment), severe abuse (i.e., punching and kicking and hitting with an object), neglect (i.e., failure to provide food and medical care and leaving a child unattended), and indirect abuse (i.e., witnessing domestic violence between parents including insulting, punching or kicking severely, and hitting with an object). The nine items asking about these experiences were adopted from the Parent-Child Conflict Tactics Scales (PCCTS) that was suggested by Straus et al. (1998) (MOGEF, 2019). Responses to questions about experiencing each type of abuse were coded as dichotomies (Yoon et al., 2019).
Polyvictimization is defined as experiencing multiple types of victimization (Finkelhor et al., 2007a), but its operational definition varies among researchers (Ford & Delker, 2018). Some studies have conceptualized polyvictimization as cumulative frequency (Andrew et al., 2015; Mustanski et al., 2016) but most others have analyzed it as a dichotomized variable (Finkelhor et al., 2007ab; Finkelhor, et al., 2011; Pereda et al., 2014). Finkelhor et al. (2011), who are distinguished researchers in the field, defined the threshold of polyvictimization as the top 10% of individuals in a sample of victimized children. In other words, the top 10% of individuals who had experienced the highest numbers of types of victimizations were identified polyvictimized group and later studies have also adopted this standard. In the sample in this study, individuals who experienced four or five types of violence in childhood fell within this threshold and were thus categorized as the polyvictimization group.
Perpetration of Child Abuse
Perpetration of child abuse was measured using 14 items in four categories (emotional abuse, minor physical abuse, severe physical abuse, and neglect), and the items were adopted from PCCTS. (1) Emotional abuse was measured with two items asking whether respondents had threatened a child with physical violence or blamed or insulted them. (2) Minor physical abuse was measured with a single item asking whether respondents had slapped a child in the face, head or body. (3) Severe physical abuse was measured with seven items asking whether respondents had hit a child with an object, pushed or shoved them, punched or kicked them severely, beaten them violently, choked or suffocated them, deliberately burned them, or threatened or hurt them with a weapon. (4) Neglect was measured with four items asking whether respondents had failed to provide a child with food, failed to provide medical care, struggled to take care of them because of substance abuse or heavy drinking or left a child unattended (MOGEF, 2019). Perpetration was coded as dichotomous (Yoon et al., 2019), and respondents were regarded as having abused a child in each category if they reported any perpetration in that category.
Perpetration and Experience of Spousal Abuse
The Domestic Violence Survey classifies perpetration and experience of spousal abuse into five major categories: emotional violence, physical violence, sexual violence, financial violence, and control. This study used 18 items in three of these five categories (emotional, physical, and sexual violence), which were adopted and revised from the Revised Conflict Tactics Scale suggested by Straus et al. (1998), to measure the perpetration and experience of spousal abuse. In order to maintain consistency with the measures of child abuse, we divided physical violence into minor and severe physical violence. (1) Emotional violence was measured with six items asking whether the respondent had ever insulted their partner, threatened them with physical violence, destroyed their partner’s personal items, threatened to hurt their pets or loved ones, threatened to hurt or kill themselves, or interrupted their sleep. (2) Minor physical violence was measured with three items asking whether the respondent had ever thrown something at a partner that could cause pain; pushed or shoved them; grabbed their arms, shoulders, or hair; or slapped their face, head, or body. (3) Severe physical violence was measured with four items asking whether the respondent had ever choked or burned their partner, threatened or hurt them with a weapon, hit them with an object, or punched or kicked them severely. (4) Sexual violence was measured with five items asking whether the respondent had ever coerced their partner into unwanted sexual activity, coerced them into unwanted sexual activity using physical force, forced them to submit to unwanted physical touch, taken video or photographs of a partner’s body or sexual acts against their will, or engaged in non-consensual sharing or uploading of such video or photographs (MOGEF, 2019). Since the main objective of this study was to investigate whether or not people have perpetrated or experienced spousal abuse, rather than the frequency of such violence, the categories were condensed into dichotomous questions asking whether respondents had perpetrated or experienced the violence in each category.
Analytic Strategies
This study used the chi-square test to model the association between family polyvictimization and subsequent family violence in Korea. Through analyzing statistical differences between the polyvictimized group and the other groups (namely, those who experienced no victimization and those who experienced 1–3 types of victimization, which means non-polyvictimization, during childhood), we aimed to investigate the hypothesis that early experiences of family polyvictimization are a good predictor of subsequent experience or perpetration of family violence. Furthermore, logistic regression analyses about the association between polyvictimization in childhood and the risk of experiencing or perpetrating family violence in adulthood (i.e., perpetrating child abuse, perpetrating spousal abuse, or experiencing spousal abuse) were conducted to substantially support the findings of the chi-square test. Gender was included in the logistic regression models as the main predictor of experiencing or perpetrating family violence. To account for the influence of sociodemographic differences in victimization or the carryover of victimization from baseline to follow-up, covariates included age, education level (middle school, high school, and college or higher), income level, and employment status.
Results
Demographic Characteristics of the Sample
Descriptive Statistics (N = 5688).
Note. 1 $ = 1133 Won (exchange rate in 2021, July 1).
Experience of Polyvictimization in Childhood by Gender
Figure 1 presents data about the experience of victimization in childhood by gender. Across all types of abuse, males were more likely to have been victimized than females. Males reported more frequent experience of emotional abuse (29.8%), minor physical abuse (47.8%), severe physical abuse (24.14%), neglect (6.32%), and indirect abuse (39.34%), while females reported less frequent experience of emotional abuse (24.07%), minor physical abuse (39.95%), severe physical abuse (16.98%), neglect (5.00%), and indirect abuse (34.41%). The gender differences in emotional abuse, minor and severe physical abuse, and indirect abuse were statistically significant (p < .001), and the gender difference was also statistically significant in neglect (p < .05). Experience of Victimization in Childhood by Gender.
The Number of Types of Victimization in Childhood by Gender.
*p < .05, **p <.01, ***p < .001.
The Types of Victimization in Childhood by Polyvictimized Group.
The Association Between Family Polyvictimization in Childhood and Subsequent Experience and Perpetration of Family Violence in Adulthood
The Association between Family Polyvictimization in Childhood and Perpetration of Child Abuse.
*p < .05, **p <.01, ***p < .001.
The Association Between Family Polyvictimization in Childhood and Perpetration of Spousal Abuse.
*p < .05, **p <.01, ***p < .001.
The Association between Family Polyvictimization in Childhood and Victimization from Spousal Abuse.
*p < .05, **p <.01, ***p < .001.
Logistic Regression Analysis for Predicting the Risk of Experiencing or Perpetrating Family Violence in Adulthood.
*p < .05, **p <.01, ***p < .001.
Discussion
Confirming prior studies in Korean context, the current study suggests the possibility of intergenerational transmission of violence within the family context. Findings indicate that polyvictimization in childhood was significantly associated with spousal abuse and child abuse in adulthood. In addition, gender differences in the same associations were found. Experiencing polyvictimization in childhood had an impact on abuse perpetration towards their child in adulthood, regardless of the type of abuse. While there were partial associations between polyvictimization and perpetration of child abuse among male, perpetration of every type of child abuse was associated with polyvictimization for female.
In current study, males were more likely to experience polyvictimization in childhood and gender differences were found in the association between polyvictimization in childhood and abuse perpetration in adulthood. Previous studies showed inconsistent results in gender differences in the experience of child abuse and neglect. Some studies found that females were more likely to be maltreated within a family (Pereda et al., 2014), while others found that males were more likely to be physically abused (Thompson et al., 2004). Other studies also no gender differences were found in the child maltreatment experiences. In other case, no gender differences were found in any type of child maltreatment, except sexual abuse (Kerig & Modrowski, 2018). Although it is not conclusive whether males are more easily victimized by child abuse in previous studies yet, some argued that there are gender differences in the responses and the effects of child abuse victimization. For example, Jung et al. (2019) found that childhood victimization experiences (physical–emotional abuse and the exposure to domestic violence in childhood) led negative consequences like perpetrating more violence. In Gómez (2011)’s study, child abuse experience predicted spousal abuse victimization and perpetration and it was found in both men and women but the magnitude were different by showing a greater association in men. This study is one of the early attempts to deepen understanding of polyvictimization experiences in childhood and how this is associated with later lives. It is premature to conclude that the findings in this study particularly focused on the population in South Korea are consistent in other contexts. More research is needed to expand our knowledge of gender differences in polyvictimization and its impact on later outcomes.
Strengths and Limitations
This study contributes to the relevant literature in several important ways. First, it strives to adapt family polyvictimization to the context of South Korea, where the prevalence of child maltreatment is noticeably high (MOGEF, 2019). Second, the current conceptualized polyvictimization by considering witnessing spousal abuse in childhood as an important type of child maltreatment has not been fully considered in previous research. Third, the study tested the potential gender differences. Fourth, the study leveraged the data from the nationally representative samples of Korean population.
Despite these strengths, there are limitations that should be considered when interpreting our findings. First, polyvictimization measure was depended on retrospective report of child abuse and neglect. Although retrospective reports accounts offer the benefit of capturing data about experiences across childhood, they have several limitations, including recall bias. Second, the study limited polyvictimization experiences within the family, omitting violence exposure outside of the family. Follow-up studies should examine more detailed possibilities of abuse, neglect, and witnessing spousal abuse within and outside of family context and its long-term effect. Third, the study dichotomized PCCTS. This may create some issues of disregarding the contextual severity of polyvictimization in childhood. The current study intended to measure any experience of childhood exposure to multiple types of victimization before the age of 18, thus created the dichotomous items of nine type abuse, neglect, and witnessing spousal abuse. This needs to be more carefully handled in the future work. At last, this study mainly focused on the population in South Korea where child abuse prevalence rates are remarkably high. Due to this limitation, it is not appropriate to generalize the results of this study, ignoring these cultural contexts. However, the findings of intergenerational transmission of maltreatment of South Korean population can be meaningful, to other countries where they have similar cultural contexts with high prevalence rates in child maltreatment.
Conclusion and Implications
There are several implications of our findings. First, practitioners should work to assess family polyvictimization experiences to understand the background of their violent behavior in adulthood. This information can be used to tailor interventions for children who have experienced child maltreatment and witnessing spousal abuse within family and to break the cycle of intergenerational transmission of violence. Second, prevention remains a top priority in the area of childhood maltreatment and spousal abuse research. Attempts need to be made to identify children who may be at higher risk of child maltreatment. Third, assessing polyvictimization would help practitioners to intervene more effectively by planning intensive strategies to mitigate the cumulative effects of polyvictimization.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
