The concept of intergenerational transmission of violence is rooted in several theoretical foundations, including social learning theory (Bandura, 1971, 1978). Social learning theory of aggression posits that children of the parents who prefer aggression to problem-solve tend to exhibit aggressive interpersonal behaviors acquired through observational learning (Bandura, 1978). Their learned violent behaviors are reinforced through family members’ approval and rewards over time, and thus aggressive interpersonal behaviors are used as an acceptable or effective means of conflict resolution (Bandura, 1978; Ehrensaft et al., 2003). In the context of the intergenerational transmission of violence, adults with a history of childhood maltreatment may adopt their learned violent behaviors toward their children when parenting-related issues arise (Fuchs, Mohler, Resch, & Kaess, 2015; Savage, Palmer, & Martin, 2014).

Guided by the intergenerational transmission of violence hypothesis, several studies have examined the long-term effects of childhood maltreatment on later parenting behaviors. Meta-analysis studies have shown that exposure to any childhood abuse and neglect was associated with an increase in harmful and maltreating parenting practices (Madigan et al., 2019; Savage, Tarabulsy, Pearson, Collin-Vezina, & Gagne, 2019). Empirical evidence also suggests that individuals who experienced varying types of childhood abuse and/or neglect were more likely to exhibit negative parenting practices compared to those without such early victimization (Fuchs et al., 2015; Kim, 2009; Lakhdir et al., 2019; Zvara et al., 2015).

Moreover, existing studies suggest the negative impact of low socioeconomic status (SES) on the intergenerational transmission of violence. Zvara and colleagues (2015) revealed that previously maltreated mothers in low-income families with lower educational attainment were more likely to use harsh intrusive parenting than maltreated mothers who were more socioeconomically advantaged. Similarly, mothers who had been maltreated during childhood and also perpetrated violence to their child reported a higher level of sociodemographic risk (i.e., an index compiling low education, single parenthood, receipt of social welfare benefits, young maternal age) and a lack of perceived family support compared to maltreated mothers who did not maltreat their child (St-Laurent, Dubois-Comtois, Milot, & Cantinotti, 2019). These findings suggest that for parents with a history of childhood maltreatment, low SES (e.g., low income, single parenthood) and lack of available resources (e.g., lack of social supports and networks) may increase the risk of continuing the cycle of intergenerational violence with abusive or neglectful parenting practices.

Table 1 presents weighted descriptive statistics of sample characteristics and key variables. On average, respondents reported that they rarely used psychological aggression (M = 1.56, SD = 0.67) and never used physical aggression and neglect based on the five-point scale (M = 1.30, SD = 0.45; M = 1.42, SD = 0.51, respectively). Approximately 5% of respondents sometimes or more often used abusive and neglectful parenting behaviors. On average, respondents rarely experienced childhood abuse (M = 2.05, SD = 1.16) and childhood neglect (M = 2.01, SD = 0.95) based on the five-point scale. Approximately 10% of respondents often or very often experienced childhood abuse and neglect. About half of the respondents reported witnessing domestic violence during childhood.

Table 3 summarizes the results of OLS regression models predicting respondents’ negative parenting behaviors (i.e., psychological aggression, physical aggression, neglect) as a function of the independent variables: childhood abuse, childhood neglect, and witnessing domestic violence. Due to high correlations among childhood exposure to violence variables (Table 2), we ran separate analyses for each independent variable. In the main effects models [Table 3 (a)], we found that a parent’s history of childhood abuse was associated with more frequent use of psychological aggression (b = 0.12, p < .001), physical aggression (b = 0.06, p < .001), and neglectful parenting (b = 0.06, p < .001). A parent’s history of childhood neglect was associated with greater psychological aggression (b = 0.13, p < .001), physical aggression (b = 0.07, p < .001), and neglectful parenting (b = 0.09, p < .001). Witnessing domestic violence during childhood was also associated with greater psychological aggression (b = 0.29, p < .001), physical aggression (b = 0.14, p < .001), and neglectful parenting (b = 0.11, p < .01).

In the two-way interaction models [Table 3 (b)], we found two significant moderating effects of living in a three-generation household on the association between childhood maltreatment and the use of abusive and neglectful parenting behaviors at the significance level of .05. First, the negative effect of a history of childhood abuse on the use of neglectful parenting was reduced for respondents living in a three-generation household (b = −0.11, p < .05). We also found that the negative effect of witnessing domestic violence on the use of psychological aggression was reduced for respondents living in a three-generation household (b = −0.33, p < .05). Although marginally significant (p < .10), living in a three-generation household moderated the associations between a history of childhood abuse and the use of psychological aggression; a history of witnessing domestic violence and the use of physical aggression; and, a history of childhood neglect and the use of neglectful parenting behaviors.

This study used data collected from 727 young parents (average age: 34 years) who had been investigated by CPS for child maltreatment but had not received ongoing CPS intervention (“deflected families”). The majority of them were African American women with low family income. Our primary aim was to test the intergenerational transmission of violence hypothesis by examining the association between respondents’ retrospective reports of childhood family violence experiences and their current parenting practices. We also examined whether living in a three-generation household moderated the intergenerational transmission of violence.

Our findings were consistent with the first hypothesis and corroborated prior research in the intergenerational transmission of violence (Madigan et al., 2019; Savage et al., 2019) demonstrating that parents who reported a history of childhood abuse, neglect, and witnessing domestic violence showed more frequent use of psychological aggression, physical aggression, and neglectful behaviors against their children. It was notable that childhood family violence experiences were prevalent among the study sample; for example, 47% of the study sample reported having witnessed domestic violence during childhood. When applying the social learning theory (Bandura, 1978), these parents may have learned violent behaviors from their families of origin and justified their use of abusive/neglectful parenting behaviors (Fuchs et al., 2015; Savage et al., 2014). Given such intergenerational connection, adults with noted disadvantages can benefit from parenting education programs focusing on positive, nurturing parent-child interactions. More importantly, policy interventions are warranted to prevent the replication of abusive and neglectful parenting behaviors across generations.

We also supported our second hypothesis by showing that, despite childhood exposure to violence, living in a three-generation household moderated the effect of childhood maltreatment on later parenting practices. Specifically, parents who had witnessed domestic violence exhibited greater psychological aggression against their child, and this association was weaker among those living in a three-generation household. In the same way, living in a three-generation household significantly moderated the association between parents’ history of childhood abuse and their use of neglectful parenting practices. These findings suggest that three-generation household arrangements may help decrease the use of abusive/neglectful parenting practices among parents of young children with low SES, ultimately promoting positive parent-child interactions and child well-being outcomes. For these adults, three-generational households may serve a means and source of relieving financial demands and allows family resources to be shared that can ultimately reduce the incidence of violence against their children (Pilkauskas & Cross, 2018).

To explore the role of grandparent coresidence as a source of social support, we estimated three-way interaction models and examined whether and how the protective role of living in a three-generation household would differ for single parents (i.e., not living with a spouse or partner) who may have greater needs than those living with a spouse or partner. We found a significant three-way interaction effect of a history of childhood abuse, living in a three-generation household, and single parenthood on neglectful parenting behaviors (Table 4, Figure 1), indicating that the protective effect of living in a three-generation household was more pronounced among single parents. This result may indicate that living in a three-generation household appears to help address parents’ unmet needs such as childcare and thus buffer the negative impact of childhood maltreatment on parenting outcomes. As previous studies suggest, grandparent coresidence appears to offer various social supports in the form of emotional, financial, informational, or instrumental assistance (Dunifon et al., 2014; Mutchler & Baker, 2009; Pittman & Boswell, 2008).

To further explore, we conducted other post-hoc analyses to support this speculation. We found that respondents living in a three-generation household tend to use less paid childcare compared to those who were not living with their parents, and this association was stronger for single parents (results available upon request). We also found that for respondents living in a three-generation household, their youngest children were younger than those who were not living with their parents. These results suggest that living in a three-generation household may be most helpful in reducing the occurrence of abusive and neglectful parenting, perhaps through easing the burden of childcare, which warrants further empirical support. Future research should also identify specific aspects of how multi-generational household arrangements are beneficial or stressful to parents and their children.

Our key result concerning the protective role of living in a three-generation household in terms of reducing the use of abusive and neglectful parenting practices shed important insight into adult relationships between grown children with histories of childhood maltreatment and their parents. Our results are somewhat in contrast to those of prior studies that have shown that adult children may continue to be distressed with their previously abusive and neglectful parents, for example, by providing care to them (Kong, 2018b; Kong & Moorman, 2015). Amongst many differences in the current study with Kong’s previous research (i.e., age of respondents and children, relationships with the perpetrating parent), the most notable are the differences in SES status and the roles of adult children and their parents that are tied to a specific phase of life. The impact that childhood family violence has on later family relationships may depend on adult children’s life stages and the concurrent role dynamics between parent-child dyads, which may interact with the needs for social support and resources. Future research should explore the benefits and risks of multi-generational living arrangements based on the past and current relationship quality among co-residing family members.

The current study has limitations. First, the key variables of exposure to childhood violence and current parenting practices were based on self-reported retrospective measures, which may involve recall errors (Macmillion, 2009). Also, childhood abuse measures lacked specific information concerning which parent perpetrated violence against respondents, limiting the assessment of potential stress associated with living in a three-generation household. For example, Kong and Martire (2019) found the negative effect of childhood maltreatment on the relationship with the previously abusive parent, but not with a non-abusive parent. For example, a history of maternal childhood abuse negatively affected adult children’s relationship quality with aging mothers, but not affected their relationship with aging fathers who were previously non-abusive. Relatedly, we cannot rule out an issue of selection bias that adults living in a three-generation household may live with the parent(s) who were less abusive or non-abusive. Second, this study was conducted with a specific high-risk group-families reported to CPS whose cases closed after a CPS investigation. This makes generalizability to other populations somewhat limited, although the nature of the primary research questions renders this population highly relevant. Lastly, the sample size is relatively small. Given the low incidence of some of the phenomena in question, the sample may have lacked statistical power for detecting some hypothesized associations.

Despite the limitations, the current study offers important conceptual contributions. First, the risk of intergenerational transmission of violence may be reduced in three-generation households, which was especially true among single parents. Such parents may relieve their parenting stress by living with their aging parents with whom they can share resources and/or help address child care needs. This overall benefit may exceed potential stress arising from closely interacting with the parents who may have been abusive or neglectful to them as children. Future research is needed to scrutinize and explain factors or contexts that are associated with the benefits of multi-generational living arrangements. Second, our findings emphasize the importance of examining complex family relationships across time and generations. Adults with a history of childhood family violence may have specific concerns and needs in their dyadic relationships with parents and offspring, which may look different in a multi-generational context (Brubaker, 1990). Future research may incorporate the family systems framework (Prest & Protinsky, 2007) to assess better how a history of childhood family violence affects family relationships and dynamics across and among multiple generations.

The lives of individuals are typically embedded in family relationships across the life course (Settersten, 2015), and despite childhood abuse and neglect, living in a three-generation household may help some parents with young children reduce their use of negative parenting practices. When assessing the effects of multi-generational living arrangements, practitioners should consider the specific context of these parent-adult children dyads, including their particular needs, expected familial roles, and SES statuses. Our findings inform practitioners and policymakers that parents with multiple socioeconomic disadvantages may rely heavily on a limited informal social network such as their parents as their major or sole source of social support. Helping the parents diversify and widen their social network and find more sources of support and resources can reduce the concentrated burden and stress in the family, preventing potential conflicts and adequately addressing their specific needs and concerns.