Intimate partner violence (IPV) is a global public health and safety issue (Black et al., 2011; Devries et al., 2013; El Sayed et al., 2020; Smith et al., 2018). Estimates reveal that over 30% of women worldwide experience some form of IPV victimization during their lifetime (Devries et al., 2013). Within the United States, over 40 million women and 30 million men experience IPV victimization (Smith et al., 2018). Additionally, Peterson et al. (2018) estimated the United States’ IPV lifetime economic burden to be just over $3 trillion. Being such a pervasive issue, it is important to gain a better understanding of the causes of IPV in order to inform policy and reduce the prevalence of such occurrences.

Much research has investigated risk factors associated with both intimate partner victimization and perpetration (Capaldi et al., 2012; Cui et al., 2013; Halpern et al., 2009; Ponti & Tani, 2019; Renner et al., 2015; Richards et al., 2017). In particular, research focused on family has frequently identified risk factors associated with experiencing later IPV. For example, studies have found that childhood maltreatment and abuse (Ørke et al., 2018; Paat & Markham, 2019; Renner et al., 2015; Richards et al., 2017), adverse childhood experiences (Fonseka et al., 2015), and inadequate parenting processes (Giordano et al., 2016; Hassija et al., 2018; Palazzolo et al., 2010; Ponti & Tani, 2019) are related to experiencing IPV. Prior research suggests a link between poor experiences within the family environment and negative behaviors throughout the life course (Gottfredson & Hirschi, 1990; Hirschi, 1969; Nye, 1958). While research has examined the relationship of varying family processes on IPV, relatively few studies have assessed how family structure may influence such outcomes (Fonseka et al., 2015; Paat & Markham, 2019; Pflieger & Vazsonyi, 2006).

Family structure has been found to be related to delinquency, with individuals from nontraditional family structures more likely to engage in delinquent behaviors (Kierkus & Hewitt, 2009; Rebellon, 2002; Wells & Rankin, 1991). Much research has focused specifically on the influence of biological father absence, linking this absence to a variety of negative life outcomes (Demuth & Brown, 2004; Ermisch et al., 2004; Heard, 2007; Juby & Farrington, 2001; Luo et al., 2012; McLanahan et al., 2013). Literature suggests that father absence inhibits parental control over offspring, leading to adverse behavioral outcomes (Gottfredson & Hirschi, 1990; Hirschi, 1969; Nye, 1958; Reiss, 1951). Despite numerous studies examining the influence of father absence on negative life outcomes, little is known about the influence, if any, father absence during the developmental period has on IPV victimization and perpetration in adulthood.

As it pertains to IPV victimization, current literature offers little explanation on how father absence may influence this experience. Maas et al. (2010) found that quality bonding (i.e., attachment) to parents acted as a protective factor against teen dating violence victimization. This suggests that father absence may act as a risk factor for IPV victimization through the lack of sufficient attachment to the child. Alternate explanations for this potential relationship may be found in other criminological theories. Drawing on general strain theory (Agnew, 1992), father absence may be perceived as the removal of a positively valued stimuli or may impose undue strain on the lives of family members, particularly their offspring. In response, individuals may resort to coping mechanisms that are antisocial in nature, such as engaging in criminal behaviors or entering unhealthy romantic relationships.

The purpose of the current study is to examine the potential long-term association between timing of father absence on intimate partner victimization and perpetration. The study will use a nationally representative sample to assess how biological father absence during different periods of development is related to both IPV victimization and perpetration in adulthood. Developmental/life-course (DLC) criminology suggests that developmental pathways are affected by life events or transitions (Elder, 1998). The influence of transitions on developmental pathways is dependent on the age of the individual when the transition occurs. The age of the child when the father leaves may be consequential in experiencing IPV in adulthood. Father absence occurring during a particular development period could shift an individual’s life trajectory in a way that may result in an increased likelihood of being in situations where they may experience IPV. As Young et al. (2020) suggest, early life stressors—particularly those occurring before age six—may take place during salient developmental periods, hindering the formation of adequate bonding and socialization. When the formation of these bonds is disrupted, there may be an increased likelihood for individuals to engage in antisocial behaviors (Hirschi, 1969). These early stressors may continue to negatively influence an individual’s life by way of cumulative disadvantage (Moffitt, 1993; Young et al., 2020). For example, an individual may experience father absence prior to being born, which may bring on feelings of rejection (Bowlby, 1969). As a result, they may engage in various forms of antisocial behavior across the life course as a means of coping with those feelings of rejection (Agnew, 2002). Moreover, engaging in antisocial behaviors can restrict prosocial opportunities throughout the life course (Moffitt, 1993). Before exploring the relationship between father absence and IPV victimization and perpetration, it is important to assess the literature concerning father absence and IPV.

Parenting and Intimate Partner Violence

Introduced by Bowlby (1969), attachment theory posits that individuals develop an attachment to primary caregivers early in life and the quality of these attachments governs the development of future relationships, as well as how the individual views themselves. Variances in attachment style are a result of the quality of care provided by the caregiver (Wang & Stalker, 2016). Inadequate caregiving—characterized by a failure to provide and fulfill emotional and physical necessities—may result in myriad issues across the life course (Fonseka et al., 2015; Giordano et al., 2016; Hassija et al., 2018; Ørke et al., 2018). When a caregiver fails to provide a secure base for the child, such as through emotional distance, abuse, or unreliability, the child develops an insecure attachment (Ainsworth et al., 1978; Bowlby, 1969; Wang & Stalker, 2016). Early attachment patterns are cyclical, affecting relationship attachment throughout the life course—from caregiver, to peers, to intimate relationships (Wang & Stalker, 2016). Thus, adequate parenting in childhood is imperative to emotional well-being and healthy relationships across the life course (Simpson et al., 2011).

Factors related to parenting practices have frequently been linked to later IPV involvement by offspring (Capaldi et al., 2012; Maas et al., 2010; Pflieger & Vazsonyi, 2006; Thomas et al., 2021; Tyler et al., 2011). Research indicates that poor parent–child relationships are linked to an increase in IPV victimization and perpetration likelihood in adolescence and adulthood. Examining factors in relation to teen dating violence victimization, Maas et al. (2010) found that teens with lower levels of bonding to their parents were more likely to experience dating violence victimization, as well as exhibit externalizing behaviors. Perceived verbal aggression by parents has also been linked to later IPV. Palazzolo et al. (2010) examined how self-reported and perceived parental verbal aggression was related to IPV experiences of their young adult children. Findings indicated that perceived verbal aggression by the same-sex parent had a significant influence on IPV victimization and perpetration for sons and daughters.

Additionally, studies have reported associations between lack of support (Pflieger & Vazsonyi, 2006), maternal indifference (Hassija et al., 2018), inadequate monitoring (Howard et al., 2003), and low parental warmth (Tyler et al., 2011) with dating violence (Howard et al., 2003; Pflieger & Vazsonyi, 2006; Tyler et al., 2011) and IPV involvement (Hassija et al., 2018). Results from an investigation of the influence of adolescent alcohol exposure, attendance of religious service, and parental monitoring revealed that more frequent monitoring by parents was significantly related to a reduction in dating violence victimization likelihood (Howard et al., 2003). Using data from the Add Health, Tyler et al. (2011) found maltreatment and low parental warmth to be positively related to both dating violence victimization and perpetration in young adulthood for both males and females. Their results also revealed that delinquency acted as a mediator between low parental warmth and dating violence perpetration, suggesting that lack of warmth may drive youth to associate with deviant peers (Tyler et al., 2011). Perceptions of coldness or hostility from parents may act as a catalyst for antisocial behaviors and set a precedent for relationship expectations. Drawing on Bowlby’s (1969) attachment theory, the rejection sensitivity model suggests that when parents fail to fulfill physical and emotional necessities in childhood, youth develop expectations of analogous behaviors of rejection and hostility in all subsequent relationships (Downey & Feldman, 1996). The expectation of rejection leads to anticipatory anxiety, causing individuals to become hypervigilant of possible rejection. Any perceived slight or rejection may lead to an emotional or behavioral reaction (Downey & Feldman, 1996; Schwartz et al., 2006).

Along this line, the cyclical manner by which parent–child relationships affects future relationships suggests that the absence of a biological parent may be viewed as a rejection—negatively influencing intimate relationships throughout the life course. Indeed, when Dutton (1994) examined the relationship between borderline personality and spousal assault, paternal rejection was the only variable to maintain statistical significance. Taken together, research indicates that the quality of parent–child relationships is important in the development and quality of subsequent relationships. While much research has focused on parent–child relationship quality, there remains a lack of information regarding the influence of family structure, and more specifically, father absence, on IPV experiences.

Father Absence and Intimate Partner Violence

Social scientists have long questioned the influence family has on human development (Bowlby, 1969; Durkheim, 1888; Freud, 1909; Harris et al., 1998; Hirschi, 1969; Merton, 1968). As discussed previously, parenting processes influence subsequent relationships across the life course. Namely, the quality of parenting processes may be influenced by the family structure. For example, research suggests family disruption inhibits the ability of parents to effectively monitor youth, through both direct and indirect control, leading to a higher likelihood of delinquent involvement (Demuth & Brown, 2004; Gottfredson & Hirschi, 1990; Nye, 1958; Reiss, 1951). While much research has examined how these processes relate to antisocial behaviors such as delinquency, literature concerning how family structure relates to IPV experiences is scarce (Halpern et al., 2001, 2009; Paat & Markham, 2019). Those who have taken family structure into consideration have found mixed results. Halpern et al. (2001) found nontraditional family structures to increase the likelihood of domestic violence victimization in adolescent boys. In particular, the absence of a father figure increased the odds of boys experiencing psychological victimization. In a later study, Halpern et al. (2009) examined IPV victimization from adolescence into young adulthood and found that continuous victimization and onset of sexual debut was associated with a nontraditional family structure. Paat and Markham (2019) discovered that residing in a two-parent household reduced the odds of experiencing domestic violence victimization or perpetration for college students. In contrast, Cui et al. (2013) found no family structure variables to be significantly related to persistent IPV between adolescence and young adulthood. Examining a range of early life adversity—including parental separation—Thomas et al. (2021) concluded that women with increased exposure were more likely to experience IPV in adulthood. These mixed results suggest that more research is needed on the influence of family structure on IPV experiences.

An area of family structure research that has garnered much attention is that of biological father absence, which has been found to be associated with a host of negative life outcomes (Ellis et al., 2003; Ermisch et al., 2004; Heard, 2007; Juby & Farrington, 2001; McLanahan et al., 2013; Stermac et al., 2002; TenEyck et al., 2021; Wells & Rankin, 1991). Despite this, little is known about the effect father absence has on IPV victimization and perpetration. Stermac et al. (2002) found absence of a father figure during childhood to be related to adult sexual assault victimization in women. Lack of a father figure was also significantly related to IPV perpetration in the form of emotional abuse in a sample of Sri Lankan men (Fonseka et al., 2015). Additionally, assessments of father-child relationship have generally found an association between paternal processes, domestic violence, and sexual behaviors (Alleyne-Green et al., 2015, 2016; Katz & van der Kloet, 2010). Paternal closeness was found to reduce the likelihood of intimate partner victimization for Black male adolescents (Alleyne-Green et al., 2015). For adolescent Black girls, paternal involvement was related to a decrease in the likelihood of engaging in risky sexual behaviors but was not related to domestic violence victimization (Alleyne-Green et al., 2016). The researchers suggested that this finding may be due in part to how involvement was measured. Using two cohorts, one from the United States and one from New Zealand, Ellis et al. (2003) examined how timing of father absence influenced daughters’ risk of risky sexual behavior. Findings indicated that compared to girls whose fathers were always present, pregnancy rates for girls whose fathers left between birth and age five were seven to eight times higher.

Taken together, current research suggests that father absence may be a risk factor for experiencing IPV during the life course. Criminological theory—specifically strain theory—may also offer some explanation for why father absence may be related to later IPV. Agnew’s (1992) general strain theory holds that individuals experience strain in various aspects of their life. Agnew (1992) noted that a major focus of strain theories are relationships with others. In particular, it is how these relationships negatively influence the individual. Strain thus results from a perceived wrongdoing by others. According to strain theorists, when strain is impactful enough, it will lead to antisocial behavior. One of the major sources of strain identified by Agnew (1992) is the actual or anticipated blockage from a positively valued goal/stimuli. Under general strain theory, blockage from a positively valued goal/stimuli does not solely refer to one that is economic. It could be from not receiving a promotion at work, not getting a particular grade on an assignment, or being rejected by a potential friend group—or in this instance, the perceived rejection by a biological father. Drawing on this, father absence may be related to IPV perpetration through the removal of a positively valued stimuli, the perception of rejection, and the subsequent coping through antisocial avenues. The strain of the father leaving at any point in time may cause individuals to engage in IPV perpetration as a means of making another individual feel a similar level of pain and frustration they felt from their father’s absence or as a way to emotionally alleviate themselves of the negative feelings surrounding their father’s absence.

In terms of IPV victimization, strain from a father’s absence may lead to coping through lack of social support. Similar to the results found by Maas et al. (2010), Agnew (1992) argues that individuals are more likely to engage in antisocial coping when they do not have access to adequate conventional social support. This lack of support is believed to make it more difficult for these individuals to handle strain. Father absence may thus be related to IPV victimization through both the absence itself and/or the diminution of the ability of the remaining caregiver to provide adequate support.

Current Study

While considerable research has focused on risk factors related to IPV, little is known about what role father absence plays in these experiences and how the timing of father absence may influence IPV victimization and perpetration in adulthood. Life-course perspective suggests that transitions or life events may influence an individual in such a way that the trajectory their life follows may be altered. More specifically, the timing of the event will influence all subsequent events (Elder, 1998). Coupled with the concept of the cyclical manner by which early relationship attachments affect later relationship attachments, paternal absence at a particular point during the life course may alter an individual’s future relationship expectations and life course trajectory. Given this, the purpose of the current study is to assess how biological father absence in adolescence is related to IPV victimization and perpetration in adulthood. Using a nationally representative sample, we investigate whether the timing of paternal absence is associated with these outcomes. Importantly, we control for a lifetime measure of general criminal behavior/delinquency and a lifetime measure of victimization, as prior research has found an overlap between general offending and victimization and IPV victimization and perpetration (Ehrensaft et al., 2003; Kennedy et al., 2016; Lussier et al., 2009; Richards & Gillespie, 2021).

Methods

Data

The current study uses data from the National Longitudinal Study of Adolescent to Adult Health (Add Health; Harris, 2009; Harris et al., 2009). The Add Health is a multi-wave longitudinal study using a nationally representative sample of youth in grades 7 through 12 during the 1994–1995 academic year. Data collection began with a sample of 90,118 students from 145 middle, junior high, and high schools. From this sample, a subset of students was selected to complete in-home surveys. Cuban and Puerto Rican students, Chinese students, Black students with college-educated parents, and students with physical disabilities were deliberately oversampled. The resulting sample consisted of 20,745 students and 17,670 parents. During wave 1 data collection, students and their parents were interviewed and asked questions regarding demographic information, family structure, risk and protective factors, education, health, and peers. Wave 2 data collection took place 1 year later with the same sample of respondents, with the exception of the students who were in the 12th grade the year prior and the sample of physically disabled students. In 2001–2002, wave 3 data collection was completed. At the time of the interviews, the respondents were 18 to 26 years old. Data collection for wave 4 took place in 2008, when respondents were between the ages of 24 and 32. Wave 5 data collection was completed in 2018. Data from waves 1 through 4 will be used in the current study. Due to the oversampling during wave 1, some individuals had a higher probability of being included in the sample than others (i.e., not randomly selected). For this reason, survey weights will be used in all analyses.

An examination of missingness revealed that most of the missings were on one or more of the dependent variables (27%) as well as the criminal behavior and victimization measures (between 53 and 56%)—this is to be expected given the measures contained multiple scales across multiple waves collected at various points in the respondent’s life. Importantly, there were no systematic missing responses (i.e., the missing values were randomly distributed). Despite dropping these cases, a robust sample size remained. Because of this, only those respondents who had no missings at any wave were included in the final sample (i.e., listwise deletion was used). This resulted in a final sample of over 7000 individuals, which was more than sufficient for statistical power (see Barnes et al., 2020).

Measures

Dependent Variables

IPV Victimization

During wave 4, respondents were asked how often in the past year had their partner: threatened them with violence; slapped, hit, or kicked them; or insisted or forced sex upon them. Responses were coded such that 0 = never, 1 = this has not happened in the past year, but it did happen before then, 2 = once in the last year of the relationship, 3 = twice in the last year of the relationship, 4 = 3 to 5 times in the last year of the relationship, 5 = 6 to 10 times in the last year of the relationship, 6 = 11 to 20 times in the last year of the relationship, 7 = more than 20 times in the last year of the relationship. Scores for each item were summed together, with higher scores representing greater IPV victimization (α = 0.77).

IPV Offending

During wave 4, respondents were asked how often in the past year they had: threatened their partner with violence; slapped, hit, or kicked their partner; or insisted or forced sex upon their partner. Responses were coded such that 0 = never, 1 = this has not happened in the past year, but it did happen before then, 2 = once in the last year of the relationship, 3 = twice in the last year of the relationship, 4 = 3 to 5 times in the last year of the relationship, 5 = 6 to 10 times in the last year of the relationship, 6 = 11 to 20 times in the last year of the relationship, 7 = more than 20 times in the last year of the relationship. Scores for each item were summed together, with higher scores representing greater IPV offending (α = 0.71).

Key Independent Variables

Biological Father Absence

During the wave 1 in-home interviews, respondents were asked what age they were when their biological father last lived with them. No respondents reported being over age 18 when they last lived with their father. Responses were coded into a series of dummy variables capturing the following: father always present, father always absent (i.e., the father left before birth), father left between birth and age 5, father left between ages 6 and 13, father left between ages 14 and 18. While these age groups may seem arbitrary, the same groupings have been used in prior research on father absence (Markowitz & Ryan, 2016; TenEyck et al., 2021). The father always present group will serve as the reference group for the analyses.

Covariates

Criminal Behavior

During wave 1, respondents were asked questions pertaining to delinquent involvement in the past 12 months. Respondents were asked about delinquent behaviors such as painting graffiti, lying to their parents, theft, damaging property, fighting, and selling drugs. The same questions were asked in wave 2, with the inclusion of questions about carrying a weapon or using a weapon in a fight. During waves 3 and 4, similar questions were asked, with the inclusion of questions about writing bad checks and using someone’s credit card without their permission. Some questions differed in order to be relevant to the respondent’s age. To construct the delinquency variable, a composite score was created from the delinquency scales constructed in all four waves (α = 0.89).

Victimization

During waves 1 and 2, respondents were asked if they had seen someone be shot or stabbed, had a gun or knife pulled on them, been shot, been cut or stabbed, or been jumped in the past 12 months. Responses were coded such that 0 = never, 1 = once, and 2 = more than once. During wave 3, respondents were asked if, in the past 12 months, they witnessed someone be shot; had a gun pulled on them; had a knife pulled on them; were shot; were stabbed; were beaten without anything being stolen; or were beaten and had something stolen. During wave 4, respondents were asked if, in the past 12 months, they had something worth more than US $50 stolen from them; witnessed someone be shot or stabbed; a knife or gun pulled on them; been shot or stabbed; been hit, choked, or slapped; or been beaten. Responses from waves 3 and 4 were coded dichotomously such that 0 = no and 1 = yes. To construct the victimization variable, a composite score was created from the victimization scales constructed in all four waves (α = 0.82).

Father Figure

During wave 1, youths were asked to provide a roster of individuals that made up their household and to best describe their relationship with each individual. To construct the variable “father figure,” a summated measure was created from all non-biological male figure responses including stepfather, adoptive father, step/adoptive father, and foster father.

Paternal Death

During wave 1, respondents were asked if their biological father was still living. The item was coded so that 0 = yes and 1 = no.

Parental Permissiveness

During wave 1, respondents were asked if their parents allowed them to make their own decisions regarding bedtime, curfew, friends, clothes, diet, and how much television they watched and what they watched. Responses were coded dichotomously such that 0 = no and 1 = yes. Items were summed together, with higher values indicating more parental permissiveness (α = 0.64).

Low Paternal Attachment

A measure of biological father attachment was created by summing four variables—taken from the wave 1 interviews—asking if respondents (1) knew anything about their biological father, (2) stayed overnight with their biological father in the last 12 months, (3) communicated with their biological father in the last 12 months, and (4) how close they felt to their biological father. All items were coded so that higher scores indicated lower attachment (α = 0.67).

Low Maternal Attachment

Maternal attachment was measured in wave 1. Respondents were asked how much they felt their mother cared for them and how close they felt to their mother. Responses were coded such that 1 = very much, 2 = quite a bit, 3 = somewhat, 4 = very little, and 5 = not at all (α = 0.69).

Maternal Disengagement

During wave 1, respondents were asked the degree to which they agreed to statements regarding the warmth of their mother, their mother’s encouragement and understanding, and their satisfaction with their communication and relationship with their mother. Responses were coded such that 1 = strongly agree, 2 = agree, 3 = neither agree nor disagree, 4 = disagree, and 5 = strongly disagree. Items were summed together with higher scores indicating higher maternal disengagement (α = 0.84).

Low Socioeconomic Status

During wave 1, respondents were asked if their mother received welfare. Responses were coded dichotomously such that 0 = did not receive welfare and 1 = received welfare.

Low Social Support

Seven items were used to create the scale for low social support. During wave 1, respondents were asked if they had fun with their families, if their family understood them, how much their family paid attention to them, and how much they felt friends and adults cared about them. Responses were coded such that 5 = not at all, 4 = very little, 3 = somewhat, 2 = quite a bit, and 1 = very much. Items were summed together, with higher scores indicating lower social support (α = 0.79).

Control Variables

Age

Age was measured during wave 4. Responses ranged from 24 to 34.

Race

Race was measured at wave 1 and was coded as a series of dummy variables where 1 indicated the respondent was either Black, Hispanic, or White (0 = no and 1 = yes).

Male

Respondent’s biological sex was obtained during wave 1. Responses were coded such that 0 = female and 1 = male. Descriptive statistics for all variables used in the current study can be found in Table 1.

Table 1 Descriptive statistics (n = 7319)
Full size table

Analytic Plan

Of interest is whether the timing of a father’s departure during the developmental period is related to IPV victimization and perpetration in adulthood. To examine these research questions, the analysis will unfold in a series of seven steps. In steps one through three, a series of negative binomial regression models will be assessed to examine the relationship between timing of father absence and IPV victimization in adulthood. The first step is to estimate a negative binominal regression model examining the relationship between IPV victimization and absent father timing. The second step to the analysis is to re-estimate the association between IPV victimization and absent father timing after controlling for the influence of additional covariates. The third step examines the relationship between IPV victimization and absent father timing after accounting for the influence of covariates and controls.

Steps four through six examine the relationship between father absence and IPV offending in adulthood using a series of negative binominal regression models. Specifically, the fourth step is to estimate a negative binominal regression model examining the association between IPV offending and absent father timing without considering the influence of additional variables. Step five is to re-estimate the association between IPV offending and the absent father timing variables after controlling for the influence of additional confounders. The sixth step entails examining the relationship between IPV offending behaviors and absent father timing after accounting for the influence of covariates and controls variables.

For step seven, difference in coefficient tests will be conducted to determine if the timing variables assessed differ significantly from one another. While results for one or more timing variables may be significantly (or not significantly) related to the outcome of interest (i.e., IPV victimization and perpetration), this does not divulge whether the timing variables differ from one another. The difference in coefficients test will allow for this distinction.

Negative binomial regression is utilized because the dependent variables—IPV victimization and offending—are skewed count data, and negative binomial regression accounts for the overdispersion (Long, 1997). Incident risk ratios (IRR) will be used to interpret the magnitude of the effect. Incident risk ratios can be calculated by converting the coefficient estimates from the negative binomial model and exponentiating them: IRR = e ßtk, where ßtk represents the estimated relationship between covariate k and Y at time t. This allows for it to be interpreted as a percentage change in the rate of the dependent variable as a function of a one-unit change in the independent variable. As an aid for interpretation, an IRR of 1.00 indicates no association, an IRR below 1.00 indicates a negative association, and an IRR above 1.00 indicates a positive association.

Results

Table 2 presents the results estimating the relationship between timing of father’s absence and IPV victimization. This question was examined in a stepwise manner with model 1 including independent variables, model 2 adds covariates, and model 3 is the full model with the inclusion of the control variables. Recall that five categories represent timing of father’s absence: (1) father always present, (2) father always absent, (3) father left between birth and age 5 (i.e., father left early), (4) father left between ages 6 and 13 (i.e., father left mid), (5) father left between ages 14 and 18 (i.e., father left late). All results presented hereafter utilize father always present as the reference category.

Table 2 Father absence timing and intimate partner violence victimization (n = 7319)
Full size table

Model 1 includes only the independent variables (i.e., timing of father’s absence) and the dependent variable, IPV victimization, in adulthood. In model 1, father always absent is positive and significantly related to IPV victimization. Specifically, the coefficient for father always absent (IRR = 1.58, p < 0.05) indicates that when children have a biological father that is completely absent, there is a 58% increase in the rate of IPV victimization during adulthood, in reference to father always present. Father absence occurring between ages 14 and 18 is related to a 63% increase in the rate of IPV victimization during adulthood (IRR = 1.63; p < 0.05).

Model 2 reports the relationship between the timing of father’s absence and intimate partner victimization, while controlling for additional covariates. As can be seen, father always absent (IRR = 1.37; p < 0.05) is associated with a 37% increase in experiencing IPV victimization as an adult. Criminal behavior (IRR = 1.01; p < 0.05) and victimization (IRR = 1.18; p < 0.05) are significant and positive. Low social support is also significant with results indicating that a one-unit increase in low social support is related to a 3% increase in the rate of intimate partner victimization (IRR = 1.03; p < 0.05).

Model 3 reports findings for the full model, including the independent variables, covariates, and control variables. Criminal behavior (IRR = 1.01; p < 0.05) and victimization (IRR = 1.16; p < 0.05) are significant and positive. Low social support remained significant with a one-unit increase in low social support being associated with a 3% increase in the rate of intimate partner victimization (IRR = 1.03; p < 0.05). The only demographic characteristic significantly related to intimate partner victimization in adulthood is being male, with results indicating that males are more likely to experience IPV victimization in adulthood by 29% (IRR = 1.29; p < 0.05).

Results reported in Table 3 examine the relationship between timing of father’s absence and intimate partner offending. Similar to Table 2, this question was examined in a stepwise manner. As can be seen in model 1, father always absent (i.e., up to the age of 18), father left early (i.e., between birth and age 5), and father left mid-adolescence (i.e., between ages 6 and 13) are positive and significantly related to intimate partner offending in adulthood. Specifically, father always absent is related to a 75% increase in intimate partner offending (IRR = 1.75; p < 0.05), father absence occurring between birth and age 5 is related to a 52% increase in being an intimate partner offender (IRR = 1.52; p < 0.05), while father absence occurring between ages 6 and 13 is related to a 43% increase in IPV offending (IRR = 1.43; p < 0.05).

Table 3 Father absence timing and intimate partner violence offending (n = 7319)
Full size table

Model 2 reports the relationship between when father absence occurs and IPV offending in adulthood, while adjusting for additional covariates. Father always absent (IRR = 1.60; p < 0.05) and father left early (IRR = 1.49; p < 0.05) are positive and significantly related to IPV offending in adulthood. Criminal behavior is significant, with results showing that a one-unit increase is related to a 3% increase in the rate of intimate partner offending (IRR = 1.03; p < 0.05). A one-unit increase in victimization (IRR = 1.06; p < 0.05) was associated with a 6% increase in intimate partner offending. Experiencing the death of a biological father was related to a 46% decrease in the rate of intimate partner offending (IRR = 0.54; p < 0.05). Low SES is related to a 53% increase in the rate of intimate partner offending (IRR = 1.53; p < 0.05). A one-unit increase in low social support is associated with a 4% increase in the rate of intimate partner offending (IRR = 1.04; p < 0.05).

Model 3 reports the findings for the full model. The results for this model indicate that father always absent increases the rate of intimate partner offending by 35% (IRR = 1.35; p < 0.05). Criminal behavior (IRR = 1.03; p < 0.05) and victimization (IRR = 1.09; p < 0.05) remain positive and significant with a one-unit increase in criminal behavior being related to a 3% increase in the rate of intimate partner offending and a one-unit increase in victimization increasing the rate by 9%. Paternal death was related to a 49% decrease in the rate of intimate partner offending (IRR = 0.51; p < 0.05). Low attachment to one’s biological father (i.e., paternal attachment) was significant with a one-unit increase being associated with a 13% decrease in the rate of intimate partner offending (IRR = 0.87; p < 0.05). Low SES is associated with a 58% increase in the rate of intimate partner offending (IRR = 1.58; p < 0.05), while a one-unit increase in low social support is related to a 5% increase in the rate of intimate partner offending (IRR = 1.05; p < 0.05). Age is significant, with results showing that a one-unit increase in age reduces the risk of intimate partner offending by 5% (IRR = 0.95; p < 0.05). White respondents are 29% less likely to report being an intimate partner offender (IRR = 0.71; p < 0.05) and being male is associated with 48% reduction in the risk of intimate partner offending in adulthood (IRR = 0.52; p < 0.05).

In order to further examine if there were differential relationships between periods of timing of father absence and IPV victimization and perpetration, a coefficient test was applied to the key independent variables in model 3 in both Tables 2 and 3. The purpose was to determine if there were any differences among the groups representing when father’s absence occurred (i.e., always absent, early, middle, late). Results revealed that there were no statistically significant differences among the coefficients measuring timing of father’s absence for either IPV victimization or offending (p > 0.05). This suggests that timing of father absence does not appear to be differentially related to IPV victimization or perpetration.

Discussion

Attachment theory suggests that inadequate attachments to caregivers in childhood may prove detrimental to future relationships (Ainsworth et al., 1978; Bowlby, 1969). The influence of relationships with caregivers has received much attention within the criminological literature (Gottfredson & Hirschi, 1990; Hirschi, 1969; Paat & Markham, 2019). In particular, the short- and long-term consequences of paternal absence on offspring behavior has been the focus of numerous studies, with many finding paternal absence to negatively affect children across the life course (Ermisch et al., 2004; Luo et al., 2012; TenEyck et al., 2021). Relatively little is known, however, about how father absence may influence IPV experiences.

The purpose of the current study was to use a nationally representative sample to investigate if the timing of father absence in adolescence was related to IPV victimization and perpetration experiences of offspring during adulthood. Overall, a few major findings were revealed. First, having a father who was absent since birth was related to intimate partner perpetration in adulthood—after controlling for lifetime measures of general criminal behavior, general victimization, and a host of additional theoretically informed covariates including parental variables and whether a father figure lived in the home. Second, there was no clear relationship between the timing of father absence during different developmental periods and either intimate partner victimization or perpetration in adulthood. Finally, father’s absence during any of the developmental periods examined was not significantly related to IPV victimization in adulthood. The findings suggest that forming any attachment with one’s biological father, regardless of periods of absence, may be more beneficial than forming no attachment at all.

Conventional criminological theory may provide a guide for understanding the results regarding intimate perpetration in adulthood. Drawing on general strain theory, father absence may be a source of strain resulting from perceived wrongdoing or rejection (Agnew, 1992). This strain may cause offspring to cope in antisocial manners, such as through the perpetration of IPV. Additionally, recall that early criminologists held parents as the primary agents of control over youth’s behavior (Hirschi, 1969). As such, absence of the parent may inhibit the development and maintenance of effective control, contributing to a youth’s involvement in delinquent and antisocial behaviors. Given that paternal absence in childhood appears to increase risk of IPV perpetration in adulthood, it may be that parental control on the part of the father failed to establish. When the biological father is never present, there may be fewer opportunities for the child to attach to the father and for the father to establish parental control.

However, results also indicate that low attachment to the non-resident biological father was related to a decrease in the risk of perpetrating IPV in adulthood, even after controlling for criminal behavior, victimization, and a host of other covariates. While it is in the opposite direction one would expect, this may be due to several reasons. First, according to Hirschi (1969) and Sampson and Laub (1993), the likelihood of antisocial behavior is reduced through quality bonding to prosocial others, while weak or nonexistent bonds may increase antisocial behavior likelihood. As it pertains to the observed results, although the level of attachment to the biological father was low, it may be that the bond was strong enough to impart a small influence on the offspring’s behavior, resulting in risk reduction. Second, it may be that the biological fathers who no longer reside with their offspring are individuals who were poor parental models and therefore, it was more beneficial for the offspring to have less exposure to them. Furthermore, it could be that the limited exposure to their biological father allowed the offspring to witness behaviors antithetical to their own moral standards, resulting in a conscious decision to act in a manner that is in opposition to that of their father.

Importantly, the current study is one of the first to examine the relationship between father absence timing and IPV victimization and perpetration while controlling for general criminal behavior and victimization throughout the life course. It is important to include a lifetime measure of general criminal behavior and victimization in IPV research as there is often much overlap between being a victim and/or perpetrator of general crime while also experiencing IPV (Richards & Gillespie, 2021; Tanskanen & Kivivuori, 2021; Tillyer & Wright, 2014). Most research, however, categorizes IPV victimization and perpetration as separate entities from general criminal behavior and victimization. While general victimization and offending were both related to intimate partner victimization and perpetration, having a biological father who was always absent remained significantly associated with IPV offending despite controlling for lifetime measures of criminal behavior and victimization.

Although the findings of this study add to the current knowledge of father absence and IPV, there are limitations that should be noted. First, there is a potential that residual confounding is not accounted for in the study, which may influence the observed relationships. While it is beyond the scope of the current study to control for all potential confounders, several theoretically informed variables were included. For example, research has found that general victimization and offending are correlated with IPV victimization and offending (Tanskanen & Kivivuori, 2021; Verbruggen et al., 2022). As mentioned previously, we thus controlled for previous and current victimization and offending—which, to our knowledge, has not been done before within this research context. Second, the measure of father absence used refers only to the father no longer residing in the same household as the child. While this measure may not assess full absence of a father, this measurement has been used in previous studies on father absence (Markowitz & Ryan, 2016; TenEyck et al., 2019). Additionally, the dichotomous measure of father’s absence may not fully capture the relationship, or role, the father may have in the child’s life. For example, a father may provide emotional support without residing in the same household as his child. While only examined partially here, future research should examine whether the reason for fathers’ absence during the developmental period differentially influences IPV victimization and perpetration in adulthood.

While much research has investigated the influence that varying family processes has on both intimate partner victimization and perpetration (Fonseka et al., 2015; Giordano et al., 2016; Hassija et al., 2018; Ørke et al., 2018; Paat & Markham, 2019; Palazzolo et al., 2010; Ponti & Tani, 2019), less is known about family structure and whether it also influences IPV. To foster knowledge in this area, the present study examined timing of father’s absence during the developmental period and whether IPV victimization and perpetration was experienced in adulthood. Results indicate that while father’s absence during a specific time period was not related to IPV victimization and perpetration, father’s complete absence from his offspring’s life was consistently related to experiences with intimate partner perpetration in adulthood. To expand, those who always had an absent father while growing up reported higher rates of being a perpetrator of IPV in adulthood.

These findings highlight the importance of considering family structure when examining IPV and the need to further examine family structure to better understand contributing factors for intimate partner victimization and perpetration. Furthermore, these findings point to the need for further research on both proximal and distal consequences of life events occurring during or before adolescence. Policies and practices focused on early prevention could be key to reducing the prevalence of IPV. Such preventive programming may be geared toward youth while providing education on IPV, alternative techniques for addressing anger, and the inclusion of resources for those experiencing IPV to seek help. Programming focused on intervention should also be implemented, as research has found that such programming may be promising for reducing subsequent abuse (Wong & Bouchard, 2020). Additionally, interventions which provide aid and resources to single-parent households may help to reduce subsequent disadvantages and household stress, while also providing the foundation for families to build stable environments that foster the building and maintenance of healthy relationships. Finally, programming directed toward fathers and mothers before and after the birth of the child should be developed. Access to psychoeducation which emphasizes healthy relationship sustainability and parenting skills is suggested. Such programming may provide long-term benefits for youth behavior and well-being as successful program outcomes may instill adequate expectations of healthy relationships early for youth.