Transitioning From Victim to Perpetrator: Testing Direct and Mediation Effects

Abstract

Some criminologists have focused on understanding the role of childhood experiences on an individual’s likelihood of offending during adulthood. Although research has consistently linked these two phenomena, little is known about the extent to which childhood victimization influences later criminal behavior. The current study analyzed self-reported data to examine the abused–abuser relationship in the context of intimate relationships. Specifically, the study examined the direct effects of victimization across three types of offending, and whether or not individual characteristics in such a relationship were significant mediators. Using a multivariate OLS regression, a significant relationship was observed between childhood victimization and adulthood offending. Likewise, results revealed that individual characteristics significantly mediated this relationship. Policy implications of these observations are discussed.

Keywords

mediation victimization criminality offending intimate violence childhood and adulthood

Introduction

The relationship between victimization and offending has been an oft-studied topic within criminology (see Gottfredson, 1981; Jennings, Piquero, & Reingle, 2012; Lauritsen & Laub, 2007; Lauritsen, Sampson, & Laub, 1991). While some studies have focused on the victim–offender overlap by assessing factors that account for an offender’s propensity to become a victim (Broidy, Day, Crandall, Klar, & Jost, 2006; Jennings et al., 2012; Maldonado-Molina, Piquero, Jennings, Bird, & Canino, 2009), others have concentrated on explaining the etiology of offending by looking at an individual’s victimization history. There is some research showing that victims can become offenders later in life, drawing on social learning theory (modeling of behavior) and general strain theory (using the strain of being victimized and then coping with aggression) (Cohen et al., 2002; Ogloff, Cutajar, Mann, & Mullen, 2012). A recent meta-analysis by Jespersen et al. (2009) revealed a significant connection between experiencing past sexual abuse and engaging in sexual offending behavior. This and similar studies lend credence to the abused–abuser hypothesis.

The abused–abuser hypothesis has been mainly tested in sexual assault situations, with the vast majority of other offenses excluded. Accordingly, the current study empirically examines the influence of childhood victimization on adulthood offending in the context of intimate partner violence (IPV). A focus on the lasting adverse consequences of childhood victimization is important for a variety of reasons, but mainly because of the devastating effects of victimization—especially sexual assault—on children (Afifi & Macmillan, 2011; World Health Organization, 2005). For example, according to the World Health Organization, children who are exposed to IPV have an increased likelihood of engaging in violent behavior as children and later as adults as compared to those who have not been exposed to IPV. Similarly, Afifi and Macmillan (2011) identified numerous adverse consequences of victimization such as poor performance in school, aggression, and criminal conduct. These consequences set the stage for problematic outcomes throughout adolescence and into adulthood. In the past, the issue of the abused becoming abusers has been approached from different methodological perspectives, including longitudinally (Salter et al., 2003), cross-sectionally (Aebi, Landolt, Mueller-Pfeiffer, Schnyder, Maiser, & Mohler-Kuo, 2015; Schraufnagel, Davis, George, & Norris, 2010), and through systematic reviews and meta-analyses (Jennings et al., 2012; Stith, Rosen, Middleton, Busch, Lundeberg, & Carlton, 2000).

The present study utilizes a cross-sectional methodological approach to assess the effects of childhood exposure to violence or abuse on the likelihood of engaging in offending as young adults. Specifically, the study achieves the following objectives. First, it examines the direct effect of childhood victimization on adult offending behavior. Second, it examines the mediation effects of demographic characteristics and other variables in the relationship between victimization and offending. This exploration is done across three major types of intimate relationship violence. By focusing on violence in intimate relationship (IRV), this study contributes to the fields of victimology and criminology in several meaningful ways. For instance, earlier research on the abused–abuser hypothesis has mainly focused on understanding whether sexual assault victimization can lead to sexual offending. Although this work has been useful, it provides limited knowledge of the relationship between offending and victimization. There are other types of victimization that individuals may experience during childhood, and one cannot rely on the current literature to know whether non-sexual victimizations will lead to violence/offending behavior later in life. The current study addresses this limitation by focusing on three main types of IRV: sexual assault, physical assault, and emotional/psychological abuse. Furthermore, this study expands on existing efforts by drawing attention to the important role of other variables in mediating the victimization–offending connection. Findings from this endeavor may serve as a springboard for developing practical policies to address a great threat to society—IRV.

Theoretical Foundation: Social Learning and the Abused–Abuser Relationship

Several developmental, psychological, and criminological theories have been developed to explain the abused–abuser hypothesis. These theories include attachment theory, which suggests that there is a positive correlation between childhood insecure attachment and negative behavioral outcomes such as aggression, violent, and criminal offending (Chakhssi, Ruiter, & Bernstein, 2013; Ford, Chapman, Connor, & Cruise, 2012; Grady, Levenson, & Bolder, 2017; Miner, Romine, Robinson, Berg, & Knight, 2014; Park, 2016; Shillkret & Shillkret, 2011); control theory, which uses power and control as precursors of violence, where more powerful individuals are able to use force or threats to secure obedience from less powerful individuals, mostly children and women (Bostock, Auster, Bradshaw, Brewster, Chapin, & Williams, 2002; Goode, 1971; Loseke, 2005); and social learning theory, which proposes that violence and aggression are learned behaviors (Fite et al., 2008; Malley-Morrison & Hines, 2004). All these theories are convincing and have been empirically tested and examined. However, this study focuses on social learning theory as its theoretical framework because of the generalizability of the theory’s propositions. In fact, social learning theory posits that violence and aggression are learned through imitation, role modeling, and observation can apply to all contexts and crime types (see McCloskey & Lichter, 2003), and it is this potential that makes the theory appealing for the current study.

Social learning theory argues that behavior is learned, and that people model their behavior based on what they observe from others (Bandura, 1973, 1977; Burgess & Akers, 1966). Children imitate behaviors they observe their family members and caregivers engaging in or are exposed to themselves; thus, there is the possibility for learning and modeling both vicariously and directly. According to this theory, violence and aggression are learned through role modeling provided by these individuals either directly or indirectly (Malley-Morrison & Hines, 2004). Malley-Morrison & Hines, 2004 believed that people become violent toward others, especially family members, because they learn violent conduct through operant conditioning and observation. They suggest that children who grow up in violent familial or non-familial environments where people respond to stress and frustration with anger, aggression, and violence have an increased risk of engaging in similar behavior as adults (Malley-Morrison & Hines, 2004). Proponents of this theory believe that family members or caregivers not only expose children to violence and its associated techniques but also train children to accept or approve the use of violence as a normal practice (Fite et al., 2008; Gelles, 1972; Hare et al., 2009).

The general propositions of this theory can be summarized as follows: first, individuals (children) will commit their first violent acts as a result of a constellation of factors, such as internalizing social definitions that approve the acts and perceptions of reward and punishment, especially how their own offenses are rewarded or punished (Akers, 1998). Second, people will continue to offend due to their desire to overcome issues associated with their own victimization experiences, such as trauma, stress, anxiety, emotional disturbances, and addiction (Veneziano, Veneziano, & LeGrand, 2000).

Several empirical studies provide support for the abused–abuser hypothesis, arguing that experience of violence/victimization as a child is related to general patterns of aggressive/violent behavior, both as a child and as an adult (Aebi et al., 2015; Burton, 2003; Burton, Miller, & Shill, 2002; DeLisi, Kosloski, Vaughn, Caudill, & Trulson, 2014; Glasser, Kolvin, Campbell, Glasser, Leitch, & Farrelly, 2001; Jespersen, Lalumiere, & Seto, 2009; Nunes, Hermann, Malcom, & Lavoie, 2013; Ogloff et al., 2012; Simons, Wurtele, & Durham, 2008; Veneziano et al., 2000; White & Widom, 2003). For example, Jespersen et al. (2009) reviewed several studies that compared sex offenders to non-sex offenders and found a significantly higher rate of past sexual abuse among sex offenders than among non-sex offenders. Similarly, Glasser et al. (2001) adopted a retrospective approach to identify whether victims of child sexual abuse would become offenders of such behavior. The authors reviewed clinical case notes involving 843 individuals and reported that male victims of child sexual abuse had a higher risk of becoming child sexual offenders later in their life. Ehrensaft, Cohen, Brown, Smailes, Chen, and Johnson (2003) conducted a prospective study involving a randomly selected sample of children and their mothers to test the effects of exposure to domestic violence between parents; the authors found that observing violence between parents predicted perpetration of partner violence in later life.

Two recent studies also provide empirical support for the relationship between victimization and offending within the context of sexual abuse/assault. Aebi et al. (2015) analyzed self-reported data from a nationally representative sample of more than 6600 ninth-graders in Switzerland, which offered credence to the argument that children exposed to violence are more likely to exhibit violent behavior themselves. Specifically, the authors observed that victims of contact sexual abuse and non-contact sexual abuse were significantly more likely to report coercive sexual behaviors. This observation was true for both male and female students. Moreover, DeLisi et al. (2014) examined a large dataset containing information on more than 2500 adjudicated juvenile male offenders incarcerated in a major southern state facility to identify potential effects of childhood sexual abuse on sexual offending and other forms of crime. Consistent with other studies, the authors observed a strong effect of victimization on sexual offending and a marginal effect on other forms of crime such as homicide and property crimes.

The above studies help to explain the etiology of offending by providing support for the abused–abuser hypothesis. Still, there may be ample opportunity for other correlates of offending to mediate the relationship between victimization and offending. Many scholars believe that certain factors serve as mediators of the victim-to-offender cycle, and such factors play a significant role in determining whether an abused person will proceed to become an abuser (Glasser et al., 2001; Lambie, Seymour, Lee, & Adam, 2002; Lisak, Hopper, & Song, 1996). Glasser et al. (2001) observed the effect of the victim’s relationship with the perpetrator on a victim’s propensity to become an abuser. Similarly, Lambie et al. (2002) revealed that arousal related to abuse, masturbating and fantasizing about the abuse, and comfort or pleasure obtained from the abuse could largely affect a victim’s likelihood of victimizing others.

Additionally, some researchers have noted that not every person who is victimized becomes a victimizer, arguing for what is called the resilience thesis (Lambie & Johnston, 2016; Leach, Stewart, & Smallbone, 2016; Salter et al., 2003; Thomas & Fremouv, 2009; Widom & Ames, 1994). Researchers have variously found insignificant percentages of offenders with prior histories of abuse/victimization. For example, Salter et al. (2003) found that about 12% of the sample they studied became sexual abusers, with the majority refraining from offending. Similarly, Leach et al. (2016) conducted a prospective longitudinal birth cohort study of more than 38,200 males with a history of maltreatment and found that only about 3% of the sexually abused boys became sexual offenders.

The relationship between the abused and the abuser is more complex than most of the existing research suggests. It is not a straightforward relationship and given the fact that extant research examining the cycle of violence has extensively focused on sexual offending, there is room for further examination of this relationship across different crime contexts/forms. Based on the key propositions of the social learning perspective, the current study examines the abused–abuser connection across three major forms of IRV by assessing (1) direct effects of victimization on offending and (2) mediation effects of individual characteristics and exposure to sexual harassment in abused–abuser relationship. The following hypotheses are tested:

1. Individuals that were exposed to sexual violence or abuse as children will have greater likelihood of engaging in criminal behavior as an adult. With this hypothesis, it is expected that

a. exposure to sexual assault/abuse will increase one’s propensity to engage in sexual assault offending in intimate relationship.

b. Exposure to sexual assault/abuse will increase one’s propensity to engage in physical assault offending in intimate relationship.

c. Exposure to sexual assault/abuse will increase one’s propensity to engage in emotional abuse offending in intimate relationship.

2. Individuals that were exposed to physical violence or abuse as children will have greater likelihood of engaging in criminal behavior as an adult. With this hypothesis, it is expected that

a. exposure to physical assault/abuse will increase one’s propensity to engage in sexual assault offending in intimate relationship.

b. Exposure to physical assault/abuse will increase one’s propensity to engage in physical assault offending in intimate relationship.

c. Exposure to physical assault/abuse will increase one’s propensity to engage in emotional abuse offending in intimate relationship.

3. Individuals that were exposed to emotional abuse as children will have greater likelihood of engaging in criminal behavior as an adult. With this hypothesis, it is expected that

a. exposure to emotional abuse will increase one’s propensity to engage in sexual assault offending in intimate relationship.

b. Exposure to emotional abuse will increase one’s propensity to engage in physical assault offending in intimate relationship.

c. Exposure to emotional abuse will increase one’s propensity to engage in emotional abuse offending in intimate relationship.

4. We hypothesized that individual characteristics and later experiences of violence/abuse will mediate the relationship between victimization and offending.

Method

Participants and Procedures

The present study used a sample of college students randomly selected from a major public university in the southern United States. Data were obtained from the selected students in the fall of 2017, enrolled in the various courses offered across all the campuses of the university. Across all the campuses, the student population was about 56% female, and the majority was of Caucasian descent. Ethical clearance for the study was obtained from the institution’s Internal Review Board, and students who volunteered to participate in the study provided informed consent. After IRB approval was obtained, the researchers requested a random sample of students from a designated office, which is a body created by the institution to centralize procedures for requesting sample panels for survey research. The office created a list of randomly selected students from across all campuses, schools, and colleges of the institution. Initially, 1000 students were invited to take part in the online survey, but at the end, 537 students provided responses to the questions used to measure different aspects of their offending behavior and childhood victimization experiences. The survey also included items about respondents’ demographic characteristics. A description of the sample used in the study is provided in Table 1. The average age of the respondents was 22 years, with the youngest being 18 years and the oldest 55 years. Most of the respondents were female (65%), single (92%), and white (81%). Seventy-eight percent of them were undergraduate students, while the rest were either graduate or professional students.

Table 1.

Descriptive characteristics of respondents in the study (n = 537)¹.

Variables	M (SD)/%
Childhood Victimization Experiences (IPV)
Emotional abuse	8.20 (3.29)
Physical assault	2.19 (.73)
Sexual assault	1.02 (.22)
Adulthood Offending/Criminality (IPV)
Emotional abuse offending	3.71 (1.17)
Physical assault offending	7.58 (1.91)
Sexual assault offending	3.08 (.61)
Offender Characteristics
Age²	22.42 (5.24)
Gender (1 = female)	68
Marital status
Married	8
Single	92
Racial background
White	81
Black/African America	10
Hispanic/Latino	3
Other races	6
Living arrangement
Live alone	13
Live with roommate	66
Live with my partner	14
Live with parent/relative	7
Sexual orientation
Other	9
Heterosexual	91
Experiences of sexual harassment (as adult)³	4.61 (2.02)
I am
An undergraduate student	78
A graduate or professional student	22

Note. M = Mean; SD = Standard Deviation; % = percentages; 1 = Missing values were less than 5%; and 2 = minimum age was 18 years and maximum was 55 years; 3 = ranging from 3 to 12.

Measures

Outcome variables

Intimate partner offending was the main outcome variable in the study. This was assessed using 13 items adapted with modification from the National Intimate Partner and Sexual Violence Survey (2010) and the National Violence Against Women Survey (2000), which was sponsored by the National Institute of Justice. The items had the same lead-in question asking respondents to indicate how frequently they had committed certain offenses against their partners. Factor analysis provided support for three different forms of IPV offending.

Sexual offending was measured with three items: “Sexual touching against your partner’s will”; “Attempted sexual penetration (vaginal, anal, or oral intercourse) against your partner’s will”; and “Sexual penetration (vaginal, anal, or oral intercourse) against your partner’s will.” The factor loadings of these items ranged from 0.74 to 0.96, with an alpha value of 0.74, indicating good internal consistency for the scale.

Physical assault offending was assessed using seven items: “You put your partner down in front of family and/or friends”; “You pushed, grabbed, or shoved your partner”; “You slapped your partner”; “You kicked or bit your partner”; “You beat your partner up”; “You hit your partner or tried to hit your partner with something”; and “You choked your partner.” The items were summed to form a scale (factor loadings ranged from 0.67 to 0.84) with an alpha value of 0.86.

Three items were used to measure emotional abuse: “You insulted or swore at your partner”; “You threatened to hit or throw something at your partner”; and “You threatened your partner with a gun or a knife.” The emotional abuse scale (factor loadings ranged from 0.68 to 0.87) had an alpha of 0.44. All the items were assessed on a four-point scale with 1 = never, 2 = once or twice, 3 = 3 to 10 times, and 4 = more than 10 times.

Independent variables

The effects of three types of childhood victimization experiences were examined as predictors of the outcome variables. Items used to measure these forms of victimization asked respondents to report on their experiences before the age of 16 years. Sexual assault victimization was measured with a single item, asking respondents to indicate the number of times parents/caregivers forced them to have sex (vaginal, anal, or oral intercourse) against their will. Two items were combined to form the physical assault victimization scale: “Parent/caregiver choked you” and “Parent/caregiver beat you up.” Emotional abuse victimization was measured using six items: “You saw your parent/caregiver push, grab, or shove your other parent/other caregiver”; “You saw your parent/caregiver put your other parent/other caregiver down in front of family and/or friends”; “You saw your parent/caregiver insult or swear at your other parent/other caregiver”; “You saw your parent/caregiver beat up your other parent/other caregiver”; “You saw your parent/caregiver choke your other parent/other caregiver”; and “Parent/caregiver threatened you with a gun or a knife.” As indicated by a factor analysis, these items measured the same underlying construct (factor loadings ranged from 0.65 to 0.89) and were combined to form an additive scale with an alpha of 0.81, suggesting good internal consistency.

Mediating variables

Several individual characteristics were included in the models as mediators. Age was measured as a continuous variable asking respondents to indicate their age (in years) at the time of survey administration. Gender was dichotomously measured as 0 = male and 1 = female. Also, marital status was measured as 0 = married and 1 = single. Race was a categorical variable with 1 = white, 2 = Black, 3 = Hispanic, and 4 = other races. The “white” category was used as the reference category. Equally, respondents’ living arrangement was categorically assessed as 1 = live alone, 2 = live with roommate, 3 = live with my partner, and 4 = with parents. Living alone was the reference category. Respondents’ sexual orientation was assessed as 0 = other and 1 = heterosexual. Finally, the effect of one’s adulthood experiences was also assessed. Experiences of sexual harassment as an adult was measured using three items that delved into whether respondents had experienced sexual harassment from their following students: For the past 12 months, how often did you experience (1) unwelcome remarks of a sexual nature from student(s); (2) unwelcome touching from student(s); and (3) repeated pressure for dates or sexual activity from student(s). These items were assessed as 1 = never, 2 = once or twice, 3 = 3 to 10 times, and 4 = more than 10 times. These items loaded on the same factor, with factor loadings ranging from .84 to .90 and 75% total variance explained. Subsequently, their responses were summed up to form sexual harassment scale, with an alpha value of .83, suggesting good internal reliability.

Plan of Analysis

To test the study’s hypotheses, several analyses were conducted. First, univariate and multivariate analyses were conducted to examine the distribution of responses and to estimate the effects of the predicting variables. Specifically, given that the three dependent variables were continuous measures, a series of ordinary least squares regression models were estimated to determine the relative impact of the independent variables on the dependent variables while controlling for other variables. Further, collinearity analysis was conducted to find out whether there was collinearity in our data that would raise concerns about multicollinearity. As indicated by the results of this analysis, none of the VIF values was greater than 10 and the tolerance values were all above .02 (none was below or closer to 0.01).

Results

Effects of Childhood Exposure to Violence on Adult Offending

This study explored the effects of childhood exposure to violence on the likelihood of one committing violence later in life in the context of an intimate relationship using multivariate analyses to account for the effects of demographic background and sexual harassment. Three separate analyses were conducted to assess these effects across three types of offending. The results of the first analysis are presented in Table 2 and these estimate the effects on sexual assault offending. Model one contains only the three independent variables and the model was significant (F = 6.91, p <.001) and it explains six percent of the variance in sexual assault offending. Exposure to sexual victimization/abuse as a child significantly and positively influenced the likelihood of engaging in sexual offending in intimate relationship (t = 3.23, p < .01). In model 2, when other variables were introduced to account for their effects on sexual offending, exposure to sexual victimization remained a significant predictor of sexual offending, although the extent of its impact reduced slightly with the addition of marital status, living arrangement, and experiences of sexual harassment as adult. In addition to the effects of sexual victimization, marital status (single), living arrangement, and experiences of sexual harassment as an adult positively influenced sexual assault offending. Specifically, being single, living with a partner (compared to living alone), and having a prior experience of sexual harassment increase the likelihood of one engaging in sexual offending in an intimate relationship.

Table 2.

The Influence of Victimization on Sexual Assault Offending (n = 341).

	Model 1			Model 2
	b (SE)	Beta	t/F	b (SE)	Beta	t/F	Tolerance	VIF
Constant	2.34 (.30)		7.72^∗∗∗	.34 (.50)		0.68
Victimization variables
Sexual victimization	1.15 (.36)	0.21	3.23^∗∗	.65 (.28)	0.12	2.32^∗	.56	1.79
Emotional abuse	.01 (.03)	0.04	0.54	–.01 (.02)	−0.04	−0.78	.60	1.66
Physical abuse	.03 (.13)	0.02	0.22	.02 (.10)	0.01	0.20	.43	2.31
Control variables
Age				.01 (.01)	0.07	1.37	.59	1.71
Gender (female)				–.02 (.12)	−0.07	−0.17	.87	1.15
Race¹
Black				–.08 (.18)	−0.02	−0.48	.94	1.06
Hispanic				–.15 (.37)	−0.02	−0.40	.98	1.02
Other race				.37 (.23)	0.06	1.57	.93	1.08
Single				.51 (.25)	0.12	2.01^∗	.44	2.26
Living arrangement²
with roommate				.23 (.15)	0.09	1.52	.44	2.27
with my partner				.44 (.22)	0.13	1.99^∗	.35	2.85
with parents				.31 (.22)	0.06	1.39	.69	1.44
Sexual orientation				–.14 (.17)	−0.03	−0.85	.91	1.10
Experiences of sexual harassment as adult				.39 (.03)	0.68	15.41^∗∗∗	.77	1.29
Model fit			6.91^∗∗∗			24.65^∗∗∗
R²			0.06			0.51

Note. 1 = White race was the reference category; 2 = Living alone was the reference category. ^∗p<.05, ^∗∗p<.01,^∗∗∗p<.001.

Table 3 reports the estimates for the effects on physical assault offending. Model 1, which examines the effects of the three independent variables without any control variable was significant (F = 18.27, p <.001) and explains 14% of the variance in physical assault offending. Exposure to sexual victimization/abuse significantly and positively influenced the likelihood of engaging in physical assault offending in intimate relationship (t = 3.61, p<.001). Moreover, exposure to physical abuse predicted physical assault offending (t = 2.12, p<.05). In model 2, when other variables were included to account for their effects on physical assault offending, exposure to sexual and physical victimizations remained significant predictors of physical assault offending. However, the strength of their individual impacts reduced slightly with the addition of these variables. This model explained about 23% of the variance in physical assault offending.

Table 3.

The Influence of Victimization on Physical Assault Offending (n = 335).

	Model 1			Model 2
	b (SE)	Beta	t/F	b (SE)	Beta	t/F	Tolerance	VIF
Constant	4.19 (.52)		8.12∗∗∗	2.27 (1.13)		2.01∗
Victimization variables
Sexual victimization	2.20 (.61)	0.23	3.61∗∗∗	1.63 (.63)	0.17	2.59∗	.56	1.80
Emotional abuse	.03 (.04)	0.05	.78	.02 (.04)	0.04	.56	.60	1.66
Physical abuse	.47 (.23)	0.16	2.12∗	.38 (.22)	0.13	1.72∗	.43	2.31
Control variables
Age				.04 (.02)	0.09	1.42	.59	1.70
Gender (female)				–.41 (.26)	−0.08	−1.55	.86	1.16
Race¹
Black				–.65 (.42)	−0.08	−1.56	.94	1.06
Hispanic				–.33 (.83)	−0.02	-.40	.98	1.02
Other race				.58 (.53)	0.06	1.10	.93	1.08
Single				1.14 (.57)	0.15	1.98	.43	2.30
Living arrangement²
with roommate				–.13 (.33)	−0.03	-.38	.45	2.23
with my partner				.93 (.50)	0.15	1.84	.35	2.86
with parents				–.40 (.51)	−0.05	-.78	.70	1.43
Sexual orientation				.03 (.39)	0.00	.07	.91	1.10
Experiences of sexual harassment as adult				.25 (.06)	0.25	4.43	.78	1.28
Model fit			18.27^∗∗∗			6.72∗∗∗
R²			0.14			0.23

Note. 1 = White race was the reference category; 2 = Living alone was the reference category. ^∗p<.05, ^∗^∗p<.01, ^∗∗∗p<.001.

Finally, Table 4 presents the results of the analysis examining the influence of the exposure variables on emotional abuse offending. Model one contains only the three independent variables and the model was significant (F = 6.98, p <.001) and it explains 6% of the variance in emotional abuse offending. Exposure to sexual victimization/abuse significantly and positively influenced the likelihood of engaging in emotional abuse offending in intimate relationship (t = 2.38, p<.05). Further, emotional abuse exposure also significantly and positively influenced the likelihood of engaging in emotional abuse offending in intimate relationship (t = 2.97, p<.01). However, when other variables were introduced to account for their effects on emotional abuse offending in model 2, the effect of exposure to sexual victimization disappeared while only emotional abuse exposure remained a significant predictor of emotional abuse offending though the extent of its impact reduced slightly with the addition of living arrangement and experiences of sexual harassment as adult. In addition to this observation, living arrangement and experiences of sexual harassment as an adult positively influenced emotional abuse offending. Specifically, living with a partner (compared to living alone) and having a prior experience of sexual harassment increases the likelihood of one engaging in emotional abuse offending in an intimate relationship.

Table 4.

The Influence of Victimization on Emotional Abuse Offending (n = 336).

	Model 1			Model 2			Collinearity Statistics
	B (SE)	Beta	t/F	B (SE)	Beta	t/F	Tolerance	VIF
Constant	2.45 (.41)		5.96∗∗∗	1.10 (.90)		1.22
Victimization variables
Sexual victimization	1.15 (.49)	0.16	2.38∗	.80 (.51)	0.11	1.59	.56	1.80
Emotional abuse	.10 (.03)	0.20	2.97∗∗	.09 (.04)	0.18	2.69∗∗	.60	1.66
Physical abuse	–.19 (.17)	−0.08	−1.08	–.23 (.18)	−0.10	−1.31	.43	2.31
Control variables
Age				.01 (.02)	0.03	.50	.59	1.70
Gender (female)				–.21 (.21)	−0.06	-.20	.86	1.16
Race¹
Black				–.28 (.33)	−0.05	-.86	.94	1.06
Hispanic				.09 (.66)	0.01	.13	.98	1.02
Other race				.34 (.42)	0.04	.80	.93	1.08
Single				.70 (.46)	0.12	1.53	.44	2.26
Living arrangement²
with roommate				–.06 (.27)	−0.02	-.23	.45	2.23
with my partner				.86 (.40)	0.19	2.14∗	.36	2.82
with parents				.25 (.41)	0.04	.61	.71	1.42
Sexual orientation				.15 (.31)	0.03	.48	.91	1.10
Experiences of sexual harassment as adult				.19 (.05)	0.25	4.20∗∗∗	.78	1.29
Model fit			6.98∗∗∗			3.48∗∗∗
R²			0.06			0.13

Note. 1 = White race was the reference category; 2 = Living alone was the reference category. ^∗p<.05, ^∗^∗p<.01, ^∗^∗^∗p<.001.

Discussion

The abused–abuser hypothesis has mainly focused on delineating the relationship between sexual victimization and sexual offending (Aebi et al., 2015; Glasser et al., 2001; Thomas & Fremouw, 2009), except for a few studies that included other forms of crime such as homicide and property crimes (DeLisi et al., 2014). These studies consistently suggest that victims of crime, especially sexual assault, have a greater likelihood of becoming offenders themselves. The present study extends existing knowledge on the abused–abuser thesis by examining the relationship between childhood victimization and adult offending across the three major types of violence in intimate relationship: sexual assault, physical assault, and emotional/psychological abuse. The goals of this study were (a) to examine the direct effects of childhood victimization on adult offending behavior and (b) to examine the mediation effects of demographic characteristics and other variables in the relationship between victimization and offending.

Results from the current analysis revealed two significant patterns. First, the study observed a direct effect of childhood victimization on adult offending, offering credence to existing findings and supporting the argument that children who are victimized during adolescence or pre-adolescence are most likely to become offenders in their adult life. The observation made in this study was true for all three types of violence. For instance, children who were sexually assaulted were found to have committed offenses such as sexual assault, physical assault, and emotional abuse against their partners as adults. Likewise, those who were either physically or emotionally abused engaged in similar offenses against their spouses or partners. These observations suggest a vicious cycle of violence that can be explained by revisiting the social learning argument about aggression and violence (see Malley-Morrison & Hines, 2004; Veneziano et al., 2000).

Social learning theorists have explained why people engage in criminal acts by arguing that criminal behavior is a learned behavior and that people become criminals through an elaborate learning process. Due to its generalizability, this theory is well suited to explain why people who have experienced childhood victimization may commit violence against their partners in intimate relationship. These people may have learned the languages and definitions of offending through their own victimization and become accustomed to such behavior, to the point that they may consider criminal offending a normal and socially accepted behavior. This may especially happen when their abusers were not punished for the crime. Another explanation could be the internal struggle that victims may undergo to overcome the negative consequences of their victimization. Research suggests that victims may cope by committing violence against others to unleash their frustration and to find internal peace (Veneziano et al., 2000). Victimizing romantic or intimate partners may therefore be viewed as a strategic option to put the past behind and move forward.

The second significant observation made in this study was regarding the mediating role that individual characteristics and experiences of sexual harassment as adult play in the relationship between victimization and offending. These variables partially and fully mediated the effects of victimization on offending. Specifically, when the variables were introduced into the models, the standardized coefficients of the victimization variables were reduced but remained significant, with the exception of sexual victimization which became insignificant in the emotional abuse offending model. One can therefore conclude that while individual characteristics have full mediation effects in the relationship between sexual victimization and emotional offending, they have partial mediation effects in the relationship between sexual and physical victimizations and sexual and physical assault offending. These observations are important especially given the role that individual characteristics play in criminal behavior. For instance, age has been recognized for decades as affecting whether an offender will desist or continue to offend throughout the life-course, the variable has not been examined to determine what role it plays in a victim’s propensity to become a perpetrator. Criminologists believe that the amount of crime an individual commit decreases after the age of 17 years, and as the person grows older, he/she may eventually stop offending. This argument is espoused by those who support the desistance perspective of criminal behavior (Cornelius, Lynch, & Gore, 2017; Hirschi & Gottfredson, 1983; Steffensmeier & Streifel, 1991; Sweeten, Piquero, & Steinberg, 2013; Ulmer & Steffensmeier, 2014; Wyse, Harding, & Morenoff, 2014). If this assertion is true, then it makes sense to suggest that age will decrease the amount of violence that a previously victimized person will commit against his/her partner or spouse.

There are two limitations to keep in mind as readers interpret the results of this study. First, the study used a cross-sectional approach in analyzing the issues discussed in this article. The disadvantage of this approach is that one cannot claim a causal relationship between childhood victimization and adult offending. To make such a claim, future research should rather use a longitudinal approach where data are collected over several time points. The second limitation has to do with the use of self-report data. Not only were the data analyzed collected at one-time point, but the study required respondents to report on their own victimization before the age of 16 years as well as their own offending behavior. Although using self-reporting is not bad, the data may be biased due to memory recall issues and inflated or deflated responses.

Despite the above limitations, the observations and findings of the current study have some policy implications. The observation that childhood victimization influences an individual’s likelihood to commit violence in intimate relationship such as sexual assault, physical assault, and emotional or psychological abuse highlights the need for early intervention to stop these incidents from escalating into serious criminal behavior in adult life. It is acknowledged that some previously victimized children do not grow up to become offenders (Lambie & Johnston, 2016; Thomas & Fremouw, 2009). However, it is important that a risk assessment strategy is put in place to identify at-risk children who have the potential of engaging in criminal activities. The adoption of modern risk assessment tools will also provide practitioners with information that will inform a child’s level and type of service needs. Given children who have a history of victimization are also more likely to experience psychological and emotional problems, it is imperative that trauma experts train service providers in the area of trauma-informed care. Attention to the unique needs of this population will not only increase responsivity to treatment but will ultimately reduce future criminal behavior.

In conclusion, the current study used college data to examine the direct and indirect effects of childhood experiences with violence on victims’ likelihood of becoming offenders. Despite its limitations, the findings obtained are significant and offer a meaningful contribution to the field of criminology; they highlight the need for effective early intervention mechanisms to identify at-risk children and help them effectively cope with the aftermath of victimization. This will eventually ensure that these children will grow up to become noncriminal adults and valuable members of the society.

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.

References

Aebi

Landolt

M. A.

Mueller-Pfeiffer

Schnyder

Maier

Mohler-Kuo

(2015). Testing the “sexually abused-abuser hypothesis” in adolescents: A population-based study. Archives of sexual behavior, 44(8), 2189-2199.

Afifi

T. O.,

Macmillan

H. L.

(2011). Resilience following child maltreatment: A review of protective factors. The Canadian Journal of Psychiatry, 56, 266-272.

Akers

R.L

(1998). Social learning and social structure: A general theory of crime and deviance. Boston, MA: Northeastern University Press.

Bandura

(1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-Hall.

Bandura

(1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.

Bostock

D. J.

Auster

Bradshaw

R. D.

Brewster

Chapin

Williams

(2002). Family violence. American academy of family physicians home study self-Assessment program.

Broidy

L. M.

Daday

J. K.

Crandall

C. S.

Sklar

D. P.

Jost

P. F.

(2006). Exploring demographic, structural, and behavioral overlap among homicide offenders and victims. Homicide Studies, 10, 155-180.

Burgess

R. L.,

Akers

R. L.

(1966). A differential association-reinforcement theory of criminalbehavior. Social Problems, 14, 128-147.

Burton

D. L.

(2003). Male adolescents: Sexual victimization and subsequent sexual abuse. Child and Adolescent Social Work Journal, 20, 277-296.

10.

Burton

D. L.

Miller

D. L.

Shill

C. T.

(2002). A social learning theory comparison of the sexual victimization of adolescent sexual offenders and nonsexual offending male delinquents. Child Abuse & Neglect, 26, 893-907.

11.

Chakhssi

de Ruiter

Bernstein

D. P.

(2013). Early maladaptive cognitive schemas in child sexual offenders compared with sexual offenders against adults and nonsexual violent offenders: An exploratory study. The Journal of Sexual Medicine, 10, 2201-2210.

12.

Cohen

L. J.

McGeoch

P. G.

Gans

S. W.

Nikiforov

Cullen

Galynker

I. I.

(2002). Childhood sexual history of 20 male pedophiles vs. 24 male healthy control subjects. The Journal of Nervous and Mental Disease, 190, 757-766.

13.

Cornelius

C. V.

Lynch

C. J.

Gore

(2017). Aging out of crime: exploring the relationship between age and crime with agent-based modeling. In: Proceedings of the agent-directed simulation symposium (p. 3). San Diego, CA, USA: Society for Computer Simulation International.

14.

DeLisi

Kosloski

A. E.

Vaughn

M. G.

Caudill

J. W.

Trulson

C. R.

(2014). Does childhood sexual abuse victimization translate into juvenile sexual offending? New evidence. Violence and victims, 29(4), 620-635.

15.

Ehrensaft

M. K.

Cohen

Brown

Smailes

Chen

Johnson

J. G.

(2003). Intergenerational transmission of partner violence: A 20-year prospective study. Journal of Consulting and Clinical Psychology, 71, 741-753.

16.

Fite

J. E.

Bates

J. E.

Holtzworth-Munroe

Dodge

K. A.

Nay

S. Y.

Pettit

G. S.

(2008). Social information processing mediates the intergenerational transmission of aggressiveness in romantic relationships. Journal of Family Psychology, 22(3), 367-376.

17.

Ford

J. D.

Chapman

Connor

D. F.

Cruise

K. R.

(2012). Complex trauma and aggression in secure juvenile justice settings. Criminal Justice and Behavior, 39, 694-724.

18.

Gelles

R.J.

(1972). The violent home: A study of physical aggression between husbands and wives. Beverly Hills: Sage.

19.

Glasser

Kolvin

Campbell

Glasser

Leitch

Farrelly

(2001). Cycle of child sexual abuse: Links between being a victim and becoming aperpetrator. British Journal of Psychiatry, 179, 482-494.

20.

Goode

W. J.

(1971). Force and violence in the family. Journal of Marriage and the Family, 33, 624-636.

21.

Gottfredson

M. R.

(1981). On the etiology of criminal victimization. The Journal of Criminal Law and Criminology, 72, 714-726.

22.

Grady

M. D.

Levenson

J. S.

Bolder

(2017). Linking adverse childhood effects and attachment: A theory of etiology for sexual offending. Trauma, Violence & Abuse, 18(4), 433-444.

23.

Hare

A. L.

Miga

E. M.

Allen

J. P.

(2009). Intergenerational transmission of aggression in romantic relationships: The moderating role of attachment security. Journal of Family Psychology, 23(6), 808-818.

24.

Hirschi

T.,

Gottfredson

(1983). Age and the explanation of crime. American Journal of Sociology, 89(3), 552-584.

25.

Jennings

W. G.

Piquero

A. R.

Reingle

J. M.

(2012). On the overlap between victimization and offending: A review of the literature. Aggression and Violent Behavior, 17, 16-26.

26.

Jespersen

A. F.

Lalumière

M. L.

Seto

M. C.

(2009). Sexual abuse history among adult sex offenders and non-sex offenders: A meta-analysis. Child Abuse & Neglect, 33(3), 179-192.

27.

Park

C. J.

(2016). Intimate partner violence: An application of attachment theory. Journal of Human Behavior in the Social Environment, 26(5), 488-497.

28.

Lambie

I.,

Johnston

(2016). “I couldn't do it to a kid knowing what it did to me”. International journal of offender therapy and comparative criminology, 60(8), 897-918.

29.

Lambie

Seymour

Lee

Adams

(2002). Resiliency in the victim-offender cycle in male sexual abuse. Sexual Abuse, 14, 31-48.

30.

Lauritsen

J.L.

Laub

J.H.

(2007). Understanding the link between victimization and offending New reflections on an old idea. In Hough

Maxfield

(Eds), Crime prevention studies: Vol. 22. Surveying crime in the 21st century (pp. 55-75). Monsey, NY: Criminal Justice Press.

31.

Lauritsen

J. L.

Sampson

R. J.

Laub

J. H.

(1991). The link between offending and victimization among adolescents. Criminology, 29, 265-292.

32.

Leach

Stewart

Smallbone

(2016). Testing the sexually abused-sexual abuser hypothesis: A prospective longitudinal birth cohort study. Child Abuse & Neglect, 51, 144-153.

33.

Lisak

Hopper

Song

(1996). Factors in the cycle of violence: Gender rigidity and emotional constriction. Journal of Traumatic Stress, 9, 721−743.

34.

Loseke

D. R.

(2005). Through a sociological lens. In Current controversies on family violence (Vol 35). Sage.

35.

Maldonado-Molina

M. M.

Piquero

A. R.

Jennings

W. G.

Bird

Canino

(2009). Trajectories of delinquency among puerto rican children and adolescents at two sites. Journal of Research in Crime and Delinquency, 46, 144-181.

36.

Malley-Morrison

Hines

(2004). Family violence in a cultural perspective: Defining, understanding, and combating abuse. Sage.

37.

McCloskey

L. A.,

Lichter

E. L.

(2003). The contribution of marital violence to adolescent aggression across different relationships. Journal of Interpersonal Violence, 18(4), 390-412.

38.

Miner

M. H.

Swinburne Romine

Robinson

B. B. E.

Berg

Knight

R. A.

(2014). Anxious attachment, social isolation, and indicators of sex drive and compulsivity. Sexual Abuse, 28, 132-153. doi:10.1177/1079063214547585.

39.

National Intimate Partner and Sexual Violence Survey (2010). Summary Report. National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Retrived from: https://www.cdc.gov/violenceprevention/pdf/nisvs_report2010-a.pdf. Accessed on April 6, 2018

40.

National Violence Against Women Survey (2000). . Full Report of the prevalence, incidence, and Consequences of violence against women. Washington D.C. U. S: Department of Justice. Reterived form: https://www.ncjrs.gov/pdffiles1/nij/183781.pdf Accessed on 6 April 2018.

41.

Nunes

K. L.

Hermann

C. A.

Renee Malcom

Lavoie

(2013). Childhood sexual victimization, pedophilic interest, and sexual recidivism. Child Abuse & Neglect, 37(9), 703-711.

42.

Ogloff

J. R. P.

Cutajar

M. C.

Mann

Mullen

(2012). Child sexual abuse and subsequent offending and victimization: A 45-year follow-up study. , Trends and Issues in Crime and Criminal Justice, 440, 1-6.

43.

Salter

McMillan

Richards

Talbot

Hodges

Bentovim

Hastings

Stevenson

Skuse

(2003). Development of sexually abusive behaviour in sexually victimised males: A longitudinal study. Lancet, 361, 471-476.

44.

Schraufnagel

T. J.

Davis

K. C.

George

W. H.

Norris

(2010). Childhood sexual abuse inmales and subsequent risky sexual behavior: A potential alcohol-use pathway. Child Abuse & Neglect, 34(5), 369-378.

45.

Shillkret

C. J.

(2011). Attachment theory. In Berzoff

Flanagan

L. M.

Hertz

(Eds), Inside out and outside in: Psychodynamic clinical theory and contemporary multicultural context (3rd ed., pp. 186-207). Lanham, MD: Rowman & Littlefield.

46.

Simons

D. A.

Wurtele

S. K.

Durham

R. L.

(2008). Developmental experiences of child sexual abusers and rapists. Child Abuse & Neglect, 32, 549-560.

47.

Steffensmeier

D.,

Streifel

(19911985). Age, gender, and crime across three historical periods: 1935, 1960, and 1985. Social Forces, 69(3), 869-894.

48.

Stith

S. M.

Rosen

K. H.

Middleton

K. A.

Busch

A. L.

Lundeberg

Carlton

R. P.

(2000). The intergenerational transmission of spouse abuse: A meta-analysis. Journal of Marriage and Family, 62(3), 640-654.

49.

Sweeten

Piquero

A. R.

Steinberg

(2013). Age and the explanation of crime, revisited. Journal of Youth and Adolescence, 42(6), 921-938.

50.

Thomas

T. A.,

Fremouw

(2009). Moderating variables of the sexual “victim to offender cycle” in males. Aggression and Violent Behavior, 14(5), 382-387.

51.

Ulmer

J. T.

Steffensmeier

(2014). The age and crime relationship: Social variation, social explanations. In The nurture versus biosocial debate in criminology: On the origins of criminal behavior and criminality. SAGE Publications Inc.

52.

Veneziano

LeGrand

(2000). The relationship between adolescent sex offender behaviors and victim characteristics with prior victimization. Journal of Interpersonal Violence, 15, 363-374.

53.

White

H. R.,

Widom

C. S.

(2003). Intimate partner violence among abused and neglected children in young adulthood: The mediating effects of early aggression, antisocial personality, hostility and alcohol problems. Aggressive Behavior, 29, 332-345.

54.

Widom

C. S.,

Ames

M. A.

(1994). Criminal consequences of childhood sexual victimization. Child Abuse & Neglect, 18, 303-318.

55.

World Health Organization (2005). WHO Multi-county study on Women’s Health and Domestic Violence against Women: Initial results on prevalence, health outcomes and women’s responses. Geneva: WHO.

56.

Wyse

Harding

Morenoff

(2014). Romantic relationships and criminal desistance: Pathways and processes. Sociological forum (Randolph, N.J.), 29(2), 365-385.