Psychopathic Traits and Recurring Victimization: An Examination Using a Community and Clinical Sample

Abstract

Few studies have identified specific characteristics of recurrent victims that distinguish them from single victims. One such characteristic that may do so is possessing psychopathic traits, given that persons with psychopathic traits are generally risk-seeking, callous, short-tempered, and lack behavioral controls. To examine this possibility, we use data from both the National Longitudinal Study of Adolescent to Adult Health (Add Health) and the MacArthur Violence Risk Assessment Study (MacRisk). We find that in both samples, psychopathic traits are able to distinguish between nonvictims and recurring victims as well as single-wave victims and recurring victims. This finding holds great promise for identifying who may be at risk of experiencing recurring victimization, ways to reduce victimization risk, as well as potential additional avenues for research in this area.

Keywords

revictimization psychopathy victimization psychopathic traits replication

Researchers have identified that a sizable proportion of victims experience more than one victimization event. For example, data from the year ending report of the Crime in England and Wales 2017 indicate that 27% of victims of violence experience more than one incident in the previous 12 months (Office for National Statistics, 2017). This finding appears to hold true irrespective of the type of criminal victimization examined. In fact, similar trends have been reported for crimes ranging from larceny (Lauritsen & Davis Quinet, 1995) to violence against women (Tjaden & Thoennes, 2006). What is also striking is the finding that recurring victims account for a disproportionate share of all victimization incidents. In their study of sexual victimization of college women, Daigle, Fisher, and Cullen (2008) found that the 7% of college women who had experienced more than one sexual victimization during an academic year experienced over 70% of all of the sexual victimization incidents. Given the disproportionate amount of victimizations experienced by recurring victims, intervention after an initial victimization event is crucial in reducing the overall amount of victimizations that occur, and that successfully intervening after an initial victimization could substantially reduce victimization rates.

In light of this implication for crime reduction and the fact that there are many persons who will be victimized once, but not again, more recent research has focused on the development and testing of theoretical explanations and the identification of risk factors to identify who is at risk of experiencing multiple victimization incidents and why. Specifically, researchers have begun to investigate what it is that differentiates individuals who experience repetitive events of victimization from those who experience a single victimization incident (and none at all). The main theoretical perspectives used to explain recurring victimization are state dependence and risk heterogeneity. At its core, the state dependence (or “boost”) perspective suggests that during and after a victimization incident, critical information is being gained by both the offender and the victim that serves to elevate or decrease victimization risk (Farrell, Phillips, & Pease, 1995; Tseloni & Pease, 2003). Such information may be how to safely exit a burgled home, that the victim did not report the victimization to the police, or that the victim is capable of resisting the attack. The risk heterogeneity perspective, on the contrary, suggests that it is persistent individual-level characteristics that, if left unchanged, will keep a target at risk for victimizations over time. That is, whatever placed a victim at risk initially will keep a victim at risk, if not changed. These characteristics could be things such as low self-control, having deviant peers, or using alcohol or drugs. Although research findings support both perspectives (Daigle, 2010; Fisher, Daigle, & Cullen, 2010; Gabor & Mata, 2004; Tseloni & Pease, 2003; Turanovic & Pratt, 2014), there remain many questions regarding what distinguishes those recurring victims from persons victimized a single time.

In fact, most research has found more similarities than differences between these two victimization groups and unexplained heterogeneity (see Fisher et al., 2010), leading Daigle and Fisher (2013) to conclude that “[m]ore advanced theorizing and rigorous measures are needed to tease out what these unexplained forces that place individuals at risk of experiencing not just one victimization, but repeated victimizations, might be” (pp. 390). That is, there must be factors that distinguish single-incident and recurring victims from each other—If there were not, then the same factors could be used to successfully predict both single and recurring victimization. In that case, one would intervene in the same ways for both single and recurring victims. Assuming that variables distinguish these two groups, there may be specific targets for primary and secondary prevention. One possible distinguishing feature between single and recurring victims could be having psychopathic traits, which has been linked to perpetrating violence (Blais, Solodukhin, & Forth, 2014; DeLisi, 2009; Walters, 2006). Given the strong link between victimization and offending and the recent application of concepts and theories germane to crime and criminology to victimization, it seems likely that having psychopathic traits plays a role in the production of recurring victimization. As such, we investigate whether psychopathic traits are a distinguishing feature of recurring victimization.

Risk Heterogeneity and Psychopathy

According to the risk heterogeneity perspective, any risk factor that places a person at risk for an initial victimization will continue to place that person at risk, if it is not changed. Many factors have been identified as risk factors for recurring victimization that meet this criterion. For example, persons who live in urban areas (Tseloni, 2000; Wittebrood & Nieuwbeerta, 2000) and those characterized by disorder (Lauritsen & Davis Quinet, 1995; Outlaw, Ruback, & Britt, 2002) are at greater risk for recurring victimization than persons living in nonurban areas and those with low levels of disorder. Individual-level characteristics are also related to recurring victimization. Younger (Gabor & Mata, 2004; Perreault, Sauve, & Burns, 2004) men (Lauritsen & Davis Quinet, 1995; Mukherjee & Carcach, 1998) who are nonmarried (Lasley & Rosenbaum, 1988; Mukherjee & Carcach, 1998; Perreault et al., 2004; Tseloni, 2000), symptomatic and suffering from mental illness (Teasdale, Daigle, & Ballard, 2014), and unemployed are at risk of recurring victimization. Other features characteristic of recurring victims are related to risky lifestyles. Specifically, recurring victims are more likely to spend time with delinquent peers (Lauritsen & Davis Quinet, 1995), be involved in delinquency and crime (Lauritsen & Davis Quinet, 1995; Outlaw et al., 2002), spend time away from home at night (Lasley & Rosenbaum, 1988; Tseloni, 2000), engage in promiscuous sexual behavior (Bramsen et al., 2013), and be high-level drinkers (Lasley & Rosenbaum, 1988).

Although these factors have been identified, the role of one specific individual-level factor—having psychopathic traits—has received little empirical attention. Although two earlier studies examining repeat victimization among women mentioned the inclusion of measures of psychopathy in their analyses, neither reported psychopathy-specific findings (Arata, 1999; Ellis, Atkeson, & Calhoun, 1982). Thus, researchers have yet to directly examine psychopathy (or psychopathic traits) as a potential differentiator between persons who are victimized once and those who are victimized multiple times.

Psychopathy, Victimization, and Recurring Victimization

The psychopathic personality is characterized by callousness, superficial charm, glibness, impulsivity, short-temperedness, and poor behavioral controls (Hare & Vertommen, 1991). More recent developments in the psychopathy literature have identified three core characteristics of psychopathy: disinhibition, boldness, and meanness (Patrick & Drislane, 2015; Patrick, Fowles, & Krueger, 2009). These characteristics are those that ease the restrictions nonpsychopathic persons have on desiring to harm others and behaviorally facilitate engagement in criminal and other antisocial behavior (Frick & White, 2008; Higgins, Kirchner, Ricketts, & Marcum, 2013). These same characteristics are likely to place persons on a path to victimization for several reasons. First, research that has investigated components that comprise psychopathy such as impulsivity (Fanti & Kimonis, 2012; Wilcox, Tillyer, & Fisher, 2009), low self-control (Pratt, Turanovic, Fox, & Wright, 2014), and callous unemotional traits (Fontaine, Hanscombe, Berg, McCrory, & Viding, 2016) has shown that these characteristics place a person at risk for victimization. A person, then, who has not just one of these traits, but a constellation of these factors or a psychopathic personality, is likely to also be at risk for victimization and recurring victimization, which has not been explored.

Second, they are the type of persons who are likely to aggress against others and, in turn, to elicit negative, aggressive behavioral responses from those they affront. Given their lack of affect and empathy, and their manipulative, lying behavior, they likely do not have strong ties and deep personal connections to others who would intervene or protect them (i.e., capable guardians), when this does happen; thus, they are ripe targets for victimization. Third, because they have a penchant for risk and inability to avoid it, they may seek situations that place them in harm’s way, hence becoming a target for victimization. In this way, those high in psychopathic traits may be vulnerable targets, as they may be in settings conducive to victimization. Fourth, research shows that psychopaths display diminished fear reactivity, which suggests that they may not cue to environmental stimuli signaling risk as others do (Newman, Curtin, Bertsch, Baskin-Sommers, 2010), and they may not learn from previous exchanges that resulted in harm or victimization. Indeed, those with a psychopathic personality are likely to learn in different ways than others (Hare & Quinn, 1971), a point we return to in our discussion of psychopathic traits and recurring victimization risk. Fifth, some research indicates that men high in psychopathic traits demonstrate lower level of cooperation in mutually cooperative interactions (Rilling et al., 2007). Although speculative, these individuals may interact in ways that create emotional discomfort for others, thus increasing the likelihood that those people react aggressively to them.

Despite these possible associations between psychopathic traits and victimization, as previously mentioned, the potential link between psychopathic traits and recurring victimization has not been directly explored. Psychopathic traits and victimization and their connection, however, has received attention in the extant literature in at least two ways. First, few studies have investigated the role that psychopathic traits may play in increasing victimization risk. Second, although not investigated directly, the biological underpinnings of psychopathy suggest that those with psychopathic personality traits are poor learners; thus, they may be at increased risk for experiencing victimization over time.

Research on Psychopathy and Victimization

Despite the possibility that possessing psychopathic traits would increase victimization, this relationship has not been widely investigated. In Silver and colleagues’ (2011) study on the victim–offender overlap, psychopathy was found to be related to victimization and offending among a sample of individuals in acute inpatient mental health facilities who were then discharged. In another study, psychopathy was examined among patients in the community in Australia who met diagnostic criteria for schizophrenia spectrum disorders and who were attending community mental health clinics. In this study, persons who had experienced serious violence had higher psychopathy scores than those who had not experienced serious violence (Dolan, O’Malley, & McGregor, 2013). Other studies have found a positive association between psychopathic traits and odds of victimization (Beaver et al., 2016). For example, in a study investigating bullying and psychopathy dimensions, all psychopathy dimensions predicted membership in bully and bully-victim groups, and impulsivity was found to predict membership in victim and bully-victim groups (Fanti & Kimonis, 2012). These studies provide some evidence to suggest that psychopathy plays a role in increasing risk for victimization; however, as noted by Dolan et al. (2013), “there is negligible literature on how psychopathy or psychopathic traits contribute to victimization” (p. 29). Of note, most of these studies used samples of persons who have utilized mental health services. Thus, it is unclear whether psychopathic traits are related to victimization in general community samples, particularly those using nationally representative data. Furthermore, psychopathic traits have not been studied as a factor that may distinguish recurring victimization from single victimization, yet it should play a role in recurring victimization insomuch as a person continues to possess these traits after an initial victimization.

Learning, Psychopathy, and Recurring Victimization

Early descriptions of psychopathy included not only personality characteristics but also difficulty in learning from experience or poor conditionability (Hare, 1968). In this way, one of the reasons that psychopaths were often entrenched in antisocial, even criminal lifestyles despite being arrested or receiving negative responses from others is because they do not respond to punishment in the same ways as nonpsychopaths. In identifying the underlying mechanism for this lack of response to punishment, researchers have identified neurological differences among psychopaths that may dampen their ability to respond to stressful or novel situations (Lorber, 2004). This finding suggests that psychopaths may be less likely to be fearful or to learn from negative experiences. For victimization and recurring victimization, this may indicate an increased likelihood to experience victimization over time. Those high in psychopathic traits may be more at risk for recurring victimization incidents as they do not learn from negative events. Hence, a victimization that may cause a nonpsychopathic person to mold their behavior in ways to reduce risk in the future may not create a similar response in a person high in psychopathic traits.

Explanations about learning ability and psychopathy have expanded to include not just a lack of experiencing or reacting to fear but also a hypersensitivity to rewards (Gorenstein & Newman, 1980). In this way, psychopaths are overly sensitive to rewarding cues in their environment. They are less likely to attend to punishment because they pursue reward—when the chance of reward is present. Thus, they are unlikely to detect risks for experiencing punishment (Arnett, 1997). This hypersensitivity to reward at the expense of recognizing punishment may be relevant for victimization. Persons high in psychopathic traits may be less likely to see danger cues in their environment, particularly when they can see benefits. An inability to see danger, especially in the presence of reward, may mean that a person does not recognize a situation as dangerous when it is (e.g., a person goes home with a stranger for a sexual encounter), and then becomes at risk for victimization. Research on risk perception confirms this possible link between risk recognition and victimization and revictimization. That is, women who have been sexually victimized have longer response latency when determining when date rape scenarios have become dangerous (Soler-Baillo, Marx, & Sloan, 2005). Coupled with a lack of learning from harm, these situations that are high-risk for victimization may be entered into repeatedly by persons high in psychopathic traits.

Current Study

Although research has begun to investigate individual differences in the risk for recurring victimization, psychopathy and psychopathic traits have not been explored as potential sources of differentiation between those who are not victims, who are victimized once, and who are victimized multiple times. This omission is particularly salient as psychopathy is linked to aggression and violence. In addition, psychopathic personality traits have been identified as being relatively stable once established, usually during childhood (Lynam, Caspi, Moffitt, Loeber, & Stouthamer-Loeber, 2007; Shaw & Porter, 2012; Silver et al., 2011), such that if psychopathy is related to an initial victimization, it may place persons at risk for subsequent victimizations, in accordance with a heterogeneity perspective. Furthermore, a person with the constellation of psychopathic personality traits is likely to place himself in risky situations, be less likely to avoid risk, have an interactional style that is prone to elicit aggression, and be unlikely to have close ties that could offer sufficient protection from harm. Considered together, these factors are likely to increase the risk of experiencing recurring victimization. The present study is the first to directly examine the role of psychopathic traits in the production of recurring victimization. We do so by examining this relationship in two separate samples.

Method

To explore the association between psychopathic traits and recurring victimization, we analyze data from two sources. First, we analyze data from the National Longitudinal Study of Adolescent to Adult Health (Add Health). The Add Health dataset features a nationally representative, community-based sample of adolescents and spans a period from 1995 to 2008. Second, we analyze data from the MacArthur Violence Risk Assessment Study (MacRisk). The MacRisk features a clinical sample of released psychiatric patients, who are followed 5 times (every 10 weeks) after release from the hospital, over the course of 1 year. While the strong external validity of Add Health ensures generalizability of those findings, the measurement validity of the clinical measures (particularly the Psychopathy Checklist–Screening Version [PCL:SV]) in the MacRisk strengthens our confidence in the operationalization of key constructs. This methodological triangulation should allow us to replicate our analyses and strengthen conclusions drawn from the current study. In addition, the approach serves as its own replication. That is, we estimate the findings with one sample and then replicate them in the other. Our approach thus bolsters confidence that any findings are real, in that they are replicated utilizing a different sampling strategy, with different measures, a different timeframe, and a different age range.

Sampling—Add Health

The Add Health data collection was accomplished through a clustered-random sample of adolescents from selected study schools, which were representative of the United States with regard to region, urbanity, school size, school type, and ethnicity. The sampling plan for Add Health has been described extensively elsewhere (see Harris, 2013). In brief, school eligibility was determined based on the presence of at least 30 students and an 11th grade within the school. Of the 80 high schools that were selected for inclusion, 52 were eligible and agreed to participate. The remaining 28 schools were replaced by similar high schools. In addition, a single feeder school (junior high or middle school) was selected for each high school. Some schools did not have a feeder school or functioned as their own feeder (that is they had a middle school within the high school). This selection resulted in 145 participating middle, junior, and high schools. Once schools were selected, students within those schools were selected for inclusion in the in-home sample. Students were selected by taking a stratified random sampling approach. Specifically, students were stratified by sex and year in school. A quota was set for each strata, based on dividing the total size of the sample by the number of strata and a random sample of students was then selected. Four complete waves of data are currently available for analysis. In Wave 1 (1995), individuals were enrolled in seventh through 12th grades. Wave 2 (1996) was administered 1 year after Wave 2. Wave 3 (2001-2002) respondents were between the ages of 18 and 26 years, and Wave 4 (2008) interviews were conducted when individuals were between the ages of 24 and 32 years.

Sampling—MacRisk

The MacArthur Violence Risk Assessment study (MacRisk) collected data from individuals released from three psychiatric inpatient hospitals, in Pittsburgh, PA; Worchester, MA; and Kansas City, MO. The sampling plan has been described extensively elsewhere (Robbins, Monahan, & Silver, 2003; Steadman et al., 1998). Patients were excluded from the sample if they were not civil admissions, were not between the ages of 18 and 40, could not speak English, were not White or African American, and were primarily diagnosed with something other than schizophrenia spectrum disorders, major depression, mania, alcohol or substance dependence disorders, or a personality disorder. Patients were also excluded if they had been hospitalized for more than 3 weeks. Eligible patients were stratified by age, sex, and race and were selected based on a random sampling strategy within each strata until a quota was met. Data collection began in 1992 and continued until 1995. After an initial baseline assessment in the hospital, each individual was followed up 5 times, approximately 10 weeks apart, for 1 year post release from the hospital.

Measures—Add Health

In the Add Health data, we created a dependent variable based on a series of items that assess physical assault victimization. These include “In the past 12 months, how often did (a) someone pull a knife or gun on you? (b) someone shot or stabbed you? (c) someone hit, choked, or kicked you? and (d) you were beaten up?” Response options were never, once, or more than once. These responses were collapsed into a dichotomous indicator of victimized or not at each of the four waves of the Add Health. We then created the dependent variable based on whether the individual reported no victimization at any wave (0), being victimized at one of the four waves (1), or being victimized at more than one wave (2).¹

Psychopathic traits were measured in the Add Health with three different measures from Wave 4.² The first captures lack of empathy and is measured by items capturing sympathizing with others’ feelings, not being interested in others’ problems, the extent to which the individual can feel others’ emotions, and not being interested in others. These items had a Cronbach’s alpha of .701. The second measure captures the impulsivity dimension of psychopathy and is measured by items that capture going with your gut when making decisions, living without thought for the future, perceptions that there is little to do to change important things in life, and the perception that others determine what the individual can do. These items had a Cronbach’s alpha of .651. Finally, we capture the antisocial dimension of psychopathy with the following items: losing one’s temper, getting angry easily, being rarely irritated (reversed), getting upset easily, and getting stressed out easily. These items had an alpha of .828. These measures mirror the measure of psychopathic traits used by Beaver and colleagues (Beaver, Barnes, May, & Schwartz, 2011; Beaver et al., 2014; Beaver, Vaughn, & DeLisi, 2013; Beaver, Vaughn, DeLisi, Barnes, & Boutwell, 2012; Boccio & Beaver, 2015). We estimated a confirmatory factor analysis (CFA) treating the items as indicators of the first-order latent variables and the three latent variables as indicators of a second-order factor, which we call psychopathic traits. The CFA produced a good fit to the data: root mean square error approximation (RMSEA) = .035, comparative fit index (CFI) = .940, Tucker–Lewis index (TLI) = .924. For multivariate analysis, we treat the second-order factor as a standard latent variable. For bivariate analyses, we summed the items to create measured variables for each of the three dimensions.

In addition to psychopathic traits, our analyses of Add Health also include controls for sex (male = 1), race (White and Black versus other), binge drinking (ever = 1, never = 0), drug use (ever = 1, never = 0), number of delinquent peers, and past year delinquency (any = 1 and none = 0). All control variables are measured at Wave 1.

Measures—MacRisk

In the MacRisk, victimization was assessed at each community-based follow-up by seven items. These measure whether the respondent reported someone had thrown something at them; they had been “pushed, grabbed, or shoved”; they had been slapped; they had been “kicked, bitten, or choked”; they had been “hit with a fist or object, or beaten up”; they had been “threatened with a knife or a fun or a lethal weapon”; or whether someone had “used a knife or fired a gun” at them. For each of the five follow-up waves, these items were coded dichotomously as 1 if the individual experienced any of these victimizations and 0 if they reported experiencing none. The dependent variable was created based on whether an individual never reported experiencing victimization at any of the five waves (0), only experienced victimization at one wave (1), or experienced victimization at more than one wave (2).

Psychopathic traits were assessed in the MacRisk based on the PCL:SV (Hart, Cox, & Hare, 1995). The PCL:SV is a shortened form of the Psychopathy Checklist–Revised and has received strong support in psychometric evaluations (see, for example, Cooke, Michie, Hart, & Hare, 1999). The PCL:SV has 12 items. These 12 dimensions assess superficial, grandiose, manipulative, lacks remorse, lacks empathy, doesn’t accept responsibility, impulsive, poor behavioral controls, lacks goals, irresponsible, adolescent antisocial behavior, and adult antisocial behavior. These 12 items are scored: 0 = not present, 1 = possibly present, or 2 = definitely present. Scores for the 12 items were summed to create the psychopathic traits measure.

In addition to the substantively interesting variables, we also control for a variety of measures in our analyses of the MacRisk data. These include sex, race, age, socioeconomic status (SES) (based on the Hollingshead & Redlich, 1958 measure), violence (1 = yes, 0 = no), Michigan Alcohol Screening Test (MAST) scores (Pokorny, Miller, & Kaplan, 1972), and Drug Abuse Screening Test (DAST) scores (Skinner, 1982).

Data Analysis

Because of the multicategory nominal measure of our outcome variable, we analyze the data using a multinomial logistic regression model in Stata (version 13) and Mplus (version 8). We use Mplus to model the effects of the second-order factor (a latent variable) on our multinomial outcome measure. We also take into account key design features of the Add Health data collection effort (the nesting of students within study schools, the stratification in the sampling plan, and weights for the oversampling). Analyses of MacRisk also utilize a multinomial logistic regression model. In both datasets, we check for selection bias using a Heckman bivariate probit model in Stata.

Results

Analyses of Add Health

As shown in Table 1, 60% of the respondents in Add Health reported not being a victim at any wave, while 28% reported being a victim in only one wave, and 13% reported being a recurring victim. Because psychopathic trait was measured as a standard latent variable, the mean was 0 and the standard deviation was 1. Most of the sample was female (54%) and White (61%). Approximately 27% reported ever binge drinking and 41% reported engaging in some delinquency in the past year. Finally, the average number of delinquent peers was 2.5. Turning to a bivariate analysis of the impact of psychopathic traits³ on recurring victimization, we ran a one-way analysis of variance. The results suggested significant variation in average scores on the lack of empathy dimension of our psychopathic traits measure across groups, F(2, 13,975) = 92.4, p < .001. The results also showed significant variation in the impulsiveness dimension, F(2, 13,975) = 122.7, p < .001, and the antisocial behavior dimension, F(2, 13,975) = 58.2, p < .001, across groups. Bonferroni post hoc tests revealed significant differences across all three groups for all three psychopathic traits measures. For nonvictims, average scores on the lack of empathy dimension were 8.5; for single-wave victims, they were 8.9, and for recurring victims, they were 9.3. For nonvictims, average scores on the impulsiveness dimension were 8.3, while single-wave victims averaged 8.7 and recurring victims averaged 9.2. Finally, for nonvictims, scores on the antisocial behavior dimension averaged 13.1, single-wave victims averaged 13.5, and recurring victims averaged 14.1. These trends indicate increasing levels of psychopathic traits across victim status, with recurring victims indicating the highest levels of psychopathic traits.

Table 1:

Descriptive Data for Add Health (n = 13,978)

Characteristic	Mean or %	SD	Range
Nonvictims	59.9%		0-1
Single-wave victims	27.5%		0-1
Recurring victims	12.6%		0-1
Psychopathic trait	0.00	1.00
Male	46.4%		0-1
White	60.7%		0-1
Black	20.0%		0-1
Binge drinking	26.7%		0-1
Drug use	29.0%		0-1
Delinquency	40.8%		0-1
Delinquent peers	2.51	2.63	0-9

Note. Psychopathic trait is a second-order factor; it has been standardized via confirmatory factor analysis.

Table 2 reports the results of our design-adjusted multinomial logistic regression analyses predicting variation in victimization experience (nonvictim and single victim versus recurring victims, which made up the omitted reference category).⁴ The psychopathic traits measure significantly distinguished nonvictims from recurring victims. Specifically, for every one standard deviation increase in psychopathic traits, the odds of being a nonvictim, compared with a recurring victim decreased by 32%. That is, people who were recurring victims had on average higher scores on the psychopathic traits measure, compared with nonvictims. Gender, race, drug use, delinquency, and having delinquent peers all significantly distinguished between nonvictims and recurring victims. That is, white individuals had higher odds and black individuals had lower odds of being in the nonvictim group than the recurring victim group, compared with other race individuals. Men had lower odds of being in the nonvictim group, compared with the recurring victim group. Those who used drugs, had delinquent peers, or engaged in delinquency all had significantly lower odds of being in the nonvictim group than in the recurring victim group.

Table 2:

Multinomial Logistic Regression Predicting Victimization in Add Health (n = 13,975)

Nonvictim	b	SE	OR	Standardized OR
Psychopathic trait	−0.039***	.058	0.68	0.75
Male	−1.353***	.074	0.26	0.61
White	0.621***	.135	1.86	1.23
Black	−0.531***	.148	0.59	0.87
Binge drinking	−0.150	.105	0.86	0.95
Drug use	−0.495***	.115	0.61	0.85
Delinquency	−1.334***	.092	0.26	0.62
Delinquent peers	−0.129***	.017	0.88	0.78
Single-wave victim	b	SE	OR	Standardized OR
Psychopathic trait	−0.143***	.061	0.87	0.84
Male	−0.867***	.084	0.42	0.59
White	0.360**	.132	1.43	1.22
Black	−0.103	.139	0.90	0.96
Binge drinking	−0.133	.105	0.88	0.93
Drug use	−0.207*	.101	0.81	0.89
Delinquency	−0.860***	.099	0.42	0.60
Delinquent peers	−0.078***	.019	0.93	0.78

Note. Excluded reference for the outcome variable is “recurring victim.” Excluded reference for race dummies is other race. OR = odds ratio.

p < .05. **p < .01. ***p < .001.

Next, we compare single-wave victims with recurring victims. The psychopathic traits measure significantly distinguished single-wave victims from recurring victims. Specifically, for every one standard deviation increase in psychopathic traits, the odds of being a single-wave victim decreased by 13%, compared with being a recurring victim. That is, recurring victims had significantly higher scores on the psychopathic traits measure than did single-wave victims, holding all other variables in the model constant. Males had significantly greater odds than females of being recurring victims (compared with single-wave victims). White individuals had significantly greater odds than other race individuals of being single-wave victims (compared with being recurring victims). Those who had delinquent peers, used drugs, or were delinquent themselves had significantly greater odds of being in the recurring victim category than the single-wave victim category.

Analyses of MacRisk

As shown in Table 3, approximately 48% of the sample experienced no victimizations, 27% experienced victimization in one wave, and 25% experienced recurring victimization. In addition, the average PCL:SV score was 8.08. Most of the subjects were male (58%), White (71%), and had not committed violence (27% were violent). Turning to a bivariate analysis, we examined whether PCL:SV scores varied as a function of victimization status using a one-way analysis of variance. The result was significant variation in the average PCL:SV score across groups (F_2,685 = 29.56, p < .001). Bonferroni post hoc tests revealed that all three groups had significantly different scores on the PCL:SV. The average PCL:SV score for nonvictims was 6.57. The average score for single-wave victims was 8.84, and the average score for recurring victims was 10.16. These results indicate an increase in PCL:SV scores across victimization status, with recurring victims showing the highest levels of psychopathic traits.

Table 3:

Descriptive Data for MacRisk (n = 688)

Characteristic	Mean or %	SD	Range
Nonvictim	47.8%		0-1
Single-wave victim	27.0%		0-1
Recurring victim	25.1%		0-1
PCL:SV	8.08	5.43	0-24
White	70.8%		0-1
Male	57.6%		0-1
SES	60.64	15.47	11-84
Violence	27.3%		0-1
MAST	0.938	1.37	0-4
DAST	0.256	0.60	0-2

Note. PCL:SV = Psychopathy Checklist–Screening Version; SES = socioeconomic status; MAST = Michigan Alcohol Screening Test; DAST = Drug Abuse Screening Test.

Table 4 reports the results of a multinomial logistic regression,⁵ examining the impact of PCL:SV scores on not being victimized and single-wave victimization, as opposed to recurring victimization (the excluded referent), holding constant sex, race, SES, violent behavior, and MAST and DAST scores. Results from this analysis demonstrate that PCL:SV scores significantly distinguished between nonvictims and recurring victims. Specifically, the odds of being a nonvictim as opposed to a recurring victim are reduced by 10% for every one-point increase on the PCL:SV. That is, recurring victims had significantly higher PCL:SV scores than did nonvictims, all else in the model held constant. Interestingly, the PCL:SV also significantly distinguishes between single victims and recurring victims. That is, for every one-point increase on the PCL:SV, the odds of being a single victim (as opposed to a recurring victim) are reduced by 4.4%. Stated differently, recurring victims had significantly higher PCL:SV scores than did single-wave victims, holding constant all of the variables in the model. Although the metric of the PCL:SV is reasonably interpretable, we turn to the standardized odds ratios for comparability to our Add Health analyses. Here, we see that a one standard deviation increase in psychopathy is associated with a 44% decrease in the odds of being a nonvictim and a 22% decrease in the odds of being a single victim, as opposed to a recurring victim. These are remarkably consistent with our Add Health analyses, where the comparable numbers are 32% and 13% respectively. In addition to psychopathic traits, MAST scores also significantly distinguish between nonvictims and recurring victims. For every one-point increase in the MAST, the odds of being a nonvictim (as opposed to a recurring victim) are reduced by about 20%. In addition, being White (compared with being Black) significantly reduces the odds of being a single victim, as opposed to a recurring victim.

Table 4:

Multinomial Logistic Regression Predicting Victimization in MacRisk (n = 688)

Nonvictim	b	SE	OR	Standardized OR
PCL:SV	−.107***	.020	0.898	0.56
White	.186	.216	1.204	1.09
Male	.070	.204	1.073	1.04
SES	−.012	.007	0.989	0.83
Violence	−.187	.364	0.830	0.95
MAST	−.231**	.077	0.794	0.73
DAST	−.107	.170	0.968	0.94
Single-wave victim	b	SE	OR	Standardized OR
PCL:SV	−.045*	.021	0.956	0.78
White	.488*	.239	1.629	1.25
Male	.383	.224	1.467	1.21
SES	.014	.008	1.014	1.24
Violence	.224	.362	1.251	1.07
MAST	−.012	.077	0.988	0.98
DAST	−.071	.174	0.931	0.96

Note. Excluded reference for the outcome variable is “recurring victim.” Excluded reference for race is “Black.” OR = odds ratio; PCL:SV = Psychopathy Checklist–Screening Version; SES = socioeconomic status; MAST = Michigan Alcohol Screening Test; DAST = Drug Abuse Screening Test.

p < .05. **p < .01. ***p < .001.

Discussion

This research examined the relationship between psychopathic traits and recurring victimization to see whether psychopathic traits distinguished between individuals victimized in a single wave from those who experienced multiple incidents. We also examined whether psychopathic traits distinguished nonvictims from single victims and recurring victims. In doing so, we explored these issues using two different data sets—one from a community sample that used a measure that captures psychopathic traits and one from a clinical sample that used a validated measure of psychopathy (PCL:SV). Such a replication-based approach strengthens our confidence in our three key findings.

First, we found that scores on the psychopathic trait measures (different dimensions and the scale) were higher for single-wave victims than nonvictims and for recurring victims than nonvictims. We also found that scores for recurring victims were higher than for single-wave victims. In examining the scores, the mean score on each dimension of psychopathy in the Add Health sample increased from nonvictim to single-wave victim and from single-wave victim to recurring victim. A similar pattern was found in the MacRisk sample—the average PCL:SV score was highest for recurring victims, next highest for single-wave victims, and lowest for nonvictims. Consistent with the research of Dolan et al. (2013), who found a link between serious violence and psychopathy scores, this finding suggests that psychopathic traits are related to victimization and recurring victimization at the bivariate level.

Second, when controlling for demographics and other risk factors, psychopathic traits were able to distinguish nonvictims from recurring victims and single-wave victims from recurring victims in the Add Health data, while the PCL:SV score was significant in the MacRisk models for both distinguishing nonvictims from recurring victims and single-wave victims from recurring victims. For the Add Health models, higher scores on the psychopathic traits measure were correlated with an increased odds of being a recurring victim, compared with both the single-wave victim and nonvictim categories. Remember that psychopathy is characterized by lack of planning, thought for the future, and agency, as well as aggressiveness, irritability, impulsivity, lack of empathy, and having a quick temper. It makes sense that persons with these types of characteristics would be more likely to be victimized than others—they are likely to engage in risky behavior, unlikely to consider the consequences of their actions, are likely to elicit negative responses from others, and are unlikely to easily learn from previous behavior. The PCL:SV score was also significant in the MacRisk sample models, in that higher scores reflect a greater likelihood to be recurring victims compared with single-time victims and recurring victims compared with nonvictims. As a comparison, a one standard deviation increase in psychopathy is associated with a 44% decrease in the odds of being a nonvictim and a 22% decrease in the odds of being a single victim, compared with being a recurring victim. Again, these findings support the notion that persons with greater psychopathic tendencies having a greater risk for victimization and recurring victimization.

The fact that psychopathic traits distinguish not only nonvictims from recurring victims, but single-wave victims from recurring victims is particularly striking. As previously noted, prior research on recurring victimization has found few individual characteristics that distinguish single from recurring victims (Fisher et al., 2010; Osborn, Ellingworth, Hope, & Trickett, 1996). The fact that psychopathy increases the likelihood of a person being a recurring victim supports a risk heterogeneity perspective. It is likely that once developed, psychopathic traits are largely immutable. If one of the factors that is associated with people becoming victimized is tied to their psychopathic traits, and if they are unable to change these traits, then they likely will be at continued risk for victimization after an initial victimization event. It is also possible that psychopaths experience compounding vulnerability (Clay-Warner, Bunch, & McMahon-Howard, 2016), which suggests that psychopathy may play a role in victimization initially and that the victimization event then works to enhance vulnerability for those high in psychopathic traits in ways unlike those not high in psychopathic traits. Perhaps psychopaths react to victimization with enhanced aggressiveness or anger that then reduces their already low behavioral inhibition and also elicits aggressive responses from others. Also possible is their inability or unwillingness to engage in behaviors that are risk reducing, as found by Turanovic and Pratt (2014) in their study on low self-control and behavioral change following a victimization. Future research should consider ways in which psychopathic traits may serve to enhance vulnerability after an initial victimization.

Another possible explanation is that an initial victimization is related to the development of psychopathy (Bernstein, Stein, & Handelsman, 1998; Borja & Ostrosky, 2013; Forouzan & Nicholls, 2015; Gobin, Reddy, Zlotnick, & Johnson, 2015; Krischer & Sevecke, 2008; Weiler & Widom, 1996). Research on the link between early childhood victimization and elevated psychopathy scores supports this assertion. In their study of juvenile offenders, Krischer and Sevecke (2008) found that early physical trauma was related to higher scores on the Psychopathy Checklist Revised–Youth Version for boys. A similar relationship was found among adults. Those with a history of childhood abuse or neglect had higher scores on the Psychopathy Checklist–Revised than matched controls (Weiler & Widom, 1996). Others have shown that physical abuse and neglect were related to a subcluster of psychopathic personality traits (Bernstein et al., 1998). If this is the case, an initial victimization is related to subsequent victimizations through its effects on psychopathy. That is, a victimization event elevates the risk of developing psychopathy, which in turn increases aggression and violence by increasing affective deficits, impulsive behavior, and lack of connectedness. These characteristics may place a person at risk of experiencing subsequent victimizations. Also important is that, for this study, we are unable to definitely conclude that having psychopathic traits leads to victimization. Instead, it is possible that victimization events precede the development of psychopathic traits. What we can conclude, however, from our research is that the two are correlated. Future research should investigate the causal links, if any, that exist.

Together, these findings may help shape victimization prevention programs. Our findings suggest that those high in psychopathic traits are at risk of experiencing victimization. Although previous research has documented the likelihood of these individuals perpetrating violence, they are also at risk of experiencing violence as victims. Instead of only viewing psychopathic traits as danger cues for violence against others, our research suggests that psychopathic traits may also be indicators of victimization risk. It is noteworthy that, unlike many other risk factors for victimization, psychopathic traits can be identified with existing screening tools (e.g., PCL:SV), and clinicians and criminal justice personnel can use these screening tools to identify those who may be at risk of experiencing victimization. Because psychopaths comprise between 10% and 15% of the offender population, having these tools available to identify those at risk among a criminal justice population may have utility (Hare, 2003).

Once identified, clinicians can work with individuals at risk of experiencing victimization on how to assess situations as risky, how to acknowledge situations as risky, and how to act with appropriate levels of resistance (Rozee & Koss, 2001). Of particular importance may be to teach those high in psychopathic traits to control their impulses and anger. Doing so may be difficult, given that psychopaths may be less attuned biologically to note danger cues in their environment, particularly when rewards are present (Arnett, 1997). Nonetheless, armed with this knowledge, prevention programs can be developed that incorporate these notions of learning that specifically target individuals who may have difficulties assessing risk, such as those high in psychopathic traits. Any program for those high in psychopathic traits should focus on the benefits of change (i.e., reward), rather than be focused on the negative aspects of their behavior. In this way, the knowledge gleaned from this study can be used to further develop individualized treatment and prevention programs to reduce victimization and recurring victimization.

Third, we found that psychopathic traits were related to recurring and single victimization for both a community and clinical sample, when measured using survey questions designed to assess personality in the case of Add Health and when measured via the PCL:SV, which has previously been positively psychometrically evaluated (Cooke et al., 1999). Doing so allowed us to show replication of our findings across multiple measurement strategies. Furthermore, using both community-based and clinical samples allows for greater generalizability of our findings. Moreover, the two data sources captured different points in the life course. While Add Health captured the developmental period spanning adolescence into early adulthood, MacRisk spanned the developmental period from early- to mid-adulthood. This form of methodological triangulation gives us great confidence in the generalizability and reproducibility of our findings.

The importance of this issue of replication cannot be understated. Psychological science is currently in the midst of a “crisis” of reproducibility (John, Loewenstein, & Prelec, 2012; Open Science Collaboration, 2015). For example, a recent publication in Science found that of 100 studies published in three of Psychology’s best journals, only 39% were rated as replicating the original result (Open Science Collaboration, 2015). One of the strengths of the approach provided here is that we began with a hypothesis—that psychopathic traits would correlate with recurring victimization. We then utilized two data sources with complementary strengths to test that hypothesis. The fact that the results mirrored each other so closely across the two analyses bolsters our belief that the hypothesis is not only supported in these two samples, but likely will be in future replications.

Despite the methodological rigor of our study, it is not without its limitations. One limitation is that we were unable to classify persons as psychopaths; rather, given the measures available across the two samples, we measured psychopathic traits in the Add Health study as has been done in previous research (Beaver et al., 2011; Beaver, Boutwell, Barnes, Vaughn, & DeLisi, 2017; Beaver et al., 2014; Beaver et al., 2013; Beaver et al., 2012) and measured psychopathy as a scale in the MacRisk study (Neumann & Hare, 2008; Skeem & Mulvey, 2001). Even though we did not classify individuals as psychopaths, our findings demonstrate that those with higher scores on psychopathic traits were more likely to be single-wave victims and recurring victims than nonvictims. In addition, the approach taken to measurement in Add Health is a limitation. The measure, while paralleling that utilized in a number of other studies, was designed to measure personality. In support of using this measure, researchers have demonstrated that measures designed to capture personality through the five-factor model (FMM) are related to individual behavioral patterns that are consistent with psychopathy (Miller, Lyman, Widiger, & Leukefeld, 2001; Miller & Lynam, 2003). Thus, we are less concerned about the limitations of the Add Health measure as we also measure psychopathic traits using the PCL:SV in the MacRisk analyses. Furthermore, the sample of MacRisk is limited to three communities. We are, however, less concerned about this limitation as Add Health is nationally representative. Similarly, the period of time at risk in MacRisk is short—only 1 year. This is countered by the extensive period of time at risk in Add Health—approximately 13 years (1995-2008).

One limitation might be the measurement strategy adopted in the current study for recurring victimization. As we dichotomized victimization at each wave, we ignore the possibility that someone could be a recurring victim during a single wave. This is, however, a conservative bias, as we are likely undercounting the number of recurring victims. Such an approach has been utilized by others in studying recurring sexual violence (Swartout, Koss, & White, 2015). The alternative would be to allow those who reported experiencing a victimization more than once in a single wave to be counted as a recurring victim. We opted not to do so, because the possibility exists that a respondent would claim they experienced a behavior twice in a single victimization event (e.g., getting punched 4 times during the course of one fight). Furthermore, the measure of victimization only captures the experience of violent victimization, and is self-reported. It is possible that other forms of victimization are experienced by those in the sample and, if assessed, may change the prevalence of victimization and recurring victimization. Furthermore, the context that surrounds the victimization is not captured by these measures. It is possible that victimization stems from initiating violence, something that may be particularly relevant for those high in psychopathic traits. Future research should consider the role that psychopathic traits may play in the production of other forms of recurring victimization and the context that surround victimization.

Conclusion

Even with these few limitations, our study is the first of which we are aware to explore the relationship between psychopathic traits and recurring victimization. The finding that psychopathic traits and PCL:SV scores predict recurring victimization versus single-wave victimization, and recurring victimization versus no victimization, may help aid in the development of prevention programs that could prevent not only first victimizations but also the recurrence of victimization over time. This reduction in recurring victimization may also reduce the use of criminal justice resources, as victims who experience more than one incident tend to use criminal justice resources at higher rates (van Dijk, 2001) and suffer greater negative psychosocial effects than others (Najdowski & Ullman, 2009).

Footnotes

Acknowledgements

Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website ().

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by Grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations.

No direct support was received from Grant P01-HD31921 for this analysis.

ORCID iD

Leah E. Daigle

Notes

Leah E. Daigle, PhD, is professor in the Department of Criminal Justice and Criminology in the Andrew Young School of Policy Studies at Georgia State University. Her most recent research has centered on repeat sexual victimization of college women and the development and continuation of victimization across the life course. She is coauthor of Criminals in the Making: Criminality Across the Life Course (Second Edition) and Unsafe in the Ivory Tower: The Sexual Victimization of College Women, which was awarded the 2011 Outstanding Book Award by the Academy of Criminal Justice Sciences, and author of Victimology: A Text/Reader and Victimology: The Essentials. Her research has also appeared in peer-reviewed journals including Justice Quarterly, Victims and Offenders, The Journal of Quantitative Criminology, and The Journal of Interpersonal Violence. She was awarded the 2014 Andrew Young School of Policy of Studies Excellence in Teaching Award.

Brent Teasdale, PhD, is a professor and chair in the Department of Criminal Justice Sciences at Illinois State University. He is a graduate of Penn State’s Crime, Law, and Justice Program. He has published extensively in the area of mental illness, violence, and victimization using quantitative methods. His coedited book Preventing Crime and Violence was published by Springer Press in 2017.

References

Arata

C. M.

(1999). Repeated sexual victimization and mental disorders in women. Journal of Child Sex Abuse, 7, 1-17.

Arnett

P. A.

(1997). Autonomic responsivity in psychopaths: A critical review and theoretical proposal. Clinical Psychology Review, 17, 903-936.

Beaver

K. M.

Al-Ghamdi

M. S.

Kobeisy

A. N.

Alqurashi

F. H.

Connolly

E. J.

Schwartz

J. A.

(2016). The association between psychopathic personality traits and victimization and exposure to violence in a sample of Saudi Arabian youth. Psychiatric Quarterly, 87, 217-228.

Beaver

K. M.

Barnes

J. C.

May

J. S.

Schwartz

J. A.

(2011). Psychopathic personality traits, genetic risk, and gene-environment correlations. Criminal Justice and Behavior, 38, 896-912.

Beaver

K. M.

Boutwell

B. B.

Barnes

J. C.

Vaughn

M. G.

DeLisi

(2017). The association between psychopathic personality traits and criminal justice outcomes: Results from a nationally representative sample of males and females. Crime & Delinquency, 63, 708-730.

Beaver

K. M.

Nedelec

J. L.

da Silva Costa

Poersch

A. P.

Monica Stelmach

Freddi

M. C.

. . . Boccio

(2014). The association between psychopathic personality traits and health-related outcomes. Journal of Criminal Justice, 42, 399-407.

Beaver

K. M.

Vaughn

M. G.

DeLisi

(2013). Nonshared environmental effects on adulthood psychopathic personality traits: Results from a monozygotic twin difference scores analysis. Psychiatric Quarterly, 84, 381-393.

Beaver

K. M.

Vaughn

M. G.

DeLisi

Barnes

J. C.

Boutwell

B. B.

(2012). The neuropsychological underpinnings to psychopathic personality traits in a nationally representative and longitudinal sample. Psychiatric Quarterly, 83, 145-159.

Bernstein

D. P.

Stein

J. A.

Handelsman

(1998). Predicting personality pathology among adult patients with substance use disorders: Effects of childhood maltreatment. Addictive Behaviors, 23, 855-868.

10.

Blais

Solodukhin

Forth

A. E.

(2014). A meta-analysis exploring the relationship between psychopathy and instrumental versus reactive violence. Criminal Justice and Behavior, 41, 797-821.

11.

Boccio

C. M.

Beaver

K. M.

(2015). Psychopathic personality traits, intelligence, and economic success. The Journal of Forensic Psychiatry & Psychology, 26, 551-569.

12.

Borja

Ostrosky

(2013). Early traumatic events in psychopaths. Journal of Forensic Sciences, 58, 927-931.

13.

Bramsen

R. H.

Lasgaard

Koss

M. P.

Shevlin

Elklit

Banner

(2013). Testing a multiple mediator model of the effect of childhood sexual abuse on adolescent sexual victimization. American Journal of Orthopsychiatry, 83, 47-54.

14.

Caldwell

M. F.

McCormick

D. J.

Wolfe

Umstead

(2012). Treatment-related changes in psychopathy features and behavior in adolescent offenders. Criminal Justice and Behavior, 39, 144-155.

15.

Clay-Warner

Bunch

J. M.

McMahon-Howard

(2016). Differential vulnerability: Disentangling the effects of state dependence and population heterogeneity on repeat victimization. Criminal Justice and Behavior, 43, 1406-1429.

16.

Cooke

D. J.

Michie

Hart

S. D.

Hare

R. D.

(1999). Evaluating the screening version of the Hare Psychopathy Checklist–Revised (PCL: SV): An item response theory analysis. Psychological Assessment, 11, 3-13.

17.

Daigle

L. E.

(2010). Risk heterogeneity and recurrent violent victimization: The role of DRD4. Biodemography and Social Biology, 56, 137-149.

18.

Daigle

L. E.

Fisher

B. S.

(2013). The recurrence of victimization: Terminology, extent, characteristics, correlates and prevention. In Davis

R. C.

Lurigio

A. J.

Herman

(Eds.), Victims of crime (pp. 371-400). Los Angeles, CA: SAGE.

19.

Daigle

L. E.

Fisher

B. S.

Cullen

F. T.

(2008). The violent and sexual victimization of college women: Is repeat victimization a problem? Journal of Interpersonal Violence, 23, 1296-1313.

20.

DeLisi

(2009). Psychopathy is the unified theory of crime. Youth Violence and Juvenile Justice, 7, 256-273.

21.

Dolan

O’Malley

McGregor

(2013). The role of psychopathic traits and substance abuse in predicting violent victimization in patients with schizophrenia spectrum disorders. Personality and Mental Health, 7, 28-38.

22.

Ellis

E. M.

Atkeson

B. M.

Calhoun

K. S.

(1982). An examination of differences between multiple-and single-incident victims of sexual assault. Journal of Abnormal Psychology, 91, 221-224.

23.

Fanti

K. A.

Kimonis

E. R.

(2012). Bullying and victimization: The role of conduct problems and psychopathic traits. Journal of Research on Adolescence, 22, 617-631.

24.

Farrell

Phillips

Pease

(1995). Like taking candy: Why does repeat victimization occur? The British Journal of Criminology, 35, 384-399.

25.

Fisher

B. S.

Daigle

L. E.

Cullen

F. T.

(2010). What distinguishes single from recurrent sexual victims? The role of lifestyle-routine activities and first-incident characteristics. Justice Quarterly, 27, 102-129.

26.

Fontaine

M. G.

Hanscombe

K. B.

Berg

M. T.

McCrory

E. J.

Viding

(2016). Trajectories of callous-unemotional traits in childhood predict different forms of peer victimization in adolescence. Journal of Clinical Child & Adolescent Psychology. Advance online publication. doi:10.1080/15374416.2015.1105139

27.

Forouzan

Nicholls

T. L.

(2015). Childhood and adolescent characteristics of women with high versus low psychopathy scores: Examining developmental precursors to the malignant personality disorder. Journal of Criminal Justice, 43, 307-320.

28.

Frick

P. J.

White

S. F.

(2008). Research review: The importance of callous–unemotional traits for developmental models of aggressive and antisocial behavior. Journal of Child Psychology & Psychiatry, 49, 359-375.

29.

Gabor

Mata

(2004). Victimization and repeat victimization over the life span: A predictive study and implications for policy. International Review of Victimology, 10, 193-216.

30.

Gobin

R. L.

Reddy

M. K.

Zlotnick

Johnson

J. E.

(2015). Lifetime trauma victimization and PTSD in relation to psychopathy and antisocial personality disorder in a sample of incarcerated women and men. International Journal of Prisoner Health, 11, 64-74.

31.

Gorenstein

E. E.

Newman

J. P.

(1980). Disinhibitory psychopathology: A new perspective and a model for research. Psychological Review, 87, 301-315.

32.

Hare

R. D.

(1968). Psychopathy, autonomic functioning, and the orienting response. Journal of Abnormal Psychology, 73, 1-24.

33.

Hare

R. D.

(2003). The Psychopathy Checklist–Revised. Toronto, OH: Multi Health Systems, Incorporated.

34.

Hare

R. D.

Quinn

M. J.

(1971). Psychopathy and autonomic conditioning. Journal of Abnormal Psychology, 77, 223-235.

35.

Hare

R. D.

Vertommen

(1991). The Hare Psychopathy Checklist–Revised. Toronto, Ontario, Canada: Multi-Health Systems, Incorporated.

36.

Hart

S. D.

Cox

D. N.

Hare

R. D.

(1995). The Hare Psychopathy Checklist: Screening Version (PCL: SV). Toronto, Ontario, Canada: Multi-Health Systems, Incorporated.

37.

Harpur

T. J.

Hare

R. D.

(1994). Assessment of psychopathy as a function of age. Journal of Abnormal Psychology, 103, 604-609.

38.

Harris

K. M.

(2013). The add health study: Design and accomplishments. Chapel Hill: Carolina Population Center, University of North Carolina at Chapel Hill.

39.

Higgins

G. E.

Kirchner

E. E.

Ricketts

M. L.

Marcum

C. D.

(2013). Impulsivity and offending from childhood to young adulthood in the United States: A developmental trajectory analysis. International Journal of Criminal Justice Sciences, 8, 182-197.

40.

Hollingshead

A. B.

Redlich

F. C.

(1958). Social class and mental illness. New York, NY: John Wiley.

41.

John

L. K.

Loewenstein

Prelec

(2012). Measuring the prevalence of questionable research practices with incentives for truth telling. Psychological Science, 23, 524-532.

42.

Krischer

M. K.

Sevecke

(2008). Early traumatization and psychopathy in female and male juvenile offenders. International Journal of Law and Psychiatry, 31, 253-262.

43.

Lasley

J. R.

Rosenbaum

J. L.

(1988). Routine activities and multiple personal victimization. Sociology and Social Research, 73, 47-50.

44.

Lauritsen

J. L.

Davis Quinet

K. F.

(1995). Repeat victimization among adolescents and young adults. Journal of Quantitative Criminology, 11, 143-166.

45.

Lorber

M. F.

(2004). Psychophysiology of aggression, psychopathy, and conduct problems: A meta-analysis. Psychological Bulletin, 130, 531-552.

46.

Lynam

D. R.

Caspi

Moffitt

T. E.

Loeber

Stouthamer-Loeber

(2007). Longitudinal evidence that psychopathy scores in early adolescence predicts adult psychopathy. Journal of Abnormal Psychology, 116, 155-165.

47.

Lynam

D. R.

Miller

D. J.

Vachon

Loeber

Stouthamer-Loeber

(2009). Psychopathy in adolescence predicts official reports of offending in adulthood. Youth Violence and Juvenile Justice, 7, 189-207.

48.

Miller

J. D.

Lyman

D. R.

Widiger

T. A.

Leukefeld

(2001). Personality disorders as extreme variants of common personality dimensions: Can the five-factor model adequately represent psychopathy? Journal of Personality, 69, 253-76. doi:10.1111/1467-6494.00144

49.

Miller

J. D.

Lynam

D. R.

(2003). Psychopathy and the five-factor model of personality: A replication and extension. Journal of Personality Assessment, 81, 168-178.

50.

Mukherjee

Carcach

(1998). Repeat victimisation in Australia. Research in Public Policy Analysis and Management, 15, 1-40.

51.

Najdowski

C. J.

Ullman

S. E.

(2009). Prospective effects of sexual victimization on PTSD and problem drinking. Addictive Behaviors, 34, 965-968.

52.

Neumann

C. S.

Hare

R. D.

(2008). Psychopathic traits in a large community sample: Links to violence, alcohol use, and intelligence. Journal of Consulting and Clinical Psychology, 76, 893-899. doi:10.1037/0022-006X.76.5.893

53.

Newman

J. P.

Curtin

J. J.

Bertsch

J. D.

Baskin-Sommers

(2010). Attention moderates the fearlessness of psychopathic offenders. Biological Psychiatry, 67, 66-70.

54.

Office for National Statistics. (2017). Crime in England and Wales: Annual trend and demographic tables: Table D6. London, England. Retrieved from https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/datasets/crimeinenglandandwalesannualtrendanddemographictables

55.

Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349, 943-951.

56.

Osborn

D. R.

Ellingworth

Hope

Trickett

(1996). Are repeatedly victimized households different? Journal of Quantitative Criminology, 12, 223-245.

57.

Outlaw

Ruback

Britt

(2002). Repeat and multiple victimizations: The role of individual and contextual factors. Violence and Victims, 17, 187-204.

58.

Patrick

C. J.

Drislane

L. E.

(2015). Triarchic model of psychopathy: Origins, operationalizations, and observed linkages with personality and general psychopathology. Journal of Personality, 83, 627-643.

59.

Patrick

C. J.

Fowles

D. C.

Krueger

R. F.

(2009). Triarchic conceptualization of psychopathy: Developmental origins of disinhibition, boldness, and meanness. Development and Psychopathology, 21, 913-937.

60.

Perreault

Sauve

Burns

(2004). Multiple victimization in Canada, 2004. Ottawa: Canadian Centre for Justice Statistics. Retrieved from https://www.publicsafety.gc.ca/lbrr/archives/cnmcs-plcng/cn000037375118-eng.pdf

61.

Pokorny

A. D.

Miller

B. A.

Kaplan

H. B.

(1972). The brief MAST: A shortened version of the Michigan Alcoholism Screening Test. The American Journal of Psychiatry, 129, 342-345.

62.

Pratt

T. C.

Turanovic

J. J.

Fox

K. A.

Wright

K. A.

(2014). Self-control and victimization: A meta-analysis. Criminology, 52, 87-116.

63.

Rilling

J. K.

Glenn

A. L.

Jairam

M. R.

Pagnoni

Goldsmith

D. R.

Elfenbein

H. A.

Lilienfeld

S. O.

(2007). Neural correlates of social cooperation and non-cooperation as a function of psychopathy. Biological Psychiatry, 11, 1260-1271.

64.

Robbins

P. C.

Monahan

Silver

(2003). Mental disorder, violence, and gender. Law and Human Behavior, 27, 561-571.

65.

Rozee

P. D.

Koss

M. P.

(2001). Rape: A century of resistance. Psychology of Women Quarterly, 25, 295-311.

66.

Silver

Piquero

A. R.

Jennings

W. G.

Piquero

N. L.

Leiber

(2011). Assessing the violent offending and violent victimization overlap among discharged psychiatric patients. Law and Human Behavior, 35, 49-59. doi:10.1007/s10979-009-9206-8

67.

Shaw

Porter

(2012). Forever a psychopath? Psychopathy and the criminal career trajectory. In Häkkänen-Nyholm

Nyholm

(Eds.), Psychopathy and law: A practitioner’s guide (1st ed., pp. 201-221). New York, NY: John Wiley.

68.

Skeem

J. L.

Mulvey

E. P.

(2001). Psychopathy and community violence among civil psychiatric patients: Results from the MacArthur Violence Risk Assessment Study. Journal of Consulting and Clinical Psychology, 69, 358-374.

69.

Skeem

J. L.

Polaschek

D. L. L.

Patrick

C. J.

Lilienfeld

S. O.

(2011). Psychopathic personality: Bridging the gap between scientific evidence and public policy. Psychological Science in the Public Interest, 12, 95-162.

70.

Skinner

H. A.

(1982). The Drug Abuse Screening Test. Addictive Behaviors, 7, 363-371.

71.

Soler-Baillo

J. M.

Marx

B. P.

Sloan

D. M.

(2005). The psychophysiological correlates of risk recognition among victims and non-victims of sexual assault. Behaviour Research and Therapy, 43, 169-181.

72.

Steadman

H. J.

Mulvey

E. P.

Monahan

Robbins

P. C.

Appelbaum

P. S.

Grisso

. . . Silver

(1998). Violence by people discharged from acute psychiatric inpatient facilities and by others in the same neighborhoods. Archives of General Psychiatry, 55, 393-401.

73.

Swartout

K. M.

Koss

M. P.

White

J. W.

(2015). Trajectory analysis of the campus serial rapist assumption. JAMA Pediatrics, 169, 1148-1154.

74.

Teasdale

Daigle

L. E.

Ballard

(2014). Trajectories of recurring victimization among people with major mental disorders. Journal of Interpersonal Violence, 29, 987-1005.

75.

Tjaden

P. G.

Thoennes

(2006). Extent, nature, and consequences of rape victimization: Findings from the National Violence Against Women Survey. Washington, DC: U.S. Department of Justice.

76.

Tseloni

(2000). Personal criminal victimization in the United States: Fixed and random effects of individual and household characteristics. Journal of Quantitative Criminology, 16, 415-442.

77.

Tseloni

Pease

(2003). Repeat personal victimization: “Boosts” or “flags”? The British Journal of Criminology, 43, 196-212.

78.

Turanovic

J. J.

Pratt

T. C.

(2014). “Can’t stop, won’t stop”: Self-control, risky lifestyles, and repeat victimization. Journal of Quantitative Criminology, 30, 29-56. doi:10.1007/s10940-012-9188-4

79.

van Dijk

J. J. M

. (2001). Attitudes of victims and repeat victims toward the police: Results of the International Crime Victims Survey. Crime Prevention Studies, 12, 27-52.

80.

Walters

G. D.

(2006). Risk-appraisal versus self-report in the prediction of criminal justice outcomes a meta-analysis. Criminal Justice and Behavior, 33, 279-304.

81.

Weiler

B. L.

Widom

C. S.

(1996). Psychopathy and violent behaviour in abused and neglected young adults. Criminal Behaviour and Mental Health, 6, 253-271.

82.

Wilcox

Tillyer

M. S.

Fisher

B. S.

(2009). Gendered opportunity? School-based adolescent victimization. Journal of Research in Crime and Delinquency, 46, 245-269.

83.

Wittebrood

Nieuwbeerta

(2000). Criminal victimization during one’s life course: The effects of previous victimization and patterns of routine activities. Journal of Research in Crime and Delinquency, 37, 91-122.