Sage Journals: Discover world-class research

Abstract

Until relatively recently, the study of victimization has been largely outside the purview of behavioral geneticists and evolutionary psychologists. Recent victimology research, however, has shown that genetic and evolutionary forces are connected to the risk of victimization. The current study expands on these findings by examining whether genetic influences differentially explain victimization in males and females. To do so, we use a sample of sibling pairs drawn from the National Longitudinal Study of Adolescent to Adult Health (Add Health; N = 4,244). The analyses revealed no significant quantitative sex differences in the etiology of adult victimization. However, the results of this study do highlight the importance of accounting for genetic factors when studying the etiology of specific types of adult victimization. We conclude by discussing the implications of the current study for future research.

Keywords

victimization sex differences evolutionary criminology behavioral genetics

Public opinion polls show that for the first time since 2016, a majority of adults in the United States are worried about crime and violence. Of the adults polled, approximately 80% of respondents indicated that they are concerned about being a victim of a violent crime, which ranks crime and victimization near the top of the list of national concerns (Brenan, 2022). Whether this concern about crime is justified depends, in part, on a number of different factors, including demographic data and where a person resides. In some places and for some people, the chances of being the victim of crime are relatively low, whereas for others the odds of being victimized can be quite high. Overall, though, more than 11 million households report being victims of a property crime annually in the United States. Meanwhile, approximately 3 million people reported being victims of at least one violent crime in 2021 (Thompson & Tapp, 2022).

Given these statistics, the fear of being victimized, and the serious consequences associated with victimization, there has been an explosion of research attempting to uncover the causes of victimization (e.g., Lauritsen & Rezey, 2018; Turanovic & Pratt, 2019). This area of literature has flourished in the last few years as scholars have called for the “revitalization” of victimization theory (Pratt & Turanovic, 2021). From these calls, scholars have expanded their theoretical lenses to be more “unapologetically interdisciplinary” (Pratt & Turanovic, 2021, p. 4) and have drawn not just from the traditional fields of criminology and sociology but also from fields such as psychology and biology (Beaver & Joyner, 2021).

Relatively recently there has been interest in examining the ways in which evolutionary psychology and behavioral genetics could be melded together to more fully understand the causes and correlates of victimization (Beaver & Joyner, 2021). Although traditionally these research areas are often seen as “at odds” with the more general victimization research, scholarship that has emerged in recent years has shown that they are not nearly as disparate as originally conceived. Indeed, the movement towards sex-specific research in victimization (e.g., Kruttschnitt & Kang, 2021) dovetails perfectly with an approach that centers on evolutionary psychology. To date, however, there has not been much research explicating either the various ways in which an evolutionary psychological approach could be applied to the understanding of victimization generally or, more specifically, to male-female differences. Against this backdrop, the current study partially addresses this gap in the literature and examines sex differences in the etiology of adult victimization using a family-based research design.

Sex Differences in Victimization

Research examining male-female differences in victimization has revealed a somewhat interesting pattern of results. Studies, for example, have shown that rates of nonviolent victimization are relatively similar for males and females, a finding that has been shown to be robust and stable across time and space, with only a slight trend towards a greater proportion of female than male victims (Rand, 2009). A very different picture emerges, however, when examining sex differences for rates of violent victimization. For violent victimization, whether males or females are disproportionately victimized depends on the type of victimization. Stark differences emerge for homicide victims, for example, where males account for approximately 80% of all victims (Gartner, 2011; Lauritsen & Heimer, 2008). Other types of violent victimizations, however, are disproportionately accounted for by female victims, such as sexual assaults, rape (e.g., Planty et al., 2013), and intimate partner homicide (e.g., Catalano et al., 2009; Lauritsen & Heimer, 2008).

There has been no shortage of research and theory exploring reasons for male-female differences in victimization. For example, within the victimization literature, explanations for sex-based differences in rates of violent victimization have included structural/societal-level factors, such as examining the gender structure of society (e.g., LeSuer, 2020; Willie & Kershaw, 2019; Xie et al., 2012), as well as broader structural changes in educational attainment and labor force participation (Cohen & Felson, 1979). Meanwhile, other explanations have focused on individual-level factors, including offending and antisocial behaviors (e.g., Jennings et al., 2010), as well as personality traits and friendship characteristics (e.g., Peterson et al., 2018). Despite this literature, there is no consensus on the causes of male-female differences in victimization and, moreover, most studies only explain a relatively small percentage of variance (∼20%–30%) when attempting to explain male-female differences in victimization.

One potential reason why research has not fully accounted for male-female differences in victimization is because not all potential causes have been examined. A possible contributor to male-female differences in victimization that has not been fully explored is the theory of evolution and, more specifically, the influence of genetic factors. To explain, genetic variants have been shown to be non-randomly selected based on their ability to improve fitness (Buss, 2009; Tooby & Cosmides, 1990). This nonrandom selection of genetic factors then contributes to the development of the brain and certain traits that can enhance fitness. Although this body of work is far from complete, there is reason to suspect that adaptations exist within our species that specifically deal with various forms of victimization (e.g., Duntley & Schackelford, 2012; Wyckoff, Buss, & Markman, 2019). Daly and Wilson (1997) and David Buss (2005), for example, both directly discuss how evolution, and the nonrandom selection of genetic variants impacts the likelihood of being a victim of homicide. Specific traits that are commonly connected to the risk of victimization, such as self-control, have also been revealed to be adaptive traits (e.g., Conroy-Beam et al., 2019) that influence the risk of being victimized (e.g., Pratt et al., 2014).

Behavioral genetic research has likewise supported the application of the theory of evolution to the study of victimization by demonstrating that victimization is partially heritable. To be specific, the results generated from these studies have underscored the importance of genetic influences in the etiology of victimization. For example, heritability estimates of adolescent victimization hovers around 40% to 50% in twin- (e.g., Beaver et al., 2009) and adoption-based samples (e.g., Beaver et al., 2013) meanwhile studies examining adult victimization have revealed estimates ranging from 15% to 51% depending on the type of victimization being studied (e.g., Hines & Saudino, 2004). Kavish et al. (2019), for instance, used the sibling subsample of the Child and Young Adult sample of the National Longitudinal Survey of Youth (CNLSY) to examine differences in genetic and environmental influences between single and chronic victimization. Their analyses revealed that while the influence of genetic factors on single violent victimization hovered around 36%, the heritability estimate of chronic violent victimization in adulthood was 51%.

Although genetic influences on victimization vary based on sample-specific characteristics, collectively the results of these studies reveal that victimization is influenced, at least to some extent, by genetic factors. As a consequence, studies examining male-female differences in victimization should account for genetic influences to maximize the amount of variance explained and to shed light on reasons for male-female differences in victimization (Barnes et al., 2014; Harden et al., 2008; Haynie & McHugh, 2003).

Perhaps the most straightforward approach to begin to examine whether differences exist in the extent to which victimization is explained by genetic influences is to analyze males and females separately. To our knowledge, there are not any studies that have estimated male-female genetic differences in variance in victimization. Despite the lack of research on this topic, there are reasons to suspect that such differences in the influence of genetic and environmental factors exist. Prior studies, for instance, investigating sex-specific pathways for victimization have identified a number of salient risk factors that account for at least part of the differences between males and females on victimization. Heritability estimates of these factors have been shown to vary between males and females (e.g., Goldstein et al., 2001; de Vries et al., 2021; Zheng & Cleveland, 2015). For example, aggression (e.g., de Vries et al., 2021; Vierikko et al., 2003), conduct problems, antisocial behavior (e.g., Goldstein et al., 2001), chronic offending (e.g., Zheng & Cleveland, 2015), and callous-unemotional traits (e.g., Bezdijan et al., 2011; Fontaine et al., 2010; Viding et al., 2007), have all been shown to have different heritability estimates for males and females. Since the factors that have been shown to partially account for male-female differences in victimization have also been shown to have different heritability estimates for males and females, it points to the possibility that there are differences in the heritability of victimization for males and females.

Current Study

Behavioral genetic models represent an important tool that has yet to be consistently integrated into the mainstream literature examining the causes and correlates of victimization. The few studies that have used such methods to investigate the etiology of victimization represent an important first step into furthering our understanding and subsequent theorizing of victimization. Yet, ignoring the possibility of sex-specific genetic pathways to victimization clouds the ability to explain how and why gender disparities exist in victimization. Against this backdrop, the purpose of the current study is to examine quantitative sex differences in the etiology of adult victimization.

Method

Data

Data for this study were drawn from the National Longitudinal Study of Adolescent to Adult Health (Add Health; Harris, 2009). Beginning in 1994, Add Health is a school-based longitudinal study of adolescents living across the United States.¹ Five waves of data have been collected spanning 20 years of development. The first and second waves of data were conducted when most of the respondents were adolescents and enrolled in either middle or high school. The third wave of data was collected in 2001 when most of the respondents had reached young adulthood. Wave 4 data were collected in 2007 while Wave 5 data were collected from 2016 to 2018 when the respondents were well into adulthood (Harris et al., 2013).

One of the most unique features of the Add Health data is that siblings were purposefully oversampled during Wave 1 data collection. To do so, supplemental samples were drawn based on the genetic relatedness of siblings in a household. During the in-school survey, the respondents were asked to indicate whether they have a sibling who live with them between the ages of 11 and 20 years old. Those respondents who indicated affirmatively that they had a sibling were selected for inclusion in the in-home Wave 1 sample. Respondents with full siblings naturally occurred in larger numbers in the core in-home sample, but those with a cotwin, half-sibling, cousin, and unrelated sibling (e.g., stepsiblings, foster and adopted children, etc.) were purposefully oversampled. During Wave 1 data collection, the decision was made to only sample two children from each household. These sibling pairs took the same questionnaires and shared the same home, school, and neighborhood environments. A total of 4,368 siblings in total were interviewed (Harris et al., 2013). Analyses comparing demographic characteristics between respondents in the main sample and the sibling subsample have revealed no significant differences in demographics between the two samples (Jacobson & Rowe, 1998).

The final analytical sample consisted of monozygotic (MZ) twin pairs, dizygotic (DZ) twin pairs, full-sibling pairs, and half-sibling pairs. All cousins and unrelated siblings were dropped from the final analytical sample as these respondents never reported the amount of time they resided in the same household. Only same-sex sibling pairs are used in the analyses to detect quantitative sex differences. The final sample size is 2,122 sibling pairs. A family-based design was used over a more traditional twin-based design for two reasons. First, previous studies have revealed that family-based designs more accurately capture the influence of shared environmental factors on the phenotype than twin-based designs. Second, using inclusive samples of both nontwin siblings and twin pairs has been shown to address some of the limitations of twin-only studies such as issues regarding the presence of assortative mating and the violation of the equal environments assumption (Medland & Hatemi, 2009).

Victimization

Wave 3 personal victimization. At Wave 3, respondents reported whether, in the past 12 months, they had a knife or gun pulled on them, been shot, jumped, or cut, had been beaten up, as well as seen someone get shot or stabbed. Responses to each of these items were coded such that 0 = no and 1 = yes. The scores across the seven measures were summed together and then recoded to a dichotomous measure so that 0 = not victimized and 1 = victimized at least one time in the last 12 months.²

Wave 4 personal victimization. The Wave 4 personal victimization measure included six items assessing the victimization experiences of the respondent. All six items contained in the Wave 4 victimization measure were binary items (0 = no, 1 = yes) and had considerable overlap with the Wave 3 victimization measure. To be specific, respondents indicated whether in the past 12 months, someone had stolen something from them that was worth more than $50, had someone pull a knife or gun on them, shot or stabbed, slapped, hit, choked, kicked, or beaten them up as well as saw someone shoot or stab another person.³ The Wave 4 items were summed together and then recoded to a binary measure so that 0 = not victimized and 1 = victimized at least once in the last twelve months. Both personal victimization measures have been used in prior work assessing etiological pathways of victimization (e.g., Barnes et al., 2012).

Stability in personal victimization. A stability personal victimization measure was created by summing together the Wave 3 and 4 personal victimization measures. The scale, therefore, ranges from 0 to 2 so that higher scores indicate greater levels of stability in victimization.

Wave 3 violent victimization. At Wave 3, respondents were asked whether, in the past 12 months, they had a gun or knife pulled on them, were jumped or beaten up, shot, or stabbed. Responses to these items were coded so that 0 = no, 1 = yes. The five items were summed together and then recorded as a binary measure so that 0 = not victimized and 1 = victimized at least once in the past 12 months.

Wave 4 violent victimization. The Wave 4 violent victimization measure included some of the same items as the Wave 3 measure. To be specific, respondents reported whether they had a knife or gun pulled on them, were beaten or jumped, or were stabbed or shot. Each item was coded as a binary measure (0 = no, 1 = yes). The Wave 4 violent victimization measure was created by summing together the three items and then recoding the variable to a dichotomous measure so that 0 = had not experienced victimization and 1 = had experienced at least one victimization in the past 12 months. The Wave 3 and 4 violent victimization measures have been used previously throughout the victimization literature (e.g., Semenza, et al., 2021).

Stability in violent victimization. Similar to the stability of the personal victimization measure, the stability in the violent victimization measure was created by summing the Wave 3 and 4 violent victimization measures. The index, therefore, ranges from 0 to 2 so that higher scores indicate greater levels of stability victimization.

Wave 3 intimate partner violence. Four items were drawn from Wave 3 to assess intimate partner violence. Respondents were asked to indicate how often in the past twelve months their partner slapped, hit, or kicked them, threatened them with violence, pushed or shoved them, threw something at them, forced themselves sexually on to them, as well as injured themselves while fighting their partner. All four items were scored based on a 7-point rating scale wherein each item was coded such that 6 indicated experiencing intimate partner violence more than 20 times in the past year. The Wave 3 intimate partner violence measure was created by summing the scores of the four items together and then dichotomizing the measure.

Wave 4 intimate partner violence. The Wave 4 intimate partner violence measure contains four items identical to the Wave 3 measure. Specifically, respondents reported how often in the past twelve months their partner slapped, hit, or kicked them, threatened them with violence, pushed or shoved them, threw something at them, injured themselves while fighting their partner, as well as had their partner force themselves on them sexually. These items were scored based on an 8-point rating scale wherein each item was coded such that 7 indicated experiencing intimate partner violence more than 20 times in the past year of the relationship. The Wave 4 measure was created by summing the scores of the four items together and then dichotomizing the measure so that 0 = not experienced intimate partner violence in the past year and 1 = having experienced intimate partner violence in the past year. Intimate partner violence measures have been used in previous investigations of victimization (e.g., Connolly, et al., 2022).

Stability in intimate partner violence. Stability in intimate partner violence was created by summing together the Waves 3 and 4 intimate partner violence measures. Therefore, the index ranges from 0 to 2 wherein 2 = experiencing intimate partner violence at both waves of data collection.

Wave 4 sexual victimization. Sexual victimization was assessed at Wave 4. Respondents were asked to indicate whether they have ever been forced, in a nonphysical way, to have any type of sexual activity against their will as well as physically forced against their will.⁴ These items were coded dichotomously (0 = no, 1 = yes). The measure, therefore, was created by summing together these two items and then recoding the variable as a dichotomous measure. This measure has been used in previous peer-reviewed studies investigating the etiology of sexual victimization (e.g., Farrell, 2020). Due to the fact these items were only measured at Wave 4, stability in the sexual victimization measure was not able to be created.

Repeat victimization. A binary repeat victimization measure was created by assessing whether respondents experienced the same types of victimizations between waves. To do so, the items contained within each victimization measure were broken out and a flag was created to indicate if a respondent experienced the victimization measured by each item at both waves. To illustrate, if an individual reported being jumped or beaten at Waves 3 and 4, they would be coded as a 1. However, if a respondent reported being jumped or beaten at Wave 3 and stabbed at Wave 4, then they would be coded as a 0 as they did not experience the same type of victimization twice.⁵ Table 1 includes the descriptive statistics for this study.⁶

Table 1.
Descriptive Statistics for Add Health Study Variables.

Full sample Male sample Female sample

Range Mean SD N Mean SD N Mean SD N

Wave 3 personal 0–1 0.101 2,103 0.167 1,001 0.050 1,102

Wave 4 personal 0–1 0.335 1,913 0.395 913 0.279 1,000

Stability personal 0–2 0.432 0.595 1,661 0.553 0.649 770 0.327 0.521 891

Wave 3 violent 0–1 0.079 2,111 0.126 1,005 0.036 1,106

Wave 4 violent 0–1 0.076 1,914 0.093 914 0.061 1,000

Stability violent 0–2 0.155 0.397 1,667 0.225 0.462 774 0.094 0.318 893

Wave 3 IPV 0–1 0.315 1,700 0.298 775 0.330 925

Wave 4 IPV 0–1 0.152 2,566 0.170 1,267 0.134 1,299

Stability IPV 0–2 0.474 0.632 1,689 0.478 0.637 770 0.471 0.629 919

Wave 4 sexual 0–1 0.130 2,154 0.036 1,006 0.213 1,148

Repeat victim 0–1 0.045 2,590 0.045 1,276 0.046 1,314

Biological sex 0–1 0.493 2,590 1.000 1,276 0.000 1,314

Wave 3 age 18–26 21.949 1.736 2,120 22.036 1.748 1,010 21.870 1.722 1,110

Wave 4 age 24–33 28.435 1.734 2,161 28.546 1.756 1,011 28.337 1.710 1,150

Note: Descriptive data were calculated from a single-entered data file.

Plan of Analysis

The analyses for the current study proceeded by conducting univariate ACE models for all ten victimization scales. The models were estimated using the OpenMx package in R. The results of these models provide direct estimates of the extent to which genetic, shared environmental, and nonshared environmental factors explain the variance in the Waves 3 and 4 victimization measures (Neale et al., 2016). To explain further, the ACE model partitions the variance of a given phenotype, in this case victimization, into three latent variables – an additive genetic component (A), a shared environmental component (C), and a nonshared environmental component (E).⁷ The components in the tables are standardized estimates representing the proportion of variance in each wave of victimization that is explained by that component. To illustrate, if the models revealed an additive genetic component of 0.00, then the estimate can be interpreted as indicating that 0% of the variance in that phenotype is explained by additive genetic factors. Conversely, if the model revealed an estimate of 1.00, then 100% of the variance in the phenotype can be explained by additive genetic factors. The shared environmental and nonshared environmental components (i.e., C and E, respectively) can be interpreted similarly.⁸ It is important to note that since the victimization measures used in the present study are dichotomous and categorical, the presented models are threshold liability models rather than traditional additive genetic variance (A), common (or shared) environmental factors (C), and specific (or nonshared) environmental factors plus measurement error (E) (ACE) models. ACE models were initially developed for the purposes of continuous measures but have since been modified to allow for the estimation of dichotomous and categorical measures (Neale, 2009). The threshold liability models rely on covariance estimates using tetrachoric correlation coefficients rather than more traditional Pearson correlation coefficients. These models gain their name from their ability to identify an underlying “threshold” within each dichotomous or categorical measure that represents the point in a normal distribution in which a respondent would move from one category to another. Despite these differences, the interpretation of the threshold model remains the same as the traditional univariate ACE model. Age was included in each of these models as a covariate.

One of the advantages of estimating ACE models to examine the etiology of victimization is that the ACE model allows for the estimation of nested models which fulfills two purposes – the first to determine the most parsimonious model, the second to examine quantitative sex differences in the etiology of the phenotype (Medland & Hatemi, 2009). Therefore, for each univariate model, two equations were estimated. The first was a baseline model in which separate additive genetic, shared environmental, and nonshared environmental variance components were estimated for males and females. The second equation constrained the variance estimates of the phenotype between males and females to be equal. The two equations were then compared based on model fit diagnostic tools including examining standard errors and corresponding p-values for each estimated component, and a log-likelihood ratio test (LRT). If the model estimating separate variance components for males and females revealed significantly better fit than the full model, then we can conclude that the etiological pathway to adult victimization is likely sex-specific. Qualitative sex differences were not estimated as sex-specific genes are not expected for victimization. Only quantitative sex differences were assessed as there is some prior research to suggest that the impact of certain risk factors differs between males and females (e.g., Peterson et al., 2018). For a balloon diagram of the ACE model, see Figure 1.

Figure 1.
Balloon diagram of the additive genetic variance (A), common (or shared) environmental factors (C), and specific (or nonshared) environmental factors plus measurement error (E) (ACE) model.

Results

The current study progressed in a series of interlinked steps. First, we examined whether gender-specific pathways to personal victimization, violent victimization, intimate partner violence, and sexual victimization are present in each wave of data by estimating a series of univariate ACE models. For each univariate model, two equations were estimated. The first series of models were baseline models in which we estimated separate additive genetic, shared environmental, and nonshared environmental variance components for males and females. We then tested for quantitative sex differences in the variance components by constraining the variance estimates of the phenotype between males and females to be equal allowing us to examine estimates for the full sample. The fit of the models was then compared with the model fit diagnostic tools previously discussed which can be seen in Table 2. On top of examining the standard errors of each estimated component, if a p-value for the model is significant (p < .01), the fit of the constrained model is significantly worse than the fit of the more complex model. Therefore, as can be seen in Table 2, the model fitting results demonstrated that no quantitative sex differences were detected for any of the measures.⁹

Table 2.
Model Fitting Results of the Univariate Models.

Base Comparison ep −2LL df AIC ΔLL Δdf p value

Wave 3 personal

1 ACEq 10 1169.46 1889 1189.46

2 ACEq ACE 7 1169.98 1892 1183.98 0.519 3 .915

3 ACE AE 6 1169.98 1893 1181.98 0.007 1 .933

4 ACE CE 6 1171.44 1893 1183.44 1.463 1 .226

5 ACE E 5 1179.60 1894 1189.60 9.619 2 .008

Wave 4 personal

1 ACEq 10 2141.68 1700 2161.68

2 ACEq ACE 7 2146.44 1703 2160.44 4.761 3 .190

3 ACE AE 6 2146.76 1704 2158.76 0.315 1 .575

4 ACE CE 6 2147.61 1704 2159.61 1.163 1 .281

5 ACE E 5 2148.86 1705 2158.86 2.421 2 .298

6 AE E 5 2148.86 1705 2158.86 2.106 1 .147

Stability in personal

1 ACEq 14 2373.81 1485 2401.81

2 ACEq ACE 11 2378.09 1488 2400.09 4.278 3 .233

3 ACE AE 10 2378.52 1489 2398.52 0.429 1 .513

4 ACE CE 10 2379.83 1489 2399.83 1.737 1 .188

5 ACE E 9 2381.82 1490 2399.82 3.732 2 .155

6 AE E 9 2381.82 1490 2399.82 3.303 1 .069

Wave 3 violent

1 ACEq 10 993.13 1896 1013.13

2 ACEq ACE 7 995.90 1899 1009.90 2.765 3 .429

3 ACE AE 6 996.71 1900 1008.71 0.810 1 .368

4 ACE CE 6 997.90 1900 1009.90 2.001 1 .157

5 ACE E 5 998.98 1901 1008.98 3.079 2 .214

6 AE E 5 998.98 1901 1008.98 2.269 1 .132

Wave 4 violent

1 ACEq 10 907.09 1701 927.09

2 ACEq ACE 7 909.94 1704 923.94 2.853 3 .415

3 ACE AE 6 910.08 1705 922.08 0.139 1 .709

4 ACE CE 6 910.35 1705 922.35 0.410 1 .522

5 ACE E 5 911.23 1706 921.23 1.288 2 .525

Stability in violent

1 ACEq 14 1296.19 1491 1324.19

2 ACE ACE 11 1298.00 1494 1320.00 1.809 3 .613

3 ACE AE 10 1302.50 1495 1322.50 4.499 1 .034

4 ACE CE 10 1302.43 1495 1322.43 4.430 1 .035

5 ACE E 9 1302.64 1496 1320.64 4.641 2 .098

Wave 3 IPV

1 ACEq 10 1907.47 1524 1927.47

2 ACEq ACE 7 1910.23 1527 1924.23 2.756 3 .431

3 ACE AE 6 1911.01 1528 1923.01 0.786 1 .375

4 ACE CE 6 1913.94 1528 1925.94 3.714 1 .054

5 ACE E 5 1918.95 1529 1928.95 8.727 2 .013

Wave 4 IPV

1 ACEq 10 1786.02 1897 1806.02

2 ACEq ACE 7 1786.55 1900 1800.55 0.532 3 .912

3 ACE AE 6 1786.81 1901 1798.81 0.261 1 .609

4 ACE CE 6 1787.26 1901 1799.26 0.710 1 .399

5 ACE E 5 1787.68 1902 1797.68 1.133 2 .568

Stability in IPV

1 ACEq 14 2376.13 1337 2404.13

2 ACEq ACE 11 2376.42 1340 2398.42 0.287 3 .962

3 ACE AE 10 2376.67 1341 2396.67 0.252 1 .616

4 ACE CE 10 2376.53 1341 2396.53 0.117 1 .732

5 ACE E 9 2380.84 1342 2398.84 4.426 2 .109

Wave 4 sexual

1 ACEq 10 1331.99 1909 1351.99

2 ACEq ACE 7 1336.48 1912 1350.48 4.490 3 .213

3 ACE AE 6 1337.49 1913 1349.49 1.011 1 .315

4 ACE CE 6 1336.88 1913 1348.88 0.403 1 .526

5 ACE E 5 1350.74 1914 1360.74 14.259 2 .001

Repeat victimization

1 ACEq 10 796.71 1916 816.71

2 ACEq ACE 7 797.22 1919 811.22 0.501 3 .919

3 ACE AE 6 797.30 1920 809.30 0.089 1 .766

4 ACE CE 6 797.28 1920 809.28 0.061 1 .804

5 ACE E 5 797.31 1921 807.31 0.099 2 .952

Notes: Bolded models represent models with best fit.

Abbreviations: Base = baseline model with sex differences, ep = estimated parameters, − 2LL = minus two times the log-likelihood, df = degrees of freedom, AIC = Akaike information criterion; best-fitting model in bold letter. ACEq = full model estimating sex-specific parameter estimates for additive genetic, shared environmental, and nonshared environmental influences, ACE = full model estimating parameter estimates for additive genetic, shared environmental, and nonshared environmental influences with parameter estimates held constant across sexes, AE = nested model including parameter estimates for additive genetic and nonshared environmental influences, CE = nested model including parameter estimates for shared environmental and nonshared environmental influences; E = nested model only capturing the influence of the nonshared environment. The sample used to conduct these analyses consisted of same-sex MZ twin, DZ twin, full-sibling, and half-sibling pairs.

Table 3 contains the results of the univariate ACE models examining the victimization measures. As can be seen in the table, the shared environment does not appear to explain variance in any of the victimization outcomes. For Wave 3 personal victimization, 46.1% of the variance for males and females was explained by additive genetic factors with the remainder (53.9%) explained by nonshared environmental factors. For Wave 4 personal victimization and the stability in personal victimization measures, all the variance is explained by nonshared environmental factors.

Table 3.
ACE Model Parameter Estimates for Victimization.

AGenetic influences CShared environment ENonshared environment N

Wave 3 personal

Male sample 0.461 [0.175, 0.702] 0.000 [0.000, 0.000] 0.539 [0.298, 0.825] 437

Female sample 0.461 [0.175, 0.702] 0.000 [0.000, 0.000] 0.539 [0.298, 0.825] 504

Wave 4 personal

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 355

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 403

Stability personal

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 277

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 344

Wave 3 violent

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 440

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 508

Wave 4 violent

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 356

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 403

Stability violent

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 281

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 345

Wave 3 IPV

Male sample 0.299 [0.091, 0.494] 0.000 [0.000, 0.000] 0.701 [0.506, 0.909] 268

Female sample 0.299 [0.091, 0.494] 0.000 [0.000, 0.000] 0.701 [0.506, 0.909] 364

Wave 4 IPV

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 427

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 516

Stability IPV

Male sample 0.203 [0.008, 0.389] 0.000 [0.000, 0.000] 0.797 [0.611, 0.992] 200

Female sample 0.203 [0.008, 0.389] 0.000 [0.000, 0.000] 0.797 [0.611, 0.992] 316

Wave 4 sexual

Male sample 0.467 [0.223, 0.673] 0.000 [0.000, 0.000] 0.533 [0.327, 0.777] 430

Female sample 0.467 [0.223, 0.673] 0.000 [0.000, 0.000] 0.533 [0.327, 0.777] 525

Repeat victimization

Male sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 435

Female sample 0.000 [0.000, 0.000] 0.000 [0.000, 0.000] 1.000 [1.000, 1.000] 527

Note: 95% confidence intervals in brackets; results are presented for the best-fitting model. The sample size is based on the number of sibling pairs.

Moving further down the table, the results of the models examining the etiology of violent victimization are presented. For Wave 3 violent victimization, Wave 4 violent victimization, and stability in violent victimization, all the variance is explained by nonshared environmental factors.

Turning to the results examining the intimate partner violence measures, for Wave 3 intimate partner violence, 29.9% of the variance for males and females was explained by additive genetic factors with the remaining variance explained by nonshared environmental factors (70.1%). All the variance in the Wave 4 intimate partner violence measure was explained by nonshared environmental factors. Finally, for stability in intimate partner violence, 20.3% of the variance for males and females was explained by additive genetic factors with the remaining 79.7% explained by the nonshared environment.

Unlike the previous measures of victimization, only one measure of sexual victimization was available in the Add Health data at Wave 4. Therefore, we were unable to examine differences between waves as well as stability in victimization. For Wave 4 sexual victimization, 46.7% of the variance was explained by additive genetic factors for both males and females, and the remaining 53.3% was explained by the nonshared environment. Lastly, for repeat victimization, 100% of the variance was explained by the nonshared environment.

Discussion

In the past few decades, studies examining the origins of criminal victimization have expanded to investigate risk factors that might explain male-female differences in victimization. As this area of research has expanded, more risk factors of victimization ranging from personality traits to environmental risk factors have been identified as being moderated by biological sex (e.g., Fisher et al., 2002; Fisher et al., 2010; Reyns et al., 2016). The research questions, therefore, have shifted away from whether there are male-female differences in victimization to what are the risk factors that distinguish these two groups? The current study added to this line of research by using behavioral genetic methods to assess whether the magnitude of genetic and environmental influences on adult victimization differs between males and females.

The results presented herein suggest that there are likely no significant quantitative genetic and environmental sex differences in the etiology of adult victimization. None of our models detected a significant difference between males and females in the proportion of variance in victimization that was explained by additive genetic, shared environmental, and nonshared environmental influences. A possible explanation for this null finding may lie in the results of studies examining the link between age and victimization. To be specific, previous research has shown that as individuals age, they are less risk-prone and less likely to engage in criminal behavior and, therefore, are less likely to be victimized (e.g., Sant’Anna et al., 2016). Due to the fact our sample at Waves 3 and 4 was well into adulthood and, on average, in their mid- to late-twenties (x̄ ≈ 28; range = 24–34) by Wave 4, it may be that this natural “aging out” effect of crime reduced the differences in risk factors between males and females in adult victimization leading to our null finding. Future research should examine whether risk factors associated with male and female victimization experiences are dependent on the age of the respondents as well as whether they engage in criminal behavior.

Additionally, genetic factors were found to significantly influence some of the victimization measures for both males and females. Specifically, the models examining Wave 3 personal victimization, Wave 3 intimate partner violence, stability in intimate partner violence, and Wave 4 sexual victimization were all revealed to be significantly influenced by genetic factors. Variations in the remaining measures of victimization were totally explained by the nonshared environment. These findings have important implications for future victimization research and theory. For example, the differences in results between waves of personal victimization and intimate partner violence suggest that even within the broad category of “adulthood,” there may be significant developmental phases that need to be accounted for in future research examining the etiology of adulthood victimization. To illustrate, approximately 46% of the variance in Wave 3 personal victimization, when the sample was in their late teens to early twenties, was explained by genetic factors. However, once that same sample reached their mid- to late-twenties, genetic factors were shown to no longer significantly impact personal victimization. A similar trend was revealed for the models examining intimate partner violence. Therefore, from these analyses, it is possible to conclude that as one ages the importance of nonshared environmental factors increases. These findings fall in line with prior research showing similar trends of influence of the nonshared environment as individuals age (e.g., TenEyck & Barnes, 2018).

It is interesting to note there are two victimization measures that did not follow a similar trend – the sexual victimization measure and violent victimization measures. To be specific, the Wave 4 sexual victimization measure was the only Wave 4 measure to reveal significant genetic influences. Further research is needed to understand why genetic factors may influence sexual victimization farther into adulthood than other types of victimization. The results examining violent victimization were also markedly different from the other types of victimization to be measured. Genetic factors were revealed to not significantly influence violent victimization at Waves 3 and 4 as well as the stability victimization measure. This result is noteworthy as the majority of evolutionary work that discusses evolutionary theory, and the non-random selection of genetic variants has focused on violent victimization (e.g., Buss, 2005; Daly & Wilson, 1997). One possible explanation for these null results may be that the results are a consequence of the sampling method and the way in which “violent victimization” is measured. Add Health is a school-based sample meaning that the initial population from which Add Health drew its sample were adolescents who were attending school. More than likely, this sample would not contain individuals who were engaging in seriously risky acts that would increase their risk for seriously violent victimizations. Therefore, the current sample may not be representative of the population that would most likely be violently victimized. Additionally, the majority of evolutionary work that has been done on this topic has focused specifically on homicide victimizations (e.g., Buss, 2005; Daly & Wilson, 1997). It is possible that the conclusions drawn from previous studies do not extend to other types of violent victimizations and, therefore, would explain the difference in results.

It is important to note that although the shared environment was not significant for males or females in any of the models in this study, these null findings do not negate the importance of contextual-level factors. Rather, the null findings suggest that future research examining the influence of potential community-based or contextual-level factors should be examined as an interactive effect with individual-level factors. To explain, the difference between shared environmental risk factors and nonshared environmental risk factors is very fine. The nonshared environment is typically thought of as differences in experiences of environmental factors between siblings, such as when siblings have different peers. However, the nonshared environment can also arise when siblings interpret the same event or experience differently from one another (Turkheimer & Waldron, 2000). For example, siblings may react differently to an event they both experience such as exposure to neighborhood violence and parents divorcing. Therefore, studies demonstrating that the effect of these structural-level factors are contingent on the individual experiencing them, is not unexpected and establishes how these factors should be investigated in the future – not as a universal constant but as contingent on the individual. This finding likely does not come as a surprise as much of the research that has examined contextual-level factors have shown them to be contingent on individual-level factors such as marital status (e.g., Dugan, 2003; Dugan et al., 2003; Xie et al., 2012).

With these findings in mind, there are a few limitations that need to be addressed in future research. First, the victimization measures used in this paper are based on self-reports which inevitably raise concerns over the validity and reliability of the measures (Krohn et al., 2010). Even so, these victimization measures have been used repeatedly in prior research and have been shown to be reliable and valid (e.g., Barnes et al., 2012). Future research would benefit from comparing the current results to those examining the etiology of official records of victimization. Lastly, the current study only examines quantitative sex differences in the etiology of victimization and does not examine potential qualitative differences. As has been discussed previously, quantitative sex differences refer to differences in the magnitude of genetic, shared environmental, and non-shared environmental influences between the sexes. In contrast, qualitative sex differences refer to sex differences in the actual genetic or shared environmental factors influencing the analyzed phenotype. Future research should use other genetically informed methods to examine the differential effect of potential risk factors of victimization for males and females.

As research on male- and female-specific pathways to victimization expands, results of behavioral genetic models, such as those presented here, may be used as a guiding framework from which to more fully specify sex-specific risk for victimization. The models herein revealed the importance of genetic and nonshared environmental influences in explaining adult victimization experiences. Findings such as those presented in the current study support calls for theoretical refinement and the introduction of interdisciplinary methods to the study of victimization, including methods used by behavioral geneticists and evolutionary psychologists.

Supplemental Material

sj-docx-1-evp-10.1177_14747049241267950 - Supplemental material for Sex Differences in the Etiology of Victimization in Adulthood

Supplemental material, sj-docx-1-evp-10.1177_14747049241267950 for Sex Differences in the Etiology of Victimization in Adulthood by Bridget Joyner-Carpanini and Kevin M. Beaver in Evolutionary Psychology

		Full sample	Male sample	Female sample
Wave 3 personal	0–1	0.101		2,103	0.167		1,001	0.050		1,102
Wave 4 personal	0–1	0.335		1,913	0.395		913	0.279		1,000
Stability personal	0–2	0.432	0.595	1,661	0.553	0.649	770	0.327	0.521	891
Wave 3 violent	0–1	0.079		2,111	0.126		1,005	0.036		1,106
Wave 4 violent	0–1	0.076		1,914	0.093		914	0.061		1,000
Stability violent	0–2	0.155	0.397	1,667	0.225	0.462	774	0.094	0.318	893
Wave 3 IPV	0–1	0.315		1,700	0.298		775	0.330		925
Wave 4 IPV	0–1	0.152		2,566	0.170		1,267	0.134		1,299
Stability IPV	0–2	0.474	0.632	1,689	0.478	0.637	770	0.471	0.629	919
Wave 4 sexual	0–1	0.130		2,154	0.036		1,006	0.213		1,148
Repeat victim	0–1	0.045		2,590	0.045		1,276	0.046		1,314
Biological sex	0–1	0.493		2,590	1.000		1,276	0.000		1,314
Wave 3 age	18–26	21.949	1.736	2,120	22.036	1.748	1,010	21.870	1.722	1,110
Wave 4 age	24–33	28.435	1.734	2,161	28.546	1.756	1,011	28.337	1.710	1,150

	Base	Comparison	ep	−2LL	df	AIC	ΔLL	Δdf	p value
Wave 3 personal
1	ACEq		10	1169.46	1889	1189.46
2	ACEq	ACE	7	1169.98	1892	1183.98	0.519	3	.915
3	ACE	AE	6	1169.98	1893	1181.98	0.007	1	.933
4	ACE	CE	6	1171.44	1893	1183.44	1.463	1	.226
5	ACE	E	5	1179.60	1894	1189.60	9.619	2	.008
Wave 4 personal
1	ACEq		10	2141.68	1700	2161.68
2	ACEq	ACE	7	2146.44	1703	2160.44	4.761	3	.190
3	ACE	AE	6	2146.76	1704	2158.76	0.315	1	.575
4	ACE	CE	6	2147.61	1704	2159.61	1.163	1	.281
5	ACE	E	5	2148.86	1705	2158.86	2.421	2	.298
6	AE	E	5	2148.86	1705	2158.86	2.106	1	.147
Stability in personal
1	ACEq		14	2373.81	1485	2401.81
2	ACEq	ACE	11	2378.09	1488	2400.09	4.278	3	.233
3	ACE	AE	10	2378.52	1489	2398.52	0.429	1	.513
4	ACE	CE	10	2379.83	1489	2399.83	1.737	1	.188
5	ACE	E	9	2381.82	1490	2399.82	3.732	2	.155
6	AE	E	9	2381.82	1490	2399.82	3.303	1	.069
Wave 3 violent
1	ACEq		10	993.13	1896	1013.13
2	ACEq	ACE	7	995.90	1899	1009.90	2.765	3	.429
3	ACE	AE	6	996.71	1900	1008.71	0.810	1	.368
4	ACE	CE	6	997.90	1900	1009.90	2.001	1	.157
5	ACE	E	5	998.98	1901	1008.98	3.079	2	.214
6	AE	E	5	998.98	1901	1008.98	2.269	1	.132
Wave 4 violent
1	ACEq		10	907.09	1701	927.09
2	ACEq	ACE	7	909.94	1704	923.94	2.853	3	.415
3	ACE	AE	6	910.08	1705	922.08	0.139	1	.709
4	ACE	CE	6	910.35	1705	922.35	0.410	1	.522
5	ACE	E	5	911.23	1706	921.23	1.288	2	.525
Stability in violent
1	ACEq		14	1296.19	1491	1324.19
2	ACE	ACE	11	1298.00	1494	1320.00	1.809	3	.613
3	ACE	AE	10	1302.50	1495	1322.50	4.499	1	.034
4	ACE	CE	10	1302.43	1495	1322.43	4.430	1	.035
5	ACE	E	9	1302.64	1496	1320.64	4.641	2	.098
Wave 3 IPV
1	ACEq		10	1907.47	1524	1927.47
2	ACEq	ACE	7	1910.23	1527	1924.23	2.756	3	.431
3	ACE	AE	6	1911.01	1528	1923.01	0.786	1	.375
4	ACE	CE	6	1913.94	1528	1925.94	3.714	1	.054
5	ACE	E	5	1918.95	1529	1928.95	8.727	2	.013
Wave 4 IPV
1	ACEq		10	1786.02	1897	1806.02
2	ACEq	ACE	7	1786.55	1900	1800.55	0.532	3	.912
3	ACE	AE	6	1786.81	1901	1798.81	0.261	1	.609
4	ACE	CE	6	1787.26	1901	1799.26	0.710	1	.399
5	ACE	E	5	1787.68	1902	1797.68	1.133	2	.568
Stability in IPV
1	ACEq		14	2376.13	1337	2404.13
2	ACEq	ACE	11	2376.42	1340	2398.42	0.287	3	.962
3	ACE	AE	10	2376.67	1341	2396.67	0.252	1	.616
4	ACE	CE	10	2376.53	1341	2396.53	0.117	1	.732
5	ACE	E	9	2380.84	1342	2398.84	4.426	2	.109
Wave 4 sexual
1	ACEq		10	1331.99	1909	1351.99
2	ACEq	ACE	7	1336.48	1912	1350.48	4.490	3	.213
3	ACE	AE	6	1337.49	1913	1349.49	1.011	1	.315
4	ACE	CE	6	1336.88	1913	1348.88	0.403	1	.526
5	ACE	E	5	1350.74	1914	1360.74	14.259	2	.001
Repeat victimization
1	ACEq		10	796.71	1916	816.71
2	ACEq	ACE	7	797.22	1919	811.22	0.501	3	.919
3	ACE	AE	6	797.30	1920	809.30	0.089	1	.766
4	ACE	CE	6	797.28	1920	809.28	0.061	1	.804
5	ACE	E	5	797.31	1921	807.31	0.099	2	.952

	AGenetic influences	CShared environment	ENonshared environment	N
Wave 3 personal
Male sample	0.461 [0.175, 0.702]	0.000 [0.000, 0.000]	0.539 [0.298, 0.825]	437
Female sample	0.461 [0.175, 0.702]	0.000 [0.000, 0.000]	0.539 [0.298, 0.825]	504
Wave 4 personal
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	355
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	403
Stability personal
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	277
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	344
Wave 3 violent
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	440
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	508
Wave 4 violent
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	356
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	403
Stability violent
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	281
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	345
Wave 3 IPV
Male sample	0.299 [0.091, 0.494]	0.000 [0.000, 0.000]	0.701 [0.506, 0.909]	268
Female sample	0.299 [0.091, 0.494]	0.000 [0.000, 0.000]	0.701 [0.506, 0.909]	364
Wave 4 IPV
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	427
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	516
Stability IPV
Male sample	0.203 [0.008, 0.389]	0.000 [0.000, 0.000]	0.797 [0.611, 0.992]	200
Female sample	0.203 [0.008, 0.389]	0.000 [0.000, 0.000]	0.797 [0.611, 0.992]	316
Wave 4 sexual
Male sample	0.467 [0.223, 0.673]	0.000 [0.000, 0.000]	0.533 [0.327, 0.777]	430
Female sample	0.467 [0.223, 0.673]	0.000 [0.000, 0.000]	0.533 [0.327, 0.777]	525
Repeat victimization
Male sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	435
Female sample	0.000 [0.000, 0.000]	0.000 [0.000, 0.000]	1.000 [1.000, 1.000]	527

Footnotes

Acknowledgments

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. Special acknowledgment is due to 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 (). No direct support was received from grant P01- HD31921 for this analysis.

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.

ORCID iD

Bridget Joyner-Carpanini

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Barnes

J. C.

Boutwell

B. B.

Beaver

K. M.

Gibson

C. L.

Wright

J. P.

(2014). On the consequences of ignoring genetic influences in criminological research. Journal of Criminal Justice, 42(6), 471–482. https://doi.org/10.1016/j.jcrimjus.2014.08.003

Barnes

J. C.

Boutwell

B. B.

Fox

K. A.

(2012). The effect of gang membership on victimization: A behavioral genetic explanation. Youth Violence and Juvenile Justice, 10(3), 227–244. https://doi.org/10.1177/1541204011429948

Beaver

K. M.

Boutwell

B. B.

Barnes

J. C.

Cooper

J. A.

(2009). The biosocial underpinnings to adolescent victimization: Results from a longitudinal sample of twins. Youth Violence and Juvenile Justice, 7(3), 223–238. https://doi.org/10.1177/1541204009333830

Beaver

K. M.

Boutwell

B. B.

Barnes

J. C.

DeLisi

Vaughn

M. G.

(2013). Exploring the genetic origins of adolescent victimization in a longitudinal sample of adoptees. Victims & Offenders, 8(2), 148–163. https://doi.org/10.1080/15564886.2012.763197

Beaver

K. M.

Joyner

(2021). Biosocial perspectives on victimization. In Pratt

T. C.

Turanovic

J. J.

(Eds.), Revitalizing victimization theory: Revisions, applications, and new directions (pp. 237–253). Routledge. https://doi.org/10.4324/9781003159629-13

Bezdijan

Raine

Baker

L. A.

Lynam

D. R.

(2011). Psychopathic personality in children: Genetic and environmental contributions. Psychological Medicine, 41(3), 589–600. https://doi.org/10.1017/S0033291710000966

Brenan

(2022). Worry about crime in U.S. at highest level since 2016. Retrieved from https://news.gallup.com/poll/391610/worry-crime-highest-level-2016.aspx

Buss

D. M

. (2005). The handbook of evolutionary psychology . John Wiley & Sons, Inc. https://doi.org/10.1002/9780470939376

Buss

D. M

. (2009). How can evolutionary psychology successfully explain personality and individual differences? Perspectives on Psychological Science, 4(4), 359–366. https://doi.org/10.1111/j.1745-6924.2009.01138.x

10.

Catalano

Smith

Snyder

Rand

(2009). Female victims of violence. U.S. Department of Justice Office of Justice Programs. Retrieved from https://digitalcommons.unl.edu/usjusticematls/7/

11.

Cohen

L. E.

Felson

(1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608. https://doi.org/10.2307/2094589

12.

Connolly

E. J.

Hayes

B. E.

Boisvert

D. L.

Cooke

E. M

. (2022). Intimate partner victimization and depressive symptoms: Approaching causal inference using a longitudinal twin design. Journal of Quantitative Criminology, 38, 517–535. https://doi.org/10.1007/s10940-021-09509-6

13.

Conroy-Beam

Roney

J. R.

Lukaszewski

A. W.

Buss

D. M.

Asao

Sorokowska

Sorokowski

Aavik

Akello

Alhabahba

M. M.

Alm

Amjad

Anjum

Atama

C. S.

Duyar

D. A.

Ayebare

Batres

Bendixen

Bensafia

, … & Zupančič

(2019). Assortative mating and the evolution of desirability covariation. Evolution and Human Behavior, 40(5), 479–491. https://doi.org/10.1016/j.evolhumbehav.2019.06.003

14.

Daly

Wilson

. (1997). Crime and conflict: Homicide in evolutionary psychological perspective. Crime and Justice, 22, 51–100. https://doi.org/10.1086/449260

15.

De Vries

L. P.

van Beijsterveldt

T. C. E. M.

Maes

Colodro-Conde

Bartels

(2021). Genetic influences on the covariance and genetic correlations in a bivariate twin model: An application to well-being. Behavior Genetics, 51, 191–203. https://doi.org/10.1007/s10519-021-10046-y

16.

Dugan

(2003). Domestic violence legislation: Exploring its impact on the likelihood of domestic violence, police involvement, and arrest. Criminology & Public Policy, 2(2), 283–312. https://doi.org/10.1111/j.1745-9133.2003.tb00126.x

17.

Dugan

Nagin

D. S.

Rosenfeld

(2003). Exposure reduction or retaliation? The effects of domestic violence resources on intimate-partner homicide. Law & Society Review, 37(1), 169–198. https://doi.org/10.1111/1540-5893.3701005

18.

Duntley

J. D.

Schackelford

T. K

. (2012). Adaptations to avoid victimization. Aggression and Violent Behavior, 17(1), 59–71. https://doi.org/10.1016/j.avb.2011.09.008

19.

Farrell

. (2020). Exploring the overlap between sexual victimization and offending among young women across neighborhoods: Does the type of force and type of offending matter? Journal of Interpersonal Violence, 35(3-4), 571–599. https://doi.org/10.1177/0886260516689778

20.

Fisher

B. S.

Cullen

F. T.

Turner

M. G.

(2002). Being pursued: Stalking victimization in a national study of college women. Criminology & Public Policy, 1(2), 257–308. https://doi.org/10.1111/j.1745-9133.2002.tb00091.x

21.

Fisher

B. S.

Daigle

L. E.

Cullen

F. T.

(2010). What distinguishes single recurrent sexual victims? The role of lifestyle-routine activities and first-incident characteristics. Justice Quarterly, 27(1), 102–129. https://doi.org/10.1080/07418820902763061

22.

Flexon

J. L.

Meldrum

R. C.

Piquero

A. R.

(2016). Low self-control and the victim-offender overlap: A gendered analysis. Journal of Interpersonal Violence, 31(11), 2052–2076. https://doi.org/10.1177/0886260515572471

23.

Fontaine

N. M. G.

Rijsdijk

F. V.

McCrory

E. J. P.

Viding

(2010). Etiology of different development trajectories of callous-unemotional traits. Journal of the American Academy of Child & Adolescent Psychiatry, 49(7), 656–664. https://doi.org/10.1016/j.jaac.2010.03.014

24.

Gartner

(2011). Sex, gender, and crime. In Tonry

(Ed.), The Oxford handbook of crime and criminal justice (pp. 348–384). Oxford University Press.

25.

Goldstein

Prescott

C. A.

Kendler

K. S.

(2001). Genetic and environmental factors in conduct problems and adult antisocial behavior among adult female twins. The Journal of Nervous and Mental Disease, 189(4), 201–209. https://doi.org/10.1097/00005053-200104000-00001

26.

Harden

K. P.

Mendle

Hill

J. E.

Turkheimer

Emery

R. E.

(2008). Rethinking time of first sex and delinquency. Journal of Youth and Adolescence, 37(4), 373–385. https://doi.org/10.1007/s10964-007-9228-9

27.

Harris

K. M.

(2009). The national longitudinal study of adolescent health (Add Health), Waves I & II, 1994-1996; Wave III, 2001-2002; Wave IV, 2007-2009 [machine-readable file and documentation]. Carolina Population Center.

28.

Harris

K. M.

(2013). The Add Health study: Design and accomplishments. Carolina Population Center. https://doi.org/10.17615/C6TW87

29.

Harris

K. M.

Halpern

C. T.

Haberstick

B. C.

Smolen

(2013). The national longitudinal study of adolescent health (add health) sibling data. Twin Research and Human Genetics, 16(1), 391–398. https://doi.org/10.1017/thg.2012.137

30.

Haynie

D. L.

McHugh

(2003). Sibling deviance: In the shadows of mutual and unique friendship effects? Criminology; An Interdisciplinary Journal, 41(2), 355–392. https://doi.org/10.1111/j.1745-9125.2003.tb00991.x

31.

Hines

D. A.

Saudino

K. J.

(2004). Genetic and environmental influences on intimate partner aggression: A preliminary study. Violence and Victims, 19(6), 701–718. https://doi.org/10.1891/vivi.19.6.701.66341

32.

Jacobson

K. C.

Rowe

D. C.

(1998). Genetic and shared environmental influences on adolescent BMI: Interactions with race and sex. Behavior Genetics, 28, 265–278. https://doi.org/10.1023/A:1021619329904

33.

Jennings

W. G.

Higgins

G. E.

Tewksbury

Gover

A. R.

Piquero

A. R.

(2010). A longitudinal assessment of the victim-offender overlap. Journal of Interpersonal Violence, 25(12), 2147–2174. https://doi.org/10.1177/0886260509354888

34.

Kavish

Connolly

E. J.

Boutwell

B. B.

(2019). Genetic and environmental contributions to the association between violent victimization and major depressive disorder. Personality and Individual Differences, 140, 103–110. https://doi.org/10.1016/j.paid.2018.05.034

35.

Krohn

M. D.

Thornberry

T. P.

Gibson

C. L.

Baldwin

J. M.

(2010). The development and impact of self-report measures of crime and delinquency. Journal of Quantitative Criminology, 26, 509–525. https://doi.org/10.1007/s10940-010-9119-1

36.

Kruttschnitt

Kang

(2021). Theoretical perspectives on the victimization of women. In Pratt

T. C.

Turanovic

J. J.

(Eds.), Revitalizing victimization theory: Revisions, applications, and new directions (pp. 163–184). Routledge.

37.

Lauritsen

J. L.

Heimer

(2008). The gender gap in violent victimization, 1973-2004. Journal of Quantitative Criminology, 24, 125–147. https://doi.org/10.1007/s10940-008-9041-y

38.

Lauritsen

J. L.

Rezey

M. L.

(2018). Victimization trends and correlates: Macro- and microinfluences and new directions for research. Annual Review of Criminology, 1, 103–121. https://doi.org/10.1146/annurev-criminol-032317-092202

39.

LeSuer

(2020). An international study of the contextual effects of gender inequality on intimate partner sexual violence against women students. Feminist Criminology, 15(1), 97–118. https://doi.org/10.1177/1557085119842652

40.

Medland

S. E.

Hatemi

P. K.

(2009). Political science, biometric theory, and twin studies: A methodological introduction. Political Analysis, 17(2), 191–214. https://doi.org/10.1093/pan/mpn016

41.

Neale

M. C.

(2009). Biometrical models in behavioral genetics. In Kim

(Eds.), Handbook of behavior genetics. Springer. https://doi.org/10.1007/978-0-387-76727-7_2

42.

Neale

M. C.

Cardon

L. R.

(1992). Methodology for genetic studies of twins and families. Springer Science+Business Media Dordrecht. https://doi.org/10.1007/978-94-015-8018-2

43.

Neale

M. C.

Hunter

M. D.

Pritikin

J. N.

Zahery

Brick

T. R.

Kirkpatrick

R. M.

Estabrook

Bates

T. C.

Maes

H. H.

Boker

S. M.

(2016). Openmx 2.0: Extended structural equation and statistical modeling. Psychomtrika, 81, 535–549. https://doi.org/10.1007/s11336-014-9435-8

44.

Peterson

Lasky

N. V.

Fisher

B. S.

Wilcox

(2018). Gendered opportunity and school-based victimization: An integrated approach. Youth Violence and Juvenile Justice, 16(2), 137–155. https://doi.org/10.1177/1541204016680402

45.

Planty

Langton

Krebs

Berzofsky

Smiley-McDonald

(2013). Female victims of sexual violence, 1994-2010. U.S. Department of Justice. Retrieved from https://fathersmanifesto.net/rape19952010.pdf

46.

Pratt

T. C.

Turanovic

J. J.

(2021). Introduction: Revitalizing victimization theory. In Pratt

T. C.

Turanovic

J. J.

(Eds.), Revitalizing victimization theory: Revisions, applications, and new directions (pp. 1–12). Routledge.

47.

Pratt

T. C.

Turanovic

J. J.

Fox

K. A.

Wright

K. A

. (2014). Self-control and victimization: A meta-analysis. Criminology, 52(1), 87–116. https://doi.org/10.1111/1745-9125.12030

48.

Rand

M. R.

(2009). Criminal victimization, 2008. U.S. Department of Justice. Retrieved from https://bjs.ojp.gov/content/pub/pdf/cv08.pdf

49.

Reyns

B. W.

Henson

Fisher

B. S.

Fox

K. A.

Nobles

M. R.

(2016). A gendered lifestyle-routine activity approach to explaining stalking victimization in Canada. Journal of Interpersonal Violence, 31(9), 1719–1743. https://doi.org/10.1177/0886260515569066

50.

Sant’Anna

Scorzafave

L. G.

Justus

(2016). Nonlinear relationship between income, age and criminal victimization in Brazil. Economia (Pontificia Universidad Catolica Del Peru. Departamento De Economia), 17(2), 185–198. https://doi.org/10.1016/j.econ.2016.05.002

51.

Semenza

D. C.

Testa

Turanovic

J. J

. (2021). Trajectories of violent victimization over the life course: Implications for mental and physical health. Advances in Life Course Research, 50, 100436. https://doi.org/10.1016/j.alcr.2021.100436

52.

TenEyck

Barnes

J. C.

(2018). Exploring the social and individual differences among victims, offenders, victim-offenders, and total abstainers. Victims & Offenders, 13(1), 66–83. https://doi.org/10.1080/15564886.2016.1268985

53.

Thompson

Tapp

S. N.

(2022). Criminal victimization, 2021. U.S. Department of Justice. Retrieved from https://bjs.ojp.gov/content/pub/pdf/cv21.pdf

54.

Tooby

Cosmides

. (1990). The past explains the present: Emotional adaptations and the structure of ancestral environments. Ethology and Sociobiology, 11(4-5), 375–424. https://doi.org/10.1016/0162-3095(90)90017-Z

55.

Turanovic

J. J.

Pratt

T. C.

(2019). Thinking about victimization: Context and consequences. Routledge.

56.

Turkheimer

Waldron

(2000). Nonshared environment: A theoretical, methodological, and quantitative review. Psychological Bulletin, 126(1), 78–108. https://doi.org/10.1037/0033-2909.126.1.78

57.

Viding

Frick

P. J.

Plomin

(2007). Aetiology of the relationship between callous-unemotional traits and conduct problems in childhood. The British Journal of Psychiatry, 190(S49), S33–S38. https://doi.org/10.1192/bjp.190.5.s33

58.

Vierikko

Pulkkinen

Kaprio

Viken

Rose

R. J.

(2003). Sex differences in genetic and environmental effects on aggression. Aggressive Behavior, 29(1), 55–68. https://doi.org/10.1002/ab.10038

59.

Willie

T. C.

Kershaw

T. S.

(2019). An ecological analysis of gender inequality and intimate partner violence in the United States. Preventive Medicine, 118, 257–263. https://doi.org/10.1016/j.ypmed.2018.10.019

60.

Wyckoff

J. P.

Buss

D. M.

Markman

A. B

. (2019). Sex differences in victimization and consequences of cyber aggression. An evolutionary perspective. Evolutionary Behavioral Sciences, 13(3), 254–264. https://doi.org/10.1037/ebs0000150

61.

Xie

Heimer

Lauritsen

J. L.

(2012). Violence against women in U.S. Metropolitan areas: Changes in women’s status and risk, 1980-2004. Criminology; An Interdisciplinary Journal, 50(1), 105–143. https://doi.org/10.1111/j.1745-9125.2011.00245.x

62.

Xie

Lauritsen

J. L.

Heimer

(2012). Intimate partner violence in U.S. Metropolitan areas: The contextual influences of police and social services. Criminology; An Interdisciplinary Journal, 50(4), 961–992. https://doi.org/10.1111/j.1745-9125.2012.00284.x

63.

Zheng

Cleveland

H. H.

(2015). Differential genetic and environmental influences on developmental trajectories of antisocial behavior from adolescence to young adulthood. Journal of Adolescence, 45, 204–213. https://doi.org/10.1016/j.adolescence.2015.10.006

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

		Full sample			Male sample			Female sample
	Range	Mean	SD	N	Mean	SD	N	Mean	SD	N
Wave 3 personal	0–1	0.101		2,103	0.167		1,001	0.050		1,102
Wave 4 personal	0–1	0.335		1,913	0.395		913	0.279		1,000
Stability personal	0–2	0.432	0.595	1,661	0.553	0.649	770	0.327	0.521	891
Wave 3 violent	0–1	0.079		2,111	0.126		1,005	0.036		1,106
Wave 4 violent	0–1	0.076		1,914	0.093		914	0.061		1,000
Stability violent	0–2	0.155	0.397	1,667	0.225	0.462	774	0.094	0.318	893
Wave 3 IPV	0–1	0.315		1,700	0.298		775	0.330		925
Wave 4 IPV	0–1	0.152		2,566	0.170		1,267	0.134		1,299
Stability IPV	0–2	0.474	0.632	1,689	0.478	0.637	770	0.471	0.629	919
Wave 4 sexual	0–1	0.130		2,154	0.036		1,006	0.213		1,148
Repeat victim	0–1	0.045		2,590	0.045		1,276	0.046		1,314
Biological sex	0–1	0.493		2,590	1.000		1,276	0.000		1,314
Wave 3 age	18–26	21.949	1.736	2,120	22.036	1.748	1,010	21.870	1.722	1,110
Wave 4 age	24–33	28.435	1.734	2,161	28.546	1.756	1,011	28.337	1.710	1,150