Victims & Offenders An International Journal of Evidence-based Research, Policy, and Practice ISSN: 1556-4886 (Print) 1556-4991 (Online) Journal homepage: http://www.tandfonline.com/loi/uvao20 Reaping What They Sow? Victim-Offender Overlap Perceptions and Victim Blaming Attitudes Christina Mancini & Justin T. Pickett To cite this article: Christina Mancini & Justin T. Pickett (2015): Reaping What They Sow? Victim-Offender Overlap Perceptions and Victim Blaming Attitudes, Victims & Offenders, DOI: 10.1080/15564886.2015.1093051 To link to this article: http://dx.doi.org/10.1080/15564886.2015.1093051 Published online: 09 Nov 2015. Submit your article to this journal Article views: 18 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=uvao20 Download by: [University of Warwick] Date: 30 December 2015, At: 07:24 Downloaded by [University of Warwick] at 07:24 30 December 2015 Victims & Offenders, 00:1–33, 2016 Copyright © Taylor & Francis Group, LLC ISSN: 1556-4886 print/1556-4991 online DOI: 10.1080/15564886.2015.1093051 Reaping What They Sow? Victim-Offender Overlap Perceptions and Victim Blaming Attitudes Christina Mancini L. Douglas Wilder School of Government and Public Affairs, Virginia Commonwealth University, Richmond, Virginia, USA Justin T. Pickett School of Criminal Justice, University at Albany, State University of New York, Albany, New York, USA Abstract: Examining the tendency to attribute blame to crime victims reveals a striking dichotomy. Some types, such as children, elicit intense emotional reactions from the public. Alternatively, others, such as the typical victims of street crimes, garner substantially less concern. According to the “just world” hypothesis, these latter groups may be perceived by the public as criminally involved, and so “blameworthy” for their victimization. We test this hypothesis—specifically, we evaluate whether perceptions of the extent of victims’ involvement in crime are associated with dispositional attributions for victimization. Data from a recent national survey (N = 760) are analyzed. To extend generalizability, we replicate results with a college sample (N = 733). Findings indicate that victim-offender overlap perceptions vary consistently by crime type. There is also consistent evidence that perceiving a larger victim-offender overlap is associated with the view that the causes of criminal victimization are, in part, dispositional—and thus that crime victims hold personal responsibility. Keywords: causal attributions for victimization, victim blaming, victim-offender overlap, victimology, violent crime INTRODUCTION Although not universal, victim blaming or the tendency to hold crime victims accountable for the offenses they suffer has long existed in society (Finkel, Address correspondence to Christina Mancini, Virginia Commonwealth University, L. Douglas Wilder School of Government and Public Affairs, 923 West Franklin Street, Richmond, VA 23284. E-mail: cnmancini@vcu.edu 1 Downloaded by [University of Warwick] at 07:24 30 December 2015 2 C. Mancini and J. T. Pickett 2001). Criminologists have applied the “just world” hypothesis (Lerner, 1980) to explain this phenomenon (Cullen, Clark, Mathers, & Cullen, 1983; Freeman, 2006; Grubb & Harrower, 2008). The central premise of the “just world” hypothesis argues that the public tends to hold certain victims accountable for the offenses committed against them because of perceived character flaws or immoral behavior. Blaming crime victims for the offenses they suffer serves a pragmatic purpose (Freeman, 2006). For instance, if individuals can point to some fault in the victim’s behavior, it may alleviate feelings of insecurity and fear of possibly experiencing the same misfortune. Likewise, it follows that perceived victim blame may be associated with a lower felt responsibility both for providing victim services and for taking steps to eliminate adverse social conditions (e.g., poverty, racial inequality) in order to prevent future victimizations. Even though prior studies (for a review, see Grubb & Turner, 2012) have identified this social phenomenon and some of its consequences (e.g., reduced reporting of crime, victim distrust of the criminal justice system, general victim harm), much of this research has concentrated on victim blaming of sexual assault and intimate partner victims. There is, for example, a large body of scholarship centered on rape myths (e.g., victims invite rape by wearing revealing clothing) and how their endorsement reduces levels of offender culpability among the public (for a review, see Suarez & Gadalla, 2010); similar scholarship has been performed regarding intimate partner abuse victim blame (Worden & Carlson, 2005; Yamawaki, Ochoa-Shipp, Pulsipher, Harlos, & Swindler, 2012). Comparatively little work exists that has systematically examined public judgments about victims and blame along a range of other dimensions (e.g., blame attributed to victims of other serious crimes such as homicide, robbery, and burglary), and (by extension) predictors of those views (see, generally, Policastro & Payne, 2013). In this context, the victim-offender overlap is a relevant theoretical framework for understanding negative judgments about crime victims. A mounting body of evidence suggests a nontrivial proportion of those victimized have also engaged in offending (e.g., Barnes & Beaver, 2012; Berg, Stewart, Schreck, & Simons, 2012; Jennings, Higgins, Tewksbury, Gover, & Piquero, 2010; Tillyer & Wright, 2014). Despite this growing scholarship, how the public views the victim-offender overlap and its implications for victim blaming remains an unexplored area. Two specific questions have yet to be explored by prior research. First, to what extent does the public assess victims as perpetrators? Second, are perceptions of the victim-offender overlap salient for predicting victim-blaming attitudes? The current study aims to address these gaps in prior research. In particular, as predicted by the just world hypothesis, among those who perceive that many victims are themselves involved in crime, victim-blaming attitudes can be expected to be intensified. Specifically, perceiving a greater Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming victim-offender overlap may increase individuals’ endorsement of dispositional attributions for victimization by leading them to believe that many victims— namely, victim-perpetrators—are part of a distinct population of deviants who stand apart from the larger population of noncriminally involved citizens. In the public mind, dispositional shortcomings are likely to be principal factors differentiating deviant victim/offenders from nondeviant victims, or “true victims,” and nonvictims. From a theoretical perspective, the two factors of interest—perceptions of the victim-offender overlap and victim-blaming attitudes—are conceptually distinct. Members of the public may simply believe that many people are involved in crime, regardless of whether they are victims. For example, the American public judges the probability of men (regardless of race) committing crimes to be relatively high, with a majority assuming it is more likely than not (on a “scale of 1 to 10”) that the average male will engage in crime by age 30 (Mears, Pickett, Golden, Chiricos, & Gertz, 2013, p. 282). Put simply, then, the perception may be that there is a large citizen-offender overlap, rather than a victim-offender overlap per se. In this case, the victim-offender overlap should be roughly invariant by crime type, because it would be less reflective of victims’ attributes than of the attributes (i.e., crime-prone) of citizens more generally. At the same time, while the public may believe there is considerable victim-offender overlap, they may simultaneously believe that victimization is still a relatively random but geographically concentrated event, and thus is not a result of victims’ dispositional factors or shortcomings. That is, even if members of the public perceive a high victim-offender overlap, they may believe this overlap is a consequence of victims and offenders sharing criminogenic environments (e.g., disadvantaged neighborhoods), rather than of victims’ personal moral failings. Given the above theoretical possibilities, the current study endeavors to systematically assess the predictors of victim-blaming attitudes. Because the relationship between perceiving victims-as-offenders on victim-blaming attitudes may vary across crime context (e.g., homicide versus burglary offense), we evaluate attitudes across three different crime categories—homicide, robbery, and burglary. Separately, it is also likely that other factors identified by earlier research—such as racial attitudes, concern and perceptions of crime, news and media consumption, and demographic characteristics such as sex (Suarez & Gadalla, 2010)—may contribute to the inclination to “blame the victim.” Thus, the current investigation also accounts for the effects of these theoretically important factors on endorsing dispositional attributions for victimization. Under that backdrop, and given calls to further explore the victim-offender overlap (Schreck, Stewart, & Osgood, 2008) as well as victim-blaming attitudes (Fox & Cook, 2011; Henning & Holdford, 2006), the current study tests whether perceiving that a larger percentage of victims are themselves 3 4 C. Mancini and J. T. Pickett Downloaded by [University of Warwick] at 07:24 30 December 2015 involved in crime influences endorsement of causal attributions for victimization. To increase external validity of our results, we test our hypotheses on two separate and large samples—a national sample (N = 760) and a sample of college students (N = 733). Study findings contribute to scholarship centered on explaining perceptions of victims and crime and the tendency to victim blame. CRIME VICTIMS AND THE VICTIM-OFFENDER OVERLAP To be sure, victim blaming is not universally applied. For example, according to some accounts, prominent crime legislation stemming in part from the victims’ rights movement in the 1980s has generally portrayed child victims as vulnerable and in need of special protections (Garland, 2001; Pickett, Mancini, & Mears, 2013). At the federal and state levels, numerous laws have been enacted with the goal of protecting “special” populations such as children and women. To illustrate, the Jacob Wetterling Act and Megan’s Law are examples of federal sex offender legislation named in honor of children murdered by sex offenders (Sample, 2006). Dubber (2002) argues that “nothing excites communal punitive reflex of all potential victims more . . . no victim is more helpless . . . than a child and more so, a girl” (p. 180). Others describe the “missing white woman/girl syndrome” (Levit, 2010, p. 761; Stillman, 2007, p. 491; see also, Britto, Hughes, Saltzman, & Stroh, 2007) as a condition in which an inordinate amount of attention is directed toward missing or kidnapped young, white, upper- to middle-class women and girls—and substantially less given to crime victims from other social and demographic groups. Yet, the tenor of this emphasis obscures a striking reality. The victims’ rights movement, with its narrow framing has subsequently overlooked the most typical victim base in the United States—young, urban, minority males (Cooper & Smith, 2011; Harrell, 2007), usually residing in economically depressed communities (Simon, 2007). Relatedly, but separately, for a victim to be “deserving” of public sympathy he or she must not have previously engaged in deviant or criminal behavior (Dubber, 2002; Simon, 2007). In the public’s view, involvement in such behaviors may be judged as precipitating the victimization incident, in turn reducing the sympathy and moral outrage that typically follow unprovoked interpersonal transgressions and injustices (Darley & Pittman, 2003). Publicly funded policies to assist crime victims— such as providing monetary compensation or psychological counseling—are illustrative of this theme. Typically, they include guidelines regarding who is eligible (i.e., “true” innocent victims) for victim services and support and who is not (e.g., victimized individuals who were high on drugs at the time of the crime).1 Yet substantial overlap in victimization and offending exists. Put simply, as Schreck et al. (2008, p. 894) have explained, current victim interventions Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming “ignore the fact that a large share of those who frequently encounter violence take on roles as victims as well as offenders.” These perceptions—although clearly delineating between victims and offenders—run counter to an established knowledge base that has identified the “victim-offender overlap.” Wolfgang (1957) is credited with being one of the first criminologists to uncover the blurred lines between victimization and offending. In his seminal study of homicide offenses in Philadelphia, nearly half of all murder victims had had criminal histories, and one in four was involved in crimes of interpersonal violence. Over the six decades since Wolfgang’s (1957) publication, a large body of scholarship has continued to unpack the victim-offender link. Although a complete review of the victim-offender overlap literature is beyond the scope of this article, we discuss a sampling of this scholarship below to provide context for the current investigation. For example, a 2012 review of 37 studies concluded that a “strong overlap” exists between offenders and victims, with 31 studies demonstrating considerable support for the victim-as-offender hypothesis (Jennings et al., 2010, p. 24). As but one specific example, Broidy and colleagues (2006) in a study of New Mexico homicide cases found that 50% of murder victims had had a criminal history (also, Pizarro, Zgoba, & Jennings, 2011; earlier, Dobrin, 2001). Separately, in a study analyzing the Philadelphia Birth Cohort data (N = 564), Wolfgang and Singer (1978) estimated that 42% of those with criminal records reported robbery victimization. Beyond these interpersonal and property crimes, research has explored the victim-offender overlap across a range of offenses, such as sex offending (Jespersen, Lalumière, & Seto, 2009), property offending (Daday, Broidy, Crandall, & Sklar, 2005), gang violence (Katz, Webb, Fox, & Shaffer, 2011), juvenile delinquency (Lauritsen & Laub, 2007; Posick, 2013; Turanovic & Pratt, 2013), bullying (Unnever, 2005), dating and intimate partner violence (Gover, Kaukinen, & Fox, 2008; Tillyer & Wright, 2014), stalking (Nobles, Fox, Piquero, & Piquero, 2009), and white-collar crime (Holtfreter, Reisig, Piquero, & Piquero, 2010). Clearly, the overlap between victims and criminals, as Pyrooz and colleagues (2014, p. 316) have recently contended, “is on the short list of criminological ‘facts.’” Despite this robust finding, a so far unaddressed dimension of the victim-offender overlap literature involves the extent to which the victimas-perpetrator perception exists in the public mind and influences causal attributions for victimization. This research gap merits attention because the just world hypothesis would suggest that such perceptions may shape popular judgments about victim blameworthiness (e.g., Freeman, 2006). Stated differently, the perceived victim-offender overlap may be a key factor explaining why some types of victims are judged as blameworthy, while others (e.g., child victims) are assessed as being innocent and in need of protection. Below we 5 6 C. Mancini and J. T. Pickett summarize prior victim blame scholarship relevant for the current endeavor, which in turn provides a springboard for our analysis in the section thereafter. Downloaded by [University of Warwick] at 07:24 30 December 2015 PRIOR VICTIM BLAME RESEARCH Our investigation focuses on general causal attributions for victimization, rather than on the factors that lead individuals to blame individual victims in specific situations. Unfortunately, however, there is a dearth of research that has examined causal attributions for victimization outside of the context of sexual assault and intimate partner abuse. For this reason, in this section, we review prior studies that have evaluated causal attributions for sexual victimization and intimate partner violence and situational victim blame. This victim blaming literature is that which is the most closely related, and thus potentially the most informative, about our specific research questions. In particular, much of the criminological focus on victim blaming has been concentrated in offenses against women (see Fox & Cook, 2011). For instance, Suarez and Gadalla (2010), in a large meta-analysis, identified 158 peerreviewed studies centered on victim blaming attitudes toward rape victims among the public over a ten-year period (1997 to 2007). Among the studies examined, several respondent characteristics (direction of findings presented parenthetically) were predictive of rape victim blame—sex (males were more likely to endorse rape myths), sexual aggression (higher levels of aggressive behavior were associated with greater victim blame), and racial attitudes (racial animus increased perceived victim culpability). Other work has investigated whether traits or behaviors of victims affect public assessments of responsibility. One review demonstrated an effect of the victim-offender relationship. Victims who know their rapist tend to be assigned more responsibility for a rape (Grubb & Harrower, 2008; see also Doerner & Lab, 2012). One other study conducted by Yamawaki and colleagues (2012) centered on understanding public perceptions of female victims of intimate partner violence. The researchers assigned Brigham Young University students (N = 194) to read vignettes that specified that the victim returned to the abuser or not and the relationship status of the couple (dating versus married). Results revealed that college students were more inclined to blame an intimate partner violence victim after learning that she returned to her abuser. Students who endorsed myths about domestic violence (e.g., “any healthy woman can successfully leave her abuser if she really wants to”) and received the vignette describing that the victim stayed with the perpetrator were most likely to blame victims. Additionally, male students were significantly more likely to assign responsibility to abuse victims compared to female students. Separately, research in this area has found greater victim blame attributed to women who violate traditional gender roles (e.g., have casual sexual Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming relationships, dress provocatively), victims who consumed alcohol prior to the crime (see Grubb & Turner, 2012, for a review), and women who do not leave their abusers (Yamawaki et al., 2012). Broadly, these findings can be contextualized under the just world hypothesis. At the heart of this perspective is the belief that crime victims are responsible for offenses committed against them because of an underlying character flaw or prior deviance. Extending this logic, victims who engage in socially proscribed behaviors—violating gender norms, consuming alcohol, or not leaving abusive relationships, for example— are susceptible to victim blaming because their previous actions signal that the victim’s fate was fitting and deserved, and thus less “unfair” given the circumstances. In turn, this cognitive process reinforces the notion that “people get what they deserve,” making the world appear more orderly and “just” (Hayes, Lorenz, & Bell, 2013). In a different direction, one of the only studies to identify which victim characteristics predict victim blame across different types of crime (rape and robbery) and in varying contexts (e.g., during jogging, hitchhiking) was conducted on a sample of Wisconsin college students (N = 160; Howard, 1984). Overall, blame was more likely to be attributed to female than to male victims depicted in the specific scenarios. The type of blame attributed also varied by sex of the victim: greater blame was attributed to the character (e.g., “trusting nature,” “carelessness”) of the female victim than to that of the male victim. In contrast, more blame was attributed to the behavior (e.g., “looked scared,” “did not try to fight back”) of the male victim than to that of the female victim.2 Here again, findings underscore that victim blaming among the public is activated by the presence or absence of perceived negative traits of victims (e.g., victims perceived as careless and not conscious about safety). Concomitantly, these results demonstrate that perceptions of victim culpability vary across gender and crime contexts. To our knowledge, only one other study to date, besides Howard’s (1984) seminal investigation, has assessed victim blaming across a range of diverse offenses. Using an experimental design, Fox and Cook (2011) tested whether attributions of blame toward various crimes (e.g., physical assault, intimate partner violence) were reduced among college students after completing a victimology course. Students in the treatment group expressed significantly less victim blaming beliefs toward the individuals described in the scenarios than those in the control group. This indicated that accurate knowledge obtained during the victimology course about victims and victim issues served to reduce these negative perceptions. The Fox and Cook (2011) study is salient because it underscores that, at least among college students, levels of victim blameworthiness are largely dependent on perceptions of crime victims, though the study did not specifically determine which types of victim perceptions increase victim blaming; it is also distinct from the current study 7 Downloaded by [University of Warwick] at 07:24 30 December 2015 8 C. Mancini and J. T. Pickett as it assessed blame toward the specific individuals depicted in the victim scenarios, rather than toward victim groups generally. Although this extant scholarship has contributed to a clearer interpretation of victim blaming and identified potential correlates of it, four gaps of this prior work are apparent. First, most prior victim blaming studies have focused on explaining negative judgments held toward female victims of sexual violence (Fox & Cook, 2011, p. 3408) and intimate partner abuse (e.g., Yamawaki et al., 2012). As a result, the degree to which victim blaming attitudes apply toward victims of other serious offenses—such as homicide, robbery, and burglary—is unknown. By extension, there has been an absence of empirical attention directed at examining the typical victim base in the United States—young, minority males—that experiences these routine forms of criminal violence (Cooper & Smith, 2011; Harrell, 2007). Put differently, this gap is relevant given that non-sex-crime scenarios (e.g., burglary, robbery offenses) may trigger markedly different stereotypes among the public than those that are activated for sexual assault and intimate partner abuse crimes. Another gap involves victim measurement. Of the small handful of studies that have examined non-sex-crime and non–intimate partner violence victims, most have restricted perceptions to victims depicted as engaging in particular behaviors in specific scenarios or vignettes. As an example Fox and Cook (2011, p. 3414), in gauging attitudes toward a victim of physical assault, used this measure: “A sober man who intentionally flirts with another man’s girlfriend at a bar shares the blame for the other man punching him.” Accordingly, the measure taps only into views held about this specific type of victim (i.e., a sober man assaulted in a bar due to a relational conflict), rather than the general population of physical assault victims, many of whom are attacked in other contexts and for reasons other than relational conflicts. An important consequence of the focus in prior studies on blame attributed to specific victims is that little is known about the public’s general causal attributions for victimization. Accordingly, it is not currently clear what leads persons to believe that crime victims are generally responsible for their victimizations, rather than to believe that victimization results from factors outside of victims’ control. Third, the two studies examining attitudes beyond sexual and intimate partner violence victimization (Fox & Cook, 2011; Howard, 1984) have relied exclusively on college students with no examination of the generalizability of their findings beyond this population. Moreover, one of those studies, Howard (1984), while examining predictors of victim blame across offenses, drew on data collected more than 30 years ago. The past three decades have witnessed dramatic social changes, including declines in crime and increases in population heterogeneity. Thus, Howard’s (1984) findings may not generalize to contemporary society. Fourth, and finally, it is unclear to what extent perceptions of the victim-offender overlap—a well-documented phenomenon in the offending Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming literature—affect judgments of blame and culpability for victimization. According to the just world hypothesis, we can expect that perceiving a larger victim-offender overlap will induce greater attributions of blame because the public is “threatened by instances of injustice” and thus “motivated to reduce this threat to maintain the appearance that the world metes out resources and ill fate as deserved” (Hafer & Bègue, 2005, p. 129); even so, this remains a speculative assertion given the absence of empirical attention. In short, our study will be the first to systematically examine victim blaming across various offenses among a sample of the public and a sample of college students. In the next section, we outline the study’s hypotheses. THE CURRENT STUDY As discussed earlier, the propensity to hold people responsible for their victimization can be understood under the just world framework. That is, individuals often equate “what is with what is right” (Hunzaker, 2014, p. 166, emphasis added). This perception, as emphasized by Rye and colleagues (2006, p. 640), “makes the world seem orderly, controllable, and stable . . . [in turn] protect[ing] the perceiver from feeling ineffective at preventing undesirable events,” and reinforces a sense of justice among the public (Freeman, 2006). Under this framework, assessments of victim culpability are thus contingent on certain perceptions about the unfortunate event and victim. We draw on this perspective in developing our hypotheses. Because the public is generally uninformed about most crime issues (Pickett, Mancini, Mears, & Gertz, 2015; Roberts & Hough, 2005) we first hypothesize that while members of the public will perceive a nontrivial amount of victim-offender overlap, they will underestimate the true extent of victim involvement in crime. Second, given the focus on investigating the victimoffender overlap, we hypothesize that those who judge a greater proportion of victims to be engaged in crime will more strongly perceive victims as culpable for their offenses. Below we discuss the data and measures used to test these hypotheses. DATA AND MEASURES We analyze data from two recent surveys to test our hypotheses concerning victim blame. The utilization of two separate samples is a unique contribution of our study. Indeed a noted limitation of prior research related to this area has been a primary focus on undergraduate students, particularly samples that are highly homogenous (e.g., Cramer, Nobles, Amacker, & Dovoedo, 2013; Yamawaki et al., 2012). The first sample is a 2013 web survey of 760 adult (18 and older) U.S. residents who were randomly selected from an opt-in online 9 Downloaded by [University of Warwick] at 07:24 30 December 2015 10 C. Mancini and J. T. Pickett panel.3 While this sample constitutes a probability sample of the panel, the panel is a nonprobability sample of the public, and thus the results are only directly generalizable to the individuals in the panel. Importantly, the panel is demographically diverse, and includes more than 400,000 individuals from all 50 U.S. states. Although not without their limitations, opt-in Internet surveys can be useful for testing theories through the modeling of attitudes and beliefs, which is our objective in this study (see Baker et al., 2013; Gosling, Vazire, Srivastava, & John, 2004). As Yeager, Krosnick, and Javitz (2009) explain, there is “tremendous value in opt-in survey data. One hundred years worth of terrifically useful social science theory testing has been accomplished using non-representative samples.” Recent research also suggests that online panelists provide more valid self-reports—being characterized by lower levels of social desirability bias and measurement error—than do other types of respondents (Chang & Krosnick, 2009; see also Kreuter, Presser, & Tourangeau, 2008). Obtaining valid measures by reducing measurement error is especially important for theory testing. For the current study, 2,770 randomly selected panelists received invitations to participate in the project, of which 867 began the questionnaire and 826 completed the survey. Accordingly, the overall participation rate was 30% and the completion rate 95%. The analytic sample is 760, after deleting cases with missing values. As online panelists, all of the respondents in the web survey previously agreed to complete surveys on a regular basis in exchange for both altruistic (i.e., donations to charity) and self-interested incentives (opportunity to win $100 in weekly drawings). Many studies find a great deal of correspondence between the multivariate relationships observed with nonprobability online samples and those that emerge with other more traditional samples (Ansolabehere & Schaffner, 2014; Bhutta, 2012; Gosling et al., 2004; Sanders et al., 2006). Even still, replication of study results, preferably within the same study, proffers many benefits. Scholars, for instance, have emphasized that including multiple studies within a single article for the purpose of replicating findings may “expand the generalizability of primary findings” and also “reduce false-positive rates of findings to a substantial degree” (Murayama, Pekrun, & Fiedler, 2014, p. 110; see also Carver, 2004; Diener, 1998). For these reasons, we examine whether our results replicate with a second sample, comprised of undergraduate college students. The student sample includes 805 undergraduates enrolled in criminology and criminal justice courses at two universities during the fall 2013 and spring 2014 semesters (notably, this sample size is relatively larger than prior studies centered on victim attributions, e.g., Fox & Cook, 2011; Yamawaki et al., 2012).4 The students received the same questionnaire that was used with the national sample of opt-in panelists. In all but one class, the students completed the surveys online. The survey was voluntary, and the students completed the questionnaires anonymously. The overall response rate was 84.9%.5 Victim-Offender Overlap Perceptions and Victim Blaming Dependent Variable Downloaded by [University of Warwick] at 07:24 30 December 2015 Dispositional Attributions for Victimization As noted above, very few previous studies have explored attitudes about victim culpability outside of the context of select crimes. Additionally, most previous studies have focused on “situational blame” or the blaming of a specific victim in a specific situation (e.g., Howard, 1984; more recently, Fox & Cook, 2011). We, however, are interested in general causal attributions concerning victimization. That is, we focus on whether individuals, in the abstract, explain victimization as a result of (1) personal shortcomings or (2) situational/environmental factors that are outside of the victims’ control. Additionally, recent research suggests attributions of victimization among the public are complex (Cramer et al., 2013). For these reasons, we developed an original measure of victimization attributions that would permit us to test our research questions. Building on the literature on crime attributions (see Gabbidon & Boisvert, 2012; Unnever, Cochran, Cullen, & Applegate, 2010),6 we conceptualize our version of victim attributions as endorsement of dispositional rather than situational causal attributions for victimization. Endorsement of dispositional attributions confers personal responsibility or blame on individuals for their life occurrences (Grasmick & McGill, 1994; Pickett & Baker, 2014). As a first step in developing the measure, we identified key dispositional shortcomings that members of the public may associate with victimization. In measuring dispositional crime attributions, researchers commonly include items tapping public views about offenders’ decision-making and behavioral choices (see Cullen, Clark, Cullen, & Mathers, 1985; Templeton & Hartnagel, 2012). In the same way, public views about victims’ decision-making and choices—specifically, that victims make poor decisions and are careless in their behavioral choices—likely constitute principal dispositional attributions for victimization. A large literature also shows that low self-control, a dispositional factor, is positively associated with victimization, because it makes it unlikely that potential victims will stay away from dangerous situations (Pratt, Turanovic, Fox, & Wright, 2014). For these reasons, we use several items to measure dispositional attributions for victimization, which capture views about the extent to which victims are careless, make poor decisions, and exhibit low self-control (e.g., by failing to stay away from dangerous places). We also include an item tapping global views about victim culpability. Specifically, in the survey, an introduction to the attribution questions asked respondents to think about “the things that cause someone to become a crime victim.” They then rated their level of agreement (“1 = strongly disagree,” “2 = disagree,” “3 = somewhat disagree,” “4 = somewhat agree,” “5 = agree,” or “6 = strongly agree”) with six items tapping causal attributions 11 12 C. Mancini and J. T. Pickett Downloaded by [University of Warwick] at 07:24 30 December 2015 Table 1: Questions measuring victim blaming and perceptions of the victimoffender-overlap. Victim Blaming 1. Global culpability (reversed): “Most people who become victims do so through no fault of their own.” 2. Poor decision-making: “A primary reason that people become victims is that they choose to put themselves in risky situations.” 3. Poor decision-making: “Many people become victims because they choose to hang out with the wrong crowd.” 4. Carelessness: “A primary reason that people become victims is that they fail to take enough steps to protect themselves.” 5. Carelessness (reversed): “People who pay a lot of attention to their surroundings are just as likely to become victims as those who don’t.” 6. Low self-control 1 : “A primary reason that people become victims is that they just won’t stay away from dangerous places.” Perceived Victim-Offender Overlap 1. “When you think about the people who are MURDERED in the U.S., approximately what PERCENT OF THESE VICTIMS—that is, how many out of 100 of these victims—would you say were involved in committing crime themselves before being victimized?” 2. “When you think about the people who are ROBBED at gunpoint in the U.S., approximately what PERCENT OF THESE VICTIMS—that is, how many out of 100 of these victims—would you say were involved in committing crime themselves before being victimized?” 3. “When you think about the people who HAVE THINGS STOLEN FROM THEIR HOMES by burglars in the U.S., approximately what PERCENT OF THESE VICTIMS—that is, how many out of 100 of these victims—would you say were involved in committing crime themselves before being victimized?” NOTE: 1 The sixth item is included because research shows that low self-control reduces the likelihood that potential victims will take steps to avoid dangerous situations and risky behaviors, and is thus positively associated with victimization (Pratt et al., 2014). for victimization. Table 1 provides the exact question wording for these items. In designing the items, we were careful to word the questions in a way that ensured they tapped causal attributions for victimization rather than perceptions of the risk factors for victimization. To illustrate, with the exception of the global culpability item, the survey questions all clearly specify some behavioral choice on the part of victims (e.g., “choose to,” “fail to,” “just won’t”) that can be distinguished from mere circumstance or bad luck (e.g., living in a bad neighborhood). Stated another way, each of our questions gauged whether respondents believed that victimization results from the specific actions (or inactions) of victims, rather than from circumstances or situations that are outside of their control (e.g., living in a high-crime neighborhood with little police presence). In the questionnaire, we included a mix of positively and negatively worded items to minimize measurement error from stylistic responding (Pickett & Baker, 2014). Accordingly, we recoded the negatively worded items so that higher values indicated greater endorsement of dispositional attributions for victimization. Stated differently, higher scores for this item equate Victim-Offender Overlap Perceptions and Victim Blaming to stronger expressions of victim culpability among respondents. We then averaged responses from the six questions. For both samples, the index demonstrated sufficient reliability (Cronbach’s alpha = .72 and .677 for the national and college sample, respectively). Downloaded by [University of Warwick] at 07:24 30 December 2015 Independent Variable Victim-Offender Overlap To our knowledge, there are no previous studies that have measured public perceptions about the victim-offender overlap. It was thus necessary to develop a method for gauging such perceptions. Previous studies examining public views about the characteristics of criminal suspects or offenders often ask respondents to estimate the percentage of these groups who share a characteristic or experience (Matsueda & Drakulich, 2009; Welch, Payne, Chiricos, & Gertz, 2011).8 Given the absence of prior research on public perceptions of the victim-offender overlap, we follow this approach in our study. Specifically, three questions asked respondents to estimate the percent of murder victims, robbery victims, and burglary victims who were involved in crime. Table 1 presents the exact question wording for these items. For the analysis, we averaged responses across the three items (Cronbach’s alpha = .76 and .78 for the national and college sample, respectively). Controls In line with prior research (e.g., Clements, Brannen, Kirkley, Gordon, & Church, 2006; Suarez & Gadalla, 2010), the effects of several other predictors were included in analysis. First, we controlled for racial resentment because previous studies have linked negative racial attitudes with greater victim blame in the case of sexual victimization (George & Martinez, 2002). Specifically, we anticipate that higher racial resentment will drive greater victim blaming. Racial resentment was measured by averaging responses from five items designed to tap racial animus (six-category index).9 We also controlled for the effects of crime-related factors and news/media exposure. For instance, we can expect those who perceive a greater personal risk of victimization to express less victim blame. That is, such individuals may be more likely to self-identify with crime victims (e.g., Dubber, 2002) and so will evaluate them as less blameworthy than individuals with lower perceived victimization risk. Here, perceived risk was measured as a standardized index10 (higher values indicate greater perceived risk). Separately, but in a similar direction, the perception that crime has increased may be associated with less victim blame. The reasoning here is that such individuals may view offending as random and typical, and so may judge victims as less blameworthy. Perceived crime 13 Downloaded by [University of Warwick] at 07:24 30 December 2015 14 C. Mancini and J. T. Pickett trend was coded along a continuum, “1 = decreased greatly,” “2 = decreased some,” “3 = stayed about the same,” “4 = increased some,” and “5 = increased greatly.” Additionally, news and crime television consumption may be associated with attitudes toward victims; some accounts suggest the type of coverage affects empathy and levels of victim blame (e.g., Anastasio & Costa, 2004). We included three indicators of news/media exposure that captured how often respondents consumed three forms of media in a typical week (0 = 0 days, 7 = 7 days): national news consumption, local news consumption, and crime television consumption. Existing investigations also suggest that respondent demographics, particularly sex, influence victim blaming in the case of rape victims; women are significantly less likely to attribute blame to sex crime and intimate partner violence victims compared to men (Suarez & Gadalla, 2010; Yamawaki et al., 2012). Drawing on this finding, we anticipate similar sex effects for victim blaming across diverse crimes. Accordingly, we accounted for respondent sex (“0 = male,” “1 = female”) in the models. We also control for the following demographic characteristics: conservatism (“1 = very liberal,” “2 = liberal,” “3 = middle of the road,” “4 = conservative,” and “5 = very conservative”), religiosity (“1 = very unimportant,” “2 = unimportant,” “3 = somewhat unimportant,” “4 = somewhat important,” “5 = important,” and “6 = very important”), family crime victimization (in the last five years; “0 = no,” “1 = yes”), personal criminal victimization (in the last five years; “0 = no,” “1 = yes”), education (for the national sample; “1 = less than high school degree,” “2 = high school degree,” “3 = some college,” “4 = associate or bachelor degree,” and “5 = graduate degree”), GPA (for the college sample; coded as: “1 = 2.0 or below,” “2 = 2.1 to 2.5,” “3 = 2.6 to 3.0,” “4 = 3.1 to 3.5,” and “5 = 3.6 to 4.0”), race (“0 = nonwhite,” “1 = white”), and age (measured continuously). Table 2 presents additional information regarding the descriptive statistics and coding of the measures. RESULTS Because no previous research has examined perceptions of the victim-offender overlap, we open our analysis by investigating the extent of such perceptions in our samples. Four patterns deserve mention. One striking observation (Figure 1 and Table 2) to emphasize is that both samples substantially underestimate the extent of victims’ involvement in crime. Recall that studies have demonstrated about half of murder victims and over 40% of property crime victims have prior criminal histories (Broidy et al., 2006; Wolfgang & Singer, 1978; for a review, Jennings et al., 2010). The respondents in both of our samples, however, estimate that on average a much smaller victim-offender overlap exists, with the national sample estimating 22% of crimes involve a victim with prior criminal involvement and the college sample, 29%. Specifically, as 15 3.14 21.91 3.53 2.43 3.33 2.38 3.29 2.00 2.80 3.59 .33 .20 3.83 — .53 .78 46.57 760 Victim blaming Perceived V-O overlap Racial resentment Perceived risk Perceived crime trend National news consumption Local news consumption Crime television consumption Conservatism Religiosity Family victim Victim Education GPA Female White Age N .82 15.89 1.31 1.12 1.11 2.53 2.68 2.20 .99 1.85 .47 .40 .96 — .50 .42 16.23 S.D. National Sample ABBREVIATIONS: S.D. = standard deviation, V-O = victim-offender. Mean Variables Table 2: Descriptive statistics. 1–6 0–100 1–6 1–7 1–5 0–7 0–7 0–7 1–5 1–6 0–1 0–1 1–5 — 0–1 0–1 18–91 Range 3.36 29.00 3.38 2.31 2.50 1.50 1.85 3.03 2.86 3.61 .45 .27 — 3.65 .54 .58 20.14 733 Mean .76 17.64 1.04 1.00 1.04 1.79 1.82 2.28 .82 1.63 .50 .44 — .97 .50 .49 2.91 S.D. College Sample Downloaded by [University of Warwick] at 07:24 30 December 2015 1–6 0–100 1–6 1–7 1–5 0–7 0–7 0–7 1–5 1–6 0–1 0–1 — 1–5 0–1 0–1 18–59 Range 16 C. Mancini and J. T. Pickett Perceived Percent of Victims Involved in Crime Downloaded by [University of Warwick] at 07:24 30 December 2015 45% 40% 35% 38.4% National Sample 33.9% College Sample 30% 26.4% 25% 22.2% 20% 18.5% 15% 13.3% 10% 5% 0% Homicide Robbery Burglary Figure 1: Perceived victim-offender overlap, by crime type. shown in Figure 1, the national sample perceives a victim-offender overlap in 34% of murders, whereas the college sample perceives the victim-offender overlap in 38% of murders. Even fewer respondents believe there is a substantial victim-offender overlap across other offenses; in particular, the national sample estimated that 19% of robbery offenses involve a victim-offender, and the college sample estimated an overlap in 26% of robberies. Separately, an overlap was perceived in 13% of burglary victimizations among the national sample, whereas college students judged a higher proportion (22%). Here then, support exists for the study’s first hypothesis—and more generally, prior work demonstrating that the public is largely misinformed about crime patterns (e.g., Pickett et al., 2015; Roberts & Hough, 2005). Having said that, and as a second observation, both samples of respondents perceive the highest victim-offender overlap among homicide victimization. This perception is notable because it is consistent with research showing a larger victim-offender overlap for violent and interpersonal victimizations such as murder (Jennings et al., 2010). Perhaps the public perceives homicide to be committed primarily within the context of gang or drug involvement, and so attributes the largest overlap for these types of victimizations. Third, there is consistency in this pattern across samples. The overall trend in victim-offender perceptions—with respondents attributing the greatest overlap for homicide crimes, the second highest for robbery, and the least for burglary offenses—is similar across both the national sample and the college sample. Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming Fourth and finally, it is apparent that college students, although generally understating the extent of the victim-offender overlap, come closer to published reports of it compared to the national sample. For instance, across each crime category, the college sample perceives a significantly higher (p < .0002, two-tailed) share of victims-as-offenders. This divergence may stem from education—that is, by dint of enrolling in criminology courses, college students perceive a higher and thus more accurate level of the victim-offender overlap. How though, if at all, do perceptions of the victim-offender overlap influence general causal attributions for criminal victimization? To examine this relationship we estimated multivariate models predicting victimization attributions among the public. Given the nature of the outcome variable, we used ordinary least squares (OLS) regression to test our hypothesis. The results of the Breusch-Pagan/Cook-Weisberg test indicate the presence of heteroskedasticity. To correct for this, all OLS models were estimated using robust standard errors. Multicollinearity did not appear to bias findings, with tolerance levels consistently above .53 and variance inflation factors (VIF) no higher than 1.89. Regression results are shown in Table 3. The first two models tested the effect of the victim-offender perception on victimization attributions across the national sample. Specifically, the first model included the controls and the second model included the controls in addition to the victim-offender overlap variable. This approach permitted us to evaluate the extent to which the victim-offender overlap belief influenced endorsement of dispositional attributions for victimization. The models were repeated for the college sample (models 3 and 4). Examination of models 1 and 3 indicates that three controls are significant in the baseline models—racial resentment, conservatism, and sex. Specifically, in line with prior research on rape myths (George & Martinez, 2002), racial resentment was associated with increased blame attributed to crime victims. We also find that greater political conservatism is associated with higher victim blame. By contrast, sex—specifically, being female—reduced endorsement of dispositional attributions for victimization. This pattern is evident across both samples—suggesting that similar social and demographic factors are associated with attitudes aligned with victim blaming among both the national sample and college students. More generally, it is in line with previous victim blame research (Suarez & Gadalla, 2010; Yamawaki et al., 2012). Beyond these social and demographic factors, what effect—if any— do perceptions of the victim-offender overlap have on endorsement of dispositional attributions for victimization? The simple answer, lending support to the study’s second hypothesis, is that perceiving a large overlap predicts greater belief in victim responsibility. To illustrate, as shown in Table 3, there is a sizable and significant positive association between perceptions of the 17 18 — .119∗∗∗ −.056 .030 .008 .009 .004 .078∗ .006 .018 −.031 −.012 — −.228∗∗∗ .040 −.000 b — (.031) (.031) (.029) (.015) (.013) (.015) (.036) (.016) (.064) (.082) (.033) — (.056) (.083) (.002) .112 760 S.E. — .190 −.078 .041 .026 .032 .011 .094 .014 .010 −.015 −.014 — −.140 .018 −.002 St. Coef. S.E. (.002) (.030) (.030) (.029) (.015) (.013) (.014) (.035) (.016) (.063) (.081) (.033) — (.055) (.081) (.002) .153 760 b .011∗∗∗ .107∗∗∗ −.068∗ .026 .004 .012 −.002 .073∗ .004 .011 −.015 −.004 — −.203∗∗∗ .036 .002 Model 2 .209 .171 −.093 .036 .013 .038 −.005 .088 .008 .006 −.007 −.005 — −.124 .018 −.042 St. Coef. — .102∗∗ −.044 −.007 .025 −.016 −.010 .135∗∗∗ .011 .043 −.057 — −.025 −.323∗∗∗ −.028 −.005 b — (.030) (.028) (.027) (.021) (.021) (.014) (.037) (.018) (.058) (.066) — (.027) (.058) (.060) (.011) .142 733 S.E. Model 3 — .140 −.058 −.010 .058 −.039 −.031 .147 .024 .028 −.034 — .032 −.212 −.018 −.018 St. Coef. .007∗∗∗ .103∗∗∗ −.049 −.007 .024 −.015 −.012 .124∗∗ .012 .020 −.056 — −.019 −.302∗∗∗ −.012 −.003 b College Sample (.002) (.029) (.028) (.027) (.020) (.020) (.014) (.036) (.018) (.057) (.065) — (.026) (.057) (.059) (.010) .171 733 S.E. Model 4 .171 .142 −.065 −.010 .056 −.036 −.035 .135 .026 .013 −.033 — −.024 −.199 −.008 −.010 St. Coef. ABBREVIATIONS: cons. = consumption; b = unstandardized regression coefficient; S.E. = robust standard error; St. Coef. = standardized regression coefficient; V-O = victim-offender. ∗ p < .05; ∗∗ p < .01; ∗∗∗ p < .001 (two-tailed). Perceived V-O overlap Racial resentment Perceived risk Perceived crime trend National news cons. Local news cons. Crime TV cons. Conservatism Religiosity Family victim Victim Education GPA Female White Age R-squared N Variables Model 1 National Sample Table 3: The effect of victim-offender overlap perceptions on victim blaming. Downloaded by [University of Warwick] at 07:24 30 December 2015 Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming victim-offender overlap and endorsement of dispositional attributions of victimization.11 This relationship emerges in both the samples. A slope difference test revealed that the effect of victim-offender overlap perceptions on victim blaming was not significantly different across the national and college samples (p = .153, two-tailed). The addition of victim-offender overlap perceptions increases the explanatory power of the models; specifically, the percentage change was calculated to be 37% and 20% for the national and college samples, respectively. Strikingly, as inspection of models 2 and 4 in Table 3 reveals, the victimoffender overlap perception exerted the strongest effect on victim attributions in the national sample (beta = .209), and the second largest (beta = .171) in the college sample. The strongest predictor of dispositional attributions among college students is sex (beta = −.199)—female college students are substantially less likely to attribute culpability to crime victims than are male students. The implication is that for understanding attitudes regarding victim culpability, victim-offender overlap perceptions appear to be a more salient consideration than political ideology, racial attitudes, or prior victimization status. Notice too that the effects of the control variables on victim attributions remain relatively unchanged in the full models after including perceptions of the victim-offender overlap. This pattern suggests that perceptions of the victim-offender overlap do not mediate the relationships between the controls, such as conservatism or sex, and perceptions of victim responsibility. These findings can be framed under the just world hypothesis. To illustrate, one interpretation is that perceiving a larger proportion of victimperpetrators undermines public sympathy and compassion for victims, because such individuals stand outside the “purely innocent” victim construct so often (albeit selectively) conveyed in contemporary society (Dubber, 2002); as a result, the public experiences difficulty in identifying with these deviant “others,” in turn, holding them accountable for their victimization. As others have explained, this process is cathartic (for a review, Hafer & Bègue, 2005). In the context of blaming attitudes toward crime victims, it is theorized to alleviate anxieties about crime and personal victimization risk by minimizing cognitive dissonance and reinforcing a sense of “fairness” (i.e., “we reap what we sow”) among the public. If individuals can distance themselves from crime victims, personal concern and fear about experiencing a similar fate can be expected to be reduced. CONCLUSION AND IMPLICATIONS Although victim blaming is an enduring social phenomenon, the literature on victimization attributions for crimes other than offenses against women is underdeveloped (e.g., Fox & Cook, 2011). Using data from both 19 Downloaded by [University of Warwick] at 07:24 30 December 2015 20 C. Mancini and J. T. Pickett a national sample and a sample of college students (see Murayama et al., 2014; earlier, Carver, 2004; Diener, 1998, identifying the benefits of this approach), the current study constitutes the first test of whether perceptions of victims-as-perpetrators affect assessments of victim responsibility. In particular, we addressed two previously unexamined research questions. First, to what extent does the public perceive a victim-offender overlap for street offenses (i.e., homicide, robbery, and burglary crimes)? Second, are victimsas-perpetrators perceptions salient for predicting victim-blaming attitudes among the public? Our results reveal several notable findings concerning these questions, which we briefly summarize below before discussing their implications. First, the respondents in both samples perceive a much smaller victimoffender overlap than what exists in reality, as documented in criminological research. Respondents attribute the largest overlap to homicide offenses. This perception is consistent with objective patterns of victim involvement in offending by victimization type (Jennings et al., 2010). One explanation here is that perhaps respondents judge homicide victimization, compared to robbery and burglary crimes, to be most likely to occur to persons regularly involved in criminal groups or routines—such as gang members or drug users. It follows that given the risky lifestyles that the public might associate with homicide victimization, the highest overlap is attributed to these offenses. Supporting prior work demonstrating generalizability of college samples (Piquero & Bouffard, 2007; Wiecko, 2010), we find consistency in the overlap perception among the national and college samples. Even so, students from this college sample—perhaps because of their greater knowledge about crime issues—were more accurate in their assessment of the victim-offender overlap than the national sample of online panelists. Separately, perceiving a larger victim-offender overlap amplified dispositional attributions toward crime victims. Indeed, the perceived victim-offender overlap emerged as either the strongest or the second strongest predictor of victim blaming across both samples. Under the just world hypothesis, this finding can be explained as respondents perceiving victim-offenders as culpable for their victimization, and so significantly more blameworthy than individuals who have not engaged in deviant behavior. Specifically, among members of the public, the perception of a larger victimoffender overlap may induce endorsement of dispositional attributions for victimization by signaling that many crime victims are not conventional, lawabiding citizens, and thus likely do not possess the same character strengths and positive personality traits as such citizens. This cognitive process may have powerful effects on the public conscience. It may, for example, alleviate personal concern about victimization risk by allowing members of the public to distance themselves from victim-offenders (“I would never commit crime, so I don’t have to worry about becoming a victim”) and, simultaneously, reinforce Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming a sense of justice—“bad things (i.e., victimization) happen to bad people (i.e., those who put themselves in danger).” Additional factors—racial resentment, conservatism, perceived victimization risk, and sex—were also associated with endorsement of dispositional attributions for victimization. In particular, racial animus was associated with negative attitudes toward crime victims across both samples. The explanation here might be that racially resentful persons, although more punitive toward criminals, may also be more apathetic toward the crime issue generally because they tend to associate it with disliked racial others (see Unnever & Cullen, 2009). It is also plausible that victim blame serves an instrumental purpose for racially resentful persons—namely, providing a ready-made justification for opposing social policies aimed at addressing “the root causes” of victimization, but which may also benefit blacks (see Unnever, Cullen, & Jones, 2008). An alternative possibility is that causal order runs in the opposite direction. Racial dispositional attributions for victimization, if victims are believed to be disproportionately black, may actually increase racial resentment (see Pickett & Baker, 2014). Conservatism also served to heighten blaming attitudes toward crime victims in the models. This finding accords with prior research revealing that conservatives tend to implicate the moral or personal shortcomings of individuals (e.g., selfishness, laziness) rather than situational factors (e.g., poverty, economic forces) as contributors to crime (Unnever & Cullen, 2009). By comparison, higher perceived victimization risk was correlated with reduced victim blame among the national sample. An “identification effect” (Dubber, 2002) potentially sheds light on this relationship. For instance, perhaps individuals who recognize or perceive that they are at greater risk for experiencing crime extend empathy to crime victims and so attribute less blame toward them than individuals who believe they are relatively safe from crime. Results also indicate, in line with previous studies, that females attribute less blame to crime victims. Women, in contrast to men, have long been characterized as being predisposed to an “ethic of care” (Steffensmeier & Allan, 1996, p. 476); for this reason, women may empathize with crime victims in turn, evaluating them as less blameworthy than men. This pattern of findings holds for both the national sample and the sample of university students. This is notable given competing arguments regarding the generalizability of college samples. One claim is that public opinion of college students significantly diverges from perceptions held among the general public (Clements et al., 2006; Payne & Chappell, 2008). However, a contrasting contention is that, on balance, the public and students express similar perceptions of crime and offending (e.g., Piquero & Bouffard, 2007; Wiecko, 2010). Collectively, our findings provide support for both perspectives. On the one hand, college students were significantly more knowledgeable about crime 21 Downloaded by [University of Warwick] at 07:24 30 December 2015 22 C. Mancini and J. T. Pickett victims than the national sample, believing, on average, that a larger victimoffender overlap exists (p < .0002, two-tailed). At the same time, attributing a larger victim-offender overlap was associated with greater victim blaming across both groups of respondents, and the magnitude of this effect was statistically insignificant across the samples (p = .153, two-tailed). We turn now to a discussion of the study’s theoretical, research, and policy implications. Recall that a central theoretical motivation for studying the victimoffender overlap concerns the dichotomy in how certain classes of victims are perceived and treated by society relative to other populations. Prior accounts have argued that some victims, such as children (Dubber, 2002) and members of the upper and middle classes (Simon, 2007), receive special protections and assistance, whereas other groups garner substantially less public and policy attention. The just world hypothesis suggests that a possible explanation for this dichotomy may be that these latter groups—most often the typical victims of street offenses (Cooper & Smith, 2011; Harrell, 2007)—are perceived as deviant, and thus culpable for the offenses committed against them—and so public concern for their welfare is substantially attenuated. Our findings are consistent with this interpretation, demonstrating that in the case of street crimes, a larger perceived victim-offender overlap is associated with increased dispositional attributions for victimization. It seems, then, that the respondents in our samples primarily considered “true victims” to be victimized persons without previous involvement in crime. From a policy standpoint this is an important finding, given that crime victims, or more accurately “true victims,” are today’s “idealized political subject . . . whose circumstances and experiences have come to stand for the general good” (Simon, 2007, p. 110). One policy implication is that efforts to broaden popular concern for victims to all persons who suffer criminal victimization, or even to offenders, may benefit from campaigns that provide information to the public about the environmental correlates of the victim-offender overlap. For example, informing citizens of the strong link between neighborhood disadvantage and the victim-offender connection (Berg & Loeber, 2011; Berg et al., 2012) may reduce victim blame by fostering awareness that both offending and victimization share situational causes, even when they occur among the same individuals. Another implication is that it may be possible to increase public support for victim services by coupling those services with criminal background checks or at least by limiting those services to victims without recent criminal records. However, as we discuss below, such approaches may also exclude those victims who are the most in need of services. Given our findings, subsequent studies are needed that complement this line of inquiry by evaluating whether, in the case of specific victimization incidents, victim-offender overlap perceptions are shaped by the characteristics of the victim (e.g., victim age, sex, socioeconomic status, and employment status) or interact with such characteristics to influence public attitudes toward Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming victims. The current study is limited in that it does not tap into perceptions of victim characteristics beyond their involvement in crime. In a similar vein, it may be worthwhile to understand nuances in victim-blaming attitudes. For example, future research might explore the extent to which dispositional attributions, which we view as a form of global blame, are correlated with victim blame in specific situations or scenarios, such as those typically used in prior research (see Fox & Cook, 2011). Not least, future studies should investigate the potential for “victim blame spillover” among the public.12 For example, if a child is accidentally killed by a drug rival of her father, might the public perceive the father as blameworthy for the child’s victimization? This dimension has important implications for determining the extent to which the public holds individuals, beyond the perpetrator, culpable for crime. In short, these avenues highlight how the current study can be extended to better understand victim-blaming attitudes. To broaden our findings, future studies should also examine the victimoffender overlap across a range of offenses. The current study contributes to prior research by examining three forms of serious “street crimes”—homicide, robbery, and burglary. Still, our findings are generalizable only to this select subgroup of offenses. It may be relevant to extend this focus even further by investigating public perceptions toward other victim populations, such as victims of hate crimes (generally, Cramer et al., 2013), white-collar crimes, bullying, or political violence (e.g., war crimes, terrorism). This direction would contribute to efforts to capture public attitudes toward diverse groups of victims and offenders. Ideally, future research will explore victim blame and its effect on public perceptions and preferences along other dimensions. Our investigation is limited, as it cannot speak to whether perceptions of the victim-offender overlap or victim-blaming attitudes affect support for victim assistance and related policies. For instance, studies could be designed to investigate whether victimblaming attitudes influence support or willingness to pay for victim services and assistance. It is conceivable, for example, that individuals who attribute blame to crime victims may be less supportive of providing them with publicly funded services, such as monetary compensation, medical aid, and psychological counseling. Not least, the extent to which victim blame is related to guardianship and protective behaviors should be examined. For instance, perceiving that victims are largely responsible—because of their bad decisions or personal deviance—for the offenses committed against them may reduce individuals’ felt need to adopt protective measures to reduce their own victimization risk. In turn, this focus may contribute to extant victimology research by revealing nuances in perceptions toward crime and their influence on guardianship among the public. Another important shortcoming of our study is that neither of the samples we analyze is a probability sample of Americans, and thus it is not 23 Downloaded by [University of Warwick] at 07:24 30 December 2015 24 C. Mancini and J. T. Pickett clear whether our findings would generalize to the larger public. As noted above, there is growing evidence that opt-in online samples commonly produce findings that are very similar to those obtained with representative samples (Ansolabehere & Schaffner, 2014; Bhutta, 2012; Simmons & Bobo, in press; Weinberg, Freese, & McElhattan, 2014). Even without this evidence, however, there is a strong reason to believe that our findings are externally valid. Namely, all of our main findings replicate across both samples. These samples differ dramatically in terms of race, age, education, and residential location. Yet in both samples, the same findings emerge: (1) there is an identical patterning of victim-offender overlap perceptions across crime types (2) we find that victim-offender overlap perceptions are positively and significantly associated with dispositional attributions for victimization, and (3) we find that racially resentful persons, conservatives, and males are all more likely to endorse dispositional attributions for victimization. Indeed, the only variable that has a significant relationship to dispositional attributions in one sample but not in the other is perceived victimization risk, but this difference is hardly notable. In the online sample, the coefficient for this variable is negative and significant (b = -.068, p = .023), whereas in the college sample it is negative but only marginally significant (b = -.049, p = .074). There is widespread agreement across disciplines that replication is an important indicator of the generalizability of findings (Baker et al., 2013; Lucas, 2003; Murayama et al., 2014). As Lucas (2003, p. 239) explains, “when a theoretical principle is supported in diverse replications, we gain confidence in the theory, and each successive test increases external validity.” That there is a high degree of replicability in our study suggests that the factors on which these two samples differ from each other (e.g., age, race, education) are unlikely to be sources of selection bias. This is corroborated by the fact that age, race, and education are not associated with victim blaming in either study. These factors also happen to be those on which both samples differ most markedly from the general public. Even still, we believe it important for future studies to attempt to replicate our study using representative samples of Americans. A final recommendation for research is to explore attitudes among criminal justice practitioners. Law enforcement officers, for instance, routinely interact with victims of crime and their actions impact the experience of crime victims. Similarly, court actors—such as prosecutors—also engage in decision-making that directly affects victims. For example, prior studies indicate that victim characteristics and behavior such as sex and perceived alcohol use influence criminal justice responses, such as the decision to arrest (Waaland & Keeley, 1985). For this reason, understanding the extent to which criminal justice professionals’ perceptions of the victim-offender overlap, and perhaps other related judgments, amplify victim blaming is important for extending criminal justice scholarship. Downloaded by [University of Warwick] at 07:24 30 December 2015 Victim-Offender Overlap Perceptions and Victim Blaming Others have argued that victim blaming attitudes are among the most influential social factors that “contribute to creating an acceptable climate for violence” (Gracia & Tomas, 2014, p. 27). Indeed, perceived stigma from the public serves to reduce reporting among crime victims (Herek, Cogan, & Gillis, 2002); in turn, underreporting simultaneously prevents the apprehension and punishment of offenders and precludes victims from seeking state and federal victim compensation funds and services. Under this context, the finding that perceiving a larger share of victim-perpetrators is associated with greater victim blaming among both samples of respondents in our study indicates cause for concern and has implications for policies designed to assist victims with their recovery. To minimize the detrimental effects of victim stigma, services and aid should be uniquely tailored and designed to better accommodate victims who may have a criminal past. Many states prohibit aid to crime victims who have engaged in prior offending or are judged as having contributed to their victimization (Dubber, 2002). As a result, crime victims who have committed offenses in the past are ineligible for services and aid distributed by the state. Additionally, as discussed earlier, crime victims who may have engaged in prior risky or criminal behavior may not report victimization to law enforcement because of fear of stigma or judgment from law enforcement or the public. Consequently, as others have observed (Schreck et al., 2008), a sizable proportion of crime victims may not receive the services and care that can assist with their recovery, and potentially reduce their risk of reoffending. Perhaps states should reconsider or modify rigid legislation that precludes distributing aid to those with previous criminal convictions or arrests. At the same time, our results suggest that popular concern for crime victims, and thus support for victim services, may be greatest when victims are not involved in crime. Perhaps the best policy approach, then, for both targeting the at-risk population and ensuring popular support for victim services might be to open such services to those with prior records, but to make participation conditional on abstinence from future offending (e.g., program participants obtain no new arrests). Continued examination of victim-blaming attitudes may also help to identify other barriers to, as well as potential strategies for, expanding aid to reach a broader population of crime victims. ACKNOWLEDGMENTS The authors are grateful to Shawn Bushway for his assistance with collecting the data. The authors also acknowledge Thomas Baker for his support with data collection and insightful comments on an earlier draft. The authors thank the Editor and three anonymous reviewers for their helpful remarks on an earlier version of the paper. 25 26 C. Mancini and J. T. Pickett FUNDING This work was funded by the University at Albany Faculty Research Awards Program (FRAP)-Category A. Downloaded by [University of Warwick] at 07:24 30 December 2015 NOTES 1. To illustrate, a provision of Florida’s victim compensation statute states, “the victim’s conduct must not have contributed to the situation that brought about his or her own injury or death” (Florida Office of the Attorney General, 2014). Additionally, Florida law stipulates that claimants must submit to a criminal background check prior to services being rendered. Specifically, for compensation to be dispensed, “the victim or claimant must not have been confined or in custody in a county or municipal facility; a state or federal correctional facility; or a juvenile detention commitment, or assessment facility; adjudicated as a habitual felony offender, habitual violent offender, or violent career criminal; or adjudicated of a forcible felony offense.” In a comparable direction, Arkansas law prohibits awarding compensation to victims who have “been convicted of a felony involving criminally injurious conduct” or engaged in “illegal activity” (e.g., victim was a minor and drinking alcohol, victim was in an “illegal place of business” such as a house of prostitution) during the commission of the offense (Arkansas Attorney General’s Office, 2015). Several other states have set forth similar restrictions for victim aid (Dubber, 2002; Murphy, 2014). 2. Howard’s (1984) measure of victim blame referred specifically to the victim depicted in the scenario. That is, respondents were asked to rate the extent to which the individual victim was to blame for the attack rather than an indicator of general victim blame (i.e., all victims of these types of crimes). 3. The sampling frame for the national survey was Survey Monkey’s Audience opt-in panel. Participation was limited to panelists who were over the age of 18 and resided in the United States. Opt-in Internet panels are preassembled (typically through diverse recruitment strategies) groups of Internet users who have previously agreed to take part in surveys. The use of opt-in panels is stronger methodologically than the more common approach of relying on crowdsourcing (Berinsky, Huber, & Lenz, 2012; Weinberg et al., 2014). 4. Substantively identical findings emerge when the analysis is estimated separately for the students at each university, with the single exception that the positive correlation between conservatism and dispositional attributions for victimization (see below) only reaches conventional thresholds for significance at one university. 5. Both surveys proceeded under Institutional Review Board approval. 6. For instance, dispositional attributions of offending (i.e., personal shortcomings) have been measured using items such as the following: “Most offenders commit crimes because they are too lazy to find a lawful way out of a bad situation”; “Most offenders commit crimes because they have bad characters” (Unnever et al., 2010, p. 441). 7. Although the alpha value for the college sample (.67) is .03 less than the conventional .70 threshold for “acceptable” reliability, we feel the measure is appropriate for two reasons. First, we constructed the measure using both positively and negatively (but recoded) valenced questions. This approach is important for reducing bias from stylistic responding, but tends to reduce reliability coefficients (see Baumgartner & Steenkamp, 2001; Pickett & Baker, 2014). Second, our study is exploratory, in that we are focusing on a broader form of victim blame than what has been examined in prior studies (see Nunnally & Bernstein, 1994, who advise that when conducting such exploratory research alpha values of at least .60 are sufficient). Victim-Offender Overlap Perceptions and Victim Blaming Downloaded by [University of Warwick] at 07:24 30 December 2015 8. For example, in studying public perceptions of police fairness, Matsueda and Drakulich (2009) asked respondents, “What percent of the black people who are suspected of committing a crime in America do you think are treated fairly by the police?” In a similar direction, Welch and her colleagues (2011) posed this question to measure ethnic typification of crime: ‘‘What percent of people who commit violent crimes in this country would you say are Hispanic?” 9. The specific bidirectional measures were: “Now thinking about African Americans in the United States, how much do you disagree or agree with each of the following statements?” “There is a lot of discrimination against blacks in the U.S. today, limiting their chances to get ahead”; “It’s really a matter of some people not trying hard enough; if blacks would only try harder they could be just as well off as whites”; “Irish, Italians, Jewish, and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors”; “Over the past few years, blacks have gotten less than they deserve”; and “Generations of slavery and discrimination have created conditions that make it difficult for blacks to work their way out of the lower class.” For both samples, the index demonstrated high reliability (Cronbach’s alpha values for the national sample = .89 and for the college sample = .80). 10. The perceived victimization risk index was comprised of five items that gauged how likely it was that respondents would experience a variety of crimes (burglary, assault, robbery, sexual assault, and homicide) over a five-year period. The index demonstrated acceptable reliability (α = .87 and .80 for the national and college sample, respectively). 11. To evaluate these effects further, ancillary analysis was conducted that examined the individual victim-blaming items on perceptions of the victim-offender overlap. Across both samples, most of the items used to create the dispositional index were statistically significant and in the expected direction. For the national sample, one variable, “carelessness (pay attention)” did not reach statistical significance; however, the remaining five items were significant and in the theorized direction. The two items that did not approach statistical significance in the college model were “global culpability” and, similar to the national model, “carelessness (pay attention).” We created a new variable that excluded the variables that did not reach statistical significance independently and reestimated all models. That is, for the national sample, the new victim-blaming variable excluded “carelessness (pay attention)”; for the college sample, the new victim-blaming variable omitted “global culpability” and “carelessness (pay attention).” For both samples, the results were substantively identical to the full version of victim blaming. For this reason, we retain the full version of dispositional attributions for victimization. We thank an anonymous reviewer for encouraging us to explore this line of analysis. 12. We thank an anonymous reviewer for suggesting this avenue of inquiry. REFERENCES Anastasio, P. A., & Costa, D. M. (2004). Twice hurt: How newspaper coverage may reduce empathy and engender blame for female victims of crime. Sex Roles, 51, 535–542. doi:10.1007/s11199-004-5463-7 Ansolabehere, S., & Schaffner, B. F. (2014). Does survey mode still matter? Findings from a 2010 multi-mode comparison. Political Analysis, 22, 285–303. doi:10.1093/ pan/mpt025 Arkansas Attorney General’s Office. (2015). Arkansas crime victims reparations board: Rules and regulations. Little Rock, AR: Author. 27 28 C. Mancini and J. T. Pickett Baker, R. J., Brick, M., Bates, N. A., Battaglia, M., Couper, M. P., Dever, J. A., . . . Tourangeau, R. (2013). Report on the APPOR task force on non-probability sampling. Retrieved from https://www.aapor.org/AAPORKentico/AAPOR_Main/media/ MainSiteFiles/NPS_TF_Report_Final_7_revised_FNL_6_22_13.pdf Downloaded by [University of Warwick] at 07:24 30 December 2015 Barnes, J. C., & Beaver, K. M. (2012). Extending research on the victim-offender overlap: Evidence from a genetically informative analysis. Journal of Interpersonal Violence, 27, 3299–3321. doi:10.1177/0886260512441259 Baumgartner, H., & Steenkamp, J. E. M. (2001). Response styles in marketing research: A cross-national investigation. Journal of Marketing Research, 38, 143–156. doi:10.1509/jmkr.38.2.143.18840 Berg, M. T., & Loeber, R. (2011). Examining the neighborhood context of the violent offending-victimization relationship: A prospective investigation. Journal of Quantitative Criminology, 27, 427–451. doi:10.1007/s10940-011-9129-7 Berg, M. T., Stewart, E. A., Schreck, C. J., & Simons, R. L. (2012). The victim-offender overlap in context: Examining the role of neighborhood street culture. Criminology, 50, 359–390. doi:10.1111/crim.2011.50.issue-2 Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon.com’s Mechanical Turk. Political Analysis, 20, 351–368. doi:10.1093/pan/mpr057 Bhutta, C. B. (2012). Not by the book: Facebook as a sampling frame. Sociological Methods & Research, 41, 57–88. doi:10.1177/0049124112440795 Britto, S., Hughes, T., Saltzman, K., & Stroh, C. (2007). Does “special” mean young, white, and female? Deconstructing the meaning of “special” in Law & Order: Special Victims Unit. Journal of Criminal Justice and Popular Culture, 14, 39–57. Broidy, L. M., Daday, J. K., Crandall, C. S., Sklar, D. P., & Jost, P. F. (2006). Exploring demographic, structural, and behavioral overlap among homicide offenders and victims. Homicide Studies, 10, 155–180. doi:10.1177/1088767906288577 Carver, C. S. (2004). Editorial. Journal of Personality and Social Psychology, 86, 95. doi:10.1037/0022-3514.86.1.95 Chang, L., & Krosnick, J. A. (2009). National surveys via RDD telephone interviewing versus the Internet: Comparing sample representativeness and response quality. Public Opinion Quarterly, 73, 641–678. doi:10.1093/poq/nfp075 Clements, C. B., Brannen, D. N., Kirkley, S. M., Gordon, T. M., & Church, W. T. (2006). The measurement of concern about victims: Empathy, victim advocacy, and the Victim Concern Scale (VCS). Legal and Criminological Psychology, 11, 283–295. doi:10.1348/135532505X79573 Cooper, A., & Smith, E. L. (2011). Homicide trends in the United States, 1980–2008: Annual rates for 2009 and 2010. Washington, DC: Bureau of Justice Statistics. Cramer, R. J., Nobles, M. R., Amacker, A. M., & Dovoedo, L. (2013). Defining and evaluating perceptions of victim blame in antigay hate crimes. Journal of Interpersonal Violence, 28, 2894–2914. doi:10.1177/0886260513488687 Cullen, F. T., Clark, G. A., Cullen, J. B., & Mathers, R. A. (1985). Attribution, salience, and attitudes toward criminal sanctioning. Criminal Justice and Behavior, 12, 305–331. doi:10.1177/0093854885012003003 Cullen, F. T., Clark, G. A., Mathers, R. A., & Cullen, J. B. (1983). Public support for punishing white-collar crime: Blaming the victim revisited? Journal of Criminal Justice, 11, 481–493. doi:10.1016/0047-2352(83)90002-8 Victim-Offender Overlap Perceptions and Victim Blaming Daday, J. K., Broidy, L. M., Crandall, C. S., & Sklar, D. P. (2005). Individual, neighborhood, and situational factors associated with violent victimization and offending. Criminal Justice Studies, 18, 215–235. doi:10.1080/14786010500287347 Darley, J. M., & Pittman, T. S. (2003). The psychology of compensatory and retributive justice. Personality and Social Psychology Review, 7, 324–336. doi:10.1207/ pspr.2003.7.issue-4 Downloaded by [University of Warwick] at 07:24 30 December 2015 Diener, E. (1998). Editorial. Journal of Personality and Social Psychology, 74, 5–6. doi:10.1037/h0092824 Dobrin, A. (2001). The risk of offending on homicide victimization: A case control study. Journal of Research in Crime and Delinquency, 38, 154–173. doi:10.1177/ 0022427801038002003 Doerner, W. G., & Lab, S. P. (2012). Victimology (6th ed.). Burlington, MA: Anderson Publishing. Dubber, M. D. (2002). Victims in the war on crime: The use and abuse of victims’ rights. New York, NY, USA: New York University Press. Finkel, N. J. (2001). Not fair! The typology of commonsense unfairness. Washington, DC: American Psychological Association. Florida Office of the Attorney General. (2014). Bureau of Victim Compensation claim form. Tallahassee, FL: Author. Fox, K. A., & Cook, C. L. (2011). Is knowledge power? The effects of a victimology course on victim blaming. Journal of Interpersonal Violence, 26, 3407–3427. doi:10.1177/ 0886260511403752 Freeman, N. J. (2006). Socioeconomic status and belief in a just world: Sentencing of criminal defendants. Journal of Applied Social Psychology, 36, 2379–2394. doi:10.1111/jasp.2006.36.issue-10 Gabbidon, S. L., & Boisvert, D. (2012). Public opinion on crime causation: An exploratory study of Philadelphia area residents. Journal of Criminal Justice, 40, 50–59. doi:10.1016/j.jcrimjus.2011.11.008 Garland, D. (2001). The culture of control: Crime and social order in contemporary society. Chicago, IL: University of Chicago Press. George, W. H., & Martinez, L. J. (2002). Victim blaming in rape: Effects of victim and perpetrator race, type of rape, and participant racism. Psychology of Women Quarterly, 26, 110–119. doi:10.1111/pwqu.2002.26.issue-2 Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-based studies? A comparative analysis of six preconceptions about Internet questionnaires. American Psychologist, 59, 93–104. doi:10.1037/0003-066X.59.2.93 Gover, A. R., Kaukinen, C., & Fox, K. A. (2008). The relationship between violence in the family of origin and dating violence among college students. Journal of Interpersonal Violence, 23, 1667–1693. doi:10.1177/0886260508314330 Gracia, E., & Tomas, J. M. (2014). Correlates of victim blaming attitudes regarding partner violence against women among the Spanish general population. Violence Against Women, 20, 26–41. doi:10.1177/1077801213520577 Grasmick, H. G., & McGill, A. L. (1994). Religion, attribution style, and punitiveness toward juvenile offenders. Criminology, 32, 23–46. doi:10.1111/crim.1994. 32.issue-1 29 30 C. Mancini and J. T. Pickett Grubb, A., & Harrower, J. (2008). Attribution of blame in cases of rape: An analysis of participant gender, type of rape and perceived similarity to the victim. Aggression and Violent Behavior, 13, 396–405. doi:10.1016/j.avb.2008.06.006 Downloaded by [University of Warwick] at 07:24 30 December 2015 Grubb, A., & Turner, E. (2012). Attribution of blame in rape cases: A review of the impact of rape myth acceptance, gender role conformity, and substance use on victim blaming. Aggression and Violent Behavior, 17, 443–452. doi:10.1016/ j.avb.2012.06.002 Hafer, C. L., & Bègue, L. (2005). Experimental research on just-world theory: Problems, developments, and future challenges. Psychological Bulletin, 131, 128–167. doi:10.1037/0033-2909.131.1.128 Harrell, E. (2007). Black victims of violent crime. Washington, DC: Bureau of Justice Statistics. Hayes, R. M., Lorenz, K., & Bell, K. A. (2013). Victim blaming others: Rape myth acceptance and the just world belief. Feminist Criminology, 8, 202–220. doi:10.1177/ 1557085113484788 Henning, K., & Holdford, R. (2006). Minimization, denial, and victim blaming by batterers: How much does the truth matter? Criminal Justice and Behavior, 33, 110–130. doi:10.1177/0093854805282322 Herek, G. M., Cogan, J. C., & Gillis, J. R. (2002). Victim experiences in hate crimes based on sexual orientation. Journal of Social Issues, 58, 319–339. doi:10.1111/ 1540-4560.00263 Holtfreter, K., Reisig, M. D., Piquero, N. L., & Piquero, A. R. (2010). Low self-control and fraud: Offending, victimization, and their overlap. Criminal Justice and Behavior, 37, 188–203. doi:10.1177/0093854809354977 Howard, J. A. (1984). Societal influences on attribution: Blaming some victims more than others. Journal of Personality and Social Psychology, 47, 494–505. doi:10.1037/0022-3514.47.3.494 Hunzaker, M. B. F. (2014). Making sense of misfortune: Cultural schemas, victim redefinition, and the perpetuation of stereotypes. Social Psychology Quarterly, 77, 166–184. doi:10.1177/0190272514521219 Jennings, W. G., Higgins, G. E., Tewksbury, R., Gover, A. R., & Piquero, A. R. (2010). A longitudinal assessment of the victim-offender overlap. Journal of Interpersonal Violence, 25, 2147–2174. doi:10.1177/0886260509354888 Jespersen, A. F., Lalumière, M. L., & Seto, M. C. (2009). Sexual abuse history among adult sex offenders and non-sex offenders: A meta-analysis. Child Abuse & Neglect, 33, 179–192. doi:10.1016/j.chiabu.2008.07.004 Katz, C. M., Webb, V. J., Fox, K., & Shaffer, J. N. (2011). Understanding the relationship between violent victimization and gang membership. Journal of Criminal Justice, 39, 48–59. doi:10.1016/j.jcrimjus.2010.10.004 Kreuter, F., Presser, S., & Tourangeau, R. (2008). Social desirability bias in CATI, IVR, and web surveys: The effects of mode and question sensitivity. Public Opinion Quarterly, 72, 847–865. doi:10.1093/poq/nfn063 Lauritsen, J. L., & Laub, J. H. (2007). Understanding the link between victimization and offending: New reflections on an old idea. Crime Prevention Studies, 22, 55–75. Lerner, M. (1980). The belief in a just world. New York, NY, USA: Plenum Press. Levit, N. (2011). Reshaping the narrative debate. Seattle University Law Review, 34, 751–765. Victim-Offender Overlap Perceptions and Victim Blaming Lucas, J. W. (2003). Theory-testing, generalization, and the problem of external validity. Sociological Theory, 21, 236–253. doi:10.1111/soth.2003.21.issue-3 Matsueda, R. L., & Drakulich, K. (2009). Perceptions of criminal injustice, symbolic racism, and racial politics. The Annals of the American Academy of Political and Social Science, 623, 163–178. doi:10.1177/0002716208330500 Downloaded by [University of Warwick] at 07:24 30 December 2015 Mears, D. P., Pickett, J., Golden, K., Chiricos, T., & Gertz, M. (2013). The effect of interracial contact on whites’ perceptions of victimization risk and black criminality. Journal of Research in Crime and Delinquency, 50, 272–299. doi:10.1177/ 0022427811431156 Murayama, K., Pekrun, R., & Fiedler, K. (2014). Research practices that can prevent an inflation of false-positive rates. Personality and Social Psychology Review, 18, 107–118. doi:10.1177/1088868313496330 Murphy, C. (2014, January 9). Crime victim compensation policies leave too many behind. Colorlines. Retrieved from http://www.colorlines.com/articles/crime-victimcompensation-policies-leave-too-many-behind Nobles, M. R., Fox, K. A., Piquero, N. L., & Piquero, A. R. (2009). Career dimensions of stalking victimization and perpetration. Justice Quarterly, 26, 476–503. doi:10.1080/07418820802427833 Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY, USA: McGraw-Hill. Payne, B. K., & Chappell, A. (2008). Using student samples in criminological research. Journal of Criminal Justice Education, 19, 175–192. doi:10.1080/10511250802 137226 Pickett, J. T., & Baker, T. (2014). The pragmatic American: Empirical reality or methodological artifact? Criminology, 52, 195–222. doi:10.1111/crim.2014.52.issue-2 Pickett, J. T., Mancini, C., & Mears, D. P. (2013). Vulnerable victims, monstrous offenders, and unmanageable risk: Explaining public opinion on the social control of sex crime. Criminology, 51, 729–759. doi:10.1111/crim.2013.51.issue-3 Pickett, J. T., Mancini, C., Mears, D. P., & Gertz, M. (2015). Public (mis)understanding of crime policy: The effects of criminal justice experience and media reliance. Criminal Justice Policy Review, 26, 500–522. doi:10.1177/0887403414526228 Piquero, A. R., & Bouffard, J. A. (2007). Something old, something new: A preliminary investigation of Hirschi’s redefined self-control. Justice Quarterly, 24, 1–27. doi:10.1080/07418820701200935 Pizarro, J. M., Zgoba, K. M., & Jennings, W. G. (2011). Assessing the interaction between offender and victim criminal lifestyles and homicide type. Journal of Criminal Justice, 39, 367–377. doi:10.1016/j.jcrimjus.2011.05.002 Policastro, C., & Payne, B. K. (2013). The blameworthy victim: Domestic violence myths and the criminalization of victimhood. Journal of Aggression, Maltreatment & Trauma, 22, 329–347. doi:10.1080/10926771.2013.775985 Posick, C. (2013). The overlap between offending and victimization among adolescents: Results from the Second International Self-Report Delinquency Study. Journal of Contemporary Criminal Justice, 29, 106–124. doi:10.1177/1043986212471250 Pratt, T. C., Turanovic, J. J., Fox, K. A., & Wright, K. A. (2014). Self-control and victimization: A meta-analysis. Criminology, 52, 87–116. doi:10.1111/crim.2014. 52.issue-1 31 32 C. Mancini and J. T. Pickett Pyrooz, D. C., Moule, R. K., & Decker, S. H. (2014). The contribution of gang membership to the victim-offender overlap. Journal of Research in Crime and Delinquency, 51, 315–348. doi:10.1177/0022427813516128 Roberts, J. V., & Hough, M. (2005). Understanding public attitudes to criminal justice. New York, NY, USA: Open University Press. Downloaded by [University of Warwick] at 07:24 30 December 2015 Rye, B. J., Greatrix, S. A., & Enright, C. S. (2006). The case of the guilty victim: The effects of gender of victim and gender of perpetrator on attributions of blame and responsibility. Sex Roles, 54, 639–649. doi:10.1007/s11199-006-9034-y Sample, L. L. (2006). An examination of the degree to which sex offenders kill. Criminal Justice Review, 31, 230–250. doi:10.1177/0734016806292929 Schreck, C. J., Stewart, E. A., & Osgood, D. W. (2008). A reappraisal of the overlap of violent offenders and victims. Criminology, 46, 871–906. doi:10.1111/ crim.2008.46.issue-4 Simmons, A. D., & Bobo, L. D. (in press). Can non-full-probability Internet surveys yield useful data? A comparison with full-probability face-to-face surveys in the domain of race and social inequality attitudes. Sociological Methodology. doi:10.1177/ 0081175015570096 Simon, J. (2007). Governing through crime: How the war on crime transformed American democracy and created a culture of fear. New York, NY, USA: Oxford. Steffensmeier, D., & Allan, E. (1996). Gender and crime: Toward a gendered theory of female offending. Annual Review of Sociology, 22, 459–487. doi:10.1146/ annurev.soc.22.1.459 Stillman, S. (2007). “The missing white girl syndrome”: Disappeared women and media activism. Gender & Development, 15, 491–502. doi:10.1080/13552070701630665 Suarez, E., & Gadalla, T. M. (2010). Stop blaming the victim: A meta-analysis on rape myths. Journal of Interpersonal Violence, 25, 2010–2035. doi:10.1177/ 0886260509354503 Templeton, L. J., & Hartnagel, T. F. (2012). Causal attributions of crime and the public’s sentencing goals. Canadian Journal of Criminology and Criminal Justice, 54, 45–65. doi:10.3138/cjccj.2010.E.29 Tillyer, M. S., & Wright, E. M. (2014). Intimate partner violence and the victim-offender overlap. Journal of Research in Crime and Delinquency, 51, 29–55. doi:10.1177/ 0022427813484315 Turanovic, J. J., & Pratt, T. C. (2013). The consequences of maladaptive coping: Integrating general strain and self-control theories to specify a causal pathway between victimization and offending. Journal of Quantitative Criminology, 29, 321–345. doi:10.1007/s10940-012-9180-z Unnever, J. D. (2005). Bullies, aggressive victims, and victims: Are they distinct groups? Aggressive Behavior, 31, 153–171. doi:10.1002/(ISSN)1098-2337 Unnever, J. D., Cochran, J. K., Cullen, F. T., & Applegate, B. K. (2010). The pragmatic American: Attributions of crime and the hydraulic relation hypothesis. Justice Quarterly, 27, 431–457. doi:10.1080/07418820902855362 Unnever, J. D., & Cullen, F. T. (2009). Empathetic identification and punitiveness: A middle range theory of individual differences. Theoretical Criminology, 13, 283–312. doi:10.1177/1362480609336495 Victim-Offender Overlap Perceptions and Victim Blaming Unnever, J. D., Cullen, F. T., & Jones, J. D. (2008). Public support for attacking the “root causes” of crime: The impact of egalitarian and racial beliefs. Sociological Focus, 41, 1–33. doi:10.1080/00380237.2008.10571321 Downloaded by [University of Warwick] at 07:24 30 December 2015 Waaland, P., & Keeley, S. (1985). Police decision making in wife abuse: The impact of legal and extralegal factors. Law and Human Behavior, 9, 355–366. doi:10.1007/ BF01044476 Weinberg, J. D., Freese, J., & McElhattan, D. (2014). Comparing data characteristics and results of an online factorial survey between a population-based and a crowdsource-recruited sample. Sociological Science, 1, 292–310. doi:10.15195/ issn.2330-6696 Welch, K., Payne, A. A., Chiricos, T., & Gertz, M. (2011). The typification of Hispanics as criminals and support for punitive crime control policies. Social Science Research, 40, 822–840. doi:10.1016/j.ssresearch.2010.09.012 Wiecko, F. M. (2010). Research note: Assessing the validity of college samples: Are students really that different? Journal of Criminal Justice, 38, 1186–1190. doi:10.1016/j.jcrimjus.2010.09.007 Wolfgang, M. E. (1957). Victim precipitated criminal homicide. The Journal of Criminal Law, Criminology, and Police Science, 48, 1–11. doi:10.2307/1140160 Wolfgang, M. E., & Singer, S. I. (1978). Victim categories of crime. Journal of Criminal Law and Criminology, 69, 379–394. doi:10.2307/1142333 Worden, A. P., & Carlson, B. E. (2005). Attitudes and beliefs about domestic violence: Results of a public opinion survey: II. Beliefs about causes. Journal of Interpersonal Violence, 20, 1219–1243. doi:10.1177/0886260505278531 Yamawaki, N., Ochoa-Shipp, M., Pulsipher, C., Harlos, A., & Swindler, S. (2012). Perceptions of domestic violence: The effects of domestic violence myths, victim’s relationship with her abuser, and the decision to return to her abuser. Journal of Interpersonal Violence, 27, 3195–3212. doi:10.1177/0886260512441253 Yeager, D., Krosnick, J. A., & Javitz, H. A. (2009). More on the problems with opt-in Internet surveys [Blog post]. Retrieved from http://abcnews.go.com/blogs/politics/ 2009/09/guest-blog-more-on-the-problems-with-optin-Internet-surveys/ 33