Victim Precipitation

Abstract

Studies assessing the overlap between victimization and offending have yet to include an examination of this phenomenon within victim precipitated homicide. In recognition of this gap in the literature, this study draws from official police data pertaining to 895 homicide incidents in Dallas, Texas to test the hypothesis that the victim-offender overlap is most prevalent within victim precipitated homicides, particularly when victims’ offending histories are considered. Results from a series of bivariate and multivariate models find that victims with criminal histories (i.e., victim-offenders) are most concentrated within victim precipitated homicides. Implications of these findings and suggestions for future research are presented.

Keywords

victim precipitation victim-offender overlap homicide homicide victimization homicide offending

Introduction

Homicide is, without question, a topic that has been examined in a variety of ways. This body of work has included theories pertaining to why offenders commit homicide, hypotheses that describe under what circumstances homicide take place, an examination of factors that may contribute to the commission of homicide and, to some extent, suppositions that suggest ways to prevent, or at least reduce, the occurrence of homicide (for a review of this research see Smith & Zahn, 1999). While there have been numerous studies evaluating and confirming the overlap between victimization and offending within interpersonal violence (see Jennings, Piquero, & Reingle, 2012, for a review), little has been written about the overlap between homicide victims and homicide offenders. This small but growing body of work has yet to investigate the linkages between the victim-offender overlap and victim precipitation, despite Wolfgang’s (1958, 1967) early homicide work suggesting the two were related.

The current research will contribute to the existing literature through an exploration of the overlap between victimization and offending within homicide, with special attention directed to victim precipitated homicides. To the authors’ knowledge, there is no known scholarly literature linking these two phenomena, the victim-offender overlap and victim precipitation, together. This study addresses this gap in the literature by examining the victim-offender overlap traditionally (i.e., victims and offenders as separate entities) and with a more nuanced approach first advocated by Mustaine and Tewksbury (2000) that assesses the similarities and differences between victims with no criminal histories, victims with criminal histories (otherwise referred to as victim-offenders), and offenders. By doing so, valuable insight into the relationship between victimization, offending, and victim precipitated violence will be gained as it is theorized that the victim-offender overlap increases the likelihood of victim precipitated homicide.

Victim-Precipitated Homicide

In very basic terms, a victim precipitated homicide is an event in which the victim’s behavior facilitated his or her own murder. By and large, the notion of victim precipitated homicide is credited to Marvin Wolfgang (1967) who argued that the victim can, at times, directly influence the outcome of a violent encounter through his or her own behavior. Defining victim precipitated homicide as an incident in which “the role of the victim is characterized by his having been the first in the homicide drama to use physical force directed against his subsequent slayer” (p. 73), Wolfgang (1967) found that slightly over one quarter of the homicides in his Philadelphia study were victim precipitated.

Initially, the idea of victims precipitating their own victimization may seem off-putting and appear to have undertones of victim blaming, especially within the context of interpersonal violence (Miethe, 1985). Take for example Amir’s (1967) study of forcible rape. Amir (1967), a student of Wolfgang, defined victim precipitated rape as

Those rape situations in which the victim actually, or so it was deemed, agreed to sexual relations but retracted before the actual act or did not react strongly enough when the suggestion was made by the offender(s). The term applies also to cases in risky or vulnerable situations, marred with sexuality, especially when the victim uses what could be interpreted as indecency in language and gestures, or constitute what could be taken as an invitation to sexual relations. (p. 495)

Critics sharply criticized Amir, and the study of victim precipitation more generally, for having engaged in a thinly veiled exercise of victim blaming (see Clark & Lewis, 1977). The end result of such criticisms was a virtual halt on victim precipitation research for nearly three decades.

To avoid accusations of victim blaming, subsequent researchers have abided by a rather stringent measure of victim precipitation; one that has relied primarily on whether or not the victim was the first to use physical force (Felson & Cares, 2005; Hannon, 2004; Wolfgang, 1967). Two exceptions include work completed by Felson and Messner (1998) and Muftić, Bouffard, and Bouffard (2007). For example, Felson and Messner (1998) resurrected the concept of victim precipitation and applied it to the study of intimate homicide. Like Wolfgang (1967), Felson and Messner (1998, p. 406) measured victim precipitated intimate homicide as “incidents in which a physical attack by the victim provokes the offender’s lethal attack.” However, they expanded on their measurement by considering the criminality of the one delivering the fatal blow (i.e., the offender). Specifically, Felson and Messner (1998) asserted that an offender without a violent past was unlikely to have initiated the homicide incident; rather, they were more likely engaging in defensive (as opposed to offensive) violence and hence the victim was precipitating the homicidal act. Overall they found some support for their position that an examination of offending histories may be relevant to the study of victim precipitated intimate homicides.

Most recently, victim precipitation has been applied to the study of nonlethal violence (Felson & Cares, 2005; Muftić et al, 2007). For instance, Muftić and colleagues (2007) examined victim precipitation within the context of nonlethal intimate partner violence (IPV). Relying on police IPV incident narratives, they found that 30% of victims were the first to use violence. In an attempt to expand on traditional measures of victim precipitation, Muftić et al. (2007) also considered prior domestic violence incidents in their victim precipitation measure. An analysis of officially recorded criminal histories revealed that 10% of victims had IPV offending histories. Subsequently, IPV incidents involving victims with official criminal histories (i.e., victim-offenders as they are called in the victim-offender literature) were coded as victim precipitated. Results from a logistic regression model provided support for their argument that within dual arrest IPV incidents, victims with a criminal history were more likely to have provoked the incident (indicating victim precipitation) than victims without a criminal history. As a whole, the victim precipitation literatures suggests an overlap between victims and offenders within incidents of victim precipitated violence; a line of inquiry yet to be empirically explored in the research literature.

The Victim-Offender Overlap

Not long after Wolfgang’s seminal work (1958, 1967), scholars studying violence began to notice a pattern. While traditionally treated as separate, distinct groups, victims and offenders appear to share many demographic and behavioral similarities (Gottfredson, 1984; Singer, 1981). Drawing from routine activities and lifestyle perspectives to explain such linkages, criminologists began to argue that “victims and offenders are often one in the same” (Sampson & Lauritsen, 1990, p. 110). Mustaine and Tewksbury (2000), however, have postulated that while a growing body of work has found that many victims have a history of offending (see Jennings et al., 2012), there still remains a sizeable proportion of victims without offending histories and these victims share little demographic or behavioral resemblance to offenders (Broidy, Daday, Crandall, Skylar, & Jost, 2006). Thus, when evaluating the overlap between victimization and offending, a consideration of three categories of individuals concerned is warranted; victims (i.e., individuals with no history of offending), offenders (i.e., individuals with no history of victimization), and victim-offenders (i.e., individuals who switch between offending and victimization behaviors). Research that has examined the possibility of three groupings finds support for Mustaine and Tewksbury’s (2000) assertion that while an overlap exists for a majority of victims and offenders, the overlap is not present for all victims (Broidy et al., 2006; Klevens, Duque, & Ramirez, 2002; Maldonado-Molina, Jennings, Tobler, Piquero, &Canino, 2010; Muftić, Finn, & Marsh, 2012; Reingle, Staras, Jennings, Branchini, & Maldonado-Molina, 2011). The current study will rely on Mustaine and Tewksbury’s (2000) categorical measure of the overlap that involves a comparison of victims, victim-offenders, and offenders.

Within the Study of Homicide

There has been less attention directed at the overlap between victimization and offending within the study of homicide (Broidy et al., 2006; Dobrin, 2001; Ezell & Tanner-Smith, 2009; Piquero, MacDonald, Dobrin, Daigle, & Cullen, 2005; Pizarro, Zgoba, & Jennings, 2011; Regoeczi, 2000; Wolfgang, 1958, 1967) than within other types of nonlethal interpersonal violence (Bjarnason, Sigurdardottir, &Thorlindsson, 1999; Daday, Broidy, Crandall, & Sklar, 2005; Jennings, Park, Tomisich, Gover, & Akers, 2011; Klevens et al., 2002; Lauritsen& Davis Quinet, 1995; Muftić, Finn, & Marsh, 2012; Mustaine & Tewksbury, 2000; Reingle et al., 2011; Sampson &Lauritsen, 1990; Schreck, Stewart, & Osgood, 2008; Silver, Piquero, Jennings, Piquero, &Leiber, 2011; Wittebrood& Nieuwbeerta, 1999). En bloc the extant homicide research demonstrates mixed support for the presence of an overlap between victimization and offending. Case in point, Regoeczi (2000) and Piquero et al. (2005) found limited evidence for the victim-offender overlap as it relates to homicide victimization. Other studies, however, find more support indicating that anywhere up to 50% of homicide victims had previous offending records, and prior offending experiences are a significant predictor of homicide victimization at both the bivariate (Wolfgang, 1967) and multivariate levels (Broidy et al., 2006; Dobrin, 2001; Ezzell & Tanner-Smith, 2009; Pizarro et al., 2011).

For instance, in one of the first empirical attempts to determine whether previous offending increases homicide victimization risk, Dobrin (2001) compared homicide victims to nonvictims and found that prior offending was a significant predictor of homicide victimization even after controlling for demographic and socioeconomic factors. More recently, Ezell and Tanner-Smith (2009) examined the extent that gang membership, alcohol and drug abuse, and arrest histories predict homicide victimization among postrelease juvenile male offenders in California. Drawing from lifestyle and routine activities theories, Ezell and Tanner-Smith (2009) provided support for their claim that homicide victimization risk is greatest among offenders who engaged in more criminogenic lifestyles.

Broidy and colleagues (2006) examined the overlap between offending and homicide victimization in New Mexico. Utilizing criminal justice, health care, and Census data, they discovered that victims and offenders shared many similar demographic characteristics, structural environments, and risky behaviors, supporting the victim-offender overlap hypothesis. They also uncovered a considerable number of victims who did not appear to overlap with offenders. Broidy et al. (2006) then compared victims with no criminal histories (victims only) to victims with criminal histories (victim-offenders) and offenders. Results from a series of post hoc subgroup comparisons lent further support to the assertion that within homicide three distinct groups exist: victims, offenders, and victim-offenders.

Lastly, and probably most relevant to the current study, Pizarro and colleagues (2011) examined 513 homicide incidents that occurred within Newark, New Jersey between 1997 and 2007. Victim and offender involvement in criminal activities such as gang membership, drug dealing, priors for violent and/or weapon offenses, property offenses, and drug offenses was measured and typologies were developed with separate groupings created for offenders and victims that were dependent on level of prior criminal involvement (e.g., high vs. low). Interactions between the different types of victims and offenders were analyzed to determine their impact on specific types of homicide (e.g., domestic, gang/drug related, escalating dispute/revenge related, and robbery related). Overall, while there was little variation between victims and offenders, the variation that did exist was predictive of different homicide types. For instance, the most criminally involved offenders and victims had the greatest likelihood of being engaged in gang- and drug-related homicides. In contrast, the least criminally involved offenders and victims had the greatest probability of being involved in a homicide of a domestic nature.

Linkages With Victim Precipitation

The growing body of literature devoted to the study of the relationship between victimization and offending typically references Wolfgang’s (1967) work on victim precipitated homicide and acknowledges his study as one of the first to identify what is now referred as the victim-offender overlap. Yet current research in this area has failed to include a discussion, or analysis, of how victim precipitation relates to the overlap between victimization and offending. This is unfortunate as victim precipitation may be useful in advancing our understanding of just how offenders become victims and how victims become offenders because, as Chen (2009, p. 119) argues, the “mechanism that leads to this association remains unclear.” Perhaps this is because most of the victim-offender overlap literature looks almost exclusively at offending histories¹ and their linkage with victimization, relying on the principle of homogamy to explain such relationships.

Little attention has been directed at the context of the criminal event. Yet, according to a state-dependency argument, “experiences can alter an individual’s risk for crime” (Lauritsen & Davis Quinet, 1995, p. 146). One could argue that victim precipitation is the prime example of state dependency. For example, Lauritsen, Sampson, and Laub (1991, p. 268) argue that “assaults by definition involve more than one person so that the risk of becoming a victim given that one is the initial offender (e.g., struck the first blow) is at a maximum.” It is plausible that this type of behavior is more likely among victim-offenders than nonoffending victims. As such, an examination of victim precipitation aids in understanding the chain of events that takes someone from being the offender to being the victim and vice versa.

Current Study

Research examining the linkage between victimization and offending has demonstrated that crime victims are also, at times, offenders; referred to as the victim-offender overlap (Jennings et al., 2012). Additionally, there are clear indications that the victim of a crime can be the initial catalyst of the criminal event; referred to as victim precipitation (Wolfgang, 1967). Both concepts, the victim-offender overlap and victim precipitation, highlight the demographic and behavioral similarities between victims and offenders. To date, researchers have failed to unify these two ideas in the study of lethal violence. The current study adds to the existing literature by assessing the overlap between victimization and offending within homicide by focusing on victim precipitated homicides. Specifically, it is speculated that within victim precipitated incidents there is more victim-offender overlap than within non-victim precipitated incidents. This hypothesis is guided by previous victim-offender overlap homicide research, which suggests that “the population of victims with no arrest record may represent a distinct group that differs significantly from the offender population and, more important, from the rest of the victim population” (Broidy et al., 2006, p. 171). It is anticipated this may be especially true within victim precipitated homicides. Subsequently, it is hypothesized that victims with a criminal history (referred to as victim-offenders) will be different than victims without a criminal history (referred to as victims) but similar to homicide suspects. Furthermore, victim-offenders will be concentrated within victim precipitated homicides.

Method

Dallas Homicide Data Set

The data used in this article came from 10 years of homicide records maintained by the Dallas, Texas, Police Department (DPD) beginning in 1988 and concluding in 1997. During that time period, 3,084 homicidal events were recorded. Since this analysis is focused exclusively on adult offending and victimization, cases in which the age of the suspect and/or victim was less than 18 years, or the victim and/or suspect age was unknown were eliminated. Deletion of juveniles and unknowns from the database yielded a final population of 1,712 homicide cases.

Victim Precipitation Measures

Homicides were dichotomously coded as victim precipitated (yes, no) based on the narrative written in the report of incident by the responding officer. In accordance with Wolfgang’s (1967) original definition, a homicide was considered to be victim precipitated if the victim committed the initial physical action that led to his or her victimization. Careful consideration was given not to make inferences from what was written in the police narrative. Instead, it was decided to use cases where determinations of victim precipitation could clearly be made. Analyzing the narrative under this premise, 438 cases (25.6%) were found to be clearly victim precipitated. Additionally, 457 cases (26.7%) were determined not to be victim precipitated. The remaining 817 cases (47.7%) were not able to be classified as victim precipitated or not.² For purposes of this study, these 817 cases were eliminated from the data set.³ The final sample includes 895 homicide incidents involving 895 victims and 1,004 suspects.⁴ Victim precipitated homicides (n = 438) represent 48.9% of the homicides included for analysis.

Victim-Offender Measures

As recommended by Mustaine and Tewksbury (2000) and replicated by Broidy et al. (2006) in their study of homicide victimization, victim-offender overlap measures were created based on the official criminal histories of victims. Individuals were categorized as a victim if they were a homicide victim without an officially recorded criminal history. Individuals were considered to be a victim-offender if they were a homicide victim with an officially recorded criminal history. It should be noted that suspects could not be considered for the victim-offender category because there were not any measures of prior victimization within the data set.

Demographic, Behavioral, and Incident Measures

The first set of variables taken from the DPD’s Homicide Unit’s electronic files reflect the demographic characteristics of the victim and the suspect(s) involved in each homicide, including age (years), sex (male, female), race (White, Non-White), criminal history⁵ (yes, no), and alcohol and drug use at time of incident⁶ (yes, no). The second set of variables center around descriptions of the homicide incident of interest, including whether the homicide occurred during a dispute or altercation (yes, no), whether a firearm was the primary method by which the victim was killed (yes, no), and whether the homicide was victim precipitated (yes, no).

As depicted in Table 1, overall Dallas homicide victims were largely male (82.6%), non-White (77.4%), and in their early 30s (M = 33.5 years, range 18-94, SD = 14.5). Similarly, the vast majority of homicide suspects were male (92.0%) and non-White (85.0%). Homicide suspects were, however, younger than homicide victims with the median age being 29.5 years (SD = 12.2, range 18-87). Overall, less than one quarter of victims (20.8%) and suspects (19.8%) had an officially recorded criminal history. Roughly one third of victims (36.6%) and one quarter of suspects (24.6%) had used alcohol at the time of the homicide incident. Drug use was less common with 26.0% of victims and 19.6% suspects having used drugs at the time of the homicide incident. The vast majority of victims (82.7%) within this sample were killed by a firearm. Roughly 4 out of 10 (42.1%) homicide incidents involved a dispute that transpired between the victim and suspect(s). Lastly, almost half (48.9%) of homicides studied were victim precipitated.

Table 1.

Descriptive Statistics of Sample (N = 1,917).

Independent variables	Percentage
Victim characteristics^a
Age in years (SD)	33.5 (14.5)
Male	82.6%
Non-White	77.4%
Criminal history	20.8%
Alcohol use	36.6%
Drug use	26.0%
Suspect characteristics^b
Age in years (SD)	29.5 (12.2)
Male	92.0%
Non-White	85.0%
Criminal history	19.8%
Alcohol use	24.6%
Drug use	19.6%
Homicide characteristics^c
Firearm	82.7%
Dispute	42.1%
Victim precipitation	48.9%

n = 913.

n = 1,004.

n = 895.

Analytic Strategy

This study set out to investigate two hypotheses largely guided by the victim-offender overlap literature. We first hypothesized that victims with criminal histories, referred to as victim-offenders, are demographically and behaviorally dissimilar from victims without criminal histories but similar to homicide suspects. Offenders involved all Individuals who were identified as suspects in the DPD database. We then hypothesized that victim-offenders are more likely to be involved in a victim precipitated homicide than victims without criminal histories or suspects. To test these hypotheses, different statistical models were generated. Bivariate statistics were calculated to examine differences in demographics (sex, race, age) and behaviors (alcohol use, drug use, victim precipitation) for victim-offenders compared to victims and suspects. Multinomial logistic regression was used to predict the type of category (victim, victim-offender, suspect) based on homicide type (victim precipitated or nonvictim precipitated), while controlling for sex, race, age, alcohol use, and drug use. We now turn to the results from these analyses.

Results

To test the first hypothesis that victims with criminal histories (victim-offenders) are different from victims without criminal histories but similar to suspects, chi-square and t tests were computed (see Table 2). As expected, victim-offenders were demographically and behaviorally dissimilar from victims. For instance, victim-offenders were more likely to be male, χ²(1) = 12.714, p < .001, non-White, χ²(1) = 6.631, p < .01, and younger, t(893) = 3.372, p < .001, than victims. Victim-offenders were also more likely to have used alcohol, χ²(1) = 2.827, p < .05, and drugs, χ²(1) = 35.144, p < .001, prior to the homicidal event than victims.

Table 2.

Bivariate Comparisons of Victim-Offenders to Victims and Suspects.

	Victims^a	Victim-offenders^b	Suspects^c
Male	80.3%	91.4%	92.0%
	χ²(1) = 12.714***		χ²(1) = .048
Non-White	75.5%	84.4%	85.0%
	χ²(1) = 6.631**		χ²(1) = .079
Age in years (SD)	34.3 (15.3)	30.3 (10.5)	29.6 (12.2)
	t(893) = 3.372***		t(964) = –.819
Alcohol use	35.3%	41.9%	24.6%
	χ²(1) = 2.827*		χ²(1) = 23.281***
Drug use	21.6%	43.0%	19.6%
	χ²(1) = 35.144***		χ²(1) = 46.743***
Victim precipitation	43.0%	71.5%	48.6%
	χ²(1) = 47.851***		χ²(1) = 33.126***

n = 709.

n = 186.

n = 1,004.

p < .05, one-tailed. **p < .01. ***p < .001.

We then compared victim-offenders to suspects. Based on the victim-offender overlap literature, we predicted that we would find more similarities than differences between the two. In terms of demographics, this was the case as victim-offenders and suspects were similar in terms of sex, race, and age. When behavior was considered, however, significant and unanticipated differences were noted. Specifically, victim-offenders were almost twice as likely to have used alcohol, χ²(1) = 23.281, p < .001, or drugs, χ²(1) = 46.743, p < .001, prior to the homicidal event compared to suspects.

Finally, we hypothesized that victim precipitated homicides would be most likely to involve victim-offenders. As indicated in Table 2, we found support for this hypothesis as there is a significantly higher percentage of victim-offenders (71.5%) involved in victim precipitated homicides than victims, 43.0%; χ²(1) = 47.851, p < .001, or suspects, 48.6%, χ²(1) = 33.126, p < .001.

To determine whether this finding would hold up controlling for other variables, including risky behaviors such as alcohol and drug use, multinomial logistic regression was used. Victim-offenders were chosen as the reference category and compared against victims in the first model (comparison 1) and against suspects in the second model (comparison 2). The predictor variable of interest was victim precipitation while the control variables included male, non-White, age, alcohol use, and drug use. While results from multinomial logistic regression models produce both coefficient estimates (B) and odds ratios (Exp(B)), for ease of interpretation we focused on the odds ratio (OR). An odds ratio greater than one indicates an increased likelihood that an outcome is associated with the comparison group (e.g., victims) as opposed to the refergroup (e.g., victim-offenders). Alternatively, an odds ratio less than one indicates a decreased likelihood that the outcome is associated with the comparison group.

As depicted in Table 3, victim precipitation had a significant effect on the dependent variable, controlling for demographic and behavioral variables in the model. Specifically, victims (OR = .349, p < .001) and suspects (OR = .610, p < .001) are less likely to be involved in a victim precipitated homicide than victim-offenders. In other words, victim precipitated homicide incidents have 65% lesser odds of having a victim, and a 39% lesser odds of having a suspect, compared to a victim-offender. Additionally, demographic and behavioral differences remain significant in the multivariate model lending support to our hypothesis that victims and victim-offenders are dissimilar. Less disparity was found between victim-offenders and suspects as only drug use remained significant in the multinomial regression model (OR = .610, p < .01).

Table 3.

Victim-Offender Overlap and Victim Precipitation: Multinomial Logistic Regression Coefficients and Change in Odds.

	Comparison 1: Victims to victim-offenders			Comparison 2: Offenders to victim-offenders
	B	ExpB	SE	B	ExpB	SE
Male	.737**	2.089	.286	−.066	.936	.299
Non-White	−.170	.843	.235	−.162	.850	.234
Age (years)	.012*	1.012	.007	−.007	.993	.007
Alcohol use	.274	1.315	.193	−.107	.899	.190
Drug use	−.905***	.405	.200	−.771***	.463	.194
Victim precipitation	−1.054***	.349	.183	−.494**	.610	.183

Note: N = 1,588. Pseudo-R² = .062. χ²(6) = 36.010, p < .001.

p < .05, one-tailed. **p < .01. ***p < .001.

Risk ratios, which were calculated from conditional probabilities, further illustrate this finding. Specifically, non-White males who are of average age and who drank and used drugs prior to the victim precipitated homicide incident were 2.4 times more likely to be a victim-offender than a victim and 1.4 times more likely to be victim-offender than a suspect. Overall, these results suggest that homicide victims with criminal histories (victim-offenders) share more in common with homicide suspects than they do with homicide victims without prior offending histories and that victim-offenders are more likely to have precipitated their own homicide compared to victims.

Discussion

This study sought to explore the overlap between victimization and offending within victim precipitated homicide; an area of study neglected in the research literature. In our first hypothesis, we set out to examine the overlap between homicide victims and suspects. Previous research, however, has found that victims are not a homogenous group. Rather, victims can be differentiated by their involvement in criminal activity as offenders. As such, we separated victims with criminal histories (referred to as victim offenders) from victims without criminal histories (referred to simply as victims). We predicted that victim-offenders would share more demographic and behavioral characteristics with homicide suspects than with homicide victims. Findings from our bivariate and multivariate models largely support this hypothesis. Victim-offenders within this sample share more demographic similarities in regards to sex, race, and age with suspects than they do victims although distinct behavioral differences between the groupings were also uncovered. Particularly, victim-offenders engaged in more risky behaviors, including drug and alcohol use prior to their death. This finding is consistent with the results of the Broidy et al. (2006) study that found that victim-offenders had more extensive drug and alcohol use histories than suspects.

Victim-offenders also demonstrated a higher likelihood of having been involved in a victim precipitated homicide. This finding remained significant in the multivariate models computed, providing support for our second hypothesis. These findings coincide with that of Pizarro and colleagues (2011) who found that criminally involved victims and offenders were more likely to be involved in homicides associated with criminal events (e.g., drug-related homicides) than victims and suspects with lesser degrees of criminal involvement. Together, these two studies suggest that within homicide there exists a subset of victims who are enmeshed in highly criminogenic and subsequently victimogenic lifestyles and that such lifestyles influence homicide type.

From a practical standpoint, not only does the integration of these two bodies of work help to elucidate the overlap between victimization and offending but also the inclusion of different offending and victimization typologies (e.g., distinguishing victims with criminal histories from victims without criminal histories) may curtail accusations of victim blaming when talking about victim precipitation. Within the victim precipitation literature, researchers have myopically focused on how the victim triggers the behavior of the offender. In reality, it is more likely that within victim precipitated acts it is victim-offenders, and not nonoffending victims, who are inciting offenders, as found in this study. Within this context there may be a lessened perception of victim blaming.

While the research presented demonstrates the presence of the victim-offender overlap within victim precipitated homicides, there are several limitations to this study. The first limitation to note is the overall generalizability of the findings. The data set employed only pertains to homicides occurring in Dallas, Texas and is rather dated. As such, this study makes no attempt to generalize the results to other metropolitan areas across the nation. While Dallas, Texas is an urban area not unlike others in the United States, there are factors that make it unique. In the early 1990s, Dallas experienced one of the largest influxes of Latino immigrants into a metropolis in the United States. Simultaneously, Dallas suffered a surge in violent crime among minorities living in the city (evidenced by the high percentage of minority offenders and victims within our sample). It is recommended, then, that future studies should be conducted with more recent homicide data sets from other regions of the nation as there is an apparent need for research in this area as it promises to be fertile ground for the study of victim precipitated homicide, the victim-offender overlap, and homicide victimization.

Additionally, we were unable to examine what influence prior victimization experiences have among suspects in the sample. As such, our victim-offender category included only victims with criminal histories; suspects with victimization histories were excluded. Prior research indicates, however, that like victims, offenders are also a heterogeneous group and that differentiation by offending and victimization experiences is an important consideration within the victim-offender overlap (Muftić & Deljkić, 2012; Reingle & Maldonado-Molina, 2012). Future, more robust examinations of the victim-offender overlap within victim precipitated homicides would benefit from the inclusion of offender victimization histories.

Although this investigation was not a direct test of lifestyle or routine activities theories, the results are supportive of arguments associated with such perspectives (Cohen & Felson, 1979; Hindelang, Gottfredson, & Garofalo, 1978). Thus it is probable that a lifestyle or routine activities theoretical lens would provide more insight into the results found here (see Klevens et al., 2002; Pizarro et al., 2011). Likewise, subcultural considerations (e.g., Latin “machismo” or Black “toughness”; see Lee, 2011, and Anderson, 1999) were given no weight in this examination. While it is likely that subcultural factors may influence victim precipitated homicides (particularly among victim-offenders), the reliance on official police data made it infeasible to gather the type of information necessary to carry out such an analysis. Future studies should include direct measures of lifestyle/routine activities and subcultural theories as well as different research methods (e.g., ethnographic surveys) to better clarify the situational context in which homicides and victim precipitated events occur.

In sum, this study found more overlap between victims and offenders in cases of victim precipitated homicides than in cases of non-victim precipitated homicides. However, similar to recent victim-offender overlap work within the study of homicide, the reported findings do not entirely support previous arguments that an overlap exists for all homicide victims. When victims without criminal histories (victims) were distinguished from victims with criminal histories (victim-offenders), results reveal that homicides that include some degree of victim precipitation are more likely to involve victim-offenders than victims. Thus the results support the hypotheses presented; that victim-offenders are different than victims and that prior offending increases the likelihood of having been the victim of a victim precipitated homicide.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Lisa R. Muftić, PhD, is an assistant professor in the Department of Criminal Justice and Criminology at Georgia State University. Her current research interests are focused on examining the overlap between offending and victimization within interpersonal violence. She has published articles in such journals as Justice Quarterly, Crime and Delinquency, Violence Against Women, and Journal of Interpersonal Violence.

Donald E. Hunt is a graduate student in the Department of Criminal Justice and Criminology at Georgia State University. His research interests include the victim-offender overlap, white-collar crime in the corporate environment, and neurobiological effects of victimization and offending.

References

Amir

(1967). Victim precipitated forcible rape. Journal of Criminal Law, 58, 493-502.

Anderson

(1999). Code of the street: Decency, violence, and the moral life of the inner city. New York, NY: W. W. Norman.

Bjarnason

Sigurdardotlir

T. J.

Thorlindson

(1999). Human agency, capable guardians, and structural constraints: A lifestyle approach to the study of violent victimization. Journal of Youth and Adolescence, 28, 105-119.

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.

Chen

(2009). The linkage between deviant lifestyles and victimization: An examination from a life course perspective. Journal of Interpersonal Violence, 24, 1083-1110.

Clark

L. M. G.

Lewis

D. J.

(1977). Rape: The price of coercive sexuality. Toronto, Ontario, Canada: Canadian Women’s Educational Press.

Cohen

Felson

(1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44, 588-608.

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.

Dobrin

(2001). The risk of offending on homicide victimization: A case control study. Journal of Research in Crime and Delinquency, 38, 154-173.

10.

Ezell

M. E.

Tanner-Smith

E. E.

(2009). Examining the role of lifestyle and criminal history variables on the risk of homicide victimization. Homicide Studies, 13, 144-173.

11.

Felson

R. B.

Cares

A. C.

(2005). Gender and the seriousness of assaults on intimate partners and other victims. Journal of Marriage and Family, 67, 1182-1195.

12.

Felson

R. B.

Messner

S. F.

(1998). Disentangling the effects of gender and intimacy on victim precipitation in homicide. Criminology, 36, 405-423.

13.

Gottfredson

M. R.

(1984). Victims of crime: The dimensions of risk. Norwich, UK: Home Office Research and Planning Unit.

14.

Hannon

(2004). Race, victim precipitated homicide, and the subculture of violence thesis. Social Science Journal, 41, 115-121.

15.

Heyman

R. E.

Smith Slep

A. M.

(2002). Do child abuse and interparental violence lead to adulthood family violence? Journal of Marriage and Family, 64, 864-870.

16.

Hindelang

M. J.

Gottfredson

M. R.

Garofalo

(1978). Toward a theory of personal criminal victimization. In Hindelang

M. J.

Gottfredson

M. R.

Garofalo

(Eds.), Victims of personal crime: An empirical foundation for a theory of personal victimization. Cambridge, MA: Ballinger.

17.

Jennings

W. G.

Park

M. R.

Tomsich

E. A.

Gover

A. G.

Akers

R. L.

(2011). Assessing the overlap in dating violence perpetration and victimization among South Korean college students: The influence of social learning and self-control. American Journal of Criminal Justice, 36, 188-206.

18.

Jennings

W. G.

Piquero

A. R.

Reingle

J. M.

(2012). On the overlap between victimization and offending: A review of literature. Aggression and Violent Behavior, 17(1), 16-26.

19.

Klevens

Dugue

L. F.

Ramirez

(2002). The victim-perpetrator overlap and routine activities: Results from a cross-sectional study in Bogota, Columbia. Journal of Interpersonal Violence, 17, 206-216.

20.

Lauritsen

J. L.

Davis Quinet

K. F.

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

21.

Lauritsen

J. L.

Sampson

R. J.

Laub

J. H.

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

22.

Lee

M. R.

(2011). Reconsidering culture and homicide. Homicide Studies, 15, 319-340.

23.

Loeber

Pardini

Homish

D. L

Wei

Farrington

D. P.

Creemers

Rosenfeld

(2005). The prediction of violence and homicide in young men. Journal of Consulting and Clinical Psychology, 73, 1074-1088.

24.

Maldonado-Molina

M. M.

Jennings

W. G.

Tobler

A. L.

Piquero

A. R.

Canino

(2010). Assessing victim-offender overlap among Hispanic youth. Journal of Criminal Justice, 38, 1191-1201.

25.

Miethe

T. D.

(1985). The myth or reality of victim involvement in crime: A review and comment on victim precipitation research. Sociological Focus, 18, 209-220.

26.

Muftić

L. R.

Bouffard

L. E.

Bouffard

J. A.

(2007). An exploratory analysis of victim precipitation among men and women arrested for intimate partner violence. Feminist Criminology, 2, 327-346.

27.

Muftić

L. R.

Deljkić

(2012). Exploring the overlap between offending and victimization within intimate partner violence in Bosnia and Herzegovina. International Criminal Justice Review. Advance online publication. doi:10.1177/1057567712448476.

28.

Muftić

L. R.

Finn

M. A.

Marsh

(2012). The victim-offender overlap, intimate partner violence, and sex: Assessing differences among exclusive victims, exclusive offenders, and victim-offenders. Crime & Delinquency. doi:10.1177/0011128712453677

29.

Mustaine

Tewksbury

(2000). Comparing the lifestyles of victims, offenders, and victim-offenders: A routine activity theory assessment of similarities and differences for criminal incident participation. Sociological Focus, 33, 339-362.

30.

Piquero

A. R.

MacDonald

Dobrin

Daigle

L. E.

Cullen

F. T.

(2005). Self-control, violent offending, and homicide victimization: Assessing the general theory of crime. Journal of Quantitative Criminology, 21(1), 55-71.

31.

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.

32.

Regoeczi

W. C.

(2000). Adolescent violent victimization and offending: Assessing the extent of the link. Canadian Journal of Criminology and Criminal Justice, 42, 493-505.

33.

Reingle

J. M.

Maldonado-Molina

M. M.

(2012). Victimization and violent offending: An assessment of the victim-offender overlap among Native American adolescents and young adults. doi:10.1177/1057567712443966.

34.

Reingle

J. M.

Staras

S. A.

Jennings

W. G.

Branchini

Maldonado-Molina

M. M.

(2011). The relationship between marijuana use and intimate partner violence in a nationally representative, longitudinal sample. Journal of Interpersonal Violence, 27, 1562-1578.

35.

Sampson

R. J.

Lauritsen

J. L.

(1990). Deviant lifestyles, proximity to crime, and the offender-victim link in personal violence. Journal of Research in Crime and Delinquency, 27(2), 110-139.

36.

Schreck

C. J.

Stewart

E. A.

Osgood

D. W.

(2008). A reappraisal of the overlap of violent offenders and victims. Criminology, 46, 871-906.

37.

Silver

Piquero

A. R.

Jennings

W. G.

Piquero

N. L.

Leiber

(2011). Assessing the violent offending and violent victimization overlap among discharged psychiatric patients. Law and Human Behavior, 35, 49-59.

38.

Singer

S. I.

(1981). Homogeneous victim-offender populations: A review of some research implications. Journal of Criminal Law and Criminology, 72, 779-788.

39.

Smith

M. D.

Zahn

M. A.

(1999). Homicide: A sourcebook of social research. Thousand Oaks, CA: Sage.

40.

Wittebrood

Nieuwbeerta

(1999). Wages of sin? The link between offending, lifestyle and violent victimization. European Journal of Criminal Policy and Research, 7, 63-80.

41.

Wolfgang

M. E.

(1958). Patterns in criminal homicide. Oxford, UK: University of Pennsylvania Press.

42.

Wolfgang

M. E.

(1967). Victim precipitated criminal homicide. In Wolfgang

M. E.

(Ed.), Studies in Homicide (p. 72-87). New York, NY: Harper & Row.