Sex Worker Homicide Series: Profiling the Crime Scene

Abstract

Contrary to popular misconceptions, offenders who kill sex workers as part of their series exhibit substantial variability in their victim selection and behavioral patterns, thus creating additional issues for the investigation of these crimes. This article first aims to outline differences in the demographics of crime scene actions present in homicide series with exclusively sex worker victims and series that includes both sex worker and non–sex worker victims, with the aim of understanding the crime scene aetiology of these two different types of series. Second, the research aims to determine between-series differences of victimology as well as crime scene action between sex worker series and mixed-victim series. Third, the research focuses on mixed-victim series and aims to determine the within-series similarities of victimology and crime scene actions, that is, what factors link sex worker victims and non–sex worker victims in the same series. Data were collected through a large-scale review of international media sources to identify solved serial homicide cases that have included at least one sex worker. Of the 83 series looked at, 44 (53%) included sex worker victims only, and 39 (47%) of the series included both sex worker and non–sex worker victims. The findings highlight the challenges that these types of crime present for investigation and the implications they have on current crime analysis research and practice, and results are discussed in line with theoretical and psychological issues relating to understanding differentiation and similarity, as well as investigative implications relating to linkage blindness and linking of serial crimes.

Keywords

linking profiling crime scene analysis sex workers serial homicide serial sexual assault classification crossover

Offender Profiling and Crime Scene Investigation

The main aim of behavioral crime scene analysis, otherwise known as offender profiling, is to analyze the way an offender commits their crime, to establish discernible patterns and behavioral subtypes and then link subtypes of crime scene actions to the most likely offender background characteristics, and to use this in criminal investigations as a primary tool for the police to narrow their suspect pool down to statistically the most likely type of offender, and/or identify and link series of crimes.

Three general interlinked areas have been the focus of this behavioral crime scene analysis and offender profiling research: individual differentiation, behavioral consistency, and inferences about offender characteristics (Canter, 2000; Salfati, 2008).

Individual differentiation aims to establish differences between the behavioral actions of offenders and identify subgroups of crime scene types.

Behavioral consistency is a key issue in profiling, specifically for understanding both the development of an offender’s criminal career and an individual’s consistency across a series of crimes—that is, whether the same subsets of actions are displayed at each crime scene over a series of offenses.

Inferences about offender characteristics based on the way an offender acts at the time of the crime are at the core of profiling and use consistency analysis as its main focus.

The current article will focus specifically on the first component of identifying crime scene patterns in those cases that are part of a series that involves sex worker victims. The aim is to establish a baseline understanding of the aetiology of crime scenes that are part of serial homicides of sex workers and to use this as a first step to explore the key salient factors that are paramount for identifying crime scenes that may be part of the same series. Alongside being able to identify actions that can be used to link a series, it is also important to know how to differentiate one series from another series. Key to this process is the question of what aspects of the crime scene are the most reliable for making these determinations.

Focusing on the Victim in Crime Scene Analysis

In much of the literature on behavioral crime scene analysis and behavioral consistency, even when it focuses on behavior, the underlying psychological focus has been on how these behaviors may reflect something about how the offender interacts with the victim. At the heart of the behavioral crime scene analysis work, therefore, has been the question of the role of the victim (Canter, 1994; Salfati, Horning, Sorochinski, & Labuschagne, 2015; Salfati, James, & Ferguson, 2008; Sorochinski & Salfati, 2010).

Proulx (2007), in his comprehensive report on the current understanding of sexual homicide, also highlights the importance of not only focusing on what offenders do during their crimes but also on how these offenders differ depending on the victim group they target. Furthermore, Chan, Beauregard, and Myers (2015), in their study of Canadian serial and single sexual homicides, found that serial offenders are significantly more likely to have specific victim-type preferences. Beauregard (2010), in a discussion paper on the contributions of sex offender research to offender profiling, also discusses crime-switching patterns and how these patterns are victim focused, but highlights that the research in the clinical arena has suffered from focusing specifically on one crime (victim) type at a time, thus not allowing for a more in-depth and broader analysis of these issues.

Offenders Who Target Sex Workers

One particular focus in the work on serial sexual offenses has been that of vulnerable victims and subgroups of highly targeted women in serial homicides such as sex workers (e.g., Abrams, Palmer, & Salfati, 2016; Salfati, 2007, 2013; Salfati et al., 2008; Salfati & Sorochinski, 2018). This body of work provides a further expansion of the discussion on the factor of victim targeting as a focus in profiling the crime scene and the potential type of offender involved.

Egger (1984), in one of the earliest writings on serial homicide, wrote that the key challenge for investigators is what he terms “linkage blindness.” Linkage blindness is a situation whereby crime scenes that are part of the same series are not identified as such due to overt factors (such as type of victim) looking dissimilar, and as such conclusions are made that they are unlinked. In the case of offenders who target different types of victims, and victimology being a key salient feature of linkage analysis, this poses a very real problem. It is, therefore, of utmost importance that the issues underlying victim selection are more fully understood. Egger (2003) goes on to point out that 65% of serial homicide victims are female, and nearly 78% of female victims of serial homicide offenders are sex workers.

A number of other studies have also shown that sex workers are a highly targeted group of victims in serial homicide (Brewer et al., 2006; Brooks-Gordon, 2006; Egger, 2003; Kinnell, 2006; Quinet, 2011; Salfati, 2009; Salfati et al., 2008), with some work suggesting that 32% of serial homicides involve sex workers (Quinet, 2011), and that in the United States when sex worker victims are killed, in at least 35% of cases, a serial offender will be involved (Brewer et al., 2006) and that between 1990 and 2016 in the United Kingdom, 15% of all solved cases of sex worker homicides were by serial offenders (Cunningham, Sanders, Platt, Grenfell and Macioti (2018).

In addition, despite the overall general decline in serial homicide cases between the 1980s and 2000s, research indicates that serial homicide cases involving sex worker victims have dramatically increased by decade (Quinet, 2011). Quinet also found that the average length of time that offenders were active was longer for cases involving sex worker victims. This undoubtedly relates to the fact that sex worker homicides are some of the most difficult to solve, which is supported by some statistics that show that 10 years on, 63% of sex worker homicides are still unsolved (Kinnell, 2006). All taken together, it highlights sex workers as a key group to focus on in the study of victimology in serial homicide. In addition, due to their occupation in the sex industry, the sexual component of the crime is implicit.

Crossover in Victim Types Among Serial Offenders Who Target Sex Workers

Recent studies, specifically on sex worker homicides, have shown that serial offenders who target sex workers also target non–sex workers. Salfati (2007, 2013) reported that 42% of series examined were composed of sex worker victims only, and 58% were mixed-victim series, that is, the series included both sex worker victims and non–sex worker victims. Abrams, Palmer, and Salfati (2016) reported that in their data set of homicides involving sex workers, 43% were apparent one-off offenses, 33% were sex worker–only series, and 24% were mixed-victim series. In addition, they found that mixed-victim series were longer and had more victims when compared with sex worker series. Salfati and Sorochinski (2018) reported that 45% of the series they analyzed included only sex worker victims, 42% included a mix of sex workers and non–sex workers, and another 13% included not only at least one sex worker victim but also at least one victim whose occupation was not known (i.e., they could have fallen into either of the two categories). Cunningham et al. (2018), in their recent study, also pointed out that “some of the serial offenders in the United Kingdom were known only to have targeted sex workers and others had both sex working and nonsex working victims” (p. 330). These initial results all go against much of the belief in the field that offenders are behaviorally consistent and display victim preferences, especially in cases where sex workers are targeted.

The knowledge that offenders do not display overt consistency causes a practical issue in terms of identifying and linking crime scenes to each other. Thus, the two key questions that need to be looked at, and that form the focus of the current article, are as follows:

Research Question 1: What are the between-series differences in victimology and crime scene action between sex worker series and mixed-victim series?

Research Question 2: What are the within-series similarities of victimology and crime scene actions in the mixed-victim series?

Interplay Between Victim Type and Crime Scene Actions

A small body of research over the last decade has provided actual baseline numbers on the fact that offenders do separate into those who target only sex worker victims and those who target both sex workers and non–sex workers in their series, highlighting the need for further research of the overt elements of the crime that may further elucidate victim crossover patterns and what factors may link different types of victims (sex workers and non–sex workers) to each other. These studies have not only started to explore some of the key factors that may help us understand what behavioral aspects of the crime link to this differentiation but have also highlighted that it is important to go deeper to fully understand the issues at hand and make them applicable to practice.

Crime Scene Variable Selection

Target and Location Selection

Salfati (2009) highlighted that although some of the precipitating circumstances to sex worker violence and homicide have been outlined in the overall literature, as well as some details regarding the perpetrator, there is as yet not a clear understanding about what the assaults consist of, or how this compares with other types of aggressions against women, and stated that one of the biggest questions is whether violence against sex workers is one of opportunity and availability because they represent specific targets to the offender. She goes on to say that to understand violence, and especially homicide, against sex workers, it is, therefore, important to understand the context of violence and sexual violence. Indeed, there may be an assumption that the murder of a sex worker is a sexually motivated offense simply because of the nature of the victim’s profession, but this may not be the case. As in the case of rape, it may not be simply a matter of sex, but rather an exertion of power.

The location, where the crime occurs and where the body of a victim is found, has been one of the key differentiating and useful factors identified in studies focusing on linking and crime classification as part of the set of planning behaviors offenders engage in (e.g., Beauregard & Field, 2008; Sorochinski & Salfati, 2010).

Crime Scene Behaviors (Sexual and Wounding)

Salfati (2007, 2013), in a U.S. sample, estimated that 43% of sex workers and 44% of non–sex workers were sexually assaulted. When the sexual element was looked at to determine whether it provided the element of consistency that ties the different types of victims together, the analysis further showed that of the 19 series examined, two series included sex worker victims who were all sexually assaulted, five series included sex worker victims who were not sexually assaulted, and 11 series included mixed victim types (sex workers and non–sex workers) and a mix of sexual assaults and no sexual assaults. The results, thus, indicate that overt sexual assault is not a simple distinguishing factor between victim types.

Salfati and Sorochinski (2018) found a general trend of non–sex worker victims being significantly more likely to have been left alive by the offender and were often sexually assaulted. Furthermore, of the non–sex worker victims who survived the attack, 81.1% were sexually assaulted, and of those who were killed, 32.4% were known to have been sexually assaulted. In terms of looking at the interplay between the different components of the crime, they hypothesized the consistency of actions for these offenders may be victim-specific. The meaning of the overt sexual activity, therefore, becomes an intricate one, with some victims perhaps representing sex, while others became sexual victims through the sexual assault itself.

Furthermore, wounding and weapon selection have also been highlighted as key differentiating factors in sexual versus nonsexual homicides (e.g., Beauregard & Martineau, 2014; Beauregard & Proulx, 2002; Chan & Heide, 2008), as well as in studies looking at behavioral consistency and linkage in homicides (e.g., Bateman & Salfati, 2007; Hickey, 2016; Sorochinski & Salfati, 2010). Although comparisons of weapon choice in sexual versus nonsexual homicides find that strangulation is the method of choice in sexually motivated crimes, studies focusing on homicide series also highlight that the type of weapon and the degree/severity of wounding may vary across series. In the context of homicides involving different types of victims (sex workers and non–sex workers), it is, thus, important to understand whether the choice of weapon may be contingent on the victim type.

Postmortem Actions and Body Disposal

What the offenders choose to do with a victim’s body after the homicide constitutes another key aspect of the crime that can help in differentiating and linking homicides (Beauregard & Field, 2008; Beauregard & Martineau, 2014; Sorochinski & Salfati, 2010). Indeed, the key aspect of the offenders’ decision making in terms of evading capture is what they do with the victim’s body after the completion of the offense (Showers & Cantor, 1985). Furthermore, behaviors such as transporting the body away from the original crime scene has important implications for the investigation in terms of delays in finding the body, delays in finding the original crime scene, and potential destruction of important evidence. It is thus important to determine whether the decision to transport the body and/or to otherwise make an effort in disposing of the body is a factor that helps us differentiate between sex worker homicides of different types.

Summary

From the review of the literature that aims to highlight the most salient features of the crime scene to look at, there are some preliminary indications that certain crime scene factors help to determine differentiation and similarity between crime scenes. These include actions that precede the crime itself (victim type and location), actions during the crime (sexual activity, theft, and wounding), and actions that occur after the assault (postmortem wounding, transportation of the body, and body disposal). However, as yet there has been no comprehensive empirical study that has pulled all this together to obtain a detailed full picture of how these interact to understand their role in helping elucidate within-series similarities as well as between-series differences.

Conclusion

Further numbers based on larger data sets on the aetiology of these types of series will allow to consolidate the literature on this type of crime. In addition, understanding between-series differences of victimology as well as crime scene action between sex worker series and mixed-victim series will allow a more empirical basis for differentiating between crime scenes. Finally, identifying the within-series similarities of victimology and crime scene actions, that is, what factors link sex worker victims and non–sex worker victims in the same series, will allow for the alleviation of linkage blindness of outwardly different looking crimes. At the heart of this research is the development of the refinement of the theoretical and psychological issues relating to understanding differentiation and similarity, as well as investigative implications relating to linkage blindness and linking of serial crimes.

Aims

The aims of this study were threefold and together form an important foundation work to establish a baseline understanding of the aetiology of crime scenes that are part of serial homicides of sex workers and to use this as a first step to explore the key salient factors that are paramount for identifying crime scenes that may be part of the same series. Alongside being able to identify actions that can be used to link a series, it is also important to know how to differentiate one series from another series. Key to this process is the question of what aspects of the crime scene are the most reliable for making these determinations.

Previous literature has established that there are series where all victims are sex workers as well as mixed series; the first step is to establish the behavioral characteristics of these two series types broadly. Next, it is necessary to understand what differentiates specifically the sex worker victims who are part of the sex worker–only series and the mixed-victim series as the key step in identifying the scope of the linkage efforts in an investigation. Finally, it is necessary to determine the similarities between the sex worker and non–sex worker victims in the mixed-victim series so as to understand whether the offender behaviors remain similar despite victim-type inconsistency. Specifically, the study aims were as follows.

Aim 1

The first aim was to outline differences in the demographics of crime scene actions present in homicide series with exclusively sex worker victims (sex worker series) and series that include both sex worker and non–sex worker victims (mixed-victim series). Specifically, the focus was on variables that span from start to finish of the crime and that have been shown to be the most salient, including type of victim targeted, location of the crime scene, presence of sexual assault, type of wounding used, whether the victim was left alive or killed, and whether the offender transported or otherwise made an effort to dispose of the victim’s body.

Aim 2

The second aim was to conduct a between-group analysis to determine whether sex worker victims in sex worker–only series and sex worker victims in mixed-victim series can be differentiated based on crime scene behaviors.

Aim 3

The third aim was to conduct a within-group analysis to determine whether the crime scenes within the mixed-victim series are similar in terms of the presence of the same factors as such similarity would provide the first step to ascertaining overall consistency in these series as the basis for linking.

Method

Data Collection

Data for this project were collected from open media sources that include local, national, and international newspapers, broadcasts, and magazines, as well as books and documentaries. Data collection took place in several stages, with first initial extensive searches through news sources databases, including LexisNexus, National Newspapers Index (ProQuest), Ethnic Newswatch, General OneFile, and others. Search terms used were different combinations of “sex worker,” “prostitute,” “call girl,” “escort,” and so forth + “homicide,” “killed,” “murder,” and so forth. After this initial search, each victim and/or offender names were searched extensively to find all available information pertaining to the case. To maximize the reliability of the information acquired on each case, a required minimum of three independent news sources was set as standard (i.e., there had to be articles coming from at least three different newspaper titles for a case to be considered complete for inclusion). Overall, the median number of sources per case was 7, and the median number of articles from across the sources per case was 15.

Information for cases was collected and coded using the Homicide Involving Prostitutes (HIP©; Salfati & Sorochinski, 2016) coding dictionary specifically developed for the purpose of this project. The dictionary contains 66 variables pertaining to victim and offender demographics, crime type and crime scene characteristics, as well as the outcome of the case. Most variables are coded dichotomously (present/absent). A total of 10 coders went through training on use of open source data and on use of the coding dictionary, followed by several rounds of interrater agreement testing that entailed groups of three to four coders coding the same three cases, and interrater agreements of at least 86% were achieved before independent coding was allowed.

Data

Series

The data set used in this study¹ included a total of 83 series, with 519 victims in total (median per series = 5) from six different countries (62—United States, 21—international, including Canada, United Kingdom, Australia, Austria, Hong Kong). Of those, 44 were series where all victims were sex workers (sex worker series) and 39 were mixed-victim series (see Table 2 for full series information).

Victim

Victims’ age was known for 89.9% of the sample, and the average age was 27.6 years (SD = 8.9). The vast majority (n = 496, 95.2%) were female, 21 victims were male, and two were transgender. Race/ethnicity was known for 349 (67%) victims. Of those, majority (61%) were White, 27% were Black, and 5% were Hispanic. Overall, 396 of them were sex workers and 91 were not (32 unknown; see Table 1 for full victim demographics).

Table 1.

Demographics of Victims and Offenders.

Victim
Age (known for 467 out of 519, 89.9%)	27.6 years (SD = 8.9)
Gender
Female	n = 496 (95.2%)
Male	n = 21 (4.0%)
Transgender	n = 2 (0.4%)
Race (known for 349 of 519, 67%)
White	n = 213 (61%)
Black	n = 94 (27%)
Hispanic	n = 19 (5%)
Other (Asian, Native, Middle Eastern)	n = 23 (7%)
Offender
Gender	N = 83 (100%)
Age at first offense (known for 80 of 83)	32 years (SD = 9.5, median = 30 years)
Race (known for 74 out of 83)
White	n = 45 (60.8%)
Black	n = 18 (24.3%)
Asian (including East Asian and Indian/Pakistani)	n = 8 (8.2%)
Hispanic	n = 3 (4.0%)

Offenders

All the offenders who committed these series were male. For the 80 (out of 83) offenders for whom this information was available, the average age at first offense was 32 years (SD = 9.5, median = 30). Race/ethnicity was known for 74 of the offenders, and of those, 60.8% were White, 24.3% were Black, 8.2% were of Asian origin (including East Asian and Indian/Pakistani), and 4% were Hispanic (see Table 1 for full offender demographics).

Results

Aim 1: Comparison of Sex Worker Series and Mixed-Victim Series

The first aim of this article was to outline the differences in the demographics of crime scene actions present overall in homicide series with exclusively sex worker victims (sex worker series) and series that includes both sex worker and non–sex worker victims (mixed-victim series). Table 2 summarizes the series composition and crime scene characteristics by type of series (sex worker series vs. mixed-victim series).

Table 2.

Series Composition and Crime Scene Characteristics by Overall Sample and Type of Series (Sex Worker Series vs. Mixed-Victim Series).

Variable	Total series(N = 83, 100%)	Sex worker series(n = 44, 53%)	Mixed-victim series(n = 39, 47%)	Significant difference^a
Total number of victims Median per series^b	519 Median = 5	183 Median = 3	336 Median = 7	**
Sex worker victim	396 (76%)	183 (100%)	213 (63%)
Alive	88 (17%)	25 (14%)	63 (19%)
Sexual assault	156 (30%)	44 (24%)	112 (33%)	*
Weapon brought	143 (49%)	45 (48%)	98 (49%)
Strangulation	223 (47%)	71 (41%)	152 (51%)	†
Stabbing	26 (6%)	8 (5%)	18 (7%)
Gunshot	57 (12%)	17 (10%)	40 (13%)
Blunt instrument	23 (5%)	17 (10%)	3 (1%)
Location found (outside)	305 (77%)	94 (67%)	211 (82%)	**
Body transported	222 (56%)	81 (59%)	141 (54%)
Disposal effort	154 (30%)	63 (42%)	91 (31%)	†

Note. Percentages shown are valid (excluding missing values). All percentages rounded up/down to the nearest whole number.

Based on chi-square test.

Median is presented as an accurate and representative measure of central tendency due to significant outliers present in the data.

†

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

The presence of sexual assault was more common in the mixed-victim series compared with sex worker series (50.2% vs. 34.6%; χ² = 7.95, p = .005). Victims in the mixed-victim series were more often strangled (50.5% vs. 40.8%; χ² = 4.16, p = .041) and more likely to be found outside (82% vs. 67%; χ² = 16.77, p < .0001). Conversely, offenders were more likely to exert an effort to dispose of the victims (e.g., cover, hide, or otherwise delay the finding of the body) in the sex worker series compared with the mixed-victim series (42.3% vs. 31.2%; χ² = 8.05, p = .018). No significant differences were identified between mixed-victim and sex worker series in terms of other types of wounding besides strangulation, the presence of live victims, whether the victim’s body was transported, and whether the offender brought a weapon to the crime scene. Importantly, the number of victims per series differed significantly between the two types of series, with mixed-victim series having more than twice as many victims as in the sex worker series (median = 7 vs. 3, respectively; p < .0001). Although the mixed-victim series included victims of different types, almost two thirds (63.4%) of the victims in this type of series are, nonetheless, sex workers. From a practical perspective, this further underlines the need to understand how crime scenes with sex worker victims differ from each other between these two types of series. This will aid in appropriately focusing investigative efforts on identifying all victims who are part of the same series (i.e., linkage). That is, once a sex worker victim is found, it is crucial to determine whether the linkage process should be limited to other sex worker victims or expanded to other types of victims.

Aim 2: Between-Group Analysis: Differentiating Crime Scene Behaviors Between Sex Worker Victims in Sex Worker–Only Series and Sex Worker Victims in Mixed-Victim Series

The second aim of the study was to determine whether specific differences in crime scene behaviors could be used to determine whether a sex worker homicide is likely to be part of a sex worker series or a mixed-victim series. To do so, a classification and regression tree (CRT) analysis was conducted (see Figure 1). CRT is a type of classification tree analysis that “splits” the data using the dependent variable until the sets (sex worker victims only or mixed victims) are as homogeneous (where as many cases as possible fit into one group) as possible. Variables are prioritized based on how significant their unique contribution is to the distinction of types in the overall model. The particular advantages of CRT for the purposes of the present study are that it is a nonparametric test and that it conducts an exhaustive search of interactions between variables to identify complex relationships between variables that would otherwise be missed and maximize prediction accuracy. The test is often used, and is particularly appropriate, for exploratory predictive analytics (Breiman, 2017) due to its hierarchical approach to variable prioritization as well as its ability to construct a decision tree to select from a large number of explanatory variables that are very critical in determining the response variable. Furthermore, CRT, unlike other classification methods, does not treat missing values as a separate classification category, which would not be helpful for classification purposes.

Figure 1.

CRT analysis differentiating sex worker victims between series type.

Dependent variable

The dependent variable was “mixed-victim series,” which was coded as 1 (present, n = 213) for each sex worker victim who was part of a mixed-victim series and 0 (absent, n = 183) if the victim was part of a sex worker series.

Independent variables

The independent variables were selected for inclusion in this analysis based on the theoretical (previous literature) and practical (investigative) significance in crime scene analysis. These variables broadly relate to three categories that span behaviors relating to behaviors that occurred before, during, and after the crime: planning/cover-up (crime scene location—inside/outside, bringing weapon, body transported, disposal effort), wounding type (strangulation, blunt force, gunshot), and victim–offender interaction (victim left alive, victim sexually assaulted).

The predictive validity of the resultant decision tree analysis was also tested using the receiver operating characteristic (ROC) analysis (see Figure 2). An ROC analysis will show the robustness of the model by providing a figure that shows a primary line indicating the 50% chance level of the sample and a line that shows how much the model actually predicts. This second line is referred to as the area under the curve (AUC) and provides the actual estimation of how much the model predicts. The larger this area is above the 50% mark, the more robust the model. The AUC value that depicts the robustness of results ranges from 0 to 1, with 0.5 representing chance-level accuracy and 1.0 representing the highest level of predictive accuracy.

Figure 2.

ROC analysis differentiating sex worker victims between series type.

Figure 1 shows the CRT decision tree. The most important predictor identified by the analysis (i.e., the factor that differentiates the most between sex worker victims from mixed-victim series and those from sex worker series) was whether the victim’s body was found outside. The next node identified by the CRT analysis was strangulation, with 68.2% of sex workers who were strangled being part of mixed series. Finally, a modest addition to the model was made by the sexual assault variable, with 65.9% of cases where it was present belonging to the mixed series.

The AUC for the model was 0.666 (p < .0001), suggesting that the predictive accuracy is significantly better than chance (see Figure 2).

The overall classification accuracy was at 62.9%, but the accuracy of predicting whether the victim was part of a mixed-victim series was substantially higher (91.1%), compared with how accurately these variables can predict that the victim is part of a sex worker series (30.1%). From the practical point of view, this means that, more often than not, when a sex worker victim is found, investigators cannot discount the possibility that the linkage search for other possible victims must include non–sex worker victims and other sex worker victims.

Aim 3: Within-Group Analysis: Ascertaining Similarity Between Sex Worker Victim and Non–Sex Worker Victim Crime Scenes in Mixed-Victim Series

The third aim of this study was to determine whether the crime scenes within the mixed-victim series are similar in terms of the presence of the same crime scene behaviors analyzed in Aim 2. Pearson chi-square was performed to determine whether crime scenes with sex worker victims and non–sex worker victims within the mixed-victim series were similar or different. As shown in Table 3, for most of the crime scene behaviors, sex worker and non–sex worker victim crime scenes differed significantly. Specifically, sex worker victims were significantly more likely to be killed by strangulation with a weapon brought to the scene, transported away from the original crime scene, and found outside. Conversely, non–sex worker victims were more likely to be attacked inside and left alive. The crime scenes of these two types of victims did not differ in whether the offender made any effort to dispose of the body (e.g., hiding, covering, burying) and whether the victim was a sexually assaulted.

Table 3.

Comparing Victim Type Within Mixed Series.

Variable	Sex worker (n = 213)	Non–sex worker (n = 91)	χ²
Victim alive	25 (12%)	35 (39%)	28.75***
Sexual assault	59 (28%)	44 (48%)	2.83
Weapon brought	50 (24%)	40 (44%)	10.61**
Strangulation	115 (54%)	25 (27%)	23.01***
Transported	114 (54%)	15 (17%)	48.63***
Location (outside)	174 (82%)	53 (58%)	23.91***
Effort to dispose	52 (24%)	13 (14%)	3.94

Note. Thirty-two victims were excluded from this comparison because it could not be confirmed whether they were in the sex trade or not.

p < .001. ***p < .0001.

Discussion

The main purpose of the present study was to shed light on the specific challenges in both the research and investigation of serial homicides involving sex worker victims. Victimology has been highlighted as one of the key factors for linking crimes both in the literature and in crime analysis practice, but as recent studies suggest, many offenders are inconsistent in their victim choice (particularly as it relates to focusing on sex worker victims) and thus create a challenge for successful linkage and subsequent solvability of the cases.

As has been summarized in this article, a small body of research over the last decade has provided actual baseline numbers on the fact that offenders do separate into those who target only sex worker victims and those who target both sex workers and non–sex workers in their series, highlighting the need for further research of the overt elements of crime that may further elucidate victim crossover patterns and what factors may link different types of victims (sex workers and non–sex workers) to each other. These studies have also started to explore some of the key factors that may help us understand what behavioral aspects of the crime link to this differentiation and have also highlighted that it is important to go deeper to fully understand the issues at hand and make them applicable to practice.

Preliminary indications that certain crime scene factors help to determine differentiation and similarity between crime scenes were identified. These include actions that precede the crime itself (victim type and location), actions during the crime (sexual activity, wounding), and actions that occur after the assault (transportation of the body and body disposal). Results of the present study begin to shed light on the full picture of how these interact to understand their role in helping identify within-series similarities as well as between-series differences. Specifically, we established the key crime scene features that have the potential for determining the linkage scope for investigators faced with a sex worker homicide and also highlighted some major challenges in this area that warrant further more detailed analysis of offenders’ behavioral patterns across series.

The study’s results highlighted that the location of the victim’s body is one of the strongest differentiators between sex worker victim series and mixed-victim series overall, and also the most salient decision-making factor in determining whether a sex worker victim is part of a mixed-victim series. Specifically, it was found that, overall, victims in the sex worker series were more often found inside, whereas victims in the mixed-victim series were more often found outside. Notably, the sex worker victims in the mixed-victim series are the most likely to be found outside of all victims looked at in this study. Body location, thus, significantly differentiated between sex worker series and mixed-victim series overall and is the most salient factor in differentiating specifically between sex worker victims in these two types of series. Within the mixed-victim series, offenders additionally treated sex worker and non–sex worker victims differently, attacking the sex worker victims outside and the non–sex worker victims inside. With the body location being the first indicator available at the onset of an investigation, this finding may have important implications for investigators. The decision of where to commit the crime and where to end it (i.e., where to leave the victim’s body) is part of planning behaviors, which has previously been found to be among the most consistent features in homicide series (e.g., Salfati et al., 2015; Sorochinski & Salfati, 2010). However, findings in this study also showed that although specific crime scene indicators, such as finding the victim’s body outside, strangulation, and sexual assault, may improve the investigator’s ability to determine the likelihood of non–sex worker victims being part of a series that also includes sex workers, these same indicators are insufficient to confidently determine that a victim is part of a consistent sex worker–only series. These results further underpin the need to determine what additional features help show differences between series, as well as similarities within series, to allow for better discrimination and linking of series.

In the initial comparison between series that consisted of sex worker victim series and mixed-victim series, it was found that mixed-victim series had almost double the number of victims per series, on average. This finding alone supports the need to expand research efforts in this area because it makes it clear that offenders who are less consistent in their victim choice seem to be able to evade capture for longer periods of time and so are able to kill more victims. Furthermore, when looking at the overall offending behaviors within those mixed-victim series, results showed that not only are these offenders inconsistent in their victim selection (sex worker vs. non–sex worker), but also they seem to be largely inconsistent in their behaviors toward these different victims. This finding highlights the importance of further research into specific behavioral patterns across series that would help account for these broad inconsistencies within the mixed-victim series, both in terms of victim selection and in terms of specific crime scene behaviors across crimes. To understand how these series may be effectively linked, it is, therefore, necessary to look more closely at the progression of both victimization patterns (e.g., at what point in the series do the offenders switch victims, do they go back and forth in the victim selection, or do they start off with one type and then move on to consistently victimizing another type of victim) and behavioral patterns (e.g., do offenders escalate or de-escalate in their crime type, going from assaulting the victim to killing the next or vice versa). Understanding these patterns will further our understanding of these complex crimes and likely bring us closer to devising behavioral models that would be effective for investigative linkage of these series (see Salfati & Sorochinski, 2019, for further discussion).

The present study is not without its limitations. First and foremost, the data were collected from media sources that are not originally intended for research and that may be missing pertinent details of the cases. Although every effort was made to ensure completeness of information and to limit the analysis to variables that were robust, missing information may affect the detailed accuracy of the analysis.

Conclusion

Although it has long been identified that sex workers are the single most targeted group of women in serial homicide (Egger, 2003; Hickey, 2016), the present study is among the first to specifically focus on understanding the characteristics that can help distinguish between different types of series that include sex worker victims and on how the victim selection pattern relates to the offender behaviors at the crime scene. The findings reported here highlight several important issues related to serial homicides that involve sex workers, and as such also more broadly violent serial offenses, and purport the need for (a) closer attention on the part of investigators to broadening their linkage efforts in serial homicide cases involving sex workers and (b) further research into understanding the different types of offenders who remain consistent versus inconsistent in their victim selection and the different behavioral patterns that these offenders exhibit across their series.

Footnotes

Acknowledgements

The authors would like to sincerely thank the many research assistants involved in this project for their help with data collection. Special mention goes to the following individuals for their extensive input on data preparation: Sneha Gupta, Rosanne Libretti, Marissa M. Abrams, Marcel Trujilo.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for this project was provided by a PSC-CUNY Award, jointly funded by the Professional Staff Congress and the City University of New York.

ORCID iD

C. Gabrielle Salfati

Notes

References

Abrams

M. M.

Palmer

Salfati

C. G.

(2016, November). Patterns of sex worker and non-sex worker victim selection: Relationship to offender characteristics. Poster presented at The American Society of Criminology, New Orleans, LA.

Bateman

A. L.

Salfati

C. G.

(2007). An examination of behavioral consistency using individual behaviors or groups of behaviors in serial homicide. Behavioral Sciences & the Law, 25, 527-544.

Beauregard

(2010). Rape and sexual assault in investigative psychology: The Contribution of sex offenders’ research to offender profiling. Journal of Investigative Psychology and Offender Profiling, 7, 1-13.

Beauregard

Field

(2008). Body disposal patterns of sexual murderers: Implications for offender profiling. Journal of Police and Criminal Psychology, 23(2), Article 81. doi:10.1007/s11896-008-9027-6

Beauregard

Martineau

(2014). No body, no crime? The role of forensic awareness in avoiding police detection in cases of sexual homicide. Journal of Criminal Justice, 42, 213-220.

Beauregard

Proulx

(2002). Profiles in the offending process of nonserial sexual murderers. International Journal of Offender Therapy and Comparative Criminology, 46, 386-399.

Breiman

(2017). Classification and regression trees. New York, NY: Routledge.

Brewer

D. D.

Dudek

J. A.

Potterat

J. J.

Muth

S. Q.

Roberts

J. M.

Woodhouse

D. E.

(2006). Extent, trends, and perpetrators of prostitution-related homicide in the United States. Journal of Forensic Science, 51, 1101-1108.

Brooks-Gordon

(2006). The price of sex: Prostitution, policy and society. Cullumpton, UK: Willan Publishing.

10.

Canter

D. V.

(1994). Criminal shadows. London, England: HarperCollins.

11.

Canter

D. V.

(2000). Offender profiling and criminal differentiation. Legal and Criminological Psychology, 5, 23-46.

12.

Chan

H. C.

Beauregard

Myers

W. C.

(2015). Single-victim and serial sexual homicide offenders: Differences in crime, paraphilias and personality traits. Criminal Behaviour & Mental Health, 25, 66-78.

13.

Chan

H. C.

Heide

K. M.

(2008). Weapons used by juveniles and adult offenders in sexual homicides: An empirical analysis of 29 years of US data. Journal of Investigative Psychology and Offender Profiling, 5, 189-208.

14.

Cunningham

Sanders

Platt

Grenfell

Macioti

P. G.

(2018). Sex work and occupational homicide: Analysis of a U.K. murder database. Homicide Studies, 22, 321-338. doi:10.1177/1088767918754306

15.

Egger

S. A.

(1984). Working definition of serial murder and the reduction of linkage blindness. Journal of Police Science and Administration, 12, 348-357.

16.

Egger

S. A.

(2003). The need to kill. Upper Saddle River, NJ: Prentice Hall.

17.

Hickey

E. W.

(2016). Serial murderers and their victims (7th ed.). Boston, MA: Cengage Learning.

18.

Kinnell

(2006). Murder made easy: The final solution to prostitution? In Campbell

O’Neill

(Eds.), Sex work now (pp. 116-140). Cullompton, UK: Willan Publishing.

19.

Proulx

(2007). Sexual murderers: Theories, assessment and treatment. In Sexual homicide and paraphilias: The correctional service of Canada’s experts forum (pp. 215-233). Ottawa, Ontario: Canadian Government Publishing. Retrieved from http://www.csc-scc.gc.ca/research/shp2007-paraphil12-eng.shtml

20.

Quinet

(2011). Prostitutes as victims of serial homicide: Trends and case characteristics from 1970-2009. Homicide Studies, 15, 74-100.

21.

Salfati

C. G.

(2007, November). Prostitutes: Vulnerable or sexual targets of serial homicide? Presented at the Annual American Society of Criminology Conference, Atlanta, GA.

22.

Salfati

C. G.

(2008). Profiling. In Cutler

B. L.

(Ed.), Encyclopedia of psychology and law (pp. 575-580). Thousand Oaks, CA: Sage.

23.

Salfati

C. G.

(2009). Prostitute homicide: An overview of the literature and comparison to sexual and non-sexual female victim homicide. In Canter

Ioannou

Youngs

(Eds.), The Psychology, Crime and Law Series. Safer sex in the city: The experience and management of street prostitution (pp. 51-68). Aldershot, UK: Ashgate.

24.

Salfati

C. G.

(2013). Linkage analysis of serial murder cases involving prostitute victims. In Helfgott

(Ed.), Criminal psychology reference set (Vol. 3, pp. 211-228). Westport, CT: Praeger.

25.

Salfati

C. G.

Horning

Sorochinski

Labuschagne

(2015). South African serial homicide: Consistency in victim types and crime scene actions across series. Journal of Investigative Psychology and Offender Profiling, 12, 83-106.

26.

Salfati

C. G.

James

A. R.

Ferguson

(2008). Prostitute homicides: A descriptive study. Journal of Interpersonal Violence, 23, 505-543.

27.

Salfati

C. G.

Sorochinski

(2016). Homicides Involving Prostitutes (HIP©) coding dictionary. In Investigative Psychology Research Unit (IPRU). Retrieved from https://www.jjay.cuny.edu/ipru/research

28.

Salfati

C. G.

Sorochinski

(2018). Sex worker victims: Consistency vs. inconsistency in victimization patterns by serial sexual homicide offenders. In Proulx

Carter

Beauregard

É.

Mokros

Darjee

James

(Eds.), Routledge international handbook of sexual homicide. Routledge. doi:10.4324/9781315212289

29.

Salfati

C. G.

Sorochinski

(2019). MATCH: A New Approach for Differentiating & Linking Series of Sex Worker Homicides and Sexual Assaults. International Journal of Offender Therapy and Comparative Criminology, 63, 1794-1824. doi:10.1177/0306624X19839279

30.

Showers

Cantor

(1985). Social cognition: A look at motivated strategies. Annual review of psychology, 36, 275-305.

31.

Sorochinski

Salfati

C. G.

(2010). The consistency of inconsistency in serial homicide: Patterns of behavioural change across series. Journal of Investigative Psychology and Offender Profiling, 72, 109-136. doi:10.1002/jip.118