Whiter Than White: The Art of Delaying Detection in Sexual Killers

Abstract

The study investigated whether different types of sexual homicide perpetrators are more or less skilled at delaying detection. A newly proposed direct/indirect typology was used alongside information about the time of arrest, the frequency of specific precautions as well as the impact of forensic strategies used by the perpetrators to examine skill at delaying detection. The results indicated that the time from the killing to the arrest, as measured in days, was longer for the direct than the indirect sexual killers. Despite the fact that the direct aggressors were better at delaying detection, overall the indirect and the direct offenders did not differ in the frequency of use of most of the precautions. However, different forensic awareness strategies were more efficacious for the direct and the indirect offenders. These results are discussed in relation to the crime scripts for the two perpetrator groups.

Keywords

sexual homicide forensic awareness police detection avoiding arrest typology

Forensic evidence refers to information gathered by various forensic science disciplines employing a scientific method that can serve as evidence in a criminal case, for example, toxicology, pathology, botany or DNA analysis. Such evidence has either class or individual characteristics. Class material shares properties with a group of items and, therefore, it narrows down the evidence to a group of people or things such as blood type of the perpetrator or a model of a firearm. Individual material shares properties with a single source and, therefore, it narrows down the evidence to an individual person or a thing (Kiely, 2001).

From the outset of the investigation, the aim for the investigative team is to make sense of the homicide incident. Drawing inferences from the material available, the goal is to answer the questions as to what had happened and why. ‘Seeking after the truth’ requires the investigators to explore various lines of inquiry whilst taking into account choices made by the perpetrators at various stages of the criminal event (Bryant, 2009). A rational-choice approach postulates that criminal behaviour is not fundamentally different from noncriminal behaviour and as such decisions made by the perpetrator are the result of a weighing process of costs versus perceived benefits (Cornish & Clarke, 1986). These decisions are taken at three phases of the criminal event: precrime, crime and postcrime. In addition to the perpetrator’s intentions and feelings prior to the sequence of events also depend on the situational components found during the commissioning of the criminal act (Beauregard & Leclerc, 2007).

Despite Burrows et al.’s observation that ‘it is clear that there is no single, universally applied model of investigation’ (2005, p. 24), Bryant (2009) postulated that the development of investigative theory has been influenced by two interrelated approaches: that is, research on what constitutes successful detective skills and ‘borrowing’ ideas from other disciplines and professions, especially those with a theoretical basis in the physical and biological sciences such as medicine. In an effort to provide a template of best practice, the Murder Investigation Manual (National Centre for Policing Excellence [NCPE], 2006) was developed. The investigative theory outlined in the manual is described under seven main headings: introduction, crime scene assessment/process, offender profiling, behaviour patterns, geographical profiling, synthesis/analysis and lines of enquiry. With regard to crime scene assessment and the evidence found at the scene, the manual suggests to organise it by three core elements of a homicide: the location, the victim and the perpetrator. Features associated with the location relate to various places of the homicide event such as the encounter between the victim and the offender, the location of the offence or the body deposition site. Features associated with the victim relate to victim activities and victim characteristics. Finally, offender features relate to the actions and the amount of evidence the perpetrator left or indeed took away from the crime scene to minimise the chances of being caught. These include but are not limited to the degree of crime scene arrangement or rearrangement; moving, arranging or hiding the body; steps taken to avoid themselves or the victim being recognised; ways of taking control over the victim; and ways of acting on the location such as choosing the scene to minimise chances of being disturbed (NCPE, 2006).

Despite their importance, not all types of evidence are of equal value (helpful) in terms of crime solvability. A body of research examining homicide clearance rates points to a range of characteristics that tend to make the case easier or harder to solve by the investigative team. For example, cases involving noncontact weapons (frequently firearms but also poison or drugs) have consistently been reported to be more difficult to solve in comparison with contact weapons such as knife, hands or a blunt object, possibly due to the potential presence of more physical evidence (Litwin, 2004; Litwin & Xu, 2007; Mouzos & Muller, 2001). Stranger homicides decrease the likelihood of being cleared due to a lower potential for identifying a suspect (Trussler, 2010), while in known victim killings, the relatives or close friends are able to provide more information (Lee, 2005; Riedel & Rinehart, 1996). Homicides committed at home or in public areas tend to be positively associated with clearances as opposed to those carried out in alleys or abandoned buildings. This is because the former are regarded as high or medium visibility areas with more witness presence (Litwin & Xu, 2007). In addition, homicides committed at home tend to involve domestic-related incidents (Beauregard & Martineau, 2017). The clearing rate of the homicide when the offender was under the influence of substances tends to be higher as under such circumstances perpetrators are possibly more apt to making mistakes during the commissioning of the crime. This includes choosing a location in view of more potential witnesses (Roberts, 2007) or leaving more physical evidence as a result of little/poor employment of forensic awareness strategies (Beauregard & Bouchard, 2010).

The impact of the use of forensic awareness strategies by the perpetrator (i.e., steps taken in a criminal event to hide evidence to avoid apprehension; Davies, 1992) on crime solvability have been investigated by Salfati and Haratsis (2001). Significant association was found between offenders who did or did not use various forensic awareness strategies and the solvability of homicides (with former being more prevalent in the unsolved cases and the latter in the solved cases). The use of arson to destroy evidence was an exception as it did not relate to the solved/unsolved case status. Beauregard and Martineau (2014) examined whether using forensic awareness strategies enabled offenders to either avoid detection or to delay apprehension (as measured by the number of days until body discovery). Surprisingly (and contrary to their working hypotheses), Beauregard and Martineau found that if the perpetrator used at least one precaution to avoid detection, the case was more likely to be solved. Moreover, precautions used by the perpetrator did not delay the discovery of the victim, and thus it did not delay the perpetrator’s apprehension. However, concealing the victim’s body did in fact delay the time of body recovery and as a consequence it delayed the perpetrator’s apprehension. The number of posthomicide precautions that was used by the perpetrator also shortened the investigation time in the study by James and Beauregard (2018).

Using one of the first and most well-known examples of a classification system of sexual murder, the organised or disorganised classification (Ressler, Burgess, & Douglas, 1988), Beauregard and Martineau (2017) expanded on the current understanding of different types of sexual murderers by examining if the behaviours associated with the organised offender were related to ability to delay or avoid police detection. Avoidance was measured by solved/unsolved case status whereas delay by number of days until body recovery. The results suggested that some behaviours have a similar effect on both delaying and avoiding detection. For example, selection of a less risky location is likely to both delay and avoid detection whereas use of a weapon and anal intercourse increased the possibility of being detected. On the contrary, some behaviours whilst having a positive effect on delaying detection had a negative effect on avoiding detection and vice versa. Overall, of a total of 12 behaviours were classed as organised: (a) offender targeted the victim, (b) offender used a con approach, (c) offender used restraints, (d) offender beat the victim, (e) offender had vaginal intercourse with victim, (f) offender had anal intercourse with victim, (g) offender used a weapon, (h) evidence of overkill, (i) offender took items from victim, (j) offender moved victim’s body, (k) offender concealed victim’s body and (l) offender selected less risky location. Out of these 12 behaviours, two (offender moved victim’s body, offender selected less risky location) were positively associated with the ability to avoid and four (offender had vaginal intercourse with victim, evidence of overkill, offender concealed the victim’s body, offender selected less risky location) with the ability to delay detection. The authors theorised that this could be because organised behaviours identified by the Federal Bureau of Investigation (FBI) are not in fact associated with police avoidance. However, despite somewhat inconsistent results the overarching aim of the study was original in its design as for the first time different types of sexual homicide perpetrators were scrutinised in relation to police avoidance.

The current research aims to combine the two approaches described above and examine (a) the use of specific forensic awareness strategies, in relation to (b) different types of sexual homicide perpetrators. We believe that accounting for the types of sexual killers is important as time and time again the research suggests that sexual homicide comprises of a diverse group of perpetrators (e.g., Stefanska, Carter, Higgs, Bishopp, & Beech, 2015). The offences should, therefore, be understood as a hybrid between sexual assault and homicide (Beauregard & Martineau, 2017) with distinguishably different scenarios where some perpetrators kill for sexual reasons, while others kill for more pragmatic reasons (Stefanska, Higgs, Carter, & Beech, 2017). As motivational aspects in sexual homicides are difficult to establish, it has been suggested that different types of perpetrators ought to be identified based on their actions at the crime scene rather than the motivational component (Beauregard & Martineau, 2017). A new classification system proposed by Stefanska et al. (2017) allows examining the behaviours within the criminal event to classify perpetrators depending on the relationship between the sexual element and the act of killing within the criminal event of the homicide. As such, Stefanska et al. (2017) proposed the direct/indirect typology. In the direct type, the sexual element and the act of killing are clearly integral, and thus the killing at some point (either pre- or post-) provided a source of sexual stimulation. In the indirect type, although the offence occurs in the sexual context, the killing was not a source for sexual arousal and the sexual aspect within the sexual assault and homicide hybrid clearly happened before the killing took place. The study also aims to improve on the measurement of the perpetrators ability to delay detection by accounting for the days it took from the killing to arrest, rather than killing to finding the victim’s body.

Using the proposed direct/indirect typology (Stefanska et al., 2017), the study examined whether direct and indirect sexual killers differ in their ability to delay detection and what specific precautions they use during the criminal event.

Method

Sample

The sample comprised of 350 perpetrators who served a custodial sentence for murder or manslaughter within HM Prison Service in England and Wales of female victims aged 14 years or above. All of the offenders were male, and the sample included only nonserial cases, that is, one or two victims without an emotional cool-off period (i.e., two victims killed at the same time or within a period indicative of a single event) were killed (Proulx, Cusson, & Beauregard, 2007). The criteria for sexual homicide included offences where a sexual element and/or a sexual motivation for the murder was evidenced, suspected or admitted, which was in line with the U.K. National Offender Management Service OASys research database. However, out of this sample, cases where the perpetrator turned himself in almost immediately after the murder (n = 49) as well as outlier cases, that is, cold cases where the perpetrators were found more than 20 years later due to advancements in DNA testing (n = 10) were excluded. This left the final sample comprising of 291 cases.

Procedure and Measures

Information was gathered from the electronic or physical files of the offenders serving a life sentence for murder or manslaughter (containing the details of the crime event) in the Public Protection Unit Database. To establish interrater reliability, 10% of the same cases were blind-coded by two raters, and the strength of agreement was excellent (Fleiss criteria, 1981; Cohen’s kappa = .91).

Delaying detection was measured by the time (counted in days) it took from the killing to arrest. In addition, the time from the killing to recovery of the body and the time from recovery of the body to the arrest were also examined. Eight forensic awareness strategies used by the perpetrator to delay detection were examined in the current study: (a) protecting own identity—included behaviours such as wearing a mask, giving a false name, disguising or altering appearance or changing residence after the crime; (b) acting on victim—included tying, gagging or blindfolding the victim or administering drugs; (c) acting on the environment—disabling lights or telephones, any security systems, obscuring access of the doors or windows; (d) removing evidence—included cleaning, disposing of the weapon used or the perpetrators clothes worn during the commissioning of the crime; (e) setting fire to the scene; (f) disposing of the victim’s body; (g) giving false statements; and (h) arranging an alibi. The strategies were based on the forensic awareness strategies used by Beauregard and Martineau (2014). All of the precautions were binary, coded as either absent or present (0 = no, 1 = yes).

As a next step, each case was assigned as belonging to either the direct or the indirect group, and this decision was based on the evidence accepted by the court at trial. The assignment process was based on the definitions provided by Carter and Hollin (2014), and it closely followed the procedure described in Stefanska et al. (2017).

In the indirect cases, three typical scenarios emerged: (a) victim was killed to eliminate the witness; (b) victim was killed trying to escape a sexual assault; (c) there was no sexual violence, but killing occurred in a sexual context (i.e., following consensual sex). Overall, 148 cases were assigned as belonging to this indirect group. In the direct cases, two typical scenarios emerged: (a) the act of killing was itself sexually gratifying or (b) the purpose of killing was to enable sexual acts to be carried out with the victim’s body. Overall, 143 cases in total were assigned as belonging to this direct group.

Analytical Strategy

Data analysis was performed in three stages. First, the analysis explored whether direct and indirect sexual killers differed in their ability to delay detection. Delaying detection was measured by the time (in days) it took from the murder to arrest although the time from the murder to recovery of the body and the time from recovery of the body to the arrest was additionally examined. Mann–Whitney U test was used as the distribution of these variables was highly skewed (with a large proportion of the cases being solved during the first few weeks) violating the normality assumption of the data (Pallant, 2007).

Second, chi-square tests examined whether direct and indirect offenders differed in the precautions they used. Third, negative binomial regressions examined the impact of forensic strategies used by the perpetrators on a number of days until arrest. Negative binomial regression is a technique designed to analyse overdispersed count data. Although it is a generalisation of Poisson regression, it loosens the restrictive assumption that the variance is equal to the mean made by the Poisson model (Beaujean & Morgan, 2016). To ensure that the dispersion parameter was accurate, the estimate value rather than the assumed dispersion parameter was used. In other words, the dispersion parameter was calculated for each of the regression models. Using the same set of variables, a series of negative binomial regressions were employed to separately look at the direct, the indirect and the whole sample of sexual killings (with the results from the whole sample provided for comparison).

Results

Descriptive statistics showed that the time from the killing to the arrest (measured in days) was longer for the direct (M = 163.2, Mdn = 7.5, n = 143) than the indirect sexual killers (M = 129.3, Mdn = 4.4, n = 147). Mann–Whitney U value was found to be statistically significant U = 8,116 (z = −3.44, r = .2), p < .01.

Perhaps not surprisingly, a body was recovered more quickly in the indirect than in the direct killings (M = 128.8, Mdn = .5, n = 148 vs. M = 162.9, Mdn = 1, n = 143, respectively), U = 8,041 (z = −3.49, p < .001, r = .2) and a perpetrator of the indirect rather than the direct killing was arrested faster after the body was recovered (M = 133.8, Mdn = 3, n = 148 vs. M = 158.7, Mdn = 4, n = 143, respectively), U = 8,772.5 (z = −2.53, r = 1.4), p < .05.

When examining specific forensic awareness strategies used by indirect and direct sexual killers, results show that only two methods used by the perpetrators significantly differed between indirect and direct cases (Table 1). Specifically, sexual killers classed as direct were significantly more likely to act upon their victims, that is, tie their victims, gag or cover their eyes or administer drugs (χ² = 4.5, p = .03). On the contrary, sexual killers classed as indirect were significantly more likely to prepare an alibi (χ² = 4.3, p = .04).

Table 1.

Distribution of Subjects in the Direct and Indirect Profile of Forensic Awareness Strategies.

Precautions used by the offender	Direct cases (n = 143) %	Indirect cases (n = 148) %	χ²	Significance
Protecting his identity	11.9	9.5	0.45	ns
Acting upon victim*	28.0	17.6	4.5	.03
Acting upon environment	17.5	14.9	0.37	ns
Removing evidence	40.6	39.2	0.05	ns
Setting fire to the crime scene	9.1	12.8	1.0	ns
Disposing of victim’s body	35.7	37.2	0.37	ns
Giving false statements	30.8	26.4	0.4	ns
Preparing alibi*	7.7	15.5	4.3	.04

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

Table 2 presents the findings of the negative binomial regression analyses with respect to the number of days it took from killing until arrest of the perpetrator for the direct, indirect cases as well as the whole sample (which was provided for comparison). All of the regression models were found to be significant at p < .001. In the indirect sample, two cases were found to have an undue influence on the results (using Cooks Distance). However, given that the exclusion of the influential observations did not alter the results, the cases were retained. Findings indicate that except for acting upon environment and preparing an alibi (which had no effect), all of the predictors increased the incident rate ratio of delaying detection for the sexual killers examined as a whole sample. The differences, however, can be noted between the direct and the indirect groups.

Table 2.

Negative Binomial Regression Analyses for the Number of Days From Killing Until Arrest.

Predictor	All SK (n = 291)	Direct SK (n = 143)	Indirect SK (n = 148)
Predictor	Exp β (SE)	Exp β (SE)	Exp β (SE)
Protecting his identity	3.32 (.15)***	2.98 (.42)**	12.94 (.54)***
Acting upon victim	3.48 (.21)***	4.01 (.27)***	0.69 (.35)
Acting upon environment	0.91 (.24)	0.32 (.38)**	2.60 (.34)**
Removing evidence	1.79 (.20)**	0.91 (.26)	2.10 (.27)**
Setting fire to the crime scene	0.40 (.30)**	0.25 (.43)**	1.13 (.41)
Disposing of victim’s body	1.77 (.32)**	2.09 (.26)**	1.54 (.25)
Giving false statements	2.24 (.21)***	2.16 (.29)**	2.91 (.27)***
Preparing alibi	1.11 (.31)	2.82 (.52)*	1.30 (.34)
Constant	8.83 (.15)***	12.99 (.21)***	4.43 (1.8)***

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

In the direct group, the time of delaying detection was likely to be longer if the perpetrators acted upon their victim (incidence rate ratio [IRR] = 4.01, p < .001), protected their identity (IRR = 2.98, p < .01), prepared an alibi (IRR = 2.82, p < .05), gave false statements (IRR = 2.16, p < .01) and disposed of the victim’s body (IRR = 2.09 times, p < .01). On the contrary, the risk of detection was higher when they acted on the environment (IRR = .32, p < .01) or set fire to the scene (IRR = .25, p < .01). Removing evidence had no effect. In the indirect group, protecting identity (IRR = 12.94, p < .001), giving false statements (IRR = 2.91, p < .001), acting upon the environment (IRR = 2.60, p < .01) and removing evidence from the crime scene (IRR = 2.10, p < .01) increased the time of delaying detection. Acting upon victim, setting fire to the crime scene, disposing of the victim’s body and preparing an alibi had no effect in this group.

Discussion

The study examined the time it took from the killing to the arrest of the offender (excluding the cases where the perpetrator turned himself in immediately after the killing). The analysis revealed that it took longer for the direct than the indirect sexual killers to be arrested after the killing. Both the time it took to recover the body and the time it took to arrest the perpetrator after the body was recovered was faster in the indirect compared with the direct homicides. However, despite the fact that the direct perpetrators were better at delaying detection, overall, the indirect and the direct perpetrators did not differ in the use of many of the precautions they employed. In fact, out of eight forensic awareness strategies noted in the current study, only two differed in the frequency they were used between the groups; direct perpetrators more frequently acted on the victim whereas indirect perpetrators more frequently prepared an alibi.

When considering the use of specific forensic precautions on the ability to delay apprehension for the two types of sexual killers, the results show that the direct perpetrators were likely to delay their apprehension if they acted upon their victim, protected their identity, prepared an alibi, gave false statements and disposed of the victim’s body. The risk of quicker detection was higher when they acted on the environment or set fire to the scene while removing evidence had no effect. On the contrary, the indirect perpetrators were likely to delay their apprehension if they protected their identity, gave false statement, acted upon the environment and removing evidence from the crime scene. Acting upon victim, setting fire to the crime scene, disposing of the victim’s body and preparing an alibi had no effect in this group.

Importance of Understanding the Crime Scripts

The results are better understood in the context of analysis of the crime scripts (particular sequence of actions) for each offender type. For the direct sexual killers, deviant sexual fantasies are a strong motivational force behind the killing, and these, often sadistic, fantasies appear to influence the behaviour of the perpetrator during the criminal event (Stefanska & Higgs, 2018; Stefanska et al., 2017). Importantly, when taking into account the hybrid between sexual assault and homicide, for these offenders the killing is integral as at some point it provided a source for sexual stimulation. This strongly suggests that the act of killing was integrated in the stage of preparation for the crime or at least a prominent feature of the fantasy world (Arrigo & Purcell, 2001; Hickey, 2001; Ressler et al., 1988). Therefore, through rehearsal, it is possible that direct perpetrators simply became more skilled at using various behaviours with the aim of avoiding detection, more specifically disposing of the victim’s body, protecting own identity, preparing an alibi and/or giving false statements. Acting upon victim (i.e., tying, gagging or blindfolding the victim or administering drugs) was the only strategy that not only delayed apprehension, but was also used more frequently by the direct aggressors. This suggests that these perpetrators are either again more skilled at using it or it is a more successful precaution although in the latter case the results would be somewhat inconsistent with the results obtained by Beauregard and Martineau (2014) who found that acting upon victim increased the chances of the case being solved. In the discussion regarding ‘overkill’, Stefanska, Higgs, Carter, and Beech (2018) noted that it is not possible to always define excessive violence as stemming from anger given that in some cases the severity of the attack could also represent more sadistic acts as indicted by Radojevic et al. (2013). Similarly, it is possible that the ‘acting upon victim’ variable has been confounded with a sexual aspect of the case, that is, the behaviour is in fact not measuring forensic awareness but rather forms an intrinsic part of the offender’s sexual fantasy.

Acting upon environment by, for example, disabling lights or telephones or security systems or obscuring access of the doors or windows appeared to be an ineffective forensic strategy that increased the risk of quicker detection in the direct group. The explanation for this could be linked to the possibility that the direct perpetrators spend more time with their victims as they enact their deviant sexual fantasies either pre- or postmortem. Although acting on the environment might be employed by the direct killers to avoid being disturbed (rather than aimed at avoiding detection), it is possible that potential witnesses had noticed something whilst the perpetrator was still at the crime scene or shortly after when the evidence had not been compromised. In other words, acting upon the environment might draw attention to an otherwise well prepared crime. A similar effect can be found in the use of arson as a strategy to dispose of evidence. Understanding the crime script of the two perpetrator types could explain why removing evidence and cleaning the crime scene delays apprehension in the indirect perpetrators but it has no effect in the direct group. We can assume that overall it is an effective precaution (if not to avoid crime then at least to slow down police work) but in the direct perpetrators the risk of leaving physical evidence linking them to the killing is simply greater.

For the indirect sexual killers, the crime scripts picture is somewhat different. Two main scenarios emerge in which the sexual assault is either premeditated or the sexual assault is not planned and in fact, the victim and the perpetrator meet in a consensual situation, for example, a bar or someone else’s home. The killing, however, is usually not planned and occurs due to situational factors, for example, the victim resisted further sexual advances or threatened to report the perpetrator for the sexual assault (Stefanska et al., 2015). Therefore, protecting one’s own identity from the start and acting upon the environment, so that the victim cannot alert anyone, is effective. Interestingly, while giving false statements successfully delays arrest, using an alibi does not, even though this strategy is effective in the direct group and the indirect aggressors use it more frequently. The explanation could very well stem from situational components in the criminal event of the indirect aggressors. Giving false statements, for example, ‘I was with her but I left at such and such an hour’ sufficiently delays arrest, however, giving false alibi has a potential to be incriminating as it often relies on other people’s statements. Although having an alibi per se was not examined, Beauregard and Martineau (2014) found that using other precautions (including an alibi) increased the likelihood of the offender apprehension.

Limitations

As the study relies on a prison population who have all been convicted, it is not possible to investigate the ability to avoid detection (as measured by solved/unsolved case status). It is important to remember that the duration of the criminal investigation is not only explained by the behaviour of the perpetrators but it also relies on the choices made by those involved in the criminal investigation (e.g., investigators, magistrates, forensic experts; James & Beauregard, 2018), something that the current study did not control for. The study did also not account for the relationship between the perpetrator and the victim even though research suggests that stranger homicides decrease the likelihood of being cleared due to a lower potential for identifying a suspect (e.g., Trussler, 2010). There are, however, other crime factors such as the location of the killing or the type of the weapon that was used, which appear to influence whether the case is easier or harder to solve by the investigative team, and controlling for all of these potential characteristics is difficult if not impossible.

Implications

Understanding the relationship between the sexual aspect and the act of killing within the criminal event of a sexual homicide is crucial for appropriate management of these perpetrators through prison or secure hospital (Carter, 2018; Stefanska et al., 2017). This is because sexual homicide consists of, as Beauregard and Martineau (2017) suggest, a hybrid of two components: the sexual assault and the killing. The newly proposed direct/indirect classification takes into account this relationship and distinguishes between the sexual killings based on the proximity of the sexual act to the killing and whether the killing was or was not a source of sexual stimulation for the offender. Importantly, despite the fact that the research has been consistently indicating that sexual killings are diverse, when examining various crime scene behaviours the offenders are often amalgamated into one group thereby losing the explanatory power to account for various scenarios. As the research distinguishing sexual killers based on the proximity of the sexual act to the killing, that is, the direct/indirect typology is in the early stages of research, it is difficult to comment on its implications for policy making or the direct help for investigation purposes. We hope, however, that this research will bring practitioners working in the field a step closer to understanding how the crime scene behavioural indicators can be useful when formulating an understanding about the offenders’ motivation and with it, factors that could be helpful in understanding the risk they may continue to pose after apprehension and conviction.

From the management perspective of sexual killers within the prison or secure hospital settings, it is perhaps not as important to concentrate on what the perpetrator had set out to do (e.g., took a kit with him or attempted to conceal his identity) as, after all, we will not always be able to determine whether the original intent was to rape or to murder the victim. In addition, the behaviour of the perpetrator might change depending on the situational components found during the commissioning of the criminal act (Beauregard & Leclerc, 2007). Some forensic awareness strategies (e.g., acting on victim) might have been confounded with a sexual aspect of the case and, therefore, they are not measuring forensic awareness but form an intrinsic part of the offender’s sexual fantasy. Within this very complex picture, the use of various strategies cannot be understood without detailed examination of the crime scripts. Fortunately, research suggests that based on the available crime scene evidence, we can reliably formulate the proximity of the sexual act to the killing and that in depth understanding of these different crime scripts may assist staff working with perpetrators of these crimes in optimising treatment planning and management (Carter, 2018; Stefanska et al., 2017).

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.

ORCID iD

Ewa B. Stefanska

References

Arrigo

B. A.

Purcell

C. E.

(2001). Explaining paraphilias and lust murder: An integrated model. International Journal of Offender Therapy and Comparative Criminology, 45, 6-31.

Beaujean

A. A.

Morgan

G. B.

(2016). Tutorial on using regression models with count outcomes using R. Practical Assessment, Research & Evaluation, 21(2), 1-16.

Beauregard

Bouchard

(2010). Cleaning up your act: Forensic awareness as a detection avoidance strategy. Journal of Criminal Justice, 38, 1160-1166.

Beauregard

Leclerc

(2007). An application of the rational choice approach to the offending process of sex offenders: A closer look at the decision-making. Sexual Abuse: Journal of Research and Treatment, 19(2), 115-133. doi:10.1177/10790632070190020

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

Martineau

(2017). The sexual murderer: Offender behaviour and implications for practice. Abingdon, UK: Routledge.

Bryant

(2009). Theories of criminal investigation. In Tong

Bryant

Horvarth

(Eds.), Understanding criminal investigation (pp. 13-34). Oxford, UK: Wiley-Blackwell.

Burrows

Hopkins

Hubbard

Robinson

Speed

Tilley

(2005). Understanding the attrition process in volume crime investigations (Home Office Research, Development and Statistics Directorate Home Office Research Study 295). London, England: Her Majesty’s Stationery Office.

Carter

(2018). New directions in assessment and treatment of sexual murderers. In Proulx

Beauregard

Carter

A. J.

Mokros

Darjee

James

(Eds.), Routledge international handbook of sexual homicide studies (pp. 601-604). Abingdon, UK: Routledge.

10.

Carter

A. J.

Hollin

C. R.

(2014). Assessment and treatment when sex is attached to a killing: A case study. In Wilcox

D. T.

Garett

Harkins

(Eds.), Sex offender treatment (pp. 286-304). Chichester, UK: John Wiley.

11.

Cornish

D. B.

Clarke

R. V.

(1986). The reasoning criminal: Rational choice perspectives on offending. New York, NY: Springer-Verlag.

12.

Davies

(1992). Rapists’ behaviour: A three aspect model as a basis for analysis and the identification of serial crime. Forensic Science International, 55, 173-194.

13.

Fleiss

J. L.

(1981). Balanced incomplete block designs for inter-rater reliability studies. Applied Psychological Measurement, 5, 105-112. doi:10.1177/014662168100500115

14.

Hickey

(2001). Serial murderers and their victims (3rd ed.). Belmont, CA: Wadsworth.

15.

James

Beauregard

(2018). How sexual murderers thwart investigations. In Proulx

Beauregard

Carter

A. J.

Mokros

Darjee

James

(Eds.), Routledge international handbook of sexual homicide studies (pp. 574-593). Abingdon, UK: Routledge.

16.

Kiely

T. F.

(2001). Forensic evidence: Science and the criminal law. Boca Raton, FL: CRC Press.

17.

Lee

(2005). The value of life in death: Multiple regression and event history analyses of homicide clearance in Los Angeles County. Journal of Criminal Justice, 33, 527-534.

18.

Litwin

K. J.

(2004). A multilevel multivariate analysis of factors affecting homicide clearances. Journal of Research in Crime & Delinquency, 41, 327-351.

19.

Litwin

K. J.

(2007). The dynamic nature of homicide clearances: A multilevel model comparison of three time periods. Homicide Studies, 11, 94-114.

20.

Mouzos

Muller

(2001). Solvability factors of homicide in Australia: An exploratory analysis. (Trends & Issues in Crime and Criminal Justice, No. 216). Canberra: Australian Institute of Criminology.

21.

National Centre for Policing Excellence. (2006). Murder investigation manual. Wyboston, UK: National Centre for Policing Excellence on behalf of ACPO.

22.

Pallant

(2007). SPSS survival manual: A step-by-step guide to data analysis using SPSS version 15. Berkshire, UK: Open University Press.

23.

Proulx

Cusson

Beauregard

(2007). Sexual murder: Definitions, epidemiology and theories. In Proulx

Beauregard

Cusson

Nicole

(Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 9-28). Chichester, UK: John Wiley.

24.

Radojevic

Radnic

Petkovic

Miljen

Curovic

Cukic

. . . Savic

(2013). Multiple stabbing in sex-related homicides. Journal of Forensic and Legal Medicine, 20, 502-507. doi:10.1016/j.jflm.2013.03.005

25.

Ressler

R. K.

Burgess

A. W.

Douglas

J. E.

(1988). Sexual homicide: Patterns and motives. New York, NY: Lexington.

26.

Riedel

Rinehart

T. A.

(1996). Murder clearances and missing data. Journal of Criminal Justice, 19, 83-102.

27.

Roberts

(2007). Predictors of homicide clearance by arrest: An event history analysis of NIBRS incidents. Homicide Studies, 11, 82-93.

28.

Salfati

C. G.

Haratsis

(2001). Greek homicide: A behavioral examination of offender crime-scene actions. Homicide Studies, 5, 335-362.

29.

Stefanska

E. B.

Carter

A. J.

Higgs

Bishopp

Beech

A. R.

(2015). Offense pathways of non-serial sexual killers. Journal of Criminal Justice, 43, 99-107. doi:10.1016/j.jcrimjus.2015.01.001

30.

Stefanska

E. B.

Higgs

(2018). Understanding the sexual element in sexual murder. In Proulx

Carter

A. J.

Beauregard

É.

Mokros

Darjee

James

(Eds.), Routledge international handbook of sexual homicide studies (pp. 104-116). Abingdon, UK: Routledge.

31.

Stefanska

E. B.

Higgs

Carter

A. J.

Beech

A. R.

(2017). When is a murder a sexual murder? Understanding the sexual element in the classification of sexual killings. Journal of Criminal Justice, 50, 53-61. doi:10.1016/j.jcrimjus.2017.03.004

32.

Stefanska

E. B.

Higgs

Carter

A. J.

Beech

A. R.

(2018). ‘We boil at different degrees’: Factors associated with severity of attack in sexual killing. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260518758333

33.

Trussler

(2010). Explaining the changing nature of homicide clearance in Canada. International Criminal Justice Review, 20, 366-383. doi:10.1177/1057567710379221