When both the police and the public become aware that a serial murderer has begun a series of homicides, one of the primary concerns is the length of time the offender remains free in the community. As such, the longer a serial murderer is active, the deeper the sense of panic is felt among the community. Drawing from the criminal career perspective, in particular, the criminal achievement concept, the duration of a homicide series may be seen as a direct measure of the serial murderer’s longevity; a proxy signal of their level of criminal achievement. The empirical study of serial murder has been undertaken only relatively recently and has coincided with the availability of databases designed to capture details about the offender’s homicide series (e.g., Campedelli & Yaksic, 2021; Quinet, 2011; Yaksic, 2015). But even after decades of study using both qualitative and quantitative methods, few researchers have questioned what factors directly influence the duration of the serial murderer’s homicide series and, in turn, their criminal achievement.
Attaining longevity extends beyond amassing a victim count that may earn the serial murderer the title of ‘prolific’. The number of victims is secondary to the serial murderer’s need to remain undetected and unconfined. Thus, it is assumed that criminal achievement can be distilled down to merely being free to kill for as long as possible. To accomplish this form of criminal achievement is a substantial effort, one more difficult in the modern age than in the heyday of serial murder (Yaksic et al., 2019). The study of what constitute a serial murderer’s criminal achievement is difficult because it is not possible to catalog the traits of uncaught serial murderers who have reached the extreme end of longevity.
The criminal achievement concept has been used to explain factors associated with offending productivity, or the ‘successful criminal’. The focus of this perspective was to examine the: 1) payoffs/rewards from crime, and 2) cost/detection avoidance (Bouchard & Nguyen, 2010; Lussier et al., 2011; Tremblay & Morselli, 2000). Within the context of serial murders, the payoffs or rewards from committing these crimes may be reflected in the number of victims or in the fulfilment of the offender’s motivation. More relevant to the current study, is the operationalization of criminal achievement in terms of the offender’s ability to minimize cost or avoiding detection throughout the commission of the murder series.
Serial murderers are distinct from non-serial murderers primarily because these offenders attempt to avoid suspicion while carrying out homicides repeatedly over time (James & Proulx, 2016; Pakkanen et al., 2015). Serial murderers are generally referred to as efficient killers given that they continually escape apprehension. But, in their endeavors to remain free to fulfill their objective, serial murderers encounter both fortuitous circumstances and environmental obstacles.
Therefore, a balance between factors, both within and outside of their control, enhances their chances of attaining criminal achievement. As Shanafelt and Pino (2012) proclaim, contingency, accident, and choice all play a role in serial murder. James and Beauregard (2020) also note that situational factors can nullify precautions. Guided by the criminal achievement concept, the aim of this study is to explore factors associated with the duration of a serial murderer’s killing series by examining offender, victim, and crime characteristics that can contribute to its longevity.
Early research identified certain behaviors – planning of offenses, targeting of strangers, controlling the victim with restraints, and hiding of the body – thought to enhance the likelihood that a serial murderer would be ‘organized’ and therefore be better at avoiding detection (Ressler et al., 1988). Although the prevailing belief that so-called ‘organized’ serial murderers are effective at sustaining persistent careers has contributed to their legend (Walters et al., 2015; Hodgkinson et al., 2017), research in recent years instead supports the notion that serial murderers prosper due to conscious learned strategies. These include premeditation and predation on vulnerable victims.
First, serial killings are generally preceded by premeditation (Brantley & Kosky, 2005; Harbort & Mokros, 2001; Adjorlolo & Chan, 2014; McNamara & Morton, 2004), and sometimes followed by efforts to delay or avoid arrest (Beauregard & Martineau, 2014b; Chopin et al., 2020; Salfati & Bateman, 2005). Additionally, insufficient planning contributes to cases being solved more easily (Balemba et al., 2014). Second, the degree to which a serial murderer’s plan impacts the duration of the ‘cooling off’ period and the pre-selection of homicide locations, which provide safety for the serial murderer and little opportunity for the victim’s escape, may lead to longer or shorter homicide series (Reale et al., 2017; Sorochinski et al., 2015). Finally, efficient serial murderers make rational choices and weigh many factors simultaneously in their assessment of opportunity when seeking victims to kill (Chopin et al., 2020; James & Beauregard, 2020). As in any other crime, some individuals are better at this process than others.
Serial murderers are generally in control of the death location and use their physical environment to create opportunities for murder, particularly in dense urban spaces that provide them with anonymity for the means to kill (Brookman et al., 2019; DeFronzo et al., 2007; Hodgkinson et al., 2017; James & Beauregard, 2020; Lynes et al., 2019). These environments may influence the serial murderer’s decision-making regarding how to dispose of bodies (Sorochinski et al., 2015). While body disposing and non-body disposing serial murderers make conscious decisions equally in choosing the location of the crime depending on their sophistication, time, and effort (Morton & McNamara, 2005; Synnott et al., 2019), some will use deliberate strategies offering the least amount of evidence (Mott, 1999). These strategies include moving the body to delay or avoid recovery as a method of detection avoidance (Reale & Beauregard, 2019), concealing the body (Beauregard & Martineau, 2014b), or leaving the body outside and exposed to the elements. Other serial murderers leave their victims “as is” after their death as a means of minimizing the chance of leaving evidence behind (Morton et al., 2014). Each of these tactics hampers forensic analysis, disrupts investigations, and leads to longer kill periods, thereby making cases harder to solve (Quinet, 2011).
Victimology also influences how vigorously police work to solve homicides—what Beauregard and Martineau (2014b) call ‘victim preferencing’. Studies have shown that demographic characteristics such as race and ethnicity influence how individuals spend their time, which impacts risky lifestyles and routines, and in turn affect their exposure to crime and victimization (e.g., Lauritsen & White, 2001; Peguero et al., 2013). The criminal opportunity theory has been used to explain the disparities in violent victimization across different racial groups, wherein minority individuals living in socioeconomically disadvantaged areas were more likely to be exposed to risky lifestyles (Cohen et al., 1981; Hindelang et al., 1978). Studies have suggested that criminal opportunities were often exploited in socially disorganized neighborhoods due to the lack of social guardianship and informal surveillance, thereby increasing the risk of victimization (Bursik & Grasmick, 2001; Schreck et al., 2002; Spano & Nagy, 2005).
In the context of criminal opportunity, some serial murderers target sex workers, the homeless, drug abusers, runaways, transients, or other marginalized populations because they understand that the victim’s social status can influence the attention garnered from investigators (Brookman et al., 2019; Reale & Beauregard, 2019). These victims are less likely to be reported missing, which can cause homicides to remain unsolved (Mott, 1999; Quinet, 2011). Even if reported, police may not place marginalized victims on a priority list (Lee & Reid, 2018). These choices in victim characteristics can enable serial murderers to continue killing for longer periods of time because the police have more difficulty linking the crimes to one offender, or they may simply not prioritize these cases.
Beyond the offender and victim characteristics, routine events, location, and social controls also play a role in the duration of a murder series. On an individual level, offender’s choices regarding the crime, such as the level of preparation, victim type, killing method, interaction with the victim, and body contribute to a murder series’ longevity disposal (Beauregard & Martineau, 2016; Quinet, 2011; Reale & Beauregard, 2019). On a societal level, the murderer’s homicide series can be influenced by routine events (e.g., daily activities) (Osborne & Salfati, 2015), location (e.g., geographic factors) (DeFronzo et al., 2007), and social controls (e.g., employment and romantic attachments) (Liem, 2013; Osborne & Salfati, 2015). Studies have also shown that situational circumstances (e.g., degraded evidence, improper police practices, and lack of investigative skills or inaction) have profound effects on the duration of the murderer’s homicide series (Balemba et al., 2014; Brookman et al., 2019; James & Beauregard, 2020; James & Proulx, 2016; LePard et al., 2015; Reale & Beauregard, 2019).
While serial murderers are thought to amass expertise and specialize in repetitive killing, the surrounding offenses of kidnapping, rape, armed robbery, and theft often occur over their series as well (DeLisi et al., 2019). Since the ulterior motives were not just the act of killing, the serial murderer is compelled to commit the act multiple times to achieve their goals, while acquiring apprehension prevention measures at the same time (Salfati & Bateman, 2005). It has been noted that the sexual serial killer, a subset of serial murderer, acquires some forensic knowledge through their experience and exhibits post-crime signs of organization (Beauregard & Martineau, 2014a; Chopin et al., 2020; Morton et al., 2014; Stefanska & Carter, 2019). Thus, this experiential journey allows most serial murderers to kill without a partner, given that the average duration of the homicide series for team perpetrators has been shown to be shorter than for those killing alone (Chai, Yaksic et al., 2021). Experience also leads some serial murderers to intentionally vary their killing methods (e.g., using blunt objects on some victims, while stabbing, strangling, and shooting others) (Beasley, 2004), use atypical methods (e.g., bombing, gassing, drowning), and select from a variety of victim types (Beasley, 2004; Harbort & Mokros, 2001; Salfati & Bateman, 2005) to create confusion and prevent separate jurisdictions from linking homicides (Sane et al., 2017).
In addition to situational circumstances, social constructs itself may influence the duration of a murder series. Discussions in the critical criminology literature have highlighted the systemic racial disparities apparent in penal institutions across the United States (US), where offenders from minority groups (e.g., Black, Hispanic, Indigenous) are being disproportionately arrested and incarcerated for violent crimes (Demuth & Steffensmier, 2004; Sampson & Lauritsen, 1997; US Department of Justice, 1993). For instance, among offenders who committed their first homicide, it is likely that offenders from minority backgrounds spent longer time in correctional institutions for nonviolent crimes (e.g., technical violations), or other crimes not related to homicide. Studies have also shown that the proliferation of arrest and incarceration among these individuals may have contributed to an increase in the likelihood of gaining criminal capital and improved criminal skills within the institution walls (Camp & Gaes, 2005; Sampson & Lauritsen, 1997; Vieraitis et al., 2007). Such experience gained through multiple arrests and spending longer time being incarcerated for unrelated crimes makes serial murderers more versatile criminals than non-serial murderers, as they are less prone to impulsive outbursts, and are apprehended less frequently for violent crimes (James et al., 2019). Scholars have surmised that the serial murderer’s level of experience with the criminal justice system can be beneficial to navigate around police tactics and strategies (US Federal Bureau of Investigation, 2008).
Although other scholars have studied some of these factors as noted above, these aspects were studied separately from one another. Factors such as preparation, victimology, experience with the criminal justice system, body disposal, and ‘luck’ in serial murders were analyzed without considering how these factors affect the murder series collectively. Thus, the findings from these studies are therefore limited in their potential implications for investigations. Drawing from the criminal achievement concept, the current study is the first to consider how these factors converge to influence the duration of the serial murderer’s series (i.e., cost/detection avoidance).
A recent study conducted by Campedelli and Yaksic (2021) reviewed several variables for potential impact on a serial murderer’s homicide series. This was the first study that examined factors predicting the end of a multiple homicide offender’s career while controlling for time duration. Results showed that young female serial murderers who had prior criminal histories, who killed alone, targeted both sexes, used multiple killing methods, and who killed in more than one state, were more likely to have a longer career. The authors noted that the race of the offender and the number of victims in a series had no impact on when a series concluded. While being the first study to investigate impacts on homicide series, it was limited in that it included spree killers and only US-based offenders operating without a partner.
While Campedelli and Yaksic’s (2021) study examined factors that influenced the duration of a murder series, its focus was to predict variables associated with the end of a criminal career (i.e., end of the murder series), in which time factor was used as a weighted variable. In contrast, the current study focuses on variables directly associated with the time factor (i.e., duration of a murder series). Thus, building on the work conducted by Campedelli and Yaksic (2021), the aim of this study is to predict the duration of the serial murder series by discerning the factors that contribute to the length of time a serial murderer is able to remain free of police detection. Additionally, the focus of this study is to extend the results’ generalizing potential by examining a broader set of data that include 56 countries from around the world in combination with a comprehensive set of offender, victim, and crime variables on a multivariate level.
The formulated research question for this study is, ‘what factors predict the duration of a serial murderer’s series?’ Factors that the offender had direct control over were divided into three phases: target selection, crime commission, and post-crime actions. These factors include offender and victim demographics, the victim-offender relationship, modus operandi, motivation, level of experience, killing method(s), and circumstances surrounding the victim’s body.
Data from the Consolidated Serial Homicide Offender Database (CSHOD) was used in this study. The CSHOD is a product of merged records from independently built databases maintained and contributed by scholars who are members of the Atypical Homicide Research Group (AHRG). The CSHOD is a filterable database designed to allow users to curate the data to fit their given inclusion and exclusion criteria and allows users to select the serial murder definition that best suits their research study. While the AHRG does not endorse one particular serial murder definition over another, the CSHOD was designed to be as comprehensive as possible and, to that end, was built using the definition put forth by the FBI where serial murder is defined as ‘the unlawful killing of two or more victims by the same offender(s), in separate events.’ (US Federal Bureau of Investigation, 2008, p. 9).
The CSHOD is comprised of records obtained through: (1) scholarly journal articles, (2) court and prison records, (3) textbooks, (4) news articles, (5) contributions from law enforcement, (6) Freedom of Information Act requests, (7) Internet searches, (8) personal files, (9) social media, and (10) true crime books. A database coordinator is responsible for overseeing the process of verifying information. The validity of the data is maintained by consistent cross-checks of old and new information against daily Internet searches for the terms ‘serial killer’, ‘serial homicide’, and ‘serial murder’. Any new additions to the database are scrutinized by at least two coders and measured against at least two independent sources. Data are only input into the database after approval by the coordinator.
While the CSHOD contains data on all subsets of serial murderers who victimize acquaintances, family members, intimate partners, and strangers in various settings (e.g., college campuses, medical or nursing facilities, private residences) and scenarios (e.g., arguments, domestic violence, robberies, sexual encounters), serial murderers must be responsible for at least two homicides, committed in separate events, to be included in the CSHOD. Additionally, homicide events must occur across at least two locations, either in a rapid, spree-like sequence or a longer time span of days, weeks, months, or years. A variety of motives identified from the offender’s personal statements and/or media reports include: (1) anger (e.g., anti-government beliefs, domestic disputes, racial hatred, revenge), (2) convenience (e.g., eliminate witnesses, undesired relatives), (3) criminal negligence (e.g., over-prescription of drugs, reckless indifference), (4) enjoyment (e.g., control, pleasure, power, thrill), (5) financial (e.g., access to spousal resources, burglary, professional contract, robbery), (6) attention (e.g., fame, celebrity), (7) avoid arrest (e.g., escape apprehension), (8) torture (e.g., inflicting pain and suffering), (9) mental illness (e.g., paranoia, schizophrenia), (10) random attacks, (11) sexual assaults, (12) multiple motives, or (13) unknown. Users of the CSHOD may sort by and remove any of the subcategories within the 13 overarching motive categories that do not fit their study aims. Data from the CSHOD have been used in recent studies that reviewed serial homicide time intervals (Yaksic, Simkin et al., 2021), spree killing (Yaksic, 2019), geographical profiling (Salafranca Barreda, 2021), careers (Campedelli & Yaksic, 2021), and teams (Chai, Yaksic et al., 2021).
The present study only considered cases of serial murder with three or more victims and homicides that occurred in at least two separate events over a period of time greater than 14 days. Serial murderers who killed two victims were excluded from the present analysis given the controversy surrounding the inclusion of these offenders as they are thought to be different from traditional serial murderers across a variety of key characteristics (Yaksic, 2018). Events must have occurred over a period greater than 14 days to abide by the parameters set forth in Schlesinger et al. (2017) and Yaksic (2019) where it was determined that serial murderers who kill their victims within an interval of 1-14 days could be considered spree killers. To avoid such confusion, the present analysis included ‘true’ serial murderers, or those who conformed most closely to the literature base. The following exclusions were made: 265 serial murderers who were part of a large organization (i.e., gang, drug or criminal enterprise), 48 offenders who were accused, 204 who were suspected, 31 who were self-proclaimed, 4 who ordered others to kill, 390 who lacked a designated motive, and 1,797 whose records contained missing information. Offenders must have had a designated motive given the subjectivity involved in assigning a motive to a series of crimes when one is not readily apparent from established sources. Victims could have shared any type of relationship with the offender, including acquaintance, family, or stranger. Guided by logic presented in previous literature on serial murder (e.g., Hickey, 2015), the analysis was limited to serial homicide series that occurred after 1949 given concerns with the veracity of data before that time, issues that arose due to substandard record keeping, a lack of definitional consensus, and varied classification schemes for serial murders. Therefore, the sample used in this study includes 1,376 persons involved in 1,258 series (i.e., 128 teams of offenders and 1,128 solo offenders).
The choice of variables was guided by previous empirical work that focused on efficient serial murderers in the context of the duration of their homicide series (see Balemba et al., 2014; Beauregard & Martineau, 2014a, 2014b; Brookman et al., 2019; Campedelli & Yaksic, 2021; Morton et al., 2014). All but two independent variables (i.e., age and number of victims) were dichotomous.
One continuous dependent variable was used in this study. This variable measures the series duration in number of days from the date of the first murder to the date of the last murder before the police arrest. The murder series lasted an average of 2,026.23 days (SD = 2,845.19; range = 14-16,266) (a mean of about five and a half years). This variable was not normally distributed, with skewness of 2.63 (SE = 0.07) and kurtosis of 5.85 (SE = 0.13).
Offender Characteristics. Offenders are often exposed to the criminal justice system before the start of their serial murder series, which can inform them on how best to navigate circumstances that arise during their criminal career. Because exposure to the criminal justice system can be difficult to measure directly, two variables were used as proxy measures: ‘previous time incarcerated’ that refers to any amount of time the offender spent under incarceration for any crime, and ‘killed prior to series’ which refers to one or more detected homicides that occurred peripheral to the offenders’ homicide series for which they were convicted, much the same as serial murderer, Edmund Kemper murdered relatives years before killing several college-aged females over the serial homicide series that resulted in his final arrest. The age range for offenders spanned 13–19 years for teens, 20–64 years for adults, and 64–70 years for seniors.
Crime Parameters. There were five primary motivations that spurred offenders forward in the present study that were identified from the offender’s personal statements or writings and/or media reports summarizing police findings. Serial homicides were committed for the following reasons: (1) financial (i.e., to acquire goods of value), (2) enjoyment (i.e., personal gratification in the form of pleasure, leisure, power, or control), (3) anger (i.e., to quell feelings of rage), (4) rape (i.e., to attain sexual dominance), and (5) other. ‘Other’ encapsulates attention (i.e., to garner fame by way of the news or entertainment media or impress others with their prowess), avoid arrest (i.e., to escape apprehension), torture (i.e., to bring about pain and suffering), and mental illness (i.e., a by-product of a disease or defect). Serial homicide perpetrators could have engaged in the homicide series for more than one of these reasons, and if so, they were coded as having multiple motives. ‘Crime dynamic’ refers to the presence or absence of a killing partner. This variable was included to control for the differences in the series’ duration between solo and team perpetrators of homicide (Chai, Yaksic et al., 2021).
Crime-commission Process Characteristics. Offenders engage in a variety of behaviors before, during, and after a crime which are designed to both enhance their chances of escaping detection. Although these behaviors are often directly captured across a range of variables in the CSHOD, the offender’s ability to plan their crime can only be measured indirectly. Two variables were used as proxy measures for planning given the premeditated behavior inherent to each action: ‘sought victim to kill’, which refers to a concerted effort made to actively stalk and isolate a victim during delineated trolling activities, and ‘victims bound/blindfolded’, which refers to efforts made by the offender to restrict the victim’s awareness or movement through the use of materials intentionally brought to the crime scene. ‘Evidence of overkill’ is present if the victim sustained at least 15 crushing (e.g., as with blunt trauma) or penetrating wounds (e.g., as with stab wounds or gunshot wounds) (Trojan et al., 2019) ‘Victims were mutilated’ refers to the presence of irreparable disfigurement to a body part caused by the offender. ‘Victim’s bodies were moved’ refers to a mechanism of body disposal where the victim’s remains were transported from the crime scene to another location after death. ‘Offenders interacted with victim’s bodies’ refers to posthumous actions that include trophy or souvenir taking, sex after death, consumption of body parts, and posing of the victim’s body. The age range for victims spanned <1 year for infants, 1–12 years for children, 13–19 years for teens, 20–64 years for adults, and 64–70 years for seniors. Victims were classified based on their role in society. ‘Street people’ refers to sex workers, the homeless, and/or chronic drug abusers.
The analytical strategy in this study followed a two-step process. The first step consisted of a bivariate comparison (i.e., Mann-Whitney U test1, Kruskal-Wallis H test, and Pearson correlation coefficient) to determine the associations between the dependent (i.e., continuous) and independent (i.e., categorical and continuous) variables. The goal was to determine which variables are the most significant predictors to use in the multivariate analysis. Second, we tested at the multivariate level the significant differences (p <.05) identified in the bivariate analysis. As suggested by several authors (see Hosmer & Lemeshow, 2013; Tabachnick & Fidell, 2019), it is not recommended to perform multivariate analyses when bivariate analyses indicate that there is no relationship between two variables.
Specifically, a negative binomial generalized estimating equation (GEE) was computed. The GEE method allows for data that have correlated responses to be modeled as a generalized linear model (Liang & Zeger, 1986; Zeger & Liang, 1986). Correlated data are used in longitudinal studies, clustering (Johnston & Stokes, 1997) and analysis of serial crimes (Ensslen et al., 2018; Hewitt & Beauregard, 2014; Hewitt et al., 2012). The GEE is particularly appropriate for the data used in this research. Since the unit of analysis is the offender, team serial murderers are associated with victims and crimes with the same characteristics. As a result, it cannot be assumed that independence exists between each observation, and the current models need to take this dependence into account.
The Quasi Likelihood under the Independence Model Criterion (QIC) was analyzed to determine the appropriate correlation structure. The QIC allows for the selection of covariates and working correlation structure simultaneously (Cui & Qian, 2007; Pan, 2001). The analysis showed that the independent correlation matrix had a lower QIC value compared to the unstructured matrix; therefore, the GEE model was estimated with the former matrix. Given that the dependent variable (i.e., duration in days) was a continuous variable and positively skewed (skewness = 2.778), the negative binomial link function was specified (see e.g., Green, 2021). The multicollinearity scores were calculated for the variables included in the multivariate analyses and indicated no variance inflation factors (VIFs) above 1.56 and tolerance not below 0.64. These results are within the acceptable threshold (Menard, 2002).
Table 1 summarizes the duration of serial homicide in the sample ranging from less than one year to more than 30 years. The majority of the series duration was brief (less than one year = 45.23%) followed by a downward trend as the number of years increases. However, the number of cases with over 30 years in the series was 8.74%, representing the second largest number of cases in the sample (see Figure 1).
Table 1: Description of the Duration of Serial Homicide Series (n = 1258)
Duration in years
Less than 1 year
6 to 10 years
11 to 15 years
16 to 20 years
21 to 25 years
26 to 30 years
More than 30 years
Duration in days: Mean
Duration in days: Median
Durations in days: SD
Duration in days: Range
Figure 1: Duration of serial homicide series
Bivariate analyses were conducted between offender, crime parameters, crime-commission process characteristics, and the duration of homicide series. Table 2 shows five offender characteristics to be significantly associated with the duration of homicide series: female offender (Mann-Whitney U = 66628.50; p = .039), offenders identified as Indigenous (Kruskall-Wallis H = 29.29, p < .001), offender previously incarcerated (Mann-Whitney U = 170909, p < .001), previously killed prior to series (Mann-Whitney U = 9436, p < .001), and offender’s age at first killing (Pearson’s r = -0.13, p < .001). Two crime parameters were significantly associated with the duration of homicide series (Table 3): other motives (Kruskall-Wallis H = 78.12, p < .001) and offenders operating in teams (Mann-Whitney U = 75302, p < .001).
Table 2: Bivariate Analysis of Offenders’ Characteristics (n = 1258)
Number of days in series (Mean)
Previous time incarcerated
Killed prior to series
Age at first killing
28.43 [SD=8.96; 8-68]
Notes. * p < .05. *** p < .001.
a Mann-Whitney U
b Kruskal-Wallis H
c Pearson’s r
Table 3: Bivariate Analysis of Crime Parameters (n = 1258)
Number of days in series (Mean)
Number of victims
6.28 [SD=8.24; 2-200]
Notes. *** p < .001.
a Mann-Whitney U
b Kruskal-Wallis H
c Pearson’s r
Table 4 shows associations between the crime-commission process variables and the duration of homicide series. In the pre-crime phase (target selection), both male and female victims (Kruskal-Wallis H = 20.65, p <.001), multiple victim’s race (Kruskal-Wallis H = 35.92, p < .001), and a combination of adults and child victims (Kruskal-Wallis H = 29.29, p <.001) were significantly associated with the duration of series homicide. In addition, the exclusive selection of street victims was significantly associated with the duration of the series (Mann-Whitney U = 65094, p = .001). In the crime phase, six characteristics showed significant associations: victims who were raped (Mann-Whitney U = 169823.50, p <.001), overkilled (Mann-Whitney U = 102870, p = .047), or mutilated (Mann-Whitney U = 80917.50, p = .012) were significantly associated with the duration of the homicide series. Among killing methods, only strangulation was found to be significant (Kruskal-Wallis H = 75.92, p = < .001). Lastly, in the post-crime phase, the moving of the victims’ body (Mann-Whitney U = 143298.50, p < .001) and offenders who interacted with the victims’ body (Mann-Whitney U = 108367.50, p = .033) were significantly associated with the duration of the homicide series.
Table 4: Bivariate Analysis of Crime-commission Process Characteristics (n = 1258)
Number of days in series (Mean)
Sought victim to kill
Both male and female
Race of victims
Target street people only
Victims were raped
Victims were tortured
Evidence of overkill
Victims were mutilated
Victims were bound / blindfolded
Victim bodies were moved
Offenders interacted with victim bodies
Notes. *** p < .001.
a Mann-Whitney U
b Kruskal-Wallis H
c Pearson’s r
Significant variables from the bivariate analyses were modeled using the negative binomial GEE method (Table 5). Goodness-of-fit for the models was assessed by the Corrected Quasi-Likelihood under the Independence Model Criterion (QICC), in which QICC adds a penalty to the quasi-likelihood based on the number of parameters within a model (Pan, 2001; Hardin & Hilbe, 2003). The QICC value was associated with the model with significant variables from all bivariate analyses (QICC = 5224.94). Five offender characteristics resulted in significant effects on the duration of the homicide series. Compared to Indigenous offenders, White (b = 1.56, S.E. = 0.71, p = .028) and Hispanic (b = 1.77, S.E. = 0.80, p = .027) offenders were more likely to have longer homicide series. Further, being previously incarcerated (b = 0.14, S.E. = 0.13, p = .003) and killed prior to series (b = 0.53, S.E. = 0.18, p = .004) led to an increase in duration of the series. Homicide series were more likely to be shorter as offender’s age increased (b = -0.04, S.E. = 0.01, p < .001).
In terms of crime parameters variables, none of the motivation factors were significant in this model. Offenders who killed alone were more likely to have longer duration series compared to offenders who operated in teams (b = 1.66, S.E. = 0.20, p < .001). For victim characteristics, the sex of the victim was not significant in predicting the duration of the homicide series. Black (b = 0.63, S.E. = 0.29, p = .027) and Asian (b = 1.81, S.E. = 0.40, p < .001) victims were more likely to lead to longer series duration. On the other hand, when victims are Indigenous, the homicide series was more likely to be shorter (b = -1.60, S.E. = 0.62, p = .010). Additionally, the evidence of rape in a serial homicide case more likely led to a longer series duration (b = 0.23, S.E. = 0.19, p = .012).
Two types of killing methods significantly predicted an increase in the duration of the homicide series: multiple methods (b = 0.85, S.E. = 0.18, p <.001) and atypical methods (b = 1.86, S.E. = 0.43, p < .001) when compared to strangulation. In the post-crime phase, the homicide series will more likely be shorter when the offender moved the victim’s body (b = -0.44, S.E. = 0.16, p = .005).
Table 5: Generalized Estimating Equations with Negative Binomial Link Function of Factors Predicting the Duration of Serial Homicides (n=1258)
95% CI for β
Previous time incarcerated
Killed prior to series
Age first killing
Crime-commission process characteristics
Both male and female
Race of victims
Target street people only
Victims were raped
Evidence of overkill
Victims were mutilated
Victim bodies were moved
Offenders interacted with victim bodies
Quasi Likelihood under Independence Model Criterion (QIC)b
Corrected Quasi Likelihood under Independence Model Criterion (QICC)b
Notes. * p < .05. *** p < .001.
Despite considerable research on serial murder series in recent years, little is known about how different factors collectively affect the duration of a killing series. Campedelli and Yaksic’s (2021) study was the first to tap into a murder series’ ‘longevity’ by examining factors that contributed to the end of the murder series while controlling for time duration. Nonetheless, the time duration aspect was not addressed (only controlled for) in their study. The purpose of the current study was to investigate the offender, victim, and crime factors that aid serial murderers in their pursuit of criminal achievement through longevity, interpreted here as the duration of time between the first and last homicides in a series.
Serial murderers often rely on experience to inform how they approach events over their career, a process that compels them to exercise forethought and prepare for future crimes. Harbort and Mokros (2001) point out that experience and ideation together generate rational, goal-oriented strategies that are more likely to lead to success. DeLisi and colleagues (2017) agree and state that perpetrators of homicides involving ideation, planning, and contemplation are more likely to be chronic offenders. Furthermore, studies have shown that a high proportion of sadistic serial offenders plan their homicides (Dietz et al., 1990).
According to the literature, planning is a core element of serial murder and a critical component of the offender’s success (Brantley & Kosky, 2005; Harbort & Mokros, 2001; Adjorlolo & Chan, 2014; McNamara & Morton, 2004) that may consist of actions taken during victim selection and/or result in the accumulation of items used in the homicide (Morton & McNamara, 2005). One method employed by serial murderers in the current study is the premeditated and conscious decision to seek a specific victim to kill. The preemptive carrying of binding materials to the crime scene is another way that serial murderers in this study plan for their crimes. Using these metrics, however, we found no association between planning and the duration of the homicide series. Although this finding contrasts with much of the literature on the subject, there are studies that question the efficacy of planning in the context of serial murder. One such study found that there was no intergroup difference in planning between ‘organized’ and ‘disorganized’ serial murderers (Prentky et al., 1989). Another study noted that planning was not always necessary to serial murders because impulsivity played a role in many haphazard killings committed as reactions to situational circumstances (Beasley, 2004).
In terms of demographics, the offender’s race has a significant effect on the duration of the murder series. Both White and Hispanic offenders were more likely to have killed for a longer period prior to the end of their series. From a critical criminology perspective, the systemic racial bias inherent within the criminal justice system could explain these findings. First, White individuals are less likely to be arrested for violent crimes (Sampson & Lauritsen, 1997; US Department of Justice, 1993), which suggests that police are less likely to focus on White suspects, leading to their prolonged murder series. Second, research has shown that Hispanic individuals were more likely to be incarcerated for property crimes and drug crimes compared to White individuals (Demuth & Steffensmier, 2004), and were more likely to receive mandatory minimum sentences compared to any other ethnic groups in the US (US Sentencing Commission, 2011). Studies from different countries have found similar findings where minority groups were more likely to receive prison sentences (in Netherlands, Wermink et al., 2018), longer incarcerations (in France, Gautron & Retière, 2016), and higher conviction rates (in Belgian, Bielen et al., 2020). Upon further inspection, approximately 70% of Hispanic offenders from our dataset were living in the US.
The criminogenic effects of incarceration and imprisonment have been widely discussed elsewhere (see Camp & Gaes, 2007; Sampson & Lauritsen, 1997; Vieraitis et al., 2007). Specifically, in our study, results suggest that longer and more frequent time in prison allowed for Hispanic offenders to gain criminal capital experience, such as detection avoidance and improving criminal skills, consequently resulting in longer delays in apprehension (Chopin et al., 2020; Vieraitis et al., 2007). Since the majority of Hispanic offenders in the current dataset is represented by the US sample, the criminogenic effects of mass incarceration of minority offenders make intuitive sense. This is further corroborated by the finding in which offenders who had previously been incarcerated were more likely to have a longer murder series. Time spent in penal institutions may have allowed for offenders to learn criminal skills, perhaps reinforcing their own criminal identities. This individualized experience may be why some offenders prefer to carry out their murder series alone. Since the likelihood of their partner implicating them or leaving behind evidence is lower, solo killers are more likely to be apprehended later than killers who operate in teams (Chai, Yaksic et al., 2021).
An alternative explanation may be attributed to the fact that since Hispanics were more likely to be convicted and incarcerated for other types of crime, the murder series may have been interrupted. This will more likely prolong the duration between the first murder to the last murder. In this case, the criminal achievement of a serial murderer was not influenced by criminal experience, but rather, contextual factors such as incapacitation.
In terms of killing methods, criminal experience may play an influential role in the murder series’ length. Results show that when the offender used a combination of multiple methods across their series (e.g., using blunt objects, stabbing, strangling, and shooting), they were able to offend for a longer period of time. Through learned experience, serial murderers may have intentionally adjusted their modus operandi to carry out their crime in attempts to delay police efforts in linking multiple murders back to them (Beasley, 2004). Alternatively, it is also possible that the combination of multiple killing methods was due to situational factors. For example, studies have shown that personal weapons (e.g., strangulation, beating) were commonly used when the victim was perceived to be physically weaker in sexual homicide cases (Chan & Beauregard, 2016). The choice of weapon was also found to be associated with the offender’s primary motivation for the crime (e.g., sexual, financial) (Chan & Li, 2020), and whether it was a weapon of opportunity (e.g., weapon found at the scene) (Salfati & Park, 2007). Police inexperience with rare modus operandi such as atypical killing methods (e.g., bombing, gassing, drowning, burning) further impedes and exacerbates the challenges in linkage analysis (Sane et al., 2017), and consequently gives the offender the opportunity to remain active longer.
In the victimization literature, criminal opportunity theory posits that minority and marginalized individuals living in socioeconomically disadvantaged areas have higher risk of being victimized (Shaw & McKay, 1942; Lauritsen & White, 2001) and are more likely to participate in risky lifestyles (Cohen et al., 1981; Hindelang et al., 1978; Peguero et al., 2013; Peguero et al., 2015). Results from the current study show that victims from minority backgrounds (i.e., Black and Asian) significantly predicted longer duration in the murder series. Marginalized victims such as sex workers, homeless persons, drug abusers, and transients were often not reported missing (or in a timely manner), which made them the preferred targets for serial murderers (Chan & Beauregard, 2019; Horan & Beauregard, 2018; Reale & Beauregard, 2019). Referred to as ‘victim preferencing’ (Beauregard & Martineau, 2014b), less attention is given to marginalized victims (Lee & Reid, 2018) by the police, subsequently allowing serial murderers to continue their crimes longer and avoid detection.
Interestingly, Indigenous victims were associated with shorter duration in the murder series. This finding contradicts the well-established issue of the high prevalence among murdered and missing Indigenous women and girls (Urban Indian Health Institute, 2018; Allas et al., 2018). We revisited our data and identified that all murder cases with Indigenous victims were committed by Indigenous offenders. We hypothesized that the crime most likely occurred within the same Indigenous community, which made it easier for police to narrow down their suspect list. It is possible that many cases of murdered and missing Indigenous women and girls remain unsolved or undetected for longer because the offenders were not Indigenous (e.g., police are less likely to suspect White offenders, jurisdiction issues). This could be a potential avenue for future research to investigate reasons for the prolonged duration of a murder series involving Indigenous communities by examining the relationship between victim and offender demographics.
Results have also highlighted several crime characteristics that were significantly associated with the murder series’ duration. Contrary to studies that suggested offenders adopted post-homicide behaviors (e.g., moving the victim’s body or concealing the victim’s body) as a method of detection avoidance and potentially prolonging the murder series (Morton et al., 2014; Reale & Beauregard, 2019), current results suggest the opposite (i.e., offenders were detected sooner). It is possible that when the offender tried to move the body after killing the victim, he was exposed to potential eyewitnesses in the area, or was more likely to leave forensic evidence during the move, which resulted in shorter time for police to recover the victim’s body (Chai, Beauregard et al., 2021). It would be interesting to examine cases where the victim’s body was moved versus cases where bodies were not moved to ascertain if there were any differences in terms of additional forensic evidence left behind.
In cases where there was evidence of sexual assaults (i.e., rape), the duration of the murder series was longer than when the victims were not raped. This finding seemed counter intuitive, as physical contact with the victims may have offered additional forensic evidence to investigators (e.g., semen left behind) (see Chopin et al., 2020). An added sexual element in homicide cases is usually characterized by unusual and irrational acts, which compounds the complexity in investigative efforts (Chopin & Beauregard, 2022, 2021a, 2021b). Thus, it is possible that the complex nature of sexual homicide cases in the current study complicated the investigation, allowing offenders to thwart or delay detection and to kill for longer duration.
While informative, there are several limitations to this study. First, the current dataset lacked information pertaining to ‘luck’, which includes situational circumstances such as degraded evidence, improper police practices, and poor or lack of investigator’s skills. Several studies have identified luck as being a crucial factor in determining whether the offender is able to escape detection. (e.g., Balemba et al., 2014; Brookman et al., 2019; James & Beauregard, 2020; Lepard et al., 2015; Rossmo, 2009; Schlesinger, 2008; Yaksic, Bulut et al., 2021). However, we are unable to discern variables from the CSHOD dataset that would allow for an adequate operationalization of the concept. Future research should collect information associated with luck, potentially from interviews with police investigators concerning the situational circumstances surrounding the delay in apprehension or detection over the murder series. Relatedly, potential interruptions in the series could not be controlled for in the current study (e.g., periods of incarceration between killings). Future research should attempt to assess the offender’s “time at-risk”; that is, time spent in the community to freely carry out the murders.
Second, while our sample is comprised of serial homicide cases from 56 countries, it was not possible to control for cultural and societal differences across countries. A growing body of research on serial murderers from countries outside of the US, however, has suggested that the behaviors of serial murderers from those countries are similar to offenders from the US (Chopin & Beauregard, 2019; Chopin & Beauregard, 2021c; Deepak & Ramdoss, 2021; Harbort & Mokros, 2001; Morton et al., 2014; Salfati et al., 2015; Sturup, 2018; Yaksic, 2022). Also, it is still not clear how homicide cases were prioritized according to which criteria for each country in the sample. Investigative practices may also vary between police agencies in different countries, which could influence the crime solving and consequently the series duration. Serial homicide may be a universal occurrence, but future research would benefit from adopting a more extensive international perspective to discern any potential cultural and societal differences.
Third, this study is based on cases that occurred between 1948 and 2016. During these 68 years, investigative techniques and forensic identification have changed and evolved, which could have affected our results. However, this possibility is limited by the fact that a large portion of the cases included in this study (i.e., 44.25%) occurred since 1990. Fourth, we were interested in crime-related factors and were unable to test non-criminal indicators that could also influence the series duration. Some life events (e.g., marriage, children) could explain a temporary break in the commission of crimes that would necessarily impact the length of the series. We also could not account for any attempted homicides that occurred during the offender’s series. While every effort was made to ensure that the date of the offender’s first homicide was accurate, it can be challenging to know if offenders are responsible for additional homicides that are unknown to police simply given the furtive nature of homicide and their efforts which are designed to conceal their actions.
Fourth, potential interaction effects were not assessed since the design of this study was exploratory in nature. Due to the sparse literature in serial murder, we were unable to confidently hypothesize potential interacting terms between different variables. Future researchers could expand on the results from the current study and attempt to identify and unpack potential interaction effects among the variables that has shown significant main effects.
Finally, because serial murderers have the advantage of being in control of many variables related to their crimes, police often do not connect their homicides until later in the series. This benefits offenders by providing what can amount to significant lead time, but this translates into a somewhat artificial extension of their longevity. Because the police were usually not aware of the early stages of the serial murderer’s activities, those offenders did not have to work as hard to avoid apprehension in those circumstances.
On a practical level, results presented in the current study offer useful information for those within law enforcement, particularly in serial murder cases. Factors from offender, victim, and crime characteristics that significantly predicted the murder series’ duration in this study can be explicitly identified in open serial homicide cases. For example, the victimology aspect has been shown to contribute to longer duration in the murder series, wherein if the victims are from a minority background (i.e., Black or Asian), it is more likely that the offender will continue killing for longer. This should prompt investigators to invest more attention and effort into the case, since offenders may be more likely targeting this group of victims specifically knowing that they may be ‘low priority’ on the police priority list.
Furthermore, investigators should pay closer attention to the aspect of criminal experience when they are in the process of narrowing down potential suspects of a murder series. Results showed that offenders who had been previously incarcerated, and who has killed prior to the series were more likely to have longer murder series, therefore suggesting that their experience with committing similar crimes (i.e., murder) and potentially the criminal justice system may be aiding in their ability to thwart police detection. Thus, when a murder series has been identified, the first potential pool of suspects should be focused on those who had already been under police radar such as criminal and prison records, or information from crime linkage databases such as the Violent Criminal Apprehension Program (FBI, n.d.).
In terms of crime scene evidence, multiple and atypical types of killing methods should not deter investigators from concluding the cases as not linked. While popular media has portrayed serial murderers to be consistent in their modus operandi, several studies have found that serial murderers often vary their killing methods across victims (Beasley, 2004; Harbort & Mokros, 2001; Salfati & Bateman, 2005; Schlesinger et al., 2017). In the current study, both killing methods have been shown to increase the murder series’ duration in comparison to strangulation, possibly an intentional decision on the offender’s part to delay and thwart police investigators. Thus, police should not be so quick to dismiss a lack of linkage between cases with unusual and different killing methods.
The current study’s results showed shorter murder series duration when the victim’s body was moved (i.e., offender was apprehended sooner). In line with previous studies, the moving of the victim’s body post-homicide increased the risk of detection for the offender in terms of potential eyewitnesses and forensic evidence left behind during the move (Beauregard & Martineau, 2016; Chai et al., 2021). Thus, upon determining that a victim’s body was moved, this should prompt investigators to be on alert for potential forensic evidence left behind by increasing the search area around the body disposal site. Additionally, police should expand their efforts to seek out and interview as many eyewitnesses as possible within a wider area in vicinity of that location.
On a theoretical level, the current study contributes empirically to the serial murder literature, specifically in terms of the series’ longevity. Our analysis extends beyond current knowledge on factors that contribute to a murder series’ success by specifically examining duration as an outcome. Additionally, factors were analyzed collectively in the current study thus providing a more comprehensive understanding of the phenomenon. The expansive and representative sample used in this study also allows for greater generalizability capabilities, particularly when compared to previous studies that were restricted geographically.
Furthermore, the current study creates several potential avenues for future research opportunities. For example, this study did not incorporate the time between the final homicide and the serial murderer’s arrest given that this period of time exists beyond the bounds of the offender’s series and outside of the scope of this study. Some offenders (12%) remain free of apprehension for a year or more after they conclude their killing series2 and future research should consider the factors that allow these particular offenders to remain free of suspicion after they completed their series. This type of longevity attained by this subgroup is often due to a self-imposed desistance from crime and may be the result of different factors than those attributed to the offenders studied in the present research. Future studies should also consider stratifying criminal achievement by victim count as there may be multiple tiers of criminal productivity among efficient serial murderers.
The preponderance of offenders who aspire to upstart a serial homicide series but fail (Yaksic, Harrison, et al., 2021) highlights the need to classify serial murderers by what they were able to ‘accomplish’ over a period so that they may be studied appropriately. From the criminal achievement perspective, the study of offending productivity has been shown to be useful in the understanding of the criminal career (Bouchard & Nguyen, 2010; Lussier et al., 2011; Tremblay & Morselli, 2000). Arguably, the criminal achievement of a serial murderer can be measured by the duration of their killing series. That is, the number of days from the first to the final kill which transpire without the arousal of police detection or the instigation of other interventions (i.e., cost/detection avoidance). Future studies should consider comparing the findings of the current study with those related to aspiring serial murderers to determine factors that may lead to the criminal achievement of the former offender group. But the serial murderer’s criminal achievements are still fallible given that they were apprehended for their crimes. By identifying factors that contribute to a serial murderer’s series longevity, investigators will be better equipped to disrupt them and prevent subsequent homicides.
Adjorlolo, S., & Chan, H. C. (2014). The controversy of defining serial murder: Revisited. Aggression and Violent Behavior. 19, 486–491.
Allas, T., Bui, M., Carlson, B., Kasat, P., McGlade, H., Perera, S., Pugliese, J., Qwaider, A., & Singh, C. (2018). Indigenous femicide and the killing state. Deathscapes: Mapping Race and Violence in Settler States. https://www.deathscapes.org/case-studies/indigenous-femicide-and-the-killing-state
Balemba, S., Beauregard, E., & Martineau, M. (2014). Getting away with murder: A thematic approach to solved and unsolved sexual homicides using crime scene factors. Police Practice and Research, 15(3), 221-233.
Beasley, J. O. (2004). Serial murder in America: Case studies of seven offenders. Behavioral Sciences & the Law. 22(3), 395–414.
Beauregard, E., & Martineau, M. (2014a). Action–reaction? Forensic evidence analysis and forensic awareness in sexual homicide. In The Routledge international handbook of biosocial criminology (pp. 605-621). Routledge.
Beauregard, E., & Martineau, M. (2014b). No body, no crime? The role of forensic awareness in avoiding police detection in cases of sexual homicide. Journal of Criminal Justice, 42(2), 213–220.
Beauregard, E., & Martineau, M. (2016). Does the organized sexual murderer better delay and avoid detection?. Journal of interpersonal violence, 31(1), 4-25.
Bielen, S., Grajzl, P., & Marneffe, W. (2020). Blame based on one's name? Extralegal disparities in criminal conviction and sentencing. European Journal of Law and Economics, 51(3), 469–521. https://doi.org/10.1007/s10657-020-09670-6
Bouchard, M. and Nguyen, H., 2010. Is it who you know, or how many that counts? Criminal networks and cost avoidance in a sample of young offenders. Justice Quarterly, 27, 130–158.
Brantley, A. C., & Kosky, R. H. (2005). Serial murder in the Netherlands: A look at motivation, behavior and characteristics. FBI Law Enforcement Bulletin, 74(1), 26–32.
Brookman, F., Maguire, E. R., & Maguire, M. (2019). What factors influence whether homicide cases are solved? Insights from qualitative research with detectives in Great Britain and the United States. Homicide Studies. 23(2), 145-174.
Bursik, & Grasmick, H. G. (2001). Neighborhoods and crime the dimensions of effective community control. Lexington Books.
Camp, S. D., & Gaes, G. G. (2005). Criminogenic effects of the prison environment on inmate behavior: Some experimental evidence. Crime and Delinquency, 51(3), 425–442. https://doi.org/10.1177/0011128704271471
Campedelli, G. M., & Yaksic, E. (2021). Survival of the recidivistic? Revealing factors associated with the criminal career length of multiple homicide offenders. Homicide Studies, Advance online publication. https://doi.org/10.1177/10887679211010882
Chan, H.C. O., & Beauregard, E. (2016). Choice of weapon or weapon of choice? Examining the Interactions between Victim Characteristics in Single-victim Male Sexual Homicide Offenders. Journal of Investigative Psychology and Offender Profiling, 13(1), 70–88. https://doi.org/10.1002/jip.1432
Chan, H. C. O., & Beauregard, E. (2019). Prostitute homicides: A 37-year exploratory study of the offender, victim, and offense characteristics. Forensic science international, 294, 196-203.
Chan, H. C. O., & Li, F. (2020). Sexual murderers' choice of weapons: Examining sexual homicides committed by males in China. Journal of Criminal Justice, 71, Advanced online publication. https://doi.org/10.1016/j.jcrimjus.2020.101703
Chai, A. M. M., Beauregard, E., & Chopin, J. (2021). “Drop the Body”: Body Disposal Patterns in Sexual Homicide. International Journal of Offender Therapy and Comparative Criminology, 65(6-7), 692–714. https://doi.org/10.1177/0306624X20931436
Chai, A. M. M., Yaksic, E., Chopin, J. (2021). One ‘two’ many: An examination of solo perpetrator vs. team perpetrator serial homicides. Homicide Studies. Advance online publication. https://doi.org/10.1177/10887679211060030
Chopin, J., & Beauregard, E. (2021a). Body dismemberment in sexual homicide cases: Lust murder or rational decision? Psychology, Crime & Law, 27(9), 869–889. https://doi.org/10.1080/1068316X.2020.1863403
Chopin, J., & Beauregard, E. (2021b). Patterns of necrophilic behaviors in sexual homicide: A criminological perspective. International Journal of Offender Therapy and Comparative Criminology, 65(15), 1676–1699. https://doi.org/10.1177/0306624X20969947
Chopin, J., & Beauregard, E. (2019). Sexual homicide: A criminological perspective. Current psychiatry reports, 21(12), 1-9.
Chopin, J., & Beauregard, E. (2021c). Sexual homicide in France and Canada: An international comparison. Journal of Interpersonal Violence, 36(19–20). https://doi.org/10.1177/0886260519875547
Chopin, J., & Beauregard, E. (2022). Sexual sadism: Its role in the crime-commission process of sexual homicide of children. Journal of Interpersonal Violence, 37(1-2), NP350-NP374. https://doi.org/10.1177/0886260520916844
Chopin, J., Beauregard, E., & Bitzer, S. (2020). Factors influencing the use of forensic awareness strategies in sexual homicide. Journal of Criminal Justice, 71. Advance online publication https://doi.org/10.1016/j.jcrimjus.2020.101709
Cohen, L. E., Kluegel, J. R., & Land, K. C. (1981). Social inequality and predatory criminal victimization: An exposition and test of a formal theory. American Sociological Review, 46, 505–524.
Cui, J. & Qian, G. (2007) Selection of working correlation structure and best model in GEE analyses of longitudinal data. Communications in Statistics—Simulation and Computation, 36(5), 987–996. https://doi.org/10.1080/03610910701539617
DeFronzo, J., Ditta, A., Hannon, L., & Prochnow, J. (2007). Male Serial Homicide: The Influence of Cultural and Structural Variables. Homicide Studies. 11(1), 3-14.
DeLisi, M., & Scherer, A. M. (2006). Multiple homicide offenders: Offense characteristics, social correlates, and criminal careers. Criminal Justice and Behavior. 33(3), 367–391.
DeLisi, M., Tahja, K., Drury, A. J., Caropreso, D., Elbert, M., & Heinrichs, T. (2017). The criminology of homicidal ideation: Associations with criminal careers and psychopathology among federal correctional clients. American Journal of Criminal Justice, 42(3), 554-573.
DeLisi, M., Bunga, R., Heirigs, M. H., Erickson, J. H., & Hochstetler, A. (2019). The Past Is Prologue: Criminal Specialization Continuity in the Delinquent Career. Youth Violence and Juvenile Justice, 17(4), 335-353.
Demuth, S., & Steffensmeier, D. (2004). Ethnicity effects on sentence outcomes in large urban courts: Comparisons among White, Black, and Hispanic defendants. Social Science Quarterly, 85(4), 994-1011.
Deepak, S. A., & Ramdoss, S. (2021). The life-course theory of serial killing: A motivation model. International Journal of Offender Therapy and Comparative Criminology, 65(13-14), 1446–1472. https://doi.org/10.1177/0306624X20981030
Dietz, P. E., Hazelwood, R. R., & Warren, J. (1990). The sexually sadistic criminal and his offenses. Bulletin of the American Academy of Psychiatry and the Law, 18(2), 163-178.
Ensslen, K., Beauregard, E., & Pedneault, A. (2018). An examination of the home-intruder sex offender. International Journal of Offender Therapy and Comparative Criminology, 62(14), 4694-4713. https://doi.org/10.1177/0306624X18778450
Gautron, V., & Retière, J.-N. (2016). La décision judiciaire : jugements pénaux ou jugements sociaux ? Mouvements, 88(4), 11–18. https://doi.org/10.3917/mouv.088.0011
Green, J. A. (2021). Too many zeros and/or highly skewed? A tutorial on modelling health behaviour as count data with Poisson and negative binomial regression. Health Psychology and Behavioral Medicine, 9(1), 436-455. https://doi.org/10.1080/21642850.2021.1920416
Harbort, S., & Mokros, A. (2001). Serial murderers in Germany from 1945 to 1995. Homicide Studies, 5(4), 311–334.
Hardin, J.W., & Hilbe, J.M. (2003). Generalized Estimating Equations. Chapman & Hall/CRC.
Hewitt, A., & Beauregard, E. (2014). Sexual crime and place: The impact of the environmental context on sexual assault outcomes. Journal of Criminal Justice, 42(5), 375-383. https://doi.org/10.1016/j.jcrimjus.2014.05.003
Hewitt, A., Beauregard, E., & Davies, G. (2012). “Catch and release”: Predicting encounter and victim release location choice in serial rape events. Policing: An International Journal of Police Strategies & Management, 35(4), 835-856. https://doi.org/10.1108/13639511211275814
Hickey, E. (2015). Serial murderers and their victims. 7th Edition. Cengage Learning.
Hindelang, M. J., Gottfredson, M. R., & Garofalo, J. (1978). Victims of personal crime: An empirical foundation for a theory of personal victimization. Ballinger.
Hodgkinson, S., Prins, H., & Stuart-Bennett, J. (2017). Monsters, madmen…and myths: A critical review of the serial killing literature. Aggression and Violent Behavior, 34, 282-289.
Horan, L., & Beauregard, E. (2018). Sexual violence against marginalized victims: Choice of victim or victim of choice?. Victims & Offenders, 13(2), 277-291.
Hosmer, D. W., & Lemeshow, S. (2013). Applied logistic regression (2nd ed.). John Wiley & Sons.
James, J., & Beauregard, E. (2020). Murderer vs investigator: Factors influencing the resolution of sexual homicide cases. Police Practice and Research, 1-15.
James, J., Beauregard, E., & Proulx, J. (2019). Sexual murderers in everyday life. Journal of Criminal Justice, 60, 64-73.
James, J., & Proulx, J. (2016). The modus operandi of serial and non-serial sexual murderers: A systematic review. Aggression and Violent Behavior, 31, 200-218.
Johnston, G., & Stokes, M. (1997). Applications of GEE methodology using the SAS system. Proceedings of the 10th Annual NorthEast SAS Users Group, 881-886.
Kraemer, G. W., Lord, W. D., & Heilbrun, K. (2004). Comparing single and serial homicide offenses. Behavioral Sciences & the Law, 22(3), 325-343.
Lauritsen, & White, N. A. (2001). Putting violence in its place: The influence of race, ethnicity, gender, and place on the risk for violence. Criminology & Public Policy, 1(1), 37–60. https://doi.org/10.1111/j.1745-9133.2001.tb00076.x
Lee, J. & Reid, S. (2018). Serial killers and their easy prey. Contexts, 17(2), 46-51.
LePard, D., Demers, S., Langan, C., & Rossmo, D. K. (2015). Challenges in serial murder investigations involving missing persons. Police Practice and Research, 16(4), 328–340.
Liang, K.-Y., & Zeger, S. L. (1986). Longitudinal data analysis using generalized linear models. Biometrika, 73(1), 13-22.
Liem, M. (2013). Homicide offender recidivism: A review of the literature. Aggression and Violent Behavior, 18(1), 19-25.
Lussier, Bouchard, M., & Beauregard, E. (2011). Patterns of criminal achievement in sexual offending: Unravelling the “successful” sex offender. Journal of Criminal Justice, 39(5), 433–444. https://doi.org/10.1016/j.jcrimjus.2011.08.001
Lynes, A., Kelly, C., & Uppal, P. K. S. (2019). Benjamin’s ‘flâneur’and serial murder: An ultra-realist literary case study of Levi Bellfield. Crime, Media, Culture, 15(3), 523-543.
McNamara, J. J., & Morton, R. J. (2004). Frequency of serial sexual homicide victimization in Virginia for a ten-year period. Journal of Forensic Science, 49(3), 529–533.
Menard, S. (2002). Applied logistic regression analysis. (2nd Ed.). Sage.
Morton, R. J., & McNamara, J. J. (2005). Serial murder. In J. Payne-James (Ed.), Encyclopedia of forensic and legal medicine (pp. 47–53). Elsevier.
Morton, R. J., Tillman, J. M., & Gaines, S. J. (2014). Serial murder: Pathways for investigations. Quantico, VA: Federal Bureau of Investigation.
Mott, N. (1999). Serial murder: Patterns in unsolved cases. Homicide Studies, 3(3), 241–255.
Osborne, J. R., & Salfati, G. (2015). Re-conceptualizing “Cooling-Off Periods” in serial homicide. Homicide Studies, 19(2), 188–205.
Pakkanen, T., Zappala, A., Bosco, D., Berti, A., & Santtila, P. (2015). Can hard-to-solve one-off homicides be distinguished from serial homicides? Differences in offence behaviors and victim characteristics. Journal of Criminal Psychology, 5(3), 216–232.
Pan, W. (2001). Akaike's information criterion in generalized estimating equations. Biometrics, 57(1), 120-125. https://doi.org/10.1111/j.0006-341X.2001.00120.x
Peguero, A. A., Popp, A. M., & Koo, D. J. (2015). Race, ethnicity, and school-based adolescent victimization. Crime & Delinquency, 61, 323–349.
Peguero, A. A., Portillos, E. L., Hong, J. S., González, J. C., Kahle, L. L., & Shekarkhar, Z. (2013). Victimization, urbanicity, and the relevance of context: school routines, race and ethnicity, and adolescent violence. Journal of Criminology, 2013, 1–14. https://doi.org/10.1155/2013/240637
Prentky, R. A., Burgess, A. W., Rokous, F., Lee, A., Hartman, C., Ressler, R., & Douglas, J. (1989). The presumptive role of fantasy in serial sexual homicide. American Journal of Psychiatry, 146(7), 887-891.
Quinet, K. (2011). Prostitutes as victims of serial homicide: trends and case characteristics, 1970-2009. Homicide Studies, 15(1), 74-100.
Reale, K., & Beauregard, E. (2019). Body recovery after the “first 48”: Implications for sexual homicide investigations. Homicide studies, 23(2), 126-144.
Reale, K., Beauregard, E., & Martineau, M. (2017). Is investigative awareness a distinctive feature of sexual sadism? Journal of Interpersonal Violence, 35(7-8), 1761-1778.
Ressler, R. K., Burgess, A. W., & Douglas, J. E. (1988). Sexual homicide: Patterns and motives. Lexington. Lexington Books.
Rossmo, D. K. (2009). Introduction. In D. K. Rossmo (Ed.), Criminal investigative failures (pp. 3–8). CRC Press.
Salafranca Barreda, D. (2021). The application of Newton and Swoope's geographical profile to serial killers. Journal of Investigative Psychology and Offender Profiling, 18(1), 68-78.
Salfati, G., & Bateman, A. L. (2005). Serial homicide: An investigation of behavioural consistency. Journal of Investigative Psychology and Offender Profiling, 2(2), 121–144.
Salfati, G., & Park, J. (2007). An analysis of Korean homicide crime-scene actions. Journal of Interpersonal Violence, 22(11), 1448–1470. https://doi.org/10.1177/0886260507305716
Salfati, G., Horning, A. M., Sorochinski, M., & Labuschagne, G. N. (2015). South African serial homicide: consistency in victim types and crime scene actions across series. Journal of Investigative Psychology and Offender Profiling, 12(1), 83–106. https://doi.org/10.1002/jip.1428
Sampson, R. J. & Lauritsen, J. L. (1997). Racial and ethnic disparities in crime and criminal justice in the United States. Crime and Justice, 21, 311-374.
Sane, M. R., Mugadlimath, A. B., Farooqui, J. M., Janagond, A.B., & Mishra, P. K. (2017). Serial murder: An unusual stereotype. Medico-Legal Journal, 85(4), 190-193.
Schlesinger, L. B. (2008). Compulsive-repetitive offenders: behavioral patterns, motivational dynamics. In R.N. Kocsis (Ed.), Serial murder and the psychology of violent crimes (pp. 15-34). Humana Press.
Schlesinger, L. B., Ramirez, S., Tusa, B., Jarvis, J. P., & Erdberg, P. (2017). Rapid-sequence serial sexual homicides. The Journal of the American Academy of Psychiatry and the Law, 45(1), 72–80.
Schreck, C.J., Wright, R. A., & Miller, J. M. (2002). A study of individual and situational antecedents of violent victimization. Justice Quarterly, 19(1), 159–180. https://doi.org/10.1080/07418820200095201
Shanafelt, R., & Pino, N. W. (2012). Evil and the common life: Towards a wider perspective on serial killing and atrocities. In S. Winlow & R. Atkinson (Eds.), New Directions in Crime and Deviancy (1st ed.) (pp. 252-273). Routledge.
Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas. University of Chicago Press
Sorochinski, M., Salfati, G., & Labuschagne, G. (2015). Classification of planning and violent behaviors in serial homicide: A cross‐national comparison between South Africa and the US. Journal of Investigative Psychology and Offender Profiling, 12(1), 69-82.
Spano, R., & Nagy, S. (2005). Social guardianship and social isolation: An application and extension of lifestyle/routine activities theory to rural adolescents. Rural Sociology, 70(3), 414-437. https://doi.org/10.1526/0036011054831189
Stefanska, E. B., & Carter, A. J. (2019). Whiter than white: the art of delaying detection in sexual killers. International journal of offender therapy and comparative criminology, 63(9), 1825-1837.
Sturup, J. (2018). Comparing serial homicides to single homicides: A study of prevalence, offender, and offence characteristics in Sweden. Journal of Investigative Psychology and Offender Profiling, 15(2), 75-89.
Synnott, J., Bakker, M., Ioannou, M., Canter, D., & Van der Kemp, J. (2019). Crime location choices: A geographical analysis of German serial killers. Journal of Investigative Psychology and Offender Profiling, 16(2), 110-123.
Tabachnick, B. G., & Fidell, L. S. (2019). Using multivariate statistics (7 ed.). Pearson.
Tremblay, P., & Morselli, C. (2000). Patterns in criminal achievement: Wilson and Abrahamse revisited. Criminology, 38, 633–659.
Trojan, C., Salfati, C. G., & Schanz, K. (2019). Overkill, we know it when we see it: Examining definitions of excessive injury in homicide research. Journal of Criminal Psychology. 9(2), 61-74.
Urban Indian Health Institute. (2018). Missing and murdered Indigenous Women and Girls. https://www.uihi.org/wp-content/uploads/2018/11/Missing-and-Murdered-Indigenous-Women-and-Girls-Report.pdf
U.S. Department of Justice. (1993). Age-specific arrest rates and race-specific arrest rates for selected offenses, 1965-1992. https://ncjrs.gov/pdffiles/Digitilization/148356NCJRS.pdf
U.S. Federal Bureau of Investigation. (2008). Serial murder: Multi-disciplinary perspectives for investigators. https://www.fbi.gov/stats-services/publications/serial-murder
U.S. Sentencing Commission. (2011). Report to the Congress: Mandatory minimum penalties in the federal criminal justice system. https://www.ussc.gov/research/congressional-reports/2011-report-congress-mandatory-minimum-penalties-federal-criminal-justice-system
Vieraitis, L.M., Kovandzic, T. V., & Marvell, T. B. (2007). The criminogenic effects of imprisonment: Evidence from state panel data, 1974-2002. Criminology & Public Policy, 6(3), 589–622. https://doi.org/10.1111/j.1745-9133.2007.00456.x
Walters, B. K., Drislane, L. E., Patrick, C. J., & Hickey, E. W. (2015). Serial murder: facts and misconceptions. Science in the Courtroom. https://patrickcnslab.psy.fsu.edu/wiki/images/5/54/Walters2015.pdf
Wermink, H., van Wingerden, S., van Wilsen, J., & Nieuwbeerta, P. (2018). Studying ethnic disparities in sentencing: The importance of refining ethnic minority measures. In J. T. Ulmer & M. S. Bradley (Eds.), Handbook on punishment decisions: Locations of disparity (pp. 239-263). Routledge, Taylor & Francis Group.
Yaksic, E. (2015). Addressing the challenges and limitations of utilizing data to study serial homicide. Crime Psychology Review, 1(1), 108–134.
Yaksic, E. (2018). The folly of counting bodies: Using regression to transgress the state of serial murder classification systems. Aggression and Violent Behavior, 43, 26-32.
Yaksic, E. (2019). Moving Past Sporadic Eruptions, Discursive Killing, and Running Amok: Recognizing the Convergence of the Serial and Spree Killer. Journal of Criminal Psychology, 9(3), 138-146.
Yaksic, E. (2022). Serial Murder. Encyclopedia of Forensic Sciences. Academic Press. Forthcoming.
Yaksic, E., Allely, C.S., De Silva, R., Smith-Inglis, M., Konikoff, D., Ryan, K., Gordon, D., Denisov, E. & Keatley, D. (2019). Detecting a decline in serial homicide: Have we banished the devil from the details? Cogent Social Sciences, 5(1), 1-23.
Yaksic, E., Bulut, T., Drakulic, C., Mooney, R., De Silva, R., Geyer, P., Wills, A., & Ranger, R. (2021). How much damage do serial homicide offenders wrought while the innocent rot in prison? A tabulation of preventable deaths as outcomes of sentinel events. Psychology, Crime & Law, 27(1), 76-88.
Yaksic, E., Harrison, M., Konikoff, D., Mooney, R., Allely, C., De Silva, R., Smith-Inglis, M., Matykiewicz, B., Giannangelo, S., Daniels, S. & Sarteschi, C. M. (2021). A heuristic study of the similarities and differences in offender characteristics across potential and successful serial sexual homicide offenders. Behavioral Sciences & the Law, 39(4), 428-449.
Yaksic, E., Simkin, M. & Roychowdhury, V. (2021). Statistical study of time intervals between murders for serial killers. Journal of Criminal Justice, 73, 1-10.
Zeger, S. L., & Liang, K.-Y. (1986). Longitudinal data analysis for discrete and continuous outcomes. Biometrics, 121-130.
Duration of serial homicide series