Sex Offender Typologies and their Potential Use in Predictive Policing
Current theories of policing have begun to include evidence-based research to aid in moving away from inadequate practices in traditional policing, such as altering patrol techniques, using “gut-instinct,” and limiting contact with the public (Hegarty, Williams, Stanton and Chernoff, 2014; Lum and Koper, 2013). Evidence-based policing includes predictive policing through developing typologies for effective profiling of suspects and hot-spot mapping of offenders with those profiles. Use of scientifically established typologies allows policing agencies to narrow their focus reducing the amount of resources necessary to find perpetrators (Wilson, Smith, Markovic and LeBeau, 2009). Within sex crimes, typologies become difficult to produce due to the heterogenic nature of this criminal group (Vandiver, 2017) so multivariate models of analysis become necessary to build statistically significant profiles of sexual offenders (Fujita, Watanabe, Yokota, Suzuki, Wachi, Otsuka and Kuraishi, 2015).
Neyroud and Weisburd (2014) discuss transforming police through research, clarifying the relationship that is needed between researchers and policing entities. Forging that relationship will aid in improving policing effectiveness, time efficiency, and financial costs. It will also reduce recidivism through improved treatment and more deterrent punishment. Police officers with educations that utilize journal reviews, relevant scientific studies, and academic assessments are also discussed as a method of expanding their knowledge base.
Inadequate data quality and quantity remains a complication; improved documentation, collection and analysis by social scientists is necessary to produce useful typologies, especially regarding sex offenders (Wilson, Smith, Markovic and LeBeau, 2009). The uniqueness of each sex offender makes consistent results and external validity difficult for researchers. The purpose of this paper is multifaceted; it will explore how sex offender typologies are used in law enforcement and how common traits of sex offenders are identified in current studies. Additionally, the paper includes a discussion of research that identifies overlapping variables in sex offender types used to identify various methods of potentially improving treatment and creating more deterrent punishment to reduce recidivism and to grow predictive policing.
Typology History of Use in Law Enforcement
The creation and use of typologies has its roots in Carl Jung’s studies of personality types in his work Psychology Types in which he states:
“I came across the problem of types; for it is one's psychological type which from the outset determines and limits a person's judgment,” (Jung, 207).
Jung then developed what he called the four psychic functions that determined and limited a person’s actions: sensation, intuition, thinking and feeling. The four functions were broken into two psychological functions: perceiving and judgment. Through these two functions a person allowed one to dominate their decision making and produce either rational or irrational actions (Jung, 1989). This laid the framework for psychologists to divide criminal motivations from non-criminal and then further into typologies of offenders. These elements of criminal profiling develop through investigating the crime itself, the crime scene, the victimology, and any known history and personality traits of the potential suspects (Webb, 2013).
As an extension of Harris’s (1977) criminal justice labeling theory, typologies have been used in criminal justice since the 1970s when the Federal Bureau of Investigation (FBI) began to teach and utilize them in techniques of profiling offenders. In 1984, the FBI created the National Center for the Analysis of Violent Crime (NCAVC) in which typology development for the use of profiling was given a top-of-the-line home base. Currently, there are several Behavioral Analysis Units that work on various types of violent crime based on the typologies developed by researchers and the NCAVC (Webb, 2013). NCAVC dictates the standard that will be set for profiling techniques and typologies for most law enforcement by publishing updated scientific evaluations of those typologies (FBI, 2016).
Evidence-based policing includes deployment of improved investigative measures that are supported by research and reproducibility (Lum and Koper, 2013). Profiling is one of the techniques that deduces traits of specific offenders per their victim choice, process of perpetration, environment selection, and, if applicable, method of body disposal. Psychological, sociological, medical and criminological backgrounds of known offenders provide an opportunity for researchers to cross-analyze characteristics and formulate typologies that can be useful in suspect prioritization. Traditionally, the five-factor model was used measuring the offenders five primary psychological traits: openness, contentiousness, extroversion, agreeableness, and neuroticism. More advanced methodologies are now utilized that incorporate those five traits, as well as many others relevant to the typologies developed.
Study methods are currently developing among researchers that will improve and identify typologies more accurately. Behavioral themes of offenders are being defined by single dependent variable studies, multivariate studies and case studies. Multivariate analysis studies have been most helpful for sex offender typologies. In these studies, unique behaviors of each offender can be overlaid and statistically analyzed to develop the most useful predictive patterns (Fujita et al., 2015). An example of this is the cluster methodology used by Vandiver and Kercher in 2004 to develop typologies of female sex offenders in Texas. The study was then replicated by Sandler and Freeman in 2007 in the New York. While the outcome typologies were different between the two locales, they shared many similar characteristics. Limitations on population size and data collection in the second study where a primary factor in their distinctions (Sandler and Freeman, 2007). The importance was the replicability of the study method; it was found to be effective and produced similar meaningful results regarding female sex offender typologies. With improved data collection methods and more sophisticated analyses (i.e., factor analysis, path analysis, and multiple analysis of variance) the typologies are improved.
Methodology & Identification of Sex Offender Typologies
To date there is no specific intervention of sex offenders prior to perpetration of any offense due to the limited empirical evidence that is available. The evidence required for psychologists to diagnose high risk levels in sex offenders is generally identified after commission of their crimes. There is no scientific consensus of explanation for offender behaviors, though prevailing theories have been developed. A growing number of researchers in studies of known and recidivating offenders have assisted in developing typologies and profiles that add to the more efficient investigation of suspects in sex crimes. Currently, the best that can be accomplished is reactive identification and treatment of offenders and the development of interventions that are most effective in reducing recidivism (Cutler, 2008).
Typologies of sex offenders are extensive and inconsistent but the following is a discussion of several studies and their methods of identifying offenders per their focus on either the type of crime, victim, or psychology. The broad typologies that exist have subtypes that are classified depending on which researcher is discussing them. Traditionally, agreed-upon categories of sex offenders include child molesters, incest offenders, rapists, exhibitionists, voyeurs and frottoirists (Cutler, 2008). From those categories, researchers have identified more specific offenders; including the most commonly accepted examples: Groth’s power, anger, sadistic/ ritualistic rapists or The Massachusetts Treatment Center typologies of opportunistic, pervasively angry, sexual gratification and vindictive rapists. These are further divided into subtypes of their own such as Groth’s power rapists including both power reassurance (a.k.a.“gentleman” rapists) and power assertive rapists (Groth, 1979).
Methods of data collection include researchers hypothesizing on the likelihood of known offenders to have committed an unknown offense prior to what they were convicted of or to hypothesize on their risk of offending again in the future. Methods such as: studying prisoner volunteers through self-report, using phallometric data disclosing deviant preferences, the Hare Psychopathy Checklist (reviewed) or PCL-R, RRASOR, SCAJ-Min and Static-99 (discussed later), and psychopathy identified by specialists during interviews (Quinsey, Rice, & Harris, 1995). Other methods include statistical analysis of crime scene evidence and comparative analysis of known prior offenses - both sexual and non-sexual, or typology archetype characteristics to identify and predict recidivism categories. The following are studies that have helped improve investigative profiling techniques in law enforcement by developing useful typologies.
Identifying Rapists and Child Molester Typologies
Quinsey et al.’s (1995) study of 178 men who were rapists and/or child molesters identified that rapists were at greater risk of recidivating than child offenders. They used inferential statistical tests that compared two functions: sexual recidivism and history of violent offense. The PCL-R and phallometric data were not significant indicators of reoffending but use of violent offense history was a significant indicator at 77% for sexual offense reconviction. For identifying psychopaths versus non-psychopaths, ‘ever married’ was a weak, but significantly predictive variable. Deviant sexual interest in rape and child molesting was also a significant predictor of recidivating. After 59 months, 28% were reconvicted for a sexual offense and 40% were reconvicted for a violent offense. While the data was limited, this study identified a need for linked typologies between violent and sexual offenders to predict recidivism.
Studies generally fail to identify how deviant behaviors are developed but simply categorizing victim and offender actions; etiology of rapists and other offenders may clarify typologies. The aim of Beauregard, Lussier and Proulx’s (2004) study on adult rapists was to investigate developmental factors to describe the etiology of rape proclivity. Their population included 118 sexual aggressors against women at the Regional Reception Center maximum-security penitentiary in Quebec, Canada between 1995 and 2000. Prisoners were given a questionnaire on: developmental factors, socioeconomic, criminal and psychologic history prior to committing the crime. Additional factors regarding the commission of the crime itself for which they were convicted, impulsive/antisocial lifestyle, and current psychological characteristics were also included. (This was a pseudo-replication of 1995 Quinsey, et al. study of general sex offenders.) Multiple regression analyses via semi-structured interviews and phallometric assessments with audio stimuli (including sounds of: mutually consenting sex, physically violent rape, rape with humiliation, nonsexual physical aggression, and nonsexual nonaggressive control stimuli) were completed.
Sexually inappropriate family environments such as spousal sexual abuse between offender’s parents, sexual abuse of the offender as a child by parents, use of pornography by parents in the offender’s presence as a child or parental encouragement of sexual abuse of another were considered. Use of pornography by the offender as a child, impulsive or antisocial behavior, and deviant fantasies during childhood/adolescence correlated positively and were found to be indicators of deviant preferences for rape. Hostility towards women in childhood was also a good predictor, implicating background culture and psychopathy of parents as potential indicators of offenders for the purpose of investigative focus points for law enforcement.
Creating Juvenile Offender Typologies
Juvenile offenders make up a third of rapists and half of all child molesters. Juvenile offenders were also found to have a higher potential to recidivate as adults (Hunter, Hazelwood, and Slesinger, 2000). Treatment and detection of juvenile offenders depend on recognizing treatment needs and level of dangerousness. Recognition can be best done by identifying typologies within populations of offenders that provide indications of abuse or potential to recidivate. In Hunter, Hazelwood and Slesinger’s 2000 study including 126 juvenile sex offenders, police investigation records were coded by a research assistant into those who committed acts against children (>5 years younger than the offender) and those who acted against peers or adults. Offenders were then coded by background, coercion level and type (violent versus verbal), substance use involved, use of weapons, victim resistance level and overall violence displayed. Four predictors were used in both peer/adult and child offenders: environmental risk (socioeconomic status, parental presence and abuse, parental criminal history and drug use, etc.), influence of drugs or alcohol, planning and control, and victim difficulty. The results allowed researchers to conclude that peer and adult offenders had significantly more planning and control involved in commission of offenses compared to child molesters.
Juveniles who assaulted peers and adults generally targeted acquaintances and were less likely to be solitary where child molesters were found to be loners and more frequently abused against siblings and other relatives. In 63.9% of assaults by child molesters were perpetrated in the victim’s residence but only 45.3% of peer and adult offenses were committed at the victim’s residence. Child molesters were more likely to be Caucasian and peer/adult offenders were more likely to be African American. Peer and adult offenders almost exclusively offended against females (93.7%) while child molesters’ victims were likely to be either female (53.2%) or male. Clear typologies of juvenile sex offenders were many and distinct per Hunter, et al. (2000) therefore potentially helpful for investigators of juvenile suspects.
Differences between Serial and Non-Serial Homicide Typologies
Meta-analysis is another methodology utilized in identifying more accurate typologies. While only 2% of murderers are serial sexual murders, they hold the media’s attention and their sensationalism inspires large of amounts of moral panic (James and Proulx, 2014). Identifying the main characteristics between serial sex offenders and non-serial sex offenders is important to monitoring potential recidivism and identifying victims that may not have yet been located.
In James and Proulx’s (2014) review using two methods of data collection 845 candidate studies were yielded, 45 of which were retained meeting inclusion criteria. The two methods of data collection included online databases and manual research of study sources. Between January 1985 and March 2013, five criminal justice and psycho-social platforms were searched encompassing eighteen databases. Two keyword blocks were used: ‘homicidal acts’ (homicide, murder, kill, lethal) and ‘sexual acts’ (sexual, molest, rape, paraphilia, erotophonophilia, necrophilia) and combination terms such as sex crime, child molest, lust killing, lethal sexual assault, sex offense, sexual violence, violent crime, sexual arousal, and sadistic violence. Forty-five studies were divided into two categories: non-serial sexual murderers (NSMs) and serial sexual murderers (SSMs). Three groups were then created based on the data identified: developmental characteristics, characteristics of adult life and criminal career and psychopathology.
SSMs were more likely than NSMs to display violent behaviors or commit violent acts, set fires, and be cruel to animals, but temper tantrums were more common by non-serial than serial murderers. Nonviolent behaviors and internal problems like enuresis, sleep problems, nightmares, chronic lying, daydreaming, and compulsive masturbation were higher in SSMs than NSMs. SSMs were found to have been more likely to be neglected or psychologically abused as children, collect pornography, consume alcohol/drugs as adults, have psychiatric history, engage in promiscuous sexual behavior, and have a negative father figure. SSMs were twice as likely to be married or in an intimate relationship, over four times less likely to be unemployed and tended to specialize and be younger at first offense than NSMs. Neuro-psychopathology testing showed that SSMs were twice as likely to have normal or elevated IQ. Compulsive masturbation and fantasizing were the most predictive traits of SSMs.
Chan, Beauregard, and Myers (2015) compared single-victim and serial sexual homicide offenders as well to identify traits specific to the typologies. Sexual homicide offenders are rare so the study aimed to investigate criminal, paraphilic and personality trait differences between single victim offenders (SVOs) and serial homicide offenders (SHOs). The data source included 73 single-victim and 13 serial sexual homicide offenders convicted of sex crimes between 1994 and 2005 then sentenced to Canadian prison. Prisoner participation was voluntary and each was interviewed in private with semi-structured interviews by a psychologist and criminologist. Ten paraphilias and victim variables identified in interviews were rated with a Computerized Questionnaire for Sexual Aggressors.
SHOs were more likely than SVOs to report fantasies and have distinct characteristics, target strangers, use structured premeditation and commit verbal humiliation of their victims. They also had narcissistic, schizoid and/or obsessive-compulsive disorder (OCD) traits, and had in paraphilias like masochism, homosexual pedophilia, exhibitionism, and/or voyeurism. Most offenders were Caucasian and single with a mean age of 33.62. Eighty percent of victims were female age with a mean age of 23.15 for SVOs, and mean age of 28.9 for SHOs. SHOs were more likely to have specialized victim preference and report deviant fantasies 48 hours prior to committing sex offense making it a useful predictive trait if verbalized. These trait identifications aid in developing typologies of homicidal sex offenders that can discern between serial and non-serial sexual offenders within criminal justice contexts.
Sullivan and Sheehan (2016) identified personality trait clusters for child molesters that fit three themes: personal history and formative experiences, predisposition to engage sexually with children, and life theories or schemas that were implicit to the acceptance of child abuse. They also identified four phases that they hypothesized lead to the formulation of becoming a child molester: formative factors (early abusive trauma to them as a child.), assimilation of formative factors (acceptance of abuse as a norm), formulation of life theories and a motivation to engage children sexually.
Sullivan and Sheehan found that 86% of their study sample of 63 male child molesters did not consider their actions against children as abuse. All participants identified a formative experience involving either a childhood sexual experience, atypical socialization, exposure to neglect, physical violence or emotional abuse. Maladaptive perceptions, developmental difficulties and activated sexual arousal were identified as three themes of transition from a victim to an offender. Five common themes were developed for the formulation of maladapted life theories: distortion, entitlement, contamination, selfishness and maliciousness. Three motivations were identified in all offender participants: sexual interest in children, personal affirmation, and power and control. Power and control covered both domination in sexual gratification and intentionally sadistic behavior.
Measurement techniques for predicting risk-assessment were discussed by Hanson and Thornton (2000) in a Canadian study of the accuracy of the Rapid Risk Assessment for Sex Offender Recidivism (RRASOR) and the Structured Anchored Clinical Judgment commonly used in studies along with the Minnesota Multiphasic Personality Inventory (SACJ-Min). Hanson and Thornton further created their own Static-99 assessment, which identified and combined static factors from other tests. They relied upon sets of child molesters from three different prisons in Canada and one in the United Kingdom (U.K.). The SACJ-Min had the lowest prediction rate for Millbrook but the best at Her Majesty’s Prison. Ultimately, the Static-99 was the best predictor overall and the RRASOR and SACJ-Min were found to be comparable. The study illustrates the complications of using measurement methods consistently across offender populations.
Selections were made based on clinical files and recidivism information. A set included only male sex offenders released between 1978 and 1993 who had participated in long-term (1-3 years) treatment at the Institute Philippe Pinel in Montreal. Another set from Millbrook Correctional Centre in Ontario used recidivism information over 15 to 30 years for child molesters released between 1958 and 1974 and who were subject to only brief treatment.
Researchers at Oak Ridge Mental Health Centre, a maximum-security treatment center in Ontario, studied offender records released between 1972 to 1993 using mental health records and referrals. Researchers from Her Majesty’s Prison Service in the U.K. utilized a follow-up study done over 16 years of 536 offenders released after 1979. A minimal number received treatment of any kind. Data was coded based on previous charges rather than number of charges.
Seto (2005) discusses the use of actuarial risk scales to improve the studies of sex offenders and recidivism in a study conducted to determine whether combining those scales (as in the study by Hanson and Thornton, 2000) is necessary. Two hundred and fifteen adult male sex offenders were subjected to the Violence Risk Appraisal Guide (VRAG), the Sex Offender Risk Appraisal Guide (SORAG), the RRASOR and the Static-99 assessment scales. Despite the extensive study, no scale was found to be any more predictive of recidivism than another though all scales were significantly useful compared to psychological interviewing. If a scale can be created for psychologists and law enforcement to identify potential offenders prior to sexual or violent offenses, then predictive policing may transform the criminal justice system.
Trait Identification to Reduce Recidivism
While predictive policing of sex offender’s identities may be more of a future endeavor, creating typologies to identify high-risk sexual recidivists may improve criminal justice methods of treatment, prisoner mediation, and reconviction consequences now. Rapists are most likely to recidivate, but pedophiles and offenders who do not complete treatments also have high of recidivism rates. On average, 13-15% of offenders will commit another sexual offense and almost 50% of offenders will commit a non-sexual offense post-release (Hanson & Bussiere, 1998). Several studies and meta-analyses have been conducted to identify risk factors, but many of the findings among them conflict.
In a study about offender persistence, Miethe and Olson (2006) hypothesized that stereotypes included high recidivism rates as reported in some studies. These stereotypes have negative consequences for investigators, victims, treatment facilitators and the control of sex offenders. Notions of sex offender specialization and recidivism are inflated in these stereotypes. Other studies are inconclusive about the degree to which recidivism occurs so Miethe and Olson completed an empirical evaluation of previous research in their study. Measures of specialization and persistence are discussed from data collected via the Department of Justice and the Bureau of Justice Statistics. The sample included 10,000 of the 38,000 offenders released from prison in 15 states in 1994 along with state and FBI records regarding demographic information and criminal histories. Up to 99 arrest cycles were created to identify transitions between arrest dates and general crimes were calculated by violent offenses, property offenses, public disorder offenses. Separate analyses were calculated for sex offenses and violent sexual offenses.
Sex offender samples showed that 62% were white, 99% were males, and they had an average of 6.8 arrests in their criminal careers. They had considerably less specialization than other criminals with only 26% having committed second sex offenses; 56% were property offenders and 61% were public order offenders in their second arrest cycles. These numbers indicate that sexual recidivist stereotyping was inaccurate regarding them being persistent sexual specialists. Within the sex offender groups, child molesters were more specialized than rapists, but approximately 70% of both rapists and child molesters offended only once. Serial sex offenders exhibited wide variability in offense pattern over arrest cycles and specialization seemed to increase over age and arrest cycle but may be inflated.
Miethe and Olson’s (2006) results concluded that 26% of offenders recidivate in general crimes; rates of second convictions for sex offenses were only around 7.7% and just 9.3% committed non-sex crimes. The study followed 389 convicted sex offenders through ten arrest cycles for an average of 15.7 years. It was found that after the third arrest cycle recidivism dropped so drastically a meaningful analysis was not possible. Offenders were accurately predicted to recidivate 71.5% over the study period and non-recidivists were predicted 65.3% of the time. Eighty-three percent of recidivists did not get convicted beyond their first felony sex offense. Miethe and Olson’s key findings are two-fold; they aid in creating accurate predictive profiles and typologies and assist in eliminating offender stereotypes in criminal justice.
Hanson and Bussiere (1998) conducted a meta-analysis taken from 61 studies to identify predictors of recidivism. They found that 13.4% of offenders commit second sexual offenses. Criminal lifestyle, psychological maladjustment and sexual deviance were three factors that were focused on as predictive qualities. A predictive typology for sexual recidivists was identified in young, single, white males who had negative relationships with a mother figure and strong phallometrically measured sexual deviances. Non-sexual recidivists were more often young, male and from a minority race and were also more likely to recidivate if they did not complete treatment. Factors such as sexual fantasy and high levels of masturbation were significant moderate predictors.
Those who did not complete treatment were at higher risk for sexual recidivism, but being sexually abused was not found to be a significant predictor, as sex offender stereotypes have often implied. Several psychological maladjustment variables (alcoholism, personality disorders, depression, etc.), however, were significant predictors confirming some stereotypes. It is noted that the predictive accuracy of most variables was low and none warranted being isolated for their own study as a specific indicator of recidivism. Few dynamic traits were also studied and further study was suggested to confirm replicability; if replicable, this would give psychologists some predictive ability with patients reporting violent fantasy that may be potential sexual or violent offenders.
Researchers have found many variables that are now known to be significant in creating accurate typologies for profiling such as race, juvenile violent offense records, history of fantasizing and deviant maternal relationships; these researchers can aid in leading police and psychologists to identification of future offenders. Since juvenile violent offense is a good predictor of adult sexual offense, psychologists may identify and treat would-be sex offenders prior to their first sexual abuse, reducing the number of victims and enabling improved predictive policing.
Typologies are necessary for profiling techniques used by psychologists and law enforcement but are difficult to apply consistently due to the heterogeneity of sex offenders. Law enforcement must work together with researchers to take evidence-based theories and typologies and apply them to investigative processes and profiles. Throughout much of policing, “gut-instinct” of officers has been used and while investigators committed to it are not completely devoid of usefulness, they are not utilizing all the available resources and tools that science can provide (Nix, 2015). Development of sex offender typologies for the use of recidivism reduction and predictive preventative measures can offer a wealth of resources that can more legitimately hold up in courts of law than “gut instinct.” The future of criminal justice will require researchers to identify replicable methods and consistent data from deviant human patterns already proven to exist on most levels by psychologists and social workers. A partnership of research and science with policing agencies is the way forward for the treatment of offenders, reduction of recidivism, improvement of deterrent consequences, and the future of predictive policing and psycho-social work (Neyroud and Weisburd, 2014).
American Psychiatric Association (1994). DSM-IV: Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Association.
Aslan, D. (2011). Critically evaluating typologies of internet sex offenders: A psychological perspective. Journal of Forensic Psychology Practice, 11(5): 406–431.
Beauregard E, Lussier P, & Proulx J (2004). An exploration of developmental factors related to deviant sexual preferences among adult rapists. Sexual Abuse: A Journal of Research and Treatment 16(2): 151-161.
Bench, L. & Allen, T. (2013). Assessing sex offender recidivism using multiple measures: A longitudinal analysis. The Prison Journal, 93(4): 411-428.
Briken, P., Hill, A., Habermann, N., Kafka, MP., & Berner, W. (2010). Paraphilia-related disorders and personality disorders in sexual homicide perpetrators. Sexual Offender Treatment, 5(1): 1-7.
Chan, H.C., Beauregard, E. & Myers, W.C. (2015). Single-victim and serial sexual homicide offenders: Differences in crime, paraphilias and personality traits. Criminal Behavior and Mental Health, 25: 66-78.
Criminal Profiling. (2010). Retrieved November 16, 2016, from https://vault.fbi.gov/Criminal Profiling
Cutler, B. ed. (2008). Sex offender typologies. Encyclopedia of Psychology and Law, 2: 716-735.
Federal Bureau of Investigation (2005). Serial Murder: Multi-disciplinary perspectives for investigators, Washington, DC: U.S. Department of Justice.
Fujita, G., Kazumi, W., Kaeko, Y., Suzuki, M., Wachi, T., Otsuka, Y., & Kuraishi, H. (2015). A multivariate model for analyzing crime scene information: Predicting stranger, solo offender, and money-oriented motive of Japanese homicides. Homicide Studies, 20(4): 295-320.
Griffith, B. (2014). Translating research to practice and building capacity to use data, research, planning and problem-solving. Translational Criminology, Spring: 9-11.
Groth, A.N., & Birbaum, A.H. (1979). Men who rape: The psychology of the offender. New York: Plenum.
Hanson, R. K., & Bussiere, M.T. (1998). Predicting relapse: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66(2): 348-362.
Hanson, R. K., & Thornton, D. (2000). Improving risk assessments for sex offenders: A comparison of three actuarial scales. Law and Human Behavior, 24: 119–136.
Harris, A. (1977). Sex and theories of deviance: Toward a functional theory of deviant type-scripts. American Sociological Review, 42(1): 3-16.
Hargreaves-Cormany, H., & Patterson, T. (2016). Characteristics of survivors of juvenile sex trafficking: Implications for treatment and intervention initiatives. Aggression and Violent Behavior, 30: 32–39.
Healy, J., Beauregard, E., Beech, A., & Vettor, S. (2014). Is the sexual murderer a unique type of offender? A typology of violent sexual offenders using crime scene behaviors. Sexual Abuse: A Journal of Research and Treatment, 28(6): 512–533.
Hegarty, T., Williams, S., Stanton, S. & Chernoff, W. (2014). Evidence-based policing at work in smaller jurisdictions. Translational Criminology, Spring: 14-15.
Hunter, J.A., Hazelwood, R.R. & Slesinger, D. (2000). Juvenile-perpetrated sex crimes: patterns of offending and predictors of violence. Journal of Family Violence, 15(1): 81-93.
Irons, R. & Schneider, J. (1996). Differential diagnosis of addictive sexual disorders using the DSM-IV. Sexual Addiction & Compulsivity 1996, 3: 7-21.
Jeglic, E. L., & Culter, B. ed. (2008). Sex offender typologies. Encyclopedia of psychology and law. Thousand Oaks, CA: Sage Publications.
James, J. & Proulx, J. (2014). A psychological and developmental profile of sexual murderers: a systematic review. Aggression and Violent Behavior, 19: 592-607.
Jung, C.G., (1989). Psychological Types. The Collected Works of C.G. Jung, 6. Kindle Ed.
Langevin, R., Curnoe, S., Fedoroff, P., Bennett, R., Langevin, M., Peever, C., Pettica, R., & Sandhu, S. (2004). Lifetime Sex Offender Recidivism: A 25-Year Follow-Up Study. Canadian Journal of Criminology & Criminal Justice. 46(5): 531-552
Lee, B. (2016). Causes and cures IX: Consequences of violence. Aggression and Violent Behavior, 30: 110–114.
Lum, C. & Koper, C. (2013). Evidence-based policing. Critical Issues in Policing, 7th ed.: 260-274.
Miethe, T.D. & Olson, J. (2006). Specialization and persistence in the arrest histories of sex offenders. Journal of Research in Crime and Delinquency, 43(3): 204-229.
Neyroud, P. & Weisburd, D. (2014). Transforming police through science. Translational Criminology, Spring: 16-18
Nix, J. (2015). Predictive policing. Critical Issues in Policing, 7th ed.: 275-288.
Quinsey, V.L., Rice, M.E. & Harris, G.T. (1995). Actuarial prediction of sexual recidivism. Journal of Interpersonal Violence, 10(1): 85-105.
Rice, M. E., & Harris, G. T. (1997). Cross-validation and extension of the violence risk appraisal guide for child molesters and rapists. Law and Human Behavior, 21, 231–241.
Robertiello, G. & Terry, K. (2007). Can We Profile Sex Offenders? A Review of Sex Offender Typologies. Aggression and Violent Behavior, 12 (5): 508-518.
Sandler, J. & Freeman, N. (2007). Typology of Female Sex Offenders: A Test of Vandiver and Kercher. Sexual Abuse: A Journal of Research and Treatment, 19: 73–89.
Seto, M. C. (2005). Is more better? Combining actuarial risk scales to predict recidivism among adult sex offenders. Psychological Assessment, 17, 156–167.
Sullivan, J. & Sheehan, V. (2016). What motivates sexual abusers of children? A qualitative examination of the Spiral of Sexual Abuse. Aggression and Violent Behavior, 30: 76–87.
Tener, D., Wolnak, J., & Finkelhor, D. (2015). A Typology of Offenders Who Use Online Communications to Commit Sex Crimes Against Minors. Journal of Aggression, Maltreatment & Trauma, 24: 319–337.
Vandiver, D.M., Braithwaite, J., & Stafford, M. (2017). Sex Crimes and Sex Offenders. New York: Routledge.
Vandiver, D.M. & Kercher, G. (2004). Offender and victim characteristics of registered female sex offenders in Texas: A proposed typology of female sex offenders. Sexual Abuse: A Journal of Research and Treatment, 16: 121-137.
Webb, D. (2013). Criminal Profiling: An Introductory Guide. Amazon Publishing. Kindle Ed.: 1-21. (ISBN-13: 978-1482055436)
Wilson, R., Smith, S.C., Markovic, J, & LeBeau, J. (2009). Geospatial technology working group meeting report on predictive policing. U.S. Department of Justice. Oct 15-16: 1-13.