Due to the magnitude of the human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) epidemic, and its profound impact on public health and social structures, an emphasis on the behavioral, social, and cultural factors associated with sexual risk and its relation to HIV transmission has been essential. However, an unanticipated artifact of disease-focused research is that much of the contemporary knowledge related to sexual behavior has been constructed in the context of HIV and other sexually transmitted infections. Recent sexual health research has made meaningful contributions to scientific understandings; however, scientists still know relatively little about sexuality in the general population in comparison with other aspects of health and human behavior.
The void in scientific understanding has allowed sexual behaviors that are incongruent with dominant social norms to be constructed as pathologies despite a lack of empirical evidence to support proposed links between behavioral variations and negative outcomes. A vivid example of this phenomenon can be seen in the social construction of sexual addiction, which proposes a threshold where sexual behavior becomes a clinical disorder. Although there has been substantial debate and controversy surrounding conceptualization of sexual behavior as an addictive phenomenon, the concept has been widely studied and measured in sexological, psychological, and public health research. Numerous sexual scientists, clinicians, and support groups have increased awareness of, and treatments for, sexually compulsive behavior. However, studies of sexual compulsivity and its associations with sexual risk behavior have primarily assessed individuals who are already considered high risk for adverse sexual health outcomes due to other characteristics (e.g., substance use, higher number of sexual partners). Furthermore, systematic documentation of adverse outcomes associated with sexual compulsivity is lacking.
Recently, there has been a conceptual shift away from addiction models, and hypersexuality is now the dominant focus of the clinical and research literature. Variations in the definition of hypersexuality exist; however, the common theme is that sexual behavior, including thoughts or fantasies, is excessive. The most recent iteration of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) did not include hypersexuality as a diagnostic category despite some evidence to support the reliability and validity of proposed diagnostic criteria. Currently, critical questions remain unanswered and controversy persists. Ongoing conceptual and methodological challenges make systematic research difficult. The scientific and clinical communities will need to address gaps in the literature through well-designed studies that test conceptual frameworks with rigorous methods and sound measures.
History of Sexual Addiction and Sexually Compulsive Behavior
The earliest descriptions of sexually compulsive behavior can be traced to Greek myths describing satyrs and the god Dionysius. The term “nymphomania,” historically used to describe female sexual excess, is also derived from Greek. The 19th century term “Don Juanism” was used in reference to male sexual excess. In the late 19th century, Krafft-Ebing presented one of the first clinical case studies describing hypersexuality and its effects on life functioning.
Sexual behavior viewed as an addictive or compulsive phenomenon is relatively recent. During the mid-to-late 20th century, published case reports described similar clinical presentations of individuals reporting out-of-control sexual behaviors. However, there was inconsistency in the terms applied as well as conceptualizations of etiology. a
a References 5, 23, 28, 30, 43, 61, 67, 79.Labels ranged from historical terms (e.g., nymphomania and Don Juanism) to moral reflections (e.g., perversions), to clinical terminology (e.g., paraphilias, compulsive sexual behavior, impulse control disorders, sexual addiction, and sexual compulsivity).
Currently, there is no universal agreement on terminology or a single accepted definition. The term hypersexuality dominates the contemporary literature but words such as sexual compulsivity, sexual impulsivity, and sexual addiction are still commonly used to reference sexual behavior that is beyond an individual’s control, leading to impairment in life functioning and negative outcomes. b
b References 11, 16, 17, 24, 33, 39, 49, 58, 72.No single treatment strategy is widely accepted; rather, approaches range from individual cognitive and behavioral therapies, to the use of psychotropic medications, to group counseling..
Sex Outside the Norm
The idea that sexual behavior can exceed a threshold to become a clinical disorder has appeared in the scientific literature with increased frequency over the past 30 years. Prior to the emergence of HIV/AIDs in the late 1970s and early 1980s, interest in the phenomenon of out-of-control sexual behavior was limited primarily to researchers and clinicians from psychiatry, psychology, and medicine. In the early 1980s, the terms “sex addict” and “sexually compulsive” appeared in popular culture and were used to reference individuals whose sexual behaviors rested outside of accepted sociocultural norms. The emergence of HIV brought attention to sexual behaviors that increased the likelihood for transmission and sexual behavior perceived to be beyond an individual’s control was identified as a risk factor.
In the decades following the initial HIV crisis, there has been a rapid proliferation of research examining the etiology, consequences, and approaches to treating out-of-control sexual behavior. Much of the literature has resulted from research focused on sexual risk-taking behaviors among gay men, pedophilic clinical samples, and self-identified sex addicts. Fewer studies have examined subclinical levels of out-of-control sexual behavior and nonclinical populations. Research has largely ignored women and populations at lower risk for HIV/sexually transmitted infection (STI). As a result, disease-focused models emphasizing the link between sexual behavior at the higher end of the behavioral continuum and adverse sexual health outcomes (e.g., HIV/STI) dominate the research literature. One consequence has been that variation in sexual behavior remains seated within a disease paradigm without sufficient acknowledgment of the methodological limitations and conceptual biases on which findings have been based.
Although documentation linking out-of-control sexual behavior to adverse psychological and sexual health outcomes exists, the construct remains controversial within the scientific community. Scholars have argued that conceptualizations of normative sexual behavior are influenced by cultural and historical understandings that reflect sociocultural mores governing behavior. Social and psychological theories are a reflection of existing norms that are unique to time and place, and much of the debate surrounding sexual compulsivity centers on the ambiguities in the definition of “out-of-control.” Historically, the universal standard for identification was behavioral frequency. In 2004, Bancroft and Vukadinovic wrote a critical review calling for scientific evidence that out of control sexual behavior is qualitatively different from normative sexual behavior that occurs at the high end of the continuum. In their critique, Bancroft and Vukadinovic maintained that it is negligent to assume that engaging in frequent sexual activity is inherently risky or problematic without documenting the occurrence of negative consequences.
Other scientists have argued that perceptions of control over sexual behavior are social constructions, and that the importance and meaning of out-of-control models might reflect broader values related to self-control and self-consciousness that are unique to the American culture. It has been suggested that diagnosis, and subsequent labeling, reflect attempts to pathologize and medicalize variations in sexual behavior. Arguments in favor of this perspective often cite the fact that homosexuality was listed as a mental disorder in the Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association until the late 1970s. This point is central to understanding the lack of consensus in the scientific community given the rapidly emerging possibilities for the expression of sexuality and the diverse range of sexualities that exist in contemporary society. There remains a need for research that takes into account that sexual behaviors and norms vary among individuals and cultural groups. Indeed, what may be viewed as problematic for one individual, or within one culture, may be normative for another. The wide variation in contemporary sexual behavior, including behavioral frequency, makes it critical to systematically link behavior to adverse outcomes. Understanding the influence of sociocultural factors will allow scientists and clinicians to avoid errors in the diagnosis and treatment of problematic sexual behaviors.
Despite some research linking out-of-control sexual behavior to adverse psychological and sexual health outcomes, the DSM-5 contains no specific diagnostic criteria or category for classification. A categorization and diagnostic criteria for Hypersexual Disorder were proposed by Kafka in 2010. Kafka characterized hypersexuality as a behavioral pattern involving repetitive and intense preoccupation with sexual fantasies, urges, and behaviors, leading to adverse consequences and clinically significant distress or impairment in social, occupational, or other important areas of functioning. Kafka also hypothesized that hypersexual individuals usually experience multiple unsuccessful attempts to control or reduce the amount of time spent engaging in sexual fantasies, urges, and behaviors in response to dysphoric mood states or stressful life events. Consistent with DSM criteria for other disorders, it was proposed that symptoms must be present for at least 6 months and occur independent of substance use, mania, or a medical condition in order for a diagnosis to be established. Kafka recommended that an operational definition should be derived from large, nonclinical community samples where a range of normative sexual behaviors can be examined. He emphasized the need to consider demographic characteristics (e.g., age, gender, education, culture) when contextualizing sexual behaviors.
The results of a field trial designed to assess the reliability and validity of the diagnostic criteria proposed by Kafka reported high interrater reliability and stability of the criterion over time. Sensitivity and specificity indices showed that criteria accurately reflected presenting problems and the diagnostic criteria demonstrated acceptable validity. Based on these findings, the Sexual and Gender Identity Disorders Working Group recommended the inclusion of hypersexual disorder as a diagnostic category; however, it was ultimately excluded from the final publication of the DSM-5. Critics of the proposed category argued that the concept of hypersexuality resulted from moral norms and psychosocial values that should have no place diagnostic decision-making.
Currently, diagnostic categorization lies in the hands of the practitioner. Options fall within one of three major categories: paraphilia, either one or more specifically identified or paraphilia not elsewhere classified; impulse control disorder not elsewhere classified; sexual disorder not elsewhere classified.
Despite the lack of specificity in diagnostic categorization, there is consensus among researchers regarding the relationship between hypersexuality and psychiatric comorbidity The literature reports consistent associations between out-of-control sexual behavior and psychological impairment and substance use disorders. a
a References 26, 46, 50, 69, 70, 86, 88.Scientists have begun to document comorbidities in nonclinical populations, community samples, and among groups at lower risk for HIV.
Raymond and colleagues reported that mood and anxiety disorders were the most consistent diagnoses among their community sample of 23 men and 2 women who self-identified as hypersexual, with 80% meeting the diagnostic criteria for an Axis I disorder at the time of data collection and 100% meeting the criteria across their lifetime. Research has also established a link between hypersexuality and trait variables associated with psychological impairment. A study examining personality, psychological, and sexuality trait variables among 510 heterosexual, bisexual, and homosexual women and men who self-reported hypersexual behavior found that hypersexual behavior was related to depressed and anxious mood states and trait impulsivity. Furthermore, higher neuroticism and lower agreeableness significantly predicted hypersexual behavior. Research examining sexual compulsivity in a sample of 235 women found that hypersexual behavior was predicted by psychoticism. A similar study conducted among men ( N = 152) reported finding that psychoticism, neuroticism, and agreeableness were significant predictors of hypersexual behavior. Pachankas and colleagues (2014) established a relationship between maladaptive cognitions and hypersexuality among gay and bisexual men.
A strong and consistent association between substance use and hypersexuality has been established. Grov et al. (2010) found that higher sexual compulsivity scores were associated with the use of ketamine, MDMA (ecstasy), gamma hydroxybutyrate (GBH), cocaine, and methamphetamines among a community-based sample of 1214 gay and bisexual men. Kalichman and Cain found that higher levels of sexual compulsivity were linked to higher usage rates of alcohol, powder cocaine, crack cocaine, and inhalants. Sutton et al. (2015) conducted chart reviews of patients referred for hypersexuality and found that there were more substance abuse disorders among those who were paraphilic when compared to other subtypes.
The findings reported above are only a few examples of research documenting the association between psychiatric comorbidity and out-of-control sexual behavior. Numerous studies conducted over the past two decades have consistently found similar links. The observed associations between out-of-control sexual behavior and psychiatric comorbidity have caused some scientists to question whether associations provide evidence that out-of-control sexual behavior is a unique psychiatric disorder, or whether it should be conceptualized as a behavioral symptom of an underlying condition. Although it is possible that out-of-control sexual behavior does warrant its own classification, it is also possible that individuals with other disturbances use sex as a means of self-medication to alleviate or temporarily escape discomfort caused by underlying distress. Research has shown that nonclinical negative mood states influence sexual interest among some individuals. The Dual Control Model (DCM) proposes that sexual arousal is influenced by two distinct psychophysiological systems: sexual excitation and sexual inhibition. In this model, individuals with low inhibition and a high propensity for excitation may be more prone to problems of hypersexuality. Systematic investigation of the DCM as a possible explanation for hypersexuality in some individuals suggests that stress and dysphoric mood state may trigger hypersexual behavior. These findings lend support to the hypothesis that some hypersexual behavior may represent attempts to escape psychological discomfort through sexual behavior.
Recent scientific attention has focused on examining psychological traits that may predispose individuals to hypersexuality. As interest in documenting the phenomenon among nonpatient populations has increased, there has been a shift toward understanding how structural dimensions of personality may influence hypersexual behavior. Prior research on the relationship between personality and addictive behavior has demonstrated positive associations among certain characteristics. Studies examining hypersexuality and sexual risk-taking have also demonstrated significant associations between specific personality traits and out-of-control sexual behavior. The Big Five model of personality developed by McCrea & Costa (1987) proposes five distinct structural dimensions of personality: Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness. Although there have been inconsistencies in the strength of relationships between personality traits and hypersexuality, studies have repeatedly found that higher levels of neuroticism and extraversion and lower levels of conscientiousness and agreeableness predict hypersexual behavior.
When examining the evidence as a whole, there is support for the idea that there may be different types of sexual compulsivity that derive from different factors. Mood state might account for hypersexual behavior among some individuals, whereas others may be more prone to out-of-control behavior due to personality traits. A proportion of people may be using sexual behavior as a form of self-medication that allows them to alleviate negative emotions for some period of time. Observed comorbidities suggest that there may be some underlying mechanism causing dysregulation in both sexual behavior and mood. Additional research among diverse clinical and community samples will add specificity to our understandings of the mechanisms that cause and support hypersexual behavior. Debate about the construct is ongoing, with many scientists favoring the perspective that hypersexuality is a natural and harmless variation in sexual behavior, while others remain convinced that it is a psychiatric condition that warrants inclusion in the DSM.
The ongoing debate regarding the construct of sexual compulsivity has resulted in varied perspectives on etiology. Many researchers have attempted to explain the development and maintenance of hypersexuality, which has resulted in several theoretical representations of causation. Early theoretical explanations include Carnes’ Addiction Model, Coleman’s Compulsive Sexual Behavior Model, and Kalichman’s Impulse Control Model. More recent additions to the literature include Bancroft’s Dual Control Model and Parson’s Syndemic Model. Parsons’ model, which aims to explain hypersexuality in gay and bisexual men, is still in the early stages of testing and will not be reviewed here. However, preliminary evidence suggests that a three-group categorization of sexual compulsivity may be one systemic factor that explains HIV risk among gay and bisexual men. Each perspective suggests etiological explanations; however, overwhelming empirical evidence favoring any of these is markedly absent.
Carnes’ Addiction Model
Carnes views sexual addiction as a chronic illness and has defined it as an extremely intense sex drive or obsession with sex. From this perspective, sex becomes the most important need and drives the individual’s behaviors. Carnes operationally defined sex addiction as a pathological relationship with a mood-altering experience. According to Carnes, the hallmark of sexual addiction is the lack of ability to control sexual feelings, thoughts, and behaviors. Rather than sex being a pleasurable act, Carnes asserts that for addicts, sex becomes a tool to ameliorate pain and relieve stress. He believes that the fear of abandonment and shame are at the core of sexual addiction. Carnes likens the biological, neurological, and physiological responses that result from sexual stimulation to responses resulting from the consumption of alcohol and other drugs.
Carnes advocates a 12-step treatment approach adapted from the Alcoholics Anonymous model. The underlying premise is that individuals are powerless over the amount or type of sexual behavior in which they engage. Recovery is only possible when the individual has successfully progressed through the 12-Step process.
Coleman’s Compulsive Sexual Behavior Model
Coleman first introduced the Compulsive Sexual Behavior (CSB) model of sexual compulsivity in 1990. Coleman theorized that CSB is a disorder characterized by intense sexually arousing fantasies, urges, and associated sexual behaviors that are intrusive, driven, and repetitive. Individuals are described as lacking control over their sexual behavior, which they may perceive as excessive. Often these individuals experience serious comorbid symptoms and associated consequences including mood disorders, somatic complaints, substance abuse or dependency, HIV or other sexually transmitted infections, unwanted pregnancy, relationship problems, domestic violence, sexual dysfunction, or child abuse. Furthermore, compulsive sexual behavior may lead to ethical, social, and legal problems, in addition to psychological distress.
Coleman conceptualizes CSBs as fitting into one of two distinct categories, either paraphilic or nonparaphilic. Whereas paraphilic CSB comprises nonnormative sexual behavior that involves both distress and recurrent fantasies, nonparaphilic CSB involves the excessive and compulsive engagement in normative sexual behaviors. The DSM-5 classifies eight paraphilic disorders: exhibitionistic disorder, fetishistic disorder, frotteuristic disorder, pedophilic disorder, sexual masochism disorder, sexual sadism disorder, transvestic disorder, and voyeuristic disorder. Although there have been attempts to declassify many of these disorders, Coleman and colleagues rely on the criteria of distress and impaired functioning to distinguish between varied sexual interests and actual psychiatric disorders. According to Coleman, paraphilic behaviors inherently impair the ability to form reciprocal love relationships and achieve a sense of well-being, thus supporting underlying pathology.
The term “nonparaphilic compulsive sexual behavior” has been used by Coleman to refer to typical sexual behavior occurring at the high end of the behavioral continuum. This conceptualization is most consistent with the phenomenon under exploration in this chapter. There is no clear diagnostic category for nonparaphilic CSB listed in the DSM-5. Nonparaphilic CSB is thought to have at least seven subtypes: compulsive cruising and multiple partners, compulsive fixation on an unattainable partner, compulsive autoeroticism (masturbation), compulsive use of erotica, compulsive use of the Internet for sexual purposes, compulsive multiple love relationships, and compulsive sexuality within a relationship. Nonparaphilic CSB is thought to be linked to a variety of adverse outcomes similar to those associated with paraphilic CSB.
Coleman and colleagues acknowledge that there is no clear division between subclinical symptoms and clinically significant conditions where diagnoses can be applied. Individuals may experience problematic sexual behaviors without meeting the clinical threshold for CSB. Despite the lack of clear demarcation, Coleman et al. provide clinical criteria for making a CSB diagnosis. According to the guidelines, criteria for CSB are met when the individual has recurrent and intense normophilic or paraphilic sexually arousing fantasies, sexual urges, and behaviors that cause clinically significant distress in social, occupational, or other areas of functioning; and these fantasies, sexual urges, and behaviors cannot be accounted for by another medical condition, substance use disorder, Axis I or II disorder, or developmental disorder. Furthermore, gender, sexual orientation, and sociocultural norms must be taken into account.
The etiology of CSB is described as complex and likely involves a variety of physiological and psychological factors. Coleman has discussed links to both neuropsychiatric conditions such as temporal lobe lesions, epilepsy, and head trauma, as well as psychological disorders, particularly anxiety and depression. Coleman contends that neuropsychiatric causes should be considered when the onset of CSB is subsequent to a trauma, surgery, illness, or the use of a substance (prescribed or not). If neuropsychiatric causes are ruled out, it is important to consider psychological factors.
The suggested treatment for CSB is a combination of pharmacotherapy and psychotherapy. According to Coleman’s treatment paradigm, psychotherapy for CSB explores environmental and psychodynamic stressors that contribute to behavioral manifestations. Clients are taught coping mechanisms to manage stress, anxiety, and depression that might trigger problematic behavior. A group treatment approach is recommended; however, group treatments are not widely available. Because there are many types of CSB, Coleman suggests tailoring therapy to individuals within the group setting. Couples and/or family therapy, in conjunction with individual or group treatment, may help facilitate healthy sexual and relational functioning. Favored pharmacological interventions include the use of selective serotonin reuptake inhibitors (SSRIs) or naltrexone. The body of evidence supporting the efficacy of SSRIs for treating compulsive sexual behavior comes from several small sample studies and case reports. Likewise, there have been few studies examining the efficacy of naltrexone for treating compulsive sexual behavior. Instead, much of the literature supporting the efficacy of naltrexone studied a variety of other disorders, including alcoholism, cocaine abuse, eating disorders, and pathological gambling.
Kalichman’s Impulse Control Model
Kalichman’s interest in out-of-control sexual behavior grew out of the desire to understand mediating factors associated with HIV risk and resistance to adopting risk-reduction strategies. According to Kalichman, sexual compulsivity is a heterogeneous psychological construct that can include a preoccupation with sexual desires and behaviors to the degree that social, occupational, and daily life function are impaired. Kalichman denotes that his conceptualization of sexual compulsivity is not synonymous with sex addiction, hypersexuality, or other clinically defined categories. Kalichman defines sexual compulsivity as the propensity to experience sexual disinhibition and undercontrolled sexual impulses and behaviors as self-identified by the individual. In addition, Kalichman believes that sexual compulsivity most likely has multiple forms and etiologies.
Because Kalichman’s conceptualization of sexual compulsivity is nonclinical, the bulk of his work has focused on documenting the relationship between sexual compulsivity and HIV/STI risk, rather than trying to articulate theoretical underpinnings. Simply, he believes that individuals can only be identified as sexually compulsive when they self-report multiple markers of sexual preoccupation and undercontrolled sexual impulses, which are likely related to a lack of impulse control.
Bancroft’s Dual Control Model
Bancroft’s Dual Control Model (DCM) was not specifically developed to explain hypersexuality but his conceptualizations and measures have been applied in research examining the phenomenon. Bancroft believes that out-of-control sexual behavior is likely the result of multiple etiologies and a small proportion of cases may have features of obsessive-compulsive disorder. However, he proposes that individual propensity toward excitation and inhibition may be applicable in many cases of hypersexuality.
The DCM hypothesizes that individual sexual arousal is dependent on responsiveness to two separate neurophysiological systems: sexual excitation and sexual inhibition. These systems are thought to be mostly adaptive and functional, and individuals will vary in their propensity toward each. The balance between proneness toward excitation or inhibition will dictate the response.
The DMC has been used as a framework for understanding low (hypo) and high (hyper) sexual response in both men and women. Those who are less responsive to inhibitory processes and more responsive to excitatory processes are thought to be the most likely to demonstrate hypersexual behavior. Although the link between higher sexual excitation and hypersexuality is consistently supported in the literature, the link between hypersexual behavior and sexual inhibition is less clear. Studies have found that higher levels of inhibition are associated with participation in risk behavior, which is counterintuitive to hypothesized relationships. To account for the findings, researchers have suggested that individuals with high levels of inhibition may engage in behaviors associated with risk because they fear losing their sexual arousal. Bancroft notes that mood and personality traits may be important considerations when accounting for individual variation in patterns of response. For example, paradoxical patterns of high arousal in depressed individuals may lend support to the idea that some proportion of people are using sex as a form of self-medication. He also recommends the use of SSRIs in cases where depression and/or anxiety are associated with hypersexuality because medications stabilize mood and may also inhibit sexual response.
A range of measures have been developed and used in research examining sexual compulsivity. The two most consistently cited measures are the Compulsive Sexual Behavior Inventory (CSBI) and the Sexual Compulsivity Scale (SCS). The Cognitive and Behavioral Outcomes of Sexual Behavior Scale (CBOSBS) was developed in response to the lack of evidence linking sexual behaviors to negative outcomes. McBride and colleagues used outcomes proposed by the Society for the Advancement of Sexual Health in an attempt to create an outcomes-based measure that moved away from traditional focuses on behavioral frequency and sexual risk-taking. Furthermore, they sought to develop a tool appropriate for use in community samples and one that would more accurately capture the experiences of women. A newer measure, The Hypersexual Disorder Screening Inventory (HDSI), was developed by the American Psychiatric Association’s DSM-5 Workgroup committee as an instrument for screening hypersexuality during a field trial to test the reliability and validity of the diagnostic category. The measure has not been widely used and initial validation was limited to patient populations. The HDSI requires additional testing before psychometric properties can be reported with confidence. Although we will not include further discussion on the HDSI in the section below, it is important for readers to be aware of this new tool.
Compulsive Sexual Behavior Inventory
The Compulsive Sexual Behavior Inventory (CSBI) was developed in response to the need for psychometrically sound measures to identify CSB. The authors believed preexisting measures failed to incorporate all major components of the phenomenon. The CSBI was intended to create a standardized, reliable, and valid assessment tool for use in clinical and research settings. For validation purposes, it was hypothesized that no significant differences would be observed in scores obtained from those with paraphilic and nonparaphilic CSB but significant differences would be observed when the clinical group scores were compared to controls.
The preliminary study of reliability and validity included three groups: individuals diagnosed with paraphilic compulsive sexual behavior recruited from a sex-offender treatment program (N = 35); individuals with nonparaphilic compulsive sexual behavior recruited via advertisements (N = 15); control participants recruited via advertisements (N = 42). The initial inventory consisted of 42 items related to sexual control and various aspects of behavior associated with paraphilic and nonparaphilic CSB. Participants rated their responses to items on a scale ranging from 1 = Very frequently to 5 = Never. A principal components factor analysis using varimax rotation was performed on the data and factor loadings that exceeded 0.60 after rotation were retained for the final scale. The reliability of the retained factors was tested using Cronbach’s alpha and the data were tested using linear discriminant function analysis to determine the scale’s ability to differentiate individuals with CSB from controls. In addition, three analyses of variance were conducted to explore mean differences for each group on identified subscales. A plot of data from the entire sample indicated that a three-factor solution, accounting for 58% of the variance, was the best fit. The first factor explained 42%, the second factor explained 10.1%, and the third factor accounted for 5.9% of the variance. The retention of items on each factor was determined by the magnitude of factor loading and subsequent assessments of face validity. A total of 28 items were retained, and the factors appeared to measure control, abuse, and violence.
Initial tests of validity were conducted using linear discriminant function analysis, testing the scale’s ability to distinguish between groups believed to have CSB from those who did not. The classification matrix was reported to have correctly identified 92% of cases, with one normal control being incorrectly classified as compulsive and six compulsives being identified as normal. Further explorations of validity used the three subscales as independent variables in analyses of variance to explore group differences. The findings indicated significant effects for group on the control subscale. Pairwise comparisons demonstrated that pedophiles scored significantly lower on the subscale when compared with the other two groups. A significant main effect was found for the violence subscale, and subsequent pairwise comparisons showed that controls differed significantly from pedophiles. Table 36.1 reports the items contained in the CSBI.