JUVENILE SEX OFFENDERS: PREDICTORS OF RECIDIVISM Except where reference is made to the work of others, the work described in this dissertation is my own or was done in collaboration with my advisory committee. This dissertation does not contain any proprietary or classified information. _________________________________________ Joan D. Williams Certificate of Approval: Chippewa Thomas Ren?e A. Middleton, Chair Assistant Professor Professor Counselor Education, Counseling Counselor Education, Counseling Psychology, and School Psychology Psychology, and School Psychology Suhyun Suh George T. Flowers Assistant Professor Interim Dean Counselor Education, Counseling Graduate School Psychology, and School Psychology JUVENILE SEX OFFENDERS: PREDICTORS OF RECIDIVISM Joan D. Williams A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn University May 10, 2007 JUVENILE SEX OFFENDERS: PREDICTORS OF RECIDIVISM Joan D. Williams Permission is granted Auburn University to make copies of this dissertation at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights. Signature of Author Date of Graduation iii VITA Joan Denise Williams, daughter of George C. and Helen S. Williams was born May 10, 1964 in Kailua, Hawaii. She graduated from Alabama State University in 1982 with a Bachelor of Arts degree in Journalism and Fort Valley State University with a Master of Science Degree in Mental Health Counseling. After serving for 12 years on active duty in the United States Air Force, she went on to attend Troy University, Montgomery where she worked toward an Education Specialist degree in Agency/ Community Counseling. She was accepted into the Doctoral program at Auburn University in Counselor Education in 2003. She is the proud mother of two children, Nicos DeAngelo Williams and Melody Helen Triplett. iv DISSERTATION ABSTRACT JUVENILE SEX OFFENDERS: PREDICTORS OF RECIDIVISM Joan D. Williams Doctor of Philosophy, May 10, 2007 (M.S., Fort Valley University, 1993) (B.A., Alabama State University, 1982) 153 Typed Pages Directed by Ren?e A. Middleton Juvenile sex offending is a growing problem in the United States. Victims and communities suffer greatly from their crimes. Because there is no one profile that fit these offenders it is hard to determine who will offend and who will not. Assessing the risk for sexual reoffense in juvenile sex offenders keeps both potential victims and communities safe. Therefore risk assessment is the most salient component of the overall assessment process of juvenile sex offenders. This study will provide both demographic and static risk factors that serve as predictors of reoffense in this population. Based on a multiple regression analysis those factors that serve as predictors in this sample are: age at time of offense, IQ score, parental marital status, severity of emotional abuse, severity of physical abuse and severity of emotional abuse. Variable from the STATIC-99 risk assessment instrument were also considered and the variable listed as predictors are? prior sex offenses, stranger victims, non-related victims, male victims, non-sexual contact v convictions, index: non-sexual convictions, prior non-sexual assault and more than four sentencing occasions. vi ACKNOWLEDGMENTS Dedicated first and foremost to God, who ordered my steps towards this purpose long before my birth. To my parents, George and Helen Williams, thank you for being the wind beneath my wings. Dad thanks for instilling in me a love of reading and mentoring me to be a critical thinker (think out side of the box, you?d say). Mom, thanks for your unconditional love. To my siblings (the rowdy bunch ? Jack, Jerome, Debbie and Pappy) thanks for encouraging and supporting my dream. To my greatest contribution to mankind, my beautiful children, Nicos and Melody ? thanks for your patience and for not having me committed during this process. Thanks to the therapist?s therapist, my best friend, Susie Brown ... we?re bad, and to my second set of eyes ? Benetta Eutsey ? for over 25 years of friendship. A special thanks to Dr. Sandra Harris, for igniting the flame; and to my major professor, Dr. Ren?e Middleton, for fanning the flame and reaching out to pull me, push me and mentor me with respect throughout this process. Mrs. Kornegay, thanks for allowing me to chase my dream and still maintain my livelihood. A special thanks to Dr. Chuck Jeffords with the Texas Youth Commission for serving as a silent committee member in the collection of my data ? I could not have done it had it not been for you. And to the man who adds spice to my life and who stands in the back to let me shine ? Ricky Triplett. Finally to all the people I lost along the way and those I found, to those who said I couldn?t or that I wouldn?t ? I did! Maybe it was because of you, or in spite of you ? but to all of you especially ? I say thanks! vii Style manual or journal used: Publication Manual of the American Psychological Association, 5th Edition. Computer software used: SPSS 13, Windows XP, Microsoft Word 2003, viii TABLE OF CONTENTS LIST OF TABLES..................................................................................................... xiii I. INTRODUCTION ......................................................................................... 1 Background.................................................................................................... 1 Rates of Recidivism ....................................................................................... 5 Purpose of the Study ...................................................................................... 6 Research Questions........................................................................................ 7 Expected Benefit to the Counseling Profession............................................. 8 Glossary of Terms.......................................................................................... 8 Limitations of the Study................................................................................. 9 II. REVIEW OF LITERATURE ........................................................................ 11 The Juvenile Sex Offender............................................................................. 11 Theories of Offending.................................................................................... 15 Cognitive Theory ............................................................................... 15 Social Learning Theory...................................................................... 18 Attachment Theory ............................................................................ 20 Studies on Risk Factors, Rates of Recidivism, and types of Abuse .............. 21 Risk Assessment Considerations and Measures ............................................ 37 Static and Dynamic Risk Factors................................................................... 40 Static Factors...................................................................................... 40 Dynamic Factors ................................................................................ 44 Risk Assessment Measures............................................................................ 47 Actuarial Assessments ....................................................................... 48 Clinical Assessment........................................................................... 49 J-RAT................................................................................................. 50 ERASOR............................................................................................ 51 J-SOAP .............................................................................................. 53 ix The Role of the Counselor ............................................................................. 55 Treatment Planning............................................................................ 57 Counseling Interventions ............................................................................... 59 Individual vs. Group .......................................................................... 59 Relevant Theories of Counseling................................................................... 62 Family Therapy.................................................................................. 62 Cognitive Interventions...................................................................... 63 Behavioral Interventions.................................................................... 66 Relapse Prevention............................................................................. 67 Chapter Summary........................................................................................... 68 III. RESEARCH METHODOLOGY................................................................... 70 Sample............................................................................................................ 71 Collection of Data .......................................................................................... 72 Variable Selection .......................................................................................... 72 Measures ............................................................................................ 73 Data Analysis ................................................................................................. 79 IV. RESULTS ...................................................................................................... 81 Characteristics................................................................................................ 81 Results............................................................................................................ 85 Analysis Procedure ............................................................................ 85 IQ Scores............................................................................................ 91 Abuse ................................................................................................. 91 STATIC-1 (Prior Sexual Offenses) ................................................... 92 STATIC-2 (Contact Sex Offenses with a Stranger) .......................... 92 STATIC-3 (Contact Sex Offenses with Non-Relative) ..................... 92 STATIC-4 (Contact Sex Offenses with Male Victims)..................... 93 STATIC-5 (Non-Contact Sex Convictions)....................................... 93 STATIC-8 (Index: Nonsexual Offenses)........................................... 93 STATIC-9 (Prior Non-sexual Assaults)............................................. 93 STATIC-10 (More than Four Sentencing Occasions) ....................... 94 Reoffense ........................................................................................... 94 STATIC TTL (Final Risk Assessment Score)................................... 94 Risk Level .......................................................................................... 95 x V. DISCUSSION ................................................................................................ 96 Implications for Counseling........................................................................... 104 Implications for Future Research................................................................... 105 REFERENCES .......................................................................................................... 107 APPENDICES ........................................................................................................... 135 Appendix A: Institutional Review Board .................................................... 136 Appendix B: Research Agreement .............................................................. 138 Appendix C: Static-99 Risk Assessment Instrument................................... 140 xi LIST OF TABLES Table 1 Recidivism Studies............................................................................. 22 Table 2 Reoffender Characteristics (Demographics)...................................... 83 Table 3 Model Summary................................................................................. 86 Table 4 ANOVA ............................................................................................. 86 Table 5 Coefficients........................................................................................ 88 Table 6 Correlations........................................................................................ 90 xii 1 I. INTRODUCTION ?The power to predict, is the power to protect? Anonymous The purpose of the study will be discussed, accompanied by a brief overview and background information on juvenile sex offending. This chapter will also discuss the research questions, as well as pertinent definitions and the limitations of this study. Background Juvenile sex offending is a growing problem in the United States. There has been an increase in both violent crimes committed by juveniles (Office of Juvenile Justice and Delinquency Prevention, 1994a) and in the reports of sexual aggression and sexual abuse (Hampton, 1995). Juveniles are responsible for 16% of all violent crimes and 17% of forcible rapes in the United States (Synder, 1999). Ageton (1983) concluded from a probability sample of 863 male adolescents 13 to 19 years of age that the rate of sexual assault per 100,000 adolescent males ranged 2 from 5,000 to 16,000. A survey of high school students revealed that 1 out of 5 had been involved in forcing sex on another and that 60% of the boys found it acceptable in one or more situations for a boy to force sex on a girl (Davis, Peck, & Storment, 1993). The most current data available indicates that from 1983 to 1993 the number of programs treating juvenile sex offenders (JSOs) increased from 20 to 800 (Freeman-Longo, Bird, Stevensen, & Fiske, 1995). For those juvenile who began or establish a pattern of sexual offending, once caught, they are remanded to treatment. Juvenile as well as adult sex offenders are mandated to treatment in secure or outpatient atmospheres based on their index offense. An index offense is the most recent sex offense for which they come into contact with a legal agency. An index offense can come in the form of a charge, arrest, conviction or rule violation (Hanson & Thornton, 1999). Index offenses can include multiple counts, multiple victims, and numerous crimes perpetrated at different times because the offender may not have been detected and apprehended (spree offending). In this case, all counts and convictions are considered part of the index offense. Once under the rehabilitative jurisdiction of the courts or mental health agencies the youth are evaluated. The evaluations determine the extent of their offending and the appropriate placement for their rehabilitation. Included in the 3 evaluation is also an assessment of risk. The assessment of risk is comprehensive and takes into consideration both offender and victim characteristics. Offender characteristics include: Severity and nature of the sexual offense, as well as, offender?s age, race, gender, grade, address, parental marital status, IQ score, abuse history, and victimization history. Victim characteristics are not as comprehensive and tend to include: Age, race, gender, type of offense, and severity of offense. The offenders are referred to treatment based on information from their comprehensive assessments. Historically, counselors were on the front line in reporting and treating victims of sexual abuse. Conversely, the role of mental health counselors are changing and their clients now include those youth and adolescents who have perpetrated sexual abuse against others (Cashwell, Bloss, & McFarland, 1995). JSOs enter treatment with a myriad of risk factors that spring from various demographic characteristics (Kelly, Lewis, & Sigal, 2004). Successful treatments include the responsibility of assessing risk. Psychometric testing and a clinical interview are typical components of the assessment process (Gerardin & Thibault, 2004). The role of clinicians in the assessment of risk is to observe, draw conclusions, and take action and to make decisions informed by explorations and formulations. Although there are clinical risk assessment instruments used on juvenile sex offenders, there are no empirically validated instruments used to 4 assess risk in this population (Prescott, 2004). The Static-99 is an empirically validated risk assessment instrument normed on adult male sex offenders (Hanson & Thornton, 1999), however its validity has been tested on juvenile sex offenders. The assessment process not only includes the risk of sexual re-offense, it also includes counseling interventions. The therapeutic relationship is a critical component of the assessment process. Therapeutic intervention is based on the presumption that it will reduce the incidences of sexual re-offense and the likelihood of the JSO going on to become an adult offender (Knopp, 1985). A major consideration in dealing with JSOs is determining how to best serve their mental health needs and their successful transition into society post-treatment, while protecting victims and communities. Limited research has focused on defining risk factors that might serve as predictors of those who would benefit most by intervention strategies aimed at preventing future sexual assaults. There are a number of factors that can contribute to re-offending among the JSO population. These factors are known in the literature as risks. Risk assessment focuses on the likelihood of recurring sexual offenses (Levenson & Morin, 2006). Assessment information is valuable to those (e.g., judges and probation officers) in the legal community who come in contact with JSOs in making decisions concerning their placement and treatment. Judgments about 5 risk have lasting impact on the offender, the victim, and the community where the offender resides or will reside after treatment. Other important considerations of risk assessment include family reunification, level of community supervision, the type of sentencing imposed and the type of treatment to be received (Fisher & Thornton, 1993). However, a major component that might well serve those who would benefit most by intervention strategies aimed at predicting future sexual assaults. Juvenile sex offenders are a diverse group with no one trait labeling them as actual or would be offenders (Rich, 2003). To date, the literature on JSOs has focused on victims, typologies, treatments, and assessment, however, very few studies have focused on factors that can predict sexual re-offense among this population. Following this path could prove to be Juvenile sex offenders are a diverse group with no one trait labeling them as actual or would be offenders (Rich, 2003). Following this path could prove to be very beneficial as the literature continues to document wide variability in rates of recidivism. Rates of Recidivism In base rate studies conducted on JSO recidivism, the rates fluctuate between 2% and 75% (Kenny, Keough, & Seidler, 2001) for sexual re-offense of JSOs apprehended for the first time. A short term study that followed offenders 6 for a year, showed a 3% recidivism rate; a JSO focus group showed rates of 2- 19%; and a Swedish study found a 20% recidivism rate in a five year follow up period (Witt, Jackson, & Hiscox, 2002). The disparity in the rates of recidivism in JSOs support the notion that it is difficult to accurately predict which individuals are at greater risk for sexual reoffending once sentences have been served or treatment has been completed. Purpose of the Study This study is designed to contribute to the limited but growing literature dedicated to understanding predictors of recidivism in juvenile sex offenders (JSOs). This study examines specific data (index offense, IQ score, parental marital status, severity of physical abuse, severity of emotional abuse, severity of sexual abuse) to determine which variables might best serve as predictors of sexual reoffense. Because there are no empirically validated risk assessment instruments for juvenile sex offenders (Prescott, 2004), select variables of the Static-99 will be used. The Static-99 is an empirically validated risk assessment instrument that has been normed on adult male sex offenders. The researcher will investigate select variables on the Static-99 to determine its effectiveness in assessing risk of sexual reoffense in JSOs. The variables include: (a) prior sex offenses, (b) prior sentencing dates, (c) any convictions for non-contact sex 7 offenses, (d) current convictions for non-sexual violence, (e) prior convictions for non-sexual violence, (f) unrelated victims, (g) stranger victims, (h) male victims. A study focusing on the factors that are likely to predict the risk of sexual re- offense, is not only appropriate, but indeed, necessary if we are to safeguard both past and future victims. Therefore this study seeks to answer the following research question: Research Question R1. Which variables will serve as significant contributors and predictors of recidivism among juvenile sex offenders? Variables under study are as follows: Age at time of offense, IQ, parental marital status, severity of physical abuse, severity of emotional abuse, severity of sexual abuse, and specific Static-99 variables (prior sex offenses, contact sex offenses with stranger, contact sex offense with non-relative, contact sex offense with male victim, non contact sex conviction, index non-sexual assault, prior non-sexual assault, and more than four sentencing occasions). R2. Will select variables on the Static-99 be useful in providing valid psychometric information to determine its effectiveness in assessing risk of sexual re-offense among youth who sexually reoffend? 8 Expected Benefit to the Counseling Profession Juvenile sexual offending is a significant problem and can be greatly benefited by both legal and professional attention by the counseling profession. Victims of sexual abuse experience clinically significant symptoms in the affective, cognitive, physical, and behavioral domains (Shaw & Lewis, 2000). Juvenile sex offenders can be treated effectively, if they are provided with specialized treatment and on-going supervision. In working with juvenile sex offenders, counselors should design treatment objectives based on comprehensive assessment. If we are able to predict those who are at greater risk of sexually reoffend, counselors can intervene and provide targeted treatment to reduce the risk. Throughout this dissertation several key terms are used in reference to JSOs and the prediction of recidivism. In the section that follows, these terms are listed with their definitions as used in this present study. Glossary of Terms Adjudicated ? found guilty by a court of law (Hanson, 1998). Juvenile Sex Offender (JSO) ? a youth between the ages of 10 and 17, who commit any sexual act with a person of any age, against the victims will or in an aggressive, exploitive, or threatening manner (Gerardin & Thibaut, 2004). 9 Recidivism ? any re-arrest for a sexual offense after release from treatment or confinement for a sexual offense (Hanson & Thornton, 1999). Risk Assessment ? The process of evaluating individuals to (1) characterize the risk that they will commit violence in the future, and (2) develop interventions to manage or reduce that risk (Boer, Hart, Kropp, & Webster, 1997). Static-99 ? The Static-99 is a brief actuarial instrument designed to estimate the probability of sexual and violent recidivism among adult males who have already been convicted of at least one sexual offense against a child or non- consenting adult (Hanson & Thornton, 1999). Sexualized behaviors ? sexualized play with dolls, putting objects into the anuses or vaginas, excessive or public masturbation, seductive behavior, requesting sexual stimulation and age inappropriate sexual knowledge (Kendal- Tackett, Williams, & Finkelhor, 1993). Limitations of the Study The archival records that were used for this study are from a single secure treatment program that treats juvenile sex offenders. The generalizability of the findings to other juvenile offenders is limited. An additional limitation was the small sample size for the regression analysis (n = 179). Static factors make the 10 Static-99 easy to score however; the lack of dynamic factors is a limitation for use with juvenile sex offenders (Hanson & Thornton, 1999; Wood & Cellini, 1999). 11 II. LITERATURE REVIEW This section will discuss literature relative to juvenile sex offending and the predictors of sexual reoffending, in particular. A brief overview on recidivism rates, theories of offending, risk factors, risk assessment instruments, and counseling interventions will be discussed. The Juvenile Sex Offender Historically, juvenile sex offending has been ignored and emphasis has focused on adult male sex offending (Becker & Hunter, 1997). Juvenile sex offending is often under reported, three out of four incidences are never reported according to the U.S. Department of Justice (Bureau of Justice Statistics, 2002). There have been debates of the proper legal term to apply to juveniles who sexually offend. Many researchers prefer to call them youth that sexually offend rather than offenders; thus placing emphasis on the behavior rather than labeling the youth as an offender (Moore, Franey, & Geffner, 1994). The term ?sexually reactive youth or adolescent? has also been used to refer to this population. A sexually reactive youth is defined as a youth who offends as a way of acting out 12 his or her own abuse victimization (Moore et al.). Regardless of the title given to Juvenile Sex Offenders (JSOs?the most commonly used name and acronym), their abusive behavior often times leave their victims broken and scarred well into adulthood. Although there is no one name or profile that fits JSOs, they do present a profile or share a set of common characteristics. Becker (1988) suggested that there are essentially four kinds of youth that sexually offend, with most offenders combining features as follows: (a) The true paraphiliac with a well-established deviant pattern of sexual arousal; (b) The antisocial youth whose sexual offending behavior is but one facet of his or her opportunistically exploiting others; (c) The adolescent compromised by a psychiatric or neurological/biological substrate disorder which interferes with his or her ability to regulate and modulate aggressive and sexual impulses; and (d) The youth whose impaired social and interpersonal skills result in turning to younger children for sexual gratification unavailable from peer groups. Juvenile sex offenders are defined as adolescents between the ages of 10 to 17 who commit non-consensual and often times forcible sexual acts against others (?Center for Sex Offender Management,? 2003). According to the Center for Sex Offender Management (1999), JSOs are responsible for one-fifth of all forcible rapes and one-half of sexual offenses against children. In recent years, 15, 500 13 male and female adolescents were charged with one or more sexual offenses (Maquire & Pastore, 2002). While ninety percent of all juvenile sex offenders are male (Department Of Psychiatry, 2004), adolescent males are responsible for half of the sexual offenses committed against male victims under the age of 12 (Ryan, 1999). Fourteen-year- old males commit 90% of sexual offenses that are perpetrated against both male and female victims (Rich, 2003). In the criminal and clinical system, JSOs are characterized as those that target children and those that target peers and adults (Rich, 2003). Those juvenile sex offenders that target adult victims, usually target women and choose victims that are strangers rather than someone they know (Gerardin & Thibault, 2004). However, more characteristically, their victims are typically female acquaintances or siblings (Zonlondek, Abel, Northey, & Jordan, 2001). A study conducted in the Netherlands by Bijleveld and Hendricks (2004) concluded that 50% of the JSOs in their study that target children were in special education classes and were victims of bullying. Sixty percent of the sample had victims age 10 or less; and 76% of their participants had only one victim. Those juvenile sex offenders that target adult victims, usually target women and choose victims that are strangers rather than someone they know (Gerardin & Thibault, 2004). 14 According to Boyd, Hagen, and Cho (2000), the type of sexual offense committed correlates with risk of recidivism. There are two types of sexual offenses: hands on and hands off. Documented reports of juvenile sexual offending are typically those that involve severe hands-on offenses such as fondling, oral sodomy and penetration (Becker & Murphy, 1998). Hands on offenses are those offenses that count for the majority of sex offenses committed by adolescents (CSOM, 1999). Hands on offenses are those offenses that involve touching and the physical sexual assault of a victim. Hands on offenses typically involve oral-genital contact and actual or attempted vaginal/anal penetration. Another form of a hands on offense is frotteurism. Frotteurism is the rubbing against a non-consenting person for the purpose of sexual gratification. This act can lead to orgasm for the person committing the act and is usually done without the victim?s knowledge. Hands off offenses include the following: obscene phone calls, theft of clothing for sexual purposes, threats of sexual harm, exhibitionism, public masturbation and distribution, public display, or depictions of sexually obscene material (Rich, 2003). These offenses target same age or older victims (Davis & Leitnberg, 1987). Creation, possession, and distribution of child pornography and bestiality are also considered hands off sexual offenses. The type of offenses committed is a consideration when determining placement and treatment for 15 JSOs. Regardless of the age or gender of the victim an even more important consideration of juvenile sex offending is why. The victim may ask, ?why did this happen to me?, a parent my ask, ?why did this happen to my child,? the parent of the offender may ask, ?why did my child do it,? and perhaps the JSO will ask, ?why did I do it.? Theories of Offending In the section that follows, we will examine several theories of counseling and psychotherapy and their relevance specific to youth who sexual offend. . Cognitive Theory According to Ryan and Lane (1997), cognitive theories have shown that nearly all sex offenders use some form of ?thinking errors? to justify their abusive behavior. Sex offenders generally think about their victims in a self- serving manner (Ward, Hudson, Johnston, & Marshall, 1997). These distorted cognitions help offenders maintain their deviant behavior (Hayashino, Wurtele, & Kleeb, 1995). Cognitive distortions are self statements that assist the offender in denying, minimizing, justifying, and rationalizing their behavior (Murphy, 1990). Implicit theory is a form of cognitive theory. Implicit theory contains a number of ideas and mental constructs to include propositions about the victims? 16 desires, beliefs, and attitudes (Ward & Keenan, 1999). In a study conducted by Williams and Finkehor (1990) incestuous fathers were found to be impaired in their cognitive capacity for empathy and bonding. The implicit theory as it relates to sexual offenders has five assumptions: (a) people are motivated by a desire for pleasure, (b) some people are superior to and more important than others, (c) the world is a dangerous place which causes other to behave abusively, (d) that the world is uncontrollable and inexorable in its actions and finally, (e) there are degrees of harm and sexual activity is beneficial and unlikely to harm a person. The first assumption states that individuals, including children, have the right to express their sexual needs (Ward, Fon, Hudson, & McCormack, 1998). This assumption is likely to yield the following distorted statements, ?The child wanted sex,? ?the child seduced me,? ?we love each other,? and ?she didn?t say no.? The second assumption states that because of the superior status the sex offender thinks he holds, he has the right to assert his need above others. Offenders believe that victims should satisfy their emotional and physical needs (Ward, Hudson, Johnston, & Marshall, 1997). Distorted cognitions you would hear from this assumption would include, ?I am justified in having sex with my daughter if my wife won?t have sex with me,? ?If I don?t do it someone else will,? and ?I?m the boss in this family? (p. 829). The third assumption is that the world 17 is a dangerous place and that it is necessary to fight back and achieve dominance and control over others. It also assumes that people as a whole are untrustworthy, rejecting and will take unfair advantage of any given situation (Ward, Hudson, Marshall, & Siegert, 1995). Cognitive distortions generated by this assumption are, ?Kids really know how to love you,? ?Kids like sex with adults because it makes them feel loved and wanted,? and ?Children give more love and acceptance than adults? (p. 830). The fourth assumption states that pertinent early learning experiences that are uncontrollable leave the offender with deviant preferences that he is unable to manage and/or control. Common distortions associated with this theory include, ?It would not have happened if I had not been sexually abused as a child,? ?It just happened,? and ?I can?t control myself? (p. 831). The final assumption of implicit theory as it relates to sex offenders states that harm has degrees ? from little at one end and extreme on the other end. Cognitive distortions include, ?She was asleep, she didn?t know what I was doing,? ?She is not a blood relative so it?s not so bad,? and ?It happened to me and it didn?t hurt me (Ward, 1999). These theories have shown promise in explaining sex offending. They further assert that cognitive restructuring should be included in the treatment of juvenile sex offenders. 18 Social Learning Theory Twenty to 50% of juvenile sex offenders have histories of physical abuse and 40 to 60 % have histories of sexual abuse (Hunter & Becker, 1998). A study that compared juvenile sex offenders with nonsexual offenders found that those that committed sex offenses had higher incidences of physical abuse in their backgrounds (Ford & Linney, 1995). Social learning theory looks at victimization as a part of the etiological explanation to sexual offending (Burton, 2003). According to social learning theory, many offenders victimize others initially to meet specific needs (Akers, 1998). Those needs, according to Bandura (1986) are the anticipated rewards they perceive their own offending will obtain. The youth then continue to offend in order to take care of an ongoing need to resolve their sexual trauma (Veneziano, Veneziano, & Legrand, 2000). Because Bandura?s (1977) theory of social learning has provided a compelling model to explain the initiation and acquisition of many human behaviors, the notion that abusive behaviors are learned through observation or experience continues to dominate hypotheses regarding the question ?Why do they do it?? The outcomes of childhood victimization are not predictable. It is clear that victimization is over-represented in the lives of those juveniles who perpetrate sexual abuse (Ryan, Miyoshi, Metzner, Krugman, & Fryer, 1996), but it is equally clear that most victims do not perpetuate abuse. A study conducted 19 by Burton, Nesmith and Badten (1997) applied the social learning theory to 287 sexually aggressive youth. They concluded that there was a correlation between being sexually victimized and the subsequent number of victims of the person that was victimized. Using a social learning theory, Ryan (1989), suggests that sexually abusive behavior by male adolescents stems from their own sexual victimization She also suggests that a ?traumatized child may become fixated on the trauma, recreating the experience in ritualistic patterns that become more elaborate, more rigid and more secret over time? (p. 10). Becker and Abel (1985) found in their study, that half of the adult offenders report offending as an adolescent and also reported that their offending escalated in severity and frequency over time. Consequently, sexual offenders who began their offending as adolescents could be distinguished from those that began offending as adults by the frequency and level of their own childhood victimization (Knight & Prentky, 1993). Although there are empirical, theoretical and clinical links to adolescent sexual abuse and subsequent sexual victimization, it is important to note that very few victimized youth go on to become offenders (Burton, Miller, & Shill, 2002). According to Ryan and Lane (1997), one of the best examples of research on a more general victim to victimizer hypothesis is Widom?s (1989) longitudinal study of 908 cases, in which she assessed sexual victimization as a risk factor for 20 further delinquency. She reports that ?... abused and neglected children have significantly greater risk of becoming delinquents, criminals and violent criminals? than non-abused children (p. 246). Attachment Theory In recent years research has began to focus on the impact of sex offender?s intimate relationships and the attachments they form in their early years (Marshall, 1989). According to attachment theory, interpersonal schemas are shaped through our experiences with others (McCormck, Hudson, & Ward, 2002). The theory further states that the most significant relationship that shapes all others is the relationship we have with our parents or caregivers. The role of the father in an individual?s offending has traditionally been seen as insignificant (Tingle, Barnard, Robbin, Newman, & Hutchinson, 1986). This may be in part to the absence of the father during early upbringing of the sex offender. However, the offenders that reported the presence of a father in their early upbringing, the relationship tended to be negative and problematic (Lisak & Roth, 1990). A comparison of adolescent sex offenders to other non delinquent adolescents (Blaske, Borduin, Henggeler, & Mann, 1989) also found that the sex offender group tended to have lower rates of positive mother-son communication. 21 The loss of a caregiver is also an important source of disruption in early interpersonal relationships (McCormack ,Hudson, & Ward, 2002). Ryan and Lane (1991) conducted a study that found that over half of the JSOs in their study had experienced some form of parental loss through, separation, divorce, and/or death. Regardless of whether one or both parents are in the home of the juvenile sex offender, there still tends to be considerable dysfunction that ultimately has the potential to damage the quality of interpersonal relationships they form (Ryan & Lane,1991). Studies on Risk Factors, Rates of Recidivism and Types of Abuse The ability to predict future acts of sexual violence is an important step in targeting those JSOs who are likely to reoffend. Understanding base rates of recidivism is essential in assessing risk (Monahan, 1981; Serin & Brown, 2000). Although it has been suggested that base rates for sexual reoffense underestimate the true numbers of offenses perpetrated by JSOs (Doren, 1998; Abel & Rouleau, 1990), they do provide a rate of minimum risk that the offenders poses to the community when released (Hagan & Gust-Brey, 2000). Table 1 contains a list of recidivism studies conducted on juvenile sex offenders (Prescott, 2004; Wolk, 2005). A brief review of these studies explore the differences among samples, methods and measures of recidivism. 22 Table 1 Recidivism Studies Citation Treatment N Follow- up Period Recidivism Measure % Sexual Recidivism % Violent Recidivism % Non- Sexual Recidivism Risk factors suggested Bijleveld & Hendriks, 2003 83 (not convicted) Looked at previous sexual offenses of solo vs. group JSOs- 76% vs. 25% Solo offending, childhood sexual abuse, neurotic impulsive Bourduin, Henggeler, Blaske, & Stein (1990) 0 8 (MST) 3 years Criminal charges 12.5 treatment; 75% compare- son N/A 25% treatment; 50% comparison Non-multi- systemic therapy Brannon & Troyer (1995 36 4 years Incarcerat- ed as adult 3% N/A 14% None reported Bremer (1992) 193 Several months to 6 years (no M specified) Self-report 11% N/A N/A No specialized treatment Gretton, McBride, Hare, O?Shaughnes sy, & Kumka 220 7-106 months Charges and convictions 15 N/A 51 High score PCL:YV Hagan & Gust-Brey, 2000 50 5 or 10 years Conviction adult 20 N/A 60 None noted Hagan & Gust-Brey, 1999 50 6 or 10 years Conviction adult 16 N/A 90 Lack of supervision post release (table continues) 23 Table 1 (continued) Citation Treatment N Follow- up Period Recidivism Measure % Sexual Recidivism % Violent Recidivism % Non- Sexual Recidivism Risk factors suggested Hagan et al., 2001 100 8 years Conviction adult 20 N/A Not reported Previous sexual offense, Delin- quency Hagan et al., 1994 50 2 years Conviction adult 10 N/A 58 No treatment, antisocial Kahn & Lafond, 1988 350 <1 month ? 6 years Unknown 9 N/A 8 No treatment, Sexual abuse history Kahn & Chambers (1991) 221 20 months Juvenile adjudicatio n 7.5 N/A 50 Younger offender age, verbal threats in index offense, denial or victim blame Lab, Shields, & Schondel 49/109 comp group 1-3 years Juvenile court contact 2.2 (tx) 3.7 (comp) N/A Juvenile court contact No treatment Langstrom & Grann (2000) 46 (age 15- 20) 72 months (M = 5 years) Convictions 20% 22% 65% Sexual deviance, male victim, > 1 victim, poor social skills (table continues) 24 Table 1 (continued) Citation Treatment N Follow- up Period Recidivism Measure % Sexual Recidivism % Violent Recidivism % Non- Sexual Recidivism Risk factors suggested Langstrom (2002) 117 115 months (M= 6.32) convictions 29.9% N/A 41.9 Previous sexual offense, offense in public place, stranger victims, 2 + occasions, 2+ victims Parks, 2004 156 1-134 months Juvenile adjudica- tion, adult convictions 6.4 N/A 30.1 Impulsive & antisocial behavior (JSOAP-II) interperson al (PCL-YV) Prentky, Harris, Frizzell,.& Righthand 75 12 months Juvenile charge 4 N/A 6.7 Sexual drive, preoccupati on, impulsive, antisocial behavior Rassmussen, 1999 170 5 years or until 19 years of age Juvenile court 14 N/A 54 More female victims, failure to complete treatment Seabloom, Seabloom, Seabloom, Barron, & Hendrick- son, 2003 114 14-24 years Convictions 0 N/A 39 No treatment Sipe, Jensen, & Everett, 1998 124 1-14 years Adult re- arrest 9.7 N/A 37 None reported (table continues) 25 Table 1 (continued) Citation Treatment N Follow- up Period Recidivism Measure % Sexual Recidivism % Violent Recidivism % Non- Sexual Recidivism Risk factors suggested Smet & Cebula, 1987 21 36 months Not reported 4.8 N/A Not reported No specialized treatment Worling & Curwen, (2000) 139 120 months Criminal charges 5% treated; 18 % non- treated 19% treated; 50% non- treated 21% treated; 50% non- treated Sexual deviance, intrusive sexual assault with children, no specialized treatment Bijleveld and Hendriks (2003) were the first to compare personality variables of group offenders (index sexual offense committed by more than one offender) and solo offenders (those that commit sexual offenses alone). In their study, archival data was collected from questionnaires done at intake. The data was used to determine levels of personality variables (neuroticism, impulsiveness, sociability, sensation seeking, intelligence and conscience formation). Inter-rater reliability scores were between .80 and .98. The sample consisted of 83 males adjudicated sex offender in the Netherlands. In addition to the findings listed in Table 1, 54% of the solo offenders committed multiple sexual offenses with multiple victims. Solo offenders were significantly older than the group offenders and scored higher on neuroticism and impulsivity. The 26 solo offenders tended to score lower on sociability and were more likely victims of sexual abuse. Borduin et al. (1990) compared a small sample of 16 adjudicated JSOs. The groups (8 each) were randomly assigned to individual therapy (IT) or multisystemic therapy (MST). Multisystemic Therapy (MST) is an intensive family and community-based treatment that addresses the multiple determinants of serious antisocial behavior in juvenile offenders (Henggeler, Mihalic, Rone, Thomas, & Timmons-Mitchell, 1998). Borduin et al. were the only researchers to compare two treatment groups with subjects assigned by random assignment. The MST group received an average of 37 hours of individual, group, and family therapy. Their therapy was based on an individualized plan that took into consideration the needs of the juvenile offender cognitive process, family dynamics, and school performance. The MST group had a sexual recidivism rate of 12.5% within a three year period. The IT group received an average of 45 hours of individual counseling. In their individual sessions, they addressed personal, family, and academic issues. The IT group had a sexual recidivism rate of 75% (which was a statistically significant difference) within a 3 year period. The results of this study suggested that therapy targeted to the multiple systems of the JSO would more likely reduce sexual reoffense than would individual therapy. 27 Brannon and Troyer (1995) examined 36 JSOs in a residential center. They were given therapy that addressed concerns other than their sexual offending as well as offender specific treatment. A re-arrest for a sexual offense was how recidivism was determined. Based on their study, 1 of the 36 JSOs were remanded to an adult correctional facility for a non-sexual reoffense. Their study suggested that JSOs would have lower sexual reoffense rates whether they received offender-specific treatment or not. They also concluded that residential treatment of 12 months or less would reduce recidivism. Bremer (1992) conducted a follow up study using 193 samples. The JSOs in his sample were in an offender-specific residential facility. He found a 6% recidivism rate for sexual reoffense based on convictions and an 11% recidivism rate based on self-report. He concluded that there was no significant difference in reoffense based on time between release and follow-up (less than 6 months to 8.5 years). He also found that there was no significant difference in sexual reoffense based on the length of stay in the treatment program (less than 30 days to 30 months). Gretton et al. (2000) is an important study because it was one of a few studies to use the Psychopathy Checklist Youth Version (PCL:VY). The PCL:YV test can be used to help clinicians understand the factors that contribute to the development of adult antisocial behavior and psychopathy. It is a 20-item rating 28 scale for the assessment of psychopathic traits in male and female offenders aged 12 to 18 and created by Robert Hare (Froth, Kosson & Hare, 2003). Robert Hare is Emeritus Professor of Psychology at the University of British Columbia, where he has taught and conducted research for some 35 years, and President of Darkstone Research Group Ltd., a forensic research and consulting firm. He has devoted most of his academic career to the investigation of psychopathy, its nature, assessment, and implications for mental health and criminal justice. This study consisted of a sample of 220 Canadian juvenile sex offenders who either confessed or who were convicted of sexual offenses. The recidivism was determined by charges and convictions. This study found a recidivism rate of 51% for general reoffense, 30% for violent reoffense, and 15% for sexual reoffense. Significant differences were noted between the groups: JSOs with high scores re-offend the most, next were JSOs with medium scores, and then JSOs with low scores. Juvenile sex offenders with high scores were more likely to escape from custody, violate rules of probation, and reoffend both violently and non-violently. The next four studies (Hagan & Gust-Brey, 2000; Hagan & Gust-Brey, 1999; Hagan et al., 2001; Hagan et al., 1988) were all conducted in Wisconsin using subjects who completed specialized sex offender treatment in a state 29 operated residential program. The subjects were all male juvenile sex offenders who were convicted of serious sexual and non-sexual offenses. Hagan and Gust-Brey (2000) reviewed the records of 50 juvenile sex offenders to determine rates and predictors of sexual recidivism. The offenders in this study were at least five years older than their victims. The five year follow up of these offenders shows a 12% recidivism compared to a 20% recidivism rate at 10 years. However, these rates were not statistically significant. This study also compared the nonsexual recidivism rates at five and ten years. At five years the recidivism rate for non sexual recidivisms was 50% at five years and at 10 years it was 60%. Hagan and Gust-Brey (1999) reviewed the same records of 50 JSOs but at five and ten years post release. The five year reconviction for general offending was 74% compared to 90% at the ten year point. For sexual recidivism the follow up rates were 12% at five years and 20% at ten years. The nonsexual recidivism rates were 50% at five years and 60% at ten years. The study that had the shortest follow up period was the one conducted by Hagen et al. (1994). The sample consisted of 50 adolescent rapist who had offended against an opposite sex, same age peer or older victim. Recidivism was measured by adult conviction within two years of release. The juvenile offenders that were reconvicted as adults showed a 10% recidivism rate for sexual offenses 30 and a 58% recidivism rate for nonsexual offenses. The authors of the study suggested that specialized treatment (treatment that is sex offender specific), was successful in reducing the level of recidivism; however, there was no comparison group. In the 2001 study conducted by Hagan et al. (2001) recidivism rates of two groups of juvenile sex offenders were compared?rapist and child molesters to a nonsexually offending group of juvenile offenders from the same residential program. The groups were compared and all were found to be at an increased risk for reoffense. The sexual reoffense rate for child molesters was 20%, for rapists it was 16%, and 10% for nonsexual offending delinquents. There were no significant differences between the JSO groups, but there was a significantly more sexual reoffending by the combine juvenile sex offender groups relative to the nonsexually offending group. In one of the most dated studies of recidivism with JSOs, Kahn and Lafond (1988) used a sample from an offender specific treatment program. This study also looked at female offenders and the anticipated goals that have been targeted for treatment. This study reported that offenders disclosed more detailed histories during therapy than in initial sessions. Fifty percent of JSOs in this study reported being victims of sexual abuse yet during therapy the rate increased to 60%. The sample consisted of 350 JSOs in a Washington State secure 31 treatment facility. Child molesters (victims at least two years younger) accounted for 80% of the sample, while offenders against adult victims accounted for 2% of the sample, 6% of the sample had mixed-aged victims. The sample also included 25 female sex offenders, all of whom were victims of childhood sexual abuse. Recidivism was examined from less than a month to six years post release. The recidivism rates were reported at 9% for sexual reoffenses and 8% for nonsexual reoffenses. For the purpose of this study recidivism was not operationally defined. Kahn and Chambers (1991) conducted a study that reviewed recidivism rates of JSOs that were reconvicted in eight community based programs and two residential programs. The sample included 221 male JSO with 86% child molesters and 14% same-age peer or adult offenders. There was a 20 month follow up period. There was a 7.7% sexual reoffense rate and a 44.8% non sexual reoffense rates. Some factors related to their sexual reoffense were: Use of verbal threat during the offense, blaming the victim or denying the offense, and age (younger) at the time of the offense. In a study conducted by Lab, Shields, and Schondel (1993) recidivism rates of JSOs in a specialized sex-offender specific program (a program designed specifically for sex offenders with the goal of preventing future victimization and striving to ameliorate the harm done by the offender to the victim of the crime), 32 were compared to JSOs in a non-specialized program. The sample consisted of 150 JSOs who were given designations of risk for sexual reoffense (low, medium, and high) . Risk designations were derived at using a clinical form of assessment that takes into consideration information received from case files and client interview. The low and medium risk JSOs were referred to the specialized program and the high risk JSOs were referred to the non-specialized program. Recidivism for both programs was measured as any juvenile court contact that was either sexual or both sexual and nonsexual. The low and medium risk JSOs that received specialized treatment had recidivism rates of 2.2% for sexual and 24% for any court contact versus 3.7% sexual and 18% any court contact for those that were considered high risk and received non-specialized treatment. Langstrom and Grann (2000) examined recidivism rates of JSO in Sweden. They had a sample size of 46 (two of whom were female) and ranged in ages from 15 to 20. In Sweden, it is important to note that individuals under the age of 15 are not considered legally responsible and were not included in the sample. The mean follow up period was five years. There was a 20% sexual reoffense rate and a 22% non-sexual reoffense rate. Factors listed as predictors of reoffense were previous sexual offenses, male victims, poor social skills, and more than one victim. 33 Langstrom (2002) conducted a follow up study with a longer follow up period and a larger sample. The study consisted of a sample size of 117 (2 females) and a follow up period of 115 months. The recidivism rate for sexual reoffense was 29.9% and for violent non-sexual reoffense the recidivism rate was 41.9%. Factors attributed to sexual reoffense were previous sex offenses, stranger victims, having more than two victims, sexual offense in a public place. Parks (2004) in an unpublished dissertation examined recidivism in JSOs as they relate to offender type. The offender types include those that offend against a child, those that offend against peers and adults, and those with mixed victims. The sample consisted of male JSOs in a secure facility in Oklahoma who received specialized sex offender specific treatment. There were 74 JSOs who offended against children (defined as victims under the age of ten and at least four years younger than the offender), 51 JSOs who offended against peers and adults, and 31 JSOs who offended against both children and adults. Recidivism was measured by juvenile and adult convictions of sexual offenses. Of the sample, 6.4% sexually reoffended and 30.1% had general reoffenses. A residential program in Philadelphia was the site of a study conducted by Prentky, Harris, Fizzell, and Righthand (2000). The sample consisted of 75 male JSOs who were given the Juvenile Sex Offender Assessment Protocol (J- SOAP) to access risk. The JSO were categorized by type according to victim 34 choice and it was determined that the groups were too small to analyze. The average length of stay for the JSOs was 24 months with recidivism measured at 12 months by re-arrest rates. The recidivism rate for sexual reoffense was 4% and the rate for non-sexual reoffense was 6.7%. Rasmussen (1999) reviewed 170 JSOs (3 were female) in Utah among first-time juvenile sex offenders. This study used regression analysis to determine statistical predictors of sexual reoffense. Recidivism was measured as reconviction within a five year follow-up. Rasmussen?s results were consistent with previous research in that 54.1 percent (N = 92) of the sample committed a new nonsexual offense, whereas only 14.1 percent (N = 24) committed a new sex offense. The relatively higher reoffense rates may reflect the comparatively long follow-up period of 5 years. The sexual recidivism risk factors included multiple female victims, failure to complete treatment and the offender having no treatment. Seabloom et al. (2003) evaluated a specialized treatment program for JSOs located in Minnesota. The program use multisystemic therapy and required family involvement. The components of the program were individual therapy, group therapy, family therapy, family psycho-educational groups, and collaboration with the juvenile probation staff. The sample included 122 JSOs and their families. The follow-up period was 14-24 years and recidivism was 35 measured as adult re-arrest and/or conviction. The results of this study reported no sexual recidivism and a 39% non sexual reoffense rate. Sipe, Jensen and Everett (1998) conducted a study that compared recidivism rates using adult records of non-violent JSOs and non sexually- offending juvenile delinquents in Idaho. The follow up period was 12 months to 14 years. The sample consisted on 124 adjudicated non-violent JSOs compared to 132 non-sexually offending adjudicated delinquents. The results show that JSOs (9.7%) were significantly more likely than the comparison group (3%) to be rearrested for a sexual offense. Smet and Cebula (1987) conducted a program evaluation of a community based treatment program in rural Wisconsin. The program was a court ordered program that used group therapy in treating JSOs. The measurement for recidivism was not clearly defined, however, in a follow up period of three years one JSO (4.8%) was convicted of sexual reoffense. Worling and Curwen (2000) compared groups of juvenile offenders from a community based specialized offender specific program in Canada. The samples were 148 JSOs (9 of which were females). Fifty-eight (including 5 females) JSOs were in the treatment group and 90 (including 4 females) were in a non- treatment group. In the treatment group they addressed the following issues: Relapse prevention, cognitive behavioral intervention, denial, victim empathy, 36 deviant sexual arousal, skills building using individual and group therapy and family therapy. The sexual reoffense rate for the treatment group was 5% and the non treatment group was 18%. The violent nonsexual reoffense rates were 19% for the treatment group and 32% for the comparison group. Non-violent reoffense rates were 21% in the treatment group and 50% in the comparison group. The overall reoffense rates for the treatment group were 35% and 54% for the comparison group. Factors listed for sexual reoffense were past or present sexual fantasies of children, grooming of child victims, and intrusive sexual acts with children. These findings support the positive benefits of therapeutic intervention to reduce recidivism or re-offenses. According to Hunter (2000) program evaluation data suggest that the sexual recidivism rate for juveniles treated in specialized programs ranges from approximately 7%?13% over follow-up periods of two to five years. Studies suggest that rates of non-sexual recidivism are generally higher (25?50%). If findings from future treatment outcome studies on juvenile sex offenders parallel those on adult offenses, sexual recidivism rates will be higher in individuals who fail to successfully complete programs. In a recently conducted study, Hunter and Figueredo (1999) found that as many as 50% of youths entering a community-based treatment program were expelled during the first year of their participation. Program failure was found to be largely attributable to failure to 37 comply with attendance requirements and/or therapeutic directives. Youths failing to comply with the program were found to have higher overall levels of sexual maladjustment (as measured on assessment instruments), and were judged possibly to be at greater long-term risk for sexual recidivism. In this study, lower levels of client denial at intake best predicted successful program compliance. Higher levels of denial were found in nonadjudicated youths. Risk Assessment Considerations and Measures Knowing the statistical profile of a JSO and the theories or explanations for why they do what they do, is not enough when thinking of the far reaching impact of their sexual offending. Assessing the risk they pose and managing that risk is also an important undertaking. According to Hart (2001), risk assessment is the ability to understand hazards and to minimize their negative consequences. The most critical and salient feature in assessing the JSO, is risk assessment (Rich, 2003), all other treatment goals are secondary to the elimination of the sexually abusive behaviors. Juvenile sex offenders, considered a unique population by both researchers and policy makers alike, are in need of specialized treatment and assessment (Fanniff & Becker, 2006). The assessment of risk focuses on the likelihood of recurring sexual offenses (Levenson & Morin, 2006). Judgments 38 about risk have a lasting impact for offenders, victims and communities. When making decisions concerning the types of treatment that would most benefit JSOs, attorneys, judges, and counselors have to evaluate the level of risk (harm) the JSOs pose to the community (Kahn & Chambers, 1991). Risk assessment will determine several important factors for the offender. Those factors include whether they can rejoin their families, the level of community supervision, the type of sentencing imposed and the type and intensity of treatment to be received (Fisher & Thornton, 1993). In retrospective studies, adult pedophiles show that between 40% and 50% began offending as juveniles (Hunter, 1990). Therefore, risk assessment is an important consideration in the treatment and management of JSOs. The literature supports the notion that the assessment of youth who sexually offend must be an on-going process and it is recommended that it be conducted at six stages: (a) pretrial/investigation, (b) pre-sentence/risk prognosis, (c) post-adjudication/ disposition, (d) needs/treatment planning and treatment evaluation, (e) release/ pre-discharge, and finally (f) monitoring/ follow up (National Task Force on Juvenile Sexual Offending, 1993). However, more commonly, risk assessment is conducted at the time of arrest or release from incarceration (Witt, Jackson, & Hiscox, 2002). 39 For JSOs, risk assessment serves three purposes: (1) to assess the likelihood of on-going dangerous behavior, (2) to determine appropriate treatment, and (3) to assess the individual?s motivation to engage in treatment (Will, 1999). In working with JSOs in assessing risk, it is important that mental health practitioners and evaluators assess risk while looking at available treatments and other factors that will affect risk in the future. Risk is assessed in many different ways. It is important to note that those given the authority and responsibility to assess risk will make judgments that have long lasting impact on the offenders, victims and communities. For counselors entrusted with this responsibility, they can best serve the client, victim and community by being familiar with the emerging research and strategies for conducting risk assessment (Prescott, 2004). To effectively assess risk, the amount of structure involved and the empirical support for the procedure are important considerations (Witt, Bosley, & Hiscox, 2002). According to Calder (2000), there is no one ideal method to assess risk. When evaluating JSOs to determine the level of risk they pose, biological and psychosocial factors should be considered. For example, sexual development, gender roles, and sexual arousal patterns should be considered (Shaw, 2000). It is also important when assessing risk with JSOs that their still-developing natures be taken into consideration (Rich, 2003). 40 Static and Dynamic Risk Factors Static Factors In dealing with risk assessment of sex offenders, it is important to consider factors that are both static and dynamic. Static factors are those things that don?t change such as number of offenses, victim selection, family history and sexual history. These are historical behaviors and experiences that will remain unaltered over time (Rich, 2003). Static factors have been the easiest to obtain because they can be located using archival data. They can also give information of the developing pathology of offenders. Static factors do not tell when an offender will reoffend nor can they tell if treatment was beneficial (Hanson, 1998). Static factors are useful in assessing overall risk levels as risk is predicated on past behavior (Rich, 2003). According to Rich, the pathway to sexual offending in JSOs tend to be developmental in that it can begin in early childhood and grow through circumstances, into adolescence and even adulthood. Although there is no way to determine if a JSO will sexually re- offend, certain predictions can be made based on static information gathered during the assessment process. Static factors include but are not limited to: Prior sex offenses, family history, and victim selection. 41 Perhaps, the strongest predictor of whether or not an offender, either juvenile or adult, will reoffend is a prior history of sexual offenses (Rassmussen, 1999). Hanson (1997) found that both prior convictions and prior charges can be predictive of new offenses. Any indication that an offender has been known to engage in a variety of sexual crimes, increases their likelihood to recidivate (Hanson & Bussiere, 1998). Adolescents and juveniles who commit two or more offenses are at a higher risk to reoffend than those who have committed a single offense (Worling & Curwen, 2001). Family violence has also been linked to juvenile sexual offending (Fagan & Wexler, 1988). According to Alijazireh (1993), family dysfunction is common in the homes of juvenile sex offenders. Monastersky and Smith (1985) concluded that studies are virtually unanimous in identifying the family as a crucial influence in the development or elicitation of the offending behavior of juvenile sex offenders. In a study conducted by Ryan and Lane (1991), only 27.8% of their sample lived with both natural parents. Studies on the family environments of JSOs have identified high rates of physical violence that is witnessed by the adolescent (Haapasalo & Hamalainen, 1996; Lewis, Shanok, & Pinus, 1981; Mio, Nanjundappa, Verleur, & Dobkin de Rios, 1986; Smith, 1988; Spacarelli, Bowden, Coatsworth, & Kim, 1997). 42 In a study on caregivers of children with sexual behavior problems, Pither et al. (1998) found that 95% had been sexually abused, 48% had been victims of physical abuse, and 11% of their sample had been found to be neglected. Early research has found that neglect is the form of abuse that is most predictive of later violent delinquency and equal to the import of sexual abuse for later sexual offending (Widom, 1995; Widon & Maxfiled, 1996). It has been further investigated and found that abusive men tend to be impulsive and that they pass their lack of impulse control to their sons through social learning or genetic mechanisms (Arias & O?Leary, 1988; Patterson & Capaldi, 1991; Hur & Bouchard, 1997). The poor impulse control may then play a role in the son?s sexual offending behavior (Caputo, Frick, & Brodsky, 1999). The family structures of JSOs tends to be very rigid and enmeshed or very chaotic with a great deal of role confusion (Knopp, 1982). According to Monastersky and Smith (2000), most studies on JSOs tend to find a family environment that is rigid with sexual pathology within a parent. In many cases, one of the parents may have demonstrated deviant behavior very similar to that of the JSO (Knopp, 1982). It is not unusual for JSOs to have been exposed to aggressive role models, substance abuse and pornography (Gerardin, & Thibault, 2004). Smith (1988) found that the severity of sex offenses committed by JSOs has been found to vary as a function of degree of violence directed toward the 43 offender?s mother. Studies of JSOs also commonly find that there has been a parental loss and/or separation (Ryan & Lane, 1991). Gender of the victim in and of itself is not predictive of sexual reoffense for JSOs; however, studies have found that male JSOs that chose male victims were more likely to commit subsequent sexual offenses (Smith & Monastersky, 1986; L?ngstr?m & Grann, 2000). The selection of stranger victims by JSOs is also predictive of future recidivism (Quinsey, Rice, & Harris, 1995). When victims are known by the offender, there tends to be grooming and time involved. With victims who are strangers, once they are identified, the offense tends to happen quickly. That makes the stranger selection a high risk factor (Worling & Curwen 2001). The dating and sexual history of JSOs has also been identified as part of the profile of the JSO that one might examine (Maxwe, Robinson, & Post, 2003). In a study conducted by Hall and Flannery (1984), adolescents who were raped were found to be in peer groups that were sexually active. According to Elliot (1994) and Weinrott (1996), juveniles who rape are not likely to continue this form of sexual aggression into adulthood; however, they may be at increased risk for non sexual recidivism. Offenders listed styles of dress as well as behaviors exhibited by the victim prior to the assault as reasons for sexual assault. A final 44 consideration of a risk profile for juvenile sex offending is their naivet? about what constitutes a sexual assault. Dynamic Factors Dynamic factors (changeable) on the other hand, look at offenders at various stages of their treatment and serve best as predictors of future behaviors. Dynamic factors are those that can change over the course of time or treatment. These factors are things such as victim empathy, motivation to change, acceptance of responsibility, level of self regulation and chemical abuse issues. Victim empathy is a factor that influences juveniles who sexually offend. Many JSOs lack the ability to feel connected to others. The inability of a sex offender to empathize with their victims has been frequently implied in the literature (Hayashino et al., 1995). According to Bumby (2000), JSOs who lack empathy are more likely to continue offending. Research has identified that between 25% and 50% of JSOs blame their victims for their behaviors of assault and hold rape-supported attitudes (Elliot, Ageton, Huizinga, Knowles, & Cantor, 1983). Kahn and Chambers (1991) conducted a study and found that subsequent convictions for sexual assault tended to happen to those adolescents who blamed their victims. In a study conducted by Lakely (1992), mythical beliefs, thinking errors and faulty attitudes of juvenile offenders in a private residential facility were 45 examined. Faulty beliefs of the assaulter fell into four categories: Pretentiousness, uniqueness, responsibility and distorted values. Pretentiousness usually occurs when the JSO gets the initial idea to offend, forms an opinion based on myth or faulty information concerning the offense, and commits the offense based on the ill gotten and/or conceived information (Yochelson & Samenow, 1976). The juveniles in this study tended to think of themselves as unique. They believed they would be caught and that rules that apply to others did not apply to them (Lakely, 1992). According to Groth (1979), offenders tend to shift blame so that the responsibility is placed on the victim or circumstances and not on the offender. The final category is that of distorted values. Yochelson and Samenow (1977) observed that criminals stretch the rules and norms that govern most human behavior. The juveniles in this study tended to take established values and distort them to justify their offending behaviors. Juvenile sex offenders tend to deny allegations of abuse and minimize the huge impact their offending behaviors have on their victims (Gerardin & Thibault, 2004). Research also shows that academic achievement and substance abuse is a profile correlate to sex offending among adolescents (Fehrenbach, Smith, Monastersky, & Deisher, 1986). JSOs tend to have learning disabilities, are in remedial classes and frequently exhibit behavior problems while at school (Awad, Saunders, & Levene, 1984). According to Gerardin and Thibault (2004), 46 JSOs are impulsive and lack social skills. Conduct disorder is the most common DSM diagnosis to adolescents that offend (Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice, 2001). Alcohol and/or drug use has been known to increase the likelihood of sexual victimization among same age peers (Maxwell, Robinson, & Post, 2003). An investigation of medical records for victims of JSOs revealed that of those victimized by peers, 26% were under the influence of drugs and/or alcohol (Jenny, 1988). Dynamic factors can be further broken down into stable and acute factors. Stable factors are those that endure for long periods of time but can be changed (e.g., deviant arousal patterns or substance abuse problems). Acute factors are those things that change rapidly following the crime such as sexual arousal and drunkenness (Hanson, 1998). Not surprisingly, victim access has been a dynamic factor known to increase the likelihood of reoffending (Hanson, & Harris, 1998). Deviant sexual interest is a major risk factor for both juvenile and adult offenders (Hansom & Bussiere, 1998). In a study conducted by Kenny, Keogh, and Seidler (2001), deviant sexual fantasies were found to be related to JSO recidivism. Deviancy can be defined as any conduct or norm that strays significantly from the standard or norm (Steele & Ryan, 1997). Sexual deviance can further be defined as sexual interest or paraphilias that produce intense sexual urges, fantasies, and/or behaviors that include but are not limited to 47 nonhuman objects, children or other non-consenting individuals, and the suffering or humiliation of self and/or others (American Psychiatric Association, 2000). According to Worling and Curwen (2000a), juveniles who self-report a sexual interest in children and those who exhibit grooming behaviors and participate in penetrative acts with younger children, present a high risk of sexually re-offending. Also, preoccupation with compulsive sexual thoughts and gestures (masturbation) are strong predictors of sexual reoffense (Epps, 1997; Lane, 1997, Prentky, et al., 2000; Ross & Loss, 1991; Steen & Monnette, 1989). Reductions in dynamic risk factors can lead to reduced recidivism rates (Rich, 2003). Risk Assessment Measures There are two main types of measurements used to assess risk. They are actuarial and clinical. Actuarial assessments are conducted by looking at risk factors that an individual has and coding those factors in a way to arrive at a score that would determine the level of risk (Beech, Fisher, & Thornton, 2003). Actuarial assessments tend to target static factors or those things that do not change (i.e. number of offenses, victim selection, and prior arrest). Clinical assessment is direct contact with the clinician and the individual for which the assessment is being done. The clinician uses his/her judgment to 48 make a recommendation about risk based on interviews, observations, live interaction, and the available history of an offender (Rich, 2003). Actuarial Assessments It is a widely held belief among researchers that actuarial assessments are more predictable than clinical judgments (Bonta, Law, & Hanson, 1998; Groggin, 1994; Hanson & Bussiere, 1996; Hood et al.; McNeil, Sandberg & Binder, 1998; Mossman, 1994). According to Hanson (2000), the predictive accuracy of predicting who will re-offend through clinical judgment is only slightly above chance levels (r = .10). Whereas actuarial assessments are governed by set questions and rules, the clinical assessment is not. Actuarial assessments are fashioned by set rules and computed using algorithms based on statistical properties (Rich 2003). In a recent study, Grove, Zald, Lebow, Snitz, and Nelson (2000) reported that mechanical predictions were 10% more accurate than clinical predictions and that on specific analysis, mechanical predictions out performed clinical predictions in 33%- 47% of examined studies. Moreover, Monahan (1996) stated that predictions about those who are most likely to commit sexual reoffense can be improved by using instruments that have been empirically validated. Actuarial assessments are primarily based on static (fixed) factors as opposed to dynamic (changeable) factors. Historically actuarial assessments that 49 have been used with adult sexual offenders have also been used for juvenile sexual offender. The problem with dual use of these instruments is that assessments that are devised for adults do not take into consideration that adolescence is a time of developmental flux (Witt et al., 2002). Clinical Assessment Clinical assessments of risk rely on an opinion regarding the likelihood of reoffense using case files, interviews and formal testing (Hanson, 1998). The unstructured clinical assessment tends to yield the least reliable results. The characteristics of this assessment are (a) clinician determines the questions and constructs to measure, (b) the administration is flexible, (c) there are potential multiple data sources, (d) there is heavy reliance on clinical interview, (e) it tends to be more intuitive with idiosyncratic algorithms for determining risk, and, (f) there is no validation or reliability of data (Witt et al., 2002). Clinical assessments of risk tend to receive most of its criticism from the fact that it is unstructured. However, guided clinical assessments have proven to be more accurate in determining risk (Dorem, 2002). A clinical assessment that is empirically guided takes into consideration base rates as well as risk factors known to be related to reoffending (Hanson, 1998). Empirically guided clinical assessment also has the added benefit of having a higher rate of accuracy, with the scientific evidence used to support the risk factors being evaluated (Worling 50 & Cuwen, 2001) and are also more systematic and tend to lead to better agreement among those involved in its administration (Boer et al., 1997). Therefore, if clinical assessments are to be credible, they must be empirically sound in order to produce reliable results. Clinical risk assessment instruments traditionally used on JSOs are the Juvenile Risk Assessment Tool (JRAT), Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR), and the Juvenile Sex Offender Assessment Protocol (J-SOAP). The J-RAT is an empirically based, structured assessment instrument designed to assess risk for sexual reoffense. This tool is primarily used in the initial stages of assigning levels of risk to offenders in residential treatment centers (Witt et al., 2002). J-RAT The J-RAT uses 12 risk domains to assess risk. They are responsibility, relationships, cognitive ability, social skills, past trauma, personal characteristics and qualities, co-morbidity and treatment, substance abuse, antisocial behaviors, patterns of sexual offending behaviors, family factors and environmental conditions. Each individual element in the domain is scored based on a level of risk or low, moderate, or high. The final overall risk assessment is assigned for the domain. It should be used in conjunction with other tools such as record reviews and clinical interviews with the offender. 51 When scoring the J-RAT, if an individual is assigned a risk level of low in any given element or domain, it is unlikely that it will contribute to the possibility of sexual re-offense. By contrast, if assigned a risk level of moderate on an element or domain, it is likely that the element and/or domain will significantly contribute to the possibility of sexual re-offense. Finally, a risk assessment level of high on any given element and/or domain suggests that the particular element and/or domain is highly likely to contribute to, influence, or allow the possibility of sexual re-offense. The elements within the domain are scored using ?+? and ?-?. If an element is given a positive rating (?+?), this means there is movement towards significant concern. An element that receives a negative rating (?-?), shows there is movement away from significant concern. If there has been no history on a specific element or domain it should be recorded as none. If an element or domain is unknown it should be rated with a ?+? and treated as high risk. ERASOR The Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR) (Worling & Curwen, 2001) is an empirically guided scale. It is designed to estimate the risk of sexual recidivism in individuals (not necessarily males) ages 12-18 who have committed a sexual offense. This scale looks at five domains that include both static and dynamic risk factors when assessing risk. They are (1) 52 sexual interest, attitudes and behaviors, (2) historical sexual assaults, (3) psychosocial functioning, (4) family/environment functioning, and (5) treatment. There is also a category that is reserved for ?other? information. This can be information that is not listed in the five domains but is of particular relevance to his/her patterns of sexual offending. It is designed to determine short term risk (3 years or less) as opposed to long term risk (Worling, 2004). The ERASOR was designed based on studies of juvenile sexual offending, published checklist and guidelines detailing risk assessment of this population and on published literature on adult sex offending recidivism (Worling & Curwen, 2000b). The items in the ERASOR look at possible pathways to re- offense (Prescott, 2004). They are scored as present, partially/possibly present, not present, and unknown. Some items have additional information that should be scored and should be given careful consideration before being scored (Worling & Curwen, 2006a). To be used most effectively, evaluators using the ERASOR should have a working knowledge of juvenile sexual offending as well as family systems. They should also know the current research as it relates to recidivism and JSOs. Worling and Curwen (2001) state the following are important considerations for those evaluating JSOs with the ERASOR: the evaluator should (a) assess multiple domains of the offender?s functioning, including sexual, intrapersonal, 53 interpersonal, familial, and biological, (b) use multiple methods of data collection to form opinions regarding risk (methods could include clinical interviews, psychological tests, behavioral observation, medical examinations, and reviews of previous case records and reports), (c) collect information from multiple sources such as the offender, the victim(s), the police, family, friends, and other mental health professionals who are familiar with the offender and his/her family, (d) collect information regarding both static (historic and unchangeable) and dynamic (variable and potentially changeable) factors, (e) always be cognizant of the validity of the information that they are using in forming risk predictions and should state any reservations or qualifications in their reports, and (f) recognize that risk assessments will become obsolete after the passage of time and/or following a change in any of the risk factors that were assessed. J-SOAP The Juvenile Sex Offender Assessment Protocol (J-SOAP) was developed by Prentky, Harris, Frizzell, and Righthand (2000). It is a check-list that is used to evaluate risk factors known by research to lead to sexual reoffense or recidivism on four subscales. The J-SOAP is an attempt at an actuarial assessment that scores offenders on both static (fixed) and dynamic (changeable) risk factors. The static factors include sexual drive/preoccupation and impulsive/antisocial 54 behavior. The dynamic factors include intervention items and community stability/adjustment. The J-SOAP was developed for the use of males adjudicated sex offenders between the ages of 12-18 (Prentky & Righthand, 2003). The J-SOAP is the most widely used instrument in the assessment of risk in youth. It can be further used for males in that specific age range who are not adjudicated but who have a history of sexually inappropriate behavior. Prentky and Righthand are clear in stating that levels of risk cannot be determined by the J-SOAP alone, and should take into consideration other methods of evaluating and assessing risk. The J- SOAP is based on 28 items with four subscales. The subscales are sexual drive/ sexual preoccupation, impulsive/antisocial behavior, clinical/treatment, and community adjustment. Scoring of the J-SOAP is done on a scale of ?0? through ?2? with ?0? meaning absence of the specific offense, ?1? meaning some evidence of the offense and ?2? showing strong evidence that the specific offense has occurred. Correct scoring of the J-SOAP is of critical importance and cannot be stated enough. The authors suggest if possible, the scales should be administered by two independent clinicians using as many sources of information as possible. Those scores should then be compared, discussing the differences. A final score, based on an agreement between the raters, should be given. The final J-SOAP 55 score is obtained by adding the scores of each subscale and producing a total score. Based on this score, a risk level of low, moderate, or high is assigned. Crimes of a sexual nature rank along with murder in the negative public reaction they receive (Quinn, 2003). Victims of sexual offenses carry their victimization into every phase of their lives including adulthood. There is no cure of sexual offending but with specialized treatment by qualified professionals, many JSOs can receive offender specific treatment. This treatment will enable the offenders to manage their deviant behavior and explore the origins of their cognitive distortions. The Role of the Counselor Sex offenders are, perhaps, the most difficult groups to treat and supervise (Stalans, 2004). From 1983 to 1993, the number of programs that treat JSOs increased from 20 to 800 (Freeman-Longo, Bird, Stevensen, & Fiske, 1995). Due to the large numbers of sexual offenses perpetrated by adolescents (Cashwell & Caruso, 1997), counselors will at some point interact and play a vital role in the rehabilitation of these offenders. Historically, counselors were on the front line in reporting and treating victims of sexual abuse. Conversely, the role of mental health counselors are changing and their clients now include those youth and adolescents who have perpetrated sexual abuse against others (Cashwell, Bloss, 56 & McFarland, 1995). The Association for the Treatment of Sexual Abusers (ATSA) believes that JSOs are amenable to treatment (ATSA, 2001). JSOs enter treatment with a myriad of risk factors that spring from various demographic characteristics (Kelly, Lewis, & Sigal, 2004). Successful treatments include the responsibility of assessing risk. Psychometric testing and a clinical interview are typical components of the assessment process (Gerardin & Thibault, 2004). The role of clinicians in the assessment of risk is to observe, draw conclusions, and take action and to make decisions informed by explorations and formulations. According to Rich (2003), transferring assessment from the clinician to the psychometrician removes the human interaction element that is essential to treatment of JSOs. Therefore, the role of the clinician is vital in risk assessment. The therapeutic relationship is a critical component of the assessment process. Therapeutic intervention is based on the presumption that it will reduce the incidences of sexual re-offense and the likelihood of the JSO going on to become an adult offender (Knopp, 1985). It is important first and foremost for the counselor to build and establish and trusting relationship with the offender. It is during this process that the counselor will gleam vital information about the offender, their family constellations, and the events and circumstances that led to 57 their offending. Motivation to change is an important often neglected component in therapy (O?Reilly, Morrison, Sheerin, & Carr, 2001). Counseling sex offenders calls for counselors to examine their feelings on sexual abuse and those who are sexually abusive (Poison & McCullom, 1995). Sex offenders are historically viewed in a more negative light than are most clients receiving services from counselors (Wodarski & Whitaker, 1989). It is imperative that counselors who work with sex offenders ensure that their biases do not negatively impact the delivery of services to this population. In a study conducted by Farrenkopf (1992), female counselors working with sex offenders tended to having feelings of paranoia, vulnerability and hyper vigilance in regards to their own and their children?s safety. Male counselors tended to feel guilt regarding the male abusive behavior. Two negative themes reported among counselors that work with sex offenders was a loss of innocence and trust (Scheela, 2001) and desensitization and emotional hardening to hearing about the abuses (Edmunds, 1997; Ellerby, 1997; Frankenkopf, 1992). Treatment Planning In working with JSOs there is a need for combined and integrated treatment approaches (Becker, 1994). The emotional, behavioral, and developmental problems presented by JSOs require an integrated, multimodal treatment program which is tailored to the individual (Becker & Hunter, 1993). 58 Hunter and Figueredo (1999) found that the degree of sexual maladjustment, denial, and the lack of a sense of accountability for one?s sexual offenses predicted failure in treatment compliance in a community-based intervention program. Juvenile sexual abusers usually present with considerable psychiatric comorbidity. A number of studies have documented the frequency and diversity of the psychiatric impairment (Becker et al., 1991; Kavoussi et al., 1988; Lewis et al., 1979; Shaw et al., 1993, 1996, 1999). Psychiatric diagnoses of conduct disorder, depression, anxiety, and substance abuse are common as well as evidence of character pathology and personality disorders. Most juvenile sexual abusers, however, do not meet diagnostic criteria for paraphilia nor do they have well- established patterns of deviant sexual arousal. An essential element in treatment planning is the evaluation of the severity of the sexual offending behavior and the risk of recurrence of sexual offending behavior. This is a difficult task even when the judgments are made by experienced clinicians (Kahn & Chambers, 1991). Determining treatability has been related to the sexual abuser?s willingness to accept accountability for his or her sexual offenses manifested by (1) admission of the sexual offense, (2) acceptance of the sexual offending behavior as a problem, (3) motivation to stop sexual offending, and (4) 59 willingness to participate fully in treatment. Other reports suggest some success with adult sexual abusers who initially denied their sexually abusive behavior (Maletzky, 1996; Schlank & Shaw, 1996). In the final stages of evaluation, the counselor must discuss the possible treatment alternatives with the client and appropriate family members and to explain to the family members what their participation in the treatment program will be. The predominant treatment approaches used with JSOs include cognitive-behavioral and psychosocial therapies and psychopharmacological interventions (Shaw, Funderburk & Schlank 1999). The treatment of the juvenile sex offender has generally focused on a number of goals (Becker, 1994; Becker & Hunter, 1997; Ryan et al., 1987): (a) confronting the offender's denial, (b) decreasing deviant sexual arousal, (c) facilitating the development of nondeviant sexual interests, (d) promoting victim empathy, (e) enhancing social and interpersonal skills, (f) assisting with values clarification, (g) clarifying cognitive distortions, and (h) teaching the juvenile to recognize the internal and external antecedents of the sexual offending behavior. Counseling Interventions Individual vs. Group Interpersonal therapies include traditional individual approaches, family therapy, group therapies, and the use of the therapeutic community. Individual 60 counseling is important with JSOs, however, family and group counseling has proven to be more effective (Breer, 1987) in terms of reducing recidivism. Individual counseling with JSOs is designed to provide holistic treatment by addressing a range of problems (Cashwell & Caruso, 1997). The first element that should be addressed in individual counseling is the JSOs denial (Davis & Leitenberg, 1987). This is done through facilitating the acceptance of responsibility by the JSO (Kahn & Lafond, 1988). Another dynamic to address individually, is victim empathy. The JSO feels empathy when he is able to re- experience the pain associated with personal victimization (Burgess, Hartman, McCormack, & Grant, 1988). This can be accomplished through the reading of victim impact statements and the use of cinema therapy or the showing and discussion of movies that depict victimization (Ryan, Lane, Davis, & Isaac, 1987). Other concerns addressed through individual counseling are: awareness of triggers, victimization issues of the offender and education about human sexuality. The therapy of juvenile sex offenders is generally characterized as one of firmness and confrontation alternating with a flexible and sympathetic stance (Muster, 1992). Confrontation is necessary to address the minimizations, denial, rationalization, and cognitive distortions which the offender presents to authority. This approach is balanced with a sensitive awareness of the offender?s 61 developmental, behavioral, and emotional problems, which not infrequently emanate from his or her own childhood history. A final and important component of individual counseling is cognitive restructuring. This occurs when a JSO?s distorted beliefs, that enable and support deviant behaviors, are confronted and challenged (Ryan et al., 1987). While historically individual therapy has been a valuable intervention, it has had limited value for the individual sex offender and probably should never be relied on as the only treatment model. The advantages of individual therapy are that it provides a greater sense of confidentiality and an opportunity to develop trust in the therapeutic process. A group setting is the preferred format in treatment programs for sexual abusers and is usually the conduit through which cognitive-behavioral modalities (such as psychoeducational, behavioral, and relapse prevention programs) are conducted. Group counseling is also an important element in the holistic treatment of JSOs. The cognitive-behavioral or relapse prevention model is the most widely used approach in treating JSOs (Freeman-Longo, Bird, Stevenson, & Fisk, 1995). Generally, the group setting is ideal for cognitive- behavioral therapies. Groups targeted for the sexually abusive behavior of youth should allow for opportunities to develop interpersonal skills and affective expression as well as exploration of sex-role issues (Rencken, 1989). 62 Relevant Theories of Counseling Family Therapy Family therapy facilitates the learning of new ways of communicating and building a support system which will help interrupt the abuse cycle and ultimately be supportive to the offender?s capacity for regulating and modulating sexual aggression (Schwartz, 1988; Sholevar & Schwoeri, 1999). Bischof et al. (1995) suggest that intervention strategies which have been proven effective in other delinquent groups may be effective with the families of adolescent sex offenders. Family therapy may be warranted in those instances where there is incest, especially when the sex offender remains in the family or will rejoin the nuclear family after treatment. Understanding family dynamics is also crucial in treating JSOs. Although crucial, it is also a difficult component (Gerardin, & Thibault, 2004). Family therapy is the least provided treatment approach due to the lack or resources in areas of time and/or cost (Rich, 2003). Because of the dynamics of the family constellation of most JSOs, it is no surprise that conducting family therapy presents a challenge for the therapist. However, according to Thomas (1997), there is a family attached to every adolescent entering treatment for sex offending, and that in and of itself is a reason to provide family therapy. 63 According to Lundrigan (2000), family therapy for youth that sexually offend should include the following components: (a) ongoing general communication, (b) ongoing general support, (c ) family meetings with and with the youth, (d) family psycho-education, (e) family education and support group, and (f) family therapy. Providing the services necessary, families are strengthened. They are able to reduce family risk factors and increase the quality of protective factors for the entire family, to include, the juvenile offender and other children within the family. While treatment has been quite successful in reducing recidivism, adolescent sex offenders are not ?cured.? Treatment endeavors are organized to facilitate the sexual abuser?s development of coping and adaptive strategies to prevent further sexual offenses. Cognitive Interventions Psychoeducational modules are didactic experiences that provide sexual abusers with information about sexuality, sexual deviancy, cognitive distortions, and interpersonal and social behaviors, as well as strategies for coping with aggressive and sexual impulses and anger management (Becker & Hunter, 1997; Green, 1988). This approach assumes that the offender has acquired a set of beliefs, attitudes, and expectancies which have shaped his or her sexual offending behavior and that the sexual behavior is maladaptive, contains 64 ?thinking errors,? and is associated with impaired communication and social skills (Johnston & Ward, 1996). These modules are taught by a therapist who often uses workbooks and homework assignments. The setting is usually a classroom, although the intervention may take place in the context of ongoing group therapy. There is an emphasis on understanding the general patterns and determinants of sexual offending behavior, sex offender characteristics, and the spectrum of sexual offenses. The psychoeducational modules usually addressed are: 1. Victim Awareness/Empathy. The focus is on understanding the effects of sexual assault on the victim, identifying cognitive distortions and myths that support the sexual assault, and promoting participation in therapeutic endeavors. 2. Values Clarification. The therapist clarifies sexual values as they relate to the cessation of exploitative sexual relationships. 3. Cognitive Restructuring. This is an attempt to correct the cognitive distortions and the irrational beliefs that support the sexual offending behavior and to replace them with reality-focused and culturally acceptable beliefs. 65 4. Anger Management. Instruction is provided to facilitate the recognition and the development of appropriate coping strategies for managing anger. 5. Assertiveness Training. Training is provided to promote more appropriate self-assertive behavior to have one?s needs satisfied in a reality-oriented and culturally acceptable manner. 6. Social Skills Training. The therapist facilitates more effective prosocial behaviors, communication skills, and interpersonal awareness. 7. Sexual Education. The therapist provides information regarding human sexuality, myths, sex roles, and variations of sexual behaviors. 8. Stress Reduction/Relaxation Management. Techniques for coping and reducing stress, anxiety, and frustration are made available to the group. 9. Autobiographical Awareness. Emphasis is on the individual developing an understanding of his or her own life trajectory and how the pattern of sexual offending behavior evolved over time. 66 Behavioral Interventions Behavioral interventions have been used to diminish deviant sexual arousal and have been reported to be varyingly successful (Dougher, 1988b). Some of the techniques are as follows: 1. Covert Sensitization. In this counterconditioning paradigm, the offender learns to extinguish pleasurable responses to sexually stimulating deviant imagery through the imagining of some negative reaction or aversive stimulus. Scenes are constructed for each offender according to his or her preferred sexual-erotic fantasies (Cautela, 1966). 2. Assisted Covert Sensitization. Aversive stimuli such as noxious odors are used to facilitate an aversive reaction (Maletzky, 1974). 3. Imaginal Desensitization. The sex offender uses relaxation techniques to interrupt the sexually stimulating imagery and to inhibit the sexual arousal cycle (McConaghy et al., 1989). 4. Olfactory Conditioning. Sexually stimulating deviant imagery is presented which is followed by the presentation of a noxious odor. 5. Satiation Techniques. This involves either verbal or masturbatory satiation. The offender is first encouraged to masturbate to ejaculation in response to socially appropriate sexual fantasies with 67 the concomitant feelings of affection and tenderness. After this experience the offender is required to masturbate to deviant sexual fantasies. If the offender becomes aroused, he or she is told to switch to an appropriate fantasy or in some instances exposed to an aversive stimulus such as ammonia (Gray, 1995). Verbal satiation requires the dictation on an audiotape of the most stimulating paraphiliac imagery for at least 30 minutes after masturbation 3 times a week. It is assumed that the paraphiliac fantasy becomes boring and subsequently extinguished (Schwartz, 1992). 6. Sexual Arousal Reconditioning. This involves the pairing of sexual arousal with appropriate nondeviant sexual stimulation or sexual fantasies. Relapse Prevention Relapse prevention was originally developed as an intervention for substance abusers but was subsequently modified for sexual abusers (Pithers et al., 1983, 1988a, 1988b; Pithers & Gray, 1996). Ninety percent of all sex offender treatment programs in North America report using relapse prevention (Pithers & Gray, 1996). This intervention strategy assumes that sexual offenses are not capricious happenings but are the product of contextual triggers and an array of emotional and cognitive precursors. The treatment process entails the explication 68 and definition of each phase of the sexual assault cycle, i.e., the unique characteristics of each offender's cycle so that the offender will be aware of the triggers which initiate the cycle so that he or she will be alerted and use new strategies for interrupting the sexual assault cycle (Ryan et al., 1987). Some of the emotional states that have been found to be important emotional triggers are boredom, social or sexual embarrassment, anger, fear of rejection, and numbness (Gray & Pithers, 1993). Proulx et al. (1996) found that ?negative moods and conflicts? such as anger, loneliness, and humiliation coincided with deviant sexual fantasies and increased masturbatory behavior. The goals of relapse prevention are to empower the offender to manage his or her own sexual life through a cognitive understanding of the antecedents of the sexual offending behavior and through the development of coping strategies with which to interrupt the sexual offending cycle. Chapter Summary The National Adolescent Perpetrator Network (1993) stressed the primary objective of interventions with juveniles who have sexually offended is community safety. Juvenile sex offenders can be treated using a variety of interventions. Becker and Hunter (1997) described the main treatment objectives as preventing further victimization, halting the development of additional 69 psychosexual problems, and helping the juvenile develop age-appropriate relationships with peers. It is important to note that a sexual abuser is never cured but is rehabilitated. There is a need to provide monitoring and follow-up with continuing services. After the termination of a course of therapeutic interventions, the offender should be maintained in a spectrum of continuing services which resonate with the severity of the sexual misbehaviors and psychopathology and which may include community-based outpatient treatment programs, specialized group homes, specialized foster care programs, and other specialized follow-up services. 70 III. RESEARCH METHODOLOGY This section describes the procedures that were used to investigate how specific variables contribute to sexual reoffense in juvenile sex offenders. This study intends to increase the knowledge on predictors of reoffense in juvenile sex offenders. The study will examine specific data (age, IQ score, parental marital status, severity of physical abuse, severity of emotional abuse, severity of sexual abuse) to determine which may better serve as predictors of sexual reoffense. Because there are no empirically validated risk assessment instruments for juvenile sex offenders (Prescott, 2004), select variables of the Static-99 will be used to determine its usefulness in assessing risk in JSOs. The Static-99 is an empirically validated risk assessment instrument that has been normed on adult male sex offenders. The variables include: (a) prior sex offenses, (b) prior sentencing dates, (c) any convictions for non-contact sex offenses, (d) current convictions for non-sexual violence, (e) prior convictions for non-sexual violence, (f) unrelated victims, (g) stranger victims, (h) male victims. A study focusing on the factors that are likely to predict the risk of sexual re-offense, is not only 71 appropriate, but indeed, necessary if we are to safeguard both past and future victims. Sample The samples for this study were collected from the closed therapy files of The Texas Youth Commission (TYC) that is located in Austin, Texas. The center serves both male and female adjudicated, sex offenders between the ages of 10-17 (for the purpose of this study we will be looking at male juvenile offenders). The treatment center provides specialized treatment for sex offenders (Prior to FY 2004, this treatment also was provided by specialized contract providers.) The center has over 700 closed therapy cases of youth with sexual behavior problems. The sex offender treatment program (SOTP) builds on the agency?s resocialization program using cognitive-behavioral strategies and a relapse prevention component. It is important to note that although the study will give descriptive information on 661 male juveniles (between the ages of 10-17 at the time of intake) who had been adjudicated for sexual offenses for the purpose of the regression analysis the SPSS program chose only those samples that had complete information in the variables of interest. The sample size for the regression analysis will be 179 (df = n-1). 72 Collection of Data After signing a research agreement with the TYC and gaining the approval of the Auburn University Institutional Review Board, the research director of the TYC granted permission for the use of archival data collected on juvenile sex offenders. The research director at the agency created an excel spreadsheet with the requested information from the closed therapy files. The information was gathered using a coding system that guaranteed anonymity of the client?s identity. Once the information was gathered it was sent to the researcher electronically. Variable Selection The purpose of this study is to find out the contributing factors of recidivism among juvenile sex offenders as well as determining the usefulness of the Static-99 in assessing risk in JSOs. The data collected from the archives were: (a) age at time of intake, (b) IQ score, (c) parental marital status, (d) severity of emotional abuses, (e) severity of physical abuse, (f) severity of sexual abuse, (g) static variable 1 the (prior sex offenses), (h) static variable 2 (contact sex offenses with stranger), (i) static variable 3 (contact sex offense with non-relative), (j) static variable 4 (contact sex offense with male victim); (k) static variable 5 (non contact sex conviction), (l) static variable 8 (index non sexual assault), (m) static variable 73 9 (prior non sexual assault), and (n) static variable 10 (more than four sentencing occasions). Measures The Static-99 was developed by R. Karl Hanson, Ph.D. of the Solicitor General Canada and David Thornton, Ph.D., at that time, of Her Majesty?s Prison Service, England. The Static-99 was created by amalgamating two risk assessment instruments (RRASOR and SACJ-MIN). The RRASOR (Rapid Risk Assessment of Sex Offender Recidivism), developed by Dr. Hanson, consists of four items, 1) having prior sex offenses, 2) having a male victim, 3) having an unrelated victim, and 4) being between the ages of 18 and 25 years old. The items of the RRASOR were then combined with the items of the Structured Anchored Clinical Judgment Minimum (SACJ-Min), an independently created risk assessment instrument written by Dr. Thornton (Grubin, 1998). The SACJ-Min consists of seven items, 1) having a current sex offense, 2) prior sex offenses, 3) a current conviction for non-sexual violence, 4) a prior conviction for non-sexual violence, 5) having 4 or more previous sentencing occasions on the criminal record, 6) being single, 7) having non-contact sexual offenses. This instrument provides explicit probability estimates of sexual reconviction, is easily scored, and has been shown to be robustly predictive across several settings using a variety of samples (Harris, Phenix, Hanson & Thornton, 2003). 74 The RRASOR was designed to predict recidivism using a small number of easily scored variables to include both prior and no prior offenses as well as victim selection, marital status and age (less than 25 years old) (Hanson & Thornton, 1999). It evaluates four static risk factors and yields five and ten year estimates of risk (Wood & Cellini, 1999). The SACJ-min was designed to look at violent and sexual recidivism in stages. Stage one considered all official convictions; stage two looked at aggravating factors, and the final stage (three) took into consideration prior treatment for sexual or violent offenses (Grubin, 1998; Hanson & Thornton, 1999). Although widely used on adult sex offenders the Static-99 (Hanson & Thornton, 1999) is an actuarial tool that has been used to assess risk on JSOs (Witt et al., 2002). It uses only static (unchangeable) factors that are known to correlate to sexual recidivism in adult males (Harris, Pheonix, Hanson, & Thornton, 2003). The Static-99 is generally used on male sex offenders who have had at least one conviction for sexual offense against a child or non-consenting adult and is designed to measure long term risk potential (Hanson & Thornton, 1999). Risk factors listed on the Static-99 have been empirically linked to sexual recidivism. The ten scored items on the Static-99 are (a) prior sex offenses, The Basic Principle: This item and the others that relate to criminal history and the 75 measurement of persistence of criminal activity are based on a firm foundation in the behavioral literature. Thorndyke (1911) stated that the ?the best predictor of future behavior, is past behavior?. Andrews and Bonta (1998) state that having a criminal history is one of the ?Big Four? predictors of future criminal behavior. More recently, and specific to sexual offenders, a meta-analytic review of the literature indicates that having prior sex offenses is a predictive factor for sexual recidivism (Hanson & Bussi?re, 1998); (b) prior sentencing dates, The Basic Principle: This item and the others that relate to criminal history and the measurement of persistence of criminal activity are based on a firm foundation in the behavioral literature. Prior sentencing occasions is a convenient method of coding the length of the criminal record; (c) any convictions for non-contact sex offenses, The Basic Principle: Offenders with paraphilic interests are at increased risk for sexual recidivism. For example, most individuals have little interest in exposing their genitals to strangers or stealing underwear. Offenders who engage in these types of behaviors are more likely to have problems conforming their sexual behavior to conventional standards than offenders who have no interest in paraphilic activities; (d) current convictions for non-sexual violence, The Basic Principle: A meta-analytic review of the literature indicates that having a history of violence is a predictive factor for future violence (Hanson & Bussi?re, 1998). The presence of non-sexual violence predicts the seriousness of damage if 76 a re-offense were to occur and is strongly indicative of whether overt violence will occur (Hanson & Bussi?re, 1998). This item was included in the STATIC-99 because in the original samples, this item demonstrated a small positive relationship with sexual recidivism (Hanson & Thornton, unpublished data); (e) prior convictions for non-sexual violence, The Basic Principle: A meta-analytic review of the literature indicates that having a history of violence is a predictive factor for future violence (Hanson & Bussi?re 1998). The presence of non-sexual violence predicts the seriousness of damage if a re-offense was to occur and is strongly indicative of whether overt violence will occur (Hanson & Bussi?re, 1998). This item was included in the STATIC-99 because in the original samples this item demonstrated a small positive relationship with sexual recidivism (Hanson & Thornton, unpublished data); (f) unrelated victims, The Basic Principle: Research indicates that offenders who offend only against family members recidivate at a lower rate compared to those who have victims outside of their family (Harris & Hanson, Unpublished manuscript). Having victims outside the family is empirically related to a corresponding increase in risk; (g) stranger victims, The Basic Principle: Research shows that having a stranger victim is related to sexual recidivism. A victim is considered a stranger if the victim did not know the offender 24 hours before the offense. Victims contacted over the Internet are not normally considered strangers unless a meeting was 77 planned for a time less than 24 hours after initial communication; (h) male victims, The Basic Principle: Research shows that offenders who have offended against male children or male adults recidivate at a higher rate compared to those who do not have male victims. Having male victims is correlated with measures of sexual deviance and is seen as an indication of increased sexual deviance (Hanson & Bussi?re 1998); (i) young, The Basic Principle: Research (Hanson, 2001) shows that sexual offending is more likely in an offender?s early adult years than in an offender?s later adult years, and, (j) single, The Basic Principle: Research suggests that having a prolonged intimate connection to someone may be a protective factor against sexual re-offending. The relative risk to sexually re-offend is lower in men who have been able to form intimate partnerships. When coding the Static-99, all ten items are totaled and scores can range from ?0? to ?12?. Scores above six are considered high risk for sexual reoffense. Evaluations of juveniles based on the Static-99 must be interpreted with caution, as there is a very real theoretical question about whether juvenile sex offending is the same phenomena as adult sex offending in terms of its underlying dynamics and the ability to affect change in the individual (Hanson & Thornton, 1999). The Static-99 has also been validated on different types of adult offenders to include rapist and child molesters (Doren, 1999). The static factors make the Static-99 78 easy to score however; the lack of dynamic factors is a limitation for use with juvenile sex offenders (Hanson & Thornton, 1999; Wood & Cellini, 1999). For the purpose of this study Static-6 variable (ever lived with an intimate partner for 2 or more years) and Static-7 variable (age at time of index offense) were removed from the variables under consideration for the following reasons: Static-6 variable-Research suggests that having a prolonged intimate connection to someone may be a protective factor against sexual re-offending. On the whole, we know that the relative risk to sexually re-offend is lower in men who have been able to form intimate partnerships. If a person has been incarcerated most of their life or is still quite young and has not had the opportunity to establish an intimate relationship of two years duration, they are still scored as never having lived with an intimate partner for two years; Static-variable 7--Research (Hanson, 2001) shows that sexual offending is more likely in an offender?s early adult years than in an offender?s later adult years. In a previous study that assessed the validity of the Static-99 with juvenile sex offenders (Pool, Liedecke, & Marbibi, 2000) two items static variable 6 (ever live with intimate partner) and static variable 7 (age) were removed. The rationale for removing the two variables was that the JSOs would already have two of the four points that were necessary to be considered ?high risk?. Another modification made to the Static-99 for use with juveniles (by the agency 79 providing the samples for the study) was that the scoring for what was considered low, moderate, or high risk changed. Hanson and Thornton?s original scoring for the Static assessment of risk was 0-1 (low), 2-3 (medium), 4-5 (medium-high), and 6 and above (high). For use with JSOs the high risk score was modified to 4 in the hopes of catching more reoffenders. In studies using the Static-99, Barbaree, Seto, Langton, and Peacock (2001) reported a Pearson correlation between total scores of .90. Harris, Rice, Quinsey, Boer, and Lang (2002) reported a Pearson correlation between total scores of .96. The conclusion to be drawn from this data is that raters would rarely disagree by more than one point on a Static-99 score. Data Analysis All data collected on the individual offenders was loaded into a Statistical Package for the Social Science (SPSS) data file for data analysis. In order to address the research question (investigating the effects of selected variable on recidivism in juvenile sex offenders) a multiple regression analysis was performed. To explore the general characteristics of the sample in terms of percentages and frequencies, descriptive analysis was employed. In order to address the research questions (investigating the effects of selected variables on recidivism) multiple regression method will be used. 80 Multiple regression, a time-honored technique going back to Pearson's 1908 use of it, is employed to account for (predict) the variance in an interval dependent, based on linear combinations of interval, dichotomous, or dummy independent variables. Multiple regression can establish that a set of independent variables explains a proportion of the variance in a dependent variable at a significant level (through a significance test of R 2 ), and can establish the relative predictive importance of the independent variables (by comparing beta weights). Multiple regression is used to determine the utility of a set of predictor variables (age at time of intake, IQ score, parental marital status, severity of emotional abuses, severity of physical abuse, severity of sexual abuse, static-1 (prior sex offenses), static-2 (contact sex offenses with stranger), static-3 (contact sex offense with non-relative), static-4 (contact sex offense with male victim), static-5 (non contact sex conviction), static-8 (index non sexual assault), static-9 (prior non sexual assault), and static-10 (more than four sentencing occasions ). The greater potential predictive power of multiple regression is seen in the absolute level of must be as good and most likely better with multiple predictors than any one of the predictors taken by itself (Licht, 1995). 81 IV. RESULTS This section will explain the results generated from both descriptive and multiple regression analysis. The sample consisted of 661 males (179 for the regression analysis) who had been court ordered to attend a secured sex offender specific treatment program after adjudication of a sexual offense. The intake dates were May 1993 through April 2004. Characteristics The age at intake ranged from 10 to 17 years of age with a mean age of 15 (SD = 1.3). The ethnic distribution was 43.6% (n = 288) Caucasian, 32.8 % Hispanic (n = 217), 22.8% African Americans (n = 151), and .8% Other (n = 5). The sample was further desegregated by parental marital status. The largest percentage (38.2%) of JSOs in the sample had parents that were divorced. One hundred and ninety (30.6%) JSOs had parents that were never married, followed by 127 (20.5%) JSOs whose parents were married. Of the remaining JSOs, 36 (5.8%) had deceased fathers, 22 (3.5%) had parents that were separated, 7 (1.1%) 82 had mothers deceased and 1 (.2%) had a parental status that was unknown. As it relates to full scale IQ scores, 48.4% (n = 310) fell into the average to normal range, 26.4% (n = 169) fell into the below average range, 11.4% (n = 73) fell into the borderline range, and 2.5% (n = 16) were considered mentally retarded. On the other end of the spectrum, 7.9% (n = 52) were considered to be of superior intelligence, 3.1% were in the very superior range. Emotional abuse was experienced by 314 of the offenders in the following degrees: none (21.7%) n = 68, mild (17.8%) n = 56, moderate (31.8%) n = 100 and severe (28.7%) n = 90. Physical abuse was experienced in the following ways by the JSOs in the sample: none (30.9%) n = 86, mild (19.1%) n = 53, moderate (34.5%) n = 96 and severe (15.5%) n = 43. As it relates to sexual abuse the sample yielded the following: none (22.7%) n = 67, mild (13.2%) n = 39, moderate (31.5%) n = 93, and severe (32.5%) n = 96. Sexual reoffense or recidivism was defined as any charge for a sexual offense after release from the program. Modifications to the scoring of the Static- 99 were made by the agency providing the data. The modification were made by eliminating Static variable 6 (ever live with intimate partner for 2 or more years) and STATIC variable 7 (age at time of offense). To include these two variables with JSOs would give them 2 of the 4 points needed to be considered high risk. Based on the modification of total points given, 62.9 (n = 415) were assigned a moderate risk level and 37.1 (n = 245) were assigned a high risk level. The sexual 83 reoffense rate for this population of 661 JSOs was 5.3% (n = 35). For the JSOs that reoffended, 42.8% (n = 15) were deemed moderate risk and 57.2% (n = 20) were considered to be high risk. The demographic characteristics for the sample that reoffended is shown in Table 2. Table 2 Reoffender Characteristics (Demographics) Variable Frequency Percent Race Caucasian 18 51.4 Hispanic 9 25.7 African American 8 22.9 Age 13 3 8.6 14 6 17.1 15 11 31.4 16 12 34.3 17 3 8.6 IQ Scale 1 (MR) 2 5.7 2 (borderline) 7 20.0 3 (below avg) 13 37.1 4 (avg) 11 31.4 5 (superior) 2 5.7 84 Table 2 (continued) Variable Frequency Percent Missing samples (none) Parental Marital Status 1 (never married) 13 38.2 2 (married) 4 11.8 3 (divorced) 13 38.2 4 (separated) 2 5.9 6 (father deceased) 2 5.9 Missing samples (1) Emotional Abuse 0 (none) 5 26.3 1 (mild) 3 15.8 2 (moderate) 7 36.8 3 (severe) 4 21.1 Missing samples (16) Physical Abuse 0 (none) 7 38.9 1 (mild) 1 5.6 2 (moderate) 9 50.0 3 (severe) 1 5.6 Missing samples (17) (table continues) 85 Table 2 (continued) Variable Frequency Percent Sexual Abuse 0 (none) 5 23.8 1 (mild) 3 14.3 2 (moderate) 5 23.8 3 (severe) 8 38.1 Missing samples (14) Results Analysis Procedure The data were analyzed by multiple regression, using as predictors age, IQ score, parental marital status, severity of emotional abuse, severity of physical abuse, severity of sexual abuse, Static-99 variables 1-5 and 8-10 (prior sex offenses, contact sex offense with a stranger, contact sex offenses with non- relative, contact sex offenses with male victim, non-contact sex offenses, index non-sexual assault, prior non-sexual assault, and more than one sentencing occasion), as well as the final risk assessment score. Of the 661 samples under consideration SPSS selected 179 (only those with no missing data from the variables of interest) to analyze using multiple regression. Tables 3 and 4 give the model summary and show the analysis of variance (ANOVA) table for the regression analysis. 86 Table 3 Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .405(a) .164 .081 .231 a. Predictors: (Constant), STATICTT, DVIQSCOR, AGE, SEXABUSE, PMS, STATIC10, STATIC8, PHYABUSE, STATIC2, STATIC5, STATIC9, STATIC3, EMOABUSE, STATIC4, RISK, STATIC1 Table 4 ANOVA (b) Model Sum of Squares df Mean Square F Sig. 1 Regression 1.693 16 .106 1.986 .017(a) Residual 8.631 162 .053 Total 10.324 178 a Predictors: (Constant), STATICTT, DVIQSCOR, AGE, SEXABUSE, PMS, STATIC10, STATIC8, PHYABUSE, STATIC2, STATIC5, STATIC9, STATIC3, EMOABUSE, STATIC4, RISK, STATIC1 b Dependent Variable: REOFFEND Using the enter method, a significant model emerged (F16,162 = 1.9686, p = .017), adjusted R square = .081. The regression model was a poor fit (R 2 = 16.4%). The model shows a 16.4% of variance in the factors (age, IQ score, 87 parental marital status, severity of emotional abuse, severity of physical abuse, severity of sexual abuse, Static-1, Static-2, Static-3, Static-4, Static-5, Static-8, Static-9, and Static-10) is explained by the linear combination of the information obtained. A large portion of variance is still not explained. However, overall relationship between the DV and IVs were statistically significant. In examining the beta coefficients for this regression, Table 5 shows the coefficient (r) and the level of significance for each variable under consideration. The items that showed significant correlations to recidivism (p < .05) were: IQ score (p = .044), emotional abuse (p = .041), prior sex offenses (p = .031), stranger victims (p = .033), non-related victims (p = .017), non-contact sex offenses (p = .011), four or more sentencing occasions (p = .012). 88 Table 5 Coefficients(a) Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) .763 .282 2.702 .008 AGE -.015 .015 -.077 -1.023 .308 DVIQSCOR -.039 .019 -.156 -2.033 .044 PMS -.006 .015 -.032 -.415 .678 EMOABUSE -.040 .019 -.195 -2.059 .041 PHYABUSE .010 .020 .046 .503 .616 SEXABUSE .010 .017 .050 .588 .558 STATIC1 .193 .089 .444 2.182 .031 STATIC2 .257 .119 .246 2.150 .033 STATIC3 .216 .089 .446 2.416 .017 STATIC4 .166 .090 .340 1.848 .066 STATIC5 .337 .131 .272 2.564 .011 STATIC8 .124 .140 .085 .889 .375 STATIC9 .144 .106 .193 1.354 .178 STATIC10 .275 .108 .298 2.540 .012 RISK -.035 .064 -.075 -.550 .583 STATICTT -.142 .088 -.720 -1.607 .110 a Dependent Variable: REOFFEND 89 The relationship between the variables in the study were examined using a Pearson Correlation. The Pearson?s correlation reflects the degree of linear relationship between two or more variables (see Table 6). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. IQ score .151* -.106** 2. Emotional abuse .576** .411** .126* 3. Physical abuse .151* .576** .352** .130* 4. Sexual abuse .411** .352** .143* .142* .158** .199** 5. Static 1 .143* .247** .530** .402** 6. Static 2 .142* .214** .302** .305** 7. Static 3 .158** .214** .584** .514** .094* 8. Static 4 .126* .199** .455** .398** 9. Static 5 .247** .307** .191** .095** 10. Static 8 .125** .186** .156** 11. Static 9 .087* .125** .166** 12. Static 10 .166** .288** .239** 13.Static fnl .530** .302** .584** .445** .307** .186** .322** .288** .836** .099* 14. Risk lvl .130* .402** .305** .514** .398** .191** .156** .266** .239** .099* 15. Reoffend -.106** .094* .095* .099* .099* ** R is significant at the 0.01 level (2-tailed) *R is significant at the 0.05 level (2-tailed) Correlations Table 6 90 91 IQ Scores There is a positive correlation between deviation IQ score and physical abuse (.130) which asserts that as the IQ score increases, the severity of physical abuse increases; there is also a negative correlation between IQ score and reoffense (-.113**) which states that as IQ scores increase, sexual reoffenses decrease. Abuse There is a positive correlation between the severity of emotional abuse and the severity of physical abuse (.576), sexual abuse (411**) as the severity of emotional abuse increases, so does the severity of physical abuse and sexual abuse. There is a positive correlations between physical abuse and IQ scores (.151*) and (.130*), which asserts that as the severity of physical abuse increases, the IQ scores increase. Physical abuse is also positively correlated with emotional (.576) and sexual abuse (.352) as well as final risk score (.130) which asserts that as the severity of physical abuse increases, the final risk score increases. Sexual abuse is positively correlated with emotional abuse (.411) and physical abuse (.352). It is also positively correlated with Static-1 (.143*), Static-2 (.142*), Static- 3(.158*) and Static-4 (.199*) which asserts that as the severity of sexual abuse increases, so does the number prior number of sex offenses, contact sex offenses with strangers, contact sex offenses with non-relatives, and so does the number of males victims (respectively). 92 Static-1 (Prior Sexual Offenses) Static variable 1 is positively correlated with parental marital status (.168**), sexual abuse (.143*), Static-5 (.247**), Static-ttl (.530**) and Risk level (.402**). This asserts that as the number of prior sex offenses increase so does the severity of sexual abuse and the number of non-contact sex offenses. There is also and increase in the Static-99 total risk score and risk level assigned to the offender. Static-2 (Contact Sex Offenses with a Stranger) Static variable 2 is positively correlated with severity of sexual abuse (.142*), Static-3, Static-ttl (.320**) and risk level (.305**). This correlations asserts that as the number of contact sex offenses with a stranger increase, there is an increase in the number of contact sex offenses with non-relatives, an increase in the STATIC-99 total risk score and an increase in risk level. Static-3 (Contact Sex Offenses with Non-Relative) Static variable 3 is positively correlated with severity of sexual abuse (.158**), Static-2 (.214**), Reoffense (.094*), Static-ttl (.584**) and with Risk level (.514**). This correlation asserts that as the number of contact sex offenses with a non-relative increases, so does the severity of sexual abuse, the number of contact sex offenses with strangers, as does the number of sexual reoffenses. There is also an increase in the Static-99 total score, and as well as an increase in the risk level. 93 Static-4 (Contact Sex Offenses with Male Victims) Static variable 4 is positively correlated with severity of emotional abuse (.126*), severity of sexual abuse (.199**), Static-9 (.087*), Static=ttl (.445**) and Risk level (.398**). This correlation states that as the number of contact sex offenses with a male victim increases so does the severity of emotional abuse and sexual abuse. There is also an increase in the number of prior non sexual assaults, the Static-99 final risk score and risk level. Static-5 (Non-Contact Sex Conviction) There is a positive correlation between Static-1 (.247**), Reoffense (.095*), Static-ttl (.307**) and Risk level (.191**). The correlation between the variables asserts that as the number of non-contact sex offenses increase so then does the number of prior sex offenses, number of sexual reoffenses, Static-99 final risk scores as well as the risk levels.. Static-8 (Index: Non Sexual Assault) There is a significant correlations between Static-8 and IQ score (.091*), Static-9 (.125**), Static-ttl and Risk level. The positive correlation asserts that as the number of index: non sexual assaults increase, so does IQ scores, prior non- sexual assaults and final risk scores. Static-9 (Prior Non-Sexual Assaults) There is a significant correlation between Static-9 and Static-4 (.087*), Static-8 (.125**), Static-10 (.166*), Static-ttl (.322**) and Risk level (.226**). The 94 correlations asserts that as the number of prior non-sexual assaults increase so then does the number of index: non-sexual assaults, number of sentencing occasions, final risk score, as well as risk level. Static-10 (More Than Four Sentencing Occasions) There is a positive correlation between Static-10 and Static-9 (.166**), Static-ttl (.288**), and Risk level (.239**). This correlation states that as the number of sentencing occasions increase so then does the number of prior: non- sexual assaults, final risk scores and risk levels. Reoffense There is a positive correlation between reoffense, IQ sore (.113**), Static-3 (.094*), Static-5 (.095*), Static-ttl (.099) and Risk levels (.099*). This correlation reveals that as the number of sexual reoffenses increase, so does the number of contact sex offenses with non-relatives, the number of non-contact sex offenses, the final risk score and there is an increase in the risk level. Static-ttl (Final Risk Assessment Score) There is a positive correlation between the final risk assessment score and Static-1 (.530**), Static-2 (.302**), Static-3 (.584**), Static-4 (.445**), Static-5 (.307**), Static-8 (.186**), Static-9 (.322**), Static-10 (.288 **), Re-offense (.099*) and risk levels (.836**). The correlations assert that as the final risk assessment score increases so does the number of prior number sex offenses, contact sex offenses with a stranger, contact sex offenses with non-relatives, contact sex offenses with 95 male victims, non-contact sex offenses, index: non-sexual assaults, prior: non- sexual assaults, and the number of sentencing occasions. It further asserts that as the final risk assessment scores increase then to, will the number of sexual reoffenses and risk levels. Risk Level There is a positive correlation between risk levels and the severity of physical abuse (.130*), Static-1 (.402**), Static-2 (.305**), Static-3 (.514**), Static-4 (.398**), Static-5 (.191*), Static-8 (.156**), Static-9 (.226**), Static-10 (.239 **), Re- offense (.099*) and Static-ttl (.836**). This correlations asserts that as the risk level increases, so does the severity of physical abuse, as does the number of prior sex offenses, contact sex offenses with strangers, contact sex offense with non- relatives, contact sex offenses with male victims, non-contact sex offenses, index: non-sexual assaults, prior: non-sexual assaults, as well as an increase in the number of sentencing occasions. The correlations further assert that there will also be an increase in the number of sexual reoffenses as well as an increase in the final risk score. 96 V. DISCUSSION The purpose of this study was to determine which specific variables were likely to be significant predictors of sexual reoffense in juvenile sex offenders. Additionally the study served to determine the usefulness of the Static-99 in assessing risk among JSOs for recidivism. It is hoped that increasing our knowledge on factors that may predict recidivism in JSOs will allow those responsible for making risk assessments to structure treatments and therapies in a manner that will lesson the likelihood of sexual reoffense. It is further hoped that an instrument for assessing risk in juvenile sex offenders will emerge that will take into consideration the developmental flux of adolescence and will make accurate predictions of risk that will safeguard victims and communities. Results from this study were concentrated on JSOs from a single secure facility and does not address JSOs that receive out-patient treatment and/or residential treatment. In keeping with the disparity in rates of recidivism in juvenile sex offenders between 2% and 75% (Kenny, Keough, & Seidler, 2001), this present study showed an average recidivism rate of 5.3%. The average recidivism rate for studies referenced in this study was 9.2%. 97 In this current study a basic question was posed. Which specified variables would result in a significant association to recidivism among juvenile sex offenders? In this study, the independent variables are age, IQ score, parental marital status, severity of physical abuse, severity of emotional abuse, severity of sexual abuse, prior sex offenses, contact sex offenses with stranger, contact sex offense with non-relative, contact sex offense with male victim, non contact sex conviction, index non-sexual assault, prior non-sexual assault, and more than four sentencing occasions. Studies have shown that many of these factors are predictive of sexual reoffense in juvenile sex offenders. However, in the present study, these variables accounted for only 16.4% of the variability in rates of recidivism. This suggests that there are still other variables that contribute significantly to sexual reoffense in this population. When we look at the variables independently, beginning with age, our finding are consistent with what the literature says about JSOs. Ninety percent of juvenile sex offenders are 14 years of age (Rich, 2003). The present study had a mean age of 15. It is important to note that regardless of the age of the offender, their abusive acts are just as damaging to the victim. A proactive stance to take in regards to the age element of juvenile sex offending is to target programs in schools and communities that are geared toward prevention. This can be done through the use of psycho-educational classes that address sexuality in programs that work with youth (i.e., boy/girl scouts, boys and girls clubs of America, etc.). 98 Many tend to think of academic achievement when the question of IQ testing is raised. As it relates to sexual offending in juveniles, academic achievement has been found to correlate to sexual offending in juveniles (Fehrenbach et al., 1986). In the present study, IQ score was shown to correlate negatively with reoffense which in essence states that as IQ scores increase, sexual reoffenses decrease. This is in keeping with what the literature says as it relates to IQ and juvenile sex offending. Another explanation of this finding is that sexual offenders do not, in fact, have lower IQs than other types of offenders, they only appear to because of an ascertainment bias. That is, because less intelligent sexual offenders more frequently become apprehended and have fewer financial resources to assist in acquittal (Cantor, Blanchard, Robichaud, & Christensen, 2005). In previous studies, parental marital status was also linked to sexual reoffense in juvenile sex offenders. In a study conducted by Ryan and Lane (1991), 27.8% of their sample of JSOs lived with both parents. This present study had similar rates in that 20.5% of the offenders lived in the home with both birth parents. The largest percentage of JSOs (38.2%) had parents that were divorced. Although you cannot logically conclude that all JSO will be from divorced families, an important consideration in assessing risk will be to ensure that family dynamics are addressed as a part of treatment in both individual and family counseling sessions. 99 The family environments of JSOs tend to be chaotic and dysfunctional (Knopp, 1982). This in turn can lead to emotional, physical and sexual abuse of the JSO. One study (Pither et al., 1998) on caregivers of children with sexual behavior problems found sexual abuse rates of 95%, physical abuse rates of 48% and emotional abuse (neglect) rates of 11%. Juvenile sex offenders in this present study experienced abuse consistent to what the literature reports for this population. If we are to lessen the likelihood of sexual reoffense in JSOs, case mangers assigned to mange aftercare programs upon release should ensure the stability of the placement. Families with histories of abuse and neglect should be mandated to attend parenting classes or be assigned a family coach to ensure a healthy transition of the JSO once they leave treatment. The risk assessment used for the JSOs in this sample was the STATIC-99. The STATIC-99 is a risk assessment instrument that was normed on adult male sex offenders (Hanson & Thornton, 1999). Its static risk factors have been empirically linked to sexual reoffense in adult sex offenders. Variables taken under consideration for this study were STATIC Variables 1-5 and 8-10. Variables 6 and 7 were removed because of the false positive scores it is likely to give in juvenile offenders (Pool, Liedecke & Marbibi, 2000). The static variable used for this present study are: prior sex offenses, contact sex offenses with a stranger, contact sex offenses with a non-relative, contact sex offenses with a 100 male victim, non-contact sex offenses, index: non-sexual assault, prior: non sexual assault, and more than four sentencing occasions. According to Rassmussen (1999) the strongest predictor of whether or not an offender, either adult or juvenile will reoffend is past behavior. Hanson and Bussiere (1998) found that any indication that a victim may engage in a variety of crimes, increases their likelihood of reoffending. In the present study of the sample that reoffended (n = 35), 24 (68.1%) had no prior sex convictions and 11 (31.4%) no more than two prior convictions. Worling and Curwen (2001) stated that juveniles who commit two or more offenses are at a higher rate to reoffend than those who have committed a single offense. It is unfortunate that there are no signs to warn us of who will offend before there is a victim, therefore, any conviction of a sexual offense should be treated as an indication that the JSO will reoffend. According to the literature on JSOs, selection of a victim that is a stranger is a high risk factor (Worling & Curwen, 2001). Stranger selection is also predictive of future recidivism (Quinsey, Rice, & Harris, 1995). In the present study of those who reoffended, only 11% (n = 4) chose a stranger victim. This is consistent with what the literature says about juvenile sex offending. Their victims tend to be siblings and people they know (Zonlondek, Abel, Northey, & Jordan, 2001). As the JSO begins his transition from treatment facility to home or community, it will be important to address victim preferences. 101 Both home and community should be taken into consideration when making recommendations for placement after release. Although gender in and of itself is not predictive of sexual reoffense in JSOs, some studies have show that male JSOs that target male victims are more likely to commit subsequent sex offenses (Langstrom & Grann, 2000; Smith & Monastersky, 1986). Also male victim selection by a male JSO can be indicative of sexual deviancy which is any sexual conduct or norm that deviates from the norm (Steele & Ryan, 1997). In the present study 65.7% (n = 23) of the offenders who reoffended had no male victims, while 34.3% (n = 12) had male victims. If there are vulnerable populations within the home and/or community in which the JSO resides, relapse prevention plans should address this element. Those responsible to supervising the JSO in the home and community should understand the seriousness of their charge as it relates to potential victims and the community at large. Non-contact sex offenses were another variable under consideration in the present study. Non-contact sex offenses are those offenses that make no physical contact and include obscene phone calls, theft of clothing for sexual purposes, threats of sexual harm, exhibitionism, public masturbation, creation and possession of pornography (Rich, 2003). These offenses are not necessarily indicative of sexual reoffense, but have been noted in the studies on JSOs. The present sample consisted of 31 reoffenders (88.6%) who had no non-contact sex 102 offenses and 4 (11.4%) who had non-contact sex offenses. Weinrott (1996) reported that JSOs are na?ve? of what constitutes a sexual assault. It is important for those working with JSOs to address in individual and group therapy what constitutes a sexual offense. One of the most predictive factors of sexual reoffense in JSOs is a history of delinquency rather it be sexual or non-sexual (Rich, 2003). In the current study Static variables 8 and 9 address the issue to previous delinquency. Static-8 (index: non-sexual assault) findings were 34 (97.15) had no non sexual assault arrest and Static-9 (prior: non sexual assault) had 30 JSOs (88.2%) with no prior: non-sexual assault and only 4 (11%) who had prior: non sexual assaults. The results of the present study do not support what the literature says in regards to a history of general delinquency and the risk of sexual reoffense. However, it is important to address all elements of delinquency especially with JSOs to determine if the delinquent acts are pathways for sexual reoffense. The final variable under consideration was Static-10 (more than four sentencing occasions). The present study found that of those JSO who reoffended, 31 (88.6%) had less than four sentencing occasions and only 4 (11.4%) had been before a judge and was sentenced at least four or more times. As with Static variables 8 and 9, the findings for this factor are in contradiction to what the literature says about sexual reoffense. Again, all sentencing occasions should 103 be addressed and evaluated in the assessment process to determine their weight in the overall assessment process. Juvenile sex offending is a growing problem in the United States. There has been an increase in both violent crimes committed by juveniles (Office of Juvenile Justice and Delinquency Prevention, 1994a) and in the reports of sexual aggression and sexual abuse (Hampton, 1995). There is no cure of sexual offending but with specialized treatment by qualified professionals, many JSOs can receive offender specific treatment and go on to live productive lives that are free of sexual offending behavior toward others. The final question posed in this research study was determining if select variables on the Static-99 would be useful in providing valid psychometric information to determine its effectiveness in assessing risk of sexual re-offense among sexually offending youth. Five of ten variables on the Static-99 were statistically significant predictors of juvenile sexual recidivism (two variables, age and single were removed because they were not applicable to JSOs). The variable that were significant were Static-1 (prior sex offenses (p = .031), Static-2 (stranger victims (p = .033), Static-3 (non-related victims (p = .017), Static-5 (non- contact sex offenses (p = .011), and Static-10 (four or more sentencing occasions (p = .012). It is important to note that two variables were eliminated as they did not apply to juvenile offenders (age and single). This being the case, five of eight 104 variables were proved to be significant predictors of recidivism in JSO. While any risk of reoffense is too high a price to pay, with revisions, the Static-99 may indeed be an instrument worthy of consideration in the assessment of risk in JSOS. A limitation of the study is the lack of dynamic factors (things that change) it may be useful to determine which dynamic factors are correlated with recidivism in JSO and incorporate them into the Static-99 to make it an empirically validated risk assessment instrument for this population. Implications for Counseling This study aimed to explore the contributing factors for sexual reoffense among juvenile sex offenders. The results indicated that several factors can predict sexual reoffense in JSO, however, the R2 (square) suggest that only 16.4% of the reoffense is explained by these factors?that leaves a large percentage that is not explained. This study has several implications to counseling and counselor education. As counselors we are use to working with those who have been hurt by the effects of sexual abuse. As the tide changes, we are now required to assist those who are doing the hurting (JSOs). As stated in the literature, risk assessment remains one of the most salient components of the assessment process for juvenile sex offenders. This study alone proves that the equation of A+B = C is not true. There are many factors not explored in this study that may 105 serve as predictor of sexual reoffense in juvenile sex offenders. If we are to safeguard victims and communities, we must ensure that we are not sending out those who present a high risk to sexually reoffend nor should we penalize those who have done the hard work of attacking and working through cognitive distortions and deserving of a second chance. This study will further serve to inform counselors and counselor educators on the importance of targeting factors that may best be used to assist in identifying more effective counseling interventions for offender treatment. Identifying effective points of entry for counseling interventions could serve to decrease the likelihood of sexual reoffense and its devastating impact on victims and communities. Not every counselor is open to the idea of providing services for this population, however, imagine what is lost if they do not receive the services necessary to address their abusive behavior ? a community, a nation, and a world ? suffer the consequences. Implications for Future Research Only limited variables were taken into consideration in this research, future research is needed to study the factors which may be more indicative of recidivism in JSOs. Other factors for possible consideration would be the roles that pornography, sexual deviance, and substance abuse play in recidivism of juvenile sex offenders. It may also be beneficial to study the impact of treatment 106 completion as well as modes of treatment (multi-systemic, individual, group, cognitive behavioral, psycho educational and pharmacological). The creation of an empirically validated risk assessment instrument to include the five Static-99 variables that were significant, but that also include dynamic factors that take into consideration the developmental flux of JSOs would assist those making accurate predictions about the risk of sexual reoffense. This study indicated that as the severity of emotional abuse decreases, reoffenses increase, this is a phenomenon that deserves continued research and investigation. 107 REFERENCES Abel, G., Rouleau, J., & Cunningham-Rathner, J. (1986). Sexually aggressive behavior. In W. Curran, A. McGarry, & S. Shah (Eds.), Forensic psychiatry and psychology: Perspectives and standards for interdisciplinary practice (pp. 289-313). Philadelphia: Davis. Ageton, S. (1983). Sexual assault among adolescents. Lexington, MA: Lexington Books. Akers, R. (1998). Social learning and social structure. Boston: Northeastern University Press. Aljazireh, L. (1993). Historical, environmental, and behavioral correlates of sexual offending by male adolescents: A critical review. Behavioral Sciences and the Law, 11, 423-440. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4 th ed., test revision). Washington, DC: Author. Arias, I., & O?Leary, K. (1988). Cognitive-behavioral treatment of physical aggression in marriage. In N. Epstein, S. E. Schlesinger & W. Dryden (Eds.), Cognitive-behavioral therapy with families (pp. 118-150). New York: Brunner/Mazel. 108 Association for the Treatment of Sexual Abusers. (2001). Practice standards and guidelines for members of the Association for the Treatment of Sexual Abusers. Beaverton, OR: Author. Awad. G., Saunders, E., & Levene, J. (1984). A clinical study of male adolescent sexual offenders. International Journal of Offender Therapy and Comparative Criminology, 28, 105-115. Bandura, A. (1986). Social Foundation of thought and action: A social cognitive theory. New Jersey: Prentice-Hall. Barbaree, H. E., Seto, M. C., Langton, C. M., & Peacock, E. J. (2001). Evaluating the predictive accuracy of six risk assessment instruments for adult sex offenders. Criminal Justice and Behavior, 28, 490-521. Becker, J. (1994). Offenders: characteristics and treatment. Future of Children: Child Sexual Abuse, 4, 176-197. Becker, J. (1990). Treating adolescent sexual offenders. Professional Psychology: Research and Practice, 21, 362-365. Becker, J. (1988). Adolescent sex offenders. Behavior Therapy, 11, 185-187. Becker, J., & Abel, G. (1985) Methodological and ethical issues in evaluating the treatment of adolescent sex offenders. In E. M. Otey & G.D. Ryan (Eds.), Adolescent sex offenders: Issues in research and treatment (pp. 109-129). Rockville, MD. Department of Health and Human Services 109 Becker, J., & Hunter, J. (1997). Understanding and treating child and adolescent sex offenders. In T. H. Ollendick & R. J. Prinz (Eds.), Advances in clinical child psychology (Vol. 19, pp. 177-197). New York: Plenum Press. Becker, J., & Hunter, J. (1993). Aggressive sex offenders. Child Adolescent Psychiatry, 2, 477-487. Becker, J., & Murphy, W. (1998). What we know and do not know about assessing and treating of sex offenders. Psychology, Public Policy, and Law, 4(1/2), 116-137. Becker, J. V., & Stein, R. M. (1991). Is sexual erotica associated with sexual deviance in adolescent males? International Journal of Law and Psychiatry, 14, 85-95. Beech, A., Fisher, D., & Thornton, D. (2003). Risk assessment of sex offenders. Professional Psychology: Research & Practice, 34(4), 339-352. Bijleveld, C., & Hendriks, J. (2004). Juvenile sexual offenders: Contrasting child abusers with peer abusers. Criminal Behavior and Mental Health, 14, 238-250. Bischof, G., Stith, S., & Whitney, M. (1995). Family environments of adolescent sex offenders and other juvenile delinquents. Adolescence, 30(117), 157-171. Blaske, D., Borduin, C., Henggeler, S., & Mann, B. (1989). Individual, family and peer characteristics of adolescent sexual offenders and assaultive offenders. Developmental Psychology, 25, 846-855. 110 Boer, D., Hart, S., Kropp, P., & Webster, C. (1997). Sexual violence risk-20: Professional guidelines for assessing risk of sexual violence. Burnaby, British Columbia, Canada: The Mental Health, Law, and Policy Institute of Simon Fraser University. Bonta, J. (1989). Native inmates: Institutional response, risk, and needs. Canadian Journal of Criminology, 31, 49-62. Bonta, J., Law, M., & Hanson, R. (1998). The prediction of criminal and violent recidivism among mentally disordered offenders: A meta-analysis. Psychological Bulletin, 123, 123-142. Borduin, C., Henggeler, S., Blaske, D., & Stein, R. (1990). Multisystemic treatment of adolescent sexual offenders. International Journal of Offender Therapy and Comparative Criminology, 39, 317-326. Boyd, N., Hagen, M. & Choo, M. (2000). Characteristics of juvenile sex offenders: A review of the research. Aggression and Violent Behavior, 5(2), 137-146. Brannon, J. M., & Troyer, R. (1995). Adolescent sex offenders: investigating adult commitment rates four years later. International Journal of Offender Therapy and Comparative Criminology, 39(4), 317-326. Breer, W. (1987). The adolescent molester. Springfield, IL: Charles C. Thomas. Bremer, J. (1992). Serious juvenile sex offenders: Treatment and long-term follow up. Psychiatric Annals, 22, 326-332. 111 Bureau of Justice Statistics (http://www.usdoj/gov) [Data file]. Department of Justice. Bumby, K. (2000). Empathy inhibition, empathy deficits, and attachment difficulties in sexual offenders. In Laws, S., Hudson, & Ward. T. (Eds.), Remaking relapse prevention with sexual offenders: A source book (pp. 143-166). Thousand Oaks: Sage. Burgess, A., Hartman, C., McCormack, A., & Grant, C. (1988). Child victim to juvenile victimizer: Treatment implications. American Journal of Family Psychiatry, 9, 403-416. Burton, D. (2003). Male adolescents: sexual victimization and subsequent sexual abuse. Child and Adolescent Social Work, 20(4), 277-296. Burton, D., Miller, D., & Shill, C. (2002). A social learning theory comparison of the sexual victimization of adolescent sexual offenders and nonsexual offending male delinquents. Child Abuse & Neglect, 26, 893-907. Burton, D., Nesmith, A., & Badten, L. (1997). Clinician?s views on sexually abusive children: A theoretical exploration. Child Abuse & Neglect, 21, 157- 170. Calder, M. (2000). The complete guide to sexual abuse assessments. Dorset, England: Russell House. Campbell, T. (2000). Sexual predator evaluations and phrenology: considering issues of evidentiary reliability. Behavioral Science and the Law, 18, 111-130. 112 Cantor, J., Blanchard, R., Robichaud, L. & Christensen, B. (2005). Quantitative Reanalysis of Aggregate Data on IQ in Sexual Offenders. Psychological Bulletin, 131(4), 555?568 Cashwell, C., & Caruso, M. (1997). Adolescent sex offenders: Identification and intervention strategies. Journal of Mental Health Counseling, 19, 336-349. Cashwell, Craig S., Bloss, Kim K., McFarland, Jill E., (1995). From Victim to Client: Preventing the Cycle of Sexual Reactivity, The School Counselor, Vol. 42, pp. 233-238. Cautela , J. (1966). Treatment of compulsive behavior by covert sensitization. Psychology Review, 16, 33-41. Cellini, H. R. (1995). Assessment and treatment of the adolescent sexual offender. In B. K. Schwartz and H. R. Cellini (Eds.), The sex offender: Vol. 1. Corrections, treatment and legal practice (pp. 6.1-6-12). Kingston, NJ: Civic Research Institute. Center for Sex Offender Management [CSOM]. (1999). Compendium of OJP- sponsored projects relating to sex offenders. Unpublished Manuscript Christodoulides, T, Richardson, G., Graham, F., Kennedy, P. & Kelly, T. (2005). Risk assessment with adolescent sex offenders. Journal of Sexual Aggression, 11(1), 37-48. Davis, F., & Leitnberg, H. (1987). Adolescent sex offenders. Psychological Bulletin, 101(3), 417-427. 113 Davis, T., Peck, G., & Storment, J. (1993). Acquaintance rape and the high school student. Adolescent Health, 14, 220-224. Department of Psychiatry. (2004). Epidemiology and treatment of juvenile sexual offending (INSERM 9906). Rouen, France: Author. Dewhurst, A., & Nielsen, K. (1999). A resiliency-based approach to working with sexual offenders. Sexual Addiction & Compulsivity: The Journal of Treatment and Prevention, 6, 271-279. Doren, D. M. (2002). Evaluating sex offenders: A manual for civil commitments and beyond. Thousand Oaks, California: Sage. Doren, D. M. (1999). Using and testifying about sex offender risk assessment instrumentation. Unpublished manuscript. Mendota Mental Health Institute. Dougher, M. (1988b). Behavioral techniques to alter sexual arousal. In B. Schwartz & H. Cellini (Eds.), A practitioner?s guide to treating the incarcerated male sex offender (pp. 109-113). Washington, DC: U.S. Department of Justice, National Institute of Corrections, U.S. Government Printing Office. Dynamic Predictors of Sexual Reoffense. [Data file]. Ottawa, ON: Department of the Solicitor General of Canada. Ellerby, L. (1997). Impact on clinicians: Stressors and providers of sex-offender treatment. In S. Emounds (Ed.), Impact: Working with sexual abusers (pp. 11- 29). Brandon, VT: Safer Society Press. 114 Elliot, D. (1994). The developmental course of sexual and non-sexual violence: Results from a national longitudinal study. Paper presented at the meeting of the Association for the Treatment of Sexual Abusers? 13 th Annual Research and Treatment Conference, San Francisco, CA. Elliot, D., Ageton, S.,Huizinga, D., Knowles, B. & Cantor, R. (1983). [National youth survey. Behavioral Research Institute]. Unpublished raw data. Emounds, S. (1997). The personal impact of working with sex offenders. In S. Edmunds (Ed.), Impact: Working with sexual abusers (pp. 51-60). Brandon, VT: Safer Society Press. Epps, K. (1997). Working with sexually abusive adolescents. In M. S. Hoghughi, S. R. Bhate, & F. Grahm (Eds.), Managing risk (pp. 35-51). London: Sage. Fagan, J. & Wexler, S. (1988). Explanations of sexual assault among violent delinquents. Journal of Adolescent Research, 3, 363-385. Fanniff, A., & Becker, J. (2006). Specialized assessment and treatment of adolescent sex offenders. Aggression and Violent Behavior, 11, 265-282. Farrenkoph, T. (1992). What happens to therapists who work with sex offenders? Journal of Offender Rehabilitation, 18, 217-223. Fehrenbach, P., Smith, W., Monastersky, C., & Deisher, R. (1986). Adolescent sexual offenders: Offender and offense characteristics. American Journal of Orthopsychiatry, 56, 225-233. 115 Ferrara, M., & McDonald, S. (1996). Treatment of the juvenile sex offender: Neurological and psychiatric impairments. Northvale, NJ: Jason Aronson. Fisher, D., & Thornton, D. (1993). Assessing risk of re-offending in sexual offenders. Journal of Mental Health, 2(2), 105-118. Ford, M., & Linney, J. (1995). Comparative analysis of juvenile sex offenders, violent nonsexual offenders and status offenders. Journal of Interpersonal Violence, 10(1), 56-70. Freeman-Longo, R., Bird, S., Stevenson, W. and Fisk, J. (1995). 1994 nationwide survey of treatment programs and models: Serving abuse-reactive children and adolescent and adult sex offenders. Brandon, VT: Safer Society Press. Froth, A., Kosson, D., & Hare, R. (2003). The Hare psychopathy checklist: Youth version manual. Toronto: Multi-Health Systems. Gerardin, P., & Thibault, F. (2004). Epidemiology of juvenile sex offending. Pediatric Drugs, 6(2), 79-89. Gray, S. (1995). A comparison of verbal satiation and minimal arousal conditioning to reduce deviant arousal in the laboratory. Child Sexual Abuse, 7, 143-154. Gray, A., & Pithers, W. (1993). Relapse prevention with sexually aggressive adolescents and children: Expanding treatment and supervision. In H. Barbaree, W. Marshall, & S. Hudson (Eds.), The juvenile sex offender (pp. 289-319). New York: Guilford. 116 Green, R. (1988). Psycho-educational modules. In B. Schwartz, & H. Cellini (Eds.), A practitioner?s guide to treating the incarcerated male sex offender (pp. 95-100). Washington, DC: US Department of Justice, National Institute of Corrections, US Government Printing Office. Gretton, H., McBride, M., Hare, R., O?Shaughnessy, R., & Kumba, G. (2001). Psychopathy and recidivism in adolescent sex offenders. Criminal Justice and Behavior, 28(4), 427-449. Groggin, C. (1994). Clinical versus actuarial prediction: A meta-analysis. Unpublished manuscript. Groth, A. (1979). Men who rape: The psychology of the offender. New York: Plenum Press. Grove, W., Zald, D., Lebow, B., Snitz, B., & Nelson, C. (2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12, 19-30. Grubin, D. (1998). Sex offending against children: Understanding the risk. Home Office Research Development and Statistics Directorate Research Findings Police Research Series (Paper 1999). London: Home Office, Research, Development and Statistics Directorate, Policing and Reducing Crime Unit. Haapasalo, J., & Hamalainen, T. (1996). Childhood family problems and current psychiatric problems among young violent and property offenders. 117 Journal of the American Academy of Child and Adolescent Psychiatry, 34, 1394- 1401. Hagan, M. & Gust-Brey, K. (2000). A ten-year longitudinal study of adolescent perpetrators of sexual assault against children. Journal of Offender Rehabilitation, 31(1), 117-126. Hagan, M., & Gust-Brey, K. (1999). A ten year longitudinal study of adolescent rapist upon return to the community. International Journal of Offender Therapy and Comparative Criminology, 43(4), 448-458. Hagan, M., Gust-Brey, K., Cho, M., & Dow, E. (2001). Eight-year comparative analyses of adolescent rapist, adolescent child molesters, other adolescent delinquents, and he general population. International Journal of Offender Therapy and Comparative Criminology, 45(3), 314-324. Hagan, M. P., King, R. P., & Patros, R. L. (1994). Recidivism among adolescent perpetrators of sexual assault against children. Young Victims, Young Offenders, 21(1-2), 127-137. Hall, E., & Flannery, P. (1984). Prevalence and correlates of sexual assault experiences in adolescents. Victimology, 9(3/4), 398-406. Hampton, H. (1995). Care of the woman who has been raped. New England Medical Journal, 332, 234-237. Hanson, R. (2000). Risk assessment. Association for the Treatment of Sexual Abusers. 118 Hanson R. (1998). What we know about sex offender risk assessment. Psychology, Public Policy and Law, 4(1-2), 50-72. Hanson, R. (1997). The development of a brief actuarial scale for sexual offense recidivism. Department of the Solicitor General of Canada Hanson, R. K. (2002). Recidivism and age: Follow-up data on 4,673 sexual offenders. Journal of Interpersonal Violence, 17, 1046-1062. Hanson, R. K. (2001). Note on the reliability of STATIC-99 as used by the California Department of Mental Health evaluators. Unpublished report. Sacramento, CA: California Department of Mental Health. Hanson, R. K., & Bussiere, M. T. (1998). Predictors of sexual offender recidivism: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348-362. Hanson, R. K., & Bussi?re, M. T. (1998). Predicting relapse: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348-362. Hanson, R., & Bussiere, M. (1996). Predictor of sexual abuse recidivism: A meta- analysis (User Report No.96-04). Ottawa, Ontario, Canada: Department of the Solicitor General of Canada. Hanson, R. K., & Thornton, D. (2002). Notes on the development of the STATIC- 2002. Unpublished manuscript. 119 Hanson, R. K., & Thornton, D. (2000). Improving risk assessments for sex offenders: A comparison of three actuarial scales. Law and Human Behaviour, 24(1), 119-136. Hanson, R. K., & Thornton, D. (1999). Static-99: Improving actuarial risk assessments for sex offenders. User Report 99-02. Ottawa: Department of the Solicitor General of Canada. Department of the Solicitor General of Canada website, www.psepc.sppcc.gc.ca Hanson, R. K., Morton, K. E., & Harris, A. J. R. (in press). Sexual offender recidivism risk: What we know and what we need to know. In R. Prentky, E. Janus, M. Seto, and A. W. Burgess (Eds.), Understanding and managing sexually coercive behaviour. Annals of the New York Academy of Sciences. Hare, R. D. (1999). Without conscience: The disturbing world of psychopaths among us. New York: Simon & Shuster. Harris, A., Phenix, A., Hanson, K., & Thornton, D. (2003). Static-99 coding rules revised. Harris, G. T., Rice, M. E., Quinsey, V. L., Boer, D., & Lang, C. (2002). A multi-site comparison of actuarial risk instruments for sex offenders. Manuscript submitted for publication. Hart, S. (2001). Complexity, uncertainty, and the reconceptulization of violence risk assessment. Paper presented at the meeting of the European Association of Psychology and Law, Lisbon, Portugal. 120 Hayashino, D., Wurtele, S., & Klebe, K. (1995). Child molesters: An examination of cognitive factors. Journal of interpersonal Violence, 10, 106-116. Henggeler, S.W., Mihalic, S.F., Rone, L.,Thomas, C., & Timmons-Mitchell, J. (1998). Multisystemic Therapy: Blueprints for Violence Prevention, Book Six. Blueprints for Violence Prevention Series (D. S. Elliott, Series Editor). Boulder, CO: Center for the Study and Prevention of Violence, Institute of Behavioral Science, University of Colorado. Hunter, J. (1990). Understanding juvenile sex offending. Washington DC: Center for Sex Offender Management. Hunter, J. A. (2000). Understanding juvenile sex offenders: research findings & guidelines for effective management & treatment. Juvenile Justice Fact Sheet. Charlottesville, VA: Institute of Law, Psychiatry, & Public Policy, University of Virginia. Hunter, J. A., & Becker, J. V. (1998). Motivators of adolescent sex offenders and treatment perspectives. Washington, DC: American Psychiatric Press, Inc. Hunter, J. & Figueredo, A. (1999). Factors associated with treatment compliance in a population of juvenile sexual offenders. Child Sex Abuse, 11, 49-67. Jenny, C. (1988). Adolescent risk-taking behavior and the occurrence of sexual assault. AJDJ, 142, 770-772. Johnston, L., & Ward, T. (1996), Social cognition and sexual offending: A theoretical Framework, Child Sex Abuse, 8, 55-80. 121 Kahn, T., & Chambers, H. (1991). Assessing reoffense risk with juvenile sexual offenders. Child Welfare, 70(3), 333-345. Kahn, T., & Lafond, M. (1988). Treatment of the adolescent sexual offender. Child and Adolescent Social Work Journal, 5, 135-148. Kavoussi, R., Kaplan, M., & Becker, J. (1988). Psychiatric diagnoses in adolescent sex offenders. Journal of the American Academy of Child and Adolescent Psychiatry, 27(2), 241-243. Kelly, S., Lewis, K., & Sigal, J. (2004). The impact of risk factors on the treatment of adolescent sex offenders. Journal of Addictions & Offenders Counseling, 24, 67-81. Kendall-Tackett, K., Williams, L., & Finkelhor, D. (1993). Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin, 113, 164-180. Kenny, D., Keogh, T., & Seilder, K. (2001). Predictors of recidivism in Australian juvenile sex offenders: implications for treatment. Journal of Research and Treatment, 13, 131-148. Knight, R., & Prentky, R. (1993). Exploring the characteristics for classifying juvenile sexual offenders. In H. Barbaree, W. Marshall & S. Hudson (Eds.), The juvenile sex offender (pp. 45-83). New York: Guilford Press. Knopp, F. (1982). Remedial intervention in adolescent sex offenses: Nine program descriptions. Orwell, VT: Safer Society Press. 122 Knopp, F. (1985). Recent developments in the treatment of adolescent offenders. In E. Otey & G. Ryan (Eds.), Adolescent sex offenders, issues in research and treatment. National Institute of Mental Health, Rockville, Maryland, DHHS Publication no. (ADM)85-13960, pp. 1-108. Lab, S., Shields, G., & Schondel, C. (1993). Research note: An evaluation of juvenile sexual offender treatment. Crime & Delinquency, 39(4), 543-553. Lake, J. (1992). Myth information and bizarre beliefs of male juvenile sex offenders. Journal of Addictions & Offender Counseling, 13(1), 1-8. Lane, S. (1997). The sexual abuse cycle. In G. Ryan & S. Lane (Eds.), Juvenile sexual offending (new and revised ed., pp. 77-121). San Francisco: Jossey-Bass. Langstrom, N. (2002). Long-term follow-up of criminal recidivism in young sex offenders: Temporal patterns and risk factors. Psychology, Crime, & Law, 8(1), 41-58. Langstrom, N., & Grann. M. (2000). Risk for criminal recidivism among young sex offender. Journal of Interpersonal Violence, 15, 855-871. Levenson, J., & Morin, J. (2006). Risk assessment in child sexual abuse cases. Child Welfare League of America, 24, 59-82. Lewis, D., Shankok, S. & Pincus, J. (1979). Juvenile male sexual assaulters. American Journal of Psychiatry, 136(9), 1194-11-96. 123 Lisak, D. & Roth, S. (1990). Motives and Psychodynamics of Self-reported, Unincarcerated Rapists. American Journal of Orthopsychiatrics, 60(2), 268- 280. Maletzky, B. (1974). ?Assisted? covert sensitization in the treatment of exhibitionism. Clinical Psychology, 42, 34-40. Maletzky, B. (1996). Denial of treatment or treatment of denial. Journal Child Sex Abuse 8, 1-5. Maletzky, B. (1996). Denial of treatment or treatment of denial. Child Sex Abuse, 8, 1-5. Maquire, K., & Pastore, A. (2002, April 21). Sourcebook of criminal justice statistics. Archived at http://www.albany.edu./sourcebook/ Marshall, W. (1989). Invited essay: Intimacy, loneliness and sexual offenders. Behaviour Research and Therapy, 27, 491-503 Maxwell, C., Robinson, A., & Post, L. (2003). The nature and predictors of sexual victimization and offending among adolescents. Journal of Youth and Adolescents, 32(6), 463-477. McConaghy, N., Armstrong, M., Blaszczynski, M., & Kidson, W. (1989). Resistance to treatment of adolescent sex offenders. Archives of Sexual Behavior, 18, 97-107. 124 McCormack, J., Hudson, M., & Ward, T. (2002). Sexual offenders? perceptions of their early interpersonal relationships: An attachment perspective. Journal of Sex Research, 39(2), 85-93. Mio, J., Nanjundappa, G., Verleur, D., & De Rios, M. (1986). Drug abuse and the adolescent sex offender: A preliminary analysis. Journal of Psychoactive Drugs, 18(1), 65-72. Moffit, T. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701 Monahan, J. (1981). The clinical prediction of violent behavior. Northvale, NJ: Jason Aronson Inc. Moore, T, Franey, K. & Geffner, R. (1994). Introduction: Assessment and treatment of youth who sexually offend: An over view. Journal of Child Sex Abuse, 13(3/4), 1-32. Morrison, T. (1998). Partnership, collaboration and change under the Children Act. In M. Adocock, R. White, & A. Hollows (Eds), Significant harm (2 nd ed.). Croydon: Significant Publications. Murphy, W. (1990). Assessment and modification of cognitive distortions in sex offenders. In W. L. Marshall, D. R. Laws & H. E. Barbaree (Eds.), Handbook of sexual assault (pp. 331-342). New York: Plenum. Muster, N. (1992). Treating the adolescent victim-turned-offender. Adolescence, 27, 441-450. 125 Nanjundappa, G., deRios, M., Mio, J. & Verleur, D. (1987). Profiles of juvenile male incest perpetrators: Preliminary treatment implications. Journal of Offender Counseling, 8, 25-31. National Adolescent Perpetrator Network. (1993). The revised report from the National Task Force on Juvenile Sexual Offending. Juvenile and Family Court Journal 44(4), 1-120. National Task Force on Juvenile Sexual Offending. (1993). The revised report on juvenile sexual offending 1993 on the national adolescent perpetration network. Juvenile and Family Court, 44, 1-120. O?Reilly, G., Morrison, T., Sheerin, D., & Carr, A. (2001). Child Abuse Review, 10, 150-169. Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice. (2001). Juveniles who have sexually offended: A review of the professional literature. Washington DC: Author. Office of Juvenile Justice and Delinquency Prevention (1994a), Juveniles and Violence. Washington, DC: US Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention (Fact Sheet 19). Parker, H., & Parker, S. (1986). Father-daughter sexual abuse: An emerging perspective. American Journal of Orthopsychiatry, 56, 531-549. 126 Parks, G. (2004). Juvenile sexual recidivism: An examination of typological differences in risk assessment. Unpublished doctoral dissertation, Walden University, Oklahoma. Pithers, W., Gray, A., Busconi, A., & Houchens, P. (1998). Caregivers of children with sexual behavior problems: Psychological and familial functioning. Child Abuse & Neglect, 22(2), 129-141. Poison, M., & McCullom, E. (1995). Therapist caring in the treatment of sexual abuse offenders. Journal of Child Sexual Abuse, 4, 21-43. Prentky, R., & Righthand, S. (2003). Juvenile sexual offender assessment protocol II. Manual NCJ202316, 1-32. Prentky, R., Harris, B., Frizzell, K., & Righthand, S. (2000). An actuarial procedure for assessing risk with juvenile sex offenders. Sexual Abuse: A Journal of Research and Treatment, 12, 71-93. Prescott, D. (2004). Emerging strategies for risk assessment of sexually abusive youth: Theory, controversy, and practice. Journal of Child Sexual Abuse, 13(3/4), 83-105. Prochaska, J., & DiClemente, C. (1983). Stages and processes of self-change of smoking: toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390-395. 127 Prochaska, J., & DiClemente, C. (1986). Towards a comprehensive model of change. In W. Miller & N. Heather (Eds.), Treating addictive behaviours. Plenum: New York. Proulx, J., Lusignan, R., & McKibben, A. (1996). Relationship between affective components and sexual behaviors in sexual aggressors. Child Sex Abuse, 8, 279-289. Quinn, K. (1992). Juvenile sex offenders. In M. G. Kalogerakis (Ed.), Handbook of psychiatric practice in the juvenile court (pp.105-109). Washington, DC: American Psychiatric Association. Quinsey, V., Rice, M., & Harris, G., (1995). Actuarial prediction of sexual recidivism. Journal of Interpersonal Violence, 10, 85-105. Rasmussen, L. A. (1999). Factors related to recidivism among juvenile sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 11(1), 69-85. Rencken, R. (1989). Intervention strategies for sexual abuse. Alexandria, VA: American Association for Counseling and Development. Rice, J. C. (1994). Logistic regression: An introduction. In B. Thompson (Ed.), Advances in social science methodology (Vol. 3, pp. 191-245). Greenwich, CT: JAI Press. Rich, P. (2003). Understanding, assessing and rehabilitating juvenile sexual offenders. Hoboken, NJ: John Wiley & Sons Inc. 128 Righthand, S., Welch, C., Carpenter, E., Young, G., & Scoular, R. (2001). Sex offending by Maine youth: Their offenses and characteristics (Part 1 & Part 2). Augusta, ME: Department of Corrections and Department of Human Services. Robers, C., Doren, D. & Thornton, D. (2002). Dimensions associated with assessments of sex offender recidivism risk. Criminal Justice and Behavior, 29, 569-589. Ross, J., & Loss, P. (1991). Assessment of the juvenile sex offender. In G. D. Ryan, & S. L. Lane (Eds.), Juvenile sexual offending: Causes, consequences, and correction (pp. 199-251). Ryan, G., & Lane, S. (1997). Juvenile sexual offending: Causes, consequences, and correction (Rev. ed.). San Francisco: Jossey-Bass. Ryan, G. (1999). Treatment of sexually abusive youth. Journal of Interpersonal Violence, 14(4), 422-436. Ryan, G., & Lane, S. (1991). Juvenile sex offending: Causes, consequences and corrections. Lexington, MA: Lexington Books. Ryan, G., Lane, S., Davis, J., & Issac, C. (1987). Juvenile sex offenders: Development and correction. Child Abuse and Neglect, 11, 385-395 Ryan, G., Miyoshi, T., Metzner, J., Krugman, R. & Fryer, G. (1996). Trends in a national sample of sexually abusive youth. Journal of the American Academy of Child & Adolescent Psychiatry, 35(1), 17-25. 129 Scheela, R. (2001). Sex offender treatment: Therapist?s experiences and perceptions. Issues in Mental Health Nursing, 22, 749-767. Schlank, A., & Shaw, T. (1996). Treating sexual offenders who deny their guilt: A pilot study. Child Sex Abuse, 8, 17-23. Schwartz, B. (1992). Effective treatment techniques for sex offenders. Psychiatric Rehabilitation, 22, 315-319. Schwartz, B. (1988). Interpersonal techniques in treating sex offenders. In A practitioner?s guide to treating the incarcerated male sex offender. Schwartz, B., Sholevar, G., & Schwoeri, L. (1999). Sexual aggression within the family. In J. A. Shaw (Ed.), Sexual Aggression (pp. 75-106). Washington, DC: American Psychiatric Press. Seabloom, W., Seabloom, M., Seabloom, E., Barron, R., & Hendrickson, S. (2003). A 14-to-24-year longitudinal study of a comprehensive sexual health model treatment program for adolescent sex-offenders: Predictors of successful completion and subsequent criminal recidivism. International Journal of Offender Therapy and Comparative Criminology, 47(4), 468-481. Serin, R., & Brown, S. (2000). The clinical use of the Hare Psychopathy check-list- revised in contemporary risk assessment. In C. G. Cacono (Ed.), The clinical and forensic assessment of psychopathy (pp. 251-268). Mahwah, NJ: Lawrence Erlbaum Associates. 130 Shaw T., Funderburk J. R., & Schlank A. M. (1999). Cognitive-behavioral treatment strategies and the integrated model for adolescent perpetrators of sexual abuse. In J. A. Shaw (Ed.), Sexual Aggression (pp. 275-304). Washington, DC: American Psychiatric Press. Shaw, J., & Lewis, J. (2000). Child on child sexual abuse: Psychological perspectives. Child Abuse & Neglect, 24(12), 1591-1600. Shaw, J. (2000). Summary of the practice parameters for the assessment and treatment of children and adolescents who are sexually abusive to others. Journal of the American Academy of Child and Adolescent Psychiatry, 39(1), 127-130. Sipe, R., Jensen, E. L., & Everett, R. S. (1998). Adolescent sexual offenders grown up: Recidivism in young adulthood. Criminal Justice and Behavior, 25(1), 109-125. Smets, A., & Cebula, C. (1987). A group treatment program for adolescent sex offenders: Five steps toward resolution. Child Abuse & Neglect, 11 (2), 247- 254. Smith, W., & Monastersky, C. (1986). Assessing juvenile sexual offender?s risk for reoffending. Criminology, Justice & Behavior, 13, 115-140. Snyder, H. (1999). Juvenile arrest 1998. OJJDP: Juvenile justice bulletin. U.S. Department of Justice. 131 Spaccarelli, S., Bowden, B., Coatsworth, J., & Kim, S. (1997). Psychosocial correlates of male sexual aggression in a chronic delinquent sample. Criminal Justice and Behavior, 24(1), 71-95. Stalans, L. (2004). Adult sex offenders on community supervision: A review of recent assessment strategies and treatment. Criminal Justice & Behavior, 31, 564-608. Steele, B. & Ryan, G. (1997). Development gone wrong. In G. Ryan & S. Lane (Eds.), Juvenile sexual offending: Causes, consequences, and correction (Rev. ed., pp. 59-76). San Francisco: Jossey-Bass. Steen, C., & Monnette, B. (1989). Treating adolescent sex offenders in the community. Springfield, IL: Thomas Books. Thomas, J. (1997). The family in treatment. In G. Ryan & S. Lane (Eds.), Juvenile sexual offending: Causes, consequences, and corrections (Rev. ed., pp 360-403). San Francisco: Jossey-Bass. Thorndyke, E. L. (1911). Animal intelligence. New York: MacMillan. Tingle, D., Barnard, G., Robbins, L., Newman, G., & Hutchinson, D. (1986). Childhood and adolescent characteristics of pedophiles and rapist. International Journal of Law and Psychiatry, 9, 103-116 Center for Sex Offender Management. (1999). Understanding juvenile sexual offending behavior: Emerging research, treatment approaches and management practices [Data file]. 132 Unincarcerated Rapist. American Journal of Orthopsychiatry, 60, 268-280. Veneziano, C., Veneziano, L., & LeGrand, S. (2000). The relationship between adolescent sex offender behaviors and victim characteristics with prior victimization. Violence and Victims, 15(4), 363-374. Ward, T., & Hudson, S. (1998). The construction and development of theory in the sexual offending area: A metatheoretical framework. Sexual Abuse: A Journal of Research and Treatment, 10, 47-63 Ward, T., Fon, C., Hudson, S., & McCormack, J. (1998) A descriptive model of dysfunctional cognitions in child molesters. Journal of Interpersonal Violence, 13, 129-155. Ward, T., Hudson, S., Johnston, L., & Marshall, W. (1997). Cognitive distortions in sex offenders: An integrative review. Clinical Psychology Review, 17, 479- 507. Ward, T., Hudson, S., Marshall, W., & Siegert, R. (1995). Attachment style and intimacy deficits in sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 7, 317-335. Weinrott, M. (1996). Juvenile sexual aggression: A critical review. Boulder CO: Center for the Study and Prevention of Violence. What Research Shows About Adolescent Sex Offenders. (2003, July). National Center on Sexual Behavior of Youth. 133 Widom, C. (1995). Victims of childhood sexual abuse: Later criminal consequences. NIJ Research in brief, Available: http://www.ncjrs.org/txtfiles/abuse.txt Widom, C., & Maxfield, M. (1996). A prospective examination of risk for violence among abused and neglected children. Annals of the New York Academy of Sciences, 794, 224-237. Will, D. (1999).Assessment issues. In M. Erooga & H. Massons (Eds.), Children and young people who sexually abuse others: Challenges and responses (pp. 86- 103). London: Routledge. Williams, L., & Finkelhor, D. (1990). The characteristics of incestuous fathers. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault (pp. 231-255). New York: Plenum. Witt, P., Bosley, J., & Hiscox, S. (2002). Evaluation of juvenile sex offenders. Journal of Psychiatry & Law, 30, 569-592. Wodarski, J., & Whitaker, D. (1989). Issues in treating sex offenders in the community. Journal of Social Work and Human Sexuality, 7, 145-155. Wolk, N. (2005). Predictors associated with recidivism among juvenile sexual offending. Unpublished Doctor of Education Dissertation, University of Houston, May 2005. 134 Wood, R. M., & Cellini, H. (1999). Assessing risk of recidivism in adult male sexually abusive violent offenders. Offender Programs, Social and Behavioral Rehabilitation in Prisons, Jails and the Community, 3, 49-50 and 60-61. Worling, J. (2004). The estimate of risk of adolescent sexual offense recidivism (ERASOR): preliminary psychometric data. Sexual Abuse: A Journal of Research and Treatment, 16(3), 235-252. Worling, J. (1995). Sexual abuse histories of adolescent male sex offenders: Differences on the basis of the age and gender of their victims. Journal of Abnormal Psychology, 104, 610-613. Worling, J., & Curwen, T. (2006a). Estimate of risk of adolescent sexual offense recidivism (ERASOR). Ontario: Thistletown Regional Center for Children and Adolescents, Ontario Ministry of Community and Social Services. Worling, J., & Curwen, T. (2001). The ERASOR. Worling, J., & Curwen, T. (2000b). Adolescent sexual offender recidivism: success of specialized treatment and implications for risk prediction. Child Abuse and Neglect, 24, 965-982. Yochelson, S., & Samenow, S. (1976). The criminal personality (Vol. 1). New York: Jason Aronson, Inc. Zonlondek, S., Abel, G., Northey, W., & Jordan, A. (2001). The self-reported behaviors of juvenile sex offenders. Journal of Interpersonal Violence, 16(1), 73-89. 135 APPENDICES 136 APPENDIX A INSTITUTIONAL REVIEW BOARD (IRB) LETTER 137 138 APPENDIX B RESEARCH AGREEMENT 139 140 APPENDIX C STATIC-99 RISK ASSESSMENT INSTRUMENT 141