Vtdoc Vermont Assessment of Sex Offender Risk Manual 2001
Download original document:
![Brief thumbnail](/media/cache/8f/1a/8f1a3b2fcb38d4f12f1231cc316e65bb.jpg)
Document text
![Brief thumbnail](/media/cache/8f/1a/8f1a3b2fcb38d4f12f1231cc316e65bb.jpg)
Document text
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
______________ Vermont Assessment of Sex Offender Risk Manual ________________ Research Edition 2001 Robert J. McGrath and Stephen E. Hoke Acknowledgements Appreciatio n is extended to Georgia Cumming and Maureen Buell of the Vermont Department of Corrections for serving as members of an expert panel during initial development of the Vermont Assessment of Sex Offender Risk (VASOR) and to Katherine Burns, Shelly Goddard, and Angela Ross for data collection and coding. In addition, appreciation is extended to Joy Livingston for assistance with conducting statistical analyses on the scale. Contact Information For Further Information Contact: Robert J. McGrath, M.A. Clinical Director Vermont Treatment Program for Sexual Aggressors Vermont Department of Corrections 105 Happy Valley Road Middlebury, Vermont 05753 E-mail: rmcgrath@sover.net Stephen E. Hoke District Director Office of Probation and Parole Vermont Department of Corrections 700 Exchange Street Middlebury, Vermont 05753 E-mail: shoke@doc.state.vt.us This VASOR Manual may also be obtained from the Center for Sex Offender Management (CSOM) website at www.csom.org. Copyright © 2002 Robert J. McGrath and Stephen E. Hoke Table of Contents Overview and Administration……………………………………………..…………….. 1 Introduction…………………………………………………………………………….. 1 Description and Administration…………………………………………..……………. 1 User Qualifications and Training…………………………………………..…………... 2 Psychometric Properties…………………………………………..……………………. 2 VASOR Cutoff Scores…………………………………………..……………………... 5 Modifications to Scoring Sheet and Instructions…………………………………….… 7 Summary…………………………………………..…………………………………… 7 Scoring Instructions………………………………………………………….…………… 9 Reoffense Risk Scale Items…………..……………………………..………………… 9 1. Prior Sex Offense Convictions………………………………………………….. 9 2. Prior Adult Convictions…………………………………………………………. 10 3. VOP's and Other Court Order Violations During Past Five Years……...…….… 10 4. Force Used During Current Offense……………………………………………... 10 5. Relationship to Victims………………………………………………………….. 11 6. Male Victim and/or History of Exhibitionism…………………………………… 11 7. Deviant Sexual Fixation…………………………………………………………. 11 8. Alcohol Abuse Use During Past Five Years…………………………………….. 12 9. Drug Abuse Use During Past Five Years………………………………………... 12 10. Address Changes During Past Year……………………………………………... 13 11. Time Employed or in School During Past Year…………………………………. 13 12. Reoffense During or After Treatment or Terminated Unsuccessfully From Treatment………………………………………………………………………… 13 13. Amenability to Treatment………………………………………………………... 13 Violence Risk Scale Items………………………………………..……………………. 14 1. Prior Convictions for Crimes Involving Violence……………………………….. 14 2. Prior Conviction for a Crime Involving a Potentially Deadly Weapon…………. 14 3. Force Used During Current Offense……………………………………………... 14 4. Sexual Intrusiveness of Current Offense………………………………………… 15 5. Physical Harm to Current Victim………………………………………………... 15 6. Victim Under Age 5, Over Age 55, or Mentally or Physically Disadvantaged…. 16 References…………………………………………..……………………………………… 17 Score Sheet……………………………………………………………………………….... 19 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Overview and Administration Introduction The Vermont Assessment of Sex Offender Risk (VASOR) is a risk assessment scale for adult male sex offenders age 18 and older. It was originally designed to assist probation and parole officers in making placement and supervision decisions (McGrath & Hoke, 1994). Because the VASOR does not provide a comprehensive survey of all factors relevant to sexual offending, it is best used as a decision aid along with professional judgement and other appropriate tools. Although reliability and validity studies are encouraging, it still should be considered an experimental instrument. Minor adjustments to the VASOR are described in this document, but it remains fundamentally unchanged since it was first developed in 1994. More detailed scoring instructions than those originally disseminated (McGrath & Hoke, 1994; Carich & Adkerson, 1995; Cumming & Buell 1997) are contained in this manual. Description and Administration The VASOR is composed of two scales, a 13-item reoffense risk scale and a 6-item violence scale. The reoffense risk scale is designed for assessing the likelihood of sexual recidivism. The violence scale is designed for assessing the nature of an individual's violence history and offense severity. The interaction of these variables, reoffense risk and violence, are considered important factors for determining an individual's overall risk level (McGrath, 1995). VASOR items were selected based on an extensive literature review (McGrath, 1991), the clinical experience of the authors, and the ease with which they could be scored using information routinely available in correctional case files. Most items concern an individual's history, such as prior offenses and gender of past victims, and, as such, are unchangeable. A few items are dynamic in nature, such as amenability to treatment and current lifestyle stability factors, and, therefore have the potential for changing over time. Empirically guided clinical judgement was used to assign weights to scale items. The scoring process ideally should include an interview with the individual, in addition to carefully reviewing correctional case file information. Scores on the two VASOR scales are plotted on a scoring grid where their intersection falls into one of three risk categories; low, moderate, or high (see VASOR scoring sheet on page 19). These risk categories can be used to inform placement and supervision decisions. Offenders who score in the "low" range (i.e., low reoffense risk score and low violence score) are generally considered appropriate for community supervision and treatment. Offenders who score in the "moderate" range may or may not be considered appropriate for community placement. Offenders who score in the "high" range (i.e., high reoffense risk score and/or high violence score) are generally considered inappropriate for community supervision and treatment. For public protection purposes, incarceration is generally recommended for offenders who score in the "high" range. Page 1 Vermont Assessment of Sex Offender Risk - Research Edition 2001 The VASOR was not designed to provide guidance about whether an individual should be incarcerated, or otherwise criminally sanctioned, for purposes of punishment or deterrence. One section of the scoring sheet, in particular, requires some explanation. The upper level of the "low" risk section that extends vertically on the left-hand side of the scoring grid is designed to identify non-contact sex offenders whose risk to reoffend may be high but who have no history of hands-on sexual offending or other violent behavior. These offenders may warrant close community supervision, but not incarceration. The VASOR risk categories were developed for use in Vermont. Other jurisdictions may develop cutoff scores for their own decision-making purposes that are different than those outlined here. User Qualifications and Training The VASOR was designed to be scored easily by probation and parole officers and correctional caseworkers. Before using the VASOR, however, it is critical that users carefully read this manual. In addition, users should complete training that includes scoring practice cases in order to optimize scoring accuracy and reliability. VASOR users should also have a basic understanding of risk factors related to sexual offense recidivism and principles of psychological assessment. An individual who understands the limitations of screening tools such as the VASOR should assume the responsibility for interpretation. Psychometric Properties To date, four studies have examined the psychometric properties of the VASOR. Overall, they provide encouraging initial data about its reliability and validity. McGrath & Hoke (1994a) Development Study The original sample used to develop the VASOR was composed of 122 adult male sex offenders. About 30% were rapists, 25% incest offenders, 35% child molesters, and 10% non-contact sexual offenders. About one third of these men were incarcerated and the remainder were supervised in the community (McGrath & Hoke, 1994a). The initial development study attempted to establish concurrent validity between VASOR risk levels (i.e., low, moderate, and high) and the clinical judgements of an expert panel (McGrath & Hoke, 1994a). The panel reviewed the case files of the 122 participants and identified those individuals they considered appropriate for community placement versus those judged to require incarceration for community safety reasons. These judgments were intended to reflect Vermont's sentencing practices in which about one-third of convicted sex offenders received a probationary sentence, one-third a short punitive sentence followed by probation, and one-third a straight incarcerative sentence. Panel members then determined each participant's VASOR score and plotted them on a scoring grid. Next, the panel divided the grid into three sections. The "low" and Page 2 Vermont Assessment of Sex Offender Risk - Research Edition 2001 "moderate" risk sections of the grid were established to set apart individuals who were deemed appropriate for community supervision and treatment and the "high" risk section for individuals believed to need incarceration. Using this method, 92.6% of the cases were appropriately assigned. The resulting scoring criteria have been used in Vermont for making sentencing recommendations and assigning community supervision levels. Packard and Gordon (1999) Concurrent Validity Study A study conducted by Packard and Gordon (1999) examined the degree to which the VASOR and several other risk instruments correlated with each other on a sample of 537 adult male sex offenders incarcerated in Washington State. All participants were enrolled in a prison sex offender treatment program and their average age at time of admission to the program was 37. About 82% of the participants were white, 7% black, 4% Hispanic, and 7% other minorities. Approximately 25% were incest offenders, 33% pedophiles, 24% hebephiles, and 19% rapists. Ninety-four percent of participants completed the treatment program. Overall, the researchers found that the VASOR correlated quite well with other risk assessment instruments studied. Minnesota Sex Offender Screening Tool (MnSOST) (Epperson, Kaul, & Huot, 1995) total scores correlated with the VASOR reoffense scores .68, p < .001, violence scores .46, p < .001, and total scores (sum of risk and violence scores) .72, p < .001. Rapid Risk Assessment for Sexual Offense Recidivism (RRASOR) (Hanson, 1997) scores correlated with VASOR reoffense scores .55, p < .001, violence scores .05, n.s., and total scores .38, p < .001. Psychopathy Checklist - Revised (Hare, 1991) scores correlated with VASOR reoffense scores .30, p<.001, violence scores .33, p < .001, and total scores .42, p < .001. The Violence Risk Appraisal Guide (Quinsey, Harris, Rice, & Cormier, 1998) scores correlated with the VASOR reoffense scores .45, p < .001, violence scores, .27, p < .01, and total scores .42, p < .001. Given that two of these instruments, the MnSOST and RRASOR, predict sexual reoffense with a moderate degree of accuracy, by inference, these data provide support for the reliability and predictive validity of VASOR. McGrath, Hoke, Livingston, and Cumming (2001) Reliability and Validity Study A recent study examined the reliability and validity of the VASOR on a sample of 172 adult male sexual offenders whose re-arrest and re-incarceration records were tracked for five years after release from Vermont prisons (McGrath, Hoke, Livingston, and Cumming, 2001). The mean age of the sample at release was 38 (SD = 10.7, range = 20-72). Almost one-half (44.2%) of participants had completed 12 or more years of education. Sixty-one participants (35%) had never been married. Consistent with Vermont's lack of racial diversity, only 1 participant (1%) was non-white. Almost one third of participants (32%) had one or more convictions for prior sexual offenses. On average, participants had been incarcerated for 55.3 months (SD = 26.6) and during that time each was encouraged to enter the Vermont Department of Corrections' sex Page 3 Vermont Assessment of Sex Offender Risk - Research Edition 2001 offender treatment program. About one half of the participants (47%) refused. Less than a third (29%) entered and completed the treatment program and slightly less than one quarter (24%) dropped out or were terminated from the treatment program. Participants were classified by offender type using definitions established by the Association for the Treatment of Sexual Abusers (Gordon et al., 1998). Based on these criteria, the sample was composed of 52 rapists (30%), 4 non-contact sex offenders (2%), 44 incest offenders (26%), and 72 child molesters (42%). Of these child molesters, 23 were boy object molesters (13%) and 49 were girl object molesters (29%). Data was obtained for all new sexual and violent reoffenses during the five-year follow-up period. The definition of sexual offenses also included any substantiation of a new sexual offense by Vermont's state child protection service agency. Violent offenses were defined as including sexual offenses as well as non-sexual violent offenses. For detailed definitions see Vermont Department of Corrections (1996). Charges were based on criminal record checks in the states where each participant was known to have resided during the study. Interrater reliability was assessed by having 12 raters each rate the same 12 cases by file review. Cases were from a data set described by McGrath, Cumming, Livingston, and Hoke (2003). Cases were a randomly selected proportionate stratified sample expected to produce a range of scores. Reliability was found acceptable for each of the scales, reoffense risk (ICC = .83) and violence (ICC = .89), as well as for the total score (ICC = .87). Most single item interclass correlation coefficients for individual items were above .80. This study also examined the concurrent and predictive validity of the VASOR. Similar to the findings of Packard and Gordon (1999), the VASOR correlated quite well with other risk assessment instruments, in this case, the RRASOR and Static -99. RRASOR scores correlated with VASOR reoffense risk scores .72, p < .01, violence scores .22, p < .01, and total scores .63, p < .01. Static-99 scores correlated with VASOR reoffense risk scores .74, p < .01, violence scores .31, p < .01, and total scores .70, p < .01. In terms of predictive validity, VASOR reoffense scores showed moderated predictive accuracy for both sexual recidivism (r = 0.41, ROC area = 0.76) and violent, including sexual, recidivism (r = 0.31, ROC area = 0.68). The predictive accuracy of the VASOR was also examined for subgroups of offenders. Predictive accuracy for sexual recidivism was preserved when the sample was divided into offenders who sexually victimized adult females (rapists, n = 52, r = .50, ROC area = 0.80) and those who victimized children (incest offenders and child molesters, n =116, r = .37, ROC area = .74). Langton et al., (2002) Reliability and Predictive Validity Study Langton and his associates (Langton, Barbaree, Harkins, Seto, & Peacock, 2002) recently evaluated the reliability and predictive validity of seven risk assessment instruments for sex offenders, one of which was the VASOR. To establish interrater reliability, two scorers completed file reviews on a prison sample of 468 sex offenders and independently scored each case. Interrater reliability for the Page 4 Vermont Assessment of Sex Offender Risk - Research Edition 2001 VASOR reoffense risk scale calculated as a Pearson correlation was .79. The mean reoffense risk score was 48.27 and the standard deviation of 20.64. These researchers also found that the VASOR had good predictive validity. Table 1 compares the VASOR with seven other sex offender risk instruments on a subset of 169 offenders using a fixed time-at-risk period of three years. Table 1. AUC's for ROC for Sexual Reoffense (n = 164) RRASOR VASOR STATIC-99 SVR-20 MnSOST -R SORAG VRAG PCL-R .77*** .75** .74*** .72*** .70** .66** .57** .46 RRASOR - Rapid Risk Assessment for Sexual Offense Recidivism (Hanson, 1997) VASOR - Vermont Assessment of Sex-Offender Risk (McGrath & Hoke, 2001) Static-99 - (Hanson & Thornton, 2000) SVR-20 - Sexual Violence Risk - 20 -(Boer, Hart, Kropp, & Webster, 1997) MnSOST-R - Minnesota Sex Offender Screening Tool - Revised (Epperson, Kaul, Hesselton, Alexander, & Goldman, 2000) SORAG -Sex Offender Risk Appraisal Guide (Quinsey, Harris, Rice & Cormier, 1998) VRAG - Violence Risk Appraisal Guide (Quinsey, Harris, Rice & Cormier, 1998) PCL-R - Psychopathy checklist-Revised (Hare, 1991) * p < .05, ** p < .01, *** p < .001 VASOR Cutoff Scores Cutoff scores on the 2001 VASOR scoring sheet are based roughly on the mean and standard deviation scores of participants in the McGrath, Hoke, Livingston, and Cumming (2001) study. The mean score for the reoffense risk scale was 43.5 (SD = 19.8, range = 5 - 115) and for the violence scale the mean was 24.5 (SD = 14.9, range = 0 - 90). With respect to the reoffense risk cutoff scores, five-year recidivism rates were calculated for three risk levels. As detailed in Table 2, almost all individuals who score below the mean are classified as "low" risk for sexual reoffense. Those who score up to approximately one standard deviation above the mean are classified as "moderate" risk. Individuals who score approximately more than one standard deviation above the mean are classified as "high" risk for sexual reoffense. As also noted in this table, there is a significant main effect for sexual and violent recidivism. In addition, there are statistically significant differences in sexual recidivism rates between individuals in the low and moderate risk levels, X2 (df 1) = 8.6, p < .01, and the low and high risk levels, X2 (df 1) = 18.6, p < .01. The differences between the moderate and high risk levels are not statistically significant. For violent recidivism, there are significant differences between low and moderate risk Page 5 Vermont Assessment of Sex Offender Risk - Research Edition 2001 levels, X2 (df 1) = 3.9, p < .05, and the low and moderate risk levels, X2 (df 1) = 11.4, p < .001. Again, differences between moderate and high risk levels are not statistically significant. Table 2. VASOR Reoffense Risk Scale Levels and Five Year Recidivism Rates Reoffense Risk Level Low Moderate High Totals a Score Range 0 - 40 41 - 60 61 - 125 0 - 125 Sample Size Sexual Recidivisma Any Violent Recidivismb 92 (53%) 43 (25%) 37 (22%) 172 (100%) 8 (9%) 12 (28%) 18 (49%) 38 (22%) 23 (25%) 18 (42%) 21 (57%) 62 (36%) Main effect, p < .001, b Main effect, p < .01. Although these results provide encouraging data about the predictive validity of the VASOR, assignment of specific percentage rates to individual sex offenders is not recommended at this time due, in part, to the small sample size in this study. With respect to violence scale cutoff scores, as detailed in Table 3, individuals who score below the mean are classified in the "low" violence category in terms of violence history and offense severity. Those who score up to approximately one standard deviation above the mean are classified in the "moderate" violence category. Individuals who score approximately more than one standard deviation above the mean are classified in "high" violence category. Because the violence risk scale was not designed to nor does it predict sexual or other types of reoffense risk particularly well, reoffense rates are not noted for this scale. Table 3. VASOR Violence Scale Levels Violence Level Score Range Low Moderate High Totals 0 - 25 26 - 40 41 - 125 0 - 125 Sample Size 100 50 22 172 (58%) (29%) (13%) (100%) As previously noted, scores on these two VASOR scales are plotted on a scoring grid where their intersection falls into one of three overall risk categories; low, moderate, or high (see VASOR scoring sheet). These risk categories can be used to inform placement and supervision decisions. Page 6 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Modifications to Scoring Sheet and Instructions As noted, although minor adjustments to the VASOR have been made, it remains fundamentally unchanged since it was first developed in 1994. More detailed scoring instructions than those originally disseminated (McGrath & Hoke, 1994; Carich & Adkerson, 1995; Cumming & Buell 1997) are described and new cutoff scores are recommended in this manual. The primary changes to the scoring sheet are minor adjustments in the cut-off scores for low, moderate and high risk levels using standard deviations to guide these decisions (McGrath, Hoke, Livingston, and Cumming, 2001). Additionally, the "Violence Risk" scale has been renamed the "Violence Scale" to indicate that its primary purpose is to quantify the severity of an individual's violence history rather than the likelihood of violent recidivism. Finally, the item "Amenability to Outpatie nt Treatment" has been renamed "Amenability to Treatment" in recognition of the fact that the VASOR can be used with both prison and community-based populations. Summary Overall, the results of these studies provide support for the use of the VASOR as an aid in making sex offender placement and supervision decisions. It is important to state again that the VASOR does not provide a comprehensive survey of all factors relevant to sexual offending. Professional judgment and other relevant tools should be used in the decision making process. Until the findings outlined here have been further replicated, the VASOR is best considered an experimental scale. Page 7 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Page 8 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Scoring Instructions The VASOR is intended only for use with males who have been convicted of a least one sex or sex-related offense when they were age 18 or older. If the offender’s current sexual offense involves multiple victims or multiple offenses, score the most serious elements of each offense. If the offender is incarcerated at the time of scoring and a scale item involves scoring behavior during a time frame (e.g., during the past five years), score for the designated period of time that he was in the community prior to his current incarceration. If the offender is in the community at the time of scoring and a scale item involves scoring behavior during a time frame (e.g., during the past five years), score for the designated period of time that he was in the community prior to the date of the assessment. Adjust for any periods of incarceration greater than six months. If scoring from a file review and information about an item is not noted or unclear, assume that the risk factor is absent or otherwise give the lowest logical score. Reoffense Risk Scale Items 1. Prior Sex Offense Convictions . Use official records and credible offender and collateral reports. Count the total number of separate prior sex and sex-related convictions, including convictions in which the sentence was deferred. Sex-related means that the underlying nature of the offense behavior was sexual, such as when the person being evaluated was convicted of a reduced charge of simple assault, but was originally charged with sexual assault. Count all juvenile and adult convictions. Count institutional rule violations and probation, parole, or conditional release violations resulting in a conviction or finding of guilt. Count as one conviction instances of multiple convictions concerning the same victim if sentencing was on the same date. Count as convictions all sex offenses committed by individuals with severe mental illness or developmental delays if the offense would have likely resulted in the offender’s conviction except for his serious mental impairment. Do not count criminal arrests or charges. Do not count substantiations of child sexual abuse by a state child protection agency. Do not count the index offense. The index offense is the most recent sex offense. The index offense may involve multiple counts and multiple victims perpetrated at different times but all having the same sentencing date. Page 9 Vermont Assessment of Sex Offender Risk - Research Edition 2001 2. Prior Adult Convictions. Use official records and credible offender and collateral reports. Count the number of convictions for non-sex-related criminal offenses. Count each conviction separately, even if sentencing for multiple convictions occurred all on the same date. Count alcohol-related driving offenses, such as Driving While Intoxicated and Careless and Negligent driving, but exclude other motor vehicle offenses. Do not count convictions for probation, parole, and other conditional release violations. 3. VOP’s and Other Court Order Violations During the Past Five Years. Use official records and credible offender and collateral reports. Count the number of convictions for probation, parole, and other conditional release violations during the past five years. Count each conviction separately, even if sentencing for multiple convictions occurred all on the same date. Conditional release violations include convictions for violations of family court orders as well as violations of criminal court orders. If the offender is incarcerated, score for the five-year period that he was in the community prior to his current incarceration. 4. Force Used During Current Offense. Use official records and credible offender, victim, and other collateral reports. If the current crime involves multiple victims or multiple offenses, score the most serious. A hands -off offense is a non-contact sexual offense such as exhibitionism, public masturbation, obscene telephone calling, voyeurism, and child pornography crimes. Do not count failure to register as sex offender. A hands -on offense is any sex offense in which the offender has physical contact with the victim but does not use the degree of force described in either of the following two categories. Force greater than necessary to gain compliance or clear threats of physical harm to victim or others includes force that was clearly greater than necessary or involved realistic threats of physical harm to gain the victim’s compliance to engage in a sexual act or refrain from reporting the act. Count if the offender kidnapped the victim. Using one’s size, position of authority, grabbing the victim, or physically positioning the victim to gain compliance by itself does not count as greater than necessary force. Use of potentially deadly weapon includes the threat or actual use of any weapon that poses potential realistic physical harm to the victim’s life. A potentially deadly weapon would include a gun, knife, stick, or other potentially lethal weapon, including the offender’s fist or feet, if they are used in such a manner as to pose a realistic loss of life to the victim. Page 10 Vermont Assessment of Sex Offender Risk - Research Edition 2001 5. Relationship to Victims . Use official records and credible offender, victim, and other collateral reports. Consider all of the offender’s sex offense victims even if the offense did not lead to an arrest or conviction. If the offender has multiple victims, count the item with highest score. Living with at time of offense is defined as living in the same residence as the victim for 30 days or more prior to an offense. A nonresidential relative is a victim whose relationship to the offender is sufficiently close that marriage would normally be prohibited, such as a divorced or separated parent, uncle, grandparent, sibling, or stepsibling. A spouse, either married or common-law, who was not living with the offender at the time of the offense would be considered a nonresidential relative. An acquaintance is a victim who was not living with the offender at the time of the offense, and is neither a “nonresidential relative” nor a “stranger,” such as a neighbor. A stranger is a victim who did not know the offender 24 hours prior to the offense. 6. Male Victim and/or History of Exhibitionism. Use official records and credible offender, victim, and other collateral reports. Score if there is credible information that the offender has committed a sex offense involving a male victim or sexually exposed himself to a victim of either gender. The offender need not have been arrested or convicted for an offense to count on this item. For the purposes of this item, possession of pornography involving boys or exhibitionism as a prelude to hands-on sexual contact with a victim would not be scored. 7. Deviant Sexual Fixation. Use official records, sexual arousal test data (i.e., phallometric or Abel Screen test data), and credible offender, victim, and other collateral reports. This item is concerned with the “degree” to which an offender’s sexual interests and behaviors are deviant versus non-deviant. The evaluator may take into consideration the age of the offender and the opportunity that he has had to engage in age appropriate sexual relationships. Deviant is defined as sexual offending activity, such as sexual contact with minors, coercive sexual activity with a person of any age, and voyeurism. Non-deviant is defined as sexual activity that is age appropriate, legal, and consensual. A score of “0” indicates that the offender’s sexual interests or behavior appear to be more nondeviant than deviant. Score “0” if data indicate that the offender has a single sexual offense victim and has a history of mutually consenting, age-appropriate sexual relationships. Page 11 Vermont Assessment of Sex Offender Risk - Research Edition 2001 A score of “5” indicates that the degree of an offender’s deviant and nondeviant sexual interests or behavior appear to be quite similar. Score “5” if the offender has undergone any sexual arousal assessment (i.e., phallometric or Abel Screen test data) in which his arousal to deviant themes was about the same as his arousal to non-deviant themes. If sexual arousal test data are not available or if they do not indicate interest in deviant sexual themes, score “5” if the offender has 2 to 4 sexual offense victims and a history of mutually consenting, age-appropriate sexual relationships. A score of “10” indicates that the offender’s sexual interests or behavior appear to be more deviant than non-deviant. Score “10” if the offender has undergone any sexual arousal assessment (i.e., phallometric or Abel Screen test data), in which his arousal to deviant themes was significantly greater than his arousal to non-deviant themes. Regardless of sexual arousal test data results, score “10” if the offender has 5 or more sexual offense victims and/or minimal or no history of mutually consenting, age-appropriate sexual relationships. 8. Alcohol Abuse During Past Five Years . Use official records and credible offender, victim, and other collateral reports. If the offender is incarcerated, score for the five-year period that he was in the community prior to his current incarceration. No problem indicates that there is no alcohol use or that alcohol use does not interfere with the offender’s functioning. Some legal or social problems indicates that the offender’s alcohol use has caused some legal or occasional social problems. Serious life disruption indicates that the offender’s alcohol use played an important role in the commission of one or more sex offenses or has caused serious and frequent life disruption such as multiple alcohol-related convictions or job loss. 9. Drug Abuse During Past Five Years . Use official records and credible offender, victim, and other collateral reports. If the offender is incarcerate d, score for the five-year period that he was in the community prior to his current incarceration. No problem indicates that there is no drug use or that drug use does not interfere with the offender’s functioning. Some legal or social problems indicates that the offender’s drug use has caused some legal or occasional social problems. Any illegal drug use should receive a score of “3” or “5.” Serious life disruption indicates that the offender’s drug use played an important role in commission of one or more sex offenses or has caused serious and frequent life disruption such as multiple drug-related convictions or job loss. Page 12 Vermont Assessment of Sex Offender Risk - Research Edition 2001 10. Address Changes During Past Year Use official records and credible offender, victim, and other collateral reports. Count the number of address changes that the offender made during the past year. If the offender is incarcerated, score for the one-year period that he was in the community prior to his current incarceration. Do not count address changes that were required by the court or other governmental body in connection to pending charges of the index offense. 11. Time Employed or in School During Past Year. Use official records and credible offender, victim, and other collateral reports. Score based on the percent of full-time that the offender was employed and/or attended school during the last year. Do not count employment that is criminal in nature (e.g., drug dealing or prostitution). If the offender is incarcerated, score for the one-year period that he was in the community prior to his current incarceration. Score “0” if offender was retired, disabled, a homemaker, or undergoing an expected seasonal lay-off during the past year. 12. Reoffense During or After Treatment or Terminated Unsuccessfully From Treatment. Use official records and credible offender and other collateral reports. This item is concerned with how well the offender has responded to “specialized sex offender treatment.” Specialized sex offender treatment is defined as treatment, provided by a trained mental health clinician, that has attempted to change factors directly linked to the offender’s sexual offending behavior, such as distorted sexual cognitions, deviant sexual arousal, emotional self-regulation deficits, and intimacy deficits. Treatment that has focused solely on problems such as depression, anxiety, marital distress, and self-esteem is considered “non-specialized.” Score “20” if the offender, at any time, has committed a new sex offense while enrolled in or following completion of a “specialized sex offender treatment” program. Score “20” if the offender has dropped out of or been terminated from a “specialized sex offender treatment” program and has not subsequently completed or is not currently enrolled in a “specialized sex offender treatment” program. 13. Amenability to Treatment. Use offender self-report. This item is concerned with whether the offender meets the typical minimal requirements for admission into a “specialized sex offender treatment” program. Partial admission means that the offender admits to and accepts at least some responsibility for committing a sex offense. Willing to participate in treatment means that the offender agrees to participate in “specialized sex offender treatment” that is approved by his probation or parole officer or correctional caseworker. Page 13 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Violence Scale Items 1. Prior Convictions for Crimes Involving Violence. Use official records and credible offender and other collateral reports. Count juvenile and adult convictions. Count the number of prior criminal convictions and violations of probation/parole for violent offenses, including prior hands-on sex and sex-related offense convictions. Count if the underlying nature of an offense was violent (e.g., Disturbing the Peace amended from Simple Assault). Count each conviction separately, even if sentencing for multiple convictions occurred all on the same date. Do not count hands-off sexual offenses such as exhibitionism and voyeurism. Do not count property crimes (e.g., Arson, Burgla ry), unless persons were occupying the property at the time of the offense. Do not count offenses involving accidental injury or death (e.g., Driving While Intoxicated with injury or death resulting) unless there was clear intent to harm the victim. 2. Prior Conviction for a Crime Involving a Potentially Deadly Weapon. Use official records and credible offender and other collateral reports. Count juvenile and adult convictions. Count the number of prior criminal convictions and violations of probatio n or parole for crimes involving a potentially deadly weapon. Count each conviction separately, even if sentencing for multiple convictions occurred all on the same date. A potentially deadly weapon includes the threat or actual use of any weapon that poses potential realistic physical harm to the victim’s life. A potentially deadly weapon would include a gun, knife, stick, or other potentially lethal weapon, including the offender’s fist or feet, if they are used in such a manner as to pose a realistic loss of life to the victim. 3. Force Used During Current Offense. Use official records and credible offender, victim, and other collateral reports. If the current crime involves multiple victims or multiple offenses, score the most serious. A hands -off offense is a non-contact sexual offense such as exhibitionism, public masturbation, obscene telephone calling, voyeurism, and child pornography crimes. Do not count failure to register as sex offender. A hands -on offense is any sexual offense in which the offender has physical contact with the victim but does not use the degree of force described in either of the following two categories. Force greater than necessary to gain compliance or clear threats of physical harm to victim or others includes force that was clearly greater than necessary to gain the victim’s compliance, or involved realistic threats of physical harm to either frighten the victim or gain the victim’s compliance. Count if the offender kidnapped the victim. Using one’s size, position of authority, pinning the victim, or physically positioning the victim to gain compliance does not by itself count as greater than necessary force. Page 14 Vermont Assessment of Sex Offender Risk - Research Edition 2001 Use of potentially deadly weapon includes the threat or actual use of any weapon that poses potential realistic physical harm to the victim’s life. A potentially deadly weapon would include a gun, knife, stick, or other potentially lethal weapon, including the offender’s fist or feet, if they are used in such a manner as to pose a realistic loss of life to the victim. 4. Sexual Intrusiveness of Current Offense. Use official records and credible offender, victim, and other collateral reports. A hands -off offense is a non-contact sex offense such as exhibitionism, public masturbation, obscene telephone calling, voyeurism, and child pornography crimes. Do not count failure to register as sex offender. Fondling is a hands-on sex offense that does not involve any of the behaviors described in the following three categories. Digital penetration, fellatio, or cunnilingus involves the offender putting one or more of his fingers in the victim’s vagina or anus or any contact of the offender’s mouth with the victim’s vagina, penis, or anus. It also includes the victim being manipulated or forced to engage in any of these behaviors towards the offender. Actual or attempted penile penetration of the vagina or anus is any contact of the offender’s penis or some object with the victim’s vagina or anus. It also includes the victim being manipulated or forced to engage in any of these behaviors towards the offender. Bizarre or ritualistic behavior includes such activities as bondage, urinating or defecating on the victim, and torture. 5. Physical Harm to Current Victim. This item concerns only physical harm to the current victim. Use official records and credible offender, victim, and other collateral reports. No medical treatment required means that the victim did not suffer any physical injury. Injury not requiring formal medical attention includes minor bruises, scratches, and abrasions for which the victim does not need the attention of trained medical personnel. Treated for injury and released means treatment by trained medical personnel either in a hospital or outpatient setting, after which the victim was rele ased. It does not include a medical exam performed solely for the purpose of obtaining evidence about a sexual offense or for preventative treatment for a sexually transmitted disease. Hospitalized means admitted to a hospital bed for the care of physical injuries sustained during a sexual offense. Page 15 Vermont Assessment of Sex Offender Risk - Research Edition 2001 6. Victim Under Age 5, Over Age 55, or Mentally or Physically Disadvantaged. Use official records and credible offender, victim, and other collateral reports. Victim under age 5 means that any of the offender’s victims were under age 5 when the offender began sexually offending against them. Over age 55 means that any of the offender’s victims were age 55 or older when the offender sexually offended against them. Mentally or physically disadvantaged includes any impairment that seriously compromised the victim’s ability to defend him or herself from abuse or to provide a credible report about the abuse to authorities. Impairments include mental retardation (i.e., Intelligence Quotient below 70), severe mental illness, and severe alcohol or drug intoxication at the time of the offense. Page 16 Vermont Assessment of Sex Offender Risk - Research Edition 2001 References Boer, D. P., Hart, S. D., Kropp, P. R., & Webster, C. D. (1997). Manual for the Sexual Violence Risk - 20. Burnaby, British Columbia: Simon Fraser Universit y. Carich, M.S., & Adkerson, D. L. (1995). Adult sex offender assessment packet. Brandon, VT: Safer Society. Cumming, G., & Buell, M. (1997). Supervision of the sex offender (pp. 145-147). Brandon, VT: Safer Society Press. Epperson, D. L., Kaul, J. D., Hout, S. J., Hesselton, D. Alexander, W., & Goldman, R. (2000, November). Cross-validation of the Minnesota Sex Offender Screening Tool - Revised. Paper presented at the 19th Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, San Diego, CA. Epperson, D. L., Kaul, J. D., Hout, S. J. (1995, October). Predicting risk of recidivism for incarcerated sex offenders: Updated development on the Sex Offender Screening Tool (MnSOST). Paper presented at the 14th Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, New Orleans, LA. Hanson, R. K. (1997). The development of a brief screening scale for sexual offense recidivism. Ottawa: Solicitor General of Canada. Hanson, R. K., & Thornton, D. (2000). Improving risk assessment for sex offenders: A comparison of three actuarial scales. Law and Human Behavior, 24, 119-136. Hare, R. D. (1991). The Revised Psychopathy Checklist. Toronto, Ontario: Multi-Health Systems. Langton, C. M., Barbaree, H. E., Harkins, L., Seto, M. C., & Peacock, E. J. (2002, October). Evaluating the predictive validity of seven risk assessment instruments for sexual offenders. Paper presented at the 21st Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, Montreal, Quebec, Canada. McGrath, R. J. (1991). Sex-offender risk assessment and disposition planning: A review of empirical and clinical findings. International Journal of Offender Therapy and Comparative Criminology, 35(4), 329-351. McGrath, R. J. (1992). Five critical questions: Assessing sex offender risk. Perspectives, 16 (3), 6-9. McGrath, R. J., Cumming, G., Livingston, J. A., Hoke, S. E. (2003). Outcome of a treatment program for adult sex offenders: From prison to community. Journal of Interpersonal Violence, 18, 3-17. McGrath, R. J. & Hoke, S. E. (1994). Vermont Assessment of Sex Offender Risk. Waterbury, VT: Author. Page 17 Vermont Assessment of Sex Offender Risk - Research Edition 2001 McGrath, R. J. & Hoke, S. E. (1994a). Unpublished raw data. Middlebury, VT. McGrath, R. J., Hoke, S. E., Livingston, J. A., & Cumming, G., (2001, November). The Vermont Assessment of Sex Offender Risk (VASOR): An initial reliability and validity study. Paper presented at the 20th Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, San Antonio, TX. Packard, R., & Gordon, A. (1999, September). An investigation of actuarial risk scales: Concordance and factor analysis. Paper presented at the 18th Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, Lake Buena Vista, FL. Quinsey, V. L., Harris, G. T., Rice, M. E., & Cormier, C. A. (1998). Violent offenders: Appraising and managing risk. Washington, DC: American Psychological Association. Vermont Department of Corrections. (1996). Sentencing options: Criminal justice system user manual for selecting sentencing programs. Waterbury, VT: Author. Page 18 Vermont Assessment of Sex Offender Risk Name DOB Rater Date Reoffense Risk Scale Violence Scale 1. 1. Prior Convictions for Crimes Involving Violence Prior Sex Offense Convictions none = 0 2. one = 10 two or more = 20 (Count prior hands-on sex offense convictions) none = 0 one = 5 two = 10 three = 15 four or more = 20 Prior Adult Convictions (Do not count items on # 1 or # 3) none = 0 one = 1 two or three = 3 four to six = 5 seven or more = 10 2. Prior Conviction for a Crime Involving a Potentially Deadly Weapon 3. VOP's and Other Court Order Violations During Past Five Years none = 0 none = 0 3. Force Used During Current Offense 4. one = 2 two or more = 5 yes = 15 hands-off offense = 0 hands-on offense = 5 force greater than necessary to gain compliance or clear threats of physical harm to victim or others = 15 use of potentially deadly weapon = 30 Force Used During Current Offense hands-off offense = 0 hands-on offense = 5 force greater than necessary to gain compliance or clear threats of physical harm to victim or others = 8 use of a potentially deadly weapon = 10 4. Sexual Intrusiveness of Current Offense living with at time of offense = 0 nonresidential relative/acquaintance = 5 stranger = 10 hands-off = 0 fondling = 3 digital penetration, fellatio, or cunnilingus = 5 actual or attempted penile penetration of vagina or anus = 10 bizarre or ritualistic behavior = 20 6. Male Victim and/or History of Exhibitionism 5. Physical Harm to Current Victim none = 0 no medical treatment required = 0 injury not requiring formal medical attention = 10 treated for injury and released = 20 hospitalized = 30 5. Relationship to Victims 7. yes = 10 Deviant Sexual Fixation single victim and history of consenting, age appropriate sexual relationships = 0 two to four victims and history of consenting, age appropriate sexual relationships = 5 five or more victims and/or little or no history of consenting, age appropriate sexual relationships = 10 8. 6. Victim Under Age 5, Over Age 55, or Mentally or Physically Disadvantaged no = 0 yes = 10 ___________ Alcohol Abuse During Past Five Years no problems = 0 some legal or social problems = 3 serious life disruptions = 5 9. Total Drug Abuse During Past Five Years no problems = 0 some legal or social problems = 3 serious life disruptions = 5 Risk Level 10. Address Changes During Past Year none = 0 one = 2 80 two or more = 5 60% or more = 0 40-59% = 2 60 under 40% = 5 Reoffense Risk Score 12. Reoffense During or After Treatment, or Terminated Unsuccessfully from Treatment none = 0 High 70 11. Time Employed or in School During Past Year yes = 20 50 40 30 13. Amenability to Treatment Moderate Low 20 full or partial admission and willing to participate in treatment = 0 denies offense or unwilling to participate in treatment = 10 10 ________ 10 2001 Research Edition Copyright © 2001 R. J. McGrath & S. E. Hoke Total 20 30 40 50 60 Violence Score 70 80