- Email: [email protected]
Assessing suspected child molesters
(‘hrld Abuse & Neglect. Vol. 17. pp. I69- 174, I993 Pnnted in the U.S.A. All rights reserved.
0145.2134/93$6.00+.00 Copyright 0 1993 Pergamcn Press Ltd.
ASSESSING SUSPECTED
CHILD MOLESTERS
JUDITH V. BECKER University of Arizona, Tucson, AZ
VERNON L, QUINSEY Queen’s University, Kingston, Ontario
Abstract-Clinical evaluations can play a useful role in making a variety of dispositional decisions concerning child molesters provided they are done carefully, using relevant techniques, and their limitations are clearly understood. Assessment instruments and techniques that can differentiate child molesters from other persons, that are related to treatment planning, and that predict the commission of future sexual offending are the most useful. Sex offense history and phallomet~c assessments are useful in planning treatment and supervision programs and in assessing risk. The Psychopathy Checklist is also a strong predictor of sexual recidivism but its usefulness in treatment planning is equivocal. Measures of attitudes and beliefs that support child molestation are suitable for designing treatment programs but their prognostic significance remains to be established. More general measures of personality and psychopathology have not been established as predictors of recidivism but can be used to address collateral clinical issues. Key Words-Child
molesters, Assessment,
INTRODUCTION THE SEXUAL ABUSE of children continues to be a major problem in our society. Sexual offending against children, as well as societal concern about this problem, frequently leads to the referral of suspected child molesters to mental health professionals for assessment of various kinds. However, child molesters rarely refer themselves for evaluations; typically, they are referred by attorneys, the court, or child protection agencies. Although these referrals are made in order to obtain answers to a variety of questions, clinical assessments are only suited to providing answers to particulars of them. Referrals may be made to: (a) determine whether or not a person has committed a specific sexual offence, (b) make evaluations in custody disputes as to whether or not the individual represents a risk to the children involved, (c) determine whether or not the individual would be a good candidate for treatment, (d) specify what the treatment needs of the individual are, and (e) determine the amount of risk an individual might present to the community (Earls, I 99 1). The question of determining whether or not a person has committed a sexual offence is not one that clinical assessment can address. There are no psychological tests or techniques that indicate whether someone has engaged in sexual behaviors with children; such questions are best left to detectives and the courts. Questions of relative risk (b and e above) can be adReceived for publication May 28, 1991; final revision received May 8, 1992; accepted May 18, 1992. Reprint requests should be sent to Judith V. Becker, Psychiatry Department, Campbell Ave., Tucson, AZ 85724. 169
University of Arizona, 1501 North
170
J. Becker and V. Quinsey
dressed by clinical assessments, albeit very imperfectly, under certain conditions. Chief among these is the availability of a corroborated history, including the person’s official criminal record. Without such a history, a clinical assessment cannot ordinarily provide an appraisal of risk. Even with this information, both the providers and consumers of clinical assessments must keep in mind the probabilistic nature of such appraisals, particularly that a proportion of the people assessed for risk will be incorrectly predicted to sexually assault a child (these are false positive errors). A review of the issues involved in prediction is provided by Quinsey and Walker (in press). Determining the treatability of child molesters or their treatment needs presents quite a different classification or decision situation for the clinician. There is, of course, the possibility of making false positive and false negative errors, but the implications of these errors (particularly of wrongly concluding that someone needs treatment or needs a particular kind of treatment) are seldom nearly so important as making errors in appraising risk. By treatment in this context, we are only referring to interventions that are designed to reduce the likelihood that a person will engage in sexual activities with children. Treatment can, of course, be required for a wide variety of other problems but only those problems that are related to the probability with which a person victimizes children are at issue here. There are a variety of assessment techniques that can be used to choose a particular treatment modality or to monitor the acquisition of therapeutic goals. Examples of these include hormonal assays, psychological tests of suitability for verbal interventions, behavioral tests of heterosocial skillfulness, tests of sexual knowledge, and knowledge of specific program content, such as relapse prevention techniques. These assessments are most commonly used when a client has already been accepted into a treatment program, and will not be commented on further here. Typically, clinical evaluations of child molesters include a clinical interview and psychological testing; more rarely they include phallometric assessment and, in some cases, polygraphy. Below, we briefly describe these methods of assessment and comment on the uses to which they may be best put. The information gathered from these various sources are integrated in order to provide a tentative explanation of why the offence occurred; that is, the clinician seeks to develop a theory to explain the specific offence by specifying the situational antecedents and personal characteristics of the offender that led to the sexual victimization. Regardless of whether the purpose of the assessment involves treatment or risk issues, clinicians are well advised to obtain as much detail about the alleged molestation(s) as possible. This can be accomplished by requesting police reports and victim statements and, where possible, directly interviewing the victim(s). These are the data to be explained by the clinician’s theory. In addition, criminal history remains one of the best predictors of future sexual recidivism among child molesters. Extrafamilial child molesters are more likely to recidivate than strictly intrafamilial child molesters, and men who choose boy victims are more likely to recidivate than men who choose girl victims; the probability of recidivism rises with each additional sexual offence (Quinsey, 1986).
THE INTERVIEW Interviews of the offender are similar to other clinical interviews except the validity of the information provided by the offender is more likely to be suspect. Frequently, the alleged child molesters distort information, falsely deny that the alleged offence occurred, and report difficulty in recalling events surrounding the offence. A noted by Earls (199 l), “There is surprisingly little empirical research concerning the reliability and validity of the information obtained in a clinical interview.” In an effort to obtain valid self-report data on individuals who committed sexual offences
Assessing child molesters
171
against children, Abel, Becker, Mittleman, Cunningham-Rathner, Rouleau, and Murphy (1987) interviewed sexual offenders under a certificate of confidentiality. Federal law goveming these certificates insured that no city, county, state, or federal agency could compel investigators to reveal the identity of the subjects participating in the project. This arrangement gave some level of confidence to the participants that the information they disclosed would not be reported to criminal justice authorities. A total of 377 men were seen who had engaged in sexual activity with children not related to them. The mean number of victims for those men who had molested young girls was 19.8 and the mean number of victims for those men who had molested young boys was 150.2. Among the 203 men who had engaged in intrafamilial sexual abuse, those who were sexually involved with female relatives had an average of 1.8. victims, and for those who had involved themselves with male relatives, the mean number of victims was 1.7. If those interviews had been conducted within a forensic context without a certificate of confidentiality, it is highly doubtful that the child molesters would have been quite so forthcoming in divulging the extent of their sexual activity with children. The clinical interview should obtain demographic information, a full history of both paraphilic and non-paraphilic behavior, history of sexual and physical abuse, social history, alcohol and substance abuse history, medical history, criminal justice history, and employment history (Becker & Kaplan, 1990). It is particularly important that the suspected molesters be questioned about paraphilic behavior as well as fantasies. Because individuals who engage in norm-violating sexual behavior are frequently reluctant to disclose sexual behaviors, it is critical that the clinician ask the interviewee about every type of paraphilia, including bestiality, coprophilia, exhibitionism, fetishism, frottage, pedophilia, public masturbation, rape, sadism, masochism, necrophilia, obscene phone calls, obscene letters, transvestism, and urolagnia. The client should be questioned as to whether he has fantasized about these categories of behavior and whether or not he has ever acted on these fantasies. It is important to note that individuals who engage in norm-violating sexual behaviors are a heterogeneous group (Quinsey, 1986). For some, the deviant sexual behavior may be part of a paraphilic interest pattern, for others the behavior may represent an overall pattern of antisocial behavior, and, for yet others, both may be involved. It is important in this connection, that Hare’s Psychopathy Checklist is a strong predictor of future sex offences among both rapists and child molesters (Quinsey, Rice, & Harris, submitted). In a small percentage of cases, individuals with major psychiatric disorders, such as psychosis may commit sexual offenses and, more frequently, persons with significant problems of alcohol abuse may sexually victimize children. It is, therefore, important to assess the presence of other clinical problems during the interview because these problems may contribute to the commission of the sexual misbehaviors.
PSYCHOLOGICAL
TESTING
Psychological tests can be divided into three categories: tests of sexual interest and attitudes, objective personality tests, and projective tests. Tests of sexual interest include such assessments as the Sexual Interest Cardsort. Barlow, Leitenberg, and Agras, ( 1969) devised a cardsort technique in which various sexual scenes were typed on cards. The individual sorts the cards into four categories according to the amount of sexual interest they elicit (Tollison & Adams, 1979). These cardsort tasks are good measures of sexual interest with cooperative subjects. Clinicians have found that sex offenders frequently have attitudes or beliefs that support their inappropriate sexual behavior. These beliefs often involve cognitive distortions. A cognition scale has been developed by Abel and his colleagues ( 1984) to measure these distortions and has been found to distinguish child molesters from non-child molesters (Able, Gore,
172
J. Becker and V. Quinsey
Holland, Camp, Becker, & Rathner, 1989). These cognitive distortions are appropriate targets for intervention as they may serve to maintain sexual behaviors toward children, although their prognostic significance remains to be established. The MMPI (Minnesota Multiphasic Personality Inventory) has been used extensively in assessment of sexual offenders. While the MMPI can be useful in clinical assessment for the purposes for which it was designed, its discriminant validity in the assessment of child molester is poor (e.g., Quinsey, Arnold, & Pruesse, 1980). Levin and Stava (1987) found methodological problems in most of the 36 MMPI studies that they reviewed and concluded that, “in general, negative or inconsistent findings outweigh those of a positive nature.” Projective tests such as Rorschach, Thematic Apperception Test, and Draw-A-Person Test, have frequently been used to assess child molesters but there is little evidence concerning the reliability or validity of these measures; the studies that have been done in this area are very methodologically weak (Quinsey, 1986). We do not recommend the use of the MMPI or projective tests for appraising the risk that child molesters present to the community or in determining treatment targets related to sexual reoffending. The results of psychological testing, like the results of interviewing, must be interpreted with caution. Dissimulation is a common problem in the assessment of child molesters, and only some psychological tests, such as the MMPI, contain scales that attempt to address faking. There is no “profile of the child molester” that can reliably be identified with standard psychological tests. A comprehensive review of questionnaires and psychological tests that have been used to assess offenders can be found in Hanson, Cox, and Woszcsyna (199 1).
PHALLOMETRIC
ASSESSMENT
In a phallometric assessment, changes in penile tumescence occasioned by a variety of sexual and nonsexual stimuli are measured with a penile plethysmograph. The relative magnitude of these responses reflect the amount of sexual interest elicited by the types of stimuli shown. Phallometric assessments of sexual age and gender preferences have excellent discriminant validity with extrafamilial child molesters and the classification accuracy of some stimulus sets and procedures has been well documented (Freund & Blanchard, 1989). Exclusively incestuous offenders are less likely to show inappropriate sexual age preferences in phallometric assessments than extrafamilial child molesters (Barbaree & Marshall, 1989; Quinsey, Chaplin, & Carrigan, 1979). Inappropriate sexual age preferences measured in phallometric assessments predict subsequent new offences against children (Barbaree & Marshall, 1988; Quinsey, Rice, & Harris, submitted; Rice, Quinsey, & Harris, 199 1). These assessments attempt to measure the degree of sexual preference among stimulus categories; they do not detect whether someone has committed a specific sexual offence. Although phallometric assessments are related to the risk of reoffending an offender presents, they are far from infallible. In particular, some child molesters can intentionally influence their responses in order to appear to have normal sexual preferences (e.g., Freund, Watson, & Rienzo, 1988); because of this possibility, deviant phallometric profiles are more easily interpreted than normal profiles. Phallometric data are useful chiefly in determining whether an offender selected a child victim because of a specific sexual interest in children, or as a matter of convenience or opportunity. Such a determination has obvious therapeutic and supervisory implications. Phallometric assessments are, however, somewhat more technical than regular clinical evaluations and require specialized apparatus. Laws and Osbom (1983) have provided a useful description of laboratory phallometric procedures; some of the technical issues in performing these assessments are commented on by Quinsey and Laws (1990).
Assessing child molesters
173
LIE DETECTION Because child molesters’ potential dissimulation often clouds their assessment, it is understandable that clinicians and supervisory authorities have sought techniques that reveal when offenders are lying about the commission of previous sexual offences and their true sexual interests. Unfortunately, because there are currently no reports of controlled studies using polygraphy in the literature, the utility of lie detection methodology in this area is impossible to assess. There are, however, a priori considerations that should give practitioners pause before adopting these techniques for regular use (Lalumiere & Quinsey, 199 1). These considerations include the unknown effects of repeated testing and the high possibility of false positive errors in some circumstances. The physiological responses measured by the polygraph are not specific correlates of lying.
CONCLUSIONS Referrals of child molesters for clinical evaluations are sometimes inappropriate, as, for example, when the purpose of the evaluation is to determine whether the referred person actually committed a sexual offence. Other referrals may involve questions for which the referring agent desires definite answers but for which only probabilistic answers are possible. Assessments of dangerousness, for example, can only provide estimates of risk and the precision with which it is possible to make these estimations varies widely according to the details of the case and the amount of information available to the clinician. Although there are limitations to any of the assessment techniques described above and client faking remains a pervasive problem, some of these techniques are better suited to assessments of child molesters than others. Those techniques that can differentiate child molesters from other persons, that are related to the planning of treatment programs, and that predict the commission of future sexual offending are, of course, the most useful. Sex offence history and phallometric assessments are useful in planning treatment and supervision programs and in assessing risk. The Psychopathy Checklist is also a strong predictor of sexual recidivism but its usefulness in treatment planning is equivocal. Measures of attitudes and beliefs that support child molestation are suitable for designing treatment programs but their prognostic significance remains to be established. More general measures of personality and psychopathy have not been established as predictors of recidivism but can be used to address collateral clinical issues. Clinicians integrate information from a variety of sources to develop a theory designed to explain the offence history of there child molester who is being assessed. Such a theory can be used to select appropriate interventions and provide a focus for supervision. Clinical evaluations can play an important and useful role in making a variety of dispositional decisions concerning child molesters, providing they are done carefully using relevant techniques and their limitations are clearly understood.
REFERENCES Abel, G. G., Becker, J. V., Mittelman, M. Cunninham-Rathner, J.,Rouleau, J., & Murphy, W. (1987). Self-reported sex crimes of non-incarcerated paraphiliacs. Journal of Interpersonal Violence, 2, 3-25. Abel, G., Becker, J., Cunningham-Rathner, J.,Rouleau, J., Kaplan, M., & Reich. J. (1984). The treatment of child molesters: A manual. Unpublished manuscript. Abel, G. G., Gore, D., Holland, C., Camp, N., Becker., & Rathner, J. (I 989) The measurement of cognitive distortions of child molesters. Annals of Sex Research. 2, 135-l 53.
174
J. Becker and V. Quinsey
Barbaree, H. E., & Marshall, W. L. (1988). Deviant sexual arousal, offense history, and demographic variables as predictors of reoffense among child molesters. Behavioral Sciences and the Law, 6, 267-280. Barbaree, H. E., & Marshall, W. L. (1989). Erectile responses among heterosexual child molesters, father-daughter incest offenders and matched nonoffenders: Five distinct age-preference profiles. Canadian Journal of Behavioural Science, 21, 70-82. Barlow, D., Leitenberg, H., & Agras, W. (1969). The experimental control of sexual deviations through manipulation of the noxious scene in court sensitization. Journal of Abnormal Psychology, 74, 596-60 I. Becker, J. V., & Kaplan, M. S. (1990). Assessment of the adult sex offender. In P. McReynolds, J. Rosen, & G. Chelune (Eds.), Advances in psychological assessment. (pp. 26 l-283). New York: Plenum. Earls, C. (199 I). Clinical issues in the psychological assessment of child molesters. In W. O’Donohue & J. Greer (Eds.), The sexual abuse of children; Theory, research and therapy. New York: Erlbaum. Freund, K., & Blanchard, R. (1989). Phallometric diagnosis of pedophlia. Journal of Consulting and Clinical Psychology, 57, 100-105. Freund, K., Watson, R., & Rienzo, D. (1988). Signs of feigning in the phallometric test. Behaviour Research and Therapy, 26, 105-I 12. Hanson, R. K., Cox, B., & Woszcsyna, C. (199 1). Sexuality, personality and attitude questionnairesfor sex offenders: A review. Cat. No. JS4- l/ 199 1- 13. Ottawa: Supply and Services Canada. Lalumiere, M. L., & Quinsey, V. L. (199 1) Polygraph testing ofchild molesters: Are we ready? Violence Update, l( 1l), 3, 9, 11; (12), 6-7. Laws, D. R., & Osborn, C. A. (1983).How to build and operate a behavioral laboratory to evaluate and treat sexual deviance. In J. G. Greer, & I. R. Stuart (Eds.). The sexual aggressor: Current perspectives on treatment. (pp.292335). New York Van Nostrand Reinhold. Lewin, S., & Stava, L. (1987). Personality characteristics of re-offenders: A review. Archives of Sexual Behavior, 16, 57-59. Quinsey, V. L. (1986), Men who have sex with children. In D. N. Weisstub (Ed.), Law and mental health: Znternationalperspectives, 2, (pp. 140-172). New York: Pergamon. Quinsey, V. L., Arnold, L. S., & Pruesse, M. G., (1980). MMPI profiles of men referred for a pretrial psychiatric assessment as a function of offense type. Journal of Clinical Psychology, 36,4 IO-4 17. Quinsey, V. L., Chaplin, T. C., & Canigan, W. F. (1979). Sexual preferences among incestuous and non-incestuous child molesters. Behavior Therapy, 10, 562-565. Quinsey, V. L., & Laws, D. R. (1990). Validity ofphysiological measures of pedophilic sexual arousal in a sex offender population: A critique of Hall, Protor, and Nelson. Journal of Consulting and Clinical Psychology, 58, 886-888. Quinsey, V. I., &Walker, W. D. (in press). Dealing with dangerousness: Community risk management strategies with violent offenders. In R. Peters, R. J. McMahon, & V. L. Quinsey (Eds.). Aggression and violence throughout the lifespan. Quinsev. V. L., Rice, M. E., & Harris, G. T. (submitted). Psychopathy, sexual deviance, and recidivism among sex - offenders released from a maximum security psychiatric institution. Rice, M. E., Quinsey, V. L., & Harris, G. T. (1991). Sexual recidivism among child molesters released from a maximum security psychiatric institution. Journal of Consulting and Clinical Psychology, 59, 38 l-386. Tollison, C. D., & Adams, H. E. (1979). Sexual disorders: Treatment, theory, and research. New York: Gardner.
Resum&-Les evaluations cliniques peuvent jouer un role utile dans la prise dune s&e de dispositions a l’egard des agresseurs d’enfants a condition qu’elles soient faites prudemment en utilisant des techniques adequates et que leurs limites soient clairement comprises. Les instruments et les techniques d’tvaluation les plus utiles sont celles qui permettent de differentier les agresseurs d’enfants des autres personnes, celles qui sont associeees au programme therapeutique et celles qui predisent la perpetuation future dune agression sexuelle. L’histoire de l’agression sexuelle et les evaluations phallomttriques sont utiles pour la planification des programmes de traitement, de supervision et pour l’bvaluation du risque. La “Psychopathy Checklist” forme un bon predicteur de recidives sexuelles mais son utilite dans la planification d’un traitement est Bquivoque. Les mesures des comportements et des croyances qui sont a la base de l’agression sexuelle des enfants conviennent pour l’tlaboration du traitement, mais leur valeur pronostique reste a etablir. Des evaluations de personnalitt ou de psychopathologic plus getterales, predictives de recidives n’ont pas ttt etablies, mais peuvent Ctre utilisees a des fins cliniques paralltlles. Resumen-Las evaluaciones clinicas pueden jugar un rol muy 6til al hater una variedad de decisiones en relation a 10s que molestan a 10s nifios siempre que Sean hechas con cuidado utilizando ttcnicas apropiadas comprendiendo claramente sus limitaciones. Las mas utiles son 10s instrumentos de evaluation y las tecnicas que pueden diferenciar 10s que molestan a 10s nifios de otras personas, las que esmn relacionadas al plan de1 tratamiento, y las que predicen que se.cometan actos futures sexualmente ofensivos. El historial de ofensas y las evaluaciones falometricas son itiles para planificar el tratamiento y para supervisi6n de programas asi coma para evaluar el riesgo. El Inventario Psiconkico es tambien un fuerte predictor de recidivism0 sexual pero su utilidad en la planificaci6n de1 tratamiento es equivoca. Las medias de attitudes y creencias que apoyan 10s actos de molestar sexualmente a 10s mifios son utiles para d&tar programas de tratamiento pero su importancia en el pronbstico no ha sido estabecida. Medidas rna.s generales de personalidad y psicopatologia no han sido establecidas coma predictoras de recidivism0 pero pueden utilizarse para enfocar aspectos clinicos colaterales.
0145.2134/93$6.00+.00 Copyright 0 1993 Pergamcn Press Ltd.
ASSESSING SUSPECTED
CHILD MOLESTERS
JUDITH V. BECKER University of Arizona, Tucson, AZ
VERNON L, QUINSEY Queen’s University, Kingston, Ontario
Abstract-Clinical evaluations can play a useful role in making a variety of dispositional decisions concerning child molesters provided they are done carefully, using relevant techniques, and their limitations are clearly understood. Assessment instruments and techniques that can differentiate child molesters from other persons, that are related to treatment planning, and that predict the commission of future sexual offending are the most useful. Sex offense history and phallomet~c assessments are useful in planning treatment and supervision programs and in assessing risk. The Psychopathy Checklist is also a strong predictor of sexual recidivism but its usefulness in treatment planning is equivocal. Measures of attitudes and beliefs that support child molestation are suitable for designing treatment programs but their prognostic significance remains to be established. More general measures of personality and psychopathology have not been established as predictors of recidivism but can be used to address collateral clinical issues. Key Words-Child
molesters, Assessment,
INTRODUCTION THE SEXUAL ABUSE of children continues to be a major problem in our society. Sexual offending against children, as well as societal concern about this problem, frequently leads to the referral of suspected child molesters to mental health professionals for assessment of various kinds. However, child molesters rarely refer themselves for evaluations; typically, they are referred by attorneys, the court, or child protection agencies. Although these referrals are made in order to obtain answers to a variety of questions, clinical assessments are only suited to providing answers to particulars of them. Referrals may be made to: (a) determine whether or not a person has committed a specific sexual offence, (b) make evaluations in custody disputes as to whether or not the individual represents a risk to the children involved, (c) determine whether or not the individual would be a good candidate for treatment, (d) specify what the treatment needs of the individual are, and (e) determine the amount of risk an individual might present to the community (Earls, I 99 1). The question of determining whether or not a person has committed a sexual offence is not one that clinical assessment can address. There are no psychological tests or techniques that indicate whether someone has engaged in sexual behaviors with children; such questions are best left to detectives and the courts. Questions of relative risk (b and e above) can be adReceived for publication May 28, 1991; final revision received May 8, 1992; accepted May 18, 1992. Reprint requests should be sent to Judith V. Becker, Psychiatry Department, Campbell Ave., Tucson, AZ 85724. 169
University of Arizona, 1501 North
170
J. Becker and V. Quinsey
dressed by clinical assessments, albeit very imperfectly, under certain conditions. Chief among these is the availability of a corroborated history, including the person’s official criminal record. Without such a history, a clinical assessment cannot ordinarily provide an appraisal of risk. Even with this information, both the providers and consumers of clinical assessments must keep in mind the probabilistic nature of such appraisals, particularly that a proportion of the people assessed for risk will be incorrectly predicted to sexually assault a child (these are false positive errors). A review of the issues involved in prediction is provided by Quinsey and Walker (in press). Determining the treatability of child molesters or their treatment needs presents quite a different classification or decision situation for the clinician. There is, of course, the possibility of making false positive and false negative errors, but the implications of these errors (particularly of wrongly concluding that someone needs treatment or needs a particular kind of treatment) are seldom nearly so important as making errors in appraising risk. By treatment in this context, we are only referring to interventions that are designed to reduce the likelihood that a person will engage in sexual activities with children. Treatment can, of course, be required for a wide variety of other problems but only those problems that are related to the probability with which a person victimizes children are at issue here. There are a variety of assessment techniques that can be used to choose a particular treatment modality or to monitor the acquisition of therapeutic goals. Examples of these include hormonal assays, psychological tests of suitability for verbal interventions, behavioral tests of heterosocial skillfulness, tests of sexual knowledge, and knowledge of specific program content, such as relapse prevention techniques. These assessments are most commonly used when a client has already been accepted into a treatment program, and will not be commented on further here. Typically, clinical evaluations of child molesters include a clinical interview and psychological testing; more rarely they include phallometric assessment and, in some cases, polygraphy. Below, we briefly describe these methods of assessment and comment on the uses to which they may be best put. The information gathered from these various sources are integrated in order to provide a tentative explanation of why the offence occurred; that is, the clinician seeks to develop a theory to explain the specific offence by specifying the situational antecedents and personal characteristics of the offender that led to the sexual victimization. Regardless of whether the purpose of the assessment involves treatment or risk issues, clinicians are well advised to obtain as much detail about the alleged molestation(s) as possible. This can be accomplished by requesting police reports and victim statements and, where possible, directly interviewing the victim(s). These are the data to be explained by the clinician’s theory. In addition, criminal history remains one of the best predictors of future sexual recidivism among child molesters. Extrafamilial child molesters are more likely to recidivate than strictly intrafamilial child molesters, and men who choose boy victims are more likely to recidivate than men who choose girl victims; the probability of recidivism rises with each additional sexual offence (Quinsey, 1986).
THE INTERVIEW Interviews of the offender are similar to other clinical interviews except the validity of the information provided by the offender is more likely to be suspect. Frequently, the alleged child molesters distort information, falsely deny that the alleged offence occurred, and report difficulty in recalling events surrounding the offence. A noted by Earls (199 l), “There is surprisingly little empirical research concerning the reliability and validity of the information obtained in a clinical interview.” In an effort to obtain valid self-report data on individuals who committed sexual offences
Assessing child molesters
171
against children, Abel, Becker, Mittleman, Cunningham-Rathner, Rouleau, and Murphy (1987) interviewed sexual offenders under a certificate of confidentiality. Federal law goveming these certificates insured that no city, county, state, or federal agency could compel investigators to reveal the identity of the subjects participating in the project. This arrangement gave some level of confidence to the participants that the information they disclosed would not be reported to criminal justice authorities. A total of 377 men were seen who had engaged in sexual activity with children not related to them. The mean number of victims for those men who had molested young girls was 19.8 and the mean number of victims for those men who had molested young boys was 150.2. Among the 203 men who had engaged in intrafamilial sexual abuse, those who were sexually involved with female relatives had an average of 1.8. victims, and for those who had involved themselves with male relatives, the mean number of victims was 1.7. If those interviews had been conducted within a forensic context without a certificate of confidentiality, it is highly doubtful that the child molesters would have been quite so forthcoming in divulging the extent of their sexual activity with children. The clinical interview should obtain demographic information, a full history of both paraphilic and non-paraphilic behavior, history of sexual and physical abuse, social history, alcohol and substance abuse history, medical history, criminal justice history, and employment history (Becker & Kaplan, 1990). It is particularly important that the suspected molesters be questioned about paraphilic behavior as well as fantasies. Because individuals who engage in norm-violating sexual behavior are frequently reluctant to disclose sexual behaviors, it is critical that the clinician ask the interviewee about every type of paraphilia, including bestiality, coprophilia, exhibitionism, fetishism, frottage, pedophilia, public masturbation, rape, sadism, masochism, necrophilia, obscene phone calls, obscene letters, transvestism, and urolagnia. The client should be questioned as to whether he has fantasized about these categories of behavior and whether or not he has ever acted on these fantasies. It is important to note that individuals who engage in norm-violating sexual behaviors are a heterogeneous group (Quinsey, 1986). For some, the deviant sexual behavior may be part of a paraphilic interest pattern, for others the behavior may represent an overall pattern of antisocial behavior, and, for yet others, both may be involved. It is important in this connection, that Hare’s Psychopathy Checklist is a strong predictor of future sex offences among both rapists and child molesters (Quinsey, Rice, & Harris, submitted). In a small percentage of cases, individuals with major psychiatric disorders, such as psychosis may commit sexual offenses and, more frequently, persons with significant problems of alcohol abuse may sexually victimize children. It is, therefore, important to assess the presence of other clinical problems during the interview because these problems may contribute to the commission of the sexual misbehaviors.
PSYCHOLOGICAL
TESTING
Psychological tests can be divided into three categories: tests of sexual interest and attitudes, objective personality tests, and projective tests. Tests of sexual interest include such assessments as the Sexual Interest Cardsort. Barlow, Leitenberg, and Agras, ( 1969) devised a cardsort technique in which various sexual scenes were typed on cards. The individual sorts the cards into four categories according to the amount of sexual interest they elicit (Tollison & Adams, 1979). These cardsort tasks are good measures of sexual interest with cooperative subjects. Clinicians have found that sex offenders frequently have attitudes or beliefs that support their inappropriate sexual behavior. These beliefs often involve cognitive distortions. A cognition scale has been developed by Abel and his colleagues ( 1984) to measure these distortions and has been found to distinguish child molesters from non-child molesters (Able, Gore,
172
J. Becker and V. Quinsey
Holland, Camp, Becker, & Rathner, 1989). These cognitive distortions are appropriate targets for intervention as they may serve to maintain sexual behaviors toward children, although their prognostic significance remains to be established. The MMPI (Minnesota Multiphasic Personality Inventory) has been used extensively in assessment of sexual offenders. While the MMPI can be useful in clinical assessment for the purposes for which it was designed, its discriminant validity in the assessment of child molester is poor (e.g., Quinsey, Arnold, & Pruesse, 1980). Levin and Stava (1987) found methodological problems in most of the 36 MMPI studies that they reviewed and concluded that, “in general, negative or inconsistent findings outweigh those of a positive nature.” Projective tests such as Rorschach, Thematic Apperception Test, and Draw-A-Person Test, have frequently been used to assess child molesters but there is little evidence concerning the reliability or validity of these measures; the studies that have been done in this area are very methodologically weak (Quinsey, 1986). We do not recommend the use of the MMPI or projective tests for appraising the risk that child molesters present to the community or in determining treatment targets related to sexual reoffending. The results of psychological testing, like the results of interviewing, must be interpreted with caution. Dissimulation is a common problem in the assessment of child molesters, and only some psychological tests, such as the MMPI, contain scales that attempt to address faking. There is no “profile of the child molester” that can reliably be identified with standard psychological tests. A comprehensive review of questionnaires and psychological tests that have been used to assess offenders can be found in Hanson, Cox, and Woszcsyna (199 1).
PHALLOMETRIC
ASSESSMENT
In a phallometric assessment, changes in penile tumescence occasioned by a variety of sexual and nonsexual stimuli are measured with a penile plethysmograph. The relative magnitude of these responses reflect the amount of sexual interest elicited by the types of stimuli shown. Phallometric assessments of sexual age and gender preferences have excellent discriminant validity with extrafamilial child molesters and the classification accuracy of some stimulus sets and procedures has been well documented (Freund & Blanchard, 1989). Exclusively incestuous offenders are less likely to show inappropriate sexual age preferences in phallometric assessments than extrafamilial child molesters (Barbaree & Marshall, 1989; Quinsey, Chaplin, & Carrigan, 1979). Inappropriate sexual age preferences measured in phallometric assessments predict subsequent new offences against children (Barbaree & Marshall, 1988; Quinsey, Rice, & Harris, submitted; Rice, Quinsey, & Harris, 199 1). These assessments attempt to measure the degree of sexual preference among stimulus categories; they do not detect whether someone has committed a specific sexual offence. Although phallometric assessments are related to the risk of reoffending an offender presents, they are far from infallible. In particular, some child molesters can intentionally influence their responses in order to appear to have normal sexual preferences (e.g., Freund, Watson, & Rienzo, 1988); because of this possibility, deviant phallometric profiles are more easily interpreted than normal profiles. Phallometric data are useful chiefly in determining whether an offender selected a child victim because of a specific sexual interest in children, or as a matter of convenience or opportunity. Such a determination has obvious therapeutic and supervisory implications. Phallometric assessments are, however, somewhat more technical than regular clinical evaluations and require specialized apparatus. Laws and Osbom (1983) have provided a useful description of laboratory phallometric procedures; some of the technical issues in performing these assessments are commented on by Quinsey and Laws (1990).
Assessing child molesters
173
LIE DETECTION Because child molesters’ potential dissimulation often clouds their assessment, it is understandable that clinicians and supervisory authorities have sought techniques that reveal when offenders are lying about the commission of previous sexual offences and their true sexual interests. Unfortunately, because there are currently no reports of controlled studies using polygraphy in the literature, the utility of lie detection methodology in this area is impossible to assess. There are, however, a priori considerations that should give practitioners pause before adopting these techniques for regular use (Lalumiere & Quinsey, 199 1). These considerations include the unknown effects of repeated testing and the high possibility of false positive errors in some circumstances. The physiological responses measured by the polygraph are not specific correlates of lying.
CONCLUSIONS Referrals of child molesters for clinical evaluations are sometimes inappropriate, as, for example, when the purpose of the evaluation is to determine whether the referred person actually committed a sexual offence. Other referrals may involve questions for which the referring agent desires definite answers but for which only probabilistic answers are possible. Assessments of dangerousness, for example, can only provide estimates of risk and the precision with which it is possible to make these estimations varies widely according to the details of the case and the amount of information available to the clinician. Although there are limitations to any of the assessment techniques described above and client faking remains a pervasive problem, some of these techniques are better suited to assessments of child molesters than others. Those techniques that can differentiate child molesters from other persons, that are related to the planning of treatment programs, and that predict the commission of future sexual offending are, of course, the most useful. Sex offence history and phallometric assessments are useful in planning treatment and supervision programs and in assessing risk. The Psychopathy Checklist is also a strong predictor of sexual recidivism but its usefulness in treatment planning is equivocal. Measures of attitudes and beliefs that support child molestation are suitable for designing treatment programs but their prognostic significance remains to be established. More general measures of personality and psychopathy have not been established as predictors of recidivism but can be used to address collateral clinical issues. Clinicians integrate information from a variety of sources to develop a theory designed to explain the offence history of there child molester who is being assessed. Such a theory can be used to select appropriate interventions and provide a focus for supervision. Clinical evaluations can play an important and useful role in making a variety of dispositional decisions concerning child molesters, providing they are done carefully using relevant techniques and their limitations are clearly understood.
REFERENCES Abel, G. G., Becker, J. V., Mittelman, M. Cunninham-Rathner, J.,Rouleau, J., & Murphy, W. (1987). Self-reported sex crimes of non-incarcerated paraphiliacs. Journal of Interpersonal Violence, 2, 3-25. Abel, G., Becker, J., Cunningham-Rathner, J.,Rouleau, J., Kaplan, M., & Reich. J. (1984). The treatment of child molesters: A manual. Unpublished manuscript. Abel, G. G., Gore, D., Holland, C., Camp, N., Becker., & Rathner, J. (I 989) The measurement of cognitive distortions of child molesters. Annals of Sex Research. 2, 135-l 53.
174
J. Becker and V. Quinsey
Barbaree, H. E., & Marshall, W. L. (1988). Deviant sexual arousal, offense history, and demographic variables as predictors of reoffense among child molesters. Behavioral Sciences and the Law, 6, 267-280. Barbaree, H. E., & Marshall, W. L. (1989). Erectile responses among heterosexual child molesters, father-daughter incest offenders and matched nonoffenders: Five distinct age-preference profiles. Canadian Journal of Behavioural Science, 21, 70-82. Barlow, D., Leitenberg, H., & Agras, W. (1969). The experimental control of sexual deviations through manipulation of the noxious scene in court sensitization. Journal of Abnormal Psychology, 74, 596-60 I. Becker, J. V., & Kaplan, M. S. (1990). Assessment of the adult sex offender. In P. McReynolds, J. Rosen, & G. Chelune (Eds.), Advances in psychological assessment. (pp. 26 l-283). New York: Plenum. Earls, C. (199 I). Clinical issues in the psychological assessment of child molesters. In W. O’Donohue & J. Greer (Eds.), The sexual abuse of children; Theory, research and therapy. New York: Erlbaum. Freund, K., & Blanchard, R. (1989). Phallometric diagnosis of pedophlia. Journal of Consulting and Clinical Psychology, 57, 100-105. Freund, K., Watson, R., & Rienzo, D. (1988). Signs of feigning in the phallometric test. Behaviour Research and Therapy, 26, 105-I 12. Hanson, R. K., Cox, B., & Woszcsyna, C. (199 1). Sexuality, personality and attitude questionnairesfor sex offenders: A review. Cat. No. JS4- l/ 199 1- 13. Ottawa: Supply and Services Canada. Lalumiere, M. L., & Quinsey, V. L. (199 1) Polygraph testing ofchild molesters: Are we ready? Violence Update, l( 1l), 3, 9, 11; (12), 6-7. Laws, D. R., & Osborn, C. A. (1983).How to build and operate a behavioral laboratory to evaluate and treat sexual deviance. In J. G. Greer, & I. R. Stuart (Eds.). The sexual aggressor: Current perspectives on treatment. (pp.292335). New York Van Nostrand Reinhold. Lewin, S., & Stava, L. (1987). Personality characteristics of re-offenders: A review. Archives of Sexual Behavior, 16, 57-59. Quinsey, V. L. (1986), Men who have sex with children. In D. N. Weisstub (Ed.), Law and mental health: Znternationalperspectives, 2, (pp. 140-172). New York: Pergamon. Quinsey, V. L., Arnold, L. S., & Pruesse, M. G., (1980). MMPI profiles of men referred for a pretrial psychiatric assessment as a function of offense type. Journal of Clinical Psychology, 36,4 IO-4 17. Quinsey, V. L., Chaplin, T. C., & Canigan, W. F. (1979). Sexual preferences among incestuous and non-incestuous child molesters. Behavior Therapy, 10, 562-565. Quinsey, V. L., & Laws, D. R. (1990). Validity ofphysiological measures of pedophilic sexual arousal in a sex offender population: A critique of Hall, Protor, and Nelson. Journal of Consulting and Clinical Psychology, 58, 886-888. Quinsey, V. I., &Walker, W. D. (in press). Dealing with dangerousness: Community risk management strategies with violent offenders. In R. Peters, R. J. McMahon, & V. L. Quinsey (Eds.). Aggression and violence throughout the lifespan. Quinsev. V. L., Rice, M. E., & Harris, G. T. (submitted). Psychopathy, sexual deviance, and recidivism among sex - offenders released from a maximum security psychiatric institution. Rice, M. E., Quinsey, V. L., & Harris, G. T. (1991). Sexual recidivism among child molesters released from a maximum security psychiatric institution. Journal of Consulting and Clinical Psychology, 59, 38 l-386. Tollison, C. D., & Adams, H. E. (1979). Sexual disorders: Treatment, theory, and research. New York: Gardner.
Resum&-Les evaluations cliniques peuvent jouer un role utile dans la prise dune s&e de dispositions a l’egard des agresseurs d’enfants a condition qu’elles soient faites prudemment en utilisant des techniques adequates et que leurs limites soient clairement comprises. Les instruments et les techniques d’tvaluation les plus utiles sont celles qui permettent de differentier les agresseurs d’enfants des autres personnes, celles qui sont associeees au programme therapeutique et celles qui predisent la perpetuation future dune agression sexuelle. L’histoire de l’agression sexuelle et les evaluations phallomttriques sont utiles pour la planification des programmes de traitement, de supervision et pour l’bvaluation du risque. La “Psychopathy Checklist” forme un bon predicteur de recidives sexuelles mais son utilite dans la planification d’un traitement est Bquivoque. Les mesures des comportements et des croyances qui sont a la base de l’agression sexuelle des enfants conviennent pour l’tlaboration du traitement, mais leur valeur pronostique reste a etablir. Des evaluations de personnalitt ou de psychopathologic plus getterales, predictives de recidives n’ont pas ttt etablies, mais peuvent Ctre utilisees a des fins cliniques paralltlles. Resumen-Las evaluaciones clinicas pueden jugar un rol muy 6til al hater una variedad de decisiones en relation a 10s que molestan a 10s nifios siempre que Sean hechas con cuidado utilizando ttcnicas apropiadas comprendiendo claramente sus limitaciones. Las mas utiles son 10s instrumentos de evaluation y las tecnicas que pueden diferenciar 10s que molestan a 10s nifios de otras personas, las que esmn relacionadas al plan de1 tratamiento, y las que predicen que se.cometan actos futures sexualmente ofensivos. El historial de ofensas y las evaluaciones falometricas son itiles para planificar el tratamiento y para supervisi6n de programas asi coma para evaluar el riesgo. El Inventario Psiconkico es tambien un fuerte predictor de recidivism0 sexual pero su utilidad en la planificaci6n de1 tratamiento es equivoca. Las medias de attitudes y creencias que apoyan 10s actos de molestar sexualmente a 10s mifios son utiles para d&tar programas de tratamiento pero su importancia en el pronbstico no ha sido estabecida. Medidas rna.s generales de personalidad y psicopatologia no han sido establecidas coma predictoras de recidivism0 pero pueden utilizarse para enfocar aspectos clinicos colaterales.