Despite social and legal consequences, child molesters engage in sexual activities with children. Speculation as to why molesters choose children as their sexual partners was examined from seven perspectives: psychodynamic, attachment, three mental functions, psychopathic, physiological, learning, and third force theories. The potential causal variables suggested by these theories were coded using information from the files of 200 male adolescent and adult child molesters. Variables were analyzed using Ward's minimum variance clustering technique to develop a psychological classification system based on child molesters' motivations. Five-cluster solutions were discovered when using either the adolescent or adult child molesters' samples. Clusters from the adolescent sample were labeled: Stable (47%), Pedophilic (13%), Disturbed Conduct (15%), Psychopathic Tendencies (10%), and Dysfunctional Families (15%). The clusters from the adult sample were remarkably similar, yet there were differences. These clusters were labeled Mentally Unstable (25%), Mentally Stable (40%), Brain Damage/Head Injuries (14%), Psychopathic Tendencies (10%), and Dysfunctional Families (11%). Using a split-half test, four of the five clusters for both samples appear to be stable. Only the clusters based on psychiatric diagnoses of pedophilia were unstable when these diagnoses were eliminated from the analysis. Although validation of the five-cluster solution using contrasting variables for the adolescent sample was weak, the five-cluster solution for the adult sample was convincingly validated.

Despite social and legal consequences, child molesters engage in sexual activities with children. Speculation as to why molesters choose children as their sexual partners was examined from seven perspectives: psychodynamic, attachment, three mental functions, psychopathic, physiological, learning, and third force theories. The potential causal variables suggested by these theories were coded using information from the files of 200 male adolescent and adult child molesters. Variables were analyzed using Ward's minimum variance clustering technique to develop a psychological classification system based on child molesters' motivations. Five-cluster solutions were discovered when using either the adolescent or adult child molesters' samples. Clusters from the adolescent sample were labeled: Stable (47%), Pedophilic (13%), Disturbed Conduct (15%), Psychopathic Tendencies (10%), and Dysfunctional Families (15%). The clusters from the adult sample were remarkably similar, yet there were differences. These clusters were labeled Mentally Unstable (25%), Mentally Stable (40%), Brain Damage/Head Injuries (14%), Psychopathic Tendencies (10%), and Dysfunctional Families (11%). Using a split-half test, four of the five clusters for both samples appear to be stable. Only the clusters based on psychiatric diagnoses of pedophilia were unstable when these diagnoses were eliminated from the analysis. Although validation of the five-cluster solution using contrasting variables for the adolescent sample was weak, the five-cluster solution for the adult sample was convincingly validated. / Arts, Faculty of / Psychology, Department of / Graduate

This study addresses the problem of the assessment of child molesters. The research questions in this study specifically address the assessment of child molester personality and its relationship to demographic, personal, criminal offense, and treatment progress variables.The purpose of the study was to replicate cluster analyses done with the Minnesota Multiphasic Personality Inventory (MMPI) in child molester populations using the revised and re-nonmed Minnesota Multiphasic Personality Inventory-2 (MMPI-2), and also, to identify other variables that are associated with particular profile patterns in child molesters.This study examined the archival records of 91 adult male child molesters who were involved in a communitybased sexual abuse treatment program between 1990 and 1993. Information analyzed included the MMPI-2, demographic, personal, and criminal offense data obtained from the child molesters' treatment files, and therapist ratings on the Treatment Progress Rating Scale which is an 11-point scale based on behaviors that the child molester reveals in treatment.The data from the MMPI-2 scales were cluster analyzed, yielding four homogeneous profile subgroups. Three of these four subgroups closely resembled those found in earlier MMPI research with child molesters. Discriminant analysis and chi square analyses were used to examine variables that differentiated the profile groups of child molesters derived in the cluster analysis. Results indicated that the derived groups significantly differed in terms of their reported history of having been sexually abused as a child. Analysis of variance (ANOVA) using the Scheffe procedure revealed that child molesters with within normal limits MMPI-2 profiles were less likely to report a history of childhood sexual abuse than were child molesters with more elevated profiles. ANOVA with repeated measures was used to analyze child molester progress in treatment as a function of MMPI-2 profile group. Although the analysis revealed that significant treatment progress was made for the overall sample, the groups did not significantly differ in terms of their treatment progress. / Department of Counseling Psychology and Guidance Services

Child sexual abusers have typically been studied as a homogeneous group, comparing them with non-child sexual abusers. The present study was designed to examine the heterogeneity of child sexual abusers on a measure of psychopathology, the Minnesota Multiphasic Personality Inventory-2 (MMPI-2), the revised version of the MMPI. MMPI-2 profiles of 106 child sexual abusers in treatment for their offenses were cluster analyzed. Three clusters were obtained from the data, suggesting that child sexual abusers are heterogeneous as measured by the MMPI-2. The clusters were F-8-7-6 (n=15), elevated 4 scale (n=29), and within normal limits (n=62). A discriminant function generated by cluster membership was able to accurately classify 98.11% of the sample.The three clusters were replicated in an independent sample (n=50) drawn from the same population. The discriminant function generated on the derivation sample was applied to the validation sample, correctly classifying 80.00% of the subjects.The mean MMPI-2 profiles for the clusters in this study did not replicate mean profiles of clusters derived from similar research using the MMPI, indicating that the two instruments may not be equivalent for use in research using multivariate statistical techniques.The three clusters were tested against amount of self-reported sexual psychopathology as measured by the Multiphasic Sex Inventory (MSI). Discriminant function analysis indicated that cluster membership did predict degree of sexual psychopathology. In essence, the more elevated the mean MMPI-2 cluster profile, the more sexual psychopathology was reported. The clusters did not differ with regard to progress in treatment (though there were definite trends toward significant group differences in the data), nor did they differ with regard to victim and offender demographics. They did, however, differ with regard to level of intelligence, with the F-8-7-6 group being significantly lower in intelligence than the other two groups.The results of this study indicate that child sexual abusers are heterogeneous with regard to their MMPI-2 profiles, and that this heterogeneity can be used to form homogeneous groups. Additionally, homogeneous groups of child sexual abusers can be expected to differ on the basis of general psychopathology and sexual psychopathology. / Department of Counseling Psychology and Guidance Services

Afrikaans text / Kindermolestering het die afgelope vyftien jaar 'n toenemende probleem in Suid-Afrika geword. Baie aandag is aan die slagoffer gegee, tenvyl die kindermolesteerder as kardinale deel van die probleem heeltemal verwaarloos is. Laasgenoemde maak hierdie navorsingsondersoek ten opsigte van die kindermolesteerder dringend noodsaaklik. en prosesse in die lewe van die kindermolesteerder, waaruit 'n ryk en indiepte beskrywing van die kindennolesteerder gevolg het. Die tweede fase is kwantitatief benader en navorser het gebruik gemaak van 'n onderhoudskedule wat aan 'n meer omvangryke groep kindermolesteerders en verkragters voorgele is om hulle reaksie op die data syfennatig te bepaal. Verkragters is gekies omdat hulle en die molesteerders 'n seksuele misdryf as gemene deler het. Navorser wou met laasgenoemde die bruikbaarheid van die skedule bepaal en verdere ondersteuning vir die ondersoek verkry. Hierdie data is rekenaarmatig verwerk en skematies uitgebeeld. Die navorsingsondersoek het aan die lig gebring dat kindermolesteerders uit disfunksionele gesinne kom, waarin hulle persoonlikheidseienskappe en gedragspatrone aangeleer het, wat kan lei tot kindermolestering. Laasgenoemde word deur bevindings ten opsigte van Hierdie navorsingsondersoek het 'n bydrae tot die wetenskaplike kennis gemaak deurdat die bruikbaarheid van pluralisme in die ondersoek uitgewys is; • ooreenkomste en verskille tussen kindermolesteerders en verkragters geYdentifiseer is; • die prominente rol van pornografie in kindermolestering aan die lig gekom het; • 'n profiel van die kindermolesteerder saamgestel is; die gedrag van die molesteerder teoreties verklaar is; • 'n behandelingstruktuur vir die oortreder saamgestel is. Aanbevelings vir die toekomstige bantering van die kindermolesteerder en verdere navorsing in die verband is gemaak. Die doel van hierdie navorsingsondersoek is om 'n omvattende beeld van die kindermolesteerder as oortreder te verkry ten einde die probleem van kindermolestering aan te spreek. Hierdie ondersoek is pluralisties benader en het in opeenvolgende fases plaasgevind. Fase een is kwalitatief benader en navorser het semi-gestruktureerde onderhoude gevoer met tien kindermolesteerders aan die hand van 'n onderhoudsgids. Aandag is gegee aan die gebeure / Over the past fifteen years child molestation has become a growing problem in South Africa. Much attention has been given to the victim while the child molester as a cardinal part ofthe problem has been completely neglected, which made this study of the child molester crucial. The purpose of this study is to obtain a comprehensive picture of the child molester as offender in order to tackle the problem of child molestation. The research was conducted in consecutive phases from a pluralistic approach. The first phase was qualitative and the researcher conducted semi$tructured interviews with ten child molesters, based on an interview guide. Attention was given to the events and processes in the lives of the child molesters. From this a rich and detailed description of the child molesters was obtained. The second phase was quantitative and the researcher used an interview schedule which was presented to a more comprehensive group of child molesters and rapists to determine their reaction to the data statistically. Rapists were chosen because they and the child molesters had a sexual offence as common denominator. By using rapists the researcher wanted to detennine the usefulness of the schedule and gain further support for the investigation. The data were processed by computer and depicted schematically. The research revealed that child molesters come from dysfunctional families, in which they acquired personality traits and behavioural patterns which can lead to child molestation. The latter is supported by findings on rapists. child molesters. From this a rich and detailed description of the child molesters was obtained. The second phase was quantitative and the researcher used an interview schedule which was presented to a more comprehensive group of child molesters and rapists to determine their reaction to the data statistically. Rapists were chosen because they and the child molesters had a sexual offence as common denominator. By using rapists the researcher wanted to detennine the usefulness of the schedule and gain further support for the investigation. The data were processed by computer and depicted schematically. The research revealed that child molesters come from dysfunctional families, in which they acquired personality traits and behavioural patterns which can lead to child molestation. The latter is supported by findings on rapists. A profile of the child molester was compiled and Bandura's social learning theory used to explain his behaviour. To deal with the problem of child molestation, a treatment schedule was developed for the child molester. This study has contributed to scientific knowledge by • showing the usefulness of pluralism in the investigation • identifying similarities and differences between child molesters and rapists • revealing the prominent role of pornography in child molestation • compiling a profile of the child molester • explaining the behaviour of the molester theoretically • developing a treatment structure for the offender Recommendations are made for the future treatment of the child molester and further research. / Criminology and Security Studies / D.Phil. (Kriminologie)

This study utilized a qualitative/phenomenological research methodology to study the development of child molestation. Five volunteer male juvenile perpetrators of child molestation and their consenting family members participated in three one-on-one in-depth interviews. The juveniles were referred by juvenile justice departments in Texas. The investigation was pursued along three paths: (1) what factors in the perpetrators' life experiences may have influenced the formation of molesting thought and behavior patterns? (2) how did the perpetrators construe social relationships? (3) in developing sexual preference, what process did the perpetrators utilize to achieve the degree of sexual arousal needed to motivate the act of molestation? During analysis, 16 categories and 9 subcategories were developed from the data; these were evaluated in order to address the lines of inquiry listed above. Factors contributing toward deviancy were identified: these included general environmental factors and relationship issues with both parents and peers. The data also elucidated the perpetrators' characteristic ways of viewing themselves and other people, as well as the strategies that they relied upon to cope with their lives. These same strategies were subsequently used in their molesting. The results revealed that the answers to the investigational questions posed above were synthesized by the individual across the lifespan, with origins in childhood when the concepts of self, others, social convention and interpersonal relationships were forming. It is likely that the typical interest in sexual expression that occurs at the onset of puberty was a catalyst that facilitated molesting behavior. The lack of ability to interact successfully with age mates appeared to influence the perpetrators' selection of children as targets. Recommendations were presented for consideration by the mental health and criminal justice communities based on the findings. Also, preventative measures for the public sector were offered.

Afrikaans text / Kindermolestering het die afgelope vyftien jaar 'n toenemende probleem in Suid-Afrika geword. Baie aandag is aan die slagoffer gegee, tenvyl die kindermolesteerder as kardinale deel van die probleem heeltemal verwaarloos is. Laasgenoemde maak hierdie navorsingsondersoek ten opsigte van die kindermolesteerder dringend noodsaaklik. en prosesse in die lewe van die kindermolesteerder, waaruit 'n ryk en indiepte beskrywing van die kindennolesteerder gevolg het. Die tweede fase is kwantitatief benader en navorser het gebruik gemaak van 'n onderhoudskedule wat aan 'n meer omvangryke groep kindermolesteerders en verkragters voorgele is om hulle reaksie op die data syfennatig te bepaal. Verkragters is gekies omdat hulle en die molesteerders 'n seksuele misdryf as gemene deler het. Navorser wou met laasgenoemde die bruikbaarheid van die skedule bepaal en verdere ondersteuning vir die ondersoek verkry. Hierdie data is rekenaarmatig verwerk en skematies uitgebeeld. Die navorsingsondersoek het aan die lig gebring dat kindermolesteerders uit disfunksionele gesinne kom, waarin hulle persoonlikheidseienskappe en gedragspatrone aangeleer het, wat kan lei tot kindermolestering. Laasgenoemde word deur bevindings ten opsigte van Hierdie navorsingsondersoek het 'n bydrae tot die wetenskaplike kennis gemaak deurdat die bruikbaarheid van pluralisme in die ondersoek uitgewys is; • ooreenkomste en verskille tussen kindermolesteerders en verkragters geYdentifiseer is; • die prominente rol van pornografie in kindermolestering aan die lig gekom het; • 'n profiel van die kindermolesteerder saamgestel is; die gedrag van die molesteerder teoreties verklaar is; • 'n behandelingstruktuur vir die oortreder saamgestel is. Aanbevelings vir die toekomstige bantering van die kindermolesteerder en verdere navorsing in die verband is gemaak. Die doel van hierdie navorsingsondersoek is om 'n omvattende beeld van die kindermolesteerder as oortreder te verkry ten einde die probleem van kindermolestering aan te spreek. Hierdie ondersoek is pluralisties benader en het in opeenvolgende fases plaasgevind. Fase een is kwalitatief benader en navorser het semi-gestruktureerde onderhoude gevoer met tien kindermolesteerders aan die hand van 'n onderhoudsgids. Aandag is gegee aan die gebeure / Over the past fifteen years child molestation has become a growing problem in South Africa. Much attention has been given to the victim while the child molester as a cardinal part ofthe problem has been completely neglected, which made this study of the child molester crucial. The purpose of this study is to obtain a comprehensive picture of the child molester as offender in order to tackle the problem of child molestation. The research was conducted in consecutive phases from a pluralistic approach. The first phase was qualitative and the researcher conducted semi$tructured interviews with ten child molesters, based on an interview guide. Attention was given to the events and processes in the lives of the child molesters. From this a rich and detailed description of the child molesters was obtained. The second phase was quantitative and the researcher used an interview schedule which was presented to a more comprehensive group of child molesters and rapists to determine their reaction to the data statistically. Rapists were chosen because they and the child molesters had a sexual offence as common denominator. By using rapists the researcher wanted to detennine the usefulness of the schedule and gain further support for the investigation. The data were processed by computer and depicted schematically. The research revealed that child molesters come from dysfunctional families, in which they acquired personality traits and behavioural patterns which can lead to child molestation. The latter is supported by findings on rapists. child molesters. From this a rich and detailed description of the child molesters was obtained. The second phase was quantitative and the researcher used an interview schedule which was presented to a more comprehensive group of child molesters and rapists to determine their reaction to the data statistically. Rapists were chosen because they and the child molesters had a sexual offence as common denominator. By using rapists the researcher wanted to detennine the usefulness of the schedule and gain further support for the investigation. The data were processed by computer and depicted schematically. The research revealed that child molesters come from dysfunctional families, in which they acquired personality traits and behavioural patterns which can lead to child molestation. The latter is supported by findings on rapists. A profile of the child molester was compiled and Bandura's social learning theory used to explain his behaviour. To deal with the problem of child molestation, a treatment schedule was developed for the child molester. This study has contributed to scientific knowledge by • showing the usefulness of pluralism in the investigation • identifying similarities and differences between child molesters and rapists • revealing the prominent role of pornography in child molestation • compiling a profile of the child molester • explaining the behaviour of the molester theoretically • developing a treatment structure for the offender Recommendations are made for the future treatment of the child molester and further research. / Criminology and Security Studies / D.Phil. (Kriminologie)

Thesis (MSc (Psychology))--University of Stellenbosch, 2005. / There is a growing need for research to facilitate a better understanding of paedophilia. This study aims to make a contribution in this regard by providing an analysis of current discourses in paedophilia research as well as a phenomenological exploration of how the male paedophile constructs his close relationships with children. Analysis reveals how the dominant positivist approach to paedophilia research embeds and perpetuates moral and often prejudicial positions that in turn influence the validity of empirical findings and theory, and tend to marginalise contradictory evidence. In an attempt to circumvent these obstacles, it is argued that the psychological need to form close relationships is a universal one.

Path models assessed different models of influential order for parental bonding; adult romantic attachment; views of self, world/others, and the future; the fear of intimacy; and cognitive distortions among child molesters and non-offending controls. Child molesters receiving sex offender treatment reported more problematic parental bonding; insecure adult romantic attachment; negative views of self, world/others, and the future; a greater fear of intimacy, and more cognitive distortions regarding adult-child sex. The predicted path models were not established as the models did not adequately fit the data. However, post hoc logistic regressions indicated that Maternal Optimal Bonding, Preoccupied attachment, and cognitive distortions regarding adult-child sex significantly predicted child molester status. Overall, the findings provide support for a multi-factorial model of child molestation derived from attachment theory. Limitations of the study and areas for future research are also discussed.

This study compared one group of pedophiles, two groups of incestuous fathers, and two groups of incestuous stepfathers on their individual MMPI scale scores and self-reported demographic data. Past research on sex offenders has shown that patterns of variables are better indicators of offender typology than single variables. A nonrandom sample of male abusers (N=177) was drawn from two separate sources, a clinic that screens offenders for treatment disposition and an unrelated outpatient treatment clinic in the Portland, Oregon area. Scores from the 13 MMPI scales as well as data from 17 demographic variables were drawn from clinic interviews and intake records of the offenders. The demographic data included age of the offender, educational level, age at first marriage, number of marriages, number of juvenile and adult arrests, number of jobs, number of past inpatient treatment periods, outpatient treatment periods, and incarcerations, alcohol consumption level, recreational drug usage, adoption status, and incidence of foster care, physical, and sexual abuse. Of these variables, several are susceptible to error in self-report, especially alcohol consumption level, which may be subject to the usefulness of a particular report to the offender. Hence, one statistical analysis was made eliminating alcohol consumption level as a variable.