\. 1L .. 1 'x‘a ( .. , ' "fi-afi‘r'fir =3?“ -r‘55‘.11'{f m: “1,9,; .gfi.‘ r31; 532;; v; ‘ ‘ e r O‘ wvx' - 031'! ‘ - g}: “ .ur "‘ '3" . a, %“*i“' .. .L 4‘ '3' ’3"? J ‘ 1:. 0,.4 1": .I 1. v. JQT‘ I % "G. gn'w l‘ 'J . f 's ’r' "W?“ A15 ‘ ., A: i ' 52'? v ' L . ‘ “*3?” .z: uaJ ‘_ 'V'fi'uf‘ic. "4”" .. “3-. £6 ~ lb- .v" .Jfizfi". '3. r." 27‘3é3 - ‘ my: ' "‘33???“ ‘ - aw an 1‘55 I r 54"“ 1t}... -- :fifk‘gt. $51: . ' ’i‘ifiiiéflz" m 5-3:;- ‘33 Ly 3" mm. ' ' ,5 . fitiéywi‘a “v '§&'%%$‘g% mama ELY-7‘3 13; «If: -‘ ,V mtg illllllllllllIIIHIIHHHINIlllllllllllllllllllilllillllllli 91293 01402 7894 This is to certify that the dissertation entitled SOCIETAL SYSTEM INTERVENTION TRAUMA TO CHILD SEXUAL ABUSE VICTIMS FOLLOWING DISCLOSURE presented by James A. Henry has been accepted towards fulfillment of the requirements for Ph.D. degree in Social Science 72%wa Major professor Date 10-14-94 MSU is an Affirmative Action/Er] ual Opportunity Institun'on 0» 12771 LIBRARY M'Chigan State University PLACE ll RETURN BOX to moon this checkout from your record. TO AVOID FINES Mum on or boron duo duo. DATE DUE DATE DUE DATE DUE C:L__J:J ICT—T _J[::L_J I II] LAG TI | | MSU IoAn Affirmative Action/Equal Opporumtty Instituton W m1 .— . _,... , E‘._._..___.__._‘_._._- SOCIETAL SYSTEM INTERVENTION TRAUMA TO CHILD SEXUAL ABUSE VICTIMS FOLLOWING DISCLOSURE BY James A. Henry A DISSERATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR or SOCIAL SCIENCE Department of Social Work 1994 ABSTRACT SOCIETAL SYSTEM INTERVENTION TRAUMA TO CHILD SEXUAL ABUSE VICTIMS FOLLOWING DISCLOSURE BY James A. Henry Over the last ten years society has finally recognized the pervasiveness of child sexual abuse and its devastating impact on children and their families. Societal systems mandated to intervene in child sexual abuse have been forced to respond to a subsequent tremendous increase in reported cases. The societal system response however, has too often been directed by past antiquated practices and theoretical speculations. Few studies on the impact of societal system intervention on sexually abused children have been initiated. This study was an attempt to discover if further traumatization occurs to sexually abused children through societal system interventions. The Traumagenic Model, developed by David Finkelhor, which explains the dynamics of trauma in child sexual abuse, was used as the theoretical framework to understand how societal system interventions can produce or reinforce the previous trauma. The primary hypothesis was; societal system interventions by mandated professionals i.e., child protective services, law enforcement, juvenile courts, and the criminal courts, exacerbate the experiences of betrayal, powerlessness, and stigmatization of sexual abuse and result in further traumatization to child victims. A secondary hypothesis was; a county with more child supportive societal system interventions is likely to have less traumatized children than counties with a more adult reactive system intervention. Ninety sexually abused children from ages 9-19 were selected from three Michigan counties who have contrasting societal system interventions. The ninety children were interviewed in a semistructured format. Two instruments were used to measure trauma and mental health, the Trauma Symptom Checklist for Children and the Child Behavior Checklist. The Intervention Stressor Inventory and the investigator's own questionnaire were administered to determine the amount of societal system stress and the children's perceptions of subsequent harm. The results of the study supported certain components of the primary hypothesis. The number of interviews children experienced and the ability to establish trust with a professional were statistically significant predictors of trauma scores. Other major system interventions, testifying and removal of the child from the home, were not statistically correlated to trauma scores. The secondary hypothesis that there would be statistically significant differences between counties in trauma scores due to the forms of societal system interventions in each county was not supported by an analysis of the data. I dedicate this study to the sexually abused children who have so courageously shared their secrets and painful realities with me over the last fifteen years. ACKNOWLEDGMENTS A dissertation reflects the author's character and commitment to the profession and most importantly to people. My life has been full of people who have challenged me to personally and professionally grow. My family and friends have surrounded me with loving care throughout my life which has precipitated my willingness to explore the dynamics and pain of child sexual abuse. I am especially indebted to my mother, a social worker herself, who has provided a model of professional excellence and personal care in working with children. I am so very thankful for the loving support of my wonderful wife Mary Kay, and children; Rachel, Jessaca, and Aaron who have been willing to experience four grueling years of study and effort. It is their encouragement that has provided light in some dark days. My friends have surrounded me with hugs, verbal support, and energy that has been so necessary in order to keep going. I would personally like to thank my friends, Rich, Roberta, Pam, Mary, Jim, and Carol. I am very grateful for my committee who have spent much time and energy in guiding me. Much thanks goes to Victor Whiteman, Thomas Luster, Dorothy Harper-Jones, and Jackie Lerner. vi TABLE OF CONTENTS CHAPTER 1: OVERVIEW OF SOCIETAL SYSTEMS ............ 1 INTERVENTIONS Introduction .................................. 1 Literature Review ............................. 3 CHAPTER II: QUESTIONS .............................. 7 Research Question ............................. 7 Trauma Theories ............................... 9 Informational-processing model ............. 9 Psychosocial model ......................... 10 Developmental model ........................ 10 Post traumatic stress disorder ............. 12 Traumagenic dynamics model ................. 15 CHAPTER III: METHODOLOGY ........................... 19 Purpose ....................................... l9 Hypothesis .................................... l9 Assumptions ................................ 24 Procedure ..................................... 28 Participants ............................... 28 Sample Characteristics ..................... 31 Design ........................................ 35 Sample Procedure .............................. 35 Measurement ................................... 41 Child Behavior Checklist .................. . 42 Trauma Symptom Checklist ................... 44 vii Intervention Stressor Inventory ............. 46 Investigator's Questionnaire ............... 48 Validity ...................................... 50 Internal Validity .......................... 50 External Validity .......................... 55 CHAPTER IV: RESULTS ................................ 60 Sample Description ............................ 60 Results ....................................... 67 Original hypothesis ........................ 70 Second hypothesis .......................... 75 CHAPTER V: DISCUSSION .............................. 88 Interviews ................................. 88 Trust ...................................... 97 Testifying ................................. 102 Removal from home .......................... 105 Overall perceptions ........................ 110 County differences ......................... 118 CHAPTER VI: IMPLICATIONS AND CONCLUSIONS ........... 123 viii LIST OF TABLES I. Correlation with Variables of Severe Trauma with Trauma Symptom Checklist ................ 51 II. Correlation with Variables of Possible Trauma with Trauma Symptom Checklist ................ 52 III. Correlation with Intervening Variables with Trauma Symptom Checklist ................ 53 IV. Sample Characteristics using Continuous Variables and County ......................... 61 V. Sample Characteristics using Nominal Variables and County ......................... 66 VI. T—test for Variables Likely to Cause Severe Trauma with Trauma Symptom Checklist .. ....... 69 VII. Who Believed Disclosure in the Family with Trauma Symptom Checklist ................ 106 VIII. Most Difficult Part of the System Interventions ................................ 112 Ix. Most Helpful Part of System Interventions .... 113 X. What Could Have Made System Intervention Easier ....................................... 114 ix LIST OF FIGURES OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO LIST OF APPENDICES A. Child Behavior Checklist ....................... 136 B. Trauma Symptom Checklist ....................... 140 C. Intervention Stressor Inventory ................ 145 D. Investigator's Questionnaire ................... 148 E. Fact Sheet .... ............... . ................. 156 F. Investigator‘s Letter to Parents ............... 157 G. Court Letters to Parents ....................... 158 H. Child Letter ................................... 163 BIBLIOGRAPHY ...................................... 164 xi CHAPTER I OVERVIEW OF SOCIETAL SYSTEM INTERVENTIONS Introduction American society's willingness to recognize that intrafamily child sexual abuse exists, despite years of resistance, appears to have finally occurred. Society now encourages children to disclose sexual abuse through prevention programs, media campaigns, and open public acknowledgment that adults do molest children. In spite of these gains within society, when sexually abused children consider disclosure they are most often confronted with a staggering personal dilemma that far exceeds their maturity. They are forced to choose between continued abuse, or betrayal of someone of great significance within their lives. The difficulty in making their decisions is frequently exacerbated by their fears of being physically harmed or their sexual abuse disclosure not being believed. Society's response to these children must go beyond recognition of their problem if children are going to have a chance to overcome the tragedy of their abuse. The investigation and subsequent follow up of child sexual abuse involves a variety of societal systems. The key designated systems for this process are, child protective services, law enforcement, and the juvenile and criminal courts. The catalyst that initiates the involvement of each of these systems is the sexual victimization of a child. Each system is mandated by society to achieve different goals in the aftermath of child victimization. Child protective services is to secure the safety of the child, law enforcement the arrest of the perpetrator, juvenile court to determine the most appropriate placement for the child, and the criminal court to decide on the guilt of the perpetrator. The investigation and interventions by each system are focused on the achievement of their own defined goals. The result is that frequently differing approaches to the child are utilized. Consequently the child is required to participate in several systems simultaneously, all having their own demands and expectations for the child. The child is forced to cope with the rigors of this process if societal intervention is to be successful. Sexual abuse victims are engulfed in a web of pain and trauma. Societal systems respond by telling children what must occur in order for them to be safe from further abuse. The directions are complicated and overwhelming for most child victims. Victimized children, given their age, potential arrested development, and the frequent absence of supportive systems, lack the emotional and psychological resources to navigate through the system. If societal systems do not address the needs of these children through supportive interventions they become a co-conspirator in the traumatization of children. Literature Review Reflecting society's past refusal to accept child sexual abuse as a social problem, there have been few questions and no identified empirical studies on the effects of system intervention on child victims until relatively recently. Fortunately questions have surfaced among experts in the child sexual abuse field as to the psychological and emotional impact of system intervention on the lives of these children (Whitcomb, Shapiro & Stellwagon, 1985: Goodman & Aman, 1990; Runyan, 1986; Runyan, Everson, Edelsohn, Hunter & Coulter, 1988; Tedisco & Schnell, 1987; Newberg. 1987). Many theories, especially regarding child testimony, indict system intervention as psychologically harmful to children (Weiss & Berg, 1982; Berliner & Barbieri, 1984; Burgess & Holsmstrom, 1976). In contrast there are those who hypothesize, using psychological theory, that interventions such as child testimony are beneficial to children. They postulate that testifying may increase the child's sense of self- efficacy, serve as coping strategy, and provide a sense of psychological closure to a traumatic experience (Pynoos & Eth, 1984; Melton, 1984). Child sexual abuse professionals are in agreement with Melton (1984) regarding his testimony before a United States Senate Committee when he concluded that there is a need for a "substantial research initiative" on the effects of child testimony and societal system intervention on children, as most of what directs the system interventions today are past practices and theoretical speculations (Tedesco & Schnell, 1987, p. 269). Studying societal system intervention is an enormous task given the numerous variables and the difficulty of separating system induced harm from harm suffered from child sexual abuse itself. A small number of researchers, concerned with the effects of system intervention have begun gathering research data on this subject (Runyan, Hunter, Everson, Whitcomb, & De V03, 1992; Tedesco & Schnell, 1987; Newberger, 1987). Preliminary results indicate the need for change in societal system interventions, as well as the need for further research (Tedesco & Schnell, 1987; Runyan, Hunter, Everson, Whitcomb, & De V08, 1992). Tedesco and Schnell (1987) conducted a study in Iowa and devised their own questionnaire on whether or not societal system intervention was harmful or helpful to child victims. They mailed 120 questionnaires and 48 were returned. Their convenience sample consisted of persons ages 4 to 22, with the average age being 13 years. The results indicated that societal system interventions were not necessarily harmful. The most significant findings were that the most harmful process to children was the number of investigatory interviews children experienced. The greater the number of interviews the more harmful the process. Second, children who had to testify viewed the process as harmful. The study, by the authors own admissions, is limited by "its sampling techniques, small number of subjects, and lack of a more objective behavior rating scale" (Tedesco & Schnell, 1987, p. 271). The study, however, does supply an introductory understanding of what children experience due to system intervention and provides a direction for further research. Another study conducted by Runyon, Everson, Edelsohn, Hunter, and Coulter (1988) examined the effects of legal interventions on one hundred sexually abused children in North Carolina. They employed a cohort study design in an attempt to learn if sexually abused children's mental health improved in response to the patterns of intervention each child experienced. Standardized testing measures were used to assess the mental health of the one hundred children that participated in the study (Runyon, et al., 1993). There were two significant findings of the study. First, children who testified in juvenile court appeared to benefit from the experience by demonstrating improved mental health, as indicated by the testing measures. The finding was unexpected to the researchers as their expectation was that testifying would increase anxiety and consequently mental health would deteriorate. Second, those children who were still awaiting juvenile court trials at the end of the five month testing interval were only 8% as likely to show improvement in depression as their counterparts who had completed their system involvement. The conclusion was that the lack of legal resolution impedes child mental health recovery due to the continued threat of change through court action (Runyon, et al., 1993). The researchers defined the need for further study to increase professionals knowledge of how legal interventions impact sexually abused children; their belief being that new information would produce legal interventions to better support children. This investigator chose to further investigate the impact of societal system interventions in response to professional experience with victimized children and the findings of the aforementioned studies that outlined the potential important benefit that could result. CHAPTER II QUESTIONS Research Questions I. Do the investigatory processes, court proceedings, and social service interventions by the socially mandated authorities i.e., child protective services, law enforcement, and the juvenile and criminal courts, in intrafamily child sexual abuse, increase the level of trauma experienced in child victims after their disclosures? II. If it is found that some or all societal interventions do increase trauma, what then is the best model for societal intervention that will minimize the risk of further trauma to the child? Discovering answers to these questions demands a clear delineation of what is meant by trauma. Stress, pain, and trauma are related but are not identical. Pain for purposes of this study is identified as a feeling state that can physically manifest itself through such occurrences as tears, minor stomach disturbances, and difficulties in swallowing. Pain is an experience of the present and is usually a spontaneous response to a stimulus. Pain need not affect behavior, although it may initially, but then quickly dissipates and does not affect normal development in children. If pain persists and begins to affect the child's behavior and future development it is redefined as trauma. Stress is similar to pain in that it is most often experienced as the result of external circumstances. It will be defined as tension or constraining force that is felt mentally, physically, or psychologically. It can alter behavior and cognitive processing temporarily. Stress is most often able to be assimilated into the internal framework of the individual without any long term consequences. Stress, like pain, can result in trauma and produce developmental disruption. Both pain and stress are usually uncomfortable but frequently cannot be avoided because they are a by-product of many different kinds of life events. Some level of stress and emotional pain will occur in any societal system intervention in child sexual abuse situations no matter how non-intrusive it intends to be (Runyon, Everson, Hunter, Whitcomb & De V08, 1992). These intrapsychic phenomena are unavoidable given the impact of child sexual abuse and the family and societal events proceeding disclosure. The question however, for this study, was whether the system by its methods of intervention generates such significant amounts of stress and pain that new trauma occurs or previous trauma is exacerbated. Trauma Theories There are a variety of theories that attempt to explain trauma and its consequences. Each find their fundamental constructs within a particular school of psychological theory. Most recognize trauma as being initiated by an external event, but each theory gives varying weights to other factors which affect the individual's experience of trauma. Information-procesging model Horowitz (1979) proposes the information-processing trauma model which has its roots within psychoanalytic thought. He views trauma as, "events which involve massive amounts of internal and external information, most of which cannot be matched with a person's cognitive schemata due to the fact it lies outside the realm of normal experience. The result is information overload; the person experiences ideas, affects, and images which cannot be integrated with the self." (Horowitz, 1979, p. 70). The consequence of this overload is a manifestation of symptoms, such as psychic numbing, flashbacks, repetitive nightmares and unwanted thoughts of the event. These symptoms occur because the event is unable to be processed by the internal coping systems which then create disequilibrium within the person. The resolution of trauma occurs when a new 10 cognitive model is constructed that integrates the previous information surrounding the event into the psychic structure which eliminates the disequilibrium. Egychoeocial meeel Another model for trauma is the psychosocial theory postulated by Green, Wilson, and Lindy (1985). It proposes that there is an, "interaction between a traumatic stressor, "normal" reactions to catastrophe, individual characteristics, and the social/cultural environment in which the trauma is experienced and in which the person recovers" (Green, Wilson, & Lindy, 1985, p. 72). This approach defines the best predictors of significant trauma as the severity of the stressor and the degree of psychosocial isolation in the environment during the recovery from the event. If the person is within a favorable recovery environment the possibilities of working through trauma are greatly enhanced. Developmental meeel Pynoos and Eth (1986) have created a model of childhood trauma from a developmental perspective. They define trauma as occurring when an, "individual is exposed to an overwhelming event resulting in helplessness in the face of intolerable danger, anxiety, and instinctual arousal. The result is painful, frightening, and distressing, and is usually followed by 11 a constellation of psychiatric symptoms." (Pynoos & Eth, 1986, p. 307). The symptoms in children include, reenactments or unknowing performance of acts similar to the traumatic occurrence (Terr, 1983); repetitive and unsatisfying play involving traumatic themes (Terr, 1981); pessimistic expectations for the future (Terr, 1983); enduring personality changes (Gislason & Call, 1982); heightened inhibition, counterphobic behavior, school problems, poor impulse control, and problems in interpersonal relations (Pynoos & Eth, 1986). Pynoos and Eth further posit that the experience and manifestation of these symptoms are significantly influenced by the child's particular developmental phase. They offer four developmental considerations in the experience of trauma and its consequences for the child. 1) The specific nature of the traumatic occurrence and its meaning to the child may vary by age of the child. 2) Children's early efforts to cope with traumatic anxiety and helplessness are a function of their maturity. 3) The developmental characteristics can help to ameliorate or impede trauma resolution. 4) There is an interplay of trauma on developmental childhood tasks such as schoolwork, play and interpersonal relations. Over time if trauma is unresolved there can be accentuation, retardations or fixations in subsequent developmental stages (Pynoos & Eth, 1985). 12 Egg; traumatic etreee disorder The preceding trauma theories as well as several others including, behavioral, psychodynamic, and object relations explanations, lay a foundation for the mental health diagnosis in the Diagmostic and Stagietical Manual 9; Mental Disorders (Third Edition) (DSM IIIR) of Post Traumatic Stress Disorder (PTSD). This is a recent standard diagnosis that was first included in the Diagmostic and Statietical Manual 9: Mental Disorders (Second Edition) in 1980. The present ng ILLR Post Traumatic Stress Disorder diagnosis, which includes nightmares, psychic numbing, intrusive imagery, and memory impairment, is reflective of adult symptoms. New data indicates that children frequently have these symptoms, but also may have different symptoms given their stage in the developmental process (Terr, 1985; Frederick, 1985). The differences include, post traumatic play and reenactment, time skew, school problems, and conduct disturbances (Terr, 1985; Frederick, 1985; Kinzie et al.,1982; Malmquist, 1986). The examination of child sexual abuse in light of the various trauma theories that serve to explain the diagnosis of Post Traumatic Stress Disorder are important because of the assistance they provide in understanding the effects of child sexual abuse. David Finkelhor (1985), a leading expert and researcher in the 13 field of child sexual abuse recognizes that there are similarities between a Post Traumatic Stress Disorder diagnosis and many of the symptoms displayed by sexual abuse victims. According to Finkelhor, viewing the consequences of child sexual abuse within the structured framework of Post Traumatic Stress Disorder has linked the trauma of child sexual abuse victims to the other types of trauma frequently recognized within society, such as exposure to war, death of a parent, and terminal childhood illness. This linkage provides for greater acceptance within the psychological community of child sexual abuse trauma, a reduction in the stigma to victims, and a broader base of research with which to connect child sexual abuse trauma. Despite the recognition of these positives, Finkelhor ultimately opposes including child sexual abuse trauma within the Post Traumatic Stress Disorder diagnosis. He cites several reasons for his position. First, there are different areas of emphasis between Post Traumatic Stress Disorder and child sexual abuse trauma. Post Traumatic Stress Disorder's primary focus is on the intrusive imagery and the decreased affect and social interaction. For child sexual abuse victims the central symptomology is the increase in fear, depression, self-blame, and sexual problems. Other researchers (Briere & Runt, 1988) have documented suicidal ideations, substance abuse, and revictimization 14 as common to child sexual abuse victims that are not contained within the parameters of Post Traumatic Stress Disorder. Second, there is evidence that many child victims may not have Post Traumatic Stress Disorder symptoms. In a study by Kilpatrick (1986) of 126 women who had been sexually abused as children only 36% had ever experienced Post Traumatic Stress Disorder symptoms. In another study of over 500 women, Hindman (1989) found a large proportion had few Post Traumatic Stress Disorder symptoms despite experiencing other significant symptoms especially in the area of sexuality (Hindman, 1989). Finkelhor's concern is that if symptoms of child sexual abuse trauma are defined under the heading of Post Traumatic Stress Disorder then many children will be viewed as not being traumatized when in fact they have been. Finkelhor also defines some important contrasting theoretical foundations between Post Traumatic Stress Disorder and child sexual abuse trauma. Post Traumatic Stress Disorder has no one consensus theory and Finkelhor takes issue with the primary theoretical explanation of Horowitz and Pynoos & Eth because of their emphasis on trauma as an event. Finkelhor argues that the trauma for child sexual abuse victims occurs within a relationship which changes meaning over time and is not defined by a single event. Further he views 15 these theories as too limiting in scope because they fail to adequately explain such phenomena specific to sexual abuse such as sexualized behavior, betrayal, and "overintegration". Overintegration is defined as the taking of a learned behavior in an abusive situation and the act of applying it indiscriminately to other situations (Finkelhor & Browne, 1985). Traumagenic dymamics model Finkelhor has developed his own trauma theory designed specifically for the phenomena of child sexual abuse. His model called the, Traumagenic Dynamics Model of Child Sexual Abuse, is eclectic and defines different dynamics that account for the variation of symptoms. The theory draws on the Post Traumatic Stress Disorder model but expands it to explain elements of child sexual abuse trauma that are not contained in within the Post Traumatic Stress Disorder framework (Finkelhor & Browne 1985). Finkelhor postulates that there are four traumagenic dynamics that account for the impact of sexual abuse. These are 1) traumatic sexualization, 2) betrayal, 3) stigmatization, 4) powerlessness. Each of these dynamics has the potential to alter the child's cognitive and emotional orientation to the world and create trauma by distorting the child's self-concept, world view and affective capacities (Finkelhor & Browne, 1985). The consequences are psychological and 16 behavioral problems that are characteristic of sexually abused children. He defines numerous internal and external symptoms as manifestations of each dynamic. 1) Traumatic sexualization occurs when a child's sexuality is shaped in developmentally inappropriate and dysfunctional ways. The result can be excessive masturbation, reenactment of sexual abuse, age inappropriate sexual behavior, and promiscuity. 2) Betrayal occurs when children discover that someone with whom they were primarily dependent has caused them harm and deceived them. Betrayal frequently manifests itself in extreme dependency and clinging behaviors in young children, and hostility and inability to trust others in future relationships as the child matures. 3) Stigmatization refers to the negative messages about the self, such as worthlessness, shamefulness, and guilt that are internally communicated within the child because of the experience. The child internalizes the distorted messages received from the abusers, other family members, and society. Consequently the child usually is emotionally isolated with poor self esteem which leads to severe acting out behaviors. Later in development the child may gravitate towards stigmatized levels of society where substance abuse and prostitution occur. 4) Powerlessness takes place when the child's will 17 and sense of self-efficacy are repeatedly overruled and frustrated as children often experience the threat of injury or psychological annihilation (Finkelhor & Browne, 1985). Excessive fears, nightmares, phobias, and hypervigilance are the result. Important to the understanding of these dynamics and their symptom manifestation is a recognition that the intensity of child sexual abuse trauma varies (Hindman, 1989; Sgroi, 1985; Whitcomb, 1990). The intensity of trauma in child sexual abuse is determined by several factors. Research indicates that the most severe levels of trauma occur when the child has had a "close" relationship to the perpetrator, the secrecy of the sexual abuse has been maintained for a long period of time, the child is less than twelve, and the support after disclosure by the non—offending parent is minimal (Hindman, 1989; Briere, 1989). Gaining an understanding of the different theories and symptoms of trauma is essential to this study because they defined how sexual abuse trauma and its aftermath are viewed. Finkelhor's Traumagenic Model is used as the theoretical basis due to its thorough explanation and specific application to child sexual abuse. Implementing Finkelhor's model also provides a vehicle to interpret the impact of societal system interventions. Finkelhor's four traumagenic dynamics 18 that affect the child prior to revealing the sexual abuse also continually impact the child following disclosure. Therefore there is great potential for societal system interventions to either heighten the traumatization process through elevated stress to the child or reduce further trauma through appropriate child centered responses. Trauma elevation or exacerbation is supported in research which indicates that child victims are extremely sensitive to further trauma after disclosure because of their predisposition to trauma and the fact that they are highly vulnerable to the responses of others (Hindman, 1989; Pynoos & Eth 1985; Finkelhor & Browne, 1985; Everstine & Everstine, 1993). CHAPTER III METHODOLOGY Purpose The purpose of this study was to determine if trauma was exacerbated by societal system interventions as evidenced by behavioral and intrapsychic symptoms of victimized children. Hypothesis The hypothesis was that certain types of societal system interventions during the investigation, court procedures, and social services processes, further traumatize sexually abused children. The primary types of interventions that were expected to heighten trauma were: 1) more than one interview of a child by police, protective services, and/or prosecutor (Tedesco & Schnell, 1987) 2) a child testifying in criminal or juvenile court (Weiss & Berg, 1982) 3) a child being removed from the mother or primary caretaker by an action of the juvenile court 4) the inability of professionals involved to provide a child emotional safety or security; 19 20 consequently the child cannot establish positive feelings towards at least one person involved in system interventions These four, although being the primary sources of potential system intervention trauma, do not constitute an exhaustive list. Other system interventions which will be discussed later, were also explored in the study so as to determine their possible effects on traumatization of the child victim. The theory supporting the hypotheses is that societal system interventions reproduce children's previous experiences of powerlessness, stigmatization, and betrayal and thereby activate previous sexual abuse trauma or initiate new trauma (See Figure 1). Repeated interviews were found by Tedesco & Schnell (1987) to be viewed as harmful by sexually abused children. Theoretically, repeated interviews of a child can Specifically reinforce stigmatization, powerlessness, and betrayal for three reasons. First, children experience repeated questions about their story as indicating that they are not being believed about the sexual abuse. Second, children are forced to repeat emotionally painful material to meet the demands of adults rather than what their needs are. Third, continued interviews strengthen children's beliefs that they have said something wrong and another interview is taking place to discover what that is. 21 Figure I ._ .1. _.,, ..... _8=::Iu=m _..-..--.:.... I \ \ Ill—«28.; _ r.\ I \\ \ \ R (I 3683-5 I £58.. 8 ) _ 3.6 3 =2... ._, was: 85.3 8.58332. .288 a 8.23.. . 1 8 x. 4 m \ 1 _ Snug—— 3.58.2.2. _ S _ _ =1 9.3.8.. - 2 E. > 1 1 gum... x. ........ _ 11111111111111 835.525. Tlfi 3:8 a , w - , I I I I 1 I 2.8?! _ 22 It is important to also draw the distinction between the interviewing process and counseling. The primary purpose of interviewing is to learn the specifics of who, what, when, and how from the child. Without a specific disclosure the system professionals are thwarted in their attempts to intervene. In contrast, the purpose of counseling is to support emotional release, help affect the cognitive and affective integration of the sexual abuse, and empower the child. Specific disclosure of the sexual abuse events are not immediately necessary in counseling and are not probed unless at the request of the child. Court testimony places a child in a highly vulnerable position of being confronted by the perpetrator as well as the defense attorney. Research, as previously cited, has produced diverse results with the work of Weiss & Berg (1982) supporting testifying exacerbating trauma, while the study by Runyon, Everson, Edelsohn, and Coulter (1988) found that juvenile court testimony did not traumatize children. This investigator believes that there is support for the findings of Weiss & Berg. The primary reasons being that the system that promised protection, then forces the child to respond to hostile questions through cross examination in an adversarial and frightening environment, the courtroom. Powerlessness, betrayal, and stigmatization are the likely by-products of such 23 testimony. The final two tenants of the hypothesis were drawn from the investigators professional experience and Finkelhor's theory as there are no identified studies which explore the effects of removal and professional trust on sexually abused children. Removal of children from their home is likely to reinforce the belief by children that they have done something wrong, thereby increasing stigmatization. When children are removed they suddenly lose any power over their previous environment. Children must then navigate in a strange home with new faces and learn the expectations by trial and error. If they are to establish trust they must do so with peOple with whom they have no past connections or relationship. Critical to the process of relationship building for children is the availability of an adult who seeks to have a positive relationship. Without the establishment of such a relationship sexually abused children internalize that there is something wrong with them and stigmatization occurs. The other possibility is betrayal which results from believing that no one can be trusted. The potential outcome of efforts by professionals to develop a relationship with children is tremendously therapeutic. The relationship with a professional offers children safety, security, and an opportunity for trust. These are all necessary 24 ingredients to begin the healing process for children. Assumptions There are several generic assumptions regarding societal systems and their interventions that were gathered from professional experience, child abuse literature, and research, that contributed to the support of the hypothesis that societal system interventions heighten trauma for sexually abused children. These are: 1) 2) Societal systems of intervention are created and defined by adults; consequently the functioning of the system is determined by adults whose view of the world is through their own model. The understanding of children's needs and how they experience their world is secondary (Everstine & Everstine, 1993). The various social service and law institutions are built around preserving a system that primarily ensures the protection of adult rights. When child sexual abuse victims have to participate in this system there are adult like demands and expectations placed on the child. Each societal system views achievement of its own goal as necessary, just, and in the best interest of children. There is limited evaluation that forces the varied societal systems to recognize that while the goals are 3) 4) 25 noble, it does not preclude the fact that the actual system interventions can revictimize the child through further traumatization. The varied societal systems usually function with cross purposes and coordination of interventions frequently is poor. This increases the probability that intervention efforts are repetitive, confusing, or overwhelming to the child. The physical safety of the child is the primary purpose from which each societal system derives its goals. Research indicates, however, that frequently it is not the physical act of sexual abuse that traumatizes children but the psychological impact of the actions of those involved with the child in the aftermath (Finkelhor & Browne, 1985; Hindman, 1989). Children therefore are especially vulnerable to further psychological trauma following sexual abuse disclosure (Finkelhor & Browne 1985;Summit, 1983). Issues of betrayal, stigmatization, and powerlessness, and sexualization are at their height and the internalization of these are extremely damaging (Finkelhor & Browne, 1985). Insensitivity or discounting of the psychological and emotional needs of the child victim negates the child's 26 inner needs and increases the likelihood of recantation of the story by the child. 5) Most children feel responsible for somehow having invited or initiated the sexual abuse themselves (Hindman, 1989; Finkelhor 8 Browne, 1986; Macfarlane, 1989, Sgroi, 1987). The internalization of guilt and shame over their actions is too often reinforced by societal system interventions that convey messages that blame the victim, such as "Why didn't you tell anyone sooner." or "We have to interview you again to make sure that you know exactly what happened when." 6) Parents are the primary people in children's lives and development. A sexual abusive action by a parent or show of support to the perpetrator by the non—offending parent does not eliminate a child's need to be in relationship to them. Societal systems are quick to openly condemn both parents negating the importance of their role in the child's life. Such actions often serve to alienate a child from the system and further support recantation of the story and internalization of trauma. This study examined the three primary phases of societal interventions, 1) investigatory events, 27 2) court proceedings, 3) social service interventions, involving four different systems: child protective services, law enforcement, juvenile court, and criminal court. Within each phase several events were listed that children often experience after sexual abuse disclosure. These specific events were taken from the Intervention Stressor Inventory (Runyon, Hunter, Hunter, Everson, Whitcomb, & DeVos, 1992) as found in Appendix C. See the Measurement section for measurement and scoring procedures. Each of the three societal intervention phases contained one primary event that was defined as an independent variable. For the investigatory event it was the number of interviews by professionals, for the court phase it was testifying in either the juvenile or criminal court, and for the social service phase it was removal from the parent or guardian. Each phase also included secondary events and these were defined as intervening variables. The dependent variable was the level of trauma to child victims following societal system intervention and was determined by scores received from testing. (see Measurement section) 28 Procedure Participants The participants for the research were defined as: a) Children who were ages 10 to 16 when the sexual abuse occurred. This age span covers two well defined developmental periods, middle childhood (10-12) and adolescence (13-16). The information gathered from these two periods was used for comparison to determine if there were differences between developmental periods in the trauma experienced from societal systems intervention. Children included in this age span are usually able to communicate their thoughts and feelings readily and with a good deal of accuracy. b) The sexual abuse was initiated by an adult living within the household at the time of the abuse. This criteria was used because when the perpetrator is a household member it demands the intervention of all the mandated systems whereas those sexual abuse cases that involve a nonhousehold member are handled only by the police and criminal court. Additionally, sexual abuse cases that involve household members are often more traumatic for the child and therefore the concern to reduce trauma through societal system intervention is even more critical. c) Sexual abuse was defined by using the standard definition outlined in the child protection law and the 29 criminal code in determining child sexual abuse. It is defined as either a fondling of the child's breasts or genitalia either above or underneath the child's clothing or any form of penetration of the genitals. Further it also included any action initiated by the adult household member where the child touched the perpetrator's genitals or breasts. d) The child's initial disclosure of the sexual abuse to an investigator was to be at least 6 months prior to the research interview for this study. This was important because the span of time allowed for a better determination of whether there had been elevated trauma rather than just a temporary reaction to a stressful experience. It also gave time for all the societal systems to intervene because frequently juvenile or criminal court proceedings are not completed for at least 6 months. e) There must have been either an abuse/neglect petition filed in juvenile court on one or both of the parents, or a warrant authorized by the prosecutors office for a criminal offense by the adult household member. There are three reasons why this criterion was necessary: a) The court records were accessible to the investigator; b) The juvenile and criminal courts provided a standardized universe from which the names of the children for the study were chosen; c) The court records provided addresses for each of the children. 30 f) All participants were chosen from the three counties described in the SAMPLE CHARACTERISTIC section where the court involvement was initiated after January 1990. This was accomplished through contact by the investigator with each chief prosecutor and court administrator, and a subsequent authorization by the prosecutor to review all juvenile and criminal court records where child sexual abuse occurred within the defined time ranges. 9) From the list of children's names compiled from court records, fifty were to be randomly selected from each county. If there was a petition filed in juvenile court and also a warrant authorized in criminal court the name was entered only once. By choosing fifty names, provision was made for up to 20 subjects who could not be located or who would refuse to participate and a legitimate sample would still be preserved. h) After the names were randomly chosen letters were to be sent to the parents/guardians of each of the 50 children. The letter was to contain an explanation of the study, the researcher's phone number, and a return envelope requesting child participation authorization. Also included in the letter was to be an understanding that each child would receive twenty-five dollars for an interview. This money provided a reimbursement for the children and parents for their time and possibly a further incentive for participation. 31 Sample Characteristice Kalamazoo County, Kent County, and St. Joseph County were the counties selected for the study. These counties were chosen for the following reasons: a) Demographic data Each of the counties have 3 very different population and demographic characteristics. fig; Joseph is a rural county consisting of several smaller towns and villages with a total population of 58,913. Its average income for two parent families is $40,522 and $16,517 for single parent families. The unemployment rate is 9.2%. It's minority child population is 5.5% of the total number of children. Sixteen percent of its children are below the poverty line. Kalamazoo County has a large city population with a diverse racial makeup and a total population of 223,411. The average income for two parent families is $52,415 and $17,762 for single parent families. The unemployment rate is 5.2%. Minority children makeup 24% of the total child population. Fifteen percent of the children are below the poverty level. gem; County is a large metropolitan area with a total population of 500,631. The average b) 32 income for married couples is $50,325 and $17,709 for single parent families. The unemployment rate is 6.2%. Twenty-two percent of the children in the county are minorities and 12% of all children are below the poverty line. Investigative methods Each of the three counties have contrasting methods of conducting investigations in child sexual abuse cases. See geeeph County uses videotaping at the initial interview with only one professional present. The videotape is used as evidence and seldom is a second interview conducted with the child until prior to the preliminary hearing in criminal court. Further St. Joseph County has immediate police involvement. Immediately following a child interview the suspect is shown the videotape and subsequently interrogated. A polygraph is then offered to the suspect. The process serves as a powerful mechanism in eliciting a confession from the perpetrator. Kalamazoo County has a three step process of interviewing, where the child protective services worker does the initial interview, followed by a joint interview with child protective services and the police, followed by an interview with the prosecutor and the child. c) 33 The suspect is typically not interviewed by the police until all three interviews have been completed, which may take one to two weeks. Kent County has a special sexual abuse police unit for the county. It works in conjunction with child protective services and accompanies child protective services on all interviews, which helps to eliminate further interviews. Law enforcement typically conducts an initial interrogation of all suspects within a short period of time following the interview of the children. Secondary Hypothesis An additional hypothesis was also examined in the study that explored the impact of the differences between county societal system interventions on trauma to the children. The hypothesis was that St. Joseph County, due to its coordinated professional interventions, has the least stressful societal system intrusion for children. Consequently, the children from St. Joseph County were hypothesized as less likely to be further traumatized than children from the other two counties. There is usually only one interview of the child in St. Joseph County which is videotaped and there is no need for further interviews until the court requires 34 a hearing. Due to immediate suspect interrogation there is a greater likelihood of a confession which frequently results in a criminal plea by the perpetrator and this usually prevents the child from having to testify. There is also a strong collaborative effort amongst child protective services, law enforcement, and the court system, which minimizes the demands on the child. Using the above rationale, it was expected that the highest trauma would be found in the children from Kalamazoo County. It has more intrusive societal system interventions as indicated by more interviews of the victim, poor collaborative effort, and a slow response time in interrogating suspects, thereby decreasing the likelihood of gaining a confession and increasing the possibility of child testimony. Kent County was expected to have fewer traumatized victims than Kalamazoo but a greater number than St. Joseph County because of the nature of their societal system interventions. The rationale for this prediction was that the professional team efforts reduce interviews, gain perpetrator confessions, and minimize the possibility of court testimony for children. Consequently they are responsive to the needs of 35 victimized children in a supportive child centered manner and this decreases the possibility of further traumatization. Design A survey design was chosen to implement the study. It was postscriptive due to the inability of the investigator to interview children at the specific time of the societal system interventions. The participants were not available and time constraints made contacts untenable. The research interviews of the children were designed to be conducted in a semistructured format. Sampling Procedure The sample was obtained by the investigator's review of the criminal and juvenile court records in each of the counties. Cases that met the criteria (see Procedure section), were identified and the necessary information documented on a standardized fact sheet (see Appendix E). The children listed on the fact sheets became the potential subjects for the study. The one exception to this process was the criminal court in Kent County which did not allow the investigator access to their records. The Victim Witness Office reviewed the criminal records and generated a list of children, based on the criteria for subjects that was distributed to their office by the investigator. 36 Once a list was obtained from each of the counties all the parents and children were mailed letters seeking their participation in the study. Three letters were contained within each envelope. One letter was addressed to the specific parent/guardian of the child sent by the investigator on Michigan State stationary (see Appendix F). The letter explained the study, requested participation by them and their children, and gave a phone number to call if they were interested. The second letter was an endorsement for the study from the specific court where the child's name was obtained and encouraged the parent and child's participation (see Appendix G). The third letter was addressed to the child and explained the study and requested their participation (see Appendix H). The last address listed in the court case file was used with an address correction request label on each envelope. The Kent County prosecutors office, for the criminal court, individually addressed the envelopes but used a set form greeting rather than individually addressing the letters. The investigator had no ability to individually address the letters since there was no access to their names or addresses. In St. Joseph County a total of fifty-six children met the outlined criteria for participation. There were thirty-one children who had only been through the criminal court and nine who had been through the 37 juvenile court. A total of sixteen children had been through both the juvenile and criminal court. Of the thirty children, out of the possible fifty-six from St. Joseph County, who participated in the study, ten or 33%, had been involved with both courts, seventeen (57%) had been involved with just the criminal court, and three (10%) had been involved only with the juvenile court. After sending out the initial letters in St. Joseph County the investigator received only four calls indicating a willingness to participate in the study. The investigator followed up the initial letters to those who had not responded after one week with phone calls where phone numbers were available. The investigator also attempted to personally contact the families at the last known address. Personal contact was attempted in all three counties in hopes of encouraging participation through personally answering questions, dispelling possible fears, and reminding the parents and the children of the opportunity. It proved to be a very helpful means of gaining participants. Second letters were also sent out to those families with whom the investigator was unable to make personal contact reminding them of the study. All of the interviews in St. Joseph county were done by the investigator. It took two months to complete the thirty interviews in St. Joseph County. Of 38 the twenty-six children who did not participate, there were ten children for which correct addresses were not available. Six additional children had moved out of the state since the abuse. Four parents refused to have their children participate and two children refused to participate. The reasons given for refusal varied from an unwillingness to bring up memories to a desire to get on with a new life. In two situations there were special needs children where the parents felt that it would be detrimental for the children to be interviewed. The remaining four children did not contact the investigator and were not able to be located despite what appeared to be a correct address. In Kalamazoo County a total of sixty-nine names were obtained through the review of the criminal and juvenile court records. There were thirty-eight children who had only been through criminal court and sixteen who had been through juvenile court. The remaining fifteen had been through both court systems. Of the thirty who participated out of a possible sixty nine, twelve (40%) had been through both court systems, thirteen (43%) had gone through just the criminal court process, and five (17%) had been in only juvenile court. The investigator followed up the initial letters in the same manner as had been done in St. Joseph County. Phone calls, personal contacts at known addresses, and a second mailing were all completed in Kalamazoo County. 39 The participants' responses to the first and second mailings were almost identical in the two counties. The investigator conducted twenty-eight of the thirty interviews. Two interviews were done by another interviewer as the investigator had previously been involved with the family in a professional capacity. It took nine weeks to interview the thirty participants from Kalamazoo County. Fourteen children were unable to be contacted in Kalamazoo County due to an incorrect address. No contact was attempted with two children because of adoptions. Six parents refused their children's participation in Kalamazoo County after contact was made. The reasons given were similar to those in St. Joseph County. A total of seventeen children received letters that were not returned and with whom the investigator had no contact. The exact numbers of potential participants in Kent County cannot be calculated due to the inability to identify the names and addresses of the children who only went through the criminal court. According to the Prosecuting Attorney's office in Kent County fifty letters were sent out. There was no way however to check on wrong addresses or to make follow up contact with those children. Consequently it is difficult to determine what the actual proportion of respondents was to the letters sent. Six children out of the potential 7‘~~g- 4O fifty did participate in the study and were exclusively from the criminal system; 7% of the total sample of the study. The Kent County Juvenile Court records yielded sixty-nine names that met the required criteria for participation. Of those sixty-nine children, twenty- four participated in the study. From the sixty-nine names there were seventeen children (57%) who participated in the study who were involved with both criminal and juvenile courts. Only seven (23%) had been involved only with juvenile court. The investigator followed the same process that had previously occurred in the other two counties in attempting to procure interviews with children. In Kent County the investigator interviewed the thirty children out of the possible one hundred and nineteen, 33% of the total sample. It took two months to finish all thirty interviews. Unlike St. Joseph County and Kalamazoo County there were three children who were under the ten year age limit. Initially the study was to exclude children under the age of ten. The investigator chose to include these children because they had received letters from the prosecutors' office and wanted very much to participate. To ensure their appropriateness prior to the interview, the investigator spoke at length with the parents to determine if their children were mature and articulate enough to 41 participate. From the Kent County juvenile court pool, there were fourteen children who had incorrect addresses. Three children had moved out of state, eight adopted children did not have correct addresses, and one child had died. Five parent/guardians refused to allow their children to participate and two children refused. There were similar reasons given in Kent as in the other two counties, except two parents indicated that their child had recanted and therefore they felt it would be inappropriate for them to be interviewed. Fourteen children appeared to have valid addresses but did not respond and no personal contact was made. Measurement Four testing instruments were administered with two purposes. The first purpose was to measure the level of trauma to children who had been sexually abused and had experienced societal system intervention. The second was to measure the amount of societal system intervention stress each child had experienced. The testing instruments were used in combination to test the validity of the original and secondary hypotheses. The Trauma Symptom Checklist (see Appendix B) was employed to measure trauma. The Child Behavior Checklist (see Appendix A), although not an instrument to measure trauma per se, was used as a supplement to 42 gain a social competence and behavioral score to help determine mental health functioning and thus assist either in confirming or negating the trauma findings from the Trauma Symptom Checklist. The Intervention Stress Inventory (see Appendix C) and investigator's questionnaire (see Appendix D) were administered to obtain a measure of societal system intervention stress. CHILD BEHAVIOR CHECKLIST (CBCL) This is a widely used instrument that, "consists of social competence and behavior problem scales which are filled out by parents or the child's primary caretaker" (Achenbach & Edelbrock, 1983, p. 6). The information obtained from the checklist is scored using scales which indicate varying syndromes. Each child receives a subscore for each scale which indicates whether the child's functioning within that particular category falls within normal or abnormal limits. The Child Behavior Checklist (CBCL) includes three competence scales, plus total competence, eight cross—informant syndromes, Internalizing, Externalizing, and total problems. The Child Behavior Checklist (CBCL) was utilized for three reasons. First, research on the Child Behavior Checklist indicates high reliability and validity. The scales were derived from parents' rating of four thousand four hundred and fifty-five clinically 43 referred children and normed on two thousand three hundred and sixty-eight nonreferred children. Achenbach and Edelbrock, creators of the Child Behavior Checklist (CBCL) have conducted numerous research studies (Achenbach & Edelbrock, 1976; Achenbach & Edelbrock, 1981; Achenbach & Edelbrock; 1983) that have well established its validity and reliability within the psychological community (Runyon, Everson, Edelsohn, Hunter, & Coulter 1988). This is important because it offers a behavior and mental health profile score which can be compared to a wide range of children seen in diverse settings. Second, the Child Behavior Checklist is frequently used to compare the large differences in scoring between normal and clinical samples as the Child Behavior Checklist has shown "high discriminatory validity for behavior problems and social competence between the two groups." (Achenbach & Edelbrock, 1983, p. 10). Although the Child Behavior Checklist does not give a specific trauma score, it does provide an assessment of the mental health and behavioral functioning of children. It further provides a basis for comparison of functioning between sexual abuse victims themselves and with the population of children in general. Third, the Child Behavior Checklist is completed by the parent/caretaker and provides an independent source of data to use for comparison. The other testing for 44 this research was compiled from questionnaires and interviews with children. The Child Behavior Checklist has also previously been used in conjunction with the Intervention Stressor Inventory and also with the Trauma Checklist for Children, both of which were used for this study. TRAUMA SYMPTOM CHECKLIST FOR CHILDREN (TSC-C) The Trauma Symptom Checklist for Children (Briere, 1989) was administered to establish a trauma score for children. "The purpose of this instrument is to assess acute childhood traumas of various types and it is especially sensitive to sexual abuse in particular" (Briere, 1989, p. 3). It was developed from the Trauma Symptom Checklist which is an instrument for adults who were molested as children. It was useful to this study because it is specifically designed to provide a trauma score for children who have been sexually abused. The Trauma Symptom Checklist for Children is consistent with the theory proposed and the symptoms described by Finkelhor. It was completed by the children in this study to determine what level of trauma the children were experiencing at the time of the testing by the investigator. Previous research results by Briere testing the reliability and validity of the Trauma Symptom Checklist (TSC-C) have been positive. There is a relatively high internal consistency at both the full scale and subscale 45 levels. In a 1991 study of fifty-six children, Briere found a total alpha of .96 and the mean alpha value for subscales of .83. The highest subscale reliability was the Depression subscale at .89 and the lowest being Sexual Concerns .68. The Trauma Symptom Checklist for Children (TSC-C) also correlated fairly well with Child Behavior Checklist done by parents, although there was greater correlation with the Child Behavior Checklist self report scores. The construct validity is suggested by; a) the TSC-C's covariation with self-reported symptomology on the Child Behavior Checklist and the Child Depression Inventory; b) its association with specific aspects of subjects' abuse history (age at first abuse, penetration); c) its longitudinal decrease as a function of abuse-specific treatment intervention (Briere, 1991). There is also good evidence that the Trauma Symptom Checklist for Children is a valid measure of sexual abuse sequelae. Further a study on the validity and reliability also highlighted the potential usefulness of child self-report measures as opposed to sole reliance on parent or adult reports. The data suggested that even young children are able to report reliably on their internal experience, and that such reports can be 46 important sources of information on child abuse effects (Briere, 1989). INTERVENTION STRESSOR INVENTORY (ISI) The purpose of the Intervention Stressor Inventory (ISI) is to "establish relative weights for the different experiences likely to be encountered by children in the process of intervention following report of child sexual abuse" (Whitcomb, et al., 1992, p. 2). The Intervention Stressor Inventory provided a vehicle for the investigator to measure the probable level of stress that the three different societal system interventions, produced in sexually abused children. It was derived from a questionnaire sent to designated professional experts working in the field of child sexual abuse. From the ninety-eight returned, Whitcomb et al., were able to give mathematical weights to the amount of perceived stress that the child would experience. The greater the mathematical weight the greater the stress. The one concern however, is that the weights given by the professionals may not correspond to the actual stress experienced by the child (Whitcomb, et al. 1992). This instrument, despite this concern, provided this study an objective measure to classify and score interventions which enabled the investigator to label the intensity of stress within a particular set of interventions. 47 The Inventory was administered by asking the child which specific experiences designated in the Intervention Stressor Inventory the child actually had experienced. The investigator then assigned the designated score to each experience as listed in the Intervention Stressor Inventory. The scores were then individually tabulated for the three phases of intervention: investigative, court, and social service, as well as total stress score. The Intervention Stressor Inventory does not specifically measure the areas of powerlessness, stigmatization, and betrayal. These potential trauma outcomes were measured through the investigator's self designed questionnaire. Validity of the instrument has been supported by previous analyses, although it has not been repeatedly tested. "Face validity has been established through the process of event selection and the subsequent rating of events by experts in the child sexual abuse field. Construct validity of the Intervention Stressor Inventory has been supported by decreased stress scores over time corresponding to the planned systematic interventions of the research project that developed the instrument" (Whitcomb, et al., 1992, p. 5). Concerns regarding the validity of the instrument centered on the subjects used by the initial ninety-eight professionals who were asked to evaluate stress. They were only asked about females and the females considered were school-age 48 adolescents. Another concern regarding validity is the ability of the instrument to measure other stress factors outside the sexual abuse and societal system interventions that contribute to the stress of the child victim. Future uses of the instrument may address this issue. INVESTIGATOR'S QUESTIONNAIRE This instrument is an open ended self report of what the child recalled about the specific societal system interventions and their internal effects following disclosure. It was designed by the investigator for several purposes. The primary benefit in using the questionnaire was its value in supplying qualitative data through a list of open ended questions. Children's personal statements are powerful sources of information from which to learn the impact of societal system intervention. Efforts were made to have questions that could be grouped together to identify responses that might reflect important patterns. Second, some questions on the investigator's questionnaire were not included on the Intervention Stressor Inventory. Specifically, there were questions regarding the child's experience of powerlessness, stigmatization, and betrayal during the three phases of societal system interventions. There was at least one question on powerlessness, stigmatization, and betrayal 49 within each intervention section of the questionnaire. Four overview questions were included at the end of the questionnaire that attempted to further clarify how societal system interventions affected the child's experience of Finkelhor's three traumagenic dynamics. The hope was that the questions could provide data to assist in either supporting or negating the investigator‘s theory that societal system interventions activate the traumatic recall of the three dynamics and thereby exacerbate trauma. Included in the questionnaire were questions that attempted to solicit the support perceived from significant others and the subsequent impact of the family response on the child. The questionnaire was also used to examine the impact on the child of a perpetrator's confession, arrest, and subsequent incarceration. Other questions addressed counseling, its length, and its perceived usefulness by the child. The final use of the questionnaire was to obtain specific demographic data from parents or guardians. The following information was requested: 1) Age of the child 2) Gender 3) Race 4) County of residence 5) Present placement of the child 6) Sex of the perpetrator SO 7) Perpetrators relationship to the child (ex. father) Validity The validity of the study design and data obtained was threatened by various factors. Efforts were taken to minimize these by building controls into the design and statistical analysis. Internal Validity History and maturation were the primary threats to internal validity. The primary method for controlling for maturation was done by selecting children who had made disclosures during a three year and a half year period of time, 1990-1993. This segment of time provided adequate numbers of children within the designated time periods to allow for comparisons to discover if there were differences in trauma scores that could be attributed to the passage of time between initial disclosure and the interview for the study. The investigator was unable to convert year of disclosure to specific months because only the year of disclosure was recorded and not the month it occurred. One way analysis of variance was run and no significant differences were found between year of disclosure and trauma scores (§_= 1.12, p = .34). The finding indicates that maturation was not a significant factor in determining the trauma score, thereby negating the passage of time as a threat to internal validity. There were several ways that history was tested to determine if it affected the results. First, the variables that were thought to be the most likely to cause severe trauma: parental relationship to perpetrator, lack of family support of the disclosure, younger than age at twelve at the time of the abuse, and over one year length of time from onset of abuse to disclosure, were run using bivariate correlation tested by Pearson's R. They were analyzed to determine if these variables significantly affected trauma scores. Trauma scores were not significantly correlated with age at abuse, age at disclosure, relationship to perpetrator, or the family response at the time of the disclosure (see Table I). Table I Correlation with Variablee f Severe Traume with TSC Variables 1. age at abuse 2. age at disclosure 3. relation to perpetrator 4. family response 5. trauma score .46 p=.00 -.09 p=.35 .21 p=.04 .07 p=.51 2. .46 p=.00 -.l4 p=.17 .39 p=.00 .13 p=.21 -.21 p=.35 -.14 p=.17 -.06 p=.53 -.07 p=.48 4. .21 p=.04 .39 =.OO -.06 p=.53 .03 p=.75 5. .07 .13 p=.21 -.O7 =.48 .03 p=.75 52 Second, the above mentioned four variables were dichotomized to expand the definitions of the categories of severe trauma to include a larger number of children to further test the possibility that these characteristics were significant factors in determining trauma scores. T-tests were run on each of the four variables with trauma score, but no statistically significant differences were discovered. Other variables that related specifically to the sexual abuse itself prior to the disclosure (i.e. number of incidents, type of sexual abuse, length of abuse, and thought to possibly affect trauma scores were also run using bivariate correlation and being tested by Pearson's R. The number of incidents was the only one that proved to be significant with an association of .20 and with p = .049 (see Table II). Table II Correlation with Variablee _f Possible Traume with TS Variables 1. 2. 3. 4. 1. Length of -- .73 -.27 .14 abuse p=.00 p=.00 p=.16 2. Number of .73 -— -.35 .20 incidents p=.00 p=.00 p=.049* 3. Type of -.27 -.35 -- -.17 sex abuse p=.00 p=.00 p=.10 4. Trauma .14 .20 -.17 -- score p=.16 p=.049* p=.10 Note. * indicates significant p<.05 53 Intervening variables that occurred after the disclosure (i.e. counseling, medical exam, and court involvement) were also run with trauma scores using bivariate correlation to determine if any affected trauma scores. However none of these variables significantly affected trauma scores (see Table 111). Table III Correlation with Intervenimq System Variablee with TSC Variables 1. 2. 3. 4. 1. Counseling -- .31 -.08 .12 p=00 p=.4l p=.25 2. Court .31 -- -.01 .17 involved p=.00 p=.88 p=.10 3. Medical exam -.08 -.01 -- .01 p=.41 p=.88 p=.92 4. Trauma score .12 .17 .01 -- p=.25 p=.10 p-.92 Of all the variables tested in examining the affect of history on validity, only number of incidents was significant and this was only at the .049 level of significance. The fact that the number of incidents had a correlation coefficient of .20 indicates that it is unlikely to have significantly impacted history. The conclusion therefore was that history did not affect the results of the trauma scores. Other lesser threats to internal validity were 54 selection, instrumentation, mortality, and interaction. Selection was controlled for by obtaining the three samples from the same processes. The fact that the investigator did not have personal access to the criminal records in Kent County did not appear to affect selection of the sample. Although there were fewer participants that went only through criminal court in Kent County than in Kalamazoo and St. Joseph Counties, the difference was not significant between counties using one way analysis of variance. The explanation for this is that many of the children whose names were in the juvenile court records also had criminal court cases as well. Those children who participated in both courts were the majority of cases in all three counties. The letters requesting participation were sent to the entire population of each county. Those who responded initially were few, approximately five in each county. Follow up was attempted with all children either by phone, mail, or personal contact. Arguments could be made that those children who either refused to participate or failed to respond were in some way different from county to county. This would appear not to be true as the reasons given by the parents and/or children for failure to participate were similar across all three counties. Consequently there was little evidence to suggest that children did not participate for different reasons in one county than another. 55 Instrumentation was controlled for by administering the standardized tests and questionnaire similarly each time they were given. The order they were administered, questions asked, and notes taken were done consistently throughout the study. The standardization of the interview process by the investigator, who did all but two of the interviews, minimized the possibility of errors in scoring the instruments. Interactions were controlled by ensuring that any time there were two children being interviewed at the same home the interviews were conducted separately. Two sets of siblings interviewed for the study were living within the same home. No contact between these children occurred during the testing in hopes of eliminating any influence that one child might have on another. No children interviewed had any contact with another child prior to being interviewed by the investigator. External Validity The primary threats to external validity were sample selection, sample size and reactivity. The fact that there was no access to the criminal records in Kent County made it impossible to document the exact percentage of those who participated out of the possible population. However, for the other five courts in the three counties, 60% of those that received a letter explaining the study participated in the study. This percentage does not take into account the fifty-five 56 children who were not able to be contacted due to incorrect addresses. The breakdown of participants in the study who received letters by county populations consisted of 75% from St. Joseph County, 53% from Kalamazoo County, and 52% from the juvenile court in Kent County. Although there was only a small percentage of contact by the investigator with the non- participants, a variety of reasons for not participating in the study were given. There was no way of knowing however, if the reasons given were in any way different than the others who did not participate and with whom no contact was made. Consequently, it cannot be known if the children who did not participate were in any way different than the respondents that did. The fact that 60% did participate increases the likelihood of the sample being representative and helps to minimize any selection bias that could have occurred. The ninety children that participated in the study comprised a small, but adequate sample size. The problems with sample size were evident in the breakdown of variables that often left only a small sample size for statistical analysis. This was further complicated when the sample was divided along county lines, experiences, or ages which meant that the sample size diminished even further. An effort was made to have no less than thirty within any group, but often this was not possible given the variable breakdowns. This was 57 especially true for the Child' Behavior Checklist which had thirty-five girls in the largest group (girls 12-16) but only nine children in the second girls group (girls 8-11), and three children in the boys group. These low numbers made it impossible to run cross county analysis using the Child Behavior Checklist as a dependent variable. In the investigator's questionnaire there were some questions that were not relevant to the experience of many children which then also reduced the sample size. In some situations statistical analysis could not be run comparing counties because of the small sample size. Reactivity was a constant threat and control very difficult. The settings used for the interviews were the children's homes due to convenience for the children and the secure environment that it was hoped the home would provide. Interviewing the children, especially the investigator's questionnaire, had the potential to evoke many unpleasant feelings and memories that could not necessarily be foreseen. Every effort was made to provide children with comfort and security. The difficulty that the setting presented was not the environment itself but rather who was present while the interview occurred. As the children were being interviewed the parents/guardians of the children under seventeen were often filling out the Child Behavior Checklist. Frequently the parents/guardians would 58 remain in the room while the interview took place. Their presence posed a dilemma for the investigator because of the uncertainty of their influence upon the children. The ideal was to not have the parent/guardian present in the room but still accessible. Most often this did not occur as the parents/guardians remained in the room where the interviewing was occurring. The investigator was reluctant to request that they leave because of their need to ensure their child's safety with the investigator. There were times when the investigator was not sure how a child was going to react given the child's non-verbal cues (i.e. restlessness, rapid breathing, lack of eye contact) and so at times it was useful to have the parent/guardian present. In order to minimize reactivity, the investigator always had the children read and answer the Trauma Symptom Checklist to themselves. This process maintained the children's privacy and yet still gave the children visual accessibility to the parent/guardian. There is no question however, that reactivity occurred in some of the oral responses by the children especially when the child would look first to the parent before answering and/or a parent would volunteer answers for the child. Every attempt was made to diminish this possibility including pretest instructions of the interviewing process to the parents/guardians and children, a reminder to the adults when they would 59 interrupt the interviewer or child, and a designated time with the parent/guardian to give an opportunity for their feelings and perceptions of the societal system impact on their child to be expressed. The issue of limited universe was a given when the research began. Using only three counties for the research greatly reduced the ability to generalize the findings. Time, energy, and lack of money prohibited the expanding of the sample. Examining three very different counties with very different system intervention was the method used to minimize the limited universe and to expand the study's implications. CHAPTER FOUR RESULTS Sample description The exploration and ultimate determination of the validity of the hypothesis that societal system intervention exacerbates trauma in sexually abused children, first required an examination of the pertinent characteristics of the sample itself. The examination was necessary to determine when and if appropriate generalizations could be made to other sexually abused populations outside the realm of this study. It also was important to discover if the sample was inherently different between counties in order to determine whether the second hypothesis, that there would be a statistically significant difference between counties in their impact on trauma through their differing societal systems interventions, was supported. One way analysis of variance and chi-square analysis were the statistical tests used to determine if the characteristics of the samples among the counties were significantly different at the .05 level of significance. The mean age of the sample was fifteen. The mode was seventeen and totaled 23.3% of the sample. The youngest was age eight and there were two nine year olds. All three of these children were from Kent 60 61 County. Nineteen percent of the children were nineteen years of age and they were the oldest in the sample. The mean age for St. Joseph County and Kalamazoo County was fifteen. Kent Count had a mean age of fourteen and a half. No statistical significant difference at the .05 level between the three counties was evident using one way analysis of variance (see Table IV). Table IV Sample Characterietics using Continuoue Variablee and County Counties Variables St. Joseph Kalamazoo Kent Age m 15.2 15.5 14.4 §Q 2.4 2.2 3.0 Age at abuse m 10.8 11.7 10.2 SQ 3.1 2.6 3.1 Age at disclosure m 13.0 13.1 12.4 SQ 1.1 1.0 .8 Note. No significant differences were discovered. The sample included eighty-one females, 90% of the total sample, and nine males, 10% of the sample . There were five males from St. Joseph County and four males from Kent County. The racial identity was eighty-one Caucasian, 90%, and eight African Americans, 9%, with one Hispanic, 1%. There were four African Americans from Kent County, three from Kalamazoo County, and one from St. Joseph County. No statistical difference was discovered in either race or sex among counties using 62 chi- square (see Table V). Fifty children (56%) were living with a non- offending parent without the perpetrator present at the time of their participation in the study. There were an additional six children (7%) who were living with a non- offending parent and the perpetrator had returned to the home. Thirty-four of the children were no longer living with a parent (38%), sixteen (18%) were living independently, ten (11%) were in foster homes, four (4%) were living with relatives, two (2%) were in institutional settings, and two (2%) were in an adoptive home. The breakdown of present placement according to county for children living with a non-offending parent mirrored the total statistics, as fifteen of the children from St. Joseph County (17%) were living with a parent, seventeen (19%) were from Kalamazoo County, and eighteen (20%) were from Kent County. One difference noted was that four of the children who were presently residing with the both the non-offending parent and perpetrator were from Kent County, two were from St. Joseph County, and there were not any such living arrangements in Kalamazoo County. Another difference observed was that ten of the children in St. Joseph County (11%) were living independently and six (7%) were doing so in Kalamazoo County, whereas there were not any in Kent County. A 63 chi-square test indicated that there was a significant difference between the counties in present placement as tested by Phi which was at the .003 level. The primary differences between counties regarding placement were in the number in independent living and living with the perpetrator at the time of the investigator's interview (see Table V). There were thirty participants (33%) who disclosed the sexual abuse in 1990. The mode was 1992, with thirty-two children (36%) disclosing in that year. Nineteen (21%) disclosed in 1991 and nine (10%) disclosed in 1993. There was no significant difference between the counties in year of disclosure using one way analysis of variance (see Table IV). Fathers were the most frequent perpetrators in the sample with thirty-two (36%) being the offenders. There were twenty stepfathers (22.2%) and eighteen perpetrators (18%) who were mothers's boyfriends. The final 20% were made up of siblings, relatives, and other household members. There was a significant difference between the three counties in relationship to perpetrator running chi-square and tested by Phi. The value was .6 and the level of significance was .0002. Although the sample size was small the statistics were valid because the expected frequency was five per cell. The statistical difference is attributable to the fact that Kent County 64 had 17 fathers (19%) that were perpetrators compared to only eight (9%) in Kalamazoo County and seven (8%) in St. Joseph County. Also reflecting the statistical difference was that Kent County had only four stepfathers (5%) who were perpetrators compared to nine (10%) in Kalamazoo County and seven (8%) in St. Joseph County. Further Kent County had only two perpetrators who were relatives or other household members whereas in St. Joseph County there were ten such perpetrators and in Kalamazoo County there were eight. The only similarity found between the counties was in offenses by mother‘s boyfriends. Kent County had seven, St. Joseph County six, and Kalamazoo County five (see Table V). The mean for the age that the abuse first began was eleven years old. The mode was fourteen (18%) with sixteen children. There were fifteen children (17%) whose abuse began at age ten. The youngest age was four with one child, followed by five children indicating the abuse began at age five. The maximum age was sixteen with two children indicating that the abuse began then. A one way analysis of variance showed that there was no statistically significant difference among the counties in the age at the onset of the sexual abuse (see Table IV). For age at disclosure the mean was thirteen years old. The mode was fifteen. The statistic revealed that on the average there was a two year gap between 65 initiation of the abuse and a subsequent disclosure. There were no significant differences among the counties using one way analysis of variance (see Table IV). For forty-eight children (53%), there were more than five incidents of sexual abuse that had taken place prior to disclosure. For thirty-five children (39%) there were more than one but less than five incidents. With only 7% of the children had there been just one incident. A chi-square analysis indicated no significant differences among the counties (see Table V). There were forty-seven children (52%) who indicated that the abuse lasted less than one year.‘ Thirty-seven children (41%) revealed that the abuse had exceeded one year. There was no significant difference discovered between the counties in length of abuse using chi-square (see Table V). There were fifty-seven children (52%) who had been sexually penetrated and thirty-two (36%) had been abused by fondling. Statistical analysis using chi—square resulted in no significant differences between counties (see Table V). 66 Table V Sample Characterietice using Nominal Variables and County Counties Variables St. Joseph Kalamazoo Kent Sex ‘ Females 27% 33% 28% Males 5% 0% 4% Race Caucasian 32% 30% 28% African Americans 1% 3% 4% Year of Disclosure "90" 11% 13% 9% "91" 6% 6% 10% "92" 11% 11% 13% "93" 6% 3% 1% Number of Incidents One 5% 1% 1% Less than five 9% 13% 16% More than five 18% 16% 14% Length of Abuse Less than six mo. 15% 13% 17% 6—12 months 5% 7% 4% Over one year 15% 13% 12% Present placement* Own home 16% 18% 19% Own home with perp. 2% 0% 4% Foster home 1% 6% 3% Independent 11% 7% 0% Type of sexual abuse Fondling 9% 14% 20% Penetration 22% 18% 12% Relationship to perp.* Father 8% 9% 18% Stepfather 8% 10% 5% Mom's Boyfriend 7% 6% 8% Extended Family 1% 7% 0% Other 9% 1% 0% Note. Percentages are based on total population of sample. * Significant difference between counties as tested by chi-square. 67 In summary the county samples were very similar. The two statistically significant differences relationship to perpetrator and present placement, and their impact on analysis will be explored in the Result and Discussion Sections. Results Before the study hypothesis could be tested it was necessary to discover if there were differences in trauma scores based on what the literature defines as sexual abuse with a higher likelihood of severe trauma. The reason being that if certain variables indicative of the sexual abuse itself were the primary influencers in the trauma scores then they could skew the results of the societal system intervention variables. The primary variables that were outlined previously by Hindman and Briere for causing severe trauma were examined. The variables: younger than age twelve at the time of abuse, at least one year of sexual abuse before disclosure, parental relationship to the perpetrator, and lack of support in the response of the family following disclosure were run with trauma score and were statistically analyzed by using Pearson correlations. There was no significant association between any of the four variables and trauma scores (see Table I). To further ensure that the four variables did not significantly influence trauma score in the sample they 68 were then dichotomized so as to collapse each variable into two groups. The first group met the criteria for the most likely to be severely traumatized for each of the four variables. The second group were those that did not meet the criteria as likely to be severely traumatized. Theoretically, the children who met the criteria as likely to be more severely traumatized should have had a higher mean trauma score than those that did not. In running t-tests on each variable comparing Group 1 (the most likely to be severely traumatized) with Group 2 (the least likely to be severely traumatized) with the Trauma Symptom Checklist score, there were no statistically significant differences between the two groups. The largest difference in scores was seen in those that were sexually abused by a father or stepfather (Group 1 N=52) compared to another adult perpetrator. There was a mean difference of eleven points with a two tailed t-test probability of .067 between Group 1 which had a 56 mean score and Group 2, N=36, with a mean score of 45 (see Table VI). The children who were abused at an earlier age, before age twelve (Group 1, N=49) had a higher mean score 49, compared to those were abused after age twelve who had a mean score of 40 (Group 2, N=40). The difference was not statistically significant. With the two other variables, length of time between abuse and 69 disclosure, and family support of the disclosure, the group expected to be more likely severely traumatized had a mean score a few points lower on the trauma measure than those who were expected to be less traumatized (see Table VI). Table VI T-test for Variablee Likely :2 Cause Severe Trauma with TSC Variables M fig 1 value Sig. Family response Group 1 46 33 -.89 .37 Group 2 52 26 Removal Group 1 51 30 -.02 .98 Group 2 51 26 Age at Abuse Group 1 48 25, -1.44 .15 Group 2 56 20 Age at Disclosure Group 1 52 33 .33 .74 Group 2 48 30 HQLQI Group 1: the more likely to be severely traumatized. In summary, the statistical analysis indicated that there were no significant differences in trauma scores between the designated severely traumatized group and those that were not. The finding suggested that the variables tested that were specific to the sexual abuse itself did not demonstrate an ability to significantly affect trauma scores thereby increasing the likelihood that any significant differences discovered in trauma scores were attributable to the societal system 70 intervention variables. Original Hypotheeie The hypothesis that societal system intervention exacerbates trauma in child sexual abuse victims was tested by examining the primary events within each of the three previously defined phases of system intervention (see Figure 1). For the investigatory phase it was the number of interviews, in the court phase it was testifying in court, and it was removal from the home in the social service phase. The fourth variable in the original hypothesis, trusted relationship with a professional, weaved its way through each phase because of it being an ongoing process. The statistical analysis of the original hypothesis was done using chi-square, one way analysis of variance, bivariate correlation, and regression, depending on the type of scale determined for the variable. The four tenants of the hypothesis, number of interviews, testifying, removal, and trusting relationship were each analyzed separately and then put together to form a model, with Trauma Symptom Checklist score, the measurement of trauma, to determine if the hypothesis was supported. The mean number of investigatory interviews per child was 2.5. The mode was three interviews for thirty-seven children (41%). Twenty-four children (27%) were interviewed twice and eighteen (20%) were 71 interviewed only once. The number of interviews was determined to be interval level data. Bivariate and partial correlations were used to analyze the data for number of interviews and trauma score. The correlation between number of interviews and trauma scores yielded a p value of .007 with the level of association being .28. After determining that number of interviews was significant, partial correlation coefficients were then computed to discover if there was a spurious or interactional relationship with another variable. The interval level variables controlled for included age, age at disclosure, year of disclosure, and testifying. No variables significantly affected the correlation between number of interviews and trauma score. The analysis demonstrated that there was a statistically significant relationship between the independent variable, number of interviews, and the dependent variable, trauma, as measured by the Trauma Symptom Checklist. In the legal phase, thirty children testified in either the criminal and/or juvenile court. Testifying was defined as interval level data as the children were scored from "0" meaning no interviews to "2" for those children who had to testify twice. The correlation between trauma and testifying was .03 with p = .73. Partial correlations were not run due to the lack of significance in the bivariate correlation. From the 72 analysis, the conclusion reached indicated that the null hypothesis was supported and that testifying did not make a difference to the children in elevating trauma. The third variable in the hypothesis was removal from the home which was the primary variable in the social service phase. For this variable children were classified into two groups and therefore it was considered as nominal data. Group 1 included the thirty-six children (40%) that were removed. Group 2 was comprised of the fifty-four children (60%) who were not. Chi-square was run and Eta yielded an association of only .002 between removal and trauma score. The analysis supported the null hypothesis that there was no significant difference in trauma scores between those who were removed and those that were not. The final variable that comprised the original hypothesis was the impact of a trusting relationship on trauma. Question 22 from the investigator's questionnaire specifically addressed the issue and the responses were determined to be ordinal data as "1" indicated no trust in a professional, "3" some trust in a professional, and "5" very much trust in a professional. Thirty-seven children (41%) indicated that they trusted a professional "very much" and thirty— six participants (40%) indicated that they trusted a professional "some". There were seventeen children (19%) who did not trust any professionals. 73 The impact of trust on children was dramatically reflected in the mean scores of the Trauma Symptom Checklist for the three groups. The thirty seven children who trusted "very much" had a mean score of 44, S2 = 24, compared to a mean score of 64, SQ = 37, for those children who trusted "none". The children who were able to trust "some" had a mean score of 52, gm = 24. The results of the one way analysis of variance on the three groups yielded, 2 = 3.21, p = .04. Post hoc tests were then run using Scheffe which indicated a significant difference between the trusting "very much" group and the "none" group. The Spearman correlation coefficient revealed an association value of -.24 and it was significant at the .02 probability level. The statistical tests supported the hypothesis that children who did establish trust with a professional had a likelihood of significantly lower trauma scores than those that did not. A model was constructed that consisted of the four primary independent variables: number of interviews, testifying, removal, and trusting relationship. Multiple regression was run on the model. The overall multiple R was .35 with the F value at 3.03. It was statistically significant at .02. Number of interviews (3 = 2.4, p = .01) and trusting relationship (e = -2.0, p = .04) continued to be statistically significant revealing that each contributed uniquely to predicting 74 trauma scores and thus supported portions of the investigator's hypothesis that certain societal system interventions can be predictors of trauma level. Other system interventions that were viewed as potentially less stressful than the primary variables defined above were also explored to discover their ability to affect trauma. The intervening variables previously defined in the Measurement section (type of court involved, disposition of perpetrator, and medical exams) were statistically analyzed as potential trauma inducers. One way analysis of variance was run on each of the variables with the Trauma Symptom Checklist but none proved to be significant. The most definitive difference was between those children who were involved only with juvenile court who had a mean score of 44, SQ = 27, and those that were involved with both criminal and juvenile who had a mean trauma score of 56, SQ = 30. A t-test was performed between the two groups but there was no significant statistical difference. The Intervention Stressor Inventory, as previously discussed, obtained a composite stressor score for each of the three phases of system intervention plus a total intervention score. It's purpose was to determine if higher stress scores in the three phases of system intervention obtained in the Intervention Stressor Inventory correlated with higher trauma scores. Bivariate correlation was run and Pearson's R showed no 75 significant correlations for any of the three phases. The highest correlation was .16 in the investigatory phase but the p value was .14. The total Intervention Stressor Inventory score was not significant as it had an association of only .05 with a probability of .639. The statistical analysis indicated that the Intervention Stressor Inventory was a poor predictor of trauma scores despite its claim to define the apparent stress of system intervention on children. These results will be examined more fully in the Discussion section. Second Hypothesis The second hypothesis, that there would be statistically significant differences in trauma scores between counties, was predicated on the belief that the system interventions were markedly different among the three counties. The validation of the second hypothesis was tested by using one way analysis of variance, the Kruskal-Wallis H test and Mann-Whitney test. The latter two are nonparametric tests that were used to further test for possible differences between counties. There were no significant mean differences amongst the counties in any of the six subscales in the Trauma Symptom Checklist: anxiety, depression, posttraumatic stress, sexual concerns, dissociation, and anger. The largest difference between counties in mean scores was two points. The posttraumatic scale had the highest 76 mean at 11.3 and the sexual concern scale had the lowest score of 5.2. The mean for the total trauma score was 51.7 for the entire sample. There were no statistical significant differences between the counties using one way analysis of variance. St. Joseph County had a mean of 50.4, SQ = 27, compared to Kalamazoo County which had a mean of 51.5, SQ = 26, and Kent County with a mean score 53.2, SQ = 31. An attempt was made to look further at the breakdown of trauma scores by county by moving from an examination of the mean to the median. The median for St. Joseph was 46.5, whereas Kalamazoo's was 49.5, and Kent was at 54. The difference between the mean and the median can be explained by the outliers and the range difference between counties (see Figure 11). The examination of the medians indicated differences between counties, although they were not statistically significant when the Kruskal-Wallis H. test was performed. In addition the Mann-Whitney test did not yield a significant difference. The findings supported the null hypothesis but called into question the possibility of type I error. The conclusion from the statistical analysis was that the county of residence was not a significant factor in determining the level of trauma to the child and therefore the null hypothesis was supported. 77 Figure II O“ 1% 014 a'o human: 160 140+ 13H 28m .88 .xooso 2:38. COUNTY 78 The statistical use of the second instrument, the Child Behavioral Checklist, for support of the findings of the Trauma Symptom Checklist was problematic. The primary difficulty was that the test's scoring criteria divided the sample into three disproportionate groups which subsequently greatly reduced the ability to run statistical analysis. The largest of the three groups were girls ages 12-16 with thirty—five. The second groups of girls ages 8-11 contained only nine, and there were only three boys that were scored on the Behavioral Checklist. The low numbers within groups were even further decreased when the counties were compared to one another. St. Joseph County had only fourteen in the girls 12—16 group, Kalamazoo County had twelve, and Kent County had only nine. In the girls 7-11 group, Kent County had the most with five, and Kalamazoo County and St. Joseph County had two each. As a result of the low numbers the best statistical analysis was run not by comparing counties or other variables with the Child Behavioral Checklist but rather by comparing the social and behavioral scores to national samples. Such an examination, although not providing useful between groups data, produced valuable information as to the characteristics of the present sample. The total social competence score for all the girls was a T—score of 35 with a raw score of 15.1. The score 79 was almost identical to the national clinical sample of 36 T-score and 15.4 raw score compared to the non- clinical that had a 50.9 T-score with a raw score of 20.8. The subscales for the sample for social competence (i.e. activities, social, and school) were each within one point of their respective categories for the clinical national sample. The total behavioral T-scale score for the girls 12-16 was higher than the national clinical sample. The girls in the sample had a mean T-score of 72.6 compared to 68.5 for the national clinical sample and 50.6 in the non-clinical sample. In the two major subscales, internalizing and externalizing, the girls 12-16 scored slightly lower than the mean of the national clinical sample. The internalizing scale had a T—score of 61 compared to 64.3 of the national clinical sample and 49.8 of the non-clinical sample. The girls 12-16 had a 60 T-score in the externalizing scale compared to a 64 national clinical score and a 49.4 non-clinical sample. The nine girls in the 7-11 range were also found to closely mirror the national clinical sample. The internalizing mean T-score was 63 compared to 67 for the national clinical sample and a 51.3 for the non-clinical sample. The externalizing T-score was also a 63 which compared to a 68 on the clinical sample and a 51 on the non-clinical sample. The total behavioral score was higher at 75 than the national clinical mean of 68.5 and 80 a non-clinical finding of 50.6. The overall statistical pattern comparison with the national sample mirrored for the girls 7-11 the pattern for the girls 12-16. Comparisons for the boys with the national sample were not done given the sample of only three and the fact that one of the boys was within a different age grouping. An attempt to learn if significant differences existed in girls 12-16 between counties was made by running one way analysis of variance. There was no significant difference between counties among any of the social or behavioral scales in the Child Behavioral Checklist. No statistical analysis was run on the girls 7—11 or boys due to the small sample size. Exploration to determine if there were significant sample differences between counties was necessary in order to test if non—system variables skewed the county results and increased the likelihood of Type I error. There were two significant differences in sample characteristics noted between the three counties, the present placement at the time of the research interview and the relationship to the perpetrator. No significant difference was found however, between the varying present placements and trauma scores when analyzed using one way analysis of variance. Due to the statistical finding on present placement and trauma score it is unlikely that present placement could in any way have 81 skewed the results between county and trauma scores. The statistically significant difference noted between counties in regards to the relationship to perpetrator was likewise analyzed with trauma score using one way analysis of variance. The outcome was that there were no significant differences in trauma scores among the children's varying relationships to the perpetrator. It was therefore determined by the investigator that the significant differences noted in the relationship to the perpetrator among the counties was not a factor that influenced trauma scores among the counties. To investigate if there were differences between counties samples as to the likelihood of being more severely traumatized due to the sexual abuse itself, the variables: age at abuse, length of abuse, relationship with perpetrator, and length of time from abuse to disclosure, were compared for each county. Using chi- square no significant differences were discovered between counties in any of the four variables. Further a frequency table yielded similar numbers within each county of more severely traumatized children. From this information the conclusion was drawn that any differences in trauma scores between counties were due to the societal system interventions and not the sexual abuse experienced by the children in the sample. The investigator also examined the primary system 82 intervention variables to discover if there were significant differences between counties in the system interventions themselves. It was believed that such information would be important in determining if there actually were system differences as the researcher had anticipated. Analysis was also done on the system interventions to discover whether there were methods within a particular county that reduced interviews and thereby served to help minimize further trauma for child victims. Among two of the four primary independent variables of system intervention statistically significant differences were noted between counties. There was a significant difference between Kalamazoo County and the other two counties in number of interviews using one way analysis of variance, E = 10.09, p = .0001, and the post hoc test, Scheffe. The mean in Kalamazoo County was 3.1, SQ = .87, compared to Kent County at 2.4, SQ = .77, and St. Joseph County with a mean of 2.0, SD = 1.1. There was also a significant difference between the counties in testifying as revealed in one way analysis of variance, E = 5.3, p_3 .008. The Scheffe test indicated significant differences between St. Joseph County and Kent County. The mean for St. Joseph County was .13, SD = .34, compared to Kalamazoo which was at .53, SQ = .68, and Kent at .67, SQ = .88. There were only four children in St. Joseph County who testified, 83 whereas there were fourteen children in Kalamazoo County and twelve in Kent County. There was no statistical difference between counties in removals as tested by one way analysis of variance. The highest number of children, twenty-one (23%), remained within their own home in St. Joseph County. Kent County had eighteen (20%) who remained home while Kalamazoo County had fourteen (16%). There were twenty children (22%) who had been removed over one year, with Kalamazoo County having eleven, Kent County five, and St. Joseph County four. There was no significant differences noted in trust between the counties as tested by chi-square. The numbers were nearly identical between counties as 20% in Kent County, 20% in Kalamazoo County, and 17% in St. Joseph County had no trusting relationship. The largest difference between counties was noted in those that trusted some as 57% of the respondents fell into this category in Kent County, while only 33% in St. Joseph County, and 30% in Kalamazoo County were in the somewhat trusting. Four possible intervening system intervention variables were also examined to determine if there were significant differences between counties in these areas. Involvement in both criminal and juvenile courts occurred with 43% of the total sample. Criminal court was involved with 40% of the children, and juvenile 84 court was involved with 17% of the total sample. There were minor differences between counties but not enough to indicate statistical significance. Medical exams following sexual abuse disclosure, were examined as an intervening variable because of the potential for intrusion and perceived violation of the child's body by the physician. Exams occurred 52% of the time for the children. The counties were nearly identical in the number of medical exams that were required. There was a rather large difference between counties in what happened to the perpetrator. Over 55% of the perpetrators plead to some type of criminal sexual conduct. In St. Joseph County 76% of the perpetrators plead, compared to 53% in Kalamazoo County, and only 37% in Kent County. Fourteen percent of all perpetrators were convicted through a trial and there were 7% who were acquitted. In 16% of the cases there were never any charges filed. Of these, 8% were from Kent County, 7% from Kalamazoo County, and 3% from St. Joseph County. Despite the observed differences between counties, chi-square revealed no statistically significant differences. Counseling was a variable that was thought to be positive because of its ability to minimize or ameliorate trauma due to its potential to provide a positive relationship, a safe environment, and an 85 opportunity for sharing of thoughts and feelings. Over 54% of all the children underwent counseling that exceeded one year. There was a large difference between the counties, as 87% of the total sample in Kent County had counseling over one year, compared to 53% in Kalamazoo County and only 30% in St. Joseph County. Twenty-three percent of the children in St. Joseph County did not have any counseling. The differences between counties were statistically significant using chi-square as indicated by Phi (.36, p=.002). The differences between counties would appear to indicate that Kent County and Kalamazoo County provided more supportive services to help minimize the trauma of sexual abuse and the subsequent stress and possible trauma of societal system intervention than St. Joseph. The Intervention Stressor Inventory score was also used as a measure to compare counties system intervention stress. The total score for investigatory events between the three counties was not significant using one way analysis of variance and the post hoc test, Scheffe. The mean for St. Joseph County was 105, SQ 65, compared to Kalamazoo County which had a mean of 138, SQ = 57, and Kent County which had a mean of 128, SQ = 56. For the court phase, using the Intervention Stressor Inventory, there was a statistical significant difference using one way analysis of variance (E = 4.49, 86 p = .014) and the post hoc test, Scheffe, between St. Joseph County and Kalamazoo County. Statistical analysis indicated a mean of 42, SQ = 81, in St. Joseph County, a mean of 112, SQ = 144, in Kent County, and a mean of 138, SQ = 148, in Kalamazoo County. The third phase, social service intervention, was consistent in following the pattern of the previous two phases with St. Joseph County having the lowest mean score at 44, SQ = 68, followed by Kent County at 65, SQ = 65, and Kalamazoo County with a mean of 83,_SQ = 89. There was not a statistically significant difference between the three groups. An analysis of the cumulative total for all three phases revealed a significant difference among the counties. The one way analysis of variance test indicated, E = 6.95, p = .002. The Scheffe post hoc test showed significant differences between St. Joseph County and Kalamazoo County. The mean for St. Joseph County was 191, SQ = 148, for Kent County it was 310, SQ = 212, and for Kalamazoo County the mean was 370, SQ = 202. The findings supported the investigator's expectations that there would be a significant difference between counties in societal system intervention stressors, with St. Joseph County being the least stressful system followed by Kent County and then Kalamazoo County. Despite the statistical findings in societal system interventions however, the trauma 87 scores were not statistically different as hypothesized. This finding will be discussed in depth in the Discussion section. CHAPTER V DISCUSSION The hypothesis, that societal system intervention exacerbates trauma, was not fully supported as a result of the statistical findings documented in the Results section. The findings did support however, two of the tenants of the hypothesis. Interviews The number of interviews was found to be significantly related to the Trauma Symptom Checklist scores. The level of association .28, was significant and was the strongest of any of the independent variables with trauma score. The level of .28 is not extremely powerful, yet in this study it is an important finding and gains strength when the numerous variables regarding the experience of sexual abuse itself and the societal system interventions are considered. Any one system variable can not be expected to be the dominant factor in the causation of trauma. Societal system interventions are secondary trauma agents, not the primary agents, the sexual abuse itself and the family dynamics that surround the disclosure are the most powerful variables of trauma. Therefore, a correlation of .28 gives substantial weight to the impact of the 88 89 investigatory interviews on trauma. Several questions on the investigator's questionnaire attempted to gain a further understanding of why and how children disclose sexual abuse and their experience of the interview process. The value of such information is that it provides possible explanations as to why the interview process was so difficult. Further, the children's answers furnish professionals with insights which can assist in developing interview intervention strategies that best respond to the needs of children as defined by the victims themselves. Twenty-nine percent of all the children stated that they first disclosed to their mother with just 2% reporting to a non—offending father. Of the remaining 69%, 27% indicated that they told a friend initially. Thirteen percent revealed the sexual abuse to a school professional. Another family member was initially confided in by 11% of the children. The reasons as to why disclosure occurred was answered by seventy-seven children and evenly distributed among four responses. It is important to note that the disclosures for a large majority of the children occurred only after they had endured the sexual abuse for an extended period of time, as the mean length between the sexual abuse and disclosure was two years. Twenty-two percent of the respondents stated they knew it was wrong and so decided to tell. Eighteen percent 90 of the children didn't want the sexual abuse to keep happening. Another 16% decided to disclose because they were angry at the perpetrator for what had occurred. Fourteen percent disclosed because they found a person with whom they could entrust their secret. The responses to, "how did the system get involved" support the findings that in the majority of cases parents were not the first people that the system had contact with because the children were not likely to disclose to them. Only thirty mothers (33%) and one father initiated system contact. The other fifty-nine children (66%) became involved with the system due to a contact made outside the parent. Twenty-eight percent of the children reported that the school initiated system contact. Relatives reported to the appropriate authorities for 13% of the children. The children's responses articulate the necessity for immediate societal system intervention. In less than one-third of the cases did the parent either receive the initial disclosure or make the first contact with the system. School personnel and friends comprised over one-half of those first told and who then made the initial system contact. This information indicates the majority of these sexually abused children looked to persons outside their parents for assistance when they decided to "cry for help". The finding emphasizes the critical link to outside systems that children must have 91 because of an unwillingness to disclose to the non- offending parent. The information speaks to the demand for education of school personnel and children on the appropriate response to a child who confides that sexual abuse has occurred. It provides professionals who interview children the knowledge that parents in the majority of cases may not be aware of the sexual abuse prior to system contact. Further, it indicates the need for professional intervention in bridging the gap of information between non-offending parents and children. The finding also heightens the responsibility of the system professional to make a careful assessment of the parental response following disclosure to ensure the physical and psychological protection of the child. The children's reactions to repeated interviews varied. Of the sixty-five children (72%) who had more than one interview, twenty five (38%) of the respondents stated that being interviewed more than once was "harmful" to them. Some of the comments that supported the "harmful" response included: "I felt victimized because so many people kept asking me the same questions over and over."; "I felt like I wasn't being believed because they kept asking me the same thing"; "I didn't want to keep talking about it, but they made me." There were sixteen children (24%) who reported that 92 being interviewed more than once was helpful to them. Some of these responses included: "It helped me to get it out of me to keep talking about it." "It didn't hurt so much after I kept having to talk about it." There were twenty-three children (26%) who reported that more than one interview had "no effect" on them. One way analysis of variance was run on this question with total trauma score used as the dependent variable to learn if there were significant differences between those who experienced repeated interviews as harmful vs. those that experienced them as helpful. Those that defined repeated interviews as harmful had a mean score of 55, SQ = 27, compared to those who stated that the interviews were helpful who scored 51, SQ = 33. Despite the 4 point difference, there was no statistically significant difference. Important to the development of alternatives in interviewing were the responses solicited from the videotaping questions. Twenty nine children (32%) were videotaped and sixty-one (68%) were not. There was a significant statistical association between county of residence and the experience of a videotape interview when chi—square was run with p = .000 with a value of .72. In St. Joseph County twenty-six children, 86% of the St. Joseph participants, were videotaped compared to Kalamazoo County and Kent County where just were 6% of 93 each counties totals were videotaped. There was no statistical difference in the children who were videotaped and those that were not when analysis of variance was run with the Trauma Symptom Checklist scores. There was however, an association of .32 (p = .003) between number of interviews and videotaping. This statistically significant association identifies videotaping as a potential mechanism for reducing the number of interviews and thereby decreasing the likelihood of further trauma through system intervention. Concerns regarding videotaping having a possible negative impact on children due to the intrusive nature of the camera were not confirmed. Children's perceptions of videotaping were primarily either "helpful" or "no effect". Fifty-three percent of the children stated that knowing the interview was being videotaped had no effect on them. Thirty percent of the children viewed videotaping as "helpful". The most common comment of those defining it as "helpful" was, "I didn't have to keep telling what happened to me." Another statement frequently made was, "It helped me because I only had to talk to one person and it was over." Thirteen percent of the children videotaped experienced it as "harmful". The primary reason given by three of the four children was, "it was embarrassing." The supportive and neutral responses 94 (83%) of the children on videotaping far outweighed the few negative perceptions and illustrate that most of the children saw videotaping as an intervention that was of potential benefit to them. An examination of the interview process through the question, "what did the interviewer do to make it easier for you" elicited several positive responses. Twenty- six percent of the seventy-six children who responded to the question, made the general statement that the interviewer was "nice" to them. Seventeen percent of the children stated that the interviewer was "patient" and that made the interview easier. The fact that professionals were willing to listen was of primary importance for 13% of the children. In contrast 17% stated that the interviewer did nothing to make the interview‘easier. The answer to the question, "Who, among the professionals, was it easiest to tell" was almost equally divided amongst fifty-one respondents. Twenty- seven percent of those that responded identified counselors as the easiest, this included both school and clinical counselors. Twenty-five percent thought that the child protective services workers were the easiest people to tell, 22% defined the police, and 22% indicated the prosecuting attorney. This finding is contrary to the Intervention Stressor Inventory which weighted police and prosecuting attorney interviews as 95 more stressful than protective services interviews. The children's responses indicate that among the system professionals there were similar abilities to establish rapport and diffuse anxieties. The children were emphatic in labeling the most difficult part of the interview process. Eighty-four percent of the total sample indicated that having to "tell the details" of what happened to them was the most difficult. The significance of the response is heightened by the fact these were spontaneous responses to an open ended question. The children overwhelmingly confirmed by their response, how painful and anxiety provoking the details of their stories were. Such a significant response needs to be translated into a directive by professionals. Gaining the child's "secret" must be pursued with the utmost sensitivity, respect, and support if fear is to be dispelled and the possibility of further trauma minimized. Supportive data as to the difficulty of disclosure was also affirmed in the responses to the question, "what was the most difficult part of all the system interventions?". This question was asked at the end of the investigator's interview as a summary question. Thirty-seven percent of the children, two times the number of children in the next closest grouping, responded by stating the "initial interview". One child described being interviewed as "a very scarey process." 96 This response seemed to capture the fear of the 84% who had to "tell the details" of their sexual abuse. It is evident that the children did not want to recount their embarrassing secrets to strangers. The system however, most often responded to such painful disclosures, by demanding that the children keep repeating their secrets at the time and direction of the professionals. The children had no power to control the process of interviewing without forfeiting their safety. This put children in the untenable position of having to tell repeatedly about their victimization in order to remain safe. Theoretically, most children feel responsible for their own abuse (Hindman, 1989; Finkelhor & Browne, 1985). Demanding that children continually repeat their abuse stories connects children with painful memories and reinforces the internalization of guilt and shame experienced in sexual abuse. The potential harmful outcome is a solidification of stigmatization and dissociation from memories and affect. The ability to develop trust in professionals is undermined when children are continually asked the questions that were previously answered. Betrayal by the "helping" professional serves to exacerbate the previous traumatic betrayal by a significant other in sexual abuse. The children's negative responses in this study to the interview process and the Traumagenic theoretical 97 model from which to interpret the data assists in clarifying why the number of interviews was found to be significant in determining trauma score. Most of the children did not want to have to continually repeat their painful secret. Research outlined by Tedesco (see literature review) supports the negative impact that interviews have on trauma. The implications for professionals who engage in the interviewing process with children are evident. Professionals need to be cognizant of just how painful and difficult disclosures are for sexually abused children and provide the necessary support to minimize their fears. It is also a necessity to develop methods to reduce the number of interviews, i.e. videotaping. The children's responses to videotaping suggest that it is a child supportive intervention that can reduce interviews and thereby minimize the likelihood of further trauma. TEES: A second tenant of the original hypothesis, that trust in a professional decreases the likelihood of trauma, was statistically supported (see Results). Finkelhor's theory on traumagenic dynamics provides the explanation for this significant finding. When trust is established it minimizes betrayal and consequently reduces the likely of further trauma to sexually abused children. The lessening of betrayal helps empower 98 children because the relationship with a trusted adult provides safety and security in an otherwise unsafe environment. The establishment of a secure environment provides an opportunity for children to develop the belief that the world can be a safe place. The consequence is a reduction in children's hostility and more affective healing. When children are listened and responded to, they become empowered within their world. The system cannot change the previous betrayal and powerlessness induced by sexual abuse, as the clinical scores from the Child Behavior Checklist indicate, but it can help to minimize trauma as indicated by the lower trauma scores of the children who developed trust. The findings from the investigator's questionnaire help confirm the positive impact that trust can have on sexually abused children. Being believed by a professional at the crisis time of disclosure is critical if trust is to be established. Following initial disclosure most of the children perceived that the professionals were supportive and believed their stories. Sixty-four children (71%) felt they were believed very much and another twenty one (23%) said that they felt they were believed some. Only five out of the ninety (5%) felt that they were not believed. The children's perceptions as to why they were believed were divided into three primary categories. 99 Twenty four of the children (27%) stated that the professionals acted like they believed their story. Twelve of the children (13%) indicated that the professionals told them that they believed what they said and another ten children (11%) felt believed because the professionals listened to what they had to say. When comparing the trauma mean scores between those that felt believed, those that felt believed some, and those that did not feel believed, a pattern was seen that was similar with the finding on trusted relationships. The mean score for the 70% who felt they were believed "very much" was 48, SQ = 25. Those 23% who felt they were believed some had a mean score of 61, SQ = 34. For the 5% who were felt they were not believed at all the mean was 60, SQ = 27. Due to the fact that the "N" was too small (N=4) for those that did not feel believed to gain any statistically significant findings, they were combined with those that felt that they were believed just some to form just one group. A t-test was then run between those that felt they were believed very much and the second group. The "T" statistic was significant, (e =-2.03, p = .045). The finding strongly supported the differences discovered in mean scores for those who trusted a professional and those that did not. It also served to add considerable weight to the importance of 100 professionals communicating a genuine acceptance of sexual abuse disclosures by children. Thirty percent of the total sample cited their counselor as the most trusted person, 16% stated the prosecuting attorney, and 11% indicated their social worker. The distribution speaks to the importance of having various professionals interface with children in order to provide different opportunities for children to develop a relationship with a professional. The responses as to why trust occurred with a particular person appeared to be predicated on a caring attitude by the professional, a willingness to listen to the child, and the frequency of contacts with the child. The responses to the question on the value of having a trusted professional reinforced the importance of trusting relationships for children. Ninety percent of the seventy—seven children who responded to the question indicated that having a trusted professional was a "great help". Only one child indicated that it was "no help" and just 6% stated that it was “some" help. Although the sample size prohibited mean comparison, it is evident by review of the percentages that a relationship with a professional provided much support and assistance to the children. It again speaks to why there was a significant difference in trauma scores between those children who trusted a professional and those that did not. 101 The long term impact of system intervention on the children's willingness to trust in the future was also examined. The results were consistent with findings that emphasized the powerful role trust played in impacting trauma scores. Fifty-six percent stated that they were more trusting of others as an outcome of system intervention, compared to 26% that stated system professionals had no effect on their future willingness to trust. Eighteen percent indicated that they were less trusting. When these responses were run with trauma scores using one way analysis of variance, a statistically significant difference was observed with the F being significant (E = 3.33, pm = 04). The "more trusting" group had a mean score of 49, SQ = 27, compared to those in the "no effect" group with a mean score of 48, SQ = 27. The "less trusting group" had a mean of 68, SQ = 27. The responses of the children regarding why their willingness to trust had increased clustered around adult responsiveness to their concerns and fears. Sixty-seven percent of the children who responded to the question stated that the professionals "did what they said they would" which led to their increased trust. In contrast however, thirteen percent reported they were now less trusting because they felt "lied to" by the professionals within the system. The findings on trust from the investigator's 102 questionnaire direct system professionals to concentrate on establishing rapport and trust as the key component in any system intervention. Professionals need to be willing to intervene in ways that respect the integrity of children, in the eyes of children. Making promises to pacify children's fears that can't be ensured by the professional do more harm than good because ultimately trust in the professional is undermined. Gaining a child's disclosure must be precipitated honestly, otherwise the child will likely experience betrayal which elevates the likelihood of further trauma. The establishment of trust as a primary goal of societal system intervention is crucial to the support of sexually abused children. Testifying The data failed to support the portion of the hypothesis that children who testified would have higher trauma scores than those that did not. The explanation as to why the scores were not significantly different can be found through the information gained from the researcher's questionnaire and Finkelhor's model of sexual abuse. Many children who responded to the questions on testifying indicated that there were some very helpful processes that empowered them and served to greatly minimize the potential trauma. Preparation for court 103 testimony, having a trusted person, and the responses of the people within the courtroom all appeared to serve as key factors in empowering children to testify. Only 34% of those children who testified found it "harmful" compared to 48% who found it "helpful". Fifty-nine percent of the children felt that testifying helped build more trust in the professionals that were involved, compared to only 1% who found it harmful to their trust in professionals. The children communicated an acute awareness of the reactions of professionals and juries to their stories. The reactions, as perceived by most of the children, were positive, which appeared to decrease the stigmatization of testifying and increase self esteem. Sixty-nine percent of the children felt better about themselves after they were finished testifying. The foundation for the unexpected positive outcome in testifying appeared to be the support they received from a trusted adult. The relationship offered security and safety which minimized the powerlessness created by testifying. Over 96% of the children stated that having a trusted person with them when they testified was helpful. The children confirmed the importance of these relationships when they identified "professional peoples' support" as the primary helpful part in going to court and testifying. Despite testifying not being validated as a trauma 104 producing agent, there are still many questions about its usefulness to children. The children who testified did not articulate an appreciation for finally getting their day in court. Some of the literature supports children testifying as a method of giving them an opportunity to finally tell their story publicly, thus making the experience therapeutic. The findings did not negate this possibility but there was little information suggesting that "telling their story" in court was helpful to them. Only one child indicated in response to the open question that the experience was helpful because, "I finally got to tell my side of the story." The discovery that testifying did not elevate trauma scores does not negate how difficult the experience was to many of the children. The majority of the children when asked how testifying affected them indicated they were either scared or felt like they were on trial. The most fearful experience was generated by having the perpetrator in the courtroom when the children had to testify. Of the twenty-six children who testified with the perpetrator in the courtroom, 90% stated that the "perpetrator scared them and they didn't want him" in the room. This finding was further supported when the children were asked how testifying could have been made easier. Sixty-two percent identified not "having the perpetrator in the courtroom" as the most important change that would have made 105 testifying easier. It is evident that professionals play a significant role in determining the impact of testifying on children. The recognition of how trust affects children lays the foundation to minimize the potentially traumatic impact of testifying. The results of the study indicate that the hostile courtroom environment creates fear in children; however, the effects of this can be greatly reduced when children have relationships with adults that they trust and efforts are made to prepare children for what occurs during testimony. This explanation would appear consistent with the study done by Runyon (1988) which discovered that children who testified in juvenile courts in North Carolina with adequate preparation and support demonstrated improvement in their mental health. Removal Seem meme The final tenant of the original hypothesis, removal of the child from the home, was likewise not confirmed. The explanation as to why rests in how the children often perceived the experience of removal. In order to understand the children's perceptions of removal it is first necessary to explore the family's response to the sexual abuse because most often the family response determined the significance and intensity of the social service interventions. Eighty-two percent of the sample indicated that 106 they were believed by someone in their family compared to 18% who felt that they were not believed by their family. The investigator's expectation, supported by the theories of Briere, Hindman and Finkelhor, was that the children who were not believed by their family would have significantly higher trauma scores. Yet the trauma mean scores for those children who were not believed by a family member was slightly lower at 46,_SQ = 33, than those who were believed by their family who had a mean of 52, SQ = 26. When the believing category was specifically delineated to indicate who actually believed, some light was shed on why the scores for those who believed may have been higher. (see Table VII). Table VII Who Believed Disclosure mm the Family with TSC Responses m SQ Cases Everyone 48 23 32 Mother 53 30 20 Grandparent 57 29 8 Sibling 64 28 7 No one 46 33 16 Father 55 26 7 When "everyone" believed, the mean trauma score was at its lowest level among those that believed. As the number of family members believing in the child's story decreased, the trauma score elevated. It would appear from these results that the greater the number of people 107 who were supportive, the less trauma experienced. This still however does not explain the lowest score which is within the "no one" believed category. One possible explanation is that these children had internal strengths for support that minimized the impact of their families not believing. Within the family the most supportive according to the children were their mothers, yet only forty-seven of the children (56%) reported this. Forty-three children (48%) had to go outside the primary caretaker to receive support after their sexual abuse disclosure. These numbers are consistent with the question (see Table VII) which indicated that approximately 58% of the mothers' believed their children after disclosure. The findings reveal that only one out of two mothers were likely to believe and support their children in what would appear to be the time of their most critical need. This is a disturbing statistic as the psychological impact of mothers' responses are normally the most significant of any relationship children have (Hindman, 1989). The support void consequently experienced by many of the children demands recognition by professionals. Efforts need to be made to mediate betrayal, encourage grief expression, create safety, and provide support. The data on family support helps create an understanding about why the process of removal did not increase the likelihood of higher trauma scores in 108 removed children. The fact that over 40% of the non— offending parents did not believe their children makes it likely that removal offered the 40% who were removed from their home an environment where they were believed by an adult and protected from harm. It helps clarify why many of the children viewed removal as "helpful". Forty-one percent of the children who were removed stated that the removal was helpful and another 24% felt it had no effect on them. The trauma scores reflected the perception of removal as helpful vs. harmful as there was a 20 point differential between mean scores, with those viewing removal as helpful having the lower trauma score. A T-test was run between the two groups with e = 1.99, p = .056. The near statistical significance supports other statistical findings that when children viewed stressful system events from a positive perspective there was a high likelihood of having lower trauma scores than those who experienced the events as negative. Safety from the perpetrator was cited 60% of the time as the reason removal was helpful. Personal safety is a key issue for all people and it is especially critical for child sexual abuse victims who have been entraped and powerless within their environment. It appears that within their home most of the removed children had no power to ensure their safety and felt betrayed by the failure of others within their home to 109 protect them. The removal likely served as an escape to safety, despite the temporary loss of family. The safety gained appeared to outweigh the temporary loss of significant others. This is an important awareness for professionals who are under the mandate to keep families together. It may even be more harmful to leave children in their home at times, according to children who have been removed, than to remove them to secure their safety. This doesn't negate that separation from family is very difficult, as 70% of those that were removed cited this reason when asked, "what was the most difficult part of removal?". The children, in their responses to questions on removal, spoke of mediating factors that influenced how the removal was experienced. Children clearly stated they should be informed and listened to by system professionals at the time of removal. Seventy-six percent of the removed children indicated that removal could have been easier if professionals would have talked to them more, listened to what they had to say about removal, and prepared them more for the removal. These responses emphasize the need to empower children through gaining their input and offering explanations. Empowerment occurred, according to the removed children, when professionals listened and responded to what they had to say about safety within their family. Children felt empowered when they received information 110 about what was happening in the system process that affected them. The responses of the children dispel the myth that children should be protected from knowing what is happening within professional arenas when it pertains to them. Protecting children, according to the children in the sample, is not about withholding information, but rather alerting them to what is occurring so that they can have input into decisions and prepare themselves for what may happen. Overall perceptions Before concluding the discussion on the impact of system interventions, an examination of additional information gained through the investigator's questionnaire is necessary to assist in clarifying the factors that the children defined as important in societal system intervention. Their perceptions provide a succinct message to professionals regarding what works and what does not for children interfacing with the system. Counseling was an intervention that received strong support from the children. Ninety-three percent of the children who participated in the study were engaged in counseling for some period of time. Sixty-four percent were in counseling that exceeded six months. The outcome of counseling was defined by 66% of the children as "helpful". Only one child felt that counseling was 111 "harmful". Twenty-six percent of the children indicated that counseling had "no effect" on them. Thirty-eight percent of the children thought that counseling was helpful because they could "talk about feelings". Another 17% felt that counseling was critical in helping them get through all that had occurred in the aftermath of disclosure. In contrast, 8% felt that the counselor did not listen to what they wanted to talk about and another 7% felt that they couldn't trust the counselor. A one way analysis of variance test was not used to statistically compare the counseling scores. The reason was that the sample size for those who did not go to counseling (N=6) was too small to statistically compare with those who did. Consequently it was not possible to statistically determine the impact of counseling on the children. However, it would appear from an examination of the positive responses to counseling that it most often provided an important and useful service to children. A Spearman correlation was run on counseling and trauma scores but proved not significant with p = .29 and an association of .11. An explanation as to why counseling was not statistically significant is that since 93% of the sample attended counseling there was no way to determine its impact because of a lack of a comparison group. The question, "What was most difficult part of the system interventions after disclosure?" yielded eleven 112 different responses from the children (see Table VII). Table VIII Most Difficult Part e: the System Interventions Responses Frequency Percent Initial interview 33 37% Response of family 14 16% Testifying 11 12% Removal from home 9 10% Think about all 7 8% that happened Having to tell people 5 5% Perpetrator went to jail 4 4% Note. Percentages based on responses to question. The fact that 37% felt that the initial interview was the most difficult, which is twice as many children as those who felt that the response of their family was most difficult, and three times as many as those who responded with testifying, weights the difficulty of the interview process as far exceeding any other societal system intervention. This information helps support why repeated interviewing was statistically significant in determining trauma scores. In contrast, the most helpful intervention was counseling with 29% of the children reporting this. Closely associated with the counseling responses were 21% who stated that, "people in the system helping me" was most useful. Both of these responses appear to demonstrate the need for a positive relationship with a professional which directly supports the statistical 113 finding that a trusting relationship reduces trauma score. (see Table IX). Table Ix Most Selpful Part 9; System Interventione Responses Frequency Percent Counseling 26 29% People helped me 19 20% Perpetrator in prison 13 14% I am safe 10 11% I got through it 6 7% Got feelings out 5 5% People believed me 4 4% Note. Percentages based on responses to question. Twenty-eight percent of the children felt that the professionals could have done nothing more to help them (see Table X). When this response is combined with the two answers above (counseling and people helped me) the result is that 78% of the children felt they had received strong support from professionals. However, 27% of the respondents communicated that they wished that "the professional people would have listened to them more." The negative responses instruct professionals as to the necessity of two-way communication in ensuring that children, who are the focus of the societal system intervention, are given a voice in determining what happens in the system process. 114 Table x What Could Have Made System Intervention Easier? Responses Frequency Percent Nothing, they did 20 29% everything Professional people listen 19 27% to me more Put perpetrator in jail 7 10% longer Less interviews 5 7% Not have to wait so long 5 7% Counseling more helpful 4 6% Note. Percentages based on responses to question. The data on the impact of system intervention on children's self-esteem revealed a connection between increased self-esteem and professional interaction. Sixty-nine percent of the children stated that they "felt better" about themselves due to the involvement of professionals within the system. Only 10% indicated that they "felt worse" about themselves. Twenty-two percent indicated that there was "no effect" on their self-esteem from system involvement. Three primary reasons were cited by the children for increased self-esteem. Thirty-four percent stated, "I was told that it was not my fault.“ This response speaks to the impact that professionals can have on minimizing the guilt and shame often experienced by sexually abused children. Twenty-one percent indicated that professionals helped "get my feelings out". Twenty-five percent of the children said that, “I learned to stand up for me!" which indicated an increase 115 in personal empowerment. There were only 15% of the children who responded negatively stating that they, "weren't treated with respect" or they "were made to do too many things." No statistical analysis was run with trauma scores due to the small numbers in the less self- esteem category. Support for the positive impact of professional relationships was reflected in the majority of the responses to the question. Although there was no statistical confirmation between increased self-esteem and lower trauma scores, the responses indicate that stigmatization from the sexual abuse was reduced by the positive interaction with system professionals. The children were asked to, "weigh out whether the systems interventions overall were more positive, negative, or about equal?" The responses by the children were very positive and far exceeded the investigator's expectations. Seventy-two percent of the sample stated that the societal system interventions were more positive compared to only 14% who experienced them as more negative. Thirteen percent indicated that the positive and the negatives were about equal. The mean trauma scores for the three groups continued to follow the pattern observed in the study, that the children who had positive perceptions of system involvement had lower trauma mean scores than those who viewed it negatively. The sixty—five children (72%) who defined the system as positive had a mean score of 48 116 compared to the thirteen (13%) that did not and had a mean score of 57. In order to gain a larger sample to run analysis, the variable was dichotomized by labeling one group as positive and the other group as not positive which combined the children who viewed the system as negative with those that saw it as neutral. The result was that the second group mean became 60. A t-test was then run to determine if there were any significant differences between the two groups, ( e = -1.76, p = .08). The difference approached significance but was not significant. The final and perhaps the most critical question, was whether or not the decision to disclose the sexual abuse was one that the children regretted. The children were very clear about their decision, as 83% indicated that they were glad they told. Thirty—five percent gave the reason that the sexual abuse would still have been going on if they had not told. Another 22% indicated that they finally got "it" out and didn't have to live with the "secret" any longer. Nineteen percent stated they were glad they told because they knew that "he would have done it to someone else." A total of 76% of the sample defined physical and emotional safety as the ultimate positive outcome that occurred through their disclosure. Only 4% of the children wished they had not told. 117 Twelve percent were unsure of whether or not they had made the right decision in disclosing. The reason cited by nine of the fifteen children was that the family response to the disclosure had made it very difficult for them. The family responses included "no one believing", "the family separating with one side not talking to the other", and "feeling responsibility for the hatred that was now a part of the family." Only two children cited system failure because "the professional people had lied" to them. The variable was dichotomized to increase the sample size by combining, "wishing you hadn't told" with "don't know if you should have told." The resulting mean score differential was 12 points as those that were glad they told had a mean score of 50 and the second group had a mean score of 62. The t-test was not significant, but again the lower mean trauma score was found in the children who had a positive system experience. The fifteen children who wished they hadn't told or who were unsure were asked, "if a friend came to you and said that they had a secret about sexual abuse and they didn't know if they should tell someone or not, what would you tell them?" All fifteen stated that they would advise their friends to tell their secret to a trusted adult. The responses, regarding the perceived importance 118 of disclosure, demonstrate the significant negative impact that living with the secret of sexual abuse has on children. All ninety children supported the necessity of disclosure even after they had been through the system. The fear generated by the sexual abuse appeared to unquestionably outweigh for the children the effects of societal system interventions. The finding speaks to the physical and psychological devastation wrought through sexual abuse. County differemeee The second hypothesis, that the three counties, due to their varying system approaches, would be statistically different in the trauma scores obtained, was not supported as previously indicated. There are a variety of explanations as to why this did not occur. The number of interviews was found to be statistically significant in determining trauma scores. Yet the correlation of .28 was not extremely powerful which indicates that number of interviews was not the major determinant in trauma scores. This is important because there was a statistically significant difference between St. Joseph County and Kalamazoo County in number of interviews. The impact of the number of interviews was not sufficient to make the trauma scores significantly different between counties. There were too many other variables that mediated the importance of 119 the number of interviews on trauma scores within the counties. Second, testifying was projected to be a significant determinant in trauma scores but this was proven incorrect. The impact of testifying on trauma scores was found to be minimal. Consequently despite there being a statistical difference between counties in testifying, with children in St. Joseph County being less likely to testify, the effect on trauma scores was minimal. The lack of statistical significance in trauma score elevation due to testifying should not however indicate that testifying was a desired process by the children. Twenty-nine of the thirty children who testified defined it as something they would rather not have done. It was viewed as the third most difficult process of all of the system interventions. It would appear that St. Joseph was the most supportive of the counties regarding the court phase because of the significant difference in children testifying. However, there is no statistical support for this conclusion. The ability of professionals to establish trusting relationships with children appeared to be consistent across all counties. Trust, as previously discussed, was significant in influencing trauma scores. An explanation for the lack of statistical difference in trauma scores between counties despite more stressful 120 interventions in Kalamazoo County and Kent County than St. Joseph County is that the majority of children within each county were able to have at least some trust with a professional which served to mediate stress, and reduce the likelihood of trauma. Therefore, the trusting relationships moved the trauma scores toward a common mean within the counties. Trusted relationships are not a structural component of any one system intervention, but rather are attitudes that can accompany any intervention no matter how stressful. The probable outcome is the ability to minimize trauma even in the most stressful of system processes. Another support for the trauma scores not being significantly different between counties was the statistical difference between St. Joseph County vs. Kalamazoo County and Kent County in counseling. Counseling was less likely to occur for children in St. Joseph County and consequently they had less of an opportunity for affect release, cognitive processing of the sexual abuse, and building a relationship with a professional all necessary processes for the remediation of trauma. It appears that a decline in trauma scores was more likely in Kalamazoo and Kent Counties than in St. Joseph County because of a greater likelihood of being in counseling. In comparing county interventions and the subsequent trauma scores it is important to recognize 121 that there was a difference between how children and adults viewed the intervention process. The hypothesis that the societal system interventions in Kalamazoo County and Kent County would be more stressful than St. Joseph was correct. The Intervention Stressor Inventory total score was statistically significantly different in St. Joseph County than in Kalamazoo and Kent Counties. The scoring on the Intervention Stressor Inventory (see Measurement) reflects adult views on trauma producing agents. The children however, did not appear to be affected by what the experts perceived as most impactful, as the correlation coefficient between the Intervention Stressor Inventory and the Trauma Symptom Checklist was only .05. There was a clear difference between how the children experienced the system vs. how the experts thought they would. Associated with this finding are the scores from the Child Behavioral Checklist. This test filled out by the parents/guardians of the children under seventeen was statistically significant with a fairly strongly association of .36 with the Intervention Stressor Inventory. This statistic supports what the investigator was expecting to find between Trauma Symptom Checklist score and the Intervention Stressor Inventory. Yet it is important to emphasize that the Child Behavioral Checklist was filled out by the parents/guardians and not the children. Adult 122 perceptions in this study were inaccurate reflections of how the children experienced system interventions. CHAPTER SIX IMPLICATIONS AND CONCLUSION The ability of children to survive the effects of sexual abuse demands a tremendous resiliency to stress, pain, and trauma. Most of the children in this study endured great fear in carrying the terrifying family secrets for at least two years prior to disclosure. Following disclosure some experienced family pressure to recant and frequently felt responsible for the separation and division in their family that subsequently occurred. The emotional and psychological scars from the sexual abuse and its aftermath have significantly impacted their development. The consequences of such traumatization are reflected in the "clinical" scores in the Child Behavioral Checklist for the children. As the data in the study confirmed however, the children demonstrated a willingness to trust in professionals, believed that their opinions and perspectives should be heard and respected by adults, and sought remediation of their guilt and shame through such system mechanisms as counseling to enhance their self esteem. These children interfaced with the adult professional world that often operates from an adult reactive stance where its objectives are to protect the 123 124 children physically, gather evidence, and prove cases. Despite these adult objectives that frequently conflict with the primary need of victimized children to heal psychologically, the majority of children articulated a support for the interventions they experienced through their interface with the system. They communicated a recommitment to their decision to disclosure of the sexual abuse. The children through their responses informed the system that their need to be "safe" was far more important than any possible system stress they experienced in the aftermath of disclosure. There is not an indictment by children of the system as the investigator anticipated, but a strong support of the system as a whole. The system in the perceptions of a large majority of the children responded to their need for trust, empowerment, and self esteem. It is important to contextualize their responses to better understand why they may have communicated what they did. Children are very tolerant of adult behavior and they are much more apt to be satisfied with difficult or even unnecessary circumstances because of their limited knowledge and lack of expectations. This recognition speaks again to the resiliency of children who take the bad with the good, or who live with pain and difficulty and yet somehow navigate through it. Sexually abused children are not in the position of knowing the 125 alternatives to how the system might better create different methods of intervention with children when disclosure occurs. They can only share with professionals their individual responses to what they experienced and then through their words and behaviors communicate what affects them and what does not. They are limited by their experience and therefore cannot draw upon knowledge gained through research and systematic study. Children can, and in this study did, tell us what they experience and it becomes the responsibility of professionals to change and create new interventions based upon what has been heard. It is the responsibility of professionals to develop societal system interventions that are child responsive, not adult reactive. This means empowering children through listening and then acting according to what is known through research, experience, and their voices. The system must protect not only the physical well being of children, but also their emotional well being. This study provides an understanding that children are significantly affected emotionally by what the system does. This study reveals that when given the opportunity children will tell professionals what is helpful and what is not. The results of the study support Finkelhor's traumagenic theory of child sexual abuse. The findings are consistent with the expectations of the model. 126 Issues of betrayal, powerlessness, and stigmatization, as indicated in the interviews and documented in the results were discovered in most of the children's experiences. Continually the children that received the antidotes of trust, empowerment, and support for self esteem, were much more likely to have lower trauma scores. In contrast, the children who did not receive the system support were likely to have higher trauma scores, indicative of heightened trauma. Finkelhor's theory necessitates an acknowledgment that the most traumatic outcome of sexual abuse is the development of the perspective that the world is a hostile place designed to render one powerless, vulnerable, and isolated. This study supports such a theory, but speaks to the willingness of children to be open to healing opportunities. It further confirms that when children receive positive support, develop trusting relationships, and are listened to, the effects of the betrayal, powerlessness, and stigmatization can be significantly minimized. The specific implications for this study direct professionals to create new and to refine previous intervention processes. The following serve as recommendations for professional consideration: 1) Coordination of efforts by system professionals to minimize the number of interviews each child must experience. The data 2) 127 reveals that the fewer interviews for children the less likelihood of further trauma. Weighing the probable traumatic cost to children of continued interviews against the convenience for system professionals of numerous interviews demands that professionals explore alternative methods to the present intervention process. Establishing a community based protocol that integrates the various professional responsibilities with fewer interviews for children requires both a team effort and a willingness to probe possibilities. The videotaping of interviews is one positive alternative that this study statistically confirms as a method to reduce interviews. Joint interviews of the child by professionals is another method that merits possible implementation with the outcome being fewer interviews for the child. Institution of trainings that both teach and reinforce the necessity for professionals to communicate with children in a honest, straightforward, and caring manner. These human processes lay the foundation for relationships that foster trust for children and ultimately help ameliorate trauma. Communication by professionals to children that they want to 3) 128 listen and are willing to be patient for their responses is very important. Overcoming previous sexual abuse betrayal requires professionals to be tolerant, understanding, and open to what children have to say. Professionals cannot force trust upon children; they can however, provide optimal opportunities for trust to develop. Professionals are a vital link for children in reestablishing that the world can be safe, adults can trusted, and that they have a voice in what happens to them. Asking sexually abused children their opinions and perceptions of the perpetrator, family, and placement serves to empower and helps overcome their experience of powerlessness in an adult world. Access to information that affects children empowers them because it creates an understanding of what is happening in the present, what are the possibilities for the future, and how they might be impacted. The children's voices in this study were emphatic that they need to know what is occurring within the social service and legal arenas. Attempts to protect children by withholding information only negates their importance and renders them powerless. Sexually abused children have been thrust into the realities of life and must be given the 4) 5) 129 opportunity to impact their future by providing input into the decisions made by professionals. Such input reinforces to children that they are important and have a right to help direct their future. Review of the manner in which professionals support children through the rigors of the legal system will help to discover creative and new options for children. Court preparation, supportive advocates, and the sharing of relevant information with children, all minimize the likelihood of exacerbating trauma to children. Every effort should be made when children testify to keep the perpetrator out of sight of the children. The fear created by courtroom contact with the perpetrator intensifies a child's uncomfortableness and creates a climate for anxiety and heightens the possibility of trauma. The succinct.message of the children in this study was, "keep the perpetrator out of the courtroom". Consultation with latency age and teenage children by professionals prior to removal to solicit their perceptions of risk will provide critical information in determining placement. Who better to know the probable actions of the non-offending parent than the victim within the 6) 7) 130 home. A decision by a professional as to the parent's ability to ensure protection must include the perceptions of the child in the home. Empowering child victims through solicitation of their perspectives bequeaths dignity and respect that only serve to reinforce self-worth for the child. It also serves to assist in the transition to an out of home placement through child participation in decisionmaking. Ultimately the child must be told that it is the professional's responsibility to make the final decision, yet every attempt must be made to not reinforce the powerlessness of the child as a pawn in an adult world. The results of this study clearly indicate that counseling is the primary system vehicle for children to release feelings, establish trust, and build self esteem. Consequently every effort should be made to link children with a counselor immediately to create the most Optimal environment in which to begin psychological and emotional healing. Such an intervention, although not ensuring the reduction of trauma does provide the best opportunity for healing to occur. Professional attitudes are the fulcrum from which the impact of societal system intervention 131 hinges. When professional attitudes nurture trust through respect, listening, and personal validation, children are likely to meet the system demands without an exacerbation of sexual abuse trauma. When trust is undermined, information withheld, and children are reinforced in their powerlessness, the likelihood of trauma will most likely increase. Children in this study are telling professionals to become sensitized to the importance of treating children with respect. Training in the investigative, legal, and social service phase must begin by educating professionals as to the long term significance that their approaches have in impacting children. Educating professionals that the success of achieving their individual system goals must be balanced with interventions that support sexually abused children in the development of trust, empowerment, and self esteem must be the number one priority of system trainers. These perspectives provide an opportunity for the social work profession to positively impact the lives of sexually abused children and the system that the children must encounter. The history of child welfare reform has always been dependent on social workers 132 advocating for the needs of victimized children. Social workers have always been at the forefront of speaking for the disenfranchised. Social workers are involved at all phases of system intervention with sexually abused children. They are also the profession most likely to have the most extensive and intense involvement following disclosure. The roles of child protective services workers, foster care workers, school social workers, court liaisons, and counselors are primarily filled by social workers. The education and philosophy of the social work profession is predicated on a systems approach to human growth and development. The amelioration of trauma due to systems interventions in child sexual abuse demands, as the study indicates, a holistic approach that considers not only the physical safety of children but also their emotional and psychological health. Social workers have received this training within their profession and must move forward and be the leaders in developing system intervention processes that support the total well being of children. No other profession is better equipped or more actively involved than social work to advocate for the needs of children through system change. The need for further study to expand on the findings of this study is indicated especially in the area of societal system interventions in child sexual 133 abuse because of the relatively few studies done. Hopefully this study can provide a building block for further study on the impact of system interventions. System professionals have a responsibility to learn how they might more positively impact children and thus minimize trauma rather than exacerbating it, which this study reveals can occur. The following are areas where further study could be very helpful: 1) 2) A comparative study of children who disclose sexual abuse but do not go through any court process compared with those children that do. Court involvement is not mandatory in sexual abuse cases and there are many cases where the juvenile and criminal courts are not accessed for a variety of reasons. Such a study would provide far more revealing data on the impact of court on sexually abused children by having a control group from which to compare, which was a limitation of this study. Interviewing sexually abused children and their parents for trauma at four month intervals following the initial system interventions. The purpose of these interviews would be to learn if trauma is exacerbated in children by certain system events while the initial intervention process is either occurring or recently finished. Further such a study would 134 provide information as to whether the impact of the system interventions to exacerbate trauma changes over time. 3) Enlarging the sample size and expanding the population from which the sample was drawn would greatly increase the significance of the results gained. The impact of such interventions as interviewing, counseling, testifying, and removal could be more fully explored through additional statistical analysis. There would be a far greater ability to then generalize the findings to other populations beyond the three counties. Included in further study would be a need to lower the age limit to learn if younger children are affected differently by systems interventions. The developmental stage a child is in can have a dramatic effect on how the child experiences potentially traumatic events. It is important to discover if younger children have a greater resiliency to trauma caused by system interventions and if so why. James Garbarino, well known expert on child abuse, authored a book entitled, What childrem eem LE1; me; The results of this study tell professionals that children have much to say regarding the impact of system interventions on them. Their responses point a 135 direction for professionals that can assist in minimizing trauma to sexually abused children of the future. In order for this to occur, professionals must be willing to accept that children are the experts of their own experiences. They can communicate what works and what doesn't. Children may not be able to identify the complex intrapsychic properties of trauma but they communicate through their words and behaviors when something is wrong. The effort of this investigator was to respect children's verbal and behavioral communications by giving them a vehicle to express them. Ninety children took the opportunity and as a result the researcher believes that much has been learned. The conclusions drawn from the interviews, the subsequent data obtained, and the statistical analysis highlight the importance of what children have to say. Professionals need to support children's painful stories of sexual abuse by intervening in ways that minimize the impact of professional intervention by listening to what children really need from them once disclosure occurs. The investigator has been deeply touched by the stories of the children who so courageously shared them. They are honored as a gift. The investigator is deeply indebted to each and the hope is that by their sharing, children in the future will receive the support they so need as they interface with the system following sexual abuse disclosure. APPENDIX A APPENDIX A Child Behavior Checklist CHILD BEHAVIOR CHECKLIST FOR AGES 4-16 For otlice use only ID a CHILD'S PARENT'S TYPE OF WORK (Pieaae be specific-Io: example. auto mechanic. high NAME achool teacher. homemaker, laborer. lathe operator. shoe uieaman. army ear'gearrr. even i! parent doea not live with child.) ETHNIC FATHER'S sex D 80’ - AGE GROUP TYPE OF W0Ric OR RACE D cm MOTHER'S TYPE OF WORK TODAY'S DATE CHILD'S BIRTHOATE THIS FORM FILLED OUT BY: no, Day Yr, Mo. Day Yr. D Mother ‘ !1 GRADE D Father (name): IN SCHOOL [:1 Other-name A relationship to child: I. PIeeee list the eporte your child moat like: to take part In. For example: swimming, baseball. skating. skate boarding. bike riding, Iiehing, etc. Compared to other children at the eeme age. about how much time doee helehe epend In each? Compared to other children ot the eeme age. how well doea helehe do each one? U None °°“" ruin:- aw m“: Don‘t am Above Know Average Average Know Average Average Average a. CI [3 C] C] [I] C] D C] b. Cl C] C] C] C] D C] D c. C] D D E] Cl C] D C] It Pic-u “simmummohobbm. Wmoimamamoim Compared to otherchIIdrenoIthe activitlee.endgamee.othertheneporta. umeegqabeuthowmuchttme aameage.hewwelldoeehelehedo For example: stamps. dolls. books. piano. doee helehe apend In each? each one? cratta. singing. etc. (Do not Include TN.) D None °°"" 1": AW I“: Don‘t Ietow Above Knew Average Average Know Average ”'00. Average I D C] D D [I D D D o. C] [I] C] C] C] C] C! C] c. D E] C] E] D D E] C] III. PIeeeeIletenyorgenIzettene.cIube. Comparedtootherchitdrenotthe hemmagroupemctiltdbetongeto. eemeege.hovvectIveIeheIeheIn U None “a" m Leee A Here I E] Cl C] D o. D C) U D c. E] El Cl C] IV. Pteeeebetenytebeerchereemehltd Comparedteotherchitdrenotthe hea. For example: paper route. babysitting. I“ 090. M ”I 0°00 W making bed. etc. carry them out? [:1 None .23: a...“ m- is. an Cl E] Cl C] a E] C] D D r» U D D D mrmmuamvemgmvreem use: ear-ave- 137 V. t.AbeuthewmenyeIeeetrIendedoeeyeurehIIdheve? C) None 2. Abernhewnienytlmeeaweekdeeamehttddothtngewtththem? C11 [3 less than 1 C] 20r3 Dior? VI. Comparedteothuotdtdrutettdelhumhowwettdoeeyourchtld: Worse Abouttheaame better s Get along with his/her brothers A sisters? D D D Get along with other children? E] D C] c. Behave with hlalher parents? C] C] D d. Play and work by himseltlherselt? E] [I] [:1 VII. 1. wwmm-torehitdrenegedeandolder: DDoeenotgoioechooi Felling Iaiovveverage Average Aboveaverage a. Reading or English C] C) C] D b. Writing C] E] C] c. Arithmetic or Math D C] C] D d. Spelling E] C] C] C] Other academic aub- e. C] U D C] iecte-ior exampie: hie tory. science. foreign I. C] D D [:1 Monaco. donor-phy- g. C] C) C] E) 2. Iamohttdtnaegeetalelaee? D No C] Yes-what kind? a Neeyourehtldeverregeatedagrade? E] No D Yes-grade and reason 4. Itasyourchtldhedanyacademtcorothergrobtemstnechoot? D No D Yes-please describe When did these grobteme start? Itevetheeeproblemeended? C] No C) Yea—when? PAGE 2 138 VIII. Below is a list of items that describe children. For each item that describes your child now or within the past 6 the 2 if the item is very true or often true of your child. Circle the 1 if the item is please circle or sometimes true of your child. it the item is not bus of your child. circle the 0. Please answer all items as well as you can, even it some do not seem to apply to your child. 0 a Not True (as far as you know) 1 = Somewhat or Sometimes True 2 an Very True or Often True PAGE 3 0 1 2 1. Acts too young for his/her age 16 o 1 2 31. Fears helehe might think or do something 0 1 2 2 Allergy (describe): bad 0 1 2 32. Feels helshe has to be perfect 0 1 2 33. Feels or complains that no one loves himlher O 1 2 3 Argues a lot 0 1 2 4 Asthma 0 1 2 34. Feels others are out to get him/her 0 1 2 35. Feels worthless or interior 50 0 1 2 ‘ ‘ . . . 2 :2::1::.i:.°:::::::.::: 2° . 1 2 .. . 0 1 2 37. Gets in many tights 0 1 2 7 B'aQQ'W' boasting , _ 0 1 2 38. Gets teased a lot 0 1 2 8 Cantconcentrate. can 1 pay attention for long 0 1 2 39. Hangs around with children who get in trouble 0 1 2 9 Can't get his/her mind off certain thoughts; obsessions(describe): O 1 2 40. Hears things that aren't there (describe): 0 1 2 10 Can’t sit still, restless. or hyperactive 25 55 0 1 2 41. Impulsive or acts without thinking 0 1 2 11. Clings to adults or too dependent 0 1 2 12. Complains of loneliness 0 1 2 42. Likes to be alone 0 1 2 43. Lying or cheating O 1 2 13. Confused or seems to be in a fog 0 1 2 14. Cries a lot 0 1 2 44. Bites fingernails 0 1 2 45. Nervous. highstrung. or tense 60 0 1 2 15. Cruel to animals 30 ' 0 1 2 16. Cruelty. bullying. or meanness to others 0 1 2 46' Nervous movements or twitching (describe). 0 1 2 17. Day-dreams or gets lost in his/her thoughts 0 1 2 18. Deliberately harms self or attempts suicide 0 1 2 47. Nightmares 0 1 2 19. Demands a lot of attention 0 1 2 48_ No( liked by other children 0 1 2 20. DOSITOYS his/her OWN things 35 o 1 2 49. constipated. doesn't move bowe|s 0 1 2 21 Destroys things belonging to his/her family 0 1 2 50. Too fearful or anxious 65 or other children 0 1 2 51. Feels dizzy 0 1 2 22 Disobedient at home 0 1 2 52. Feels too guilty 0 1 2 23. Oisobedient at school 0 1 2 53. Overeating 0 1 2 24. Doesn't eat well 0 1 2 54. Overtired 1 2 25. Doesn't get along with other children 40 0 1 2 55- Overweight 70 1 2 . ‘ f l ‘I it h ' 26 Doesn ' seem to ee gu1 tya ermlsbe avmg 56. Physical problems without known medical 1 2 27. Easily jealous 1 2 causzzcnes or ains 1 2 28. Eats or drinks things that are not food 0 a. p (describe)- 0 1 2 b. Headaches ' — 0 1 2 c. Nausea. feels sick 0 1 2 d. Problems with eyes (describe): 1 2 29. Fears certain animals. situations. or places, 0 1 2 e. Rashes or other skin problems 75 other than school (describe): 0 1 2 f. Stomachaches or cramps 0 1 2 g. Vomiting, throwing up 0 1 2 h. Other (describe): 1 2 30. Fears gomg to school 45 Please see other side o - NotTmehefarasyouknow) 139 1 - SomewhatorSometImesTrue 2 - VeryTnleorOttenTnie 01 2 57. 0 1 2 58. 0 1 2 59. 0 1 2 80. 0 1 2 61. 0 1 2 62. 0 1 2 83. 0 1 2 84. 0 1 2 65. 0 1 2 66. PTwsically attacks people Picks nose, skin, or other parts of body (describe): Plays with own sex parts in public 16 Plays with own sex parts too much Poor school work Poorly coordinated or clumsy Prefers playing with older children 20 Prefers playing with younger children Refuses to talk Repeats certain acts over and over. compulsions (describe): Runs away from home Screams a lot 25 Secretive. keeps things to self Sees things that aren't there (describe): Self-conscious or easily embarrassed Sets fires Sexual problems (describe): Showing 011 or clowning Shy or timid Sleeps less than most children Sleeps more than most children during day and/or night (describe): Smears or plays with bowel movements 35 Speech problem (describe): Stares blankly Steals at home Steals outside the home Stores up things he/she doesn't need (describe): 40 1 2 84. 87. 89. 91. 92. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. Strange behavior (describe): Strange ideas (describe): Stubborn, sullen. or irritable Sudden changes in mood or feelings Sulks a lot 45 Suspicious Swearing or obscene language Talks about killing self Talks or walks in sleep (describe): Talks too much 50 Teases a lot Temper tantrums or hot temper Thinks about sex too much Threatens people Thumb-sucking 55 Too concerned with neetness or cleanliness Trouble sleeping (describe): Truancy. skips school Underactive. slow moving. or lacks energy Unhappy, sad. or depressed 80 Unusually loud Uses alcohol or drugs (describe): Vandalism Wets self during the day Wets the bed 65 Whining Wishes to be of opposite sex Withdrawn. doesn't get involved with others Worrying Please write in any problems your child has that were not listed above: 70 PLEASE BE SURE YOU HAVE ANSWERED ALL ITEMS. PAGE A UNDERLINE ANY YOU ARE CONCERNED ABOUT. APPENDIX B APPENDIX B Trauma Symptom Checklist for Children How often do each of these things happen to you? (circle the 1) Bad dreams or nightmares. O 1 2 3 Never Sometimes Lots of times Almost all 2) Feeling afraid something bad might happen. 0 1 2 3 Never Sometimes Lots of times Almost all 3) Scary ideas or pictures just pop into my head. 0 l 2 3 Never Sometimes Lots of times Almost all 4) wanting to say dirty words. 0 1 2 3 Never Sometimes Lots of times Almost all 5) Pretending I am someone else. 0 1 2 3 Never Sometimes Lots of times Almost all 6) Arguing too much. 0 1 2 3 Never Sometimes Lots of times Almost all 7) Feeling lonely. 0 l 2 3 Never Sometimes Lots of times Almost all 8) Touching my private parts too much. 0 1 2 3 Never Sometimes Lots of times Almost all 9) Feeling sad or unhappy. O l 2 3 Never Sometimes LOts of times Almost all 10) Remembering things that happened that I didn't like. 0 l 2 3 Never Sometimes Lots of times Almost all 11) Going away in my mind, trying not to think. 0 l 2 3 Never Sometimes Lots of times Almost all 12) Remembering scary things. 0 l 2 3 Never Sometimes Lots of times Almost all 140 of of of of of of of of of of of of number): the the the the the the the the the the the time time time time time time time time time time time time 141 13) Ranting to yell and break things. 0 1 2 3 Never Sometimes Lots of times Almost all 14) Crying. O 1 2 3 Never Sometimes Lots of times Almost all 15) Getting scared all of a sudden, and don’t know why. 0 1 2 3 Never Sometimes Lots of times Almost all 16) Getting mad and can't calm down. 0 1 2 3 Never Sometimes Lots of times Almost all 17) Thinking about having sex. 0 1 2 3 Never Sometimes Lots of times Almost all 18) Feeling dissy. O 1 2 3 Never Sometimes Lots of times Almost all 19) Wanting to yell at people. 0 1 2 . 3 Never Sometimes Lots of times Almost all 20) Wanting to hurt myself. 0 1 2 3 Never Sometimes Lots of times Almost all 21) Wanting to hurt other people. 0 1 2 3 Never Sometimes Lots of times Almost all 22) Thinking about touching other people's private parts. 0 1 2 3 Never Sometimes Lots of times Almost all 23) Thinking about sex when I don't want to. 0 1 2 3 Never Sometimes Lots of times Almost all 24) Feeling scared of men. V O 1 2 3 Never Sometimes Lots of times Almost all 25) Feeling scared of women. 0 1 2 3 Never Sometimes Lots of times Almost all of of of of of of of of of of of of of the the the the the the the the the the the the the time time time time time time time time time time time time time 142 26) Hashing myself because I feel dirty inside. 0 1 2 Never Sometimes Lots of times 27) Feeling stupid or bad. 0 1 2 Never Sometimes Lots of times 28) Feeling like I did something wrong. 0 l 2 Never Sometimes Lots of times 29) Feeling like things aren't real. 0 1 2 Never Sometimes Lots of times .30) Forgetting things, can’t remember things. 0 1 2 Never Sometimes Lots of times 31) Feeling like I’m not in my body. 0 1 2 Never Sometimes Lots of times 32) Feeling nervous or jumpy inside. 0 l 2 Never Sometimes Lots of times 33) Feeling afraid. O l 2 Never Sometimes Lots of times Almost Almost Almost Almost Almost Almost Almost Almost 34) Not trusting people because they might want sex. 0 l 2 Never Sometimes Lots of times Almost all all all all all all all all 3 all 35) Can’t stop thinking about something bad that happened 0 1 2 Never Sometimes Lots of times 36) Getting into fights. O l 2 Never Sometimes Lots of times 37) Feeling mean. 0 ' 1 2 Never Sometimes Lots of times Almost Almost Almost 3 all 3 all 3 all of of of of of of of of of to of of of the the the the the the the the the the the the time time time time time time time time time time time time 143 38) Pretending I’m somewhere else. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 39) Being afraid of the dark. 0 1 2 ' 3 Never Sometimes Lots of times Almost all of the time 40) Getting scared or upset when I think about sex. 0 l 2 3 Never Sometimes Lots of times Almost all of the time 41) worrying about things. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 42) Feeling like nobody likes me. O 1 2 3 Never Sometimes Lots of times Almost all of the time 43) Remembering things I don't want to remember. 0 l 2 3 Never Sometimes Lots of times Almost all of the time 44) Having sex feelings in my body. 0 l 2 3 Never Sometimes Lots of times Almost all of the time 45) My mind going empty or blank. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 46) Feeling like I hate people. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 47) Can't stop thinking about sex. 0 1 2 3 Never . Sometimes Lots of times Almost all of the time 48) Trying not to have any feelings. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 49) Feeling mad. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 50) Feeling afraid somebody will kill me. 0 1 2 3 Never Sometimes Lots of times Almost all of the time 144 51) Wishing bad things had never happened. 0 1 2 Never Sometimes Lots of times 52) Wanting to kill myself. 0 1 2 Never Sometimes Lots of times 53) Daydreaming. O 1 2 Never Sometimes Lots of times 54) Getting upset when people talk about sex. 0 1 2 Never Sometimes Lots of times Almost Almost Almost Almost all all all all of the of the of the of the time time time time APPENDIX C Appendix C Intervention Stressor Inventory Table 1 Geometric Mean Stress Ratings for Investigative Events with Modifying Factors Events Score Polygraph (victim) 88 Lineup 82 Defense interview 81 Joint interview 75 Pretrial deposition 74 Medical exam 71 Law officer interview 50 Prosecutor interview 49 Social worker interview 35 Modifying Factors Presence of perpetrator 108 Presence of unsupportive parent 53 Child videotaped 10 Anatomical dolls used 8 Guardian Ad Litem present -11 Presence of supportive parent -14 Child debriefed following event -16 Child-oriented setting used -16 Child prepared well for event -20 145 146 Table 2 Geometric Mean Stress Ratings for Testimony and Court Proceeding Events With Modifying Factors Events Score Testimony in juvenile/criminal court 111 News media reveals family name 111 Testimony before grand jury hearing 94 Testimony at preliminary hearing 86 Presence at disposition or sentencing 69 Competency examination 58 Modifying Factors Cross-examination for more than 1 hour 88 Harsh questioning of the child 88 Perpetrator present and visible during testify 77 Unsupportive non-offending parent present 47 Perpetrator present but shielded 35 Moderate questioning-difficult to follow 21 Proceedings videotaped 7 Proceedings in closed court 0 Anatomical dolls used to demonstrate -9 Court proceedings over closed circuit TV -11 Bench trial only -12 Child debriefed following event -17 Judge sets clear rules for interrogation -18 GAL or victim advocate present -18 Mild cross-examination with no confusion -22 Other support person present -21 Child well prepared for event -23 Proceeding occurred in judge's chamber -25 Presence of supportive non-offending parent -29 147 Table 3 Geometric Mean Stress Ratings for Social Services Interventions with Modifying Factors Events Score Perpetrator returns to child's home 149 Child receives inpatient psychiatric care 120 Child placed in shelter care 113 Child placed in foster care 100 Child removed from home to live with relatives 65 Perpetrator is removed from child's home 45 Modifying Factors Child must change schools as result of placement 41 Child is well prepared for event -21 Child has active support person -27 APPENDIX D APPENDIX D Investigator's Questionnaire I. DEMOGRAPHIC DATA (This data will be gathered prior to the interview with the child from the parent or guardian.) 1) Age of the child? 2) Sex of the child? 3) Race of the child? 4) County of residence? 5) Economic status of the parents? 6) Present placement of the child? 7) Sex of the perpetrator? 8) Relationship of the perpetrator to the child? 9) Length of the abuse? 10) Length of time since initial disclosure? II. INTRODUCTION :9 CHILD I am trying to learn about how protective services, police, and court involvement affects children. I will be asking you many questions about your experiences with protective services, police, and the courts. Please answer the questions the best you can. III. INVESTIGATIVE PHASE 11) Who did you first tell about the sexual abuse that happened to you? Why did you tell him/her first? 12) How did protective services or the police become involved? 13) Since protective services and/or the police have become involved about how many times have you been interviewed? By who? 148 14) 15) 16) 17) 18) 19) 20) 22) 149 How do you feel being interviewed more than once affected you? Explain. 1 3 5 very no very harmful effect helpful Of the professionals who was it easiest to tell about the sexual abuse? Why? Of the professionals who was it the most difficult to tell what happened? Why? Were there any interviews that were videotaped? Which ones? How did you feel being videotaped affected you? 1 3 5 very no very harmful effect helpful What were the most difficult parts of the interview or interviews? What did the most the interviewer do to make the interview easier for you? What could have been done to make the interview more helpful for you? How much did you trust the professional people who said' they wanted to help you? Explain. l 3 5 none some very much 150 23) Was there a particular professional person you trusted more than the others? Explain. 24) When protective services and the police first got involved did you feel that they believed what happened to you? Why or why not? 1 3 5 did not some very much believe 25) After you told them what happened to you, what did you want them to do? How much effect do you think you had on what they did after you told about the sexual abuse? 1 3 5 no some much dif. difference III. COURT PROCEEDINGS 26) Did you testify in court? 27) How many times? 28) In what court or courts did you testify? 29) How well prepared were you for what went on during your testimony? Explain. 1 2 3 4 5 6 7 8 9 10 very some very little much 30) Was there someone you trusted with you when you testified? Who? 31) If there was a trusted person there with you, how helpful was that to you? Why? 1 3 5 no some very help helpful 151 32) How do you feel that testifying in court affected you? How come? 1 3 5 very no very harmful effect helpful 33) Do you feel that what you said during the court proceedings had an effect on the people who listened? Why or why not? 1 3 5 no some great effect effect 34) How did having to testify affect your trust in the people who were trying to help you? Explain. l 3 5 harmful some very effect helpful 35) After you testified how did you think the people who. heard your testimony felt about you? Why or why not? 1 3 5 much no dif. much worse better 36) Was the person who sexually abused you in the courtroom when you testified? 37) Was the person who sexually abused you convicted of the offense? 38) Was the person who sexually abused you sentenced to jail or prison? 39) What did you think about what happened to the person? Why 1 2 3 4 5 6 7 8 9 10 didn't didn't glad want that matter about to happen what happened 40) What were the most difficult parts of going to court? Why 152 41) What were parts of the court procedures were the most helpful to you? Why? 42) What would you have like to have happened that would have made it easier for you during the court procedures? Why IV. SOCIAL SERVICE INTERVENTIONS 43) After you told about the sexual abuse was there any one in your family that believed you? Who? 44) Who was the most helpful person within your family after you told about the sexual abuse? What did they do that helped you to feel that way? 45) How did you feel that your family felt about you telling about what happened? How come? 1 3 5 very didn't very mad at you care glad 46) What could your family have done to help you feel more supported? 47) Who was the most helpful and trusted professional after you told? What did they do that helped you to feel that way? 48) If there was a trusted professional how much did it help you to know that you could trust him/her? 1 3 S no some great help help 49) Were you removed from your home? 50) 51) 52) 53) 54) 55) 56) 153 How do you feel that the experience of removal affected you? Explain. 1 3 5 very no very harmful effect helpful Do you feel that the professional within the system listened and were concerned about whether you wanted to be removed or not? Why of why not? 1 3 5 did not some listened listen concerned If you were removed how do feel that the removal affected you? How come? Has it changed with time? 1 3 5 very no very harmful effect helpful What was the most difficult part for you in being removed from your home? How come? What was the most helpful part for you in being removed from your home? How come? What could have been done to make the experience of removal easier for you? If you remained in your own home how do you feel that the remaining within your own affected you after you told? ' How come? Has it changed? 1 3 5 very no very harmful effect helpful 57) 58) 154 What was the most difficult part for you in remaining home? How come? What was the most helpful part for you in remaining home? How come? 59) How did your placement affect your ability to trust 60) 61) 62) 63) IV. 64) 65) outside professionals? Explain. 1 3 5 less no much more trusting effect trusting How long after you told about the sexual abuse did you begin counseling? How long did you remain in counseling? How do you feel that the experience of counseling affected you? Explain? l 3 5 very no very harmful effect helpful What was the most helpful thing that happened in counseling? Explain. OVERVIEW Of everything that happened after you told what were the most difficult things for you? Of everything that happened after you told what were the most helpful things for you? 155 66) What could have been done by the professional people involved i.e. cps, police, and courts that would have been helpful to you? 67) What effect do you think the involvement of the different professionals had on your willingness to trust other people? Explain. 3 . 5 very no very harmful effect helpful 68) What effect do you think the involvement of the different professionals had on your belief that what you have to say is important and matters to others? Explain. l 3 5 very no very harmful effect helpful 69) What effect do you think the involvement of the different professionals had on how you feel about yourself? Explain. l 3 5 felt worse no felt better about myself effect about myself 61) Looking back over all that has happened since you told, what do you feel about your decision to tell? Explain. l 3 5 wish I don't glad hadn't told know I told 62) How would you weigh out the total societal system intervention? 1 3 5 more positive neutral more negative APPENDIX E i) 2) 3) 4) 5) 6) 7) a) 9) 10) i 1) i2) 13) 14) )5). 16) I?) id) 19) 20) 2 I) 22) 23) 24) 25) 26) 27) 2B) 29) APPENDIX E Fact Sheet NAME DOB SEX M F RACE C B O COUNTY OF RESIDENCE K K!) S.J. PARENT'S NAME MOST RECENT ADDRESS DATE OF DISCLOSURE TIME SINCE DISCLOSURE AGE AT TIME OF DISCLOSURE AGE AT TIME OF ABUSE LENGTH OF ABUSE TYPE OF CSC ID , 2D 3D 4D SEX OF PERP RELATION OF PERP TO CHILD COURT INVOLVEMENT C J DATE WARRANT ISSUED CHARGES AT TIME OF WARRANT A'I'I'NY. CA RO " OF INTERVIEWS INTERVIEWERIS) CHILD TESTIFY? Y N WHAT HEARINGS? SUPPORT PRESON AVAILABLE? Y N WHO? DATE OF CC RESOLUTION FINAL DISPO OF CC DATE OF .JC RESOLUTION FINAL DISPO OF JC 30) TYPE OF PLACEMENT 3 I) 32) 33) 34) 3'5) 36) LENGTH OF PLACEMENT NAP. SUPPORTIVE? V N PRE-SENTENCE RECOMMENDATION SENTENCE RECEIVED VISITATION ARRANGEMENTS PERP COUNSELING ARRANGEMENTS 156 N.A.P APPENDIX F APPENDIX E Investigator's Letter to Parents MICHIGAN STATE u N l v E R s I T Y Dear Parent/Guardian: Greetings! My name is Jim Henry. I am a social worker who has been working with children for over 15 years. My area of specialty is child sexual abuse. During my career I have seen some professionals be extremely helpful to children who have experienced sexual abuse and others unfortunately who have not. I believe it is very important that professionals learn what is most helpful and what is harmful to children who have been sexually abused when they try and help. Who better to inform professionals of what is helpful and what is not then the children who have been through it. I am now conducting a research study for my dissertation at Michigan State University, to learn how system interventions (child protective services, police, and the courts) affect children when they have been sexually abused by an adult household member. I am planning to do this by interviewing children from the ages of 10-20. who have experienced system interventions following a disclosure of sexual abuse. I invite you and your child to use this opportunity to tell how you experienced interventions by protective services, police, and the courts by participating in this study. I. will be conducting interviews of at least 90 children. ages 10-20, from St. Joseph, Kalamazoo, and Kent Counties. These interviews will include 3 sets of questions for the children and one set of questions for the parent/guardian. There will be 99 mm; regarding details of the past sexual abuse itself. I am interested strictly in how the child experienced the system interventions and their possible effects. All information obtained will be confidential. The final results will include no names and the conclusions will be available to all participants. For the time and energy required the children will receive $25.00. I have included a letter to your child explaining the study. Interviews Will be done at your borne for your convenience unless you would rather have the interview done elsewhere. To arrange an interview or to ask further questions regarding the study please contact the number below after 4pm on weekdays and anytime during the weekend. Please contact me as soon as possible. If a long distance call is required please reverse the charges to me. I hope you will consider granting permission for your child to participate in this study. The information gained will be used to help create the best ways possible rn the future to help children who have been sexually abused. Jim Henry M.S.W. (91g) 345-22“) Flag; g_a_ll gm; 4 pm. 157 APPENDIX G APPENDIX G Court Letters .......... Prosecuting Attorney .................. wads-raw December 14. 1993 mam Re: Child Sexual Assault Research Study Dear Ms. This letter is written to inform and encourage you and your child's participation in a research study being done by Mr. James Henry. Mr. Henry is currently employed in Kalamazoo's child protection system. Additionally. he is working on his Doctorate degree from Michigan State University in the field of child sexual abuse. Jim is a respected professional with a reputation for integrity and sensitivity in dealing with children. The purpose of Jim's research study is to learn how children who have been sexually abused are affected by their contacts with child protective services. police. attorneys and judges. I am normally reluctant to endorse any study. However. Jim's research is intended to help future children who become victims. Jim has assured me that you and your child will be treated with sensitivity and respect. The study will provide children with an opportunity to express what they think and feel regarding their experiences as a viccim. The results will help my Office and other professionals to improve our service to future victims during a most difficult time in their life. Obviously. participation in Hr. Henry's research study is completely voluntary. However. I'd ask that you please give consideration to participating. Thank you. incerely. A owl“, :n :17WhmnlmqpnAwwue-Kmmnmneluanpmeunv m 6080588 (616) 383-6000 FAX NUMBER (616) 383-0475 158 159 THOMAS E. SHUMAKER. Judge of Probate Ruby A, mm" ST. JOSEPH COUNTY PROBATE COURT Register of Probate PO. Box 190 Dennis L. DeVore Centrevllle. MI 49032-0190 Juvenlle Court Dlrector Probate (616) 467-5538 Juvenile (616) 467-5561 December 7. 1993 Dear Parents: This letter is written to inform and encourage you and your child's participation in a research study being done by Jim Henry. Jim has undertaken a very important study to learn how children who have been sexually abused are affected by system interventions by child protective services. police. attorneys. and judges. We are very aware of the sensitive nature of your child's victimi— zation. The court is extremely hesitant to endorse any study. Yet we support this particular study because we are confident in Jim's professionalism and further we see a tremendous benefit to future children who must encounter the court's invervention following sexual abuse disclosure. Jim has worked many years in the field of child sexual abuse and is presently working on his PhD. He is seeking to help make the system more helpful to children rather than harmful. His reputation and his work speak to his honesty and integrity. He will treat you and your child sensitively and with respect. This study will provide your child with a vehicle to express what she/he thinks and feels regarding the system interventions that were experienced. The results will help this court and other professionals develop interventions that are child responsive during a most difficult time. Please consider participating! Sincerely: 43%.. Dennis L. DeVore Court Director 31. 310521111 Q'Lnuntg @ffirr of the lfirnsrruting Attarnrg PO. Box 250 Centrcville. MI 49032-0250 (616) 467-5547 PROSECUTING ATTORNEY Jeffrey C. Middleton ASSISTANT PROSECUTORS Douglas K. Fisher Jeanette Jackson Charles E. Herman Robert K. Pattlson Dear This letter is written to inform and encourage you and your child's participation in a research study being done by Jim Henry. Jim has undertaken a very important study to learn how children who have been sexually abused are affected by systan interventions by child protective services. police. attorneys. and judges. We are very aware of the sensitive nature of your child's victimization. The court is extremely hesitant to endorse any study. Yet we support this particular study because we are confident in Jim's professionalism and further. we see a tremendous benefit to future children who must encounter the court's intervention following sexual abuse disclosure. Jim has worked many years in the field of child sexual abuse and is presently working on his Ph.D. Heris seeking to help make the systen more helpful to children. rather than harmful. His reputation and his work speak to his honesty and integrity. He will treat you and your child sensitively and with respect. This study will provide your children with a vehicle to express what they think and feel regarding the system interventions that were experienced. The results will help this court and other professionals develop interventions that are child responsive during a most difficult time. Please consider participating Yours truly. ( W. W Jeffrey . Middleton Prosecuting Attorney JM/se 161 first: of mimigan 3" Donald R. Halstead '44 ' . . CHIEF JUDGE mlgmgfugi'am James 8. Casey 1400 Gull Road Carolyn H. VVIlliams ”tabla! (1101111 Kalamazoo. Micnngan 49001 JUDGES or PROBATE mum Df K818111330!) (616) 385-6000 r - JUVENILE DIVISION ”‘ (616) 385 8588 December 1993 Dear This letter is written to inform and encourage your participation in a research study being done by Jim Henry. Jim has undertaken a very important study to learn how children who have been sexually abused are affected by system interventions by child protective services. police. attorneys, and judges. Jim has worked many years in the field of child sexual abuse and is presently working on his PHd. The Court is supportive of his efforts. He is seeking to help us make the system more helpful to children rather than harmful. His reputation and his work speak to his honesty and integrity. He will treat you and your child sensitively and with respect. Jim's goal is to help make the system more supportive to children after sexual abuse disclosure. This study will provide children with a way to express what they think and feel regarding the system interventions that they experienced. The results will help this Court and other professionals develop interventions that are child responsive during a most difficult time. We are very aware of the sensitive nature of a child's victimization. The Court is extremely hesitant to endorse any study. Yet, I support this particular study because I am confident in Jim's professionalism and further, I see a tremendous benefit to future children who must encounter the Court's intervention following sexual abuse disclosure. Please consider participating! Sincerely, a--- ‘\ fr .\ _ ) _ ‘ ’é‘t" #31 ”pg eg- ‘- ——_. Douglas W. 81 de Court Administrator 162 KENT COUNTY JUVENILE COURT John P. Steketee ChieIJudge Donald J. DeYoung 150] Cedar Street N.E. J“d9°°‘ P'°b°‘° Grand Rapids, MI 49503-1390 Janet A. Haynes “"9" °‘ pmb'm Phone (616) 774-3700 Hanaruth H. Carpenter FAX (616) 242-6534 Judge 0! Ptohotc February 1994 Dear: This letter is written to inform and encourage your participation in a research study being done by Jim Henry. Jim has undertaken a very important study to learn how children who have been sexually abused are affected by system interventions by child protective services, police, attorneys, and judges. Jim has worked many years in the field of child sexual abuse and is presently working on his PhD. The Court is supportive of his efforts. He is seeking to help us make the system more helpful to children rather than harmful. His reputation and his work speak to his honesty and integrity. He will treat you and your child sensitively and with respect. Jim's goal is to help make the system more supportive to children after sexual abuse disclosure. This study will provide children with a way to express what they think and feel regarding the system interventions that they experienced. The results will help this Court and other professionals develop interventions that are child responsive during a most difficult time. We are very aware of the sensitive nature of a child's victimization. The Court is extremely hesitant to endorse any study. Yet, I support this particular study because I am confident in Jim's professionalism and further, I see a tremendous benefit to future children who must encounter the Court's intervention following sexual abuse disclosure. Please consider participating! Sincerely, «l Ja k Roedema Court Director .31-"K POFDF'M nilm'fil '4 ("N'H Q'WVI -" V‘HN M’r‘l "fiv'whm HiIN In! (If (7)1!” ‘u‘lx'lrr‘f EDWARD M MFDENIYlRl‘. SIIIInIiIIInIIIII‘IIt of Detention ,’ LMVII) “ORR. Altnnmv Roieten APPENDIX H APPENDIX H Child Letter MICHIGAN STATE IJN Iv ER SII'Y Greetings: My name is Irm Henry. I have worked with kids for many years, both as a teacher and a social worker. Sometimes I have seen situations that an adult does something to sexually harm a child. When Other adults try and help in these situations they can do so without knowing what kids need or want. I believe it is very important to hear what kids have to say about how they want to be helped when sexual abuse occurs. I want to hear what you have to say. I am doing a study that involves over 90 . children so that I can learn how children are affecred by such people as child protective service workers, police. attorneys, and judges, who try and help kids. The purpose of the study is to find the best way to help kids once they have told their story about sexual abuse. I believe the way I can do this is to talk to you and other kids about their experiences with professional people. I, or the people I have trained to help me, will not be asking you any quesrions about the sexual abuse itself. At the end of all the interviews I will be writing a paper and be doing trainings for professionals on how to help and support kids after they share their story of sexual abuse. I have interviewed 75 children already. The interviews take only 35 minutes. ' If you choose to participate I will set up an interview most likely at your home. I will be asking you three different sets of questions. The quesrions I will be asking all have to do with your thoughts and feelings regarding how professionals treated you once you told someone what had happened If any quesn'ons are too uncomfortable you do not have to answer them. There will be no names used in my paper so that all information you give me will be confidential. For your time and willingness to participate I will be giving you $25.00. Please contact me by telephone at 345-2300 as soon as you can if you are interested or have questions. Please reverse the charges to this number. Please call after 4:00 pm. during weekdays and anytime over the weekend. Thanks for your consideration! I hope to hear from you soon! Have a good day! Jim Henry M.S.W. 163 BIBLIOGRAPHY BIBLIOGRAPHY Achenbach, T. M. & Edelbrock, C.S. (1979). The child behavior profile: 11. Boys aged 12-16 and girls aged 6— 12 and 12—16. Journal 9; Consulting and Clinical PsychologyL 31L 223—233 Achenbach, T.M. & Edelbrock C. (1981). Behavioral problems and competencies reported by parents of normal and disturbed children aged 4 through 16. Social Research and Child Development. 46. 7-65. Achenbach, T.M. a Edelbrock, C.S. (1983). Manual for the Child Behavior Checklis; and the Revised Child Behavior Profile. Burlington, Vt.: University of Vermont. Berliner, L. & Barbieri, M.K. (1984). The testimony of the child victims of sexual assault. Journal g; Social Issues, 40, 125-137. Blume, E.S. (1990). Secret Survivors: Uncovering Incegt and Its Aftereffects in Women. New York, New York: John Wiley and Sons. Briere, J., Lanktree, C. & Hernandez, P. (1991). Further data 9n the Trauma Symptgm Checkligt £9; Children: ReliabilitY.validity. and ggnsitivity 39 treatment. Paper present at the annual meeting of the American Psychological Association, San Francisco, CA, August, 1991. Briere, J. (1992). Child Abuse Trauma. Newbury Park, California: Sage Publications. Burgess, A. & Holstrom, L. (1976). Coping behavior of the rape victim. American Journal Psychiatry. 134: 413- 418. Deblinger, E. (1989). Posttraumatic stress in sexually abused, physically abused, and non abused children. Child Abusg and Neglect, 13, 403-407 Everstine, D.S. & Everstine, L. (1993). The Trauma Response: Treatment for Emotional Injury. New York, New York: W.W. Norton & Company. Faller, K.C. (1991). What happens to sexually abused children identified by protective services? Children and Youth Servicgg Review! 13, 101-111. Figley, C.R. (1985). Trauma and its Wake: Volume I; New York, New York: Brunner/Mazel, Inc. 164 165 Figley, C.R. (1986). Trauma and its Wake: Volume II. New York, New York: Brunner/Mazel, Inc. Finkelhor, D. & Browne, A. (1985). The traumatic impact of child sexual abuse: A conceptualization. American Journal 9; Orthopsychiatry. 55. 530-41. Finkelhor, D. (1986). A Sourcebook pm Child Sexual Abuse. Beverly Hills, California: Sage Publication, Inc.. Garmezy, N. (1984). Children under severe stress: Critgue and commentary. American Journal 9; Psychiatry. 140, (3), 358-363. Green, B.L., Wilson, J.P. & Lindy, J.D. (1985) Conceptualizing post-traumatic stress disorder: A psychosocial framework. In C.R. Figley, Trauma and imp wake: The study and treatment 9: post-traumatic §tre§§ disorder. New York, New York: Brunner/Mazel, Inc.. Hancock, M. (1987). Child ngual Abuse. Wheaton. 111.: Harold Shaw Publishing. Herman, J. (1992). Trauma and Recovery. New York, New York: Basic Books. Hindman , J. (1989) Just Bgfore Dawn. Ontario, Oregon: AlesAndria Associates. Horowitz, M.J. (1979) Psychological response to serious life events. In K.Peterson, M.Prout and R.Schwarz, (1989), Post-Traumatic Stre§§ Disorder: A Clinicians Guide. New York, New York: Plenum Press. Monette, D.R., Sullivan, T.J. & DeJong, C.R. (1990). Applied Social Research: Tool for Human Services. Chicago, Illinois: Holt, Rinehart and Winston, Inc. Newberger, E. (1987). Prosecution: A problematic approach to child abuse. Journal 9; Interpersonal Violence. March, 112-117. Ochberg, F.M. (1988). Post-Traumatic Therapy and Victim§ 9; Violence. New York, New York; Brunner/Maze, Inc.. Patten, S., Gatz, Y., Jones, B. & Thomas, D. (1989). Posttraumatic Stress Disorder and the Treatment of Sexual Abuse. Social Work. May 1989, 197-202. Peterson, C.G. & O'Shanick, G.J. (1986). g§ychiatric Aspects pf Trauma. New York, New York: Karger Publishing 166 Peterson, K.C., Prout, M.F. & Schwarz, R.A. (1991). Post-Traumatic Stress Disorder: A Clinician's Guide. New York, New York: Plenum Press. Pynoos, R. & Eth, S. (1984). The child as witness to homicide. Journal pf Social Issues, 40, 87— 108. Pynoos, R. & Eth, S. (1986). Witness to violence: The child interview. Journal pf the American Academy pi Child Psychiatry. 25, 306-319 Pynoos, R. & Nader, K. (1989). Children's memory and proximity to violence. Journal 9: the American Academy pf Child and Adolescent Psychiatry, 28, (2), 236-241. Runyon, D.K., Everson, M., Edelsohn, 6., Hunter, W. &Coulter, M. (1988). Impact of legal intervention on sexually abused children. Journal pf Pediatrics. 113. 647-653. Runyon, D.K. (1991). The emotional impact of societal intervention into child abuse. In G. Goodman and B. Bottoms, (1993), Child VictimsL Child Witnesses: Understanding and Improving Testimony. New York, New York: Guilford Press Runyon, D.K., Everson, M., Hunter, D. & DeVos,E. (1992) The Interysntion Stressors Inventory: A measure pf mpg stress 9: interysption for sexually abused children. Unpublished manuscript from the Department of Social Medicine, Pediatrics and Psychiatry at the University of North Carolina at Chapel Hill School of Medicine. Scrignar, C.B. (1988). Post Traumatic Stress Disorder: Diagposis, Treatmsnt. and Legal Issues. New Orleans, Louisiana: Bruno Press. Sgroi, S. (1988). Vulnerablg Populations; Evaluation and Treatment pf Sexually Abused Children and Adult Survivgms Volums I; Tedesco, J & Schnell, S. (1987). Children's reactions to sexual abuse investigation and litigation. Child Abuse and Neglect, 11, 267—272. Terr, L. (1990). Too Scarsp mp Cry: Psychic Traums mp Childhood. New York, New York: Harper Row, Publishers Weiss, D.H. 8 Berg, R.F. (1982). Child victims of sexual assault: Impact of court procedures. Journal 9; the American Academy pf Child Psychiatry. 21, 513-518. Whitcomb, D., Shapiro, E., & Stellwagon, L. (1985). When the victim is a child. Washington, DC: U.S. 167 Department of Justice. Wilson, J.P. (1989). Trauma: Transformation and Healing. New York, New York: Brunner/Mazel, Inc. Wyatt, G.E. & Powell, G.J. (1988). Lasting Effects 9: Child Sexual Abuse. Newbury Park, California: Sage Publications. Yamamoto, K. (1979). Children's ratings of the stressfulness of experiences. stelopmsntal Psychology, 15, 581-582 "llllllllIIIITIIIIIIIII“