USF Libraries
USF Digital Collections

A pathway to child sex trafficking in prostitution :

MISSING IMAGE

Material Information

Title:
A pathway to child sex trafficking in prostitution : the impact of strain and risk-inflating responses
Physical Description:
Book
Language:
English
Creator:
Reid, Joan
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Human trafficking
Commercial sexual exploitation of children
Child abuse
Runaways
Family strain
Dissertations, Academic -- Criminology -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Victims of child sex trafficking in prostitution in the United States are often overlooked, misidentified, and among the most underserved type of child victim of crime. The majority of previous research on child sex trafficking has been conducted without a theoretical framework or reliable sampling methods. In this study, a schematic composed of a series of stepping-stones from childhood abuse to prostitution, which has been described by gendered pathways researchers, served as a sensitizing template for the study's development of a strain-reactive pathway into child sex trafficking. Agnew's general strain theory provided the primary theoretical basis for the proposed pathway, supplying both explanations of the generative factors of the pathway and the mechanisms operating within the life trajectory terminating in child sex trafficking in prostitution. Based on this theoretical framework, this study utilized structural equation modeling to examine the pathway by investigating the effects of caregiver strain, child maltreatment, and risk-inflating responses to strain on vulnerability to victimization in child sex trafficking in prostitution. Four structural equation models, incorporating different forms of child maltreatment, were assessed using data from a matched sample of 174 minority females who were residents of one U.S. city and participated in a longitudinal study on the effects of child sexual abuse. Findings show that the occurrence of child maltreatment including child neglect, child physical abuse, and juvenile sexual victimization increased with caregiver strain. Consequentially, neglected and abused children were more likely to have engaged in the risk-inflating responses of running away and earlier initiation of drug or alcohol use, and they also reported higher levels of relational shame. Both running away and early initiation of substance use impacted vulnerability to victimization in child sex trafficking in prostitution. Lastly, implications of the findings related to protection and intervention strategies that are projected to obstruct the progression of minors along the analytically identified pathway into child sex trafficking in prostitution are presented for criminal justice professionals, child protection investigators, and social service providers.
Thesis:
Dissertation (PHD)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Joan Reid.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004567
usfldc handle - e14.4567
System ID:
SFS0027882:00001


This item is only available as the following downloads:


Full Text

PAGE 1

A Pathway to Child Sex Trafficking in Prostitution: The Impact of Strain and Risk Inflating Responses by Joan A. Reid A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Crim inology College of Behavioral and Community Sciences University of South Florida Major Professor: Christine Sellers, Ph.D. Shayne Jones, Ph.D. Christopher Sullivan, Ph.D. Thomas Mieczkowski, Ph.D. Randy Borum, Psy.D. Date of Approval: June 24, 2010 Keywords: human trafficking, commercial sexual exploitation of children, child abuse, runaways, family strain Copyright 2010, Joan A. Reid

PAGE 2

Acknowledgements I would like to acknowledge and thank my committee chair and each committee member for the ir g enerous gift of time, thoughtful and critical review of dr afts, perceptive suggestions, and fitting words of encouragement at the most opportune moments. Without these contributions this dissertation would not have been possible. Dr. Christine Sellers, as major professor and committee chair provided the p ri ncipal guidance for this dissertation Dr. Shayne Jones offered insightful recommendations, enhancing the application of the theoretical framework as well as giving guidance regarding the methodological approach. Dr. Chris topher Sullivan supplied skillful expertise concerning the appropriate utilization of quantitative methods and analyses. Dr. Thomas Loughran, although not a committee member, provided advice on supple mentary analyses of the data that re sulted in sharper clarity of the study findings. Lastly and perhaps most of all, I am grateful to those who motivated me to complete this project by telling me of their perilous journey through childhood, showing me by their enthusiasm and love of life tha t childhood innocence may be stolen, yet the innocent child survives. The data used in this dissertation were made available by the Inter University Consortium for Political and Social Research (ICPSR) ; and have been used by permission. Data from the Ri sk Factors for Violent Victimization of Women in a Major Northeastern City, 1990 1991 and 1996 1997 were originally collected by Jane A. Siegel and Linda M. Williams. Neither the collector of the original data, the funder, the ICPSR or its agents or employ ees bear any responsibility for the analyses or interpretations presented here.

PAGE 3

i Table of Contents List of Tables ................................ ................................ ................................ .................. iii iv List of Figures ................................ ................................ ................................ .................. v vi Abstract ................................ ................................ ................................ .......................... vi viii Chapter 1: Introduction ................................ ................................ ................................ .... 1 Selection of Terms ................................ ................................ ............................... 2 Purpose of the Study ................................ ................................ ........................... 4 Composit ion of the Chapters ................................ ................................ ................ 6 Chapter 2: Child Sex Trafficking Victimization in the United States ................................ 10 Defining Child Sex Trafficking ................................ ................................ ............ 11 The Number of Child Sex Trafficking Victims in the United States ...................... 15 The Ages of Child Sex Trafficking Victims ................................ .......................... 19 Risk Factor s for Victimization in Child Sex Trafficking ................................ ........ 21 Victim Initiation into Child Sex Trafficking in Prostitution ................................ .... 28 Consequences of Victimization in Child Sex Trafficking ................................ ..... 30 The Next Step: Focused Research ................................ ................................ .... 33 Chapter 3: Current Status of Theoretical Models ................................ ........................... 36 Theoretical Explanations of Child Sex Trafficking ................................ ............... 36 Theories of Criminology Applied to Victimization ................................ ................ 39 Theories of Juvenile Victimization and Revictimization ................................ ....... 43 Theories of Population Heterogeneity and State Dependence ........................... 48 Chapter 4: Theoretical Framework ................................ ................................ ................ 59 Gendered Pathways Theory ................................ ................................ .............. 60 Gender and Life Course Theory ................................ ............................. 61 Typology of Female Pathways of Offending ................................ ........... 65 Concept of Blurred Boundaries ................................ ............................... 67 Child Victimization Pathway into Exploitation in Prostitution ................... 68 Pathways Typology as a Sensitizing Conc ept ................................ ......... 69 ................................ ................................ ......... 71 General Strain Theory and Victimization ................................ ................. 73 Consequences of Caregiver and Family Strain ................................ ....... 74

PAGE 4

ii The Strain of Child Maltreatment ................................ ............................ 77 Neglect ................................ ................................ ....................... 78 Physical Abuse ................................ ................................ ........... 79 Sexual Abuse ................................ ................................ .............. 80 Child Maltreatment, Negative Emotion, and R epeat Victimization ........... 81 Risk Inflating Responses to Strain ................................ .......................... 84 Current Study: Identifying a Strain Reactive Pathway to Victimization in Child Sex Trafficking ................................ ................................ .................... 87 Chapter 5: Method ................................ ................................ ................................ ......... 94 General Methodological Overview ................................ ................................ ..... 94 Study Participants and Data Collectio n Procedures ................................ ........... 98 Measures ................................ ................................ ................................ ......... 101 Caregiver Strain ................................ ................................ ................... 101 Child Neglect ................................ ................................ ...................... 104 Child Physical Abuse with Injury ................................ ........................... 105 Juvenile Sexual Victimization ................................ ............................... 106 Age of Initial Drug or Alcohol Use ................................ ......................... 108 Running Away ................................ ................................ ...................... 109 Relational Shame ................................ ................................ ................. 109 Prostituted as a Minor ................................ ................................ ........... 111 Analytic Strategy ................................ ................................ .............................. 112 Measurement Models ................................ ................................ ........... 113 Structural Models ................................ ................................ ................. 113 Weighted Least Squares Mean and Variance Adjusted Estimator ........ 115 Supplementary Models ................................ ................................ ......... 118 Chapter 6: Results ................................ ................................ ................................ ....... 120 Preliminary Analyses ................................ ................................ ....................... 120 Multivariate Analyses ................................ ................................ ....................... 128 Child Neglect Mode l ................................ ................................ ............. 129 Child Physical Abuse Model ................................ ................................ 135 Juvenile Sexual Victimization Model ................................ ..................... 140 Child Maltreatment Model ................................ ................................ ..... 145 Effect of Relational Shame ................................ ................................ ... 151 Chapter 7: Discussion and Conclusion ................................ ................................ ........ 159 Gene ral Overview of the Findings ................................ ................................ .... 160 Policy Implications of the Stu dy ................................ ................................ ........ 169 Limitations of the Study ................................ ................................ .................... 177 Contribution s to Criminological Research ................................ ......................... 183 Di rections for Future Research ................................ ................................ ........ 187 Conclusion ................................ ................................ ................................ ....... 191 List of References ................................ ................................ ................................ ........ 194 About the Author ................................ ................................ ............................... End Page

PAGE 5

iii List of Tables Table 1: Previous Studies on U.S. Youth Sexua lly Exploited while Homeless .............. 26 Table 2: Consequences Commonly Observed in Victims of Child Sex Trafficking ........ 31 Table 3: Descriptive Statistics of Study Variables ................................ ....................... 121 Table 4: Bivariate Correlations of All Study Variables ................................ ................. 123 Table 5: Cross Tabulati ons of Prostituted as a Minor by Juvenile Sexual Victimization ................................ ................................ ................................ .. 124 Table 6: Cross Tabulations of Prostituted as a Minor by Running Away ..................... 125 Table 7: Differences in Means of Initial Age of Drug/Alcohol Use ............................... 126 Table 8: Cross Tabulations of Prostituted as a Minor by Relational Shame ................ 127 Table 9: Unstandardized, Standardized, and Critical Ratio Values for the Child Neglect Model Displayed in Figure 4 ................................ ............................ 131 Table 10: Unstandardized, Standardized, and Critical Ratio Values for the Child Physical Abuse Mo del Displayed in Figure 6 ................................ .............. 136 Table 11: Unstandardized, Standardized, and Critical Ratio Values for the Juvenile Sexual Victimization Model Displayed in Figure 7 ......................... 141

PAGE 6

iv Table 12: Unstandardized, Standardized, and Critical Ratio Values for the Child Maltr eatment Model Displayed in Figure 9 ................................ .................. 147 Table 13: Observed Probabilities of Bring Prostituted as a Minor Based on Running Away an d Level of Relational Shame ................................ ........... 152 Table 14: Collinearity Statistics of the Regression of Being Prostituted as a Minor on Runn ing Away, Relational Sham e, and the Interaction Term ................. 154 Table 15: Logistic Regression of Being Prostituted as a Minor Regressed on Running Away, Relational Sham e, and the Interaction Term ...................... 155 Table 16: Collinearity Statistics of the Regression of Bein g Prostituted as a Minor on Age of Initial Drug or Alcohol Use, Relational Sham e, and the Interaction Term ................................ ................................ ......................... 157 Table 17: Logistic Regression of Being Prostituted as a Minor Regressed on Age of Initial Drug or Alcohol Use, Relational Shame, and the Intera ction Term ................................ ................................ ................................ ........... 158

PAGE 7

v List of Figures Figure 1: Strain R eactive Pathway to Victimization in Child Sex Trafficking in Prostitution ................................ ................................ ................................ .... 92 Figure 2: Hypothesized Strain Reactive Pathway to Victimization in Child Sex Trafficking in Prostitution Including th e Forms of Child Maltreatment ............. 97 Figure 3: Child Neglect Measurement Model ................................ .............................. 129 Figure 4: Child Neglect Structural Equation Model ................................ ...................... 132 Figure 5: Child Physical Abuse and Juvenile Sexual Victimization Measurement Model ................................ ................................ ................................ .......... 135 Figure 6: Child Physical Abu se Structural Equation Model ................................ ......... 137 Figure 7: Juvenile Sexual Victimization Structural Equation Model ............................. 143 Figure 8: Child Ma ltreatment Measurement Model ................................ ..................... 146 Figure 9: Child Maltreatme nt Structural Equation Model ................................ ............. 148 Figure 10: Observed Probabilities of Being Prostit uted as a Minor Based on Running Away a nd Level of Relational Shame ................................ ........... 153

PAGE 8

vi A Pathway to Child Sex Trafficking in Prostitution: The Impact of Strain and Risk Inflating Responses Joan A. Reid Abstract Victims of child sex trafficking in p rostitution in the United States are often overlooked, misidentified, and among the most underserved type of child victim of crime. The majority of previous research on child sex trafficking has been conducted without a theoretical framework or reliable sa mpling methods. In this study, a schematic composed of a series of stepping stones from childhood abuse to prostitution, which has been described by gendered pathways researchers served as a sensitizing template for rea general strain theory provided the primary theoretical basis for the proposed pathway, supplying both explanations of the generative factors of the pathway and the mecha nisms operating within the life traj ectory terminating in child sex trafficking in prostitution. Based on this theoretical framework, this study utilized structural equation modeling to examine the pathway by investigating the effects of caregiver strain, child maltreatment, and risk inflat ing responses to strain on vulnerability to victimization in child sex trafficking in prostitution. Four structural equation models, incorporating different forms of child maltreatment, were assessed using data from a matched sample of 174 minority female s who were residents of one U.S. city and participated in a

PAGE 9

vii longitudinal study on the effects of child sexual abuse. Findings show that the occurrence of child maltreatment including child neglect, child physical abuse, and juvenile sexual victimization i ncreased with caregiver strain. Consequentially, neglected and abused children were more likely to have engaged in the risk inflating responses of running away and earlier initiation of drug or alcohol use, and they also reported higher levels of relation al shame. Both running away and early initiation of substance use impacted vulnerability to victimization in child sex trafficking in prostitution. Lastly, implications of the findings related to protection and intervention strategies that are projected to obstruct the progression of minors along the analytically identified pathway into child sex trafficking in prostitution are presented for criminal justice professionals, child protection investigators, and social service providers.

PAGE 10

1 Chapter 1 Introdu ction Operating in the shadows, the sex trafficking of minors in the United States is considered among the most difficult forms of child maltreatment to expose or investigate (Estes & Weiner, 2001, 2005; Williams & Frederick, 2009). In an era of unprecede nted technological accessibility, coupled with greater assurance of anonymity, the number of children exploited by the commercial sex industry continues to escalate (Cooper, 2005a; Farr, 2005; Finkelhor, Mitchell, & Wolak, 2005; Free & Collins, 2005; Hughe s, 2002, 2005; Malesky, 2005; Trifiletti, 2005). Yet, their tragedy remains virtually invisible within a society reluctant to acknowledge the crime or its victims ( Dalla, 2000; DeMarco, 2004; Fang, 2005; Farley, 2003; Kreston, 2000, 2005; Vieth & Ragland, 2005; Williams & Frederick, 2009; Whitehead, 2008). While using a tiered system to assess and sanction other countries based on their attempts to combat human trafficking (DeStefano, 2007; U.S. Department of State [DOS], 2009), the United States is fail ing to combat trafficking in prostitution of U.S. children in its major cities (Estes & Wiener, 2001, 2005; Flowers, 1998, 2001; Wilson & Dalton, 2008) and in rural America (Heineman, Shelton, & Anton, 2006; Vieth & Ragland, 2005). Among the populations o f child victims of crime, sexually exploited minors entrapped in prostitution are the most marginalized overlooked by society, by law enforcement, by social services, and by scholarly research (Clawson & Dutch, 2008; Vieth & Ragland, 2005; Williams & Fre derick, 2009).

PAGE 11

2 Selection of Terms A variety of labels have been applied to minors used for prostitution in the illegal commercial sex industry, from child/juvenile/teen prostitute or sex worker to child sex trafficking victim (Grant, 2005). The conflic ting terminology prevents the identification of victims, impedes efforts to recover victims from traffickers, hinders criminal investigations and prosecutions of traffickers, and hampers the delivery of victim services (Grant, 2005; Shared Hope Internation al [SHI], 2008). Ultimately, this disparity in labeling results in discrimination in the treatment of the minor victims of sex trafficking, depending on the particular form of sexual exploitation to which the minor has been subjected (SHI, 2008). In man y instances when a prostituted minor comes to the attention of state or local authorities, instead of the offense being treated as a case of child victimization, the crime is often classified as a public nuisance crime by law enforcement and the minor is a rrested (Adelson, 2008; Albanese, 2007; Federal Advisory Committee on Juvenile Justice [FACJJ], 2007; Finkelhor & Ormond, 2004; Mitchell, Finkelhor, & Wolak, 2009; National Center for Missing and Exploited Children [NCMEC], n.d.). In contrast, the Federa l Trafficking Victims Protection Act (TVPA, 2000) and its reauthorizations (Trafficking Victims Protection Reauthorization Act [TVPRA] of 2003, 2005, 2008) have defined all minors involved in commercial sex acts (e.g., prostitution, pornography) as victims of trafficking, including minors who are U.S. citizens or lawful The U.S. Dep artment of Justice Model State Anti Trafficking Criminal Statute states that

PAGE 12

3 buying the sexual services of a minor constitutes child sex trafficking even if the involvement of a third party or trafficker profiting from the sex act is not established (Adels on, 2008; U.S. Department of Justice [DOJ], n.d.). e they falsely sanitize the brutality of this ignoring the exploitive role of the buyer and the sex trafficker (Grant, 2005; Hanna, 2002; Vieth & Ragland, 2005). Research has noted that sex traffickers control the majority of the girls in prostitution in the United States (Albanese, 2007; Estes & Weiner, 2001, 2005; NCMEC, 2008 ) and most commonly exploit younger girls and runaways (Clawson, 2009; Sheridan & VanPelt, 2005; Williamson & Cluse Tolar, 2002; Vieth & Ragland, 2005). Utilizing language analysis, researchers have examined the effects of word selection in the print med ia to describe sexually exploited children (Goddard, De Bortoli, Saunders & Tucci, 2005). Placing terms on a continuum, they found that: to adult prostitution, rather th inherits the adult prostitution discourse, it moves along the continuum and away from the concepts of child sexual abuse and victimization. The child is constructed as an accomplice to his or her own sexual abuse, the effect of which is a redefinition of the offence and the offender (p. 286).

PAGE 13

4 Such terminology is employed by the print media in order to spare readers from the This the minimization of the crime committed against them (Goddard et al., 2005, p. 278). In accordance with the results from this language analysis and adhering to the terms used in the TVPA of 2000 and the subsequent reauthorizations, in this paper the term chosen to denote minors victimized by commercial sexual exploitation is child sex trafficking victim. In this paper the term chosen for those who profit by sexually exploiting children is sex traffickers Purpose of the Study The scant information gleaned from the few existing studies on the topic of child sex trafficking has consistently exposed both individual and community risk markers that affect the like lihood of a minor becoming entrapped in prostitution (Estes & Wiener, 2005; Clawson, 2009). However, research has not yet established, nor endeavored to explore, how these risks develop and interact to create the heightened vulnerability observed in certa in minors (Williams & Fred erick, 2009). Theoretical propositions have not been proffered to advance a fuller understanding of the process that generates The purpose of the current study is to design and assess a probable explanation of the processes inducing the entrapment of victims in child sex trafficking, specifically focusing on prostituted girls. This particular population of sex trafficking victims was chosen because girls face higher risk for entrapment in sex trafficking than boys (Clawson, 2009) and prostituted minors are the most marginalized and overlooked type

PAGE 14

5 of child sex trafficking victim (Williams & Frederick, 2009). Incorporating the typically observed risk markers of such victims a nd circumspectly drawing from the propositional components of sound criminological theory, this study delineates a process or pathway into victimization in child sex trafficking in prostitution. As a history of escalating victimizations and multiple chil dhood adversities has been frequently documented in child sex trafficking victims (Albanese, 2007; Clawson, 2009; Estes & Wiener, 2005), two theories that articulate cogent explanations concerning the cumulative effects of harmful events or conditions are examined and provide the foundation for the theoretical framework of this study. Specifically using descriptive research from gendered pathways theory as an initial starting point and applying key propositions of general strain theory, this study first c rafts and then analytically tests a theoretically based pathway. A sample of high risk individuals who possess several primary, identified risk factors for victimizati on in child sex trafficking was used to assess the proposed pathway. Strategically util izing data collected from 174 females of minority and lower socio economic status living in an urban environment, the research based theoretical framework is tested, endeavoring to determine how certain risk factors merge to increase vulnerability to victi mization in child sex trafficking. The data used for this study were originally co llected for use in a longitudinal study of sexually abused children (Siegel & Williams, 2001a, 2001b). The members of the original sample of sexually abused girls were later matched to similar individuals without such abuse histories. Hence, the total sample includes girls with prior sexual abuse histories as well as those who did not have such a history at the outset of the study. By exploring the process by which girls fr om this sample became entrapped in

PAGE 15

6 prostitution as minors, this study provides empirically supported information that may be useful in the development of effective prevention education for at risk youth and in guiding the professionals engaged in serving t hem. The findings from the study also add to the limited body of available empirical knowledge, essential for the creation of effective therapeutic interventions for prostituted minors. Composition of the Chapters To provide a more complete understanding of the crime explored by this study, the second chapter briefly presents an overview of child sex trafficking. Within this chapter, information is reviewed that has been gathered on child sex trafficking victims in the United States, both international an d domestic. Prior research focusing on U.S. minors entrapped in prostitution is covered, including figures projecting the number of child sex trafficking victims in the United States, the estimated average age of minors at initiation into sex trafficking, the empirically supported risk factors, and the broad spectrum of short term and long term consequences endured by child sex trafficking victims. As little criminological theory has been applied to the specific problem of child sex trafficking in prostitu tion (Bender, in press; Graycar & McCusker, 2007), the third chapter explores previous research regarding the victimization of minors, applying such research to this elusive and inadequately studied crime against children. First, the chapter reviews the c urrent status of general theoretical explanations of child sex trafficking, identifying the crucial role to be filled by criminological theory. Next, studies investigating the interconnectedness of offending and victimization are reviewed, endeavoring to determine whether current criminological theories predicting offending

PAGE 16

7 are sufficient to encompass and explain the victimization of minors. The latter half of the third chapter reviews the empirically explored explanations of heightened susceptibility to victimization of children and adolescents, highlighting how these explanations might apply to child sex trafficking victims. Finally, as numerous reports have indicated that the majority of minors entrapped in sex trafficking have experienced prior victim izations (Albanese, 2007; Estes & Weiner, 2001, 2005), the last section of the third chapter outlines two major perspectives framing persistence in certain behaviors or the episodic repetition of specified events, including a discussion of their conceptual contribution to understanding the repeat victimization of minors. As the objective of this study is to progress beyond the use of a risk factor rm and Sampson, 2006, p. 2) by which minors become entrapped in prostitution. This chapter includes the review of two criminological theories, gendered pathways theory and general strain theory, including a discussion of their application to child sex tr afficking victims. Using a descriptive pathway followed by girls into prostitution previously documented by gendered pathways researchers as a preliminary blueprint and drawing upon essential theoretical components of general strain theory, a model pathw ay is further defined, explicating the developmental sequences and processes contributing to vulnerability to victimization in child sex trafficking in prostitution. The fifth chapter covers the methodological plan used for testing the theorized pathway pr esented in the fourth chapter. Given that the great majority of research on child sex trafficking victims has been qualitative in nature, the current study utilizes

PAGE 17

8 structural equation modeling to identify a pathway into child sex trafficking in prostitut ion. Structural equation modeling is used to assess whether the propositions hypothesized within this study can withstand statistical testing. This quantitative extension in methodology represents an important advancement in research on the topic of chil d sex trafficking. From a research perspective, it is important to offer multivariate, quantitative evidence of the existence of the posited pathway into child sex trafficking In this manner, t he study quantitatively identifies a pathway of escalating v ictimization into child sex trafficking in prostitution and provides further validation of the previous findings on this topic that were based primarily on qualitative research methods As noted, the pathway was tested utili zing data drawn from a longitud inal study of African American girls, many of whom had documented sexual abuse histories, who were followed into adulthood. The fifth chapter includes a detailed description the study participants and the data collection procedures. The measures incorpor ated in the structural equation model are reviewed, accompanied by assessments of their validity and reliability. Lastly, the analytic strategy used in the current study is outlined, including the steps followed in identifying and assessing the four struc tural equation models included in the study that are each reflective of the theorized pathway. The results of the current study are presented in Chapter 6. First, the descriptive statistics of the study sample are reported to provide a greater understand ing of the data being utilized to test the suitability of the proposed models. In addition, bivariate analyses are reported, noting the strength, direction, and statistical significance of the associations between the 15 observed variables selected for in clusion in the structural equation models with the objective of evaluating the necessity and appropriateness of conducting further analysis. In the latter half of the chapter, results from the multivariate

PAGE 18

9 analyses are reported. Assessments of all four models are covered, including results from: (a) the measurement model, (b) the structural equation model, and (c) the supplementary model exploring the effect of negative emotion. The results from the analyses of each of the four models are sequentially r eviewed. Chapter 7 includes a discussion of the results of the analyses. Within this chapter, the implications stemming from the findings such as recommendations for new policies, prevention strategies, and the application of effective treatment intervent ions are considered. Limitations of the current study are also described, several contributions of this study to scholarly literature are highlighted, and suggestions for further research are offered.

PAGE 19

10 Chapter 2 Child Sex Trafficking Victimization in the United States Child sex trafficking is classified within the general crime category of trafficking in persons or human trafficking (DOS, 2008, 2009; Fong & Cardoso, in press; TVPA, 2000). First and foremost c onsidered an acute human rights issue, hum an trafficking deprives thousands of men, women, and children of their fundamental rights to human dignity and personal freedoms (DOS, 2009; Farrell & Fahy, 2009; Kelly, 2005; Lehti & Aromaa, 2007). Child sex trafficking is deemed to be a particularly int olerable form of human trafficking, due to the natural and inherent vulnerability of children, possessing both physical and psychological weakness and immaturity (International Labour Organization [ILO], n.d., 2008; United Nations, 1995; Vieth & Ragland, 2 005). As a global economic and health risk, sex trafficking of women and children has been investigated and found to contribute to the transmission of disease, creating overwhelming health costs to local communities, and ultimately sustaining poverty by h indering economic and social development (Agrusa, 2003; Leung, 2003; Silverman et al., 2006, 2007). In some underdeveloped countries the prostitution of children contributes significantly to the gross national product, and the lives of children are sacrif iced for t he short term economic benefit (Farr, 2005; Leth, 2005). Providing enormous financial profits for organized crime, trafficking in children has been exposed

PAGE 20

11 as a source of corruption and violence 1 (Belser, 2005; DOS, 2009; Farr, 2005; Graycar & M cCusker, 2007; Kelly, 2005; Raymond, Hughes, & Gomez, 2010; Shlyk, 2007; Studnicka, 2010; Williams, 2008 ). Accordingly, trafficking in persons, whether adults or children, has been declared to be more than a human rights issue; instead, it is now perceive d as a complex problem involving the criminal justice systems throughout the world (Cooper, 2005b; Finckenauer & Schrock, 2003; Friesendorf 2010; Kreston, 2005; Lebov, 2010; Musto, 2009; Wilson & Dalton, 2008; Zhang, 2009). As if slowly awakening to th e nightmarish reality of this most egregious violation of human rights, child sex trafficking is now preeminently featured as a critical concern of the international community (Cooper, 2005a; ILO, 2008; United Nations, 1995). However, the definition of c hild sex trafficking remains ambiguous and many misconceptions regarding this form of child maltreatment still exist (Adelson, 2008; Clawson, 2009; Laczko & Gramegna, 2003; Mitchell et al., 2009; Musto, 2009; Schauer & Wheaton, 2006; Skilbrei & Tveit, 2008 ) To provide the most clarity and precision, the legal definitions provided within the Trafficking Victims Protection Act (TVPA) of 2000 will serve as the primary basis for defining child sex trafficking victimization within this paper. Defining Child S ex Trafficking TVPA (2000) delineated child sex trafficking as a severe form of trafficking in p ersons. 1 Annual profits reaching $5 billion dollars are estimated from the prostitution of children worldwide (United Nations, 1995).

PAGE 21

12 which a commercial sex act is induced by force, fraud or coercion 2 or in which the person induced to perform such act is under 18 years of age; or (b) the re cruitment, harboring, transportation, provision, or obtaining of a person for labor or services, through the use of force, fraud or coercion, for the purpose of subjecting that person to involuntary servitude, peonage (where someone is held against their w ill to pay off a debt), is one of the many forms of trafficking in persons (DOS, 2008, 2009) and is categorized as a severe form of trafficking (TVPA, 2000). If an individu al has not reached the age of 18, the use of force, fraud, or coercion is not required for any involvement in commercial sex industry to be deemed sex trafficking. Furthermore, according to the U.S. Department of Justice Model State Anti Trafficking Crimi nal Statute, buying the sexual services of a minor constitutes child sex trafficking, and the buyer is considered guilty of trafficking when there is no third party or trafficker profiting from the sex act (Adelson, 2008; DOJ, n.d.). According to this def inition, child sex trafficking would comprise all 2 The U.S. Department of Health and Human Services [HHS] defined force, fraud, and coercion as follows: Force involves the use of rape, beatings and confinement to control victims. Forceful violence is used especially during the early stages of victimization, known as the control. Fraud often involves false offers that induce people into trafficking situations. For example, women and children will reply to advertisements promising jobs as waitresses, m aids, and dancers in other countries and are then trafficked for purposes of prostitution once they arrive at their destinations. Coercion involves threats of serious harm to, or physical restraint of, any person; any scheme, plan or pattern intended to c ause a person to believe that failure to perform an act would result in serious harm to or physical restraint against any person; or the abuse or threatened abuse of the legal process (HHS, n.d.a, para. 7 9).

PAGE 22

13 from the initial recruitment of the minor into commercial sex exploitation to the buying of the sexual services of th e minor (Adelson, 2008; Kreston, 2005; Moossy, 2009). In the TVPA, minors are defined as all individuals who have not yet reached the age of 18. This age demarcation of childhood is in agreement with the definition of a child by the Convention on the Ri ghts of the Child (Office of the United Nations High Commissioner for Human Rights, 1989) and in The Worst Forms of Child Labour Convention No. 182, which was ratified by 169 member states of the International Labour Organization of the United Nations (ILO n.d.). According to the ratified convention, this definition of a child is binding regardless of varied or distinct existing national legislation. Even if countries have legalized prostitution, they must prohibit commercial sexual exploitation of any individual under the age of 18 (ILO, n.d.). The TVPA of 2000 also provided a clear definition of commercial sexual activity, of value is given to or received by any pers child sex trafficking victims would include all minors exploited in any commercial sexual activity. Commercial sexual exploitation of children may include the selling of minors for prostitution, the production of pornography, str ipping and nude dancing, or live sex shows. The for the sexual act of the minor (TVPA, 2000, Sec. 103(3)). The commercial exchange can involve payment or compensation in the form of money, services, or goods (TVPA, 2000). The ILO incorporated a similar characterization, stating that the commercial

PAGE 23

14 and specifies forms of commercial sexual exploitation of children as (1) procuring or offering children for prostitution, (2) child sex tourism, (3) the production, promotion, and distribution of child pornography, and (4) the use of children in private or public sex shows (ILO, n.d.). Summarizing, child sex trafficking includes the entire continuum of acts involved in recruiting, transporting, harboring, maintaining, and procuring individuals, who have not reached the age of 18, for the purpose of commerc ial sexual exploitation (Kreston, 2005; Moossy, 2009). Due to the conventional meaning of the term trafficking, a widespread misconception is that child sex trafficking involves the clandestine movement of children across international or interstate bord ers (Logan, Walker, & Hunt, 2009). Although the movement of child sex trafficking victims is commonly practiced to hamper detection of criminal activities by local law enforcement, it is not a necessary component of child sex trafficking (Clawson, 2009). A ccording to the definitions distinguishing trafficking from suggest that 25% of chi ld sex trafficking victims never leave home, but instead are exploited by family members (Estes & Weiner, 2001, 2005). Within the context of child sex trafficking, the term trafficking applies to the commercial component of the crime or exchange of paymen t, referring to the financial profit gained from the sex acts of minors. In drug, arms, or other types of illegal trafficking, merchandise is bought and sold or sell trafficker or brothel owner. Yet, unlike illegal transactions involving drugs or arms, child sex trafficking victims may also be sold over and over by traffickers without being moved

PAGE 24

15 from one location to another (Farr, 2005; Finckenauer & Schrock, 2003; Logan et al., 2009; Venkatraman, Jacob, & Henley, 2005). Lastly, the most commonly held misconception regarding child sex trafficking victims in the United States is that these victims cannot be U.S. citizens or legal residents, but rather it is assumed that child sex trafficking victims are minors who have been illegally transported across international borders (Adelson, 2008; Fong & Cardoso, in press; Logan et al., 2009). However, the ILO recognized that victims may be Although the TVPA of 2000 was primarily targeted at protecting international victims trafficked into the Unite d States, it was not intended to exclude the protection of domestically trafficked victims (Adelson, 2008). In legislative deliberations prior to the difference for many A merican girls, mostly runaways who are then victimized by the Moreover the subsequent reauthorizations, the Trafficking Victims Protection Reauthorization Act (TVPRA) of 20 03, 2005, and 2008, provided supplementary resources for the fight against human trafficking, including the provision of federal monies for programs to assist state and local law enforcement efforts in combating regional cases of human trafficking and to e xpand victim assistance programs to trafficking victims who are U.S. citizens or legal residents. The Number of Child Sex Trafficking Victims in the United States As a result of its secret and criminal nature, data on the number of international child se x trafficking victims in the United States remain elusive and obscured (Albanese,

PAGE 25

16 cknowledging the complexities involved in attaching a firm figure to this illegal form of child labor, by us ing standardized methodological guidelines to ensure comparability across countries and over time the ILO estimated that annually 1.8 million children are exploited worldwide in the commercial sex industry (ILO, n.d., 2008). From the passing of the TVPA of 2000 until September 2007, a total of 142 minors had been identified as international trafficking victims in the United States, completed the lengthy legal process of obtaining victim status from Federal government, and subsequently obtained the special ly designated trafficking victim visa (T visa) in Estes and Wiener (2001, 2005) estimated the number of international child sex trafficking victims in the United States to be 17,000; others have estimated the number of international victims trafficked into the United States every year to be 14,500 to 17,500, including both children and adults (HHS, n.d.b). Many believe that these mp, 2008b). Several associated figures support that assumption. Every year, approximately 100,000 unaccompanied children Most are directly returned to their country of ori gin and approximately 7,000 to 10,000 HHS, 2008). The majority of the apprehended minors are Mexicans or youth from Central America (Barnett, 2004). Many of these detaine d youth report that they are seeking to join distant relatives, such as aunts or cousins, who have already entered into the United

PAGE 26

17 States legally or illegally (Barnett, 2004). Others make the journey to the United States in search of a better life, as com these minors, such as whether they are victims of trafficking and in danger of revictimization (Barnett, 200 higher number of undocumented minors do successfully c ross U.S. borders undetected. So, while the number of identified child victims of international sex trafficking remains small, concern exists that a larger population of international child victims remains unidentified and hidden. Obtaining reliable estimates of the number of domestic child sex trafficking victims in the United States is similarly elusive and uncertain. However, the available figu efforts toward prevention and victim protection. For example, during 2008 the National Center for Missing and Exploited Children (NCEMC) received more than 102,000 hotline ti ps regarding child sexual exploitation and analyzed more than nine million child pornographic images attempting to locate and rescue child victims (NCEMC, 2008). Of the approximately 800,000 U.S. children who are reported missing each year, it is estimate d that 58,000 children are abducted by non family members ( Sedlak, Finkelhor, Hammer, & Schultz, 2002, p. 10 .) 3 The leading motive in such abductions is sexual and 3 A ily perpetrator takes a child by the use of physical force or threat of bodily harm or detains a child for at least 1 hour in an isolated place by the use of physical force or threat of bodily harm without lawful authority or parental permission; or when a child who is younger than 15 years old or is mentally incompetent, without lawful authority or parental permission, is taken or detained by or voluntarily accompanies a nonfamily perpetrator es the intention to keep the child permanentl

PAGE 27

18 the majority of the victims are female teenagers (Finkelhor, Hammer, & Sedlak, 2002, p. 7, 10). Furthermore, based on calculations of risk factors prevalent in certain populations of U.S. minors, it has been estimated that 10% or runaways are at risk for victimization in domestic commercial sexual exploitation (Este s & Weiner, 2001). The National Runaway Switchboard (NRS) estimates that from 1.6 to 2.8 million children run away in the United States every year (NRS, n.d.) and the U.S. homeless youth population is estimated to be between 1.3 to 1.5 million (Baron, 200 3). Calculations based on these figures provide reason for concern, cautioning that every year approximately 130,000 to 420,000 U.S. minors may be in danger of entrapment in sexual exploitation in the United States. Other estimates of the number of U.S minors exploited in prostitution range from 100,000 to three million (Fong & Cardoso, 2009; Friedman, 2007; Vieth & Ragland, 2005). Two supplementary figures are available that contribute further information on the number of domestically trafficked min ors in the United States. Ironically among the governmental agencies serving minors, juvenile justice agencies are most likely to keep data or track child sex trafficking victims as they continue to be detained as juvenile delinquents (Clawson, 2009; SHI, 2008; Halter, in press; Mitchell et al., 2009; Synder & Sickmund, 2006 ). For example in 2003, 1,400 youth were arrested for prostitution in the United States. Of these arrested youth, 69% were female and 14% were 14 years old or younger (Synder & Sickmu nd, 2006). Although the overall arrest rate for juveniles is in decline, the number of arrests for prostitution has increased 31 percent between 1994 and 2003 (Synder & Sickmund, 2006). In addition, s ince its inception in June 2003

PAGE 28

19 through October 2009, the Federal Innocence Lost Initiative has recovered over 900 sexually exploited minors, rescued as a result of prostitution stings in various cities across the United States (Federal Bureau of Investigation [FBI], n.d.). Despite the fact that the number of recovered minor victims, those who were either internationally or domestically trafficked, remains relatively low, concern exists that many more minors are in danger of being victimized or currently are entrapped in child sex trafficking. One explanat ion for the low number of identified and rescued minor victims is that when the authorities are investigating child sex crimes, their primary objective is the apprehension of the criminals, not the rescue of victims (Jones, 2005; Taylor & Quayle, 2003). A s a result, the number of recovered victims does not necessarily accurately reflect the magnitude of the problem. For instance, although it is known that hundreds of thousands of images of child pornography exist on the Internet (NCMEC, 2008), as of 2004 only 261 victims had been identified and removed from their abusive situation (from Interpol presentation by McCulloch, cited in Jones, 2005). The Ages of Child Sex Trafficking Victims Equally concerning are reports regarding the young ages of children bei ng victimized in sex trafficking (Associated Press, 2009; Clawson, 2009; Farr, 2005; Gray, 2005; Taylor & Quayle, 2005). The number of pornographic images of children, even infants, being sexually battered available via the Internet is escalating exponent ially (Cooper, 2005c; DeMarco, 2004; Hughes, 2002, 2005; Jones, 2005; Taylor & Quayle, 2005). For example, Taylor and Quayle (2005) compared the abusive images of children available in 1999 to those available in 2003. In 2003 the images were of high qual

PAGE 29

20 toddlers and babies . more abusive, sometimes showing bondage and other sadistic advances pornographic images sent via the Internet were limited to scanned copies of magazines or child pictures from medical reference books, the pornographic images of children currently available are high quality digital moving images with clear audio recordin gs (Jones, 2005; Taylor & Quayle, 2005). The falling age of child victims of pornography suggests an impending lowering of the age of prostituted minors. For once children or adolescents become victimized in one form of sexual exploitation, they tend to b e multiply victimized in the various forms (Klain, Davies, & Hicks, 2001; Hughes, 2005; Taylor & Quayle, 2005). A research estimate of the average age of entry into prostitution for girls in the United States is 12 to 14 (Boxill & Richardson, 2007; Clawso n, 2009; Estes & Weiner, 2001, Gray, 2005; Williams & Frederick, 2009). Boys have been found to enter at even younger ages than girls ( Curtis, Terry, Dank, Dombrowski, & Khan, 2008 ; Estes and Weiner, 2001). In the most recently reported federal child pros titution sting, the youngest victim rescued was 10 years old (Stone, 2009). Due to a fear of contracting AIDS or other sexually transmissible infections (STI), minors (F ang, 2005; Farr, 2005; Hanna, 2002; Hughes, 2005; Kreston, 2000, 2005). Those seeking to have sex with younger minors believe that due to their lack of prior sexual intercourse they will not yet be infected with STI (Farr, 2005; Fang, 2005). The reality is far different; family members or sex traffickers have most likely previously sexually abused the minors, even the youngest (Albanese, 2007). Due to the greater

PAGE 30

21 likelihood of younger minors suffering anal or vaginal tearing during intercourse, plus poss essing an underdeveloped genital tract with less efficient natural protectants against infections than adults, younger minors face greater risk for contracting and subsequently passing on STI (Farr, 2005; Leth, 2005; Kreston, 2001, 2005; Hughes, 2002). In addition, buyers of sex with children will often pay more for sex without a condom and young victims lack the power or even awareness to demand protected or safe sex, increasing their risk for infection (Farley, 2004; Farr, 2005; Leech, 2002; Leth, 2005; Raymond, 2004; Raymond et al., 2010; Sheridan & VanPelt, 2005). Beyond this myth presuming a low risk of STI in younger minors, those who sexually abuse children advocate other myths or rationalizations that justify the sexual exploitation of young minors including the fallacious notions that entrapped children are nothing more than chattel and entirely impervious to the abusive experience, that younger children will not remember the abuse, and that children take pleasure in the sexual experience (Cooper, 2005a; Farr, 2005). Risk Factors for Victimization in Child Sex Trafficking Despite the concentration of international policy and intense media attention on the topic of global sex trafficking in women and children (for review, see Farrell & Fahy, 2009), r & Aromaa, 2007; Schloenhardt, Beirne, & Corsbie, 2009; Zhang, 2009). The prostitution of minors is considered to be the most understudied form of child sex trafficking (Williams & Frederick, 2009). The relative lack of research focused on this form of child sex trafficking is considered to be due to a number of factors: (1) the hidden nature of

PAGE 31

22 the crime, (2) the intimidation and fear tactics used by sex traffickers that ensure the silence of the prostituted minors, and (3) the common entrapment in prostitution of marginalized youth who have not yet gained significant public concern (Williams & Frede rick, 2009). The few empirical studies that have investigated the risk factors for victimization by sex traffickers have found shared qualities in the psychosocial histories and environmental circumstances of both internationally and domestically traffic ked minors (for reviews, see Clawson, 2009; Estes & Weiner, 2001, 2005; see also, Priebe & Suhr, 2005; Williams & Frederick, 2009). These commonalities include individual risk factors such as youthfulness, inadequate education, limited opportunities for employment, lack of family support, history of sexual abuse, and mental, emotional, or physical disabilities (Clawson, 2009). In a three year study investigating the commercial sexual exploitation of children in the United States, Estes and Weiner (2001, 2005) reviewed risk markers such as family dysfunction, domestic violence, substance abuse and mental illness in family members, history of child physical or sexual abuse, previous sexual assault, being a runaway or thrownaway youth, drug dependency, gang membership, immaturity, and poor sexual decision making (see also, Clawson, 2009; Raphael, 2004 ). In general, females have been found to be at higher risk for sex trafficking and are more likely than males to report trading sex to benefit a trafficker, pimp, boyfriend, or relative (Clawson, 2009; Curtis et al., 2008; Estes & Weiner, 2001, 2005; Saewyc, MacKay, Anderson, & Drozba, 2008). However, both male and female homeless youth are at risk for sexual exploitation (Estes & Weiner, 2001, 2005; Saewyc e t al., 2008). Runaway, homeless boys who self identify as gay, bisexual, and

PAGE 32

23 transgender/transsexual reported high levels of sexual exploitation (Clawson, 2009; Curtis et al., 2008; Estes & Wiener, 2001, 2005). No particular race or ethnicity has been fo und to be at greatest risk for entrapment in prostitution, although being a member of minority group has been found to increase risk in certain countries (Clawson, 2009). In the United States, African American minors are more likely to be arrested due to being prostituted than minors of other races who are being prostituted and have a more difficult time escaping (Clawson, 2009; Flowers, 2001). Being a member of a certain socioeconomic class does not seem to prevent a minor from becoming exploited in the illegal commercial sex industry, but poverty has been found to be a risk factor in a number of studies (for reviews, see Clawson, 2009; Estes & Weiner, 2001, 2005). Girls from lower socioeconomic status were noted as more susceptible to entrapment and con sidered less likely to be able to escape the exploitive situation (Lloyd, 2005). entrapment into prostitution include residing in an urban environment characterized by high c rime and elevated levels of police corruption (Clawson, 2009). Estes and Wiener (2001, 2005) also noted similar situational conditions as risk factors, including the existence of an adult prostitution market and the presence of large numbers of transient males (e.g., conventioneers, tourists, truckers, or military personnel). As previously mentioned, researchers have found that the majority of minors ensnared in prostitution reported a history of childhood sexual abuse or were abandoned by their families (Clawson, 2000; Estes & Weiner, 2001, 2005; Hanna, 2002; FACJJ, 2007; Farley, Lynne, & Cotton, 2005; Layden & Smith, 2005; Raphael, 2004; Rotheram

PAGE 33

24 Borus, Mahler, Koopman & Langabeer, 1996; Whitbeck, Hoyt, Yoder, Cauce, & Paradise, 2001). Researchers foc used on the problem of child sex trafficking have noted consistent patterns in the sexual victimizations of minors, finding that the majority of child sex trafficking victims have experienced sexual victimizations that have escalated in severity (Estes & W einer, 2001, 2005; Albanese, 2007). Albanese (2007) described the commonly reported steps within the escalation of victimization from childhood sexual abuse to child sex trafficking, initiating with sexual abuse by a family member to exploitation in porn ography and ending with entrapment in prostitution. Runaway, homeless, and sheltered youth have been found to be at high risk of entrapment in sex trafficking (for reviews, see Clawson, 2009; Estes & Weiner, 2001, 2005). Unlike the situation in less econo mically developed countries where minors are on the streets due to poverty, war, or parental death, U.S. minors primarily run away from home or are thrown out by their caregivers due to neglect, abuse, or domestic conflict (for review, see Baron, 2003; see also Cauce, Stewart, Whitbeck, Paradise, & Hoyt, 2006; Farrow, 2005; Kempf Leonard & Johansson, 2007; Molnar, Shade, Kral, Booth & Watters, 1998; Vieth & Ragland, 2005; Yoder, Whitbeck, & Hoyt, 2001). Research conducted by Peled and Cohavi (2009) on the meaning of running away for U.S. girls found that girls who ran away perceived leaving home not as a choice, but as an inevitability. The primary reason given by girls for leaving was the failure of their significant family relationships, particularly wi th their mothers (Peled & Cohavi, 2009; see also, Clawson, 2009). The girls in the study felt that they had no alternative and

PAGE 34

25 Studies focused on runaway and homeless youth have consistently gathered information on the level of abuse experienced by youth prior to running or being thrown out. Studies have found that 16% 81% of the youth reported child physical abuse and child sexual abuse was reported by 5% 50% (for reviews see B aron, 2003; Clawson, 2009; Tyler & Johnson, 2006a, 2006b). Over 50% of homeless U.S. youth report some form of child abuse and many report multiple types of abuse, multiple offenders, and repeated abuse (for reviews, see Baron, 2003; Clawson, 2009; Tyler & Johnson, 2006a; 2006b). Such youth, having experienced an upbringing characterized by neglect, and psychological, sexual, and physical abuse, are at high risk for further victimization and entrapment by traffickers (Arata, 2002; Clawson, 2009; Farley & Kelly, 2000; Farrow, 2005; Halcon & Lifson, 2004; Jankowski, Leitenberg, Henning, and Coffey, 2002; Roth & Newman, 1993; Saewyc & Edinburgh, 2010; Saewyc, et al., 2008; Scaer, 2001; Scott, Wolfe, & Wekerle, 2003; Tyler & Johnson, 2006a; Whitbeck et al., 20 01). Over the past two decades, numerous researchers have documented the percentage of runaways who are further victimized by commercial sexual exploitation as defined by the TVPA of 2000 (see Table 1). The terms used for commercial sexual exploitation are not identical in all 23 studies included in Table 1, but each term includes the concept of payment in money or goods for the sex act of a youth. As previously noted, buying the sexual services of a minor constitutes trafficking, and the buyer is consi from the sex act (Adelson, 2008; U.S. Department of Justice [DOJ], n.d.). Thus, the commercial sexual activity that researchers have commonly termed survival sex or tr ading sex for money, shelter, food, or goods would be considered sex trafficking and

PAGE 35

26 the buyer deemed a sex trafficker if the individual being paid for the sex act is under the age of 18 (Adelson, 2008). Table 1. Previous Studies on U.S. Youth S exually Exploited while H omeless 1 st Author (Date ) Location Sample Size Age Research Findings Remafedi (1987) Minneapolis, MN 29 gay homeless youth 15 19 17% were involved in transactional sex (exchange of sex for money or drugs) Yates (1988) Hollywoo d, CA 110 runaway youth 12 24 26% reported survival sex Rotheram Borus (1990) New York, NY 59 homeless youth 12 18 15% engaged in transactional sex (exchange of sex for money/drugs) Rotheram Borus (1992) New York, NY 206 homeless youth 11 18 13% of mal es & 7% of females exchanged sex for money or drugs Greenblatt (1993) Hollywood, CA 93 homeless youth 13 17 33% traded sex for money, food, drugs Kipke (1995) Los Angeles, CA 409 street youth 12 23 43% reported survival sex Kipke (1997) Hollywoo d, CA 732 homeless youth 12 23 46% reported survival sex & prostitution Unger (1998) Los Angeles & San Diego, CA 245 street youth 12 23 8% (12 15 years) & 13% (16 23 years) reported prostitution 1% reported pornography Booth (1999) Denver, CO 244 homeless youth 12 19 22% had engaged in survival sex Clatts (1999) New York, NY 929 homeless youth 12 23 25% reported prostitution 3% reported pornography

PAGE 36

27 Table 1 Continued Ennett (1999) Washington, DC 327 homeless youth 14 21 46.6% reported survival sex Greene (1999) U.S. National sample 1,159 homeless or sheltered youth 12 21 27.5% of street youth/9.5% of sheltered reported survival sex Noel (2000) Northwestern U.S. city 536 homeless youth 13 20 8.7% reported survival sex De Ros a (2001) Hollywood, CA 296 homeless youth 12 23 21.5% reported survival sex Moon (2001) San Francisco 204 homeless youth 14 21 12% reported survival sex Tyler (2001) Seattle, WA 372 homeless youth 13 21 11% reported survival sex Wagner (2001) Seattle, WA 289 homeless youth 13 22 41% reported survival sex or prostitution Halcon (2004) Minneapolis, MN 208 homeless youth 15 22 20% exchanged sex for money, drugs, or others needs Milburn (2006) Los Angeles, CA 618 homeless youth 12 20 6% reporte d engaging in sex work Owen (2006) Minnesota 151 homeless youth 11 21 12% of minors reported exchanging sex for food, etc. Solorio (2006) Los Angeles, CA 688 homeless youth 12 20 12% reported trading sex for money Tyler (2007) Midwestern U.S. city 428 homeless or sheltered youth 16 19 13% had traded sex for food, shelter, money or drugs Saewyc (2010) Minnesota 68 runaway girls 12 15 13.8% had been prostituted Note: Terms for sexual exploitation are the terms used by the original researchers

PAGE 37

28 Within this review of studies displayed in Table 1, a cluster of the studies reported that from one quarter to almost one half of youth were sexually exploited while homeless. The only study using nationally representative samples of U.S. homeless or sheltered youth reported that 27.5% of homeless youth and 9.5% of sheltered youth had engaged in survival sex or the exchange of sex for subsistence needs (Green, 1999, p. n commercial sexual exploitation of children in the U.S., which predicted that 10% 15% of validated by the majority of these studies. Victim Initiation into Child Sex Trafficking in Prostitution While abductions of U.S. minors for exploitation in prostitution have been documented (Wilson & Dalton, 2008), in the majority of observed cases sex traffickers have used manipulation to recruit and entrap disadvantaged, needy, and possibly drug addicted minor victims (Albanese, 2007; Anderson & Michaelson, 2007; Kennedy, Klein, Bristowe, Cooper, & Yuille, 2007; Royal, 1998; Spidel, Greaves, Cooper, Herv, Hare, & Yuille, 2007; Stone, 2009; Williams & Frederick, 2009; Williamson & Cluse Tolar, 2002). Researchers have estimated that sex traffickers control from 50% 90% of the girls in prostitution in the United States (Albanese, 2007; Estes & Weiner, 2001, 2005; NCMEC, 2008). Sex traffickers, or pimps, are most commonly linked t o younger girls and runaways who lack housing, food, and clothing (Clawson, 2009; Sheridan & VanPelt, 2005; Williamson & Cluse Tolar, 2002; Vieth & Ragland, 2005). Emerging research on sex traffickers has suggested that traffickers employ recruiters to sp y out needy youth or runaways by frequenting their typical or routine locations (Albanese, 2007; Anderson &

PAGE 38

29 Michaelson, 2007; Gray, 2005; Vieth & Ragland, 2005). Once dependent on the sex traffickers, minors are coerced into prostitution 4 (Albanese, 2007; Priebe & Suhr, 2005; Williams & Frederick, 2009). Numerous researchers have noted the strong influence of peers in the process of initiation into commercial sexual exploitation (Curtis, et al., 2008; Tyler & Beal, 2010; Williams & Frederick, 2009). Hom eless youth often band together and form a sort of social network, a kind of substitute family (Baron, 2003; Baron, Kennedy, & Forde, 2001; Farrow, 2005; Hagan & McCarthy, 1997). Through peer connections, minors are introduced to methods of surviving on t he streets by more experienced runaways (Curtis, et al., 2008; Williams & Frederick, 2009). In some studies, homeless youth social networks have been found to have a protective effect, with those outside of the network engaging in higher levels of risky b ehaviors (Ennett, Bailey, & Federman, 1999). Further research has revealed that prostituted and psychologically conditioned minors may help their traffickers acquire new recruits from among sheltered youth when they return or are returned by law enforcem ent or child protective services to unsecured locations, such as group homes or runaway shelters (Clawson & Grace, 2007; SHI, 2008). Such recruiting tactics are similar to the common practice employed by international traffickers who use victims as recrui ters. Previously trafficked women will 4 Traffickers have been known to initiate the exploitation process by recording minors in pornography, thereby endeavoring to normalize participation in sexualized behaviors for new recruits (Cooper, 2005c; Klain et al., 2001; Raymond et al., 2010; Stark & Hodgson, 2003). The traffickers subsequently use the pornographic images as blackmail to shame and threaten minors rticipation in further sex acts (Cooper, 2005c; Klain et al., 2001; Stark & Hodgson, 2003). Other studies based on personal accounts of exploitation revealed much more brutal initiation tactics, such as violent and prolonged episodes of rape and physical assault (Albanese, 2007; Farr, 2005; Vieth & Ragland, 2005; Wilson & Dalton, 2008).

PAGE 39

30 be sent by traffickers back to their home village to recruit new victims with the promise of (Farr, 2005; Sethi, 2010). Some ch ildren who are exploited through sex trafficking never leave home, but are victimized and trafficked by their parents, siblings, or boyfriends (Albanese, 2007; Estes & Weiner, 2005; Raymond et al., 2010). When the origin of victimization is someone who is Understandably, this type of betrayal results in greater psychological distress, as the exploiter fills the role of both protector and perpetrator (Finkelhor & Browne, 1985; Herman, 1992; Putna m, 2006; Scaer, 2001). These children are at high risk for exploitation over many years (Estes & Weiner, 2005). Highest risk for children in exploitive families occurs when there is no detection of the sexual exploitation by law enforcement or child prot ective services, if the family frequently moves to avoid detection, and when the family environment includes violence, substance abuse, serious mental illness, and sexual promiscuity (Estes & Weiner, 2005). Consequences of Victimization in Child Sex Traffi cking The pervasive and lasting consequences suffered by minors as a result of being exploited in sex trafficking can best be categorized into physical injuries, psychological harm, and social impairments. Below, Table 2 lists the most commonly observed p roblems among child sex trafficking victims noted by various researchers, clinicians, and social service providers. Frequently observed medical problems include STI, unwanted and high risk teen pregnancy, physical injuries from beatings and rapes, malnour ishment, and drug related complications (Clawson & Grace, 2007; Clawson,

PAGE 40

31 Salomon, & Grace, 2008; Cooper, 2005c; Estes & Wiener, 2005; Fadel, 2005; Leth, 2005; Saewyc & Edinburgh 2010; Vieth & Ragland, 2005). Table 2. Consequences* Commonly O bserved in Victims of Child Sex T rafficking 1 st Author (Date ) Physical Injuries Psychological Harm Social Impairments Cooper (2005c) STI including HIV, drug overdose & death, unwanted pregnancy, genital and bodily injuries Dissociative d isorders, PTSD, depression, low self esteem Behavioral problems, interpersonal problems Estes (2005) Long term health problems and injuries, teen pregnancy, substance abuse Long term emotional injuries Interpersonal difficulties, high likelihood of involv ement in adult prostitution, and other involvement in delinquency and crime Fadel (2005) STI including HIV, hepatitis, unwanted pregnancies, pelvic inflammatory disease, infertility, malnourishment PTSD, low self esteem, shame Lack of education, lack of e mployment skills, behavioral problems Layden (2005) Depression, anxiety, panic disorders, PTSD, eating disorders, substance abuse disorders, personality disorders Interpersonal difficulties, lack of life skills (e.g., employment skills, problem solving s kills, parenting skills, communication skills) Leth (2005) Skin diseases, STI, health consequences due to teenage pregnancy, death, malnourishment, drug problems, general poor health Mood disturbances, self mutilation, suicidal behavior, feeling of inferi ority, self blame, guilt, aggressiveness, depression Marginalization

PAGE 41

32 Table 2 Continued Vieth (2005) STI including HIV, drug addiction, death, malnourishment Low self esteem, self destructive behaviors, self mutilation, suicidal tendencies Marginali zation, criminal involvement Clawson (2007) Injuries from beatings and rapes, including broken bones and wounds, malnutrition, STI, infertility, unwanted pregnancy Posttraumatic Stress disorder (PTSD), psychosomatic health complaints, drug addiction, s hame and guilt, depression, anxiety, self destructive behavior, suicidal ideation and attempts, trauma bonding to trafficker Inability to trust, interpersonal difficulties, lack of life skills Saewyc (2010) STI, teen pregnancy Depression, PTSD, self harmi ng behaviors, suicidal ideation and attempts, substance abuse Low level of family or school support Note: Determining if a condition is a consequence, correlate, or antecedent of child sex trafficking may not be clear due to prior childhood victimizations (Cooper, 2005c). The observed psychological harms include a lengthy list of mental health disorders, such as depression, anxiety, dissociative disorders, substance abuse, suicidal ideations and gestures, and posttraumatic stress disorder (PTSD) (Clawson & Grace, 2007; Cooper, 2005c; Fadel, 2005; Layden & Smith, 2005; Leth, 2005; Saewyc & Edinburgh 2010; Vieth & Ragland, 2005). Notably, low self esteem, guilt, and shame are often mentioned as emotional consequences (Clawson & Grace, 2007; Cooper, 2005c; Fadel, 2005; Leth, 2005; Saewyc & Edinburgh 2010; Vieth & Ragland, 2005). Social impairments include interpersonal relationship and behavioral problems, lack of

PAGE 42

3 3 life skills, and marginalization (Clawson & Grace, 2007; Cooper, 2005c; Estes & Wiener, 2005 ; Fadel, 2005, Layden & Smith, 2005; Leth, 2005; Saewyc & Edinburgh 2010; Vieth & Ragland, 2005). The Next Step: Focused Research As evidenced by this review of the extant literature, primarily based on case study research, information regarding the rec ruitment of child sex t rafficking victims, entrapment experiences, and common risk factors has been collected (Albanese, 2007; 2007; Priebe & Suhr, 2005; Williams & Frederick, 2009). Yet, the few empirical studies on human trafficking have generally been conducted without the use of valid sampling Collett, 2005; Graycar & McCusker, 2007; Kelly, 2005; Lehti & Aromaa 2007; Musto, 2009; Tyldum, & Brunovskis, 2005; Williams & Frederick, 2009; Zhang, 2009). Due to a lack of sufficient available and reliable data, researchers have not employed quantitative analytic strategies to investigate the problem of child sex tra fficking (Musto, 2009; Zhang, 2009). As a result, information is sorely lacking regarding models of vulnerability or common pathways into child sex trafficking; most o entrapment in prostitution (Williams & Frederick, 2009). Researchers have noted that the process of entrapment in child sex trafficking may differ across the various forms of exploitation and across the societal settings or environmental conditions of the victims (Bales, 2007; Kelly, 2005; Williams & Frederick, 2009; Wilson & Dalton, 2008; Winterdyk & Reichel, 2010; Zhang, 2009). In other words,

PAGE 43

34 the routes to victimization in international child sex trafficking may differ from the pathways commonly exp erienced by domestically trafficked victims (Wilson & Dalton, 2009; Zhang, 2009; see also, Wikstrm and Sampson, 2006). Therefore, it has been suggested that researchers may be most effective at finding functional solutions to the problem of child sex tra fficking by purposefully focusing on particular forms of sex trafficking in combination with specific types of victims, thereby exposing common pathways followed by selected types of individuals into certain forms of exploitation. Based on recommendation s regarding the type of research strategies and methodologies deemed most appropriate, this study purposefully has concentrated on domestically trafficked girls involved in prostitution. The focus on domestically trafficked girls was chosen because such m inors comprise the largest group of child sex trafficking victims in the United States (Clawson, 2009; Estes & Weiner, 2005). Investigating minors entrapped in prostitution was selected due to the relative scarcity of research focused on this particular f orm of child sex trafficking (Williams & Frederick, 2009). Beyond an insufficiency of empirically based research and the unavailability of systematically collected data, few theories have been proffered to advance the understanding of child sex traffick ing in prostitution, such as predicting potential pathways or life trajectories. Absent a sound theoretical framework, research impotently depends on atheoretically selected observations or correlations in the data (Wikstrm, 2008) to explain child sex tr afficking, limiting the progress of solution development. In light of the need for a sound theoretical foundation on which to build further understanding of child sex trafficking, the next chapter is focused on the theoretical

PAGE 44

35 explanations of victimizati on that may assist in elucidating and explaining this often overlooked crime against children.

PAGE 45

36 Chapter 3 Current Status of Theoretical Models The beginning of this chapter provides a brief overview of the current status of general theoretical explanatio ns of child sex trafficking, identifying the vital and yet underdeveloped role of criminological theory. Next, studies investigating the intersection of offending and victimization are reviewed, endeavoring to determine whether current criminological theo ries predicting offending are sufficient to encompass and explain juvenile victimization, and if not, pinpoint their shortcomings. The latter half of the chapter reviews the empirically explored explanations of the heightened susceptibility to victimizat ion of children and adolescents, highlighting how these explanations might apply to child sex trafficking victims. Finally, as numerous reports have indicated that the majority of minors entrapped in sex trafficking have experienced prior victimizations ( Albanese, 2007; Estes & Weiner, 2001, 2005), the last section of the chapter outlines two major perspectives framing persistence in certain behaviors or the episodic repetition of specified events, including a discussion of their conceptual contribution to understanding repeat juvenile victimization. Theoretical Explanations of Child Sex Trafficking After a thorough review of research on human trafficking, rk

PAGE 46

37 y women Beyond acknowledging poverty as a universal push factor, feminist theories have offered additional and reasoned explanations for the sex trafficking of women and girls, arguing tha t patriarchal gender arrangements prominent in many cultures support the & Trudeau, 2006; Jeffreys, 2009 2010; Morash, 2006; Raymond et al., 2010). For instance, the d evaluation of girls as economic burdens may lead to their abandonment by their caregivers, even the selling of girls to traffickers. Also, severely limited availability of legitimate employment opportunities for females may force girls and women into sexu ally exploitive relationships, as such arrangements may provide the only viable option for survival or escape from intolerable conditions (Clawson, 2009; Farr, 2005; Morash, 2006). Feminist theories have been particularly attentive to the global industri alization of prostitution and the sex trade that has normalized the use of the bodies of girls and young women to generate enormous profits, with the sex trade becoming embedded within the mainstream corporate business sphere in most countries around the w orld (Jeffreys, 2009). The globalization of the sex industry along with the ease of air travel has facilitated international sex trafficking as well as other forms of forced commercial sexual exploitation (Jeffreys, 2009). These changes in the scale and scope of the international sex trade has produced a new form of sexual colonization by sanctioning

PAGE 47

38 the women of poor countries can be sexually bought by men from rich (Jeffreys, 2009, p. 6). Such explanations of victimization in sex trafficking based in feminist theory do not fully account for the sex trafficking of boys (Dennis, 2008; Saewyc et al., 2 008; Snchez Taylor, 2001, 2006 ). Yet, as previously noted, due to the varied nature of child sex trafficking, distinct and divergent explanations may be needed to account for the different types of victims. Therefore, pervasive inequities between genders due to the devaluing of girls and women may be a gen erative source of the disproportionately high rate of entrapment in child sex trafficking experienced by girls in comparison to boys 5 (Clawson, 2009; Jeffreys, 2009, 2010). Arguably, the crime of child sex trafficking fits best within the traditional dom ain of the field of criminology, as it involves human actors behaving in violation of social norms, resulting in victimizations that engage a comprehensive set of criminal justice responses: police, courts, other agencies of formal social control designate d to establish justice and mete out punishments (Cooper, 2005b; Friesendorf 2010; Kreston, 2005; Musto, 2009; Wilson & Dalton, 2008; Finckenauer & Schrock, 2003; Lebov, 2010 ; Zhang, 2009). However, the field of criminology has yet to fully explore or a ttempt to explain this crime due its transnational features and because many of the contributory behaviors that facilitate child sex trafficking do not map neatly onto conventional crime categories ( Graycar & McCusker, 2007) 5 Global estimates suggest 98% of all victims of forced commercial sexual exploitation are female, with 40 50% estimated to be minors (ILO, 2005). The ILO defines forced labor as labor worker (ILO, 2005, p. 6).

PAGE 48

39 Nevertheless, the remedy to the lack of theoretical models explaining child sex trafficking may exist within the field of criminology. The premise of this study is that criminological theories will render a more complete elucidation of victim susceptibility and the life adversities that place certain minors, specifically certain U.S. girls, on a high risk pathway toward entrapment in child sex trafficking, and thereby possess the potential for illuminating sorely needed solutions. Theories of Criminology Applied to Victimization A s the primary purpose of the current study is to obtain a greater understanding of child victims of sex trafficking, rather than sex traffickers, it is relevant and crucial to consider whether mainstream criminological theories can provide a suitable theor etical framework for such an inquiry. As the field of criminology is relatively d eficient of theories focusing solely on victimization, and even fewer theories are dedicated to explicating the problem of child and adolescent victimization (Finkelhor, 2007 ; Lauritsen & Laub, 2007), criminological theories purposed on explaining delinquency and criminality may be required to explain victimization in child sex trafficking in prostitution. Beyond the fact that child sex trafficking is a crime and most crimin ological theories are considered generally applicable across various forms of crime, the usefulness of criminological theories for explaining victimization in child sex trafficking seems fitting and appropriate for several other reasons. Until recently, b eing prostituted as a minor was viewed as a form of juvenile delinquency. Fuller comprehension of this criminal phenomenon, along with landmark Federal legislation regarding human trafficking such as the passing of the Trafficking Victims Protection Act i n 2000, resulted in a recent paradigm shift regarding the culpability and legal status of prostituted minors.

PAGE 49

40 Prostituted minors are no longer viewed as delinquents, but instead they are now considered by many to be victims of criminal exploitation. Cons idering this history, until very recently current criminological theories had been regarded as sufficient to explain child sex trafficking in prostitution as it was viewed as a form of delinquency. Thus, these theories are likely to continue to have funct ional application to this problem regardless of the fact that the activity is now considered to be a form of victimization. Moreover, many have argued that some theories useful for explaining crime and criminality should also be effective at explaining vi ctims and victimization; and in fact, some suggest that criminality cannot be fully understood in isolation from victimization and vice versa (Lauritsen & Laub, 2007; Lauritsen, Sampson, & Laub, 1991; Ousey, Wilcox, & Brummel, 2008; Schreck, 1999; Schreck, Stewart, & Osgood, 2008). In a quest to understand crime victims and offenders by investigating the overlap and divergence of the two groups, recent studies have investigated patterns in the risk factors of offenders and victims (Baron, 2003; Chen, 200 8; Cuevas, Finkelhor, Turner, and Ormond, 2007; Hay & Evans, 2006; Holtfreter, Reisig, Piquero, & Piquero, 2010; Katz, 2000; Manasse & Ganem, 2009; Tyler & Johnson, 2006a, 2006b; Schreck et al., 2008). For example, Cuevas et al. (2007) endeavored to ident ify a typology of juvenile delinquent victims, seeking to understand pathways and interrelations between delinquency and victimization. Using research based grouping criteria, these researchers identified a typology of delinquent victims with three subtyp es, representing 23% of the youth in their nationally representative sample. Challenging the common assumption that juvenile delinquents and juvenile victims are largely overlapping populations (Bender, in press; Lauritsen & Laub, 2007), the findings from this study indicated that fewer youth could be classified as both delinquent and victim in

PAGE 50

41 comparison to the number of youth who could be classified as either one or the other, with little or no overlap. These findings also indicated that delinquency is more likely to follow victimization, rather than victimization occurring as a consequence of delinquency (see also, Feiring, Miller Johnson, & Cleland, 2007; Hollist, Hughes, & Schaible, 2009; Hay & Evans, 2006; Lansford et al., 2007; Lo, Kim, & Church, 20 08; Katz, 2000; Manasse & Ganem, 2009; Mersky & Reynolds, 2007; Tyler & Johnson, 2006a, 2006b). Several studies, focused on the directionality of the correlation between victimization and offending, found that victimization was an antecedent of delinquenc y, with the relationship mediated by depression or anger (Hay & Evans, 2006; Manasse & Ganem, 2009). However, support for converse directionality, from delinquency to victimization, has also been found (Baron, 2003; Chen, 2008). A study by Schreck et al. (2008) strove to isolate differences in the propensity for involvement in violent offending versus susceptibility to victimization by violent crime. propensity toward offending versus susceptibility to victimization. Moreover, they found that once a particular pathway, either of victimization or offending, was triggered, the specified roles (i.e., victim or offender) remained consistent across time. The majority of ri sk factors included in this study were able to predict the likelihood of involvement in a violent incident, but these factors were not able to differentiate the specific role taken within the criminal event. Only aging and getting drunk were found to infl uence the shifting from one tendency to another, indicating that among the varied risk factors included in the study (e.g., gender, race, poverty, school commitment, parental attachment, emotional distress, peer deviance), getting older and frequently gett ing drunk were the only predictors of role differentiation across time. This study suggests

PAGE 51

42 that although victims and offenders share common risk markers that propel them toward involvement in a criminal event, hence explaining the commonly observed corre lation between offending and victimization (for reviews, see Bender, in press; Lauritsen & Laub, 2007), there may also exist different factors or mechanisms that determine whether an individual begins and persists in the role of victim or offender. While commonalities in risk factors of victims and offenders have been documented thereby indicating that theories of criminality may also elucidate causes of victimization, sufficient theory modification based on the pivotal component of the specified role take n in the criminal event, i.e., perpetrator or victim, would seem to be essential. As noted by Farrington (2000), the inherent problem in using the risk factor fa These research findings regarding the association of offending and victimization imply that similar risk factors are affecting both initiation and persistence in offending an d victimization; yet, differentiating which essential and enduring role is taken may not be possible when research attempts to form depictions of victims and offenders by using the limited lens of the risk factor paradigm. Working from the premise that ri sk factors may only be revealing spurious correlates of causes (Farrington, 2000), explicating mechanisms by which risk markers are bringing about certain outcomes supported by sound theoretical framework becomes of paramount importance (Pawson & Tilley, 1 997; Wikstrm and Sampson, 2006; Wikstrm, 2008). However, only a scant amount of research has drawn upon criminological theory to build an understanding of the contexts and mechanisms producing the observed, elevated levels of victimization experienced b y minors (Baum, 2005), especially when compared to the amount of research and

PAGE 52

43 theory invested in understanding the problem of juvenile delinquency (Finkelhor, 2007; Lauritsen & Laub, 2007; Macmillan, 2001; Ousey et al., 2008). Theories of Juvenile Victimiz ation and Revictimization The behavioral scientists and criminological researchers investigating the heightened level of vulnerability and susceptibility to victimization experienced by children and adolescents initially focused their research efforts on exposing the highly detrimental and lasting effects of victimization during childhood and adolescence (for review, see Arata, Langhinrichsen Runtz, 1993; Fagan, 2005; Finkelhor, 2007; Ford, Stockton, Kaltm an, & Green, 2006; Herman, 1992; Ireland, Smith & Thornberry, 2002; Kaukinen & Demaris, 2005; Macmillan, 2001; Putnam, 2006; Scaer, 2001; Solomon & Heide, 1999). Similar to reports regarding the consequences of entrapment in child sex trafficking, the det rimental effects of childhood victimization result in a broad array of consequences, impairing victims physically, psychologically, and socially (for reviews see Arata et al., 2007; Macmillan, 2001; see also Robst, 2008). In spite of conclusive findings v alidating the critical consequences of child maltreatment and victimization due to the pervasive impairments of essential developmental processes, the study of the victimization of minors has remained sorely underdeveloped and neglected (Finkelhor, 2007; M acmillan, 2001). Theories have not been focused on illuminating why particular minors are victimized more than others, neither identifying the buffering effects of protective factors, nor detecting the interactive effects or cumulative costs of multiple r isk markers (Finkelhor, 2007; Macmillan, 2001; Schreck & Fisher, 2004).

PAGE 53

44 The majority of previous criminological research on juvenile victimization has focused on testing the veracity of concepts based in the routine activity theory of victimization, prim arily investigating whether increased exposure to crime, resultant of youth engaging in risky behaviors with delinquent peers (i.e., partying, drinking, staying out late, gang membership, running away, homelessness) may result in higher rates of victimizat ion (for reviews, see Baron, 2003; Finkelhor, 2007; Lauritsen & Laub, 2007; Schreck et al., 2008; see also, Chen, 2009; DeLisi, Barnes, Beaver, & Gibson, 2009; Higgins, Jennings, Tewksbury, & Gibson, 2009; Taylor, Freng, Esbensen, & Peterson, 2008; Wittebr ood & Nieuwbeerta, 2000). Classified as a criminal opportunity theory, routine activity theory places an importance on changing social contexts in explaining the opportunity for crime (Cohen & Felson, 1979, 2006). Routine activity theory postulates that opportunity for crime by a motivated offender increases due to elevated target suitability and an absence of capable guardianship Cohen and Felson (1979, p. 589) emphasized the importance of control, implying that the loosening of control over the oppor tunity for the convergence of a motivated offender and a suitable target without capable guardians will increase the likelihood that violations will occur. Strongly analogous to concepts postulated by routine activity theory, researchers have also applied control theories to victimization, seeking to explain victimization based on an absence of social control, such as poor parental supervision or a deficiency of collective safeguarding by delinquent peers (Chen, 2009; Lauritsen & Laub, 2007; Piquero, MacDo nald, Dobrin, Daigle, & Cullen, 2005; Schreck, 1999; Schreck et al., 2006; Schreck et al., 2008; Stewart, Elifson, & Sterk, 2004; Vincke, 2007 ). The lack of capable guardianship, in the form of social controls, has been

PAGE 54

45 empi rically supported as a significant issue when addressing juvenile victimization (Chen, 2009; DeLisi et al., 2009; Lauritsen & Laub, 2007; Schreck et al., 2008). Suggesting essential modification of routine activity or lifestyle theories when applied to youth, Finkelhor (2007) articulated that the risk inflating dynamics embedded in routine activity theory should be adapted and considered as environmental conditions rather than problems in the lifestyle choices or routine activities of minors. For exampl e, a child may be at risk for victimization due to caregiver neglect or poor supervision, conditions that have been found to often serve as prerequisites to child victimization (for review, see Schreck et al., 2008). The presence of such risk inflating co nditions may be affecting the likelihood of victimization, rather than risky lifestyle choices or routine activities of the minor. The majority of minors have no control over, nor the opportunity to change their residence, family members, or neighborhoo d peers; their school placement or schoolmates; their place of employment or coworkers; or their routes or modes of transportation (Agnew, 1985). All of these noted settings are considered possible sources of victimization risk according to routine activi ty theory (Cohen & Felson, 1979, 2006; Felson, 2001, 2002; Karmen, 2010; Menard & Huizinga, 2001). The inherent characteristics common to minors related to their age and legal status, such as reliance and dependency upon adults for basic necessities, soci al and psychological immaturity, even possessing a smaller physical stature, are essential factors to be considered when modifying or developing an inclusive explanation of juvenile victimization (Finkelhor, 2007; Steinberg & Scott, 2003).

PAGE 55

46 Beyond specif ics regarding the pervasive and lasting detrimental effects of childhood victimizations (for reviews, see Arata et al., 2007; Macmillan, 2001 ), an additional consistent finding across studies of juvenile victimization is that the majority of victimized you th are multiply and repeatedly victimized (Arata et al., 2007; Barnes, Noll, Putnam, & Trickett, 2009; Classen, Palesh, & Aggarwal, 2005; Finkelhor, Ormrod, Turner, & Hamby, 2005; Menard & Huizinga, 2001), indicating a strong and positive link between past and future victimization. For instance, b ased on self reported incidents of child maltreatment, one study found that less than five percent of respondents reported a single form of abuse occurring in isolation from all other types (Ney, Fung & Wickett, 1 994). Researchers have confirmed that varied forms of child and adolescent victimizations often occur in tandem (Arata et al., 2007). Links have been observed between neglect and sexual abuse; neglect and physical abuse; physical, sexual, and psychologi cal abuse; child maltreatment and sexual exploitation or assault; domestic violence and child abuse; and child physical abuse and dating violence (Arata et al., 2007; Classen et al., 2005; Estes & Weiner, 2005; Fang & Corso, 2007; Herrenkohl, Sousa, Tajima Herrenkohl, & Moylan, 2008; Macmillan, 2001). Prompted by such empirical findings regarding the high rate of revictimization among minors, numerous researchers and theorists have proposed that perhaps juvenile victimizations should not be regarded as i solated incidents but rather as an evolving state or developing condition, in which victimized minors do not experience only one traumatic incident but rather endure a series of victimizations (Finkelhor et al., 2005, p. 20; Perry, Hodges, & Egan, 2001; Ou sey et al., 2008). Finkelhor, Ormond, and Turner (2007) suggested that the observed clustering of juvenile victimizations may be a

PAGE 56

47 there may be a snowballing effec t of victimizations, with each escalating the risk for future victimizations. Similarly, Schneider (2001) reported that the victimization risk increases with each victimization incident and multiple victimizations occurring within a short period of time i s common. Certain victimizations, such as child maltreatment (e.g., neglect, psychological, sexual, or physical abuse) may function as initiation or gateway victimizations due to their traumatic impact, breaching natural childhood protectorates and opening the way to future and possibly more severe types of victimization (Finkelhor et al., 2007, p. 162). For example, it has been clinically observed and empirically validated that child maltreatment and sexual victimization produce an acute lack of self prot ection capabilities, as victims no longer perceive that they have options, overlook warning signs of impending danger, and thereby become highly vulnerable to revictimization (Herman, 1992; Messman Moore & Brown, 2006; Seligman, 1975; Simons & Whitbeck, 19 91). Such deficits in self protective reactions may not be solely psychological, but instead due to neurobiological damage resulting from prior abuse (De Bellis, 2001; Heide & Solomon, 2006; Putnam, 2006; Scaer, 2001). The noted similarity in patterns of revictimization across the majority of child sex trafficking victims further substantiates this concept of victimization of minors as an intensifying condition (Albanese, 2007; Estes & Weiner, 2001, 2005; Simons & Whitbeck, 1991; Williams and Frederick, 2 009). Although the strong association between past and future victimization has been observed, the primary mechanisms, whether biological, psychological, social, ecological, or behavioral, that elevate risk for repeat victimization have not yet been fully identified (for reviews, see Arata, 2002; Barnes et al., 2009; Classen et al., 2005; Finkelhor, 2007;

PAGE 57

48 Fisher, Daigle, & Cullen, 2009; Messman Moore & Long, 2003; Ousey et al, 2008) Previous victimization is not an infallible predictor of future victimiz ation, as not every victimized child suffers further victimization. As converging and diverging pathways into and out of victimization have been observed, identifying common pathways could illuminate the protective and promotive factors that influence rep eat victimization. Clarifying the contributory mechanisms that drive an escalation in victimization, as has been observed in minor victims of sex trafficking (Albanese, 2007; Estes & Weiner, 2001, 2005), could assist in demystifying the processes by which many become entrapped in a perpetuating cycle of victimization. Theories of Population Heterogeneity and State Dependence When addressing persistence in offending, criminologists have drawn upon two distinct, yet possibly compatible, contributory proc esses in explaining the continuation of various kinds of behaviors: state dependence and population heterogeneity (Ezell & Cohen, 2005; Heckman, 1981, 2000; Nagin & Paternoster, 1991, 2000; Savage, 2009) As these processes have been used to explain the c ontinuation of a varied set of behaviors or specified events (e.g., episodic poverty, accidents, unemployment), they would seem to be beneficial in explaining not only persistent offending, but also repeat victimization (Nagin & Paternoster, 1991; Ousey et al., 2008). A helpful analogical portrayal of the mechanics of these two processes, featuring the possession of red and blue balls, was given by Heckman (1981, 2000) and further developed and applied to criminal propensity by Nagin and Paternoster (1991 2000). The red balls represent propensity for choosing certain behaviors or experiencing specified events such as unemployment, victimization, or accidents, and the blue balls

PAGE 58

49 reflect prosocial or protective qualities. Reflecting the population heteroge neity process, red and blue balls are distributed inequitably among individuals in the population. The greater the proportion of red balls that one possesses, the greater the chance is that they will choose a certain behavior or experience a specified even t. Each time a red or blue ball is chosen or experienced, it is then duly replaced. Each individual maintains his/her initial proportion of red balls, so the propensity for certain behaviors or experiencing specified events remains stable over time. Conv ersely, illustrating state dependence, red and blue balls are initially distributed equally among individuals, resulting in each individual having the same initial propensity for choosing certain behaviors or experiencing specified events. In this scenari o, through the utilization of person draws a red ball, and experiences the event, additional new red balls are added to his urn. Subsequent outcomes are affected by previous outcomes because the choice This represents the process by which propensity for crime (or victimization) would intensify or strengthen via the process of state depe ndence. A third, mixed option is also imaginable, in which the initial distribution of red and blue balls is varied within the population, and individuals accrue red or blue balls at varying rates due to prior experiences. Traditionally, the prominent cri minological theories had been built upon the premise of population heterogeneity, focusing primarily on between individual differences (Farrington, 2008). Such theories advocated the existence of a causal trait (e.g., impulsivity, psychopathy, criminal pr opensity, limited cognitive ability) in certain individuals that affects their probability of committing or refraining from crime throughout

PAGE 59

50 their lifetime (Bushway, Brame, & Paternoster, 1999; Farrington, 2008; Nagin & Paternoster, 1991, 2000; Paternoster Dean, Piquero, Mazerolle, & Brame, 1997 ; Piquero, Farrington & Blumstein 2003; Savage, 2009). Similar to features typically associated with developmental theories of crime, most proponents of crime propensity or trait based theories note that childhood familial environments characterized by impaired attachment, harsh punishment, chaotic family upbringing, and/or parental conflict are sources of these traits (Agnew, 2004; Caspi, Lyman, Moffitt, & Silva, 1994; Gottfredson and Hirschi, 1990; Moffitt, 1993; Savage, 2009). According to Nagin and Paternoster (1991, 2000), the prominent theories in criminology that are clearly congruent with this position are control theories (Gottfredson & Hirschi, 1990), theories of personality development (Caspi et al., 19 94), and genetic, biological, or individual constitution theories (Wilson & Herrnstein, 1985). Several of these theories, particularly control theories, purport to not only offer an explanation for offending and persistent offending but also for victimiza tion, the association of offending with victimization, and even revictimization; and control theories have been so tested in criminological research (Campbell Augustine, Wilcox, Ousey, & Clayton, 2002; Chen, 2009; Holtfreter et al., 2010; Lauritsen & Laub, 2007; Piquero et al., 2005; Schreck, 1999; Schreck et al., 2006; Schreck et al., 2008; Schreck, Wright, & Miller, 2002; Stewart et al., 2004). The basic assumptions of control theories in criminology are: (1) that all individuals pursue pleasure and avo id pain, providing a constant motivation for crime; (2) without the presence of restraining factors individuals will commit offending behaviors; and (3) attachments to others or social bonds are the source of restraining factors

PAGE 60

51 (Gottfredson, 2006). Contr ol theories differ according to the emphasis placed on self control and whether social controls are conceived as functioning dependently or independently of self control theory has been empirical ly supported through numerous studies across ages, groups, cultures, crimes, behaviors related to crime, and time periods (Gottfredson, 2006; Pratt, 2009). According to self control theory, the formation of the stable trait of self control occurs via par ent child socialization or the reciprocal bonding process from birth to age eight (Gottfredson & Hirschi, 1990; Gottfredson, 2007). From these early attachments or social bonds, an individual gains self term gain fo r long control theory asserts that delinquency and victimization are the result of a propensity of both victims and offenders to act impulsively and ignore long term consequences (G ottfredson & Hirschi, 1990; Schreck et al., 2008). Impulsivity can result from control (Gottfredson & Hirschi, 1990). This deficiency in self control can linger even into adulthood, leaving the individual with a lasting inability to control impulsivity. Accordingly, the observed correlation between past and future victimization, past and future offending, or even the frequently observed correlation between victimization and offendin g (for review, see Chen, 2009; Lauritsen & Laub, 2007) is not causal, but rather spurious ( Paternoster et al., 1997 ; Sampson & Laub, 1992). A number of objections can be raised to applying the notion of population heterogeneity to crime victims, such as the concern of victim blaming or inappropriately

PAGE 61

52 reassigning the origins of an action to the human agency of the recipient of the action rather than the perpetrator (Boney McCoy & Finkelhor, 1995; Hyman & Steward, 2004; Lauritsen & Laub, 2007). For examp le, Gottfredson and Hirschi argue that between offender differences in the quantity of crime committed exist, but there are not qualitatively distinct types of offenders; all offenders can be placed on one continuum of criminal propensity (Piquero et al., 2003). Therefore, explanations of victimization based on the population heterogeneity paradigm could result in viewing all victims the same, i.e., as possessing a fixed and causal trait of victim propensity (Beaver, Boutwell, Barnes & Cooper, 2009; Laurit sen & Laub, 2007), and subsequently attributing the severity and noted by Lauritsen and Laub (2007), population heterogeneity may not solely result from between indi vidual differences, but also could reflect shared social conditions that Miller, 2008a, 2008b; Ousey et al., 2008; Wikstrm, 2008). Viewing the inquiry from an orthogon al perspective, studies focusing on hunting patterns of sex offenders have noted that the immediate social and economic conditions of victims such as being on the street, obviously needy, and with apparent vulnerability on of victims (Beauregard, Proulx, & Rossmo, 2007; see also, Stevens, 1998). This is congruent with the limited information gathered on strategies used by sex traffickers, which highlights their tendency to entrap younger, disadvantaged, and needy runawa y girls (Clawson, 2009; Sheridan & VanPelt, 2005; Williamson & Cluse Tolar, 2002; Vieth & Ragland, 2005). Moreover, life course criminologists criticize theories based on population heterogeneity for failing to address within individual changes, neglectin g to investigate the effect of childhood and

PAGE 62

53 adolescent environments and behaviors later in life, or to account for the modifying effects of adult social bonding and transitions (Farrington, 2008; Sampson & Laub, 1992, 1993). Challenging the assumptions of population heterogeneity, the state dependence explanation for persistence in behaviors such as crime contends that something inherent in committing crime or experiencing the event critically changes an individual and their life circumstances to such a degree that it alters their likelihood of future offending or of reexperiencing a future occurrence of a comparable event (Nagin & Paternoster, 1991, 2000; Piquero et al., 2003 ). In application to criminal behavior, the state dependency paradigm predicts that crime will beget more crime, either through producing beneficial rewards that pull or entice an individual to commit more crime or through spawning a cascading collection of consequences that pushes a person further into a criminal lifestyle (Nagin & Paternoster, 1991, 2000; Piquero et al., 2003; Paternoster et al., 1997 ). Numerous criminologists have conceptualized the process of state dependence. For instance, Sampson and Laub (2004, p. 133) labeled this adverse compounding of criminal behavior a According to Nagin and Paternoster (1991, 2000), the prominent theories in criminology that are most congruent with this position are labeling theory (Becker, 1963; onal theory of crime, social learning theory (Akers, 1985), and general strain theory (Agnew, 1992). In addressing the issue of persistent offending, Agnew predicted interactive and potentiating effects between has consequences which contribute to further

PAGE 63

54 example of the state dependence process affecting persistent offending viewed through the theoretical lens of general strain theor y, Agnew (2004, 2005a, 2006a) proposed that family strains such as child abuse or neglect could lead a child to run away, become a gang member, or use illegal substances in hopes of resolving intolerable strain. Repetition of offending can be anticipated in this example; a criminal act (i.e., illegal drug use) effectively relieved intolerable strain, reinforcing the likelihood that drug use will delinquent behavior; the perceived beneficial effect of that behavior (and ultimately drug also emerge as risk markers due to drug use, such as increased contact with delinquent peers, inv olvement in the juvenile justice system, or financial demands to pay for drugs (Agnew, 2004; Slocum, Simpson, & Smith, 2005). Further consequences of delinquency, such as rejection by parents, teachers, neighbors, and/or conventional peers, were theorized to foster strain, weaken social control, and escalate associations with delinquent peers (Agnew, 2004, 2007). Additionally, punishment resulting from delinquency, especially if perceived as undeserved or excessive, was assumed to result in anger and frus tration, thereby promoting further delinquency (Agnew, 2000; Hagan & Foster, 2003). Kim, Conger, Elder, and Lorenz (2003) conducted one of the rare, longitudinal studies investigating the connection between strain and persistent offending and concluded th at stressful life events and delinquency have a reciprocal relationship. Also, Hagan and Foster (2003) investigated delinquency as a stressor and found that the consequential negative emotions of anger and depression led to subsequent drinking problems in males. Accordingly, these studies found support for the proposition that delinquency can be

PAGE 64

55 both consequence and source of strain, with the resulting adversity due to offending mediating the association between past and future offending (Hagan & Foster, 2003; Kim et al., 2003). Thus, strain and offending exacerbate each other in a toxic cycle, as offending is perpetuated and powered by the strain it produces. This strain offending strain loop or process of persistent offending may mirror the process un derlying the observed elevated risk of revictimization experienced by victimized children and adolescents, including child sex trafficking victims. However debatable the association of victimization with population heterogeneity may be (Lauritsen & Laub, 2 007), few would object to the idea that due to the potentiating effects of prior victimizations, a state of heightened vulnerability to further victimization may result. As previously noted, numerous theorists and researchers have proffered the notion th at juvenile repeat victimization may be the result of a condition or state, created by the effects of prior victimization or child maltreatment (Finkelhor et al., 2005; Perry et al., 2001). When describing the revictimization of children or adolescents, d evelopmentally corrosive cumulative disadvantage based in childhood adversity and abuse has been empirically observed (Ireland et al., 2002; Macmillan, 2001; Putnam, 2006; Scaer, 2001; Solomon & Heide, 1999). The limited amount of criminological research that has tested the influence of population heterogeneity and state dependence in explaining the association between juvenile offending and victimization has found that both processes affect the association (for reviews, see Baron, 2003, Chen, 2008; Lauri tsen & Laub, 2007; see also Katz, 2000; Macmillan, 2001; Tyler & Johnson, 2006a, 2006b). These results are similar to the findings regarding the effect of state dependence and population heterogeneity on

PAGE 65

56 persistence in offending, which also suggest that bo th processes are influential (for review, see Nagin & Paternoster, 2000; see also, Ezell & Cohen, 2005; Mason & Windle, 2002; Paternoster et al., 1997; Piquero et al., 2003). However, even less is known about the processes affecting repeat victimization ( Fisher et al., 2009; Finkelhor, 2007; Lauritsen & Laub, 2007; Ousey et al., 2008). Findings by Ousey et al. (2008) from a study on repeat victimization during adolescence suggest that victimization alters social bonds and peer associations thereby increas ing risk of further victimization. Yet these underdeveloped (Ousey et al., 2008). Similarly, after rev iewing research on victimization specific factors underlying state Based on the appeal for further investigation into state depen dent processes underlying victimization (Lauritsen & Laub, 2007; Ousey et al., 2008) and the prevailing amount of research validating the common experience of revictimization by minors (Classen et al., 2005; Finkelhor et al., 2005; Menard & Huizinga, 2001) two criminological theories were examined by this study, endeavoring to understand vulnerability to victimization in child sex trafficking. Both theories, gendered pathways de pendence paradigm for explaining continuation of behaviors or specified events (Nagin & Paternoster, 1991, 2000) and therefore they are appraised in this study for relevance and application to the escalation in victimization often observed in the histories of child sex trafficking victims (Albanese, 2007; Estes & Weiner, 2001, 2005). While these two criminological theories have been developed primarily to explain

PAGE 66

57 delinquency and offending, their key theoretical propositions that are reviewed in the next ch apter allow for purposeful adaptation and application to victims and victimization. As a life course feminist theory applied to criminal offending, gendered pathways theory supports the perspective of feminist theory regarding the detrimental devaluing of women and girls due to patriarchal beliefs and practices (Chesney Lind & Paska, 2004; Miller & Mullins, 2006, 2009; Steffensmeier & Allan, 1996), commonly theorized to be a generative force escalating vulnerability in women and girls to sex trafficking (C lawson, 2009; Farr, 2005; Jeffreys, 2009, 2010; Morash, 2006). Particularly attuned to the disproportionate rate of sexual abuse experienced by girls, gendered pathways theory er 38). Focusing on how gendered social scripts interact with child maltreatment, gendered pathways theory predicts a pathway leading from childhood victimization to s uccessive and escalating victimizations (Giordano, Deines, & Cernkovich, 2006; Miller & Mullins, 2006, 2009). General strain theory describes underlying state dependence dynamics that may foster repeat victimization (Agnew, 2004, 2007). Strain and offend ing have been found to sustain one another, with offending becoming progressively more likely due to the strain it generates (Hagan & Foster, 2003; Kim et al., 2003). Perhaps, this cyclical strain offending process found to result in persistent offending is indicative of the process underlying the observed elevated risk of revictimization experienced by children and adolescents, including child sex trafficking victims. With specific application to the topic of childhood victimization, general strain theo ry offers both a sound explanation of

PAGE 67

58 the probable origins of a familial context of characterized by childhood maltreatment and explain escalating victimizations that terminate in victimization in prostitution (Agnew, 2009; Broidy & Agnew, 1997; Miller & Mullins, 2009). The next chapter includes a discussion of these two theories and their application to child sex trafficking in prostitution in the United States.

PAGE 68

59 Ch apter 4 Theoretical Framework Two criminological theories, gendered pathways theory (GPT) and general strain theory (GST), are reviewed in this chapter, and their application to the problem of victimization by child sex trafficking in prostitution is consi dered. GPT is a feminist theory centered on understanding the common contexts and causes of female delinquency (Miller & Mullins, 2009) and may be useful in explaining pathways into exploitation in child sex trafficking. Feminist theory has frequently of fered explanations for the high rate of entrapment of women and girls in sex trafficking based on the devaluation of females and the global industrialization of prostitution (Clawson, 2009; Farr, 2005; Jeffreys, 2009, 2010; Morash, 2006). Focusing on detr imental sequences arising from familial abuse, GPT illuminates a pathway of stepping stones leading from initial childhood victimization to successive and escalating victimizations, terminating with entrapment in prostitution (Giordano et al., 2006; Miller & Mullins, 2006, 2009). Similarly aligned within the state dependence perspective, GST offers both a sound explanation of the probable origins of childhood strain and a useful construal of the potent evolution of individual strain in caregivers and thei r children that may facilitate a caustic continuum of victimizations, and ultimately lead to entrapment of maltreated minors in sex trafficking in prostitution. Following these theoretical reviews, a plausible

PAGE 69

60 model of the processes contributing to elev ated vulnerability for victimization in child sex trafficking in prostitution is offered. Gendered Pathways Theory Projected to be beneficial in explaining life contexts and sequences that elevate the likelihood of victimization in child sex trafficking, GPT is a life course theory highlighting contexts, events, developmental sequences, and choices over the life course that result in female delinquency and criminality (Chesney Lind & Paska, 2004; Daly, 1992, 1998; Heimer & Kruttschnitt, 2006; Miller, 2008 a; Miller & Mullins, 2006, 2009; Morash, 2006; Steffensmeier & Allan, 1996). According to GPT, female pathways into delinquency and criminal activities may be the result of social and psychological pressures that are thought to be distinct from typical ma le pathways into crime (Alarid & Cromwell, 2006; Belknap & Holsinger, 2006; Heimer & Kruttschnitt, 2006; Miller, 2008a; Miller & Mullins, 2009; Steffensmeier & Allan, 1996). Mainstream criminological theories are not considered inapplicable to females, ye t it is not assumed that criminogenic factors are influencing males and females in the same way or with the same intensity (Giordano et al., 2006; Heimer & Kruttschnitt, 2006; Miller & Mullins, 2006, 2009; Steffensmeier & Allan, 1996). The crucial pressure s that are considered key to female understanding of the effect of gender within the theoretic al explanations of offending is well supported, as differences in the rates, explanations, sequencing, seriousness, and types of male and female offending and offenders have been well documented within criminological research (Alarid & Cromwell, 2006; Ches ney Lind & Irwin, 2008; Heimer &

PAGE 70

61 De Coster, 1999; Heimer & Kruttschnitt, 2006; Mason & Windle, 2002; Mnard, Anderson, & Godboldt, 2009; Miller & Mullins, 2006, 2009; Steffensmeier & Allan, 1996; Steffensmeier & Schwartz, 2009). Gender and Life Course Th eory Gender has been shown to affect key components of life course theory: social embeddedness, intergenerational links, and human agency (for review, see Miller & Mullins, 2009). The review of research that follows regarding gender and offending across t he lifespan supports the notion that gender matters and that gendered explanations of juvenile victimization may be both relevant and necessary for Within the sphere of social embeddedness (i .e., social ties to others), gender has been shown to profoundly affect offending. Close relationships with those involved in crime and intimate partner violence are important gendered criminogenic influences that markedly affect female offending (Brown & Bloom, 2009; Miller & Mullins, 2009; Morash, 2006; Steffensmeier & Allan, 1996). Research provides support for the supposition that social bonds influence female offending patterns more than male patterns (for reviews, see Bender, in press; Hartman, Turn er, Daigle, Exum, & Cullen, 2009; see also, Agnew, 2005b; Benda, Harm, & Toombs, 2005; Daigle, Cullen, & Wright, 2007; Giordano et al., 2006; Haynie, Steffensmeier, & Bell, 2007; Payne, Gottfredson, & Kruttschnitt, 2005). For example, although involvement in romantic relationships with others who engage in crime has been noted as increasing offending for males and females, the influence seems to be greater on females (Benda et al., 2005; Giordano et al., 2006; Haynie et al., 2007; Simons, Stewart, Gordon, Conger, & Elder, 2002). Considering social bonds as a

PAGE 71

62 protective factor, studies using school involvement or religious connections as measures of informal social control have noted that these types of social control provide greater protection for girls th an boys (Agnew, 2005b; Daigle et al., 2009; Payne et al., 2005). Furthermore, GPT research has highlighted the impact of intergenerational links on offending, exploring consequences of childhood victimization, family adversity, and difficulty providing fo r dependents (Alarid & Cromwell, 2006; Gaarder & Belknap, 2002; Haynie et al., 2007; Katz, 2000; Salisbury & Van Voorhis, 2009). Girls detained by the juvenile justice system recorded higher scores than boys on risk domains involving traumatic childhood e xperiences and family dysfunction (Cauffman, Lexcen, Goldweber, Shulman, & Grisso, 2007; Gavazzi, Yarcheck, & Chesney Lind, 2006). Experiencing physical and sexual abuse has been noted as a gendered risk factor influencing repeat offending (Benda, 2005). In the case of teenage pregnancies, due to the basic biological and reproductive differences between genders, girls continue to carry the primary responsibility in child rearing and this has been shown to affect types of offending across the life course ( Heimer & Kruttschnitt, 2006; Heimer, Wittrock, & Unal, 2006; Miller & Mullins, 2006; Simpson & Gibbs, 2006; Steffensmeier & Allan, 1996). Conversely, researchers considering protective factors have observed that family attachment, strong bonds to family, and parental support are greater protective factors for girls than boys (for reviews, see Kroneman, Loeber, Hipwell, & Koot, 2009; Hartman et al., 2009). Beyond gendered effects on offending due to social ties, life course theory describes the role of hu are located and develop within particular communities and historical periods. Within the

PAGE 72

63 given constr aints of the historical, geographical, and social contexts, individuals make choices (Maher, 1997). This coupling of constrained social regulation and human agency is a distinctive aspect of life course theory (Elder & Giele, 2009). Institutional, societ al, and cultural processes are known to result in gender inequality, constricting female choice and resulting in differences in offending across gender (Adler, 1975; Gaardner & Belknap, 2002; Heimer & De Coster, 1999; Lynch, 1996; Katz, 2000; Messerschmidt 1993; Maher, Dunlap, Johnson, & Hamid, 2006; Miller, 2008b; Miller & Mullins, 2006; White, 2009). For example, the disparity in the choices available to men and women and the resulting economic inequality generate long term effects, as women continue to be economically marginalized, resulting in the feminization of poverty (Brown & Bloom, 2009; Gaarder & Belknap, 2002; Heimer et al., 2006; Lynch, 1996; Morash, 2006; Steffensmeier & Allan, 1996; Widom, 2000). For example, the combining of mothering, econ omic marginality, and domestic violence has led some women to choose crime to manage economic pressure and avoid homelessness (Ferraro & Moe, 2006). Based on this review of research, the primary premise of GPT that gender and gender relations are fundame ntal to understanding patterns of offending behavior has found empirical support (Miller & Mullins, 2009). Social ties, intergenerational and familial links, and constrained choice contribute to female patterns of offending across the lifespan. However, a single dimensional focus highlighting only gender and gender inequality is also insufficient when considering informal and institutionalized discrimination. GPT argues that a multifaceted approach to understanding inequality is required, which necessita tes the recognition of the compounding effects of gender, race, and class (Bryant Davis, Chung, & Tillman, 2009; Maher, 1997; Miller, 2008b; Miller & Mullins, 2009; Morash, 2006; Simpson, Yahner, & Dugan, 2008; Slocum et al., 2005).

PAGE 73

64 Minority women and gir ls experience more constraints on their exercise of agency. Women and girls are further marginalized within the illegal economies operating in disadvantaged communities, affecting the types of crimes committed (James, Johnson, Raghavan, & Woolis; 2004; Ma her, 1997; Miller & Mullins, 2009; Morash, 2006; Steffensmeier & Allan, 1996). Although age is not often discussed as a sphere of disadvantage that intersects with the other three domains of race, class, and gender, there are many reasons for its inclus ion in this equation of discrimination. Much like the historical subjugation and oppression of women, only in the last two centuries have children gained recognition as persons, previously considered as property, unprotected, and to be treated as their pa rents saw fit ( deMause, 1998; Shelman & Lazoritz, 2005) Children physically, psychologically, and legally are dependent on adults for their survival (Agnew, 1985; Finkelhor, 1984, 2007; Kempe, 1985). Therefore, the agency of children of both genders is uniquely constrained by their status of dependency, although girls have been found to encounter higher levels of disadvantage and more severe choice constraints than boys (Chesney Lind, 1989; Hanna, 2002; Miller & Mullins, 2006, 2009; Steffensmeier & Alla n, 1996). This magnification of inequality based on the intersection of gender and age points toward the need for gendered theories of juvenile victimization (Finkelhor, 2007; Miller & Mullins, 2009). Although a fully developed theory of female juvenile victimization has not been framed, the victimization of girls and the resulting consequences have been examined by feminist scholars and researchers attempting to uncover patterns in the life experiences of female offenders (Miller & Mullins, 2009).

PAGE 74

65 Typ ology of Female Pathways into Offending GPT has been purposefully substantiated through qualitative research, detailing the disadvantages and societal conditions encountered by women, primarily documented in the words and life stories of female offenders ( for reviews, see Morash, 2006; Salisbury & Van Voorhis, 2009; Simpson et al., 2008; see also, Miller, 2008a, 2008b). In the 1990s feminist researchers initially identified a typology of life course trajectories or pathways by charting the life experiences of female offenders into patterns of offending and desistence (Daly, 1992; Miller & Mullins, 2009). Differing pathways into specified types of offending were theorized to result from particular life histories, which are composed of distinctive risk facto rs and developmental sequences marked by certain types of troubled transitions (Miller & Mullins, 2009). One of these identified pathways closely corresponds to the findings from numerous case studies reporting the life histories of child sex trafficking victims (Albanese, 2007; Clawson, 2009; Estes & Wiener, 2005). Similar to several developmental theories of offending, GPT does not hold to an assumption of a general cause of crime; rather, GPT postulates that a typology of pathways toward offending exi st for different types of offenders (Moffitt, 1993; Patterson, 1993; Morash, 2006; Steffensmeier & Allan, 1996). Initial GPT research conducted by research resulted in a five class typology of female offenders: (1) streetwomen, (2) harmed and hurting women, (3) battered women, (4) drug connected women, and (5) streetwomen possesses particula r relevance to prostituted minors, as this pathway was

PAGE 75

66 charted by examining the offending histories of women who fled abusive homes and become involved in criminal activities (e.g., prostitution, drug dealing). In addition, GPT research suggests that pat hways into crime are age graded, with different ages of onset of offending predicting different pathways into crime (Simpson et al., 2008). For instance, a pathway into prostitution is predicted for younger girls who take to the street to escape childhood abuse and a pathway marred by intimate partner violence is more likely for women who begin offending as adults (Simpson et al., 2008). Recent studies have employed quantitative methods to determine if the typology of female pathways into offending, largel y supported by findings of qualitative research methods, can be replicated and validated by advanced statistical modeling or analyses (Gavazzi et al., 2006; Johansson & Kempf Leonard, 2009; Salisbury & Van Voorhis, 2009). For example, Salisbury and Van Vo orhis (2009) employed path analysis, those originally identified by Daly (1992): (1) a childhood victimization pathway, (2) a relational pathway, and (3) a social and hu man capital pathway. The childhood victimization pathway that was identified using path analysis closely resembles the pathway of streetwomen from the original typology documented by Daly (Salisbury & Van Voorhis, 2009) and is comparable to the research f indings regarding the risk markers and life histories that were commonly reported by girls who were victimized in sex trafficking (Clawson, 2009; Estes & Weiner, 2005). Namely, a female pathway to incarceration was analytically identified beginning with childhood victimization that contributed to historical and current forms of mental illness and substance abuse (Salisbury & Van Voorhis, 2009, p. 541).

PAGE 76

67 Concept of Blurred Boundaries Perhaps the most important concept that GPT has contributed to feminist c riminology is the notion of the blurred boundaries that transect victimization and offending (Alarid & Cromwell, 2006; Chesney Lind & Pasko, 2004; Daly, 2006; Ferraro, 2006; Miller & Mullins, 2006, 2009; Moe, 2004; Steffensmeier & Allan, 1996). This blurr ing of the boundaries between victimization and offending was the original fundament of all the predicted gendered pathways into offending (Miller & Mullins, 2009). GPT researchers contend that the black and white, sharply distinct labels of victim or off ender serve as inadequate descriptors of women or girls, who due to a context of repetitive abuse, respond by offending (Ferraro, 2006). Offenses committed in the midst of such mitigating factors may be an attempt to escape the situation alive or to preve nt further repeated episodes of victimization (Moe, 2004). Criminal prosecutions in such situations are considered criminalization of self protection (Daly, 2006; Miller & Mullins, 2009). Sexual exploitation in prostitution is an often noted example of t his blurring of boundaries. Numerous researchers have illuminated the commonly followed pathway from an abusive childhood environment into entrapment in prostitution as a minor (Estes & Weiner, 2001, 2005; Farley et al., 2005; Gilfus, 1992; Hanna, 2002; S ilbert & Pines, 1982; Simons & Whitbeck, 1991; Widom & Kuhns, 1996; Williams & Frederick, 2009; Wilson & Widom, 2010). As prior victims of child sexual abuse, such girls are less likely to know how to protect themselves from sexual exploitation (Finkelhor & Browne, 1985; Meiners Levy, 2006; Rutter, 1989) and are at high risk for developing into easy prey for a predatory child sex trafficker. Chesney Lind (1989, p. 24) notes that runaway girls

PAGE 77

68 ntly they are often sexually exploited. Subsequently, these minors are often considered culpable for their own sexual exploitation in prostitution and punished for the crime committed against them (Adelson, 2008; Albanese, 2007; Halter, in press; Mitchell et al., 2009). Child Victimization Pathway into Exploitation in Prostitution Adolescent females are considered uniquely vulnerable to sexual exploitation due to socialization into gendered scripts, with males being expected to take the sexual initiative a nd young females expected to acquiesce (Chesney Lind, 1989; Hanna, 2002; Miller & Mullins, 2006, 2009; Steffensmeier & Allan, 1996). During adolescence, patterns of male domination and female subordination are amplified. This intensification of gender in equality during adolescence is predicted to result in higher levels of mental health issues for abused and disparaged girls, such as depression, low self esteem, suicidal ideation, and feelings of personal worthlessness that girls hope to change through ga ining male approval (Belknap & Holsinger, 2006; Chesney Lind & Shelden, 2004; Giordano et al., 2006). These entrenched societal roles or gendered scripts, coupled with a lack of sided sexu with sex traffickers, who commonly masquerade as boyfriends or rescuers (Hanna, 2002, p. 12). In summary, Giordano et al. (2006) outlined a prototypical sequence of life adversities that typify the pathway followed by many girls into prostit ution: 1) sexually abused by a stepfather or other male relative, 2) a girl runs away from home to escape the abuse, 3) she engages in illegal activity to survive life on the streets, 4) thereby coming in contact with exploitive males who foster or demand further illegal activity, such

PAGE 78

69 as prostitution, and 5) she consequently self medicates by using drugs or alcohol to cope with the abuse, exploitation, and other degrading experiences. Pathways Typology as a Sensitizing Concept Within a typology of female offenders, GPT charted a life course schematic of probable sequences that may occur within discriminatory and abusive contexts to mark out a pathway for girls into prostitution. Primarily using qualitative methods of research, GPT provided precursory emp irical evidence of associations between certain life events, survival strategies, and future outcomes, i.e., sexually abused girls run away only to be exploited in prostitution. In the social sciences, researchers have often used schemas or templates as st arting points to orient researchers toward a reasonable line of inquiry and to provide a structure for organizing and understanding observations or data (Bowen, 2006; King, 1998; King, Carroll, Newton, & Dornan, 2002; Patton, 2002). Rarely, if ever, do re commonly used to steer inductive qualitative research and also to guide so me exploratory quantitative research, alerting researchers to important aspects of the phenomenon under study (Bowen, 2006; Patton, 2002). Beyond providing singular concepts that bring focus to research, sensitizing concepts may also produce innovative c lassification systems made up of categories that

PAGE 79

70 classify a phenomenon into distinct parts (Patton, 2002). The first purpose of such sensitizing classification systems or typologies is descriptive, distinguishing aspects of a phenomenon based on the patte rns that appear in the data (Patton, 2002). Using this series of patterns distilled into contrasting themes that create alternative ideal (Patton, 2002 p. 459 ). One of the ideal types within the GPT typology of female offenders partially illuminated a series of steps or a pathway that adolescent girls may follow, thereby elevating their vulnerability for entrapment in prostitution (Giordano et al., 2006). Yet, many questions were left unanswered concerning how the pattern of events within this prototype were triggered, maintained, and merged together to escalate victimization (Lauritsen & Laub, 2007; Miller & Mullins, 2009). One danger of sensitizing typologies is that the typology may begin to better 460). And while sensitizing concepts may assist the researcher to simplify complex problems by bringing certain dimensions of the problem into focus, concurrently other dimensions may be ignored and fade into the background. GPT has primarily focused on providing an explanation for the disproportionately high percent of child sex trafficking victims who were girls in comparison to the percent of victims who were boys due to patriarchal societal values that lead to disparaging social scripts, the devaluation and victimization of girls and women, and economic inequality between genders (C hesney Lind, 1989; Jeffreys, 2009; Miller & Mul lins, 2006, 2009; Steffensmeier & Allan, 1996 ). However, little empirical evidence or theoretical explanations have been offered by GPT to explain why certain girls who possess known risk factors for child sex trafficking (i.e., history of sexual victimiza tion) and face multifaceted inequality based on the intersection

PAGE 80

71 of gender, age, ethnicity, and social class (i.e., minority girls from low income families) are entrapped in sex trafficking while other girls with similar backgrounds do not become victims o childhood is not an infallible predictor of future victimization, as not every victimized girl suffers further victimization (Classen et al., 2005). Neither does every girl of minority status from a lower socio economic neighborhood become a victim of child sex trafficking in prostitution (Siegel & Williams, 2001a). Encouraging the continual advancement of research, Patton (2002) noted that once qualitative case studies have been used to develop descriptive typologies, it is then drive repeated and escalating victimization for certain girls is needed in order to demystify the processes by which many become entrapped in a perpetuating cycle of victimization. So while GPT described a linkage between childhood victimization and future sexual exploitation, gender ed pathways researchers have rarely investigated the mechanisms producing the elevated risk for repeat victimization in certain girls. Simply stated, GPT substantiated a well worn path followed by girls into prostitution. Yet, GPT fell short of explicati ng how or where the path begins or why certain girls continue along the path terminating in child sex trafficking in prostitution, while others do not. As an association between certain life events and victimization experience s have been observed and described in the personal histories of girls, research efforts should follow in decomposing the empirically documented relationship into generative risk

PAGE 81

72 factors and seek to identify the underlying processes influencing the outcome of escalating victimization and sexual exploitation (Farrington, 2000; Patton, 2002; Shadish, Cook, & Campbell, 2002). Projected to be useful in revealing the underlying and dynamic state dependence mechanisms that elevate the likelihood of victimization psychological theory that focuses primarily on the pressures that push a person toward crime. Life strains, by causing, influencing, and interacting with negative emotions, ag gressive personality traits, and criminogenic social learning, are predicted to result in dysfunctional coping, such as delinquent behaviors (Agnew, 1992, 2001, 2002, 2006a). Agnew (2001) theorized such behavioral responses to strain are chosen based on t heir perceived functionality, i.e., their assumed likelihood of terminating, reducing, or escaping stressful conditions. GST defines strain as events or conditions disliked by individuals (Agnew 1992, 2006a). These strains can involve the blocking of a ccess to a valued goal, the actual or threatened loss of something valued, or the actual or threatened presentation of feared or disliked stimuli (Agnew, 1992, 2006a). Strain theory also emphasizes the role of negative emotions, which may interact with ne gative events to promote delinquency (Broidy & Agnew, 1997). In further development of the GST, Agnew (2001, 2006a) postulated that certain types of strain, specifically those of greater magnitude, those seen as unjust, those that weaken social bonds, and those that foster criminogenic supportive beliefs critically increase the likelihood of delinquency. Numerous studies have supported the propositions of GST, finding associations between strain and

PAGE 82

73 delinquency (for reviews, see Agnew, 2006b; Ellis & Sav age, 2009; Froggio, 2007; Ireland, Rivera, & Hoffman, 2009; Slocum et al., 2005). General Strain Theory and Victimization The primary purpose of the current research is the analysis of the effect of strain on the likelihood of minors becoming entrapped in prostitution, which has been defined as a type of criminal victimization. 6 An emphasis on the damaging effects of strain resulting in consequential outcomes other than delinquency is not outside of the sphere of GST. Many forms of childhood victimizat ion have been defined as types of strain (e.g., parental rejection, harsh or erratic discipline, child abuse and neglect) (Agnew, 2009). Strains or victimizations frequently beget further strain, thus implying that strain can be both a precursor and conseq uence of victimization (Agnew, 2006a). Previous researchers have examined the link between childhood and familial strains and the risk for criminal victimization (Arata, 2000, 2002; Barnes et al., 2009; Classen et al., 2005; Esbensen, Huizinga, Menard, 1 999; Reid & Sullivan, 2009a; Savolainen, Sipila, Martikainen, & Anderson, 2009; Schreck & Fisher, 2004; Tyler & Johnson, 2006b; Widom & Wilson, 2009). For example, Tyler and Johnson (2006b) conducted a longitudinal study with a sample of youth who had exp erienced child maltreatment and whose caregivers had been investigated by child protective services. The researchers identified a path from child maltreatment to physical assault by 6 As noted in Chapter 3, this criminal activity or phenomenon was previously considered to be a form of juvenile delinqu considered sufficient and even purposefully design to encompass and explain such occurrences.

PAGE 83

74 someone other than a caregiver, mediated by running away, early sexual in volvement, and delinquency (Tyler & Johnson, 2006b). Agnew (2006a) outlined such a pathway from childhood strain to victimization in his discussion regarding the effects of early family environment, highlighting that emotionally distressed individuals, c proposition that maltreated youth are inclined toward involvement in harmful environments could be understood as the unfortu nate byproduct of the immaturity of youth interacting with intolerable levels of abuse, resulting in the selection of risky or hazardous escape routes (see also, Casey et al., 2010; Romeo, 2010; Romer, 2010; Steinberg, 2010). Furthermore, the dependency o selection and personal agency in the situation (Finkelhor, 1984, 2007), with the choices generally available to girls even further constrained (Chesney Lind, 1989; Hanna, 2002; Jeffreys, 2010; Miller & Mullins, 2006; St effensmeier & Allan, 1996). Consequences of Caregiver and Family Strain Considering the origins of child maltreatment that may predispose a child to experiencing a progression of repeated victimizations, GST postulates that strain is the manifest mechani sm triggering certain caregivers to fail to provide needed protection or to abuse their children (Agnew, 1992; Agnew, Rebellon, & Thaxton, 2000) thereby implying an transgenerational transmission of strain from caregiver to child. Caregiver adversities an d strains have often been identified as key risk markers, increasing the likelihood of child abuse and neglect (for review, see Agnew et al., 2000). A range of strains applicable to caregivers have been found to result in poor parenting and the

PAGE 84

75 perpetrati on of child maltreatment, including: marital conflict, spousal abuse, maternal depression, lack of social support, parental criminality, unsupportive family structure, and economic hardship (Agnew et al., 2000). Agnew et al. (2000) further theorized tha t caregiver strain produces negative emotions in the caregiver, which are commonly expressed by their withdrawal from or neglect of nurturing responsibilities. As a result of such withdrawal, support and care for the child decays as the parent neglects th e child, failing to monitor or emotionally engage with the child (Agnew et al., 2000). Child neglect and failure to nurture or monitor a child may lead to negative behaviors in the child. Such problem behaviors in children are also noted as a source of caregiver strain, resulting in the harsh treatment of children (Agnew et al., 2000); but the directionality of the relationship is unclear (for review, see Hoeve et al., 2009; see also, Gershoff, 2002; Hipwell et al., 2008; Jaffee, Caspi, Moffitt, & Taylor 2004; Knutson, DeGarmo, Koeppl, & Reid, 2005; Patterson, 1993). Supporting such findings regarding the detrimental effects of strain on caregivers, numerous behavioral scientists have observed that caregivers are often abusive or neglectful due to physic al or mental illness, drug or alcohol abuse, incarceration, psychiatric disorders, social isolation, domestic violence, drug abuse, and divorce or separation of spouses (Barnett, Miller Perrin & Perrin, 2005; DiLauro, 2004; Ford, Chapman, Mack & Pearson, 2 006; Herrenkohl et al., 2008; Turner, Finkelhor & Ormrod, 2007). For example, in a study of the relationship of parental psychiatric disorders and the occurrence of childhood physical and sexual abuse, those with a parent diagnosed with depression, mania, or schizophrenia had a two to threefold increase in rates of

PAGE 85

76 childhood abuse, and those with a parent diagnosed with antisocial personality disorder had a six to sevenfold increase in rates of childhood abuse (Walsh, MacMillan, & Jamieson, 2002). Addi tionally, the family context in which child sexual abuse typically occurs has been characterized as less cohesive, more socially isolated, and more dysfunctional than families of children who are not abused ( Fassler, Amodeo, Griffin, Clay, & Ellis, 2005; M adonna, Van Scoyk, & Jones, 1991). As was previously noted, Agnew (2006a) held that it is common for various strains to cluster together (see also, Slocum et al., 2005). The clustering of strains is more likely to elicit negative emotional states, which produces an inability to cope with strain (Agnew, 1992, 2006a). The clustering of adversity may overwhelm the family unit, leaving the children exposed and unguarded. Effective parenting demands energy, time, and emotional stability; such resources are often depleted due to an accumulation of stressors (Agnew et al., 2000). The detrimental effects of caregiver strain on child victimization was noted in recent research examining juvenile victims, with reported findings showing that members of the most se verely victimized class of minors were more likely being located within families characterized by very high levels of adversity (e.g., caregiver imprisonment, employment difficulties, problematic drug or alcohol use, conflict between caregivers, homelessne ss) (Reid & Sullivan, 2009a). The class of children more likely to be located in such families had experienced relatively high levels of maltreatment within the home, elevated levels of victimization by peers, as well as other forms of criminal victimizat ion. Estes and Weiner (2005) identified many of these highly problematical caregiver strains as empirically observed risk factors contributing to sexual exploitation. Caregiver risk factors commonly observed in child sex trafficking

PAGE 86

77 victims were family dysfunction, domestic violence, and parental instability in the form of parental drug dependency and severe mental health problems (Estes & Weiner, 2005). According to GST, caregiver strain does not directly result in a child participating in criminal ac tivity, such as illegal drug use or sexual exploitation; rather, the effects of caregiver strain are mediated by the harmful events of child maltreatment (Agnew et al., 2000). Loeber, Hipwell, Battista, Sembower, and Stouthamer Loeber (2009) identified a mediation model from maternal problems or strain (i.e., mental health and conduct problems) to daughter strain, partially mediated by harsh punishment, inconsistent discipline, and low warmth, supporting the conceptualization of the transgenerational tran sference of strain transmitted by means of child maltreatment. The Strain of Child Maltreatment an apt definition of child maltreatment. Poor parenting, in the form of neglect and abuse, is considered a major source of juvenile strain leading to delinquency (Agnew et al., 2000). As previously mentioned, the various forms of child maltreatment hav e been found to result in a broad range of impairments, including involvement in delinquency (for review, see Arata et al., 2007; see also, Carson, Sullivan, Cochran, & Lersch, 2009; Colman, Kim, Mitchell Herzfeld, & Shady, 2009; Ford et al., 2006; Goodkin d, Ng, & Sarri, 2006; Hay, 2003; Hay & Evans, 2006; Heide & Solomon, 2006; Hildyard & Wolfe, 2002; Hollist et al., 2009; Lansford et al., 2007; Loeber et al., 2009; Mersky & Reynolds, 2007; Moran, Vuchinich, & Hall, 2004; Sullivan, Farrell, & Kliewer, 2006 ).

PAGE 87

78 Maltreated children may endure a series of abusive relationships as they mature (Agnew, 2006a) due to the formation in childhood of a dysfunctional template of relationship functioning. Bowlby (1973, 1980) asserted that primary relationships in chi ldhood with caregivers form an enduring template of the relationship process. Utilizing meta analysis of numerous studies, Fraley (2002) investigated the stability of this initial template across various types of relationships, including familial, romanti c, and partner relationships. Fraley found that a moderately stable relational pattern extends until approximately age 20. For African American children in lower income families, research suggests that the mother/child relationship is the primary determi nant of this enduring relational pattern ( Liang, Williams, & Siegel, 2006). Such findings indicate that being maltreated as a child may initiate a sequence of abusive relationships that persists into adulthood. Neglect. Straus and Kantor (2005) note tha t there are various dimensions of child neglect including emotional neglect, physical neglect, and neglect in supervision. Although neglect is often perceived as less damaging than abuse, researchers have found that being neglected as a child is often more damaging than being abused (Bloom, 2000; Hildyard & Wolfe, 2002). Those neglected as children suffer more severe cognitive deficits and language delays than physically or sexually abused children (Bloom, 2000; Heide & Solomon, 2006; Hildyard & Wolfe, 200 2; Sullivan, 2000). Neglected children have been found to suffer from greater levels of social withdrawal and have more limited peer interactions relative to physically abused children (Hildyard & Wolfe, 2002). Similarly to abused children, victims of ch ild neglect learn that they cannot count on others to respond to their needs and have a diminished sense of self worth (Bloom, 2000; Hildyard & Wolfe, 2002; Solomon & Heide, 2005). Possessing low

PAGE 88

79 self worth, they often perceive that they are unlovable, de serving of abuse, and feel incapable of preventing mistreatment from occurring (Bloom, 2000; Hildyard & Wolfe, 2002). Lack of a nurturing caregiver/child relationship has been shown to significantly affect the likelihood of children experiencing other ty pes of maltreatment beyond neglect (Finkelhor, 1994; Hildyard & Wolfe, 2002; Rikhye et al., 2008). Rikhye found an eightfold increase in the rates of child abuse among those who reported emotional neglect by their mothers. When families fail to provide n eeded protection, affection, and guidance, children are left to seek such resources from others outside of their family. Research has shown that such neglect opens up a needy child to abuse by others (Benedict & Zautra, 1993; Hildyard & Wolfe, 2002). Res earch focused on victim selection by sex offenders concluded that factor associated with choice of specific victims . a child with family problems, without supervision, always on the street and in need of h Beauregard et al., 2007, p. 455). Such findings suggest that child neglect may facilitate the occurrence of manifold victimizations, with neglect opening a gateway to further types of abuse. Physical abuse. Child physical abuse is generally defined as "any nonaccidental physical injury to the child" and can include striking, kicking, burning, or biting the child, or any action that results in a physical impairment of the child ( Child Welfare Information Gateway, 2007). Almost all physical abuse of children is perpetrated by caregivers (Bottoms, Nielsen, Murray, & Filipas, 2003). Children are smaller, weaker, and vulnerable to harm due to the physical strength of a much larger adult (Barnett et al., 2005). Very young children, especially those und er the age of four, are particularly

PAGE 89

80 vulnerability to severe injury or death due to physical abuse (HHS, 2010). Common indicators of child physical abuse are broken bones, bruises, lacerations, burns, head trauma, and radiological evidence of repetitive f ractures (Canter, Rosenfeld, Butt, & Botash, 2008; Miller Perrin & Perrin, 1999). The long term effects of physical abuse are dependent on whether the injuries are fatal, whether the injuries are enduring or permanent (i.e., brain damage from shaken baby syndrome), and whether the child is exposed to further trauma. Child physical abuse has been found to be associated with attachment problems, aggression, low self esteem, depression, anxiety, PTSD, substance abuse, poor academic performance, suicidal ide ation, and further victimization (Barnett et al., 2005; Fagan, 2005; Kolko, 2002; Miller Perrin & Perrin, 1999; see also Berenson & Andersen, 2006; Bottoms, et al., 2003). Sexual abuse. According to the American Medical Association (1992), childhood sex ual abuse consists of contact abuse ranging from fondling to rape and also noncontact abuse, such as modeling inappropriate sexual behavior, forced involvement in child pornography, or exhibitionism. Numerous researchers have concluded that child sexual a buse is a principal risk factor for various mental health disorders; the most commonly reported disorders among child sexual abuse survivors are PTSD, low self esteem, depression, suicide, alcohol and drug problems, and eating disorders (for review, see Ba rnes et al., 2009; see also, Briere & Runtz, 1993; Finkelhor, 1986; Solomon & Heide, 1999). Several studies have shown that the age of the victim shows an inverse relationship with subsequent mental health problems, i.e., the younger the victim, the more severe the damage (Berenson & Andersen, 2006; Cicchetti, Rogosch,

PAGE 90

81 Gunnar, & Toth, 2010; Downs, 1993; Kaukinen & DeMaris, 2005; Scheeringa & Zeanah, 2001). Other studies have found that sexual abuse or assault during adolescence elevates the risk of sexual revictimization in adulthood more than sexual abuse during childhood (Classen et al., 2005; Humphrey & White, 2000). Also, adolescents have been found to experience more self blame for being abused than those victimized as younger children ( Hunter, Goodw in, & Wilson, 1992). As prior victims of sexual abuse, abused girls may face greater vulnerability to sexual exploitation (Clawson, 2009; Estes & Weiner, 2005; Finkelhor & Browne, 1985; Meiners Levy, 2006; Rutter, 1989; Summit, 1983). Children brought up in an environment of child sexual abuse often have a psychological incapacity to set or recognize sexual boundary violations, either due to continual violations of boundaries during childhood or because any attempts to establish sexual boundaries during c hildhood resulted in punishment (Herman, 1992; Rutter, 1989; Summit, 1983). Child Maltreatment, Negative E motion and Repeat Victimization GST highlights the role of negative emotion, intensifying the relationship between the individual for action, lower inhibitions, and crea Piquero & Cullen, 2009, p. 41). The role of negative emotion has been examined in numerous studies, with the majority of studies confirming the influence of negative emotion on the relationship between strain and delinque ncy (Agnew, Brezina, Wright, & Cullen, 2002; Carson et al., 2009; Hagan & Foster, 2003; Hay, 2003; Hay & Evans,

PAGE 91

82 2006; Hollist et al., 2009; Jang & Johnson, 2003; Jennings, Piquero, Gover, & Perez, 2009; Manasse & Ganem; 2009; Mersky & Reynolds, 2007; Moon, Hays, & Burton, 2009; Moon, Morash, McCluskey, & Hwang, 2009; Simons, Chen, Stewart, & Brody, 2003; Tittle, Broidy, & Gertz, 2008). Studies examining the intensifying effect of negative emotion on the relationship between strain and delinquency have foc used primarily on anger and depression. However, Broidy and Agnew (1997) hypothesized women or girls typically respond to strain with self denigratory emotions, namely depression, guilt, anxiety, and shame. Gender differences in emotional and behavioral responses to childhood abuse have been substantiated, with males found more likely to respond to abuse with externalizing symptoms, such as aggression, conduct problems, and violence, and with females found more likely to respond to maltreatment with inter nalizing symptoms, such as depression, self blame, and suicidal ideation (for reviews, see Bender, in press; Grant et al., 2009; see also, Cauffman et al., 2007; Jang, 2007; Turner, Finkelhor, & Ormrod, 2006). Hay (2003) conducted one of the few studies c omparing gendered responses to strain and found that both boys and girls responded to sexual abuse with feelings of anger, but as predicted by Broidy and Agnew (1997) anger responses in girls were more likely to be accompanied by feelings of guilt and depr ession (see also, Cauffman et al., 2007; Daigle et al., 2007; Feiring et al., 2007). The role of shame and shaming in reducing delinquency has been explored and tested in criminological research using a variety of theoretical perspectives (e.g., Ahmed, Har ris, Braithwaite, & Braithwaite, 2001; Braithwaite, 1989, 2000; Grasmick & Bursik, 1990; Harris, 2003; Harris & Maruna, 2006). For example, in early research on gender

PAGE 92

83 roles and delinquency, Morris (1964, 1965) hypothesized and tested the proposition that girls are less likely to engage in delinquent behaviors than boys due to stronger internal feelings of shame produced by engaging in delinquent behaviors and from greater social disapproval of offending experienced by delinquent girls compared to delinque nt boys. In addition to studies investigating shame as an informal deterrent, derogatory self attitudes resulting from distressing and unresolved self devaluing experiences have been the central focus of numerous studies testing a general theory of devi ant behavior (for review, see Kaplan & Lin, 2000). Correspondingly, studies focused on sexual revictimization have hypothesized and documented an association between shame resulting from previous maltreatment and an elevated risk of future victimization ( for Arata, 2000, 2002; Filipas & Ullman, 2006; Messman, Ward, & Brown, 2009; Van Bruggen, Runtz, & Kadlec, 2006; Whiffen & MacIntosh, 2005). Finkelhor and Browne (1985) used the term stigmatization to describe the process by which victims of child sexual abuse may develop a negative self identity, typified by feelings of shame and worthlessness. This negative self identity based in shame is theorized to result in substa nce abuse, risky sexual behavior, delinquency, and sexual revictimization. Beyond stigmatization, another empirically supported effect of sexual abuse is traumatic sexualization or the dysfunctional linking of sex with approval and love that is also theor ized to lead to risky sexual behavior and sexual revictimization (Finkelhor & Browne, 1985; Stockdale et al., 2002). Recently, researchers have suggested that stigmatization or shame, theorized to be a unique result of sexual abuse (Finkelhor &

PAGE 93

84 Browne, 1985), may instead be a consequence of all forms of child maltreatment (Bloom, 2000; Solomon & Heide, 2005; Van Bruggen et al., 2006 ). According to GST, child maltreatment may produce feelings of shame in girls, increasing the likelihood that they will react to negative events or strain (Broidy & Agnew, 1997). Additionally, Finkelhor and Browne (1985) th eorized that shame may increase vulnerability of abused girls to sexual re victimization. In application to child sex trafficking victims, research has shown that abused girls who run away are more likely to be victimized by sexual exploitation in prostitution than those who run away but have not been previously abused ( Saewyc & Edinburgh, 2010; Saewyc, et al., 2008; Scott et al., 2003; Tyler & Johnson, 2006a) As has been extensively explored and confirmed in studies on sexual revictimization, perhaps shame is the negative emotion driving this increased vulnerability to revictimization in child sex trafficking victims (Arata, 2000, 2002; Classen, et al. 2005; Filipas & Ullman, 2006; Messman, Ward, & Brown, 2009; Stockdale et al., 2002; Van Bruggen et al., 2006; Whiffen & MacIntosh, 2005). Risk Inflating Responses to Strain Child maltreatment is considered a strain that is most likely to be perceived as unjust and high in magnitude (Agnew, 2001). Dysfunctional responses are predicted to occur when strains are perceived as unjust and when the gaps between desires and reality are high in magnitude (Agnew, 1992, 2001). The resulting disappointment and frus tration is theorized to lower inhibitions and prompt individuals toward action (Agnew et al., 2009). Behavioral responses to strain are chosen based on their perceived likelihood of reducing or ending distressing conditions (Agnew, 2001).

PAGE 94

85 As an example of a behavioral response to strain, maltreatment could lead a child or adolescent to run away in hopes of resolving intolerable strain (Agnew, 2006a). Running away is often reported as a result of maltreatment and a predictor of delinquency (Bender, in p ress; Chesney Lind, 2001). Children commonly run from neglectful and abusive circumstances (for review, see Baron, 2003). Children escape from their abusive homes with the hope and belief that life away from home, even a life on the streets, will be bett er than the life of abuse that they have experienced (Estes & Weiner, 2001; Sheridan & VanPelt, 2005). Children who are traumatized have not developed coping skills to deal with such strains (Bloom, 1999; Romer, 2010). Moreover, developmental research h as shown that during adolescence, individuals are strongly swayed by short term outcomes rather than possible long term consequences, amplifying immaturity in judgment and poor decisions (Casey et al., 2010; Cauffman & Steinberg, 1995; Romeo, 2010; Romer, 2010; Steinberg, 2010). So, the experience of victimization or abuse triggers an inherent motivation to escape further abuse. Yet, adolescents are generally devoid of the necessary knowledge, psychological maturity, or resources available to most adults in similar circumstances (Romer, 2010). This naturally occurring aversion to pain or desire to escape strain in collusion with navet of youth inflates the likelihood that adolescents may choose risky escape routes such as running away or select unreliab le rescuers such as sex traffickers masquerading as boyfriends (Agnew 2004, 2007; Hanna, 2002; Steinberg, 2010; Steinberg & Scott, 2003). Moreover, runaway girls, lacking legitimate employment opportunities or financial resources to provide for even thei r most basic needs, are highly vulnerable to exploitation (Estes & Wiener, 2005; Wilson & Widom, 2010).

PAGE 95

86 Rather than physically leaving home by running to the streets, children may use substances with psychotropic properties to create a temporary illusion of escape from the effects of victimization (Harrison, Fulkerson, & Beebe, 1997; Morash, 2006). Abused children learn to shut out pain and escape through various mechanisms (e.g., dissociation, self mutilation, substance abuse) (Summit, 1983). The paink illing or euphoric effects of drugs or alcohol provide an abused child with an instant method of escape (Harrison et al., 1997; Saah, 2005; Spear & Varlinskaya, 2010). Abusers may even provide the means of dissociation; sex offenders and traffickers have been known to drug victims to sedate, confuse, or silence them (Farley, 2003; Leth, 2005; Cooper, 2005c; Miller, 2008b). Consequently, it should not be surprising to find that those who have experienced maltreatment initiate substance use at a younger age and are at increased risk for multiple substance use (Arata et al., 2007; Carson et al., 2009; De Bellis, 2001; Harrison et al., 1997; Hollist et al., 2009; Kilpatrick et al., 2003; Lo et al., 2008; Menard, 2002; Moran et al., 2004; Sartor, Agrawal, McCut cheon, Duncan, & Lynskey, 2008). Substance use may serve as a coping mechanism, allowing maltreated children to survive repeated victimization or block the emotional pain of past victimization (Bender, in press; Morash, 2006; Spear & Varlinskaya, 2010; Su mmit, 1983). Substance abuse problems, both alcohol use and drug use, have consistently shown a strong association with delinquency, although the sequencing and directionality of the relationship remains unclear (for review, see Bender, in press; Cauffma n et al., 2007). The use of alcohol and drugs has also been linked to sexual victimization (Abbey, 2002; Messman Moore, Coates, Gaffey, & Johnson, 2008; Messman Moore et al., 2009; Miller, 2008b). Beyond impairing motor skills that increase vulnerability to victimization, substances also impair cognitive functioning and the decision making

PAGE 96

87 processes (Abbey, 2008; Davis, Stoner, Norris, George, & Masters, 2009). Drug dependency has been noted as a possible risk factor among child sex trafficking victims ( Clawson, 2009; Estes & Weiner, 2001, 2005; Stone, 2009). These two risk inflating responses to strain, running away and substance use, are predicted to result in a heightened state of vulnerability to entrapment in child sex trafficking. Traffickers often present themselves as caring and strong boyfriends who will rescue the minors from abuse (Hanna, 2002; Albanese 2007). Runaway and maltreated minors can be easily seduced by fraudulent promises of love, safety, and affection (Hanna, 2002). As prior vi ctims of abuse, runaway or drug dependent girls are less likely to be able to protect themselves from sexual exploitation and may be at high risk for becoming the prey of child sex traffickers (Arata, 2002; Herman, 1992; Messman Moore & Brown, 2006; Seligm an, 1975; Scaer, 2001; Scott et al., 2003; Simons & Whitbeck, 1991). Current Study: Identifying a Strain Reactive Pathway to Victimization in Child Sex Trafficking The majority of previous research on victimization in child sex trafficking has employed a risk factor approach when investigating the origins of the problem. As presented in Chapter 2, the common risk factors for victimization in child sex trafficking have been repetitively gathered, primarily through extensive interviewing of victims (for rev iew, see Clawson, 2009; Estes & Wiener, 2001, 2005). The frequently noted individual risk markers include female gender, lack of family support, history of abuse or sexual victimization, family dysfunction such as domestic violence, substance abuse and

PAGE 97

88 me ntal illness in family members, being a runaway or thrownaway youth, and drug dependency (Clawson, 2009; Estes & Weiner, 2005; Raphael, 2004 ). As previously discussed in Chapter 3, using a risk factor approach to explain victimization may prove ineffecti ve at illuminating origins or generative processes because it is limited to highlighting risk markers, which may or may not be determinant factors (Farrington, 2000; Wikstrm, 2008). Research on child sex trafficking has yet to provide explanations of why certain minors who have experienced known risk factors for sex trafficking or possess common attributes observed in child sex trafficking victims become entrapped, while others do not. As this study endeavors to advance knowledge regarding child sex traf ficking victimization, it is necessary to move beyond the identification of risk factors to a fuller understanding of how the previously identified risk factors originate and combine to escalate vulnerability to victimization. In order to progress past a risk factor approach, Wikstrm (2008) recommends that researchers theoretically carry out more advanced analytical work with the aim of identifying emphasis in the original). From the presented review of GPT and GST, many similarities between the two theoretical explanations for juveniles involvement in delinquency were identified that may assist in illuminating why certain minors become victimized in child sex trafficking. Both theories (1) are consistent with the state dependence paradigm for explaining persistent offending or the repetition of specified events (Agnew, 1992, 2009; Miller & Mullins, 2009; Nagin & Paternoster, 1991, 2000); (2) focus on adverse pressures tha t push an individual toward delinquency or increase vulnerability to victimization (Agnew, 1992;

PAGE 98

89 Steffensmeier & Allan, 1996); (3) assert that differences exist between males and females in both the types of strain encountered and in the responses to strai n, resulting in varied pathways into delinquency or victimization (Adler, 2009; Broidy & Agnew, 1997; Miller & Mullins, 2009; Steffensmeier & Allan, 1996); (4) consider victimization, including child maltreatment, as a major risk factor or strain promoting entanglement in criminal environments, particularly for females (Agnew, 2001, 2006a; Agnew et al., 2000; Steffensmeier & Allan, 1996); and (5) highlight responses, motivated by a desire to cope with, escape, or survive adversity, that induce further delin quency and vulnerability to victimization (Agnew, 2001; Daly, 2006; Steffensmeier & Allan, 1996). As was reviewed in this chapter, GPT pioneered a sensitizing typology (Patton, unraveling and charting the life histories of female offenders, linking the effects of past victimization with subsequent vulnerability to victimization in criminal activities (Daly, 2006; Miller & Mullins, 2009). This perilous linkage of past victimizati on with further vulnerability resulted in the GPT conceptualization of blurred boundaries between victimization and offending (Daly, 2006; Miller & Mullins, 2009). Child sex trafficking victims are often poignant examples of this blurring of the boundarie s between victim and offender, as these victims have often been previously abused or abandoned by their families (Clawson, 2009; Estes & Wiener, 2005) and subsequently considered culpable for their own sexual exploitation in prostitution (Adelson, 2008; Al banese, 2007; Halter, in press; Mitchell et al., 2009). Additionally, GPT highlighted the potency of societal devaluation of girls during adolescence and socialization into gendered scripts that may encourage girls to acquiesce to the manipulation and dem ands of sex traffickers (Belknap &

PAGE 99

90 Holsinger, 2006; Chesney Lind, 1989; Chesney Lind & Shelden, 2004; Giordano et al., 2006; Miller & Mullins, 2006, 2009; Steffensmeier & Allan, 1996). Constructively narrowing the theoretical focus onto a very specific p roblem, the child victimization pathway into prostitution within the GPT typology of female offenders was valuable to the current study as a sensitizing concept (Blumer, 1954; Patton, 2002), setting forth an initial template for further exploration of this criminal exploitation of girls. However, the origins and mechanisms operating within the child victimization pathway were not adequately explained by GPT. Furthermore, GPT did not offer a clear explication of the connection between abusive treatment and individual choices of escape, nor expl ain the role of self denigratory emotion resulting from childhood a criminal environment or the potentiating effect of negative emotion on vulnerability to revictimization. Impo rtantly, GST provided a hypothesis of the generative factors, i.e., caregiver problems and adversity, from which a familial environment characterized by child maltreatment is most likely to originate (Agnew et al., 2000). Thereby, GST offered a sound expl anation of the linkage between the previously observed risk factors in child sex trafficking victims of family dysfunction and childhood abuse. In addition, GST more fully explicated the link between individual strain and the resulting behavioral choices of escape and also explained the pivotal role of negative emotion. In light of this theoretical review, by building upon the previously documented information regarding risk factors found to increase vulnerability to child sex trafficking and drawing fro

PAGE 100

91 prostitution (Wikstrm and Sampson, 2006, p. 2). GST provided reasoned explanations of the origins and generative mechanisms that may escalate vulnerability in certain girls. This theorized process, by which girls are conditioned and made acutely vulnerable to victimization in child sex trafficking, was labeled a strain reactive pathway to victimization in child sex trafficking Although GPT supplied a sensitizing typology useful in directing initial research efforts toward a reasonable line of inquiry for understanding the topic of child sex trafficking in prostitution (Daly, 1992; Giordano et al., 2006; Salisbury & Van Voorhis, 2009), all of the key predictions of the strain reactive pathway were drawn from the more fully developed theoretical propositions of GST (see Figure 1). The pathway also incorporated and connected many of the commonly observed ri sk factors in child sex trafficking victims in the United States (Estes & Weiner, 2005). The key propositions of the current study are: (A) Caregiver strain (e.g., domestic violence, mental health problem s, substance abuse) is expected to produce a detrimenta l familial context increasing the likelihood of the occurrence of child maltreatment, including child neglect, physical abuse, and sexual abuse. (B) Next, in response to the strain of child maltreatment, it is hypothesized that victimized girls are more like ly to choose risk inflating, even delinquent, pathways of escape. Victimized girls may run away to physically escape maltreatment or begin to use drugs or alcohol to temporarily escape abuse and its effects.

PAGE 101

92 (C) Self denigratory negative emotion is theorized experiences of child maltreatment. (D) In addition, this negative emotion, resulting from previous childhood victimization experiences, is theorized to detrimentally inflate vulnerability to victimization in sex trafficking of escap ing girls, i.e., girls who run away or use drugs or alcohol to escape the pain of victimization. (E) Consequently, these risk inflating choices of escape taken by abused vulnerability to r evictimization in child sex trafficking. Figure 1 Strain Reactive Pathway to V ictimization in Child Sex Trafficking in Prostitution Caregiver Strain Risk Inflating Responses C hild Sex Trafficking Prostitution Child Maltreatment Negative Emotion A B E D C

PAGE 102

93 As noted above and displayed in Figure 1, (A) reflects the proposition by GST that caregiver adversiti es and strains increase the likelihood of child abuse and neglect (for review, see Agnew et al., 2000). Point (B) reflects the propositions offered by GST that maltreated girls may engage in risk inflating responses to maltreatment, such as running away o r using drugs or alcohol, motivated by a desire to cope with, escape, or survive adversity (Agnew, 2001). Point (C) indicates the hypothesis set forth by GST that the strain of child maltreatment results in negative emotion, specifically noting that femal es respond to str ain with self denigratory emotions (Broidy & Agnew, 1997). Point (D) reflects the proposit ion by GST that self denigratory emotions in girls resulting from childhood maltreatment have a moderating effect, increasing the effect of childhoo d strain on vulnerability to revictimization (Broidy & Agnew, 1997). Point (E) is drawn from running away, which are predicted to induce further vulnerability to victimiza tion (Agnew, 2001). The next chapter describes the method used to assess the suitability of the theorized pathway for explaining the entrapment of girls in sex trafficking in prostitution.

PAGE 103

94 Chapter 5 Method The purpose of this chapter is to present t he method used to test the suitability of the strain reactive pathway described in Chapter 4 that was designed to explore the effects of caregiver strain, child maltreatment, and risk inflating responses to strain on vulnerability to victimization in child sex trafficking in prostitution. The pathway was assessed using data drawn from a sample of 174 females of minority and lower socio economic status living in an urban environment who we re participants in a longitudinal study investigating the effects of childhood abuse (Siegel & Williams, 2001a, 2001b). The current study utilized a quantitative method of statistical analysis, structural equation modeling, in order to analytically identify and estimate the posited pathway into child sex trafficking in pro stitution. General Methodological Overview As was previously reviewed, the majority of research on child sex trafficking victims had been qualitative in nature, based on retrospective data gathered from in The proc ess of identifying prior victims of child sex trafficking in prostitution, interviewing them at length, and then organizing the interviews so that they revealed patterns in th e occurrences of victimizations effectively documented the common risk markers am ong

PAGE 104

95 victims. This prior body of research was essential in guiding the formation of the theorized pathway tested in this study. However as was recommended by numerous researchers reporting a severe deficit of methodologically reliable research on sex tra Musto, 2009; Zhang, 2009), this study deviated methodologically from the past precedent of relying on qualitative methods of data collection and analysis. From a research perspective it was considered important to offer multivariate, quantitative evidence of the existence of the theorized pathway into child sex trafficking (Salisbury & Van Voorhis, 2009; Wikstrm, 2008) By u sing quantitative methods of data analysis, it was possible to precisely specify and analytically test the proposed pathway and study hypotheses as well as statistically interpret the results of the analyses. For example, based on the use of inferential statistics it was possible to determine how probable it was that the observed associations between the variables included in the models occurred by chance alone. Ultimately, the use of quantitative methods of analysis in this study not only provided explicit validation of the proposed pathway but also offered further validation of the previously accr ued information regarding child sex trafficking victims. The method of multivariate analysis used in this study was structural equation modeling (SEM). SEM is a statistical modeling tool that is useful for theory testing or development (Asparouhov & Mut hn, 2008; Kline, 2005). SEM can be used in a variety of ways to test or develop theory: (a) in a strictly confirmatory manner with one model proposed and tested, (b) in an exploratory manner by testing alternative models with several theoretical models being tested and compared to determine the best fitting

PAGE 105

96 model, or (c) in a developing model approach with an initial model tested and subsequent modifications made to improve fit (Garson, 2009; Kline, 2005). As SEM can be employed in an exploratory manne r for theory development, numerous models based on the propositions pre sented in Chapter 4 were developed and tested. All of the models evaluated the elements of the strain reactive pathway by examining the direct and indirect effects of caregiver strain, child maltreatment, and risk inflating responses on vulnerability to child sex trafficking in prostitution (see Figure 2). SEM was projected to be particularly beneficial in testing the proposed models, as it is commonly used to identify and estimate rel ationships and mediating effects between observed and latent variables (Kline, 2005). The latent variables included in the various models were caregiver strain, child neglect, and child maltreatment. These latent variables were each identified using con firmatory factor analysis (CFA) from several observed indicators (see Measures). The remaining variables in the models, including running away as a minor, age of initial drug or alcohol use, and being prostituted as a minor, were drawn from observed indic ators. Initially, separate models were specified using SEM in order to independently examine the effect of caregiver strain on each type of child maltreatment, including child neglect, sexual abuse, and physical abuse (see Figure 2). In addition, these mo dels allowed for the identification of the direct effects of each form of childhood maltreatment on the behavioral or risk inflating responses of maltreated girls and the indirect effects of each of the forms of child maltreatment on vulnerability to victi mization in child sex trafficking in prostitution.

PAGE 106

9 7 Figure 2 Hypothesized Strain Reactive Pathway to Victimization in Child Sex Trafficking in Prostitution Including the Forms of Child Maltreatment After testing the models examining the effects of each type of child maltreatment in isolation, a model including multiple types of child maltreatment was identified and evaluated. As examining each type of child maltreatment in isolation may result in an overestimation or underestimation (i.e., suppression) of its effects, the combining of the types of child maltreatment into one model was also necessary (Kline, 2005; Wilson & Widom, 2010). In addition, the consequent development of negative emotion (measured as relational shame) due to the experience of child maltreatment and its potential moderating effect (Baron & Kenny, 1986; Holmbeck, 1997, 2002, 2006) on vulnerability to victimization in prostitution for girls who run away or initiate early drug or alcohol use was exp lored through t he specification and examination of supplementary models. Lastly, testing for the existence of moderating or interaction effects of negative emotion was further investigated using logistic regression (Jaccard, 2001). Caregiver Strain Child Neglect Risk Inflating Responses Child Sex Trafficking Prostitution Sexual Abuse Physical Abuse (Forms of Child Maltreatment) Ne gative Emotion Relational Shame

PAGE 107

98 Study Participants and Data Collecti on Procedures As mentioned in Chapter 2, due to the hidden nature of the crime of child sex trafficking and the complexities involved in gaining access to its victims, available data on this highly victimized p 2008a). However, previously collected data from a longitudinal study investigating the effects of childhood sexual abuse were available for use in testing the theorized pathway. The data were dra wn from primarily urban, low income, African American females (Siegel & Williams, 2001a, 2001b). Living in an urban environment, being of minority status, poverty, and being female have all been found to be risk markers for child victims of sex traffickin g in prostitution (Estes & Wiener, 2005; Clawson, 2009). McCahill, Meyer, and Fischman (1979) collected the original data between 1973 and 1975 as a part of a larger study on the after effects of sexual assault. All 206 girls included in the original ph ase of data collection were brought to the emergency room of one particular municipal hospital where they underwent forensic examinations and received medical treatment due to being sexually abused. 7 The original participants were later interviewed two su bsequent times, in 1990 1991 and again in 1996 1997, with the objective of gaining a greater understanding of the consequences of child sexual abuse. During the follow up interviews in 1996 1997, the original sample of abused females was matched with a s et of comparable females who received medical treatment 7 During this time period, it was mandated that all residents of this large northeastern U.S. ci ty brought to one particular municipal hospital emergency room for treatment and collection of

PAGE 108

99 at the same hospital for a reason other than sexual abuse. 8 By searching the pediatric emergency room records of the same municipal hospital, matches were made based on race, age, and date of hospita l visit (Siegel & Williams, 2001a, 2003). Data from the follow up interviews conducted in 1996 1997 were analyzed in this study. During the 1996 1997 follow up interviews, researchers contacted a tot al of 238 women. Of these women, 174 agreed to be int erviewed. 9 Interviewees were paid $35 for participating. Four female interviewers, trained in conversing on sensitive topics, conducted the face to face interviews that lasted approximately three hours ( Siegel & Williams, 2001a ). After establishing rapp ort, a series of questions about sexual history (Siegel & Williams, 2003, p. 911). To assist with memory recall, the interviewers defined the time frame of the questi ons and established memorable milestones for each & Williams, 2001a). Fifty percent of the 174 women included in this phase of data collection were from the origi nal sample of child sexual abuse victims and the other 50% were from the matched comparison group (Siegel & Williams, 2003, p. 911). As previously explained, members of the original sample were selected based on their childhood experience of 8 In b oth Wave 2 and Wave 3 comparison groups were identified and matched to the original sample based on the hospital records. However only during Wave 3 were interviews conducted with the comparison group. No significant differences were found in race, age at time of hospital visit, or median family income during 1973 1975 (based on census tract data) between the original or comparison groups (Siegel & Williams, 2001a). 9 A significantly greater percentage of the women in the original sample of childhood sex ual abuse victims than of the comparison group were located and agreed to the interview. Researchers concluded that this difference was due to the wave of interviews that occurred in 1990 1991, in which updated contact information was collected from the o riginal victims (Siegel & Williams, 2001a).

PAGE 109

100 sexual abuse. Not surprisingly, and in concurrence with the rate of self report of child sexual abuse in the general population ( Briere & Elliot, 2003) 31% of those in the matched comparison group reported sexual victimization before they reached the age of 13. As w as done by previous researchers utilizing these data (Siegel & Williams, 2003), rather than deleting data from the participants who self reported child sexual abuse from the analyses, as they are inappropriate for inclusion in the matched or comparison gro victims. In their previous research, Siegel and Williams (2003) ran analytic tests comparing the official victims of sexual abuse (the initial sexual abuse victims examined at the hospital) and the self reported victims. Statistically significant differences were not found in the characteristics or elements (e.g., perpetrator was a family member or stranger, penetration during sexual abuse) of their sexual victimizations bef ore the age of 13 (Siegel & Williams, 2001a). The ethnic diversity of the full sample of females used in these analyses consists of 89% African American, 7% Caucasian, 2% Hispanic, 1% Native American, and 2% described their ethnicity as biracial. At the t ime of the follow up interviews in 1996 1997, the average age of the sample was 31.6 years old. The average age of the sample participants at the time of their hospital visit was 8.4 years old. The average family median income based on census tract info rmation at the time of the collection of the original data (1973 1975) was categorized as lower income 10 i.e., $7,491 was the 10 The median household income in 1975 in the United States was $13,720, with $8,779 estimated to be the median income for African American families in 1975 (U.S. Bureau of the Census, 1977). The threshold of p overty for a family of six in 1975 was $7,316 (U.S. Bureau of the Census, 2009).

PAGE 110

101 median income for the original sample and $7,265 was the median income for the comparison group (Siegel & Williams, 2001a). Measu res Caregiver Strain As shown in Figure 2, the caregiver strain latent variable was specified from four drinking problem, (2) drug problem, (3) mental health problem, and (4) domestic violence commonly been included in research or surveys exploring the effects of life adversity on youth or family functioning ( Finkelhor, Hamby, Ormrod, & Turner, 2005; Turner & Butler, 2003). In addition, these observed indicators of caregiver strain were commonly noted as risk factors among minors who had been victimized in child sex trafficking in prostitution (Estes & Wiener, 2005; Clawson, 2009). Suc h information regarding the regarding caregiver/child relationships has found that for African American children in lower income families, the mother/child relationship is typ ically the primary caregiver relationship (for review, see Liang et al., 2006). The observed indicators of caregiver adversity were based on responses to four questions, asking each participant questions regarding her biological mother or the person that the participant considered her mother figure while a youth: (1) Was there ever a time while you were growing up when she drank heavily or had a drinking problem? (2) Did she ever have a problem using drugs while you were growing up? (3) While you were gro wing up, did she ever have any serious emotional or mental

PAGE 111

102 problems such as being depressed a lot, or having an uncontrollable temper, or being very fearful of things such as being afraid to go out or anything else? (4) D id you ever witness your parents (d efined as a couple, a man and woman living together with you, who were parents to you) hit or throw things at one another? Although the reliability for this index was lower than the desired alpha level of .70 (Nu nnally, 1978) or adversities are not necessarily correlated with each other (Lin & Mieczkowski, 2010; Mazerolle, 1998). This variation is typically due to the fact that certain types of adversity occur relatively infrequently. In this index of caregiver adversity, the f requencies of the four observed indicators revealed a fair amount of variation; for example, domestic violence between parents was reported by 45% of the participants, while only 17% reported that their mother had a drug problem. Straus and Kantor (2005) argued that low internal consistency between items does not necessarily mean that an instrument is may be low due to Straus & Kantor, 2005, p. 25) Another reason that the alpha may be low for this index of child ne glect is that it contained a dependent not only on the magnitude of the 64). Lastly, the use of dichotomous or coarsely cat egorized data has been found to result in an attenuation of correlations between observed variables (Bollen & Barb, 1981). As a consequence, measures of internal consistency based upon inter item correlations of dichotomized items, such as the items inclu ded in this index, will be lower

PAGE 112

103 than if the items included in the index possessed a greater number of categories of response (Barrett, 2010). Additionally, principal components analysis (unrotated factor solution) was run using SPSS version 14.0 (SPSS, 20 05) with four items included in this index of caregiver strain, resulting in a range of factor loadings from .49 to .78. SPSS (2005) provides two assessments for evaluating the adequacy of correlation matrices for factor analysis. The Kaiser Meyer Oklin (KMO) sampling adequacy is an index that compares the magnitudes of the observed correlation coefficients to the magnitudes of the partial correlation coefficients. The KMO should be greater than .60 (Dzuiban & Shirkey, 1974). The KMO for the items include d in the caregiver strain latent variable was .66, reaching the recommended value. Also, Bartlett's test of sphericity is used to test the null of sphericity was significant (p = .000), as recommended (Dzuiban & Shirkey, 1974). Five of the six inter item correlations were statistically significant with correlations ranging from .13 to .39, with an average inter item correlation of .26. Only the coefficient of dom estic violence with drug problem was non significant, = .13, p = .08. These two items were the indicators with the greatest variation in frequency (45% vs. 17%). W using confirm atory factor analysis (CFA) with the Mplus program (Muthn & Muthn, 1998 2007) prior to their inclusion in the full structural equation model. This step in the analytic plan further assessed the level of item loading using the weighted least squares mean and variance adjusted (WLSMV) estimator which is useful in factor analysis with

PAGE 113

104 ordinal and dichotomous variables because it is robust to violations of multivariate normality (see Analytic Strategy and r esults described in Chapter 6). Child Neglect Also shown in Figure 2, the child neglect latent variable was specified from five observed indicators: (1) insufficient parental supervision; (2) insufficient food; (3) inadequate medical care; (4) inadequate parental care due to drug or alcohol use; and (5) l ack of love. Although operationally defining child neglect is considered one of the greatest challenges in researching child maltreatment, these indicators cover the key dimensions of child neglect noted by researchers including: failure to provide health care, inadequate supervision, inadequate nutrition, and failure to provide emotional nurturing (Barnett et al., 2005; Hildyard & Wolfe, 2002; Straus & Kantor, 2005). to five questions, asking whether her parents (defined as persons that the participant considered her mother figure or her father figure while a youth ) ever: (1) had to leave her home alone, even when they thought an adult should be there; (2) were unable to make sure she got the food she needed; (3) were not able to make sure she got to a doctor or hospital when she needed to; (4) were so drunk or high they had a problem taking care of her; and (5) were so caught up with their own problems that they were unable to show or tell her that they loved her. These interview questions pertaining to child neglect were similar to those used in previous research on child maltreatment (Finkelhor, Hamby, Ormrod, & Turner, 2005; Hildyard & Wolfe, 2002; Turner & Butler, 2003). No was coded adequate level of internal consistency (Nunnally, 1978) As with the measure of

PAGE 114

105 caregiver strain, the items were further analyzed using the WLSMV estimator t hat is designed for analyzing models with non normally distributed categorical indicators because it is robust to violations of multivariate normality (see Analytic Strategy and results described in Chapter 6). Child Physical Abuse with Injury Physical a buse is generally defined as "any nonaccidental physical injury to the child" and can include striking, kicking, burning, or biting the child, or any action that results in a physical impairment of the child ( Child Welfare Information Gateway, 2007). The harm standard that is generally used for defining child physical abuse states that Perrin & Perrin, 1998, p. 61). Based on such definitions Did any punishments by your parents (defined as persons that the participant considered her mother figure or her father figure while a youth) cause physical injury which required for inclusion in the proposed model. The interviewer did not inquire further as to whether or not medical attention was actually sought for the injury. This one item measure of child physical abuse was not cons idered as reliable for assessing the occurrence of child physical abuse as a measure based on various indicators or obtained from multiple sources (Malinosky Rummell & Hansen, 1993). However, the child physical abuse measure used in this study was similar to measures used by previous researchers. For example, a prospective study on child abuse by Brown, Cohen, Johnson, and Salzinger (1998), which gathered child abuse data during

PAGE 115

106 the same time period as the data gathered for this study, used a very similar measure for nyone they lived with ever hurt them physically so that they were still injured or bruised the next day, could not go to In a s tudy of the effects of abuse based on the level of severity, Leserman et al. (1997) found that women who had experienced more severe forms of physical abuse such as abuse that they experienced as life threatening suffered from greater psychological consequ ences. Milder, less violent forms of physical abuse did not result in negative outcomes. So although this measure of child physical abuse could be considered relatively severe and thereby may exclude some children who were physically abused without bein g injured to the point of needing medical attention, in light of previous research the measure used in this study was considered adequate for identifying those who experienced child physical abuse at a severity level that would affect their risk for revict imization or other types of consequential impacts. The observed Juvenile Sexual Victimization fondling), force, or sexual contact with someone who was five years older than the Incidents that did not involve genital contact were not counted as sexual abuse for the purpose of this study (Banya rd, Williams, & Siegel, 2003, p. 339). Adolescent sexual sexual contact and force or involved genital contact that she considered nonconsensual

PAGE 116

107 with someone five or mor In a review of research definitions and measures of sexual abuse, the one used in this study is generally considered intermediate, with some researchers using a broader designation that includes noncontac t sexual victimization (e.g., exhibitionism) and nongenital contact (e.g., kissing, hugging) and other studies using a narrower definition of sexual abuse as only rape usually including oral, anal, or vaginal penetration (Roodman & Clum, 2001). The defin ition of sexual abuse used in research was found to influence the reported effect sizes of sexual abuse on psychological consequences in the predicted direction, i.e., the broader definitions were related to lower effect sizes in revictimization and more n arrow definitions were related to higher effect sizes (Roodman & Clum, 2001). Thus, noncontact sexual victimizations have not been found to result in the same long term psychological consequences or result in the same level of elevated revictimization ris k as the more severe forms of sexual victimization (Leserman et al., 1997; Roodman & Clum, 2001). In view of previous research findings, the definition for sexual abuse used in this study was considered adequate to measure victimization from sexual abuse, detecting those who were more likely to experience elevated risk for revictimization or other negative impacts as a result of sexual abuse. In previous research studies, sexual victimization occurring prior to reaching the age of 18 has been included as o ne measure, rather than including child and adolescent sexual victimizations as separate variables ( Muehlenhard, Highby, Lee, Bryan, & Dodrill, 1998; Senn, Carey, Vanable, Coury Doniger, & Urban, 2007; Siegel & Williams, 2001). Although these two conditio ns are not considered to be equivalent or conceptually identical, research has yet to determine at which developmental stage the

PAGE 117

108 time of the sexual victimization ( Classen et al., 2005; Downs, 1993; Humphrey & White, 2000; Kaukinen & DeMaris, 2005 ). However, research has consistently found that multiple incidents of sexual victimization or cumulative trauma across life stages results in a greater likelihood of revictimizati on, as well as increasing the likelihood of numerous other mental health disorders (Arata, 2002; Classen et al., 2005; Humphrey & White, 2000) Therefore, juvenile sexual victimization was coded in a manner that captured the ion as well as the effects of experiencing any sexual victimization as a minor (Siegel & Williams, 2001, p. 24). Participants who reported Participants reporting either being sexually abused as a child (i.e., before the age of 13) or as an adolescent (i.e., while aged 13 reported sexual victimization during both childhood and adolescence or double victims dicator was included in the models as an ordered categorical variable rather than as a continuous variable. Age of Initial Drug or Alcohol Use Participants in the study reported the age they first used drugs and the age they first used alcohol. These two variables were moderately correlated ( r = .49, p < .001). They also contained a relatively high amount of missing data compared to the other variables that were included in the study models, i.e., information on initial age of drug use was missing from 34 cases or 20% of the sample and information on the initial age of alcohol use was missing from 16 cases or 9% of the sample. As both the use of drugs and alcohol have been found to increase vulnerability to sexual victimization,

PAGE 118

109 including victimization in child sex trafficking in prostitution (Abbey, 2002; Clawson, 2009; Estes & Weiner, 2005; Messman Moore et al., 2008; Messman Moore et al., 2009; Miller, 2008b), the reported initial age of alcohol use and the reported initial age of drug use were pooled i nto one variable. The youngest reported age from either variable was recorded as the initial age of drug or alcohol use. The combining of the two variables resulted in a slight decrease in the amount of missing data. The combined variable had missing dat a from 14 cases or 8% of the sample. The reported age of initial drug or alcohol use was included in the models as a continuous variable. Running Away The observed indicator for running away was dichotomous with participants in the study reporting whet her or not they ran away from home before the age of 18. No Relational Shame Siegel and Williams (2001a, 2003) collected responses from an array of s (1985) Traumagenic Theory of Child Sexual Abuse: traumatic sexualization (i.e., a dysfunctional linking of sex with love or the inappropriate use of sex to obtain approval) and stigmatization (i.e., negative self concept, typified by feelings of shame an d scale measured the level of shame in the form of self denigratory beliefs and behaviors pertaining to romantic/sexual relationships. Full or partial versions of the Belief Inventory have been regularly used as research instruments to measure self denigratory beliefs resultant of sexual abuse (Edmond, Rubin, & Wamback, 1999; Leach, Freshwater,

PAGE 119

110 Aldridge, & Sunderland, 2001; Pitts & Waller, 1993; Price, Hilsenroth, Petre tic Jackson, & Bonge, 2001; Reid & Sullivan, 2009b; Siegel & Williams, 2003; Waller, Ruddock, & Pitts, 1993). postulation that females are likely to resp ond to strain with self den igratory emotions, such as shame. In addition, this measure of relational shame was consistent with propositions within gendered pathways theory regarding the contextual effects created by societal pressures relevant to adolescent females, specifically th at girls experience a process of devaluation during adolescence and may perceive male approval as a way to resolve such feelings of personal worthlessness (Belknap & Holsinger, 2006; Chesney Lind & Shelden, 2004; Giordano et al., 2006). Shame pertaining t o romantic/sexual relationships in girls may trigger acquiescing to the manipulation and demands of sex traffickers (Belknap & Holsinger, 2006; Chesney Lind, 1989; Chesney Lind & Shelden, 2004; Giordano et al., 2006; Miller & Mullins, 2006; Steffensmeier & Allan, 1996). To construct this measure, participants were asked whether the following statements were true or false for them all or most of the time: (1) in your opinion, no man would care for you without a sexual relationship; (2) in your opinion, onl y bad, worthless guys would be interested in you; (3) you use sex to get something you want or need; (4) you find yourself in awkward sexual situations; (5) you get into trouble because of your sexual behavior; (6) you control others through the use of sex Responses were combined to create a scale with values ranging from 0 to 6. Responses to the individual items included in the scale were not available in the data. Therefore, it was not possible to specify a latent variable from the individual scale item s. Higher scores on the scale

PAGE 120

111 indicate higher levels of shame in the form of self denigratory beliefs and behaviors. many observations clustered at zero resulting in a highly skewed distribution. As this variable was not normally distributed, the variable was included in the models as an ordered categorical variable rather than as a continuous variable. Prostituted as a Minor The observed indicator, prostituted as a minor was dichotomous with participants During the interview with the study participant s, there was no inquiry as to whether a sex trafficker or a need such as food, shelter, or drugs, drew the minor into prostitution. However as clarified earlier, both situations are considered commercial sexual exploitation of a child and the procurer of sex with a minor would be legally considered a sex trafficker if there were no involvement of a third party (i.e., pimp, family member) profiting from the sex act of the minor (Adelson, 2008; DOJ, n.d.). In addition, prior studies have found that the maj ority of girls involved in prostitution in the United States were under the control of sex traffickers (Albanese, 2007; Estes & Weiner, 2005). Moreover, the effects of both child sex trafficking involving a third party sex trafficker and the forms of chil d commercial sexual exploitation that frequently have been labeled survival sex or trading sex have been found to be highly detrimental (Albanese, 2007; DOS, 2008, 2009; Estes & Weiner, 2005; FACJJ, 2007; Inciardi, 1993). In light of previous findings and the broad legal definition of sex trafficker, the measure of being

PAGE 121

112 prostituted as a minor available in the data was considered sufficient for use in the study models. Analytic Strategy The analytic strategy of this study involved several ordered steps. D escriptive statistics of the demographic attributes of the sample and observed variables included in the structural equation models were examined to clarify key characteristics of the study sample. As the study sample was purposefully collected and involve d a matched research design, the descriptive statistics were not representative of general population parameters. Bivariate statistics of the observed variables included in the models were computed, reporting the strength, direction, and level of statisti cal significance for each of the associations between all of the variables included in the study. As nominal, ordinal, and continuous variables in were included in the model, the coefficients were computed using Pearson product moment correlation, point b iserial correlation, V or phi. In order to further examine the relationships between several key study variables, additional bivariate analyses using the Chi square test of independence were conducted. All proposed mo dels were tested via structural equation modeling (SEM) using the Mplus program (Muthn & Muthn, 1998 2007). SEM functions similarly to multiple regression, but uniquely allows for the modeling of associations between observed variables and latent variab les (Kline, 2005). Distinct from multiple regression, SEM allows for the assessment of a system of equations. SEM integrates factor analysis to identify latent variables with path analysis to estimate the st ructural model. As is commonly done when emplo ying SEM, two steps were included in the analytic process:

PAGE 122

113 (1) the measurement model that identifies and validates the latent variables was tested independently, and (2) the fully proposed SEM was tested (Anderson & Gerbing, 1988). Measurement Models Facto r analysis is a multivariate statistical technique that can be used for several purposes within research. Confirmatory factor analysis (CFA) is commonly used to confirm hypotheses or preconceptions about data patterns and as a foundational step of model s pecification within SEM called the measurement model (Kline, 2005). Factor analysis is based on the premise that shared variance among two or more manifest indicators or variables may possibly indicate the existence of some underlying, latent factor (Kim & Mueller, 1978). Factors may refer to observable phenomena or cause indicators, but more typically reflect latent or hypothetical constructs (Kline, 2005). By utilizing CFA to identify latent variables, measurement error in the model is reduced through the use of several indicators for each of the latent variables (Kline, 2005). So in this first step of the analysis, CFA was used to identify and estimate the measurement model and measurement paths, specifying the structure of each of the latent variabl es. The findings of good model fit, along with strong and significant loadings of the manifest indicators on their respective latent variables, were used to determine the quality of the measurement model (Kline, 2005). Structural Models Only after the a nalyses and assessment of the measurement models were each of the full structural equation models assessed to determine whether the theorized model adequately projected the actual observed relationship patterns in the data (Kline, 2005). A collection of t ests was used to assess various aspects of model fit of both the

PAGE 123

114 measurement models and the structural models. The appraisal of the various model fit indicators was based on guidelines provided by Kline (2005), Hu and Bentler (1999), and Yu (2002). Chi square is a statistic that tests the degree of misfit between the proposed model and a null model (Hu & Bentler, 1999). A non significant chi square suggests that the proposed model fits the data adequately. Other indicators assessing model fit were also used, including: normed chi square (NC) of two or less; Comparative Fit Index (CFI) of .95 or higher; Tucker Lewis Index (TLI) of .95 or higher; Root Mean Square of Approximation (RMSEA) of .06 or less; and Weighted Root Mean Square Residual (WRMR) of .9 0 or less (Hu & Bentler, 1999; Kline, 2005). With binary outcomes, as in the proposed model, slightly more stringent guidelines of CFI of .96 or higher and RMSEA of .05 or less have been found to indicate good model fit (Yu, 2002). The CFI and the TLI ar e incremental fit indexes and report the improvement in model fit by comparing the specified model with a baseline model (Hu & Bentler, 1999). The RMSEA and WRMR are considered absolute fit indexes and assess how well the specified model replicates the ob served patterns in the data (Hu & Bentler, 1999). Additionally, Mplus provides modification indices to assist in evaluating model fit (Muthn & Muthn, 1998 2007). Different from the model fit indicators noted above that provide an evaluation of global model fit, modification indices are useful in evaluating models for specific points of ill fit (Brown, 2006). A modification index for a specific parameter notes the expected decrease in the overall model Chi square test statistic (i.e., how the model ov erall fit would improve) if the particular parameter was adopted or fixed in the specified model (Brown, 2006). Models that adequately represent the observed patterns in the data, or have good fit, should produce modification indices that are small in ma gnitude or less than 4.00 (Brown, 2006). Modification indices with values

PAGE 124

115 greater than 4.00 were evaluated in the study models to determine if there were particular areas of poor fit (Brown, 2006). Finally, the key proposed structural paths, both direct and indirect, were assessed to determine whether they provided support for the study hypotheses. Standardized linear regression coefficients were used to report the coefficients for the key structural paths for all continuous observed indicators and late nt variables, except for the dichotomous dependent variables. Physical abuse, running away, and being prostituted as a minor were dichotomous variables, and consequently the estimate for these variables took the form of a conditional probability based on the value(s) of predictor variables (Liao, 1994). Probit coefficients were reported for those relationships. Therefore, the magnitudes of the coefficients corresponding to continuous or latent variables were not directly comparable to the coefficients co rresponding to dichotomous variables. All of the unstandardized coefficients, standardized coefficients, and standard errors for all model estimates were reported. Weighted Least Squares Mean and Variance Adjusted Estimator As the measurement models and t he structural models proposed for this study included observed indicators that were ordinal and dichotomous, the analyses of all models relied on an estimator that was suited to such data. The weighted least squares mean and variance adjusted (WLSMV) esti mator available in the Mplus program (Muthn & Muthn, 1998 2007) was utilized because of its potential to fit models containing ordinal and dichotomous variables (i.e., noncontinuously distributed measures).

PAGE 125

116 Most structural equation models are estimate d using a maximum likelihood (ML) estimation method that rests on key assumptions including the assumption of multivariate normal distributions among the indicators within the model (Brown, 2006; Kline, 2005). However, the WLSMV estimation method used in the analyses of this study does not make the assumption of multivariate normality (Kline, 2005; Muthn, 1993). Using the WLSMV estimator eliminates the need to use other remedial strategies for indicator non normality, such as bootstrapping, item parcelin g, or data transformation (Brown, 2006). Specifically, the WLSMV estimator is designed to provide more accurate parameter and standard error estimates, as well as chi square values, for situations in which data are coarsely categorized and/or non normal ly distributed (DiStefano & Hess, 2005; Flora & Curran, 2004; Muthn, 1993; Muthn, du Toit, & Spisic, 1997). WLSMV adjusts the estimates by employing a weighting process that uses the diagonal of the weight matrix in the estimation rather than the full we ight matrix, resulting in robust standard errors for use in the hypothesis tests (DiStefano & Hess, 2005; Flora & Curran, 2004; Muthn et al., 1997). In simulation studies the WLSMV estimator has been shown to perform well, producing accurate test statis tics, parameter estimates, and standard errors that are much less vulnerable to the effects of increasing model complexity and decreasing sample size in comparison to other methods of estimation (Brown, 2006; Flora & Curran, 2004). Further unpublished sim ulation studies conducted by Muthn indicated that when using the WLSMV estimator, sample sizes of 150 to 200 were sufficient for medium sized models, e.g., models with 10 to 15 indicators (cited in Brown, 2006, p. 389; see

PAGE 126

117 also, Bentler & Chou, 1987). Wi th sample sizes of 200, WLSMV has been shown to perform well with data that includes variables with floor or ceiling effects (Brown, 2006). Therefore, the use of the WLSMV estimator was considered to be suitable for this study based on the study sample si sized models incorporating up to 15 indicators, including dichotomous and ordered categorical variables. When using the WLSMV estimator with censored or categorical outcomes, Mplus handles missing data by using a complex method of pair wise deletion (Muthn & Muthn, 1998 2007). The process of considering missing data that is used by Mplus with the WLSMV estimator entails four steps (Muthn & Muthn, 1998 2007). The four steps include univariate probi t regression of all cases with data, bivariate probit regression with all available paired data, estimation of the weight matrix, and fitting the model using weighted least squares. The first two steps are similar to the missing at random technique for mi ssing data used with the maximum likelihood (ML) estimator in Mplus, thereby the process results in a superior handling of missing data than the use of simple pair wise deletion (Muthn & Muthn, 1998 2007). Although this technique for handling missing da ta is not considered to be as effective as others offered by Mplus that are not applicable when using the WLSMV estimator, the use the WLSMV estimator in the study models was considered essential despite this limitation (Brown, 2006; Muthn & Muthn, 1998 2007). Mplus also provides information on missingness within the models by reporting the proportion of data present and the availability of paired data between each of the 15 indicators included in the models (Muthn & Muthn, 1998 2007). The 15 indicato rs included in the models had minimal missing data (see Table 3 in Chapter 6) and only minor losses in covariance coverage were noted in the models.

PAGE 127

118 Different from structural equation modeling or confirmatory factor analysis using continuous variables, the residual variances of categorical indicators are not identified nor estimated when using the WLSMV estimator because the observed variances of the indicators are not analyzed (Brown, 2006). In this type of analysis, measurement error for categorical indi cators reflects the remainder of 1 minus the squared standardized item loading (Brown, 2006). The WLSMV estimator could not be used for analyzing interaction effects within structural equation models, as Mplus computes interaction effects using the ML e stimator (Muthn & Muthn, 1998 2007). Therefore, it was necessary to analyze the theorized moderating effects of relational shame using binary logistic regression. The analyses of the interaction effects are described in the next section. Supplementary Models Two of the study hypotheses incorporated the variable measuring relational shame. First, it was hypothesized that relational shame would be a consequence of experiencing child maltreatment, including child neglect, child physical abuse, and juveni le sexual victimization. Second, it was theorized that relational shame, resulting from childhood victimization experiences, would detrimentally inflate vulnerability to victimization in sex trafficking for escaping girls, i.e., girls who had run away or who had initiated drug or alcohol use to escape the pain of abuse. Based on the timing and content of the interviews with the participants, no information regarding the onset of relational shame was collected and available in the data. However, the expl oration of relational shame as a consequence of child maltreatment and as an amplifying source of vulnerability to being prostituted as a minor

PAGE 128

119 was pursued based on previous research and the theoretical framework of the study. Consequently, the effects o f relational shame were explored through the use of supplementary structural equation models to prevent the temporal ambiguity of this variable from introducing a potential threat to internal validity into the primary structural models (Shadish et al., 200 2). The supplementary models were analyzed in an identical manner as the primary structural models using the same model estimator, tests of model fit, and reporting of standardized linear regression coefficients or probit coefficients for the key structura l paths. The supplementary models incorporating the relational shame variable were useful for testing the hypothesis that relational shame was more likely to be experienced by girls who had been maltreated than by girls who had not been maltreated. Last ly, in order to test for the theorized interaction or moderating effects of relational shame, two binary logistic regression analyses were conducted. In the first logistic regression, being prostituted as a minor was regressed on running away, relational shame, and an interaction term of those two independent variables (Jaccard, 2001). In the second analysis, being prostituted as a minor was regressed on initial age of drug or alcohol use, relational shame, and an interaction term of those two independent variables.

PAGE 129

120 Chapter 6 Results The primary objective of this study is to identify and assess a pathway into victimization by child sex trafficking in prostitution, specifically exploring the impact of a family context characterized by caregiver strain and child maltreatment and the consequences of subsequent individual risk inflating responses to such strain. Prior to assessing the proposed structural equation models, preliminary analyses were conducted. First, the descriptive statistics of the study sample were examined to gain greater understanding of the data being utilized to test the suitability of the proposed models. In addition, bivariate analyses were computed to examine the strength, direction, and statistical significance of the associatio ns between the 15 observed variables selected for inclusion in the structural equation models with the objective of evaluating the necessity and appropriateness of conducting further analysis. Preliminary Analyses Descriptive statistics of the observed var iables included in the study are summarized in Table 3. Due to the original data collection procedures and matched research design (i.e., matching to an original sample of all child sexual abuse victims), the percentage of the sample that reported sexual victimization as a minor is not reflective of general population rates (Briere & Elliot, 2003). Twenty six percent of the sample reported no juvenile sexual victimization. Fifty five percent reported juvenile

PAGE 130

121 sexual victimization, either during childhood or during adolescence. Of those reporting sexual victimization during one life stage (i.e., during childhood or during adolescence), 85% reported victimization during childhood or before the age of 13 and 15% reported sexual victimization during adolescen ce or between 13 and 17 years of age. Eighteen percent of the sample reported being sexually victimized during both childhood and adolescence. Table 3. Descriptive Statistics of Study V ariables (N= 174) Variables M SD Range Proportion of Missing Data 1. Mother with a drinking problem .30 .46 0 1 .03 2. Mother with a drug problem .18 .38 0 1 .02 3. Mother with a mental health problem .39 .49 0 1 .02 4. Domestic violence in home .45 .50 0 1 .00 5. Lack of supervision .35 .48 0 1 .03 6. Lack of food .11 .32 0 1 .03 7. Lack of medical care .07 .25 0 1 .03 8. Poor care due to drug or alcohol use .13 .33 0 1 .03 9. Lack of love .42 .49 0 1 .03 10. Child physical abuse .09 .29 0 1 .11 11. Juvenile sexual victimization .92 .67 0 2 .00 12. Ran away .4 0 .49 0 1 .00 13. Initial age of drug or alcohol use 14.35 3.35 4 26 .08 14. Relational shame 1.16 1.65 0 6 .01 15. CST prostitution .12 .33 0 1 .00

PAGE 131

122 Other key descriptors of the study sample were the presence and level of the observed variables consid ered to be risk inflating responses to strain. Potential risk inflating behaviors included in the model were running away from home and the initial age of drug or alcohol use. The self reported level of relational shame was included in the model as a po ssible risk inflating emotional response to strain. Forty percent of the participants reported that they ran away from home before they reached 18 years of age. The average reported initial age of drug or alcohol use was 14.3 years old and the most commo n initial age was 16 years old. The range of ages was from 4 years old to 26 years old, with 25% reporting that they began using drugs or alcohol by the age of 13, 50% by the age of 14, and 75% by the age of 16. Lastly, the average level of relational s hame was 1.16 out of a possible score of 6 on the scale. Fifty six percent of the participants reported no relational shame, 31% reported lower levels of relational shame (scored 1 3), and 13% reported higher levels of relational shame (scored from 4 6). Results for all bivariate analyses are displayed in Table 4. Statistically significant relationships were found among many of the observed variables, with small to moderate effect sizes. For example, the indicators of caregiver strain were significantly related to the various types of child maltreatment. The types of child maltreatment were negatively related to the initial age of drug or alcohol use and positively related to running away. Relational shame was significantly related to nine of the variab les, including several indicators of child neglect and juvenile sexual victimization. Being prostituted as a minor was significantly associated with seven of the variables, including juvenile sexual victimization, running away, initial age of drug or alco hol use, and relational shame.

PAGE 132

Table 4. Bivariate Correlations of All Study Variables ( N = 174 ) p < .05, ** p < .01 Note : Coefficients were comp uted using phi or point biserial correlation. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Mother w/drinking problem 2. Mother w/drug problem .20* 3. Mother w/mental healt h problem .37** .20** 4. Domestic violence .25** .13 .37** 5. Lack of supervision .24** .23** .21** .12 6. Lack of food .15 .28** .08 .21** .29** 7. Lack of medical care .15 .20* .09 .15 .31** .53* 8. Poor care drugs/alcohol .36** .35** .26** .20** .29** .43** .34** 9. Lack of love .30** .13 .41** .36** .33** .27** .26** .41** 10. Child physical abuse .14 .08 .24** .10 .20* .12 .11 .37** .31** 11. Juvenile se x victimization .17* .17* .22** .27** .25** .16* .18* .16* .20** .19* 12. Ran away .05 .18* .14 .18* .41** .21** .22** .20** .34** .18* .40** 13. Initial age of drug/alcohol .13 .19* .22** .14 .23** .16* .19* .16* .30** .23** .26** .29** 14. Relational shame .10 .12 .21** .21** .32** .08 .23** .15 .31** .13 .27** .34** .24** 15. CST prostitution .08 .16* .10 .06 .09 .17* .13 .15 .16* .16 .26** .20** .20* .28** 123

PAGE 133

124 In order to further examine the relationships of several key study variables additional bivariate analyses using the Chi square test of independence were conducted. Tables 5 8 display the associations of juve nile sexual victimization, running away, initial age of drug or alcohol use, and the reported level of relational shame with the study outcome variable, being prostituted as a minor. Table 5. Cross Tabulations of Prostituted as a Minor by Juvenile Sexua l Victimization Juvenile Sexual Victimization Total None Either During Childhood or Adolescence Both Prostituted Minor No 45 (98%) 85 (88%) 23 (72%) 153 (88%) Yes 1 ( 2%) 11 (12%) 9 (28%) 21 (12%) Total 46 (26 %) 96 (55 %) 32 (18 %) 174 (100%) Note: 2 (2) = 12.05, p < 0.01 As shown in Table 5, youth who experienced juvenile sexual victimization were more likely to have reported being prostituted as a minor, 2 (2) = 12.05, p < 0.01. Notably, only one participant or 2% of tho se who reported no juvenile sexual victimization also reported being prostituted as a minor. In contrast, 11 participants or 12% of those who reported being sexually victimized as a child or as an adolescent reported being prostituted as a minor. Further more, nine participants or 28% of those

PAGE 134

125 and adolescence reported that they were prostituted as a minor. Importantly, the majority of those who experienced juvenile se xual victimization did not report being prostituted as a minor. Nevertheless, these results suggest that being sexually victimized as a minor may increase the likelihood of further victimization in prostitution. Table 6. Cross Tabulations of Prostituted as a Minor by Running Away (N=174) Running Away Total No Yes Prostituted Minor No 97 (93%) 56 (80%) 153 (88%) Yes 7 ( 7%) 14 (20%) 21 (12%) Total 104 (60 %) 70 (40 %) 174 (100%) Note: 2 (1) = 6.94, p < 0.01 The r esults in Table 6 indicate that girls who ran away were more likely to have been prostituted than those who did not run away, 2 (1) = 6.94, p < 0.01. Twenty percent of the girls who ran away reported being prostituted; comparatively, only 7% of those who did not run away reported being prostituted. Although only 40% of the full sample reported running away as a minor (see Table 3), twice as many youth who ran away reported being prostituted compared to the number of non runaways who reported being prosti tuted (14 participants vs. 7 participants). Once more, the majority of girls in

PAGE 135

126 the sample, whether they ran away or not, did not report being prostituted as a minor. Nevertheless, these results suggest that running away as a minor may increase the likel ihood of victimization in prostitution. Table 7. Difference in Means of Initial Age of Drug/Alcohol Use (N=160) Initial Age of Drug or Alcohol Use Count Mean SD Prostituted Minor No 141 14.6 3.33 Yes 19 12.5 2.98 Note: t (158) = 2.55, p < 0. 05 The results displayed in Table 7 indica te that there was a significant difference in the average age of initial drug or alcohol use specified by those who reported being prostituted and those who did not, t (158) = 2.55, p < 0.05. The average age o f initiation of drug or alcohol use for the whole sample was 14.4 (see Table 3), non prostituted girls' average age at initial drug or alcohol use was 14.6, and the prostituted girls' average age at initial drug or alcohol use was 12.5.

PAGE 136

127 Table 8. Cross T abulations of Prostituted as a Minor by Relational Shame (N=172) Level of Relational Shame Total None Low High Prostituted Minor No 92 (95%) 44 (83%) 15 (68%) 151 (88%) Yes 5 ( 5%) 9 (17%) 7 (32%) 21 (12%) Total 97 (56 %) 53 (31 %) 22 (13 %) 172 (100%) Note: 2 (2) = 13.52, p < 0.01 To simplify the examination of cross tabulations of relational shame and being prostituted as a minor, the ordinal variable measuring relational shame was condensed from seven levels into three level s. The participants in the study who reported no shame (scores ranging from 1 relational shame (scores ranging from 4 participants or 56% reported no shame, 53 participants or 31% reported low levels of shame, and 22 participants or 13% reported high levels of shame. As shown on Table 8, participants who reported higher leve ls of relational shame were more likely to have reported being prostituted as a minor, 2 (2) = 13.52, p < 0.01. Notably only five participants or 5% of the sample that reported no relational shame reported being prostituted as a minor. In contrast, 17% of the participants with low levels of relational shame reported being prostituted as a minor and 33% of the participants

PAGE 137

128 with high levels of relational shame reported being prostituted as a minor. These results indicate that relational shame may increase the likelihood of victimization in prostitution. In summary, the results of the preliminary analyses indicated that caregiver strain was associated with the occurrence of child maltreatment. Additionally, associations were found between the various types of child maltreatment and running away, initial age of drug or alcohol use, and relational shame. Furthermore, variables considered to be risk inflating behavioral responses to strain (i.e., running away, initial age of drug or alcohol use) were both fou nd to have a significant effect on the likelihood of having been prostituted as a minor. Lastly, the level of relational shame was found to influence the likelihood of having been prostituted as a minor. This collection of findings adequately supports t he study assumptions that caregiver strain, the strain of child maltreatment, and risk inflating responses to strain converge, elevating the likelihood of victimization in child sex trafficking in prostitution, and thereby provided justification for furthe r analysis of the data utilizing structural equation modeling. Multivariate Analyses As noted in Chapter 5, the analytic strategy outlined for this study included the identification and assessment of four distinct models designed to explain victimization in child sex trafficking in prostitution. The first model included only child neglect as the form of child maltreatment to be assessed, the second model featured only child physical abuse, the third model only juvenile sexual victimization, and the fourt h model incorporated all three forms of child maltreatment. The assessments of each of the proposed models specified: (a) the measurement model, (b) the structural equation model, and (c) the supplementary model exploring the effect of relational shame (s ee

PAGE 138

129 Analytic Strategy in Chapter 5). The results from the analyses of the four models are reviewed sequentially in the sections that follow. Child Neglect Model As shown in Figure 3, the measurement model for the two latent variables incorporat ed into the child neglect model evidenced an adequate fit to the data. Although the Chi square was narrowly significant, 2 (16) = 26.57, p = .05, all other indices demonstrated good fit (NC = 1.66; CFI = .97; TLI = .97; RMSEA = .06; WRMR = Figure 3 Child Neglect Measurement Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (16) = 26.57, p =.05; NC =1.66; CFI =.97; TLI =.97; RMSEA =.06; WRMR=.80 .80** Lack of Love .85**

PAGE 139

130 .80). The loa dings of the indicators on the latent variables were statistically significant and strong, ranging from .64 to .91. As displayed in Figure 3, MPlus also provided the correlation coefficient between the two identified latent variables. These results provid ed support for the measurement model, which suggests a reasonable foundation for testing structural relationships. T he structural equation model featuring child neglect evidenced adequate fit to the data with a non significant Chi square, 2 (27) = 36.55 p = .10 (see Table 9). All other indices demonstrated good model fit (NC = 1.35; CFI = .97; TLI = .97; RMSEA = .05; WRMR = .78). The sole modification index (MI) with a value greater than 4.00 was a possible modification of lack of medical care with la ck of food (MI = 4.99). The suggested modification of a parameter allowing these two indicators of child neglect to co vary was the only MI over 4.00 noted in any of models included in the study. As this modification would not make substantive changes in the models and the value was relatively small, such modification to the models was not performed. Next, the structural paths linking key latent and observed variables relevant to the child neglect model hypotheses were examined to assess the proposed relat ionships (see Figure 4). Standardized linear regression coefficients are reported for all dependent variables in the model except for the dichotomous variables included in the model. Probit coefficients are reported for those relationships. The unstanda rdized coefficients, standardized coefficients, and standard errors for all model estimates are reported in Table 9.

PAGE 140

131 Table 9. Unstandardized, Standardized, and Critical Ratio Values for the Child Neglect Model Displayed in Figure 4 (Standard Errors in Pa rentheses; N = 174) Parameter Estimate Unstd. Coeff./Probit Std. Coeff. Critical Ratio ( Z ) Measurement Model Estimates Strain by Domestic Violence .84 (.30) .64 2.75 Strain by Alcohol Problem 1.20 (.40) .70 2.81 Strain by Drug Problem .88 (.33) .59 2.65 Strain by Mental Health Problem 1.32 (.37) .74 3.54 Neglect by Poor Supervision .61 (.21) .64 2.94 Neglect by Lack of Food 1.65 (.56) .81 2.94 Neglect by Lack of Medical 2.14 (.96) .87 2.22 Neglect by Poor Care because of Drug/Alcohol U se 2.58 (1.06) .91 2.44 Neglect by Lack of Love 1.91 (.66) .85 2.91 Strain with Neglect .56 (.18) .80 3.05 Structural Model Estimates Neglect on Caregiver Strain .93 (.31) .76 2.99 Running Away on Neglect .80 (.23) .62 3.51 Initial Age of D/A Use on Neglect 1.46 (.42) .44 3.46 Prostituted Minor on Running Away .34 (.17) .37 1.96 Prostituted Minor on Age of D/A .10 (.05) .29 1.90 Indirect Effect Prostituted Minor on Caregiver Strain .38 (.17) .27 2.25 Indirect Effect Prostituted Minor on Neglect .41 (.18) .35 2.35 Note: 2 (27) = 36.55, p =.10; NC=1.35; CFI=.97; TLI =.97; RMSEA=.05; WRMR=.78

PAGE 141

132 Caregiver strain was positively related to child neglect with a larg e direct effect size ( std. coefficient = .76, p < .01). The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by neglect, running away, and initial age of drug or alcohol use revealed a significant and positive relat ionship with a moderate effect size ( std. coefficient = .27, p < .05). Running Away ( R 2 =. 39) Age of Initial Drug/Al. Use ( R 2 = .19) CST/ Prostitution ( R 2 =. 27) 76** .44** .62** .29 .37* Figure 4. Child Neglect Structural Equation Model Not e: p < 0.05, ** p < 0.01, two tailed Note: 2 (27) = 36.55, p =.10; NC =1.35; CFI =.97; TLI =.97; RMSEA =.05; WRMR =.78 Caregi ver Strain Child Neglect ( R 2 =. 58)

PAGE 142

133 Child neglect was positively related to running away with a large effect size ( std. coefficient = .62, p < .01) and negatively related to age of initial drug or alcohol use with a m oderate effect size ( std. coefficient = .44, p < .01). The combined indirect effect of child neglect on being prostituted as a minor as mediated by running away and age of initial drug or alcohol use revealed a significant, positive relationship with a m oderate effect size ( std. coefficient = .35, p < .05). Running away had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .37, p < .05). Age of initial drug or alcohol use h ad a negative relationship with being prostituted as a minor ( std. coefficient = .29), with a critical value slightly lower than the predetermined significance level for the analysis ( z = 1.90, p = .06) The standardized coefficients of each key path a nd the proportion of variance explained in the endogenous variables are noted in Figure 3 and were as follows: child neglect, R 2 = .58; running away, R 2 = .39; initial age of drug or alcohol use R 2 = .19; and being prostituted as a minor, R 2 = .27. In th e child neglect model the structural paths connecting the key latent and observed variables supported the study hypotheses, with only one proposed path coefficient failing to achieve the predetermined significance value for a two tailed hypothesis test. Th e final step in the analysis of the child neglect model included the exploratory examination of the origins and effects of relational shame (see Analytic Strategy in Chapter 5). As planned, a supplementary child neglect structural equation model was analy zed to examine: (1) the effects of child neglect on the level of reported relational shame; and (2) to explore the effects of relational shame on the likelihood of being

PAGE 143

134 prostituted as a minor, seeking to discern whether relational shame may intensify the effect of running away and the initial age of drug or alcohol use on being prostituted as a minor. The child neglect model including relational shame also evidenced adequate fit to the data with a non significant Chi square, 2 (31) = 41.96, p = .09. The o ther indices demonstrated good model fit (NC = 1.20; CFI = .97; TLI = .97; RMSEA = .05; WRMR = .79). The structural paths linking key latent and observed variables relevant to the child neglect model hypotheses were again examined to assess the proposed relationships. Child neglect was positively related to relational shame with a large effect size ( std. coefficient = .50, p < .01). The associations of child neglect with running away ( std. coefficient = .64, p < .01) and initial age of drug or alcohol u se ( std. coefficient = .44, p < .01) were minimally affected by the inclusion of relational shame in the model. The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by neglect, running away, initial age of drug/alc ohol use, and relational shame was minimally increased in this supplementary model ( std. coefficient = .30, p < .05). The combined indirect effect of child neglect on being prostituted as a minor as mediated by running away, age of initial drug or alcoh ol use, and relational shame was also minimally increased by the addition of relational shame ( std. coefficient = .40, p < .05). Relational shame had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .37, p < .05). Running away no longer had a significant relationship with being prostituted as a minor with a reduced effect size ( std. coefficient = .19). Also, the effect size of age of initial drug or alcohol use on the likel ihood of being prostituted as a minor decreased ( std.

PAGE 144

135 coefficient = .21). The proportion of variance explained of being prostituted as a minor in this supplementary model was R 2 = .31. Child Physical Abuse Model As shown in Figure 5, the measurement mo del for the latent variable, caregiver strain, contained in the child physical abuse model evidenced an adequate fit to the data. The Chi square was significant, 2 (2) = .34, p = .84, all other indices demonstrated good fit (NC = .17; CFI = 1.00; TLI = 1. 05; RMSEA = .00; WRMR = .16). Three of the four loadings of the indicators on the latent variable were statistically significant, ranging from .44 to .87. One indicator, mother with a mental health problem, was not statistically significant, yet showed t he highest loading among the four observed indicators, r = .87, z = 1.77, p = .07. These results provided support for the measurement model, which suggests a reasonable foundation for testing structural relationships. Figure 5. Child Physical Abuse a nd Juvenile Sexual Victimization Measurement Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (2) = .34, p =.84 ; NC =.17; CFI = 1.00; TLI = 1.05; RMSEA =.00; WRMR =.16

PAGE 145

136 Table 10. Unstandardized, Standardized, and Critical Ratio Values for Child Ph ysical Abuse Model Displayed in Figure 6 (Standard Errors in Parentheses; N = 174) Parameter Estimate Unstd. Coeff./Probit Std. Coeff. Critical Ratio ( Z ) Measurement Model Estimates Strain by Domestic Violence .84 (.31) .62 2.68 Strain by Alco hol Problem 1.19 (.45) .68 2.68 Strain by Drug Problem .63 (.27) .44 2.34 Strain by Mental Health Problem 2.21 (1.25) .87 1.77 Structural Model Estimates Child Physical Abuse on Strain .88 (.31) .56 2.82 Running Away on Child Physical Abus e .56 (.15) .56 3.82 Initial Age of D/A Use on Child Physical Abuse 1.51 (.32) .55 4.67 Prostituted Minor on Running Away .30 (.16) .32 1.90 Prostituted Minor on Age of D/A .10 (.05) .28 1.89 Indirect Effect Prostituted Minor on Careg iver Strain .28 (.13) .19 2.11 Indirect Effect Prostituted Minor on Child Physical Abuse .32 (.12) .33 2.63 Note: 2 (14) = 10.57, p =.72; NC =.78; CFI=1.00; TLI=1.04; RMSEA=.00; WRMR=.56

PAGE 146

137 As shown in Table 10, t he structural equation model featu ring child physical abuse evidenced adequate fit to the data with a non significant Chi square, 2 (14) = 10.57, p = .72. All of the other indices demonstrated good model fit (NC = .78; CFI = 1.00; TLI = 1.04; RMSEA = .00; WRMR = .56). Next, the structural paths linking key latent and observed variables relevant to the child neglect model hypotheses were examined to assess the proposed relationships (see Figure 6). As mentioned in the results of the previous model, standardized li near regression coefficients are reported for all dependent variables in the model except for the dichotomous variables Running away and being prostituted as a minor are both Running Away (R 2 =. 31) Age of Initial Drug/Al. Use (R 2 = .40) CST/ Prostitution (R 2 =. 24) 56** .55** .56** .28 .32 Figure 6. Child Physical Abuse Str uctural Equation Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (14) = 10.57, p =.72; NC =.78; CFI =1.00; TLI =1.04; RMSEA=.00; WRMR= .56 Caregiver Strain Physical Abuse (R 2 =. 31)

PAGE 147

138 dichotomous variables. Therefore, probit coefficients are reported for those r elationships (Liao, 1994). The unstandardized coefficients, standardized coefficients, and standard errors for all model estimates are reported in Table 10. Caregiver strain was positively related to child physical abuse with a large direct effect size ( std. coefficient = .56, p < .01). The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by child physical abuse, running away, and initial age of drug or alcohol use revealed a significant and positive relationship w ith a moderate effect size ( std. coefficient = .19, p < .05). Child physical abuse was positively related to running away with a large effect size ( std. coefficient = .56, p < .01) and negatively related to age of initial drug or alcohol use with a larg e effect size ( std. coefficient = .55, p < .01). The combined indirect effect of child physical abuse on being prostituted as a minor as mediated by running away and age of initial drug or alcohol use revealed a significant, positive relationship with a moderate effect size ( std. coefficient = .33, p < .05). Running away had a positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .32) with a critical value slightly lower than the predetermined si gnificance level for the analysis ( z = 1.90, p = .06) Age of initial drug or alcohol use had a negative relationship with being prostituted as a minor ( std. coefficient = .28), with a critical value slightly lower than the predetermined significance le vel for the analysis ( z = 1.89, p = .06) The standardized coefficients of each key path and the proportion of varianc e explained in the endogenous variables are noted in Figure 5 and were as follows: child physical abuse, R 2 = .31; running away, R 2 = 31; initial age of drug or alcohol use R 2 =

PAGE 148

139 .40; and being prostituted as a minor, R 2 = .24. In the child physical abuse model the structural paths connecting the key latent and observed variables supported the study hypotheses, with two of the proposed path coefficients narrowly failing to achieve the predetermined significance value for a two tailed hypothesis test. The final step in the analysis of the child physical abuse model included the exploratory examination of the origins and effects of relatio nal shame (see Analytic Strategy in Chapter 5). As planned, a supplementary child physical abuse structural equation model was analyzed to examine: (1) the effects of child physical abuse on the level of reported relational shame; and (2) to explore the e ffects of relational shame on the likelihood of being prostituted as a minor, seeking to discern whether relational shame may intensify the effect of running away and the initial age of drug or alcohol use on being prostituted as a minor. The child physica l abuse model including relational shame also evidenced adequate fit to the data with a non significant Chi square, 2 (18) = 16.71, p = .54. The other indices demonstrated good model fit (NC = .93; CFI = 1.00; TLI = 1.01; RMSEA = .00; WRMR = .63). The s tructural paths linking key latent and observed variables relevant to the child neglect model hypotheses were again examined to assess the proposed relationships. Child physical abuse was positively related to relational shame with a large effect size ( st d. coefficient = .54, p < .01). The associations of child physical abuse with running away ( std. coefficient = .62, p < .01) and initial age of drug or alcohol use ( std. coefficient = .52, p < .01) were minimally affected by the inclusion of relational s hame in the model. The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by child physical abuse, running away, initial age of

PAGE 149

140 drug/alcohol use, and relational shame was minimally increased in this supplementary mode l ( std. coefficient = .23, p < .05). The combined indirect effect of child physical abuse on being prostituted as a minor as mediated by running away, age of initial drug or alcohol use, and relational shame was also minimally increased by the addition of relational shame to the model ( std. coefficient = .41, p < .01). Relational shame had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .35, p < .05). The effect size of the relationship of running away with being prostituted as a minor was reduced and non significant ( std. coefficient = .19). Also, the effect size of age of initial drug or alcohol use on the likelihood of being prostituted as a minor decreased ( std. coef ficient = .18). The proportion of variance explained of being prostituted as a minor in this supplementary model was R 2 = .30. Juvenile Sexual Victimization Model As shown in Figure 5, the measurement model for the latent variable, caregiver strain, in corporated into the juvenile sexual victimization model evidenced an adequate fit to the data. The Chi square was significant, 2 (2) = .34, p = .84, all other indices demonstrated good fit (NC = .17; CFI = 1.00; TLI = 1.05; RMSEA = .00; WRMR = .16). Thre e of the four loadings of the indicators on the latent variable were statistically significant, ranging from .44 to .87. One indicator, mother with a mental health problem, was not statistically significant, yet showed the highest loading among the four o bserved indicators, r = .87, z = 1.77, p = .07. These results provided support for the measurement model, which suggests a reasonable foundation for testing structural relationships.

PAGE 150

141 Table 11. Unstandardized, Standardized, and Critical Ratio Values for Juvenile Sexual Victimization Model Displayed in Figure 7 (Standard Errors in Parentheses; N = 174) Parameter Estimate Unstd. Coeff./Probit Std. Coeff. Critical Ratio ( Z ) Measurement Model Estimates Strain by Domestic Violence .84 (.31) .62 2.68 Strain by Alcohol Problem 1.19 (.45) .68 2.68 Strain by Drug Problem .63 (.27) .44 2.34 Strain by Mental Health Problem 2.21 (1.25) .87 1.77 Structural Model Estimates Sexual Victimization on Strain .67 (.20) .52 3.41 Running Away on Sex ual Victimization .69 (.13) .63 5.43 Initial Age of D/A Use on Sexual Victimization 1.21 (.22) .43 5.62 Prostituted Minor on Running Away .37 (.15) .40 2.50 Prostituted Minor on Age of D/A .10 (.05) .28 1.87 Indirect Effect Prostitute d Minor on Caregiver Strain .26 (.12) .19 2.19 Indirect Effect Prostituted Minor on Sexual Victimization .39 (.13) .38 2.95 Note: 2 (15) = 18.07, p =.26; NC = 1.20; CFI =.98; TLI =.98; RMSEA =.03; WRMR = .69 As shown in Table 11, t he structural equation model featuring juvenile sexual victimization evidenced adequate fit to the data with a non significant Chi square, 2 (15) =

PAGE 151

142 18.07, p = .26. The other model fit indices also demonstrated good model fit (NC = 1.20; CFI = .98; TLI = .98; RMSEA = .03; WRMR = .69). Next, the structural paths linking key latent and observed variables relevant to the juvenile sexual victimization model hypotheses were examined to assess the prop osed relationships (see Figure 7 ). The unstandardized coefficients, s tandardized coefficients, and standard errors for all model estimates are reported in Table 11. Caregiver strain was positively related to juvenile sexual victimization with a large direct effect size ( std. coefficient = .52, p < .01). The combined indir ect effect of caregiver strain on being prostituted as a minor as mediated by juvenile sexual victimization, running away, and initial age of drug or alcohol use revealed a significant and positive relationship with a small effect size ( std. coefficient = .19, p < .05). Juvenile sexual victimization was positively related to running away with a large effect size ( std. coefficient = .63, p < .01) and negatively related to age of initial drug or alcohol use with a moderate effect size ( std. coefficient = .42, p < .01). The combined indirect effect of juvenile sexual victimization on being prostituted as a minor as mediated by running away and age of initial drug or alcohol use revealed a significant, positive relationship with a moderate effect size ( std. coefficient = .38, p < .01). Running away had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .40, p < .05). Age of initial drug or alcohol use had a negative relationshi p with being prostituted as a minor ( std. coefficient = .29), with a critical value slightly lower than the predetermined significance level for the analysis ( z = 1.87, p = .06)

PAGE 152

143 .63** The standardized coefficients of each key path a nd the proportion of variance explained in the endogenous variables are noted in Figure 7 and were as follows: juvenile sexual victimization R 2 = .27; running away, R 2 = .40; initial age of drug or alcohol use R 2 = .18; and being prostituted as a minor, R 2 = .31. In the juvenile sexual victimization model the structural paths connecting the key variables supported the study Running Away ( R 2 =. 40) Age of Initial Drug/Al. Use ( R 2 = .18) CST/ Prostitution ( R 2 =. 31) 52** .43** .28 .40* Figure 7. Juvenile Sexual Victimization Structural Equation Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (15) = 18.07, p =.26; NC=1.20; CFI =.98; T LI =.98; RMSEA =.03; WRMR =.69 Caregiver Strain Sexual Victimization ( R 2 =. 27)

PAGE 153

144 h ypotheses, with only one of the proposed path coefficients narrowly failing to achieve the predetermined significance value for a two tailed hypothesis test. 11 The final step in the analysis of the juvenile sexual victimization model included the exploratory examination of the origins and effects of relational shame (see Analytic Strategy in Chapter 5). As planned, a supplementary j uvenile sexual victimization structural equation model was analyzed to examine: (1) the effects of juvenile sexual victimization on the level of reported relational shame; and (2) to explore the effects of relational shame on the likelihood of being prosti tuted as a minor, seeking to discern whether relational shame may intensify the effect of running away and the initial age of drug or alcohol use on being prostituted as a minor. The juvenile sexual victimization model including relational shame also evide nced adequate fit to the data with a non significant Chi square, 2 (19) = 24.66, p = .17. The other indices demonstrated good model fit (NC = 1.30; CFI = .97; TLI = .97; RMSEA = .04; WRMR = .73). The structural paths linking key latent and observed vari ables relevant to the juvenile sexual victimization model hypotheses were again examined to assess the proposed relationships. Juvenile sexual victimization was positively related to relational shame with a large effect size ( std. coefficient = .50, p < 01). The associations of juvenile sexual victimization with running away ( std. coefficient = .65, p < .01) and initial age of drug or alcohol use ( std. coefficient = .43, p < .01) were minimally affected by the inclusion of relational shame in the model. The combined indirect effect of caregiver 11 To ascertain that the combination of childhood sexual abuse and adolescent sexual victimization into one measure (see Measures in Chapter 5) was not severely affecting th e structural model or path coefficients, models were run with these indicators individually. Slight differences in model fit indicators, resulting in inferior model fit were noted in the adolescent sexual victimization only model. Very minimal changes in some path coefficients were found in the modified models.

PAGE 154

145 strain on being prostituted as a minor as mediated by juvenile sexual victimization, running away, initial age of drug/alcohol use, and relational shame was minimally increased in this supplementary model ( std. c oefficient = .23, p < .05). The combined indirect effect of juvenile sexual victimization on being prostituted as a minor as mediated by running away, age of initial drug or alcohol use, and relational shame was also minimally increased by the addition of relational shame to the model ( std. coefficient = .43, p < .01). Relational shame had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .36, p < .05). The effect size of the relationship of running away with being prostituted as a minor was reduced and non significant ( std. coefficient = .23). Also, the effect size of age of initial drug or alcohol use on the likelihood of being prostituted as a minor decreased ( std. coef ficient = .22). The proportion of variance explained of being prostituted as a minor in this supplementary model was R 2 = .35. Child Maltreatment Model As shown in Figure 8, th e measurement model for the la tent variables in the child maltreatment model evidenced an adequate fit to the data. The Chi square was non significant, 2 (24) = 31.89, p = .13. All other indices also demonstrated good fit (NC = 1.32; CFI = .98; TLI = .98; RMSEA = .04; WRMR = .77). The loadings of the indicators on the latent va riables were statistically significant and strong, ranging from .45 to .90. As displayed in Figure 7, MPlus also provided the correlation coefficient between the two latent variables. These results provided support for the measurement model, which suggest s a reasonable foundation for testing structural relationships.

PAGE 155

146 .66** As shown in Table 12, t he structural equation model featuring child maltreatment evidenced adequate fit to the data with a non significant Chi squ are, 2 (34) = 48.06, p = .06. All of the other indices demonstrated good model fit (NC = 1.41; CFI = .96; TLI = .97; RMSEA = .05; WRMR = .83). Caregiver Strain Child Maltreatment Alcohol Problem Drug Problem Domestic Violence Mental Health Problem .75** Poor Supervision Lack of Food .80** Lack of Medical .86* Poor Care Drug/Alcohol .90 .58** .64** .65** .69** .85** Figure 8 Child Maltreatment Measurement Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (24) = 31.89, p =.13; NC =1.32; CFI =.98; TLI =.98; RMSEA =.04; WRMR =.77 Lack of Love Juvenile Sexual Victimizati on Child Physical Abuse .45** .82**

PAGE 156

147 Table 12. Unstandardized, Standardized, and Critical Ratio Values for Child Maltreatment Model Displayed in Figure 9 (Standard Errors in Parentheses; N = 174) Parameter Estimate Unstd. Coeff./Probit Std. Coeff. Critical Ratio ( Z ) Measurement Model Estimates Strain by Domestic Violence .84 (.30) .6 5 2.75 Strain by Alcohol Problem 1.12 (.38) 69 2.95 Strain by Drug Problem .88 (.31) .5 8 2.69 Strain by Mental Health Problem 1.33 (.35) .7 5 3.85 Maltreatment by Poor Supervision .61 (.21) .64 2.94 Maltreatment by Lack of Food 1.52 (.50) .80 3.07 Maltreatment by Lack of Medical 1.96 (.81) .86 2.41 Maltreatment by Poor Care due to D/A 2.45 (.93) .90 2.64 Maltreatment by Lack of Love 1.88 (.60) .85 3.13 Maltreatment by Physical Abuse 1.02 (.37) .66 2.81 Maltreatment by Sexual Victimization .59 (.17) .45 3.54 Strain with Maltreatment .60 (.19) .82 3.16 Structural Model Estimates Maltreatment on Caregiver Strain .90 (.28) .77 3.13 Running Away on Maltreatment .93 (.24) .67 3.93 Initial Age of D/A Use on Maltreatment 1.55 (.41) .45 3.75 Prostituted Minor on Runni ng Away .37 (.17) .41 2.12 Prostituted Minor on Initial Age of D/A .10 (.06) .28 1.80 Indirect Effect Prostituted Minor on Caregiver Strain .44 (.19) .31 2.31 Indirect Effect Prostituted Minor on Child Maltreatment .49 (.09) .40 2.48 No te: 2 (34) = 48.06, p =.06; NC = 1.41; CFI =.96; TLI =.97; RMSEA =.05; WRMR =.83

PAGE 157

148 Next, the structural paths linking key latent and observed variables relevant to the child maltreatment model hypotheses were examined to assess th e proposed relationships (see Figure 9). As mentioned in the results of the previous models, standardized linear regression coefficients and probit coefficients are reported for the key paths in the model. The unstandardized coefficients, standardized co efficients, and standard errors for all model estimates are reported in Table 12. Running Away ( R 2 =. 45) Age of Initial Drug/Al. Use ( R 2 = .21) CST/ Prostitution ( R 2 =. 32) 77** .45** .67** .28 .41* Figure 9. Child Maltreatment Structural Equation Model Note: p < 0.05, ** p < 0.01, two tailed Note: 2 (34) = 48.06, p =.06; NC =1.41; CFI =.96; TLI =. 97; RMSEA =.05; WRMR = .83 Caregiver Strain Child Maltreatment ( R 2 =. 58)

PAGE 158

149 Caregiver strain was positively related to child maltreatment with a large direct effect size ( std. coefficient = .77, p < .01). The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by maltreatment, running away, and initial age of drug or alcohol use revealed a significant and positive relationship with a moderate effect size ( std. coefficient = .31, p < .05). Child maltreatment w as positively related to running away with a large effect size ( std. coefficient = .67, p < .01) and negatively related to age of initial drug or alcohol use with a moderate effect size ( std. coefficient = .45, p < .01). The combined indirect effect of c hild maltreatment on being prostituted as a minor as mediated by running away and age of initial drug or alcohol use revealed a significant, positive relationship with a moderate effect size ( std. coefficient = .40, p < .05). Running away had a signific ant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .41, p < .05). Age of initial drug or alcohol use had a negative relationship with being prostituted as a minor ( std. coefficient = .28 ), with a critical value slightly lower than the predetermined significance level for the analysis ( z = 1.80, p = .08) 12 12 As the effect of the initial age of drug or alcohol use on vulnerability to being prostituted as a minor had not reach the predetermined level for statistical significance in the models, furth er analyses were conducted in the child maltreatment model to determine if the combining of the two variables, age of initial drug use and age of initial alcohol use (see Measures in Chapter 5) had severely affected the structural model or the effect size of the path coefficient and the statistical significance. Structural equation models were assessed with the individual indicators, age of initial drug use and age of initial alcohol use, entered independently. Slight differences in model fit indicators, resulting in inferior model fit were noted in the model that included the initial age of drug use variable. No such changes in model fit were found in the structural model including the initial age of alcohol use indicator. The path coefficients from th e initial age of use of the two substances (examined individually rather than as a combined variable) and being prostituted as a minor were stronger and statistically significant in both models (initial age of alcohol use, std. coefficient = 37, z = 2.35, p < .05 ; initial age of drug use, std. coefficient = 32, z = 2.07, p < .05 ) compared to the model with the combined variable ( std. coefficient = .28, z = 1.80, p = .08).

PAGE 159

150 The standardized coefficients of each key path and the proportion of variance explained in the endogenous variables are noted in Figu re 8 and were as follows: child maltreatment, R 2 = .59; running away, R 2 = .45; initial age of drug or alcohol use R 2 = .21; and being prostituted as a minor, R 2 = .32. In the child maltreatment model the structural paths connecting the key latent and ob served variables supported the study hypotheses, with only one proposed path coefficient narrowly failing to achieve the predetermined significance value for a two tailed hypothesis test. The final step in the analysis of the child maltreatment model inclu ded the exploratory examination of the origins and effects of relational shame (see Analytic Strategy in Chapter 5). As planned, a supplementary child maltreatment structural equation model was analyzed to examine: (1) the effects of child maltreatment on the level of reported relational shame; and (2) to explore the effects of relational shame on the likelihood of being prostituted as a minor, seeking to discern whether relational shame may intensify the effect of running away and the initial age of drug or alcohol use on being prostituted as a minor. The child maltreatment model including relational shame also evidenced adequate fit to the data with a non significant Chi square, 2 (37) = 50.00, p = .08. The other indices demonstrated good model fit (NC = 1.35; CFI = .97; TLI = .97; RMSEA = .05; WRMR = .83). The structural paths linking key latent and observed variables relevant to the child maltreatment model hypotheses were again examined to assess the proposed relationships. Child maltreatment was po sitively related to relational shame with a large effect size ( std. coefficient = .50, p < .01). The associations of child maltreatment with running away ( std. coefficient = .68, p < .01) and initial age of drug or alcohol use ( std.

PAGE 160

151 coefficient = .46, p < .01) were minimally affected by the inclusion of relational shame in the model. The combined indirect effect of caregiver strain on being prostituted as a minor as mediated by maltreatment, running away, initial age of drug or alcohol use, and relationa l shame was minimally increased in this supplementary model ( std. coefficient = .33, p < .05). The combined indirect effect of child maltreatment on being prostituted as a minor as mediated by running away, age of initial drug or alcohol use, and relati onal shame was also minimally increased by the addition of relational shame to the model ( std. coefficient = .43, p < .05). Relational shame had a significant and positive direct relationship with being prostituted as a minor with a moderate effect size ( std. coefficient = .36, p < .05). Running away no longer had a significant relationship with being prostituted as a minor with a reduced effect size ( std. coefficient = .23). Also, the effect size of age of initial drug or alcohol use on the likelihood of being prostituted as a minor decreased ( std. coefficient = .22). The proportion of variance explained of being prostituted as a minor in this supplementary model was R 2 = .35. Effect of Relational Shame Further analyses were necessary to test the study hypothesis that relational shame may intensify the effect of behavioral responses to strain (running away and initial age of drug or alcohol use) on vulnerability to being prostituted as a minor. First, to examine the combined effects of running awa y and relational shame, the empirical probabilities of being prostituted as a minor based on the two vulnerability factors were calculated.

PAGE 161

152 To facilitate and simplify the examination of the observed probabilities, the relational shame variable was conde nsed and recoded from seven levels into three levels of shame (scores ranging from 1 levels of shame (scores ranging from 4 6) increase in the observed probability of being prostituted across both vulnerability factors (see Table 13). Also, those participants who ran away and reported high levels of relational shame had the highest probabilit y of being prostituted as a m inor, with a .41 probability. In comparison, the probability of being prostituted as a minor for those participants who did not run away and reported no relation al shame was .01. The observed probabilities of being prostitute d as a minor based on these two factors are also displayed in Figure 10. Table 13. Observed Probabilities of Being Prostituted as a Minor Based on Running Away and Level of Relational Shame (N = 172) Level of Relational Shame Total None Low High Ran Away No .01 .21 .00 .07 Yes .15 .12 .41 .20 Total .05 .17 .32 .12

PAGE 162

153 F igure 10 Observed Probabilities of Being Prostituted as a Minor Based on Running Away and Level of Relational Shame (N = 172) In order to detect an interac tion or moderating effect, logistic regression with the dependent variable, being prostituted as a minor regressed on running away, relational shame, and an interaction term of those two independent variables was conducted. One of the key assumptions of r egression analysis is the absence of perfect or high multicollinearity. High multicollinearity does not bias the estimates of the coefficients, but can lead to an increase in Type II error due to an inflation of the standard errors of the regression coeff icients. 1.0 0.5 0.0 Probability of Being Prostituted as a Minor None Lo w High Level of Relational Shame Ran Away No Yes

PAGE 163

154 Including an interaction term in a general linear model can drastically increase the level of multicollinearity. The product term (running away by relational shame) is a function of the observed variables (running away and relational shame), thus correlations of the observed variables with the product term are usually high. To reduce collinearity the continuous predictor variable included in the model was centered prior to computing the interaction term ( Aiken & West, 1991). Table 14. Colline arity Statistics of the Regression of Being Prostituted as a Minor on Running Away, Relational Shame, and the Interaction Term Collinearity Statistics Tolerance VIF Running Away .99 1.13 Relational Shame .39 2.54 Interaction Term of Running Away by Relational Shame .41 2.44 Table 14 reports the collinearity statistics of tolerance and the Variance Inflation Factor (VIF). The cut off value for VIF is 2.5, and any value above 2.5 may indicate high multicollinearity. Two of the VIF values were ne ar or higher than 2.5. The VIF values indicate the degree that covariates in the model are predicted by the other variables included in the model. The cut off value for tolerance is .40, and any value below that indicates high multicollinearity. Two of the tolerance values were near or below .40, which indicates that a high proportion of variability in each predictor is explained by its linear relationship with the other predictors. These results indicate high multicollinearity

PAGE 164

155 among the predictor varia bles as would be expected due to the inclusion of the interaction term. Violations of other key assumptions of regression analysis were not noted in the results of the regression diagnostics. The Normal Probability Plot of the regression of the standardiz ed residuals approximately charted the residuals in a diagonal line from bottom left to top right, indicating no major deviations from normality. The scatterplot of the standardized residuals found that the residuals were roughly distributed in a rectangu lar shape with no observed pattern, indicating no violations of the assumptions (Tabachnick & Fidell, 2001). Only one case had a studentized residual over 3.00 and no ha ving undue influence over the results (Tabachnick & Fidell, 2001). Table 15. Logistic Regression of Being Prostituted as a Minor Regressed on Running Away, Relational Shame, and the Interaction Term p < 0.05, ** p < 0.01, two tailed. Note: 2LL = 112.69; 2 (3), = 14.96** Variable B SE B e B Constant 2.63 .42 .07 Running Away .94 .57 2.57 Relational Shame 1.22* .57 3.37 Interaction term of Running Away by Relational Shame .48 .69 .62

PAGE 165

156 Table 15 reports that the data fit the model well, 2LL = 112.69, 2 (3), = 14.96, p < .01. The Nagelkerke R 2 is .16, indicating that the covariates account for a small amount of the variance in the dependent variable. However, the interaction term was not found to be statistically significant, so no mo deration effect is evident based on this logistic regression analysis. In summary, the estimated probabilities of being prostituted as a minor based on these two vulnerability factors seemed to indicate that high levels of relational shame intensified the effect of running away on vulnerability to being prostituted as a minor in comparison to no shame. However, the results of the logistic regression including the interaction term provided no evidence of a moderating effect of relational shame on the link b etween running away and being prostituted as a minor. Although the statistical significance of the path coefficient of initial age of drug and alcohol use and being prostituted as a minor was found to be slightly lower than the predetermined level for sta tistical significance in this study, the estimated effect sizes of the path coefficients ( std. coefficients ranging from 22 to 29) were of sufficient magnitude to justify the examination of these variables to determine if there was a moderating effect of relational shame amplifying the likelihood of being prostituted for girls who began to use substances at an earlier age. As with the previous regression model, regression diagnostics were examined to identify violations of the key assumptions of regre ssion analysis. Table 16 reports the collinearity statistics of tolerance and the Variance Inflation Factor (VIF). The low VIF values indicate that covariates in the model were not predicted by the other variables included in the model. The tolerance val ues also indicate no problem with collinearity.

PAGE 166

157 These results indicate that there is no problem with multicollinearity among the predictor variables included in this model. Table 16. Collinearity Statistics of the Regression of Being Prostituted as a Min or on Age of Initial Drug or Alcohol Use, Relational Shame, and the Interaction Term Collinearity Statistics Tolerance VIF Initial Age of Drug/Alcohol Use .94 1.07 Relational Shame .86 1.16 Interaction Term of Running Away by Relational Shame .92 1.09 Further regression diagnostics were examined. The Normal Probability Plot of the regression of the standardized residuals approximately charted the residuals in a diagonal line from bottom left to top right, indicating no major deviations from nor mality. The scatterplot of the standardized residuals found that the residuals were roughly distributed in a rectangular shape with no observed pattern, indicating no violations of the assumptions (Tabachnick & Fidell, 2001). Three cases had studentized residuals no extreme cases were having undue influence over the results (Tabachnick & Fidell, 2001). As with the prior logistic regression, to reduce collinearity t he continuous predictor variables included in the model were centered prior to computing the

PAGE 167

158 interaction term (Aiken & West, 1991). Table 17 reports that the data fit the model well, 2LL = 97.86, 2 (3), = 17.73, p < .001. The Nagelkerke R 2 is .21, ind icating that the covariates account for a small amount of the variance in the dependent variable. However, the interaction term was not found to be statistically significant, so no moderation effect is evident based on this logistic regression analysis. 13 Table 17. Logistic Regression of Being Prostituted as a Minor Regressed on Age of Initial Drug or Alcohol Use, Relational Shame, and the Interaction Term p < 0.05, ** p < 0.01, two tailed. Note: 2LL = 97.86; 2 (3), = 17.73** 13 L ogistic regression was also conducted with the variable measuring the initial a ge of alcohol use to explore if the inclusion of this variable rather than the combined variable (see Measures in Chapter 5) would alter the results, but the findings were similar to the results reported in the regression including the combined variable an d were also non significant. Variable B SE B e B Constant 2.28 .30 .10 Initial Age of Drug/Alcohol Use .15 .09 .86 Relatio nal Shame .35* .15 1.42 Interaction term of Initial Age of Drug/Alcohol Use by Relational Shame .01 .04 .99

PAGE 168

159 Chapter 7 Discussion and Conclusion The pu rpose of this study is to develop and subsequently assess a theoretically based explanation of the processes elevating vulnerability to victimization in child sex trafficking, specifically focusing on prostituted girls. This particular population of sex trafficking victims was chosen because girls face higher risk for victimization in sex trafficking than boys (Clawson, 2009) and prostituted minors are the most marginalized and least studied group of child sex trafficking victims (Williams & Frederick, 2009). Incorporating the previously documented risk markers of such crime victims and circumspectly drawing from propositions of sound criminological theory, this study projects a pathway into victimization in child sex trafficking in prostitution. The suitability of the pro jected pathway was subsequently evaluated utilizing structural equation modeling. This chapter includes a thorough discussion of the results of the analyses detailed in Chapter 6. Within this chapter, the implications stemming from the findings such as r ecommendations for new policies, prevention strategies, and the application of effective treatment interventions are considered. Limitations of the current study are also described, several contributions of this study to scholarly literature are highlight ed, and suggestions for further research are offered. The current study effectively postulated and empirically identified a pathway into victimization in child sex trafficking in prostitution. Specifically, the study tested five theoretically based and i nterlocking hypotheses that framed the proposed strain reactive

PAGE 169

160 pathway into victimization in child sex trafficking in prostitution. First, caregiver strain (e.g., domestic violence, mental health problems, substance abuse problems) was predicted to produ ce a detrimental familial context increasing the likelihood of the occurrence of child maltreatment, including child neglect, physical abuse, and sexual abuse. Second, in response to the strain of child maltreatment, it was hypothesized that victimized gi rls are more likely to choose risk inflating, even delinquent, pathways of escape. The theorized pathway predicted that maltreated girls would be more likely to run away in comparison to non abused girls or they would begin to use drugs or alcohol at an e arli er age. Third, self denigratory negative emotion was theorized to result from from previous childhood victimization experiences, was theorized to detrimentally inf late vulnerability to victimization in sex trafficking of escaping girls, i.e., girls who had run away or who had begun to use drugs or alcohol at an early age to escape the pain of victimization. And fifth, these risk inflating responses or escape routes from strain expected to increase their vulnerability to revictimization in child sex trafficking. General Overview of the Findings The systematic analyses of the project ed pathway, including the identification of four structural equation models, substantiated the suitability of the theorized pathway for explaining vulnerability to child sex trafficking in prostitution and partially supported the key predictions of this st udy. In the four models assessed in this study the proportion of variance explained in vulnerability to being prostituted as a minor ranged from R 2 = .24 to R 2 = .35. Recent research examining the etiology of adult prostitution that incorporated

PAGE 170

161 compara ble individual risk factors, such as experiences of child maltreatment and running away, noted similar proportions of variance explained for adult prostitution in the study that also utilized structural equation modeling (Wilson & Widom, 2010). Beginning with the effects of caregiver strain on child maltreatment, this study caregiver, such as experiencing domestic violence or having a substance abuse or mental health pro blem, impacts the likelihood of the occurrence of child maltreatment, including child neglect, physical abuse, and juvenile sexual victimization. The study results indicated that the girls with strained mothers were more likely to have been neglected and abused than the girls whose mothers experienced less strain. In each model assessed by the study, there was a strong and positive relationship between caregiver strain and the form of child maltreatment included in the model, whether child neglect, child physical abuse, or juvenile sexual victimization (with std. coefficients ranging from .54 to .76). This association of caregiver strain with all types of child maltreatment supports previous studies on the detrimental effects of strain on caregivers that h ave reported that caregivers are often abusive or neglectful due to mental illness, drug or alcohol abuse, domestic violence, or other types of strain or adversity (Barnett et al., 2005; DiLauro, 2004; Ford et al., 2006; Herrenkohl et al., 2008; Turner et al., 2007). Differences found in the models seem to indicate that the occurrences of child neglect were more fully accounted for by caregiver strain than were the incidences of the other two types of child maltreatment. The proportion of the variance ex plained in child neglect by caregiver st r ain ( R 2 = .58) was approximately double the proportion of variance explained in the two other types of child maltreatment, i.e., R 2 = .27 in child

PAGE 171

162 physical abuse and R 2 = .31 in juvenile sexual victimization. This variation in the findings could be due to differences in the types of variables included in the model. The child neglect measure was a latent variable identified by five observed indicators. Latent variables are less influenced by measurement error than observed variables (Kline, 2005). The other two measures of maltreatment were observed variables; therefore, the amount of variance explained for these types of maltreatment may have been attenuated due to reduced reliability of those measures (Kline, 200 5). Thus, this disparity in the amount of variance explained by caregiver strain could be a result of differences in the quality of the variables measuring maltreatment. Another explanation of the disparity in the results could be that child physical abu se and juvenile sexual victimization are more likely to result from a greater number of generative factors, beyond caregiver strain, in comparison to child neglect. Such a finding would be in agreement with previous research regarding the sources of child maltreatment. Child neglect has often been found to be a consequence of a caregiver or family being strained and overwhelmed, resulting in lack of care and supervision (Agnew et al., 2000; Benedict & Zautra, 1993; Hildyard & Wolfe, 2002). Although child physical abuse has often been associated with similar risk factors (Walsh et al., 2002), additional risk factors have also been noted such as a caregiver with an impulsive or explosive temperament, difficult temperament of the child, and cultural approval of harsh and abusive parenting (Barnett et al., 2005). Although caregiver strain and family dysfunction have been associated with sexual abuse (Finkelhor, 1994; Madonna et al., 1991; Walsh et al., 2002; Terry & Tallon, 2004), additional factors have been found to increase risk for sexual abuse or assault. Sexual offenders generally select children based on vulnerability, i.e., children who are needy, shy, from divorced or single parent

PAGE 172

163 homes, trusting, or young (Barnett et al., 2005; Terry & Tallon, 200 4). As evidenced by this listing of vulnerability factors, caregiver strain may increase risk for sexual abuse but some risk factors are unrelated to caregiver strain. Also, caregivers almost always perpetrate child neglect; however, sex offenders may b e encountered in a wide range of (Shakeshaft, 2004; Snyder 2000; Terry & Tallon, 2004). Perhaps the disparity in the proportion of variance explained by caregiver strain is si gnifying that although caregiver strain is a significant risk factor for all types of maltreatment, child physical abuse and juvenile sexual victimization may result from a wider and more divergent set of sources while child neglect may be more readily pre dicted by caregiver strain. Caregiver strain was also shown to have a modest, indirect effect on vulnerability to being prostituted as a minor mediated by maltreatment, running away, and age of initial drug or alcohol use in all models (with std. coeffici ents ranging from .19 to .31). This finding of a moderate indirect relationship between caregiver strain and being prostituted as a minor provides support for the prior research that has noted family dysfunction as a risk marker inflating vulnerability to victimization in child sex trafficking in prostitution (Clawson, 2009; Estes & Weiner, 2005). In addition, this finding regarding the indirect effects of caregiver strain on later victimization provides further validation of a transgenerational effect of strain, with strain being partially transferred from mother to involvement in criminal events would seem to be mediated by the experience of child maltreatment and its linge ring, long term liability (Loeber et al., 2009; Macmillan, 2001; Agnew et al., 2000).

PAGE 173

164 Confirming the study hypotheses, the models substantiated the proposition that maltreated girls are more likely to run away and initiate drug or alcohol use at an ear lier age than girls who are not abused. The impact of all forms of child maltreatment on the likelihood of these two behavioral responses or escape mechanisms supports the study hypothesis that abused girls may attempt to escape strain by initiating vari ous behavioral, even delinquent, responses (Agnew, 2001, 2006a; Agnew et al., 2000; Steffensmeier & Allan, 1996). This finding regarding the effect of maltreatment on running away validates previous research showing that minors commonly run away from abus ive environments (for review, see Baron, 2003). Further, the finding that maltreated children initiated drug or alcohol use at an earlier age supports prior studies regarding the use of substances by maltreated children to cope with abuse (for review, see Bender, in press). Concerning the prediction that risk inflating choices made by abused minors increases their vulnerability to being prostituted, one of the predicted behavioral responses to strain, running away, was found to increase the likelihood of entrapment in prostitution as a minor in three of the four models, with statistically significant path coefficients. In the child neglect model, the juvenile sexual victimization model, and in the model that included a composite measure of all three forms of child maltreatment, running away directly impacted the likelihood of being prostituted as a minor. This finding supports previous reports asserting that runaways are at heightened risk for commercial sexual exploitation (Clawson, 2009; Este s & Weiner 2005). The association noted between child maltreatment and being prostituted as a minor, facilitated by running away, supports prior research that has shown running away to be both a result of maltreatment and a predictor of future involvement in crimi nal environments, including

PAGE 174

165 child sex trafficking (Bender, in press; Chesney Lind, 2001; Estes & Weiner, 2001; Sheridan & VanPelt, 2005). The findings regarding the effects of the other behavioral response to strain included in the projected pathway, an earlier initiation of drug or alcohol use, were equivocal. The inverse relationship between initial age of drug or alcohol use and the likelihood of victimization in prostitution was present as predicted. However the pathway coefficient between initial a ge of drug or alcohol use and being prostituted as a minor did not reach statistical significance at the level that had been predetermined for the study in any of the models. So although the models supported the study hypothesis and previous research by f inding that abused minors were more likely to begin to use substances at an earlier age than those who were not maltreated (Bender, in press; Harrison et al., 1997; Morash, 2006), a statistically significant direct effect of initial age of drug or alcohol use on vulnerability to victimization in child sex trafficking in prostitution was not found. Strictly adhering to the statistical results that failed to reject the null hypothesis at an alpha level of .05, these findings would indicate that the age of in itial drug or alcohol use does not generate vulnerability to being prostituted as a minor. On the other hand the small to medium effect sizes of the associations found between the two variables in the four primary models ( std. coefficients ranging from 22 to 29) were narrowly non us very little (if anything) about the practical significance or relative impact of the effect & Cooper, 2003, p. 1 emphasis in the original). As the path coefficients were statistically significant when the individual variables reporting the initial age of alcohol use and the initial age of drug use

PAGE 175

166 were entered into the model (i.e., initial age of alcohol use, std. coefficient = 37, z = 2.35, p < .05 ; initial age of drug use, std. coefficient = 32, z = 2.07, p < .05), it would seem reasonable to conclude from this collection of findings that an earlier initiation into drug or alcohol use increa ses vulnerability to being prostituted as a minor. Within the models that contained a single type of child maltreatment, the model including juvenile sexual victimization explained the greatest amount of variance in being prostituted as a minor ( R 2 = .31 v s. R 2 = .24 for the child physical abuse model and R 2 = .27 for the child neglect model). This finding confirming the detrimental impact of juvenile sexual victimization validates a large body of research that has demonstrated that prior sexual victimizat ion is a key predictor of repeat sexual victimization (for review, see Classen et al., 2005) and supports research regarding the substantial damage that being sexually abused as a minor eventually yields in the lives and futures of those consequently explo ited in prostitution (e.g., Silbert & Pines, 1982; Simons & Whitbeck, 1991 ). However juvenile sexual victimization was not the only form of child maltreatment shown to have an effect on the likel ihood of being prostituted, rather all of the forms of chil d maltreatment included in the models showed a significant and moderate indirect effect (with std. coefficients ranging from .33 to .38) on the likelihood of girls being prostituted, mediated by running away and age of initial drug or alcohol use. This ge neralized effect of all the forms of child maltreatment, both when entered into the models singularly or when entered as observed indicators to identify a single latent child maltreatment construct, indicates that all forms of child maltreatment harmfully impact a child, not only sexual victimization. This finding concurs with the growing body of

PAGE 176

167 research supporting the notion that there is a universally harmful effect of all types of child maltreatment (Finkelhor, Ormrod, & Turner, 2007; Van Bruggen et al ., 2006; Wilson & Widom, 2010). The detection of a persisting effect of child maltreatment on involvement in further abusive and exploitive relationships may be indicative of conceptualizations previously proffered by theorists and researchers that maltr eated children are likely to endure a series of abusive relationships as they mature, due to the formation in childhood of a dysfunctional template of relationship functioning (Agnew, 2006a; Bowlby, 1973, 1980; Finkelhor & Browne, 1985; Fraley, 2002). The results of the supplementary models incorporating the relational shame variable supported the hypothesis that self denigratory negative emotion, in the form of relational shame, was more likely to be experienced by girls who had been maltreated than by gir ls who had not been maltreated. This association of child maltreatment with relational shame validates the proposition by Agnew and Broidy (1997) that girls may respond to strain with se lf denigratory emotion. The finding that relational shame results fr om all forms of child maltreatment offers further validation to the assertion that stigmatization or shame, once theorized to be a unique result of sexual abuse (Finkelhor & Browne, 1985), may instead be a consequence of all forms of child maltreatment (St uewig, & McCloskey, 2005; Van Bruggen et al., 2006 ). Relational shame was also found to have a moderate impact on likelihood of being prostituted as a minor, validating previous research regarding the endangering effects of shame on the likelihood of sex ual revictimization (Arata, 2000, 2002; Classen, et al., 2005; Filipas & Ullman, 2006; Messman et al., 2009; Stockdale et al., 2002; Van Bruggen et al., 2006; Whiffen & MacIntosh, 2005).

PAGE 177

168 In further analysis, the predicted interaction between the age of initial drug or alcohol use and relational shame on vulnerability to being prostituted while a minor was not substantiated using logistic regression (see Table 17). Also, the predicted interaction between running away and relational shame on vulnerabilit y to being prostituted while a minor was not validated (see Table 15). However, the quantity of prostituted minors within the sample along with the lack of variation and collinearity between variables (i.e., running away, relational shame, and being pros tituted as a minor) may have distorted the findings (i.e., all of the girls who had high levels of shame or scored 4 6 on the scale of relational shame and were prostituted had also run away). In supplementary analysis, the empirical probability of being prostituted as a minor for girls who ran away and had higher levels of shame (.41) was found to be much greater than the probability of being prostituted for girls who did not run away and reported no relational shame (.01). Although thi s finding of a gr eater observed probability of being prostituted for girls who ran away and reported higher relational shame was simply an exploratory examination of this single sample, it provided introductory support for the possibility of an aversive cumulative effect o f this combination of risk factors, rather than a moderating or interaction effect. Thus, relational shame may provide partial explanation of prior findings, which have noted that previously abused runaways are at greater risk for sexual exploitation in p rostitution than those without an abuse history (Saewyc & Edinburgh, 2010; Saewyc, et al., 2008; Scott et al., 2003; Tyler & Johnson, 2006a) Conclusive evidence for a moderating effect of relational shame, which amplifies the effect of running away on th e likelihood of being

PAGE 178

169 prostituted as minor, was not revealed by this study; however, further research into this hypothesis would seem to be warranted. Policy Implications of the Study The implications of this study are relevant to policies related to chi ld victims of sex trafficking, including prevention and intervention strategies. Along the analytically identified pathway into child sex trafficking in prostitution, there seems to exist critical junctures primed for either protective intervention or ele vated endangerment. Rather than simply surveying the perilous journey of vulnerable minors into entrapment into child sex trafficking in prostitution knowing the damaging consequences that await them (for review, see Table 2), the ultimate purpose of this study was to provide needed information for the implementation of empirically informed strategies for intervening and obstructing the forward movement of minors along this strain reactive pathway. The identification of a pathway into child sex traffickin g traversed by certain girls offers essential details needed for the creation and application of effective policies and practices. The paragraphs that follow offer several specific suggestions for improved intervention at various critical junctures along the pathway into child sex trafficking in prostitution identified by this study. Based on the findings indicating that caregiver strain increases the vulnerability of girls to victimization, initiating with child maltreatment and terminating in exploitati on in prostitution, child sex trafficking prevention strategies would be most effective if targeted at severely strained families and their children. Law enforcement officers, child protection investigators, and child protection service providers are mos t likely to be the primary responders to crises within such families; therefore, they would most likely be

PAGE 179

170 responsible for providing the first phase of intervention and protection from an escalation in victimization for maltreated minors. Skillful and emp irically based intervention by law enforcement and child protection professionals could impede the progress toward victimization in child sex trafficking for vulnerable girls living in strained families by protecting such girls from maltreatment and provid ing support to their caregivers. The availability of sufficient resources and training is vital to ensure that the first responders to this most vulnerable population are adequately equipped for the task they have been given (Vieth, 2005). The caseloads of child protection investigators often exceed recommendations; in New York, for example, the average number of investigations per child protection investigator was reported as 20 per month (Vieth, 2005; Yackoub & Moss, 2007). Investigations of child mal treatment generally involve interviews with children, parents, school staff, and medical personnel; the inspection the site of the abuse for evidence; and the collection of information from various sources or databases (Vieth, 2005). Perhaps this heavy ca seload among child protection investigators explains why the majority of reported cases of abuse are not substantiated (HHS, 2010). The quantity and difficulty of the responsibility placed on child protection investigators may also help to explain the fi nding that child victims of unsubstantiated abuse were just as likely to be genuine victims of abuse as those whose cases were substantiated, as well as the high rate of recidivism within 12 months found among both substantiated and unsubstantiated cases, 22% and 13% respectively (Fuller & Neito, 2009; Vieth, 2005). The arduous caseload of investigators may not permit thorough investigations, nor allow time for the provision of needed resources to strained families and children in an effort to prevent furt her occurrences of abuse or the intensification of victimization. The escalation of child victimization identified in the strain reactive

PAGE 180

171 pathway could be prevented by the timely provision of needed services to adversely affected caregivers such as paren ting classes, substance abuse treatment, mental health therapy or medication, or assistance to victims of domestic violence. Beyond a reduction in caseloads, further recommendations for improving the responses of child protection investigators have been de tailed, such as providing better training and including a course specific to child protection investigations into the curriculum of educational programs preparing future child protection professionals (Vieth, 2005). Undergraduate and graduate social work programs seldom prepare future child protection investigators for the realities and obstacles they will encounter (Parent, for the number of families in trouble. So me of the cases would require a battalion of cops, doctors, and social workers to handle; instead there are two kids fresh out of Quindlen in Parent, 1996, p. vii). Speciali zed courses designed to provide training in child protection investigations have been skillfully developed and are currently taught in numerous universities in the United States (Vieth, 2005); widespread adoption of these specialized courses into education al programs that prepare future child protection investigators would be a relatively simple step that could be taken toward providing more effective intervention for troubled or strained caregivers and maltreated children, thereby increasing protection for abused minors and decreasing the likelihood that they will become future victims of child sex trafficking. Similar to the recommendations for specialized education and training intended to prepare social workers for child protection investigations, crimin al justice education

PAGE 181

172 and training programs in child protection are necessary as criminal justice professionals are often the first responders in child abuse cases (Vieth, 2005). For example in Florida, child protection investigation units for each of the counties are operated under the investigating allegations of child abuse against a care provider, such as a parent or teacher. When a child abuse report is received, a law enforcement officer will respond along with a child protection investigator. If a child protection investigator is not available (i.e., child protection investigators are only scheduled for work on Monday to Friday from 9 a.m. to 5 p.m. in Florida), a law enforcement officer will respond to the call alone and a child protection investigator will follow up on the abuse report during the next business day. Child protection investigators working in partnership with law enforcement officers are tasked with de termining whether the allegations of abuse against a care provider can be substantiated and whether the child should be removed from the residence. Alternatively, law enforcement officers in Florida, not child protection investigators, are fully responsi ble for investigating all child abuse allegations involving any non care provider, such as abuse allegations against neighbors, relatives other than caregivers, or acquaintances. Not only are law enforcement officers the first responders to abused childre n or adolescents; they also are often the first responsible authority to interact with runaways (SHI, 2008). As this study has found that runaways face heightened vulnerability to entrapment in child sex trafficking in prostitution, training for law enfor cement regarding the potential dangers faced by runaways and how best to intervene and protect such minors is essential. As the first and primary representative of the criminal justice system, if police officers are unaware of the vulnerabilities and dang ers faced by runaways or if the officers are uninformed of effective intervention

PAGE 182

173 strategies and safe shelter options it is possible that runaway minors will be exploited while on the streets. In light of the findings of this study that maltreated minors within strained families are at heightened vulnerability for involvement in child sex trafficking in prostitution and recognizing the extensive responsibilities of law enforcement officers in child protection investigations and their regular interactions w ith vulnerable runaways, training for future criminal justice professionals in child protection investigation seems to be warranted (Vieth, 2005). As shown in this study, maltreated minors are more likely to run away than minors without an abuse history. As noted earlier, the study findings also indicate that runaways are more vulnerable to being prostituted. Therefore based on these two findings, it would follow that maltreated minors may benefit from information regarding safe options in the event the y need to escape further maltreatment. Personally tailored safety plans are considered to be part of best practice protocols, and are crafted with previously abused women involved in domestic violence to promote safety for them and their children during f Services, 2003). This protocol encouraging preparation for a crisis through the use of a personally tailored safety plan, including a precise plan of who to call in an emergency and where to go to find safety, could reduce the likelihood that maltreated minors would run away and thereby become vulnerable to victimization in child sex trafficking in prostitution. Implementation of mandatory personal safety planning concerning what to do in a crisis should be included in the case planning provided for every abused child who is old enough to collaborate in such preparation.

PAGE 183

174 This study has validated previous reports indicating that vulnerability to exploitation and victimization is elevated for minors who run away (Clawson, 2009; Estes & Wiener, 2005). Runaway minors have few options for obtaining shelter or other basic necessities, which opens them up to exploitation and involvement in criminal activities ( Clawson, 2009; Sheridan & VanPelt 2005; Williamson & Cluse Tolar, 2002; Vieth & Ragland, 2005; Wilson & Widom, 2010) Providing safe shelter for runaways seems to strain reactive pathway that could prevent runaways from being further victimized and exploited. When caregivers are no longer able to provide sufficient care or protection, community must become an adequate guardian if a plan for the prevention of child sex trafficking of such minors is to succeed. Resourcefully, the National Safe Place initiative is an example of a youth intervention program that could be replicated in communities across the United States that would most likely lead to a reduction in the numbers of minors who run away only to be victimized in sex trafficking. This initiative meaningfully widens the doors of a ely to be able to access help whenever and wherever they need it. The Safe Place initiative provides a network of locations that are clearly marked as safe havens for youth in crisis (Safe Place, n.d.). Public facilities, such as fire stations, schools, or libraries, as well as appropriately screened locations such as fast food restaurants or convenience stores, display a distinctive yellow and black Safe Place sign letting a runaway know that safe shelter and services are available for them. This initia tive creates a community safety net of highly visible and safe options for vulnerable youth. In some communities, the

PAGE 184

175 public transit system has become a part of this safety net for youth and has taken on the role of providing mobile safe havens. The Safe Place network connects minors to immediate help and safety, as well as offering supportive social services to both the minors and their families. A number of findings from the current study are relevant to professionals who provide social services such a s mental health or substance abuse counseling to childhood maltreatment into child sex trafficking in prostitution highlights the need for effective mental health treatm ent for neglected and abused minors. For example, maltreated children were found to begin substance use at an earlier age thereby potentially increasing their vulnerability to further victimization. Such findings highlight the importance of substance and addictions counseling for abused and neglected minors. The results of this study reiterated the warning previously sounded by many child victimization researchers that being sexually abused increases vulnerability to future sexual victimization (for revi ew, see Classen et al., 2005) In this study being neglected and physically abused were also found to affect vulnerability to sexual exploitation, so the provision of counseling services should not be limited to minors who have been sexually abused. Spe cifically, family counseling for maltreated minors should be considered essential due to the findings of this study indicating that caregiver strain increases vulnerability to further endangerment and victimization. Therapeutic models, such as Multisystem ic, Functional Family, and Trauma Focused Cognitive Behavioral Therapy, involving both individual and family counseling, should be selected when

PAGE 185

176 planning treatment for such minors (Dulmus & Hilarski, 2002; Heyman & Slep, 2007; Saunders, Berliner, & Hanson, 2003). In light of the findings from the exploratory examination of the effects of relational shame, treatment for minors who have been sexually victimized, physically abused, or neglected would likely be more effective if they included strategies cent ered on reprocessing cognitions of not being worthy of healthy relationships, altering ideas of being undeserving of love, and shaping new beliefs about possessing intrinsic value and having equal importance relative to others. Maltreated minors also woul d benefit from information on how the psychological effects of childhood neglect or victimization (e.g., feelings of worthlessness in relation to others or a recurrent resignation to further abuse) may place them at an elevated risk for revictimization or involvement in exploitive relationships. To facilitate such treatment, the development of training materials and treatment strategies for therapists and child advocates are needed to assist such professionals in identifying and resolving these vulnerabili ties in abused or neglected minors. Currently, child sex trafficking in prostitution is often categorized as a public nuisance crime by state and local law enforcement. Child sex trafficking victims who have been criminally exploited continue to be pros ecuted by laws designed to protect them (Albanese, 2007; Curtis et al., 2008; FACJJ, 2007; Florida Department of Juvenile Justice [FDJJ]; 2006; Gray, 2005; Halter, in press; SHI, 2008). Arresting and adjudicating a minor victim of sex trafficking as a juv enile delinquent severely limits the availability and quality of victim services that could be provided to the minor as a crime victim by social service providers (SHI, 2008). By empirically depicting the common

PAGE 186

177 vulnerabilities of child sex trafficking vi ctims due to undeserved life experiences, the findings of this study may prove useful in curtailing the typical attribution of blame onto child sex trafficking victims for crimes committed against them and instead encourage the institutions responsible for protecting these vulnerable minors to embrace a more victim centered orientation (Halter, in press; Kreston, 2005; Williams & Frederick, 2009). In summary, the primary implications of the findings of this study have concentrated on enhancing the preparati on, development, and provision of resources for child protection investigators, law enforcement officers, and child protection social service providers. Protection and delivery of social services for strained families and maltreated children are presently coordinated by these three groups of professionals working in tandem. The victimization of vulnerable minors in child sex trafficking will likely continue unabated unless the progression of minors along the strain reactive pathway is obstructed. Accompl ishing this task of protecting minors from victimization in child sex trafficking in prostitution can be advanced by providing better preparation and adequate support for those who are already battling in the trenches nearest the front (Vieth, 2005). Limit ations of the Study As child sex trafficking victims are among the most hidden and inaccessible of victims, this study has provided a greater understanding of the origins and processes creating vulnerability to child sex trafficking in prostitution by anal yzing previously collected and available data. Although the availability of these data was beneficial, it also produced several limitations. The data were collected from African American females living in the United State s who took part in a longitudinal study on the effects of

PAGE 187

178 child sexual abuse. Accordingly, the findings of this study can only be applied to girls who are victims of domestic sex trafficking in the United States, to the exclusion of international, adult, or boy victims. The lack of mino rs of both genders in the sample limits the application of the findings to girls, preventing a comparison of genders and precluding an assessment of whether the proposed pathway is gendered or if certain boys may also follow a similar path into child sex t rafficking. Also, the sample lacked cultural diversity, limiting its generalizability to other races and ethnicities. However, minority girls living in low income families and in an urban environment possessed several individual and community risk marke rs for entrapment in child sex trafficking in prostitution (Clawson, 2009; Estes & Wiener, 2005), so the inclusion of such participants in the study was also considered a key advantage. By using a high risk sample, the study was to be able to detect the m echanisms that were operating to put certain high risk youth in greater danger to victimization in sex trafficking. However, as the result of the use of this high risk sample, descriptive statistics describing the characteristics of the sample were not re presentative of U.S. minors. Despite the careful selection and assessment of the indicators for the latent variables and the observed indicators, it is not possible to guarantee that these measures fully captured the variables of interest (Shadish et al., 2002). For example, the primary variable of interest in this study, child sex trafficking in prostitution, was not the focal point of the prior studies for which these data were originally collected. As a result, information was not collected concerning whether a trafficker or a need to survive drew the victim into prostitution. As was noted earlier, both situations are considered to

PAGE 188

179 be the commercial sexual exploitation of a child and the procurer of sex with a minor would be legally considered a sex trafficker, if no third party trafficker were involved (Adelson, 2008; DOJ, n.d.). In addition, prior studies have found that the majority of girls involved in prostitution in the United States were under the control of traffickers (Albanese, 2007; Estes & Weiner, 2005). Most importantly, the effects of both child sex trafficking involving a third party trafficker and the forms of child commercial sexual exploitation frequently labeled survival sex or trading sex have been found to be highly detrimental ( Albanese, 2007; DOS, 2008, 2009; Estes & Weiner, 2005; FACJJ, 2007; Inciardi, 1993). Therefore, identifying a pathway into such victimization was considered to be beneficial, whether or not a trafficker or third party profited in each and every case. Anot her variable of interest in the current study, the measure selected to detect relational shame, may not have exclusively or entirely assessed that negative emotion. The measure was originally created based on an index of items projected to reflect stigmat ization and traumatic sexualization resulting from sexual victimization (Finkelhor & Brown, 1985; Siegel & Williams, 2001a). However, this measure corresponded well to the devaluation experienced by adolescent girls in relation to males theorized by gende red pathways theory (Belknap & Holsinger, 2006; Chesney Lind & Shelden, 2004; Giordano et al., 2006) and to the prediction by general strain theory that girls respond to stra in with self denigratory emotion (Broidy & Agnew, 1997). Further limitations of the study arose from the original data collection procedures. Although the child sexual abuse measure was initially collected prospectively, most of the indicators of childhood maltreatment were collected from the participants based on their recollections of their childhood during follow up interviews. Such retrospectively

PAGE 189

180 collected data are not considered as accurate or reliable as data collected and documented at the time of its occurrence (Butz, 1981). Recall of past life events can be incomplete or in accurate, especially when investigating traumatic or stressful childhood occurrences (Stokes, Dritschel, & Bekerian, 2008; Williams, 1994, 1995). To offset these limitations, the researchers collecting the data were trained in asking questions on sensiti ve topics prior to conducting interviews with the study participants (Siegel & Williams, 2001a). During the interview, the interviewers initially established rapport with the participants and subsequently asked a series of questions about experiences in c Williams, 2003, p. 911). The interviewers were trained in handling indefinite responses to difficult or sensitive questions in order to determine whether the particip ants simply needed time to recall an event or whether they actually did not remember (Siegel & Williams, 2001a). In addition, recalling episodic memories can be faulty due to telescoping, i.e., assigning specific events to a specific time period when the y actually occurred prior to or after that period (Croyle & Loftus, 1992, 1993). To assist interviewees with memory recall and avoid telescoping, the interview was comprised of sections with sets of questions regarding different developmental life periods For questions regarding past events the interviewers defined the time frame and established memorable milestones which the questions referred (Siegel & Williams, 20 01a, 2003). Beyond concerns with memory recall, the inclusion of retrospective and cross sectional data into the structural equation models introduced uncertainty regarding the

PAGE 190

181 temporal relationships between the measures. For example, information on the s pecific ages of the minors when they ran away or experienced child physical abuse were not noted in the data, precluding the assessment of the effect of the specific age of these childhood experiences on the likelihood of being prostituted. Although the chronological order of these events or e xperiences during childhood was unclear due to the lack of data regarding the specific ages of the participants, running away was positioned and assessed in the models as a consequence of child maltreatment rather th an an antecedent based on a large body of previous research on runaways indicating that many flee abusive homes (for review, see Baron, 2003). More importantly, the measure of relational shame could have been specified in the models as a result (or a cor relate) of child sex trafficking in prostitution rather than as a predictor or moderator increasing vulnerability to child sex trafficking because no information regarding the onset of relational shame was collected from the study participants. However, n umerous researchers have found shame or stigmatization to be a long lasting consequence of experiencing the various forms of child maltreatment included in the model (Arata, 2000, 2002; Barnett et al., 2005; Berenson & Andersen, 2006; Bloom, 2000; Bottoms, et al., 2003; Classen, et al., 2005; Fagan, 2005; Filipas & Ullman, 2006; Kolko, 2002; Messman et al., 2009; Miller Perrin & Perrin, 1999; Solomon & Heide, 2005; Stockdale et al., 2002; Van Bruggen et al., 2006; Whiffen & MacIntosh, 2005) Moreover, stud ies focused on sexual revictimization have found shame to be a mediator of the relationship between child sexual abuse and sexual revictimization ( for review, see Classen, et al., 2005; see also, Arata, 2000, 2002; Filipas & Ullman, 2006; Messman et al., 2 009; Stockdale et al., 2002; Van Bruggen et al., 2006; Whiffen & MacIntosh, 2005) Therefore, the exploration of relational shame as a predictor of being

PAGE 191

182 prostituted was pursued based on previous research and the theoretical framework of the study. Howev er, the effects of relational shame were explored through the use of supplementary structural equation models to prevent the temporal uncertainty of this variable from creating a potential threat to the internal validity of the primary models (Shadish et a l., 2002). As this study utilized secondary data analysis, limitations in model specification may be present possibly resulting in omitted variable bias. For example, there was a lack of data regarding paternal strain. As a result, only indicators of mat ernal strain were included in the model as representative of the level of caregiver strain. However, previous studies regarding caregiver/child relationships have found that mother/child relationships were typically considered the primary caregiver relati onships among African American children in lower income families (for review, see Liang et al., 2006). The inclusion of latent variables in structural equation models has been shown to reduce measurement error, yet two inaccuracies are possible when using latent variables (Kline, 2005). First, the naming fallacy is the unsound notion that the name given to a latent variable definitively reflects the identified underlying construct. Second, reification is an error made by treating the constructs derived fr om these latent labels as if they correspond directly to measurable qualities or to real existing entities. A further limitation of structural equation modeling is that alternative models may be specified that adequately fit or replicate the observed patt erns in the data (Kline, 2005). The use of the WLSMV estimator in assessing the models was essential (see Analytic Strategy in Chapter 5), yet employing this estimator also introduced several limitations including the use of a complex pair wise deletion t echnique for handling missing data that is not

PAGE 192

183 considered to be as effective as other techniques available with other types of estimators (Brown, 2006; Muthn & Muthn, 1998 2007). Contributions to Criminological Research This segment of the chapter th oroughly reviews the important advancements in understanding of the origins of vulnerability to child sex trafficking accomplished by this study through the development, application, and empirical assessment of a distinct problem focused theoretical framew ork derived from criminological theory. Prior research on child sex trafficking had employed a risk factor approach when investigating the origins of victimization and had been built primarily on case study research, gathering information based on the re ported experiences of prior victims (Estes & Weiner, 2001, 2005; Kennedy et al., 2007; Priebe & Suhr, 2005; Williams & Frederick, 2009). These studies were descriptive in nature and lacked a sound theoretical framework, limiting the progress of solution d evelopment. In light of the need for a theoretical foundation on which to build a greater ors reactive pathway was partially foreshadowed by the child victimization pathway included within a sensitizing typology of female pathways into offending pioneered b y gendered pathways theory (Daly, 1992; Giordano et al., 2006; Salisbury & Van Voorhis, 2009). This descriptive schematic from the sensitizing typology was made up of a series of stepping stones marked by troubling transitions from childhood a buse into pr ostitution (Giordano

PAGE 193

184 strain reactive pathway. General strain theory served as the theoretical basis for the proposed pathway supplying both explanations of the origins and th e mecha nisms operating within the life trajectory terminating in child sex trafficking in prostitution (Agnew et al., 2000). Whether criminological theory could prove to be useful in explaining victimization in child sex trafficking was one of the inquirie s embedded within this study. The research findings support the study premise that criminological theories could be beneficial in rendering a more complete elucidation of victim susceptibility and the life adversities that place certain minors, specifical ly certain U.S. girls, on a high risk pathway toward entrapment in child sex trafficking. The application of criminological theory to the problem of child sex trafficking represents an important advancement in the understanding of this crime against child ren. Although the problem of child sex trafficking fits soundly within the domain of criminology, the issue has not received a great deal of attention from criminological theorists, nor has it been a central focus of criminological research. The theoret ically based pathway assessed by this research was framed by several key propositions from criminological theory. The pathway was empirically identified and key theoretical hypotheses were supported by the multivariate analyses. Detecting support for the theoretically based pathway implies that criminological theories can prove useful in understanding victimization in child sex trafficking. The results of this study val idate the stated expectation that certain criminological theories are not only capable of elucidating the causes of juvenile delinquency, but that these same theories can also contribute to

PAGE 194

185 the understanding the origins of specific types of juvenile victimizations, such as child sex trafficking. In considering the empirical support found for the theoretical propositions assimilated together to frame the pathway into child sex trafficking, several findings deserve highlighting as useful contributions to the existing body of scholarly literature on the subject matter. In previous studies ge ndered pathways theory and general strain theory had been tested as competing theories to explain juvenile delinquency (Belknap & Holsinger, 2006). However, in this study the two theories were used effectively as complementary in order to develop fuller c omprehension of vulnerability to victimization in child sex trafficking. Although gendered pathways theory has been primarily supported through the use of qualitative research methods (for reviews, see Morash, 2006; Salisbury & Van Voorhis, 2009; Simpson et al., 2008; see also, Miller, 2008a, 2008b), this study adds to the emergent body of research using quantitative methods that has found partial support for the sensitizing typology depicting distinctive pathways followed by females enmeshed in criminal environments and activities (Gavazzi et al., 2006; Johansson & Kempf Leonard, 2009; Salisbury & Van Voorhis, 2009). Furthermore, the findings from the study compellingly substantiated the key propositions of general strain theory regarding the consequenc es of family strain. As was implicitly proposed by Agnew et al. (2000) in a discussion on the effects of caregiver strain on children resulting in juvenile delinquency, this study provided convincing evidence for a conceptualization of the transference of strain from caregiver to child transmitted by means of child neglect and abuse that consequently triggers a strain reactive response in the maltreated child and ultimately leads to victimization in child sex

PAGE 195

186 trafficking. This finding supporting a transge nerational transmission of strain provides substantial validation for general strain theory and the long lasting and potent influence of strain on families and children. In addition, the limited availability of reliable data had previously restricted the application of advanced quantitative analysis to enhance the understanding of increased vulnerability to child sex trafficking faced by certain minors. Comparative analyses examining pathways of both victims and nonvictims of child sex trafficking were l acking. Few longitudinal or developmental studies had looked at the process of entrapment in child sex trafficking. As a result, prior research on the topic had not yet established how risks develop and interact to create the heightened vulnerability obs erved in certain minors (Williams & Frederick, 2009). For example, although family dysfunction and caregiver adversity, such as caregiver substance abuse or mental health problems, had been commonly noted as risk markers for child sex trafficking victims in prostitution (Clawson, 2009; Estes & Wiener, 2005), how these risks operated or functioned to create vulnerability had not been examined. The use of structural equation models, rather than single equation models, allowed for such extended specificat ion of the theorized mechanisms proposed by this study. One example of the benefits of this methodological approach was the finding that running away functioned as both a consequence and predictor of sexual victimization. This observation that running aw ay functions as a conduit of revictimization validates findings by previous researchers on runaways that have shown that abused girls who run away are more likely to be sexually exploited in prostitution than those who run away

PAGE 196

187 but have not been previous ly abused ( Saewyc & Edinburgh, 2010; Saewyc, et al., 2008; Scott et al., 2003; Tyler & Johnson, 2006a) In summary, by using structural equation modeling this study advanced the exploration of the problem of child sex trafficking in the United States. Th e findings of the study have provided greater comprehension of victim vulnerability by empirically identifying a theoretically framed pathway by which certain minors became entrapped in prostitution. Concurrently, the analyses supplied additional empirica l validation of the findings from previous studies focused on child sex trafficking victims that had been based primarily on qualitative methods of research. In looking ahead, additional targeted research using varied, nationally representative, and even t ransnational samples is needed in order to confirm and possibly extend the findings regarding this strain reactive pathway. As mentioned, studies are needed to determine whether this pathway is specific to girls or would apply across both genders. Althou gh this study explored the role of the negative emotion, further investigation into the role of negative emotions, such as guilt, stigmatization, and shame, in elevating vulnerability to victimization is warranted. Yet, these initial suggestions for futu re research only represent a small fraction of the scholarly research that is needed regarding this poorly understood and often overlooked crime against children. Directions for Future Research Indisputably, scholarly research in the area of child sex traf ficki ng is severely 2007; Kelly, 2005; Lehti & Aromaa, 2007; Musto, 2009; Tyldum & Brunovskis, 2005;

PAGE 197

188 Williams & Frederick, 2009; Zhang, 2009). Efforts to comb at child sex traffickin g will continue to be extremely inadequate, limited to blindly battling an undefined adversary and haphazardly applying remedies to victims, until an empirically validated knowledge base becomes available to guide and enhance protection and intervention ef forts. Empirically based information is crucial to enlighten the public and potentially at risk youth about the dangers of child sex trafficking and expose inaccuracies within popular culture that glamorize the pimp/ho relationship (Farley, 2003; Kreston, 2000, 2005; Dalla, 2000, 2006) Emerging research on sex traffickers has suggested that traffickers employ recruiters to spy out needy youth or runaways by frequenting their typical or routine locations (Albanese, 2007; Anderson & Michaelson, 2007). The psychological and mental immaturity inherent in all minors (Cauffman & Steinberg, 1995; Steinberg & Scott, 2003; Steinberg, 2010) prognosticate that minors face severe limitations in their capability to detect exploitative motives or defend against the ma nipulation of sex traffickers and recruiters. An accurate knowledge base is necessary for the preparation of prevention e ducation targeting at risk minors, warning them about the tactics used by sex traffickers or their recruiters to entrap victims Due to previous research findings indicating an association between psychopathy and the common characteristics of sex traffickers or pimps (Spidel et al., 2007), further research endeavors focused on determining whether traffickers generally possess psychopat hic tendencies would be beneficial and possibly provide answers regarding the nature of the trafficker and child sex trafficking victim relationship. Babiak and Hare (2007) hypothesized that qualities of psychopaths may facilitate their proficiency and s uccess in criminal activities. Possessing charm and social

PAGE 198

189 manipulation, psychopathic sex traffickers may succeed at recruiting unsuspecting and vulnerable minors. In addition, the common reporting of ruthless treatment of sex trafficking victims corres disregard for others ( Albanese 2007; Raphael, Reichert, & Powers, 2010). Exploring the common characteristics or psychological profiles of sex traffickers could provide beneficial information o n likely crime targets and the tactics that facilitate entrapment into child sex trafficking, such as whether sex traffickers are manipulating minor victims using psychopathic offenders (Spidel et al., 2006, p. 163). prostitution (Albanese, 2007; Priebe & Suhr, 2005; Raphael et al., 2010; Williams et al., 2008). Far different from the glamorized, 2003; Kreston, 2000, 2005; Dal1a, 2000, 2006), o nce a child is initially entrapped by a trafficker a grooming proce produce intense loyalty and trauma bonding the victim fear a Dutton & Painter, 1981, 1993; James, 1994; Herman, 1992, 2003; van der Kolk, McFarlene, & Weisaeth, 2007). According to the interviewed law enforcement officials, prosecutors, and social service providers, the missing key to prosecuting child sex traffickers and rescuing the victims is unlocking the trauma induced bond that develops between them (SHI, 2008). while we lack sufficien t information about the nature of the relationship that these youth have with

PAGE 199

190 their pimps, clearly there is a strong bond between them that will require a sustained and Researchers ha ve not yet conducted in depth investigations into the nature of the exploitative relationship or trauma bonding between sex traffickers and minor victims, nor has a theoretical framework useful in explaining this phenomenon been developed or assessed. Co nducting research focused on understanding the psychological stronghold that traffickers maintain over minor victims is a vital step needed to ensure the successful prosecution of traffickers and the implementation of effective therapeutic interventions co ncentrated on rescuing and restoring child sex trafficking victims. In view of the need to identify and dissolve trauma bonds between sex traffickers and minor victims, a standardized instrument is needed for detecting or measuring the level of trauma bon ding present in a relationship between a minor and an abusive exploiter, such as a sex trafficker. Such an instrument could be developed based on the trauma bonding items included within the post traumatic stress index developed by Carnes (1997) and could be constructed similarly to the Stockholm Syndrome Scale used for measuring Stockholm Syndrome in young women in dating relationships (Demarest, 2009; Graham et al., 1995). The newly designed instrument could be adapted for both clinical and research pur poses. The included items would detect and measure trauma bonds, i.e., the presence and level of dysfunctional psychological connections to people who are dangerous, shaming, or exploitative (Carnes, 1997). Information provided by such an instrument wou ld assist in answering the research questions regarding the psychological processes generated by the actions of the sex

PAGE 200

191 traffickers and how minors may continue to experience the influence of this bonding with a trafficker even after being separated from the exploitive situation. Preventing and combating human trafficking are among the most vexing problems currently facing the law enforcement community. The federal government has invested millions of dollars in these efforts, as well as hundreds of state and local law enforcement personnel (General Accounting Office, 2007). Yet, evaluations to ascertain the effectiveness of policies and programs that respond to child sex trafficking victims are almost completely absent. Addressing deficiencies in resear ch on this topic, Laczko involved in the fight against trafficking, such as service providers, law enforcement 14). Evaluation research is necessary to assist policy makers in implementing effective strategies and to guide social service providers toward best practices focused on rescuing and restoring victims. Such evaluations are vitally important for law enfor cement and other professionals working in the criminal justice system to facilitate and aid their efforts to protect victims and prosecute traffickers (GAO, 2007). Conclusion Estes and Weiner (2005) stated that although the many faces of children that exp erience sexual exploitation are continually changing, the socio psychological dynamic remains constant; an overpowering offender controls the child for sexual, economic, or other benefit. Speaking out about the plight of such vulnerable minors, the Feder al Advisory Committee on Juvenile Justice (2007) reported,

PAGE 201

192 Juveniles involved in prostitution tend to be the most vulnerable of youth. Many have histories of victimization, trauma, abuse, and neglect. Many have experienced homelessness and other forms of abandonment. Many become involved in prostitution to meet basic needs when social service providers and/or other responsible adults have failed to provide (p. 28). The results of this study suggest that abused and neglected minors, already bearing a dispro portionate amount of strain and adversity due to harmful family conditions, unfairly become the very minors who are most likely to endure further victimization in child sex trafficking in prostitution. Imparting information regarding the compounding effec ts of family adversity, child maltreatment, and its undeserved consequences may help to equip law enforcement officials, child protection professionals, and child advocates with crucial information needed to craft effective safeguards to prevent or impede victimization in child sex trafficking thereby preventing future harm to vulnerable minors. The development and analytic assessment of a theoretical framework useful in explaining victimization in child sex trafficking that can be repeatedly applied and evaluated in future research was a main objective of this study. The incorporation of criminological theory into the theoretical framework of this research revealed that the field of criminology is able to adeptly provide a greater understanding of the pr oblem of child sex trafficking in the United States. As a result, this study signifies that the remedy to the neglect by academic research of the topic of child sex trafficking may lie within the field of criminology. Although abolishing this crime again st children in the United States will require an interdisciplinary effort, the necessity of extensive involvement and

PAGE 202

193 participation by criminologists and criminal justice professionals in the efforts to protect minors from victimization in sex trafficking should not be underestimated. This study indicates that criminological theory and research can make meaningful contributions to understanding and eradicating the crime of child sex trafficking.

PAGE 203

194 List of References Abbey, A. (2002). Alcohol related sexu al assault: A common problem among college students. Journal of Studies on Alcohol, Supplement No. 14 118 128. Abbey, A. (2008). Alcohol and sexual violence perpetration. Applied Research Forum. National Online Resource Center on Violence Against Women. R etrieved January 8, 2010, from http://new.vawnet.org/Assoc_Files_VAWnet/AR_AlcPerp.pdf Adelson, W. J. (2008). Child prostitute or victim of trafficking. University of St. Thomas Law Journal, 6, 96 127. Adler, F. (1975). Sisters in crime: The rise of the ne w female criminal New York: McGraw Hill. Quality child welfare practices with families affected by domestic violence: A strategic plan. New York: NYC Administration of 2010 from http://www.nyc.gov/html/acs/downloads/pdf/domestic_violence_plan.pdf Agnew, R. (1985). A revised strain theory of delinquency. Social Forces, 64, 151 167. Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Crimino logy, 30 47 87. Agnew, R. (2000). The contribution of social psychological strain theory to the explanation of crime and delinquency. In F. Adler & W. S. Laufer (Eds.), Advances in criminological theory, vol. 6: The legacy of anomie theory (pp. 113 137). New Brunswick, NJ: Transaction Publishers. Agnew, R. (2001). Building on the foundation of general strain theory: Specifying the types of strain most likely to lead to crime and delinquency. Journal of Research on Crime and Delinquency, 38 319 361. rain t heory: Specifying the Agnew, R. (2002). Experienced, vicarious, and anticipated strain: An exploratory study on physical victimization and delinquency. Justice Quarterly 19 603 632.

PAGE 204

195 Agnew, R. (2004). Stability and change in crime over the life course: A s train theory explanation. In T. P. Thornberry (Ed.), Advances in criminological theory, vol. 7: Developmental theories of crime and delinquency (pp. 101 132). New Brunswick, NJ: Transaction Publishers. Agnew, R. (2005a). Juvenile delinquency: Causes and co ntrol Los Angeles: Roxbury Publishing. Agnew, R. (2005b, November). Religion, gender and delinquency. In M. Zahn (Ed.), the meeting of the American Society of Criminology, Tor onto, Canada. Retrieved September 13, 2009, from http://girlsstudygroup.rti.org/docs/2005_ASC_C&C.ppt.pdf Agnew. R. (2006a). Pressured into crime: An overview of general strain theory Los Angeles: Roxbury Publishing. Agnew, R. (2006b). General strain the ory: Current status and directions for further research. In F. T. Cullen, J. P. Wight, & K. R. Blevins (Eds.), Advances in criminological theory, vol. 15: Taking stock: The status of criminological theory (pp. 101 123). New Brunswick, NJ: Transaction Publi shers. Agnew, R. (2007). Strain and violent behavior. In D. J. Flannery, A. T. Vazsonyi, & I. D. Waldman (Eds.), The Cambridge handbook of violent behavior and aggression (pp. 519 529). New York: Cambridge University Press. Agnew, R. (2009). The contributi on of mainstream theories to the explanation of female delinquency. In M. Zahn (Ed.), The delinquent girl (pp. 7 29). Philadelphia: Temple University Press. Agnew, R., Brezina, T.,Wright, J. P., & Cullen, F. T. (2002). Strain, personality traits, and delin quency: Extending general strain theory. Criminology, 40 (1), 43 72. Agnew, R., Piquero, N. L., & Cullen, F. T. (2009). General strain theory and white collar crime. In S. S. Simpson, & D. Weisburd (Eds), The criminology of white collar crime (pp. 35 60). N ew York: Springer Publishing. Agnew, R., Rebellon, C. J., & Thaxton, S. (2000). A general strain theory approach to families and delinquency. In G. L. Fox & M. L. Benson (Eds), Families, crime, and criminal justice (pp. 113 138). New York: JAI Press, Inc. Agrusa, J. F. (2003). AIDS and tourism: A deadly combination. In T. G. Bauer, & B. McKercher (Eds.), Sex and tourism: Journeys of romance, love, and lust (pp. 167 180). Binghamton, NY: Haworth Hospitality Press.

PAGE 205

196 Ahmed, E., Harris, N., Braithwaite, J., & B raithwaite, V. (2001). Shame management through reintegration. Cambridge: Cambridge University Press. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Akers, R. L. (1985). Deviant beha vior: A social learning approach (3 rd ed.). Belmont, CA: Wadsworth Publishing. Alarid, L. F., & Cromwell, P. (2006). and victimization. Los Angeles: Roxbury Publishing. Albanese, J. (2007). Commercial sex ual exploitation of children: What do we know and what can we do about it? (NCJ 215733). Washington, DC: Author. Retrieved January 11, 2009, from http://www.ncjrs.gov/pdffiles1/nij/215733.pdf American Medical Association (1992 ). Diagnostic and treatment gu idelines on child sexual abuse. Chicago, IL: Author. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two step approach. Psychological Bulletin, 103 411 423. Anderson, L. E. (Producer), & Michael son, B. (Producer) (2007). Bait and switch: The terror and tragedy of child prostitution [Motion Picture]. (Available from the Arizona Foundation for Legal Services and Education, 4201 N. 24 th Street, Suite 210, Phoenix, AR 85016 6288, http://www.azflse.o rg/index.cfm) Arata, C. M. (2000). From child victim to adult victim: A model for predicting sexual assault. Child Maltreatment, 5, 28 38. Arata, C. M. (2002). Childhood sexual abuse and sexual revictimization. Clinical Psychology: Science and Practice, 9 (2), 135 175. Arata, C. M., Langhinrichsen correlates of multi type maltreatment among urban youth. Child Abuse & Neglect, 31, 392 415. Asparouhov, T., & Muthn, B. (2008). Exploratory structural equation modeling. Retrieved May 8, 2010, from http://www.gseis.ucla.edu/faculty/muthen/articles/Article_129.pdf Associated Press (2009, November 15). Authorities suspend search for missing N.C. girl: Police spent day scouring part of state for 5 year o ld. Retrieved on November 16, 2009 from http://www.msnbc.msn.com/id/33945506/ns/us_news crime_and_courts/?GT1=43001

PAGE 206

197 Babiak, P., & Hare, R. D. (2007). Snakes in suits: When psychopaths go to work. New York: HarperCollins Publisher. Bales, K. (2007). What p redicts global trafficking? International Journal of Comparative and Applied Criminal Justice, 31 269 280. Banyard, V. L., Williams, L. M., & Siegel, J. A. (2003). The impact of complex trauma and depression on parenting: An exploration of mediating risk and protective factors. Child Maltreatment, 8 (4), 334 349. Barnes, J. E., Noll, J. G., Putnam, F. W., & Trickett, P. K. (2009). Sexual and physical revictimization among victims of severe childhood sexual abuse. Child Abuse & Neglect, 33, 412 420. Barnett, D. (2004, December). No child left behind: New rules for unaccompanied minor illegal aliens. Backgrounder, 13 1 6. Washington, DC: Center for Immigration Studies. Barnett, O., Miller Perrin, C. L., & Perrin, R. D. (2005). Family violence across the life span: An introduction (2 nd ed.). Thousand Oaks, CA: Sage Publications. Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Per sonality and Social Psychology, 51 1173 1182. Baron, S. W. (2003). Street youth violence and victimization. Trauma, Violence, & Abuse, 4, 22 44. role of background, sub cultural and situational factors. Justice Quarterly 18 759 789. Barrett, P. (2010). Technical whitepaper #8: Likert coarse categorization and Pearson correlations. Retrieved April 29, 2010, from http://www.pbarrett.net/techpapers/likert.pdf Baum, K. (20 05). Juvenile victimization and offending, 1993 2003 (N CJ Publication No. 209468). Washington, DC: National Institute of Justice. Beauregard, E., Proulx, J., & Rossmo, K. (2007). A descriptive model of the hunting process of serial sex offenders: A ratio nal choice perspective. Journal of Family Violence, 22 (6), 449 463. Beaver, K. M., Boutwell, B. B., Barnes, J. C. & Cooper, J. A. (2009). The biological underpinnings in adolescent victimization. Youth Violence and Juvenile Justice, 7, 223 238.

PAGE 207

198 Becker, H. S. (1963). Outsiders: Studies in the sociology of deviance. New York: The Free Press. Belknap, J. & Holsinger, K. (2006). The gendered nature of risk factors for delinquency. Feminist Criminology, 1, 48 71. Bender, K. (in press). Why do some maltreated you th become juvenile offenders? A call for further investigation and adaptation of youth services. Children and Youth Services Review. Bentler, P. M. & Chou, C. P. (1987). Practical issues in structural modeling. Sociological Methods and Research, 16(1) 78 117. Benda, B. B. (2005). Gender differences in Life Course theory of recidivism: A survival analysis. International Journal of Offender Therapy and Comparative Criminology, 49 325 342. Benda, B. B., Harm, N. J., & Toombs, N. J. (2005). Survival analysis of recidivism of male and female boot camp graduates using life course theory. Rehabilitation Issues, Problems, and Prospects in Boot Camp, 87 113. Benedict, L. W., & Zautra, A. J. (1993). Family environmental characteristics as risk factors for childhood sexual abuse Journal of Clinical Child Psychology, 22 (3), 365 374. Berenson, K. R., & Andersen, S. M. (2006). Childhood physical and emotional abuse by a parent: Transference effects in adult interpersonal relations. Personality and Social Psychology Bull etin, 32, 1509 1522. Belser, P. (2005). Forced Labour and Human Trafficking: Estimating the Profits. Geneva: International Labour Office. Retrieved January 04, 2010, from http://www.lastradainternational.org/lsidocs/80%20Profiting%20from%20forced% 20labour %20and%20trafficking%20(ILO).pdf Bloom, S. L. (2000). Email from America: The neglect of neglect: Part I. Psychotherapy Review, 2 (5), p. 208 210. Blumer, H. (1954). What is wrong with social theory? American Sociological Review, 18, 3 10. Bollen, K. A., & American Sociological Review, 46 232 239. Boney McCoy, S., & Finkelhor, D. (1995). Prior victimization: A risk factor for child sexual abuse and PTSD related symptomology among sexually a bused youth. Child Abuse and Neglect, 19 1401 1421.

PAGE 208

199 Booth, R. E., Zhang, Y., & Kwiatkowski, C. F. (1999). The challenge of changing drug and sex risk behavior of runaway and homeless adolescents. Child Abuse & Neglect, 23, 1295 1306. Bottoms, B. L., Niel son, M., Murray, R., & Filipas, H. (2003). Religion related child physical abuse: Characteristics and psychological outcomes. Journal of Aggression, Maltreatment & Trauma, 8 87 114. Bowen, G. A. (2006). Grounded theory and sensitizing concepts. Internatio nal Journal of Qualitative Methods, 5, 12 23. Bowlby, J. (1973). Attachment and loss, Vol. II, Separation: Anxiety and anger. New York: Basic Books. Bowlby, J. (1980). Attachment and loss, Vol. III, Loss: Sadness and depression. New York: Basic Books. Boxi ll, N. A., & Richardson, D. J. (2007). Ending sex trafficking of children in Atlanta. Affilia, 22, 138 149. Braithwaite, J. (1989). Crime. Shame and reintegration. Melbourne: Cambridge University Press. Braithwaite, J. (2000). Shame and criminal justice. C anadian Journal of Criminology, 42, 281 298. Briere, J. N., & Runtz, M. (1993). Childhood sexual abuse: Long term sequelae and implications for psychological assessment. Journal of Interpersonal Violence, 8 (3), 312 330. Briere, J. N., & Elliot, D. M. (2003 ). Prevalence and psychological sequence of self reported childhood physical and sexual abuse in general population. Child Abuse and Neglect 27(10), 1205 1222. Broidy, L., & Agnew, R. (1997). Gender and crime: A general st r ain theory perspective. Journal of Research in Crime and Delinquency, 34, 275 306. Brown, T. (2006). Confirmatory factor analysis for applied research. New York: Guilford Press. Brown, M., & Bloom, B. (2009). Reentry and renegotiating motherhood: Maternal identity and success on parole. Crime and Delinquency, 55, 313 336.

PAGE 209

200 Brown, J., Cohen, P., Johnson, J. G., & Salzinger, S. (1998). A longitudinal analysis of risk factors for child maltreatment: Findings of a 17 year prospective study of officially recorded and self reported child abuse a nd neglect. Child Abuse and Neglect, 22, 1065 1078. Bryant Davis, T., Chung, H., & Tillman, S. (2009). From the margins to the center: Ethnic minority women and the mental health effects of sexual assault. Trauma, Violence, & Abuse, 10, 330 357. Bushway, S ., Brame, R., & Paternoster, R. (1999). Assessing stability and change in criminal offending: a comparison of random effects, semiparametric, and fixed effects modeling strategies. Journal of Quantitative Criminology, 15, 23 61. Butz, W. P. (1981). Why co llect retrospective data? Santa Monica, CA: Rand Corporation. Campbell Augustine, M, Wilcox. P., Ousey, G.C., & Clayton, R.R. (2002) Opportunity theory and adolescent school based victimization. Violence and Victims, 17, 233 253. Canter, J., Rosenfeld, J. H., Butt, N. K., & Botash, A. (2008). Physical and sexual abuse of children. In F. E. Gallahue & L. D. Melville (Eds.), Emergency care of the abused (pp. 1 73). Cambridge: Cambridge University Press. Carnes, P. J. (1997). The betrayal bond: Breaking free of exploitive relationships. Deerfield Beach, FL: Health Communications, Inc. Carson, D. C., Sullivan, C. J., Cochran, J. K., & Lersch, K. M. (2009). General strain theory and the relationship between early victimization and drug use. Deviant Behavior, 30, 54 88. Casey, B. J., Jones, R. M., Levita, L., Libby, V., Pattwell, S. S., Ruberry, E. J., et al. (2010). The storm and stress of adolescence: Insights from human imaging and mouse genetics. Developmental Psychobiology, 52 225 235. Caspi, A., Lynam, D. R., Moffitt, T. E., and Silva, P. A. (1994). Are some people crime prone? Replications on the personality crime relationship across counties, genders, races, and methods. Criminology, 32. 163 195. Cauce, A. M., Stewart, A., Whitbeck, L. B., Paradise, M. & Hoyt, D. R. (2006). Girls on their own: Homelessness in female adolescents. In D. J. Bell, S. L. Foster, & E. J. Mash (Eds.), Handbook of Behavioral and Emotional Problems in Girls (pp. 439 461. New York: Kluwer Academic/Plenum Publishers.

PAGE 210

201 Cauffman, E. Lexcen, F. J., Goldweber, A, Shulman, E. P., & Grisso, T. (2007). Gender differences in mental health symptoms among delinquent and community youth. Youth Violence and Juvenile Justice, 5, 287 307. Cauffman, E., & Steinberg L. (1995). The cognitive and a ffective influences on adolescent decision making. Temple Law Review, 68 1763 1789. Chen, W. (2009). The link between juvenile offending and victimization. Youth Violence and Juvenile Justice, 7 119 135. Chesney place: Toward a feminist model of female delinquency. Crime and Delinquency, 35 5 29. Chesney In C. M. Renzetti, & L. Goodstein (Eds.), Women, crime, and criminal justice (pp. Chesney Lind, M, & Irwin, K. (2008). Beyond bad girls: Gender, violence and hype New York: Routledge. Chesney Lind, M., & Pasko, L. (2004). The female offender: Girls, women, and crime (2 nd ed.). Thousand Oaks, CA: Sage Publications. Chesney Lind, M., & Shelden, R. G. (2004). Girls, delinquency, and juvenile justice (3 rd ed.). Belmont, CA: Thomson Wadsworth. Child Welfare Information Gateway (2007). State statutes series: Definitions of child abus e and neglect. Retrieved January 15, 2010, from www.childwelfare.gov/systemwide/laws_policies/statutes/define.cfm Cicchetti, D., Rogosch, F. A., Gunnar, M. R., & Toth, S. L. (2010). The differential impacts of early physical abuse and sexual abuse and inte rnalizing problems on daytime cortisol rhythm in school aged children. Child Development, 81, 252 269. Classen C. C., Palesh, O. G., & Aggarwal, R. (2005). Sexual Revictimization: A review of the empirical literature. Trauma, Violence, & Abuse, 6 103 129. Clatts, M. C., & Davis, W. R. (1999). A demographic and behavioral profile of homeless youth in New York City: Implications for AIDS outreach and prevention. Medical Anthropology, 13, 365 374. Clawson, H. J. (2009). Human trafficking into and within the U nited States: A review of the literature. Washington, DC: Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Human and Heath Services. Retrieved December 25, 2009, from http://aspe.hhs.gov/hsp/07/HumanTrafficking

PAGE 211

202 Clawson, H. J., & Dutch, N. (2008). Addressing the needs of victims of human trafficking: Challenges, barriers, and promising practices Washington, DC: Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Human and Heath Services. Retriev ed December 25, 2009, from http://aspe.hhs.gov/hsp/07/HumanTrafficking/Needs/ib.pdf Clawson, H. J., & Grace, L. G. (2007). Finding a path to recovery: Residential facilities for minor victims of domestic sex trafficking. Washington, DC: Office of the Assis tant Secretary for Planning and Evaluation, U.S. Department of Human and Heath Services. Retrieved December 25, 2009, from http://aspe.hhs.gov/hsp/07/HumanTrafficking Clawson, H. J., Salomon, A., & Grace, L. G. (2008). Treating the hidden wounds: Trauma tr eatment and mental health recovery for victims of human trafficking. Washington, DC: Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Human and Heath Services. Retrieved December 25, 2009, from http://aspe.hhs.gov/hsp/07/Hu manTrafficking Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44 588 608. Cohen, L.E., & Felson, M. (2006). Routine activity theory. In Cullen, F. T. & Agnew, R. (Eds. ) Criminological theory: Past to present, (3 rd ed., pp. 433 443). New York: Oxford University Press. Colman, R. A., Kim, D. H., Mitchell Herzfeld, S., & Shady, T. A. (2009). Delinquent girls grown up: Young adult offending patterns and their relation to ea rly legal, individual, and family risk. Journal of Youth and Adolescence, 38, 355 366. Cooper, S. W. (2005a). A brief history of child sexual exploitation. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 1 24). St. Louis, MO: GW Medical Publishing. Cooper, S. W. (2005b). Human trafficking for sexual exploitation: The role of Interpol. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, an d Internet crimes against children (pp. 735 744). St. Louis, MO: GW Medical Publishing. Cooper, S. W. (2005c). Medical analysis of child pornography. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 213 242). St. Louis, MO: GW Medical Publishing.

PAGE 212

203 Croyle, R. T., & Loftus, E. F. (1992). Improving episod ic memory performance of survey respondents. In J. M. Tanur (Ed.), Questions about questions: Inquiries into the cognitive bases of surveys (pp. 95 101). New York: Russell Sage. Croyle, R. T., & Loftus, E. F. (1993). Recollection in the kingdom of AIDS. In D. G. Ostrow & R. Kessler (Eds.), Methodological in AIDS Behavioral Research (pp.163 180). New York: Plenum. Cuevas, C. A., Finkelhor, D., Turner, H. A., & Ormrod, R. K. (2007). Juvenile delinquency and victimization: A theoretical typology. Journal of In terpersonal Violence, 22 1581 1602. Curtis, R., Terry, K., Dank, M., Dombrowski, K., & Khan, B. (2008). The commercial sexual exploitation of children in New York City, Volume 1: The CSEC population in New York City: Size, characteristics, and needs (NCJ 225083). Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice. Daigle, L. E., Cullen, F. T., & Wright, J. P. (2007). Gender differences in the predictors of juvenile delinquency: Assessing the generality specificity debate. Youth Viol ence and Juvenile Justice, 5, 254 286. Dalla, R. L. (2000). Exposing the pretty woman myth: Qualitative examination of the lives of female street walking prostitutes. Journal of Sex Research, 37 (4), 344 353. Dalla, R. L. (2006). Exposing the pretty woman m yth: A qualitative investigation of street level prostituted women. Lanham, MD: Lexington Books. and problems of representation. Southern California Review of Law and Women (1), 497 535. Daly, K. (1998). Gender, crime, and criminology. In M. Tonry (Ed.), The handbook of crime and punishment (pp. 85 108). New York: Oxford University Press. Daly, K. (2006). Feminist thinking about crime and justice. In S. Henry & M Lanier (Eds.), The essential criminology reader (pp. 205 213). Boulder, CO: Westview Press. awareness of and discomfort with sexual assault cues: Effects of alcohol consumption and relationship type. Violence Against Women, 15 1106 1125. De Bellis, M. (2001). Developmental traumatology: A contributory mechanism for alcohol and substance abuse. Psychoneuroendocrinology, 27 155 170.

PAGE 213

204 DeLisi, M., Barnes, J. C., Beaver, K. M, & Gibson, C. L. (2009). Delinquent gangs and adolescent victimization revisited. Criminal Justice and Behavior, 36 808 823. DeMarco, R. (2004). The war on child pornography? Who are the victims? Computer Crime Research Center. Retrieved October 11 2009, from http://www.crime research.org/news/16.06.2004/433/ Demarest, R. A. (2009). The relationship between Stockholm Syndrome and Posttraumatic Stress Disorder in battered women." Student Pulse Academic Journal 1.11. Retrieved May 16, 2010, from: deMause, L. (1998). The history of child abuse. The Journal of Psychohistory, 25 (3), 1 13. Dennis, J. P. (2008). Women are victims, men make choices: The invisibility of men and boys in the global sex trade. Gender Issues 25, 11 25. DeRosa, C. J., Montgomery, S. B., Hyde. J., Iverson, E., & Kipke, M. D. (2001). HIV risk behavior and HIV testing: A comparison of rates and associated factors among homeless and runaway adolescents in two cities. AIDS Education and Prevention 13, 131 148. DeStefano, A. M. (2007). The war on human trafficking: U.S. policy assessed New Brunswick, NJ: Rutgers University Press. DiLauro, M. D. (2004). Psychosocial factors associated with types of child maltreatment. Child Welfare, 83, 69 99. DiS tefano, C., & Hess, B. (2005). Using confirmatory factor analysis for construct validation: An empirical review. Journal of Pychoeducational Assessment, 23, 225 241. Downs, W. R. (1993). Developmental considerations for the effects of childhood sexual abus e. Journal of Interpersonal Violence 8 (3), 331 345. Dulmus, C. N., & Hilarski, C. (2002). Children and adolescents exposed to community violence. In L. A. Rapp Paglicci, A. R. Roberts, & J. S. Wodarski (Eds.), Handbook of violence (pp. 129 147). New York: John Wiley & Sons. Dutton, D. G., & Painter, S. L. (1981). Traumatic Bonding: The development of emotional attachments in battered women and other relationships of intermittent abuse. Victimology: An International Journal, 7(4), 139 155. Dutton, D. G., & Painter, S. (1993). Emotional attachments in abusive relationships: A test of traumatic bonding theory. Violence & Victims, 8 105 119.

PAGE 214

205 Dziuban, C. D., & Shirkey, E. C. (1974). On the psychometric assessment of correlation matrices. American Educational Re search Journal, 11 211 216. Edmond, T., Rubin, A., & Wamback, K. G., (1999). The effectiveness of EMDR with adult female survivors of childhood sexual abuse Social Work Research, 23 (2), 103 116. Elder, G. H., & Giele, J. Z. (Eds.)(2009). The craft of lif e course research. New York: Guilford Press. Ellis, S. & Savage, J. (2009). Strain, social support, and persistent criminality. In J. Savage (Ed.), The development of persistent criminality (pp. 71 89). New York: Oxford University Press. Ennett. S. T., Bai ley, S. L., & Federman (1999). Social network characteristics associated with risky behaviors among runaway and homeless youth. Journal of Health and Social Behavior, 40, 63 78. Esbensen, F., Huizinga, D., Menard, S. (1999). Family context and criminal vic timization in adolescence. Youth & Society, 31, 168 198. Estes, J. R., Azaola, E., & Ives, N. (2005). The commercial sexual exploitation of children in North America. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 297 335). St. Louis, MO: GW Medical Publishing. Estes, J. R., & Weiner, N. A. (2001, r evised 2002). The Commercial Sexual Exploitation of Children in the U.S., Canada and Mexico. Philadelphia: University of Pennsylvania School of Social Work, Center for the Study of Youth Policy. Retrieved January 15, 2009, from http://www.sp2.upenn.edu/~re stes/CSEC_Files/Complete_CSEC_020220.pdf Estes, J. R., & Weiner, N. A. (2005). The commercial sexual exploitation of children in the United States. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social s cience aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 95 128). St. Louis, MO: GW Medical Publishing. Ezell, M. E., & Cohen, L. E. (2005). Desisting from crime : Continuity and change in long term crime patterns of serious chronic offenders. New York: Oxford University Press.

PAGE 215

206 Fadel, F. B. (2005). Exiting route: A peer support model for exiting and healing programs. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 965 978). St. Louis, MO: GW Medical Publishi ng. Fagan, A. A. (2005). The relationship between adolescent physical abuse and criminal offending: Support for an enduring and generalized cycle of violence. Journal of Family Violence, 20 279 290. Fang, Bay (2005, October 14). Young lives for sale U.S News & World Report, 139 (15), 30 34. Farley, M. (2003). Preface: Prostitution, trafficking, and traumatic stress. In M. Farley (Ed.), Prostitution, trafficking and traumatic stress (pp. 17 32). Binghamton, NY: Haworth. if legalized or decriminalized. Violence Against Women, 10, 1087 1125. Farley, M., & Kelly, V. (2000). Prostitution: A critical review of the medical and social sciences literature. Women & Criminal Just ice 11 (4), 29 64. Farley, M., Lynne, J., & Cotton, A. J. (2005). Prostitution in Vancouver: Violence and the Transcultural Psychiatry, 42 (2). 242 271. Farr, K. (2005). Sex trafficking: The global market in women and c hildren. Worth Publishing: New York. Farrell, A., & Fahy, S. (2009). The problem of human trafficking in the U.S.: Public frames and policy responses Journal of Criminal Justice, 31, 617 626. Farrington, D. P. (2000). Explaining and preventing crime: the globalization of knowledge The American Society of Criminology 1999 Presidential Address. Criminology, 38 1 24. Farrington, D. (2008). Introduction to integrated developmental and life course theories of offending. In D. Farrington (Ed.), Advances in c riminological theory, vol. 14: Integrated developmental and life course theories of offending (pp. 313 364) New Brunswick, NJ: Transaction Publishers. Farrow, J. A. H. (2005). Psychosocial context leading juveniles to prostitution and sexual exploitation. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children ( pp. 337 348). St. Louis, MO: GW Medical Publishing.

PAGE 216

207 Fassler, I. R., Amodeo, M., Griffin, M. L., Clay, C. M., & Ellis, M. A. (2005). Predicting long term outcomes for women sexually abused in childhood: Contribution of abuse severity versus family environme nt. Child Abuse & Neglect, 29, 269 284. Federal Bureau of Investigations (FBI) (n.d.). Retrieved on November 22, 2009 from http://www.fbi.gov/innolost/innolost.htm Federal Advisory Committee on Juvenile Justice (FACJJ; 2007). Retrieved January 11, 2009, f rom http://www.facjj.org/annualreports/ccFACJJ%20Report%20508.pdf Feiring, C., Miller Johnson, S., & Cleland, C. M. (2007). Potential pathways from stigmatization and internalizing symptoms to delinquency in sexually abused youth. Child Maltreatment, 12, 2 20 232. Felson, M. (2001). The routine active approach: A very versatile theory of crime. In R. Paternoster & R. Bachman (Eds.), Explaining criminals and crime (pp. 43 46). Los Angeles: Roxbury Publishing Company. Felson, M. (2002). Crime and everyday life (3 rd ed.). Thousand Oaks, CA: Sage Publications. Ferraro, K. J. (2006). Neither angels nor demons. Boston: Northeastern University Press. Ferraro, K. J., & Moe, A. M. (2006). The impact of mothering on criminal offending. In L. F. Alarid & P. Cromwell (Ed s.), crime and victimization (pp. 79 92). Los Angeles: Roxbury Publishing. Filipas, H. H., & Ullman, S. E. (2006). Childhood sexual abuse, coping responses, self blame, posttraumatic stress disorder, and adult se xual revictimization. Journal of Interpersonal Violence, 21 (5), 652 672. Finckenauer, J. O., & Schrock, J. (2003). Human trafficking: A growing criminal market in the U.S. In A. M. Troubnikoff, (Ed.), Trafficking in women and children: Current issues and developments (pp. 31 38). New York: Nova Science Publishers. Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: The Free Press. Finkelhor, D. (1986). A sourcebook on child sexual abuse Beverly Hills, CA: Sage Publications. Finke lhor, D. (1994). Current information on the scope and nature of child sexual abuse. The Future of Children, 4 (2), 31 53.

PAGE 217

208 Finkelhor, D. (2007). Developmental victimology: The comprehensive study of childhood victimization. In Davis, R. C., Luirigio, A. J. & Herman, S. (Eds.). Victims of crime (3 rd ed., pp. 9 34). Thousand Oak, CA: Sage Publications. Finkelhor, D. & Browne, A. (1985). The traumatic impact of childhood sexual abuse: A conceptualization. American Journal of Orthopsychiatry, 55, 530 541. Finke lhor, D., Hamby, S. L., Ormrod, R., & Turner, H. A. (2005). The Juvenile Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse and Neglect 29 383 412. Finkelhor, D., Hammer, H., & Sedlak, A. J. (2002). Nonfamily abducted chi ldren: National estimates and characteristics. (NCJ Publication No. 196467). National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children. Washington, DC: Offices of Juvenile Justice. Finkelhor, D., Mitchell, K., & Wolak, J. (2005). O nline victimization: What youth tell us. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution and Internet crimes against children (pp. 437 468). St. Louis, MO: GW Medical Publishing. Finkelhor, D., & Ormrod, H. A (2004, June). Prostitution of juveniles: Patterns from NIBRS. Juvenile Justice Bulletin. Washington D.C.: Office of Juvenile Justice and Delinquency Prevention. Retrieved December 25, 2009, from http://www.ncjrs.gov/pdffiles1/ojjdp/203946.pdf Finkelhor, D., Ormrod, H. A., & Turner, H. A. (2007). Poly victimization and trauma in a national longitudinal cohort. Development and Psychopatho logy, 19, 149 166. Finkelhor, D., Ormond, H. A., Turner, H.A., & Hamby, S. L. (2005). The victimization of children and youth: A comprehensive, national survey. Child Maltreatment, 10 (1), 2 25. Fisher, B. S., Daigle, L. E., & Cullen, F. T. (2009). What distinguishes single and recurrent sexual victims? The role of lifestyle routine activities and the first incident characteristics. Justice Quarterly online access. Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative models of est imation for confirmatory factor analysis with ordinal data. Psychological Methods, 9, 466 491. Florida Department of Juvenile Justice (FDJJ; 2006). 2006 outcome evaluation report. Retrieved September 14, 2009, from http://www.djj.state.fl.us/Research/OE/20 06/2006_OE.pdf

PAGE 218

209 Flowers, R. B. (1998). The prostitution of women and girls. Jefferson, NC: McFarland & Company. Flowers, R. B. (2001). Runaway kids and teenage prostitution Westport, CT: Praeger Publishers. Fong, R., & Cardoso, J. B. (in press). Child huma n trafficking victims: Challenges for the child welfare system. Evaluation and Program Planning, doi: 10.1016/j.evalprogplan.2009.06.018. Ford, J. D., Chapman, J., Mack, M., & Pearson, G. (2006). Pathways for traumatic child victimization to delinquency: Implications for juvenile and permanency court proceedings and decisions. Juvenile and Family Court Journal, 57 (1), 13 26. Ford, J. D., Stockton, P., Kaltman, S., & Green, B. L. (2006). Disorders of extreme stress (DESNOS) symptoms are associated with type and severity of interpersonal trauma exposure in a sample of healthy young women. Journal of Interpersonal Violence, 21 1399 1416. Fraley, R. C. (2002). Attachment stability from infancy to adulthood: Meta analysis and dynamic modeling of developmental m echanisms. Personality and Social Psychology Review, 6 (2), 123 151. Free, K. A., & Collins, M., K. (2005). Resources of the Exploited Child Unit at the National Center for Missing & Exploited Children. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. K ellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 203 212). St. Louis, MO: GW Medical Publishing. Fri esendorf, C. (2010). Pathologies of security governance: Efforts against human trafficking in Europe. In L. Territo, & G. Kirkman (Eds.), International sex trafficking of women & children: Understanding the global epidemic (pp. 87 112) Flushing, NY: Loose leaf Law Publications. Friedman, S. A. (2007). Alternative report to the initial report of the United States of America to the UN Committee on the rights of the child concerning the optional protocol to the convention on the rights of the child on the sale of children, child prostitution, and child pornography. Brooklyn, NY: ECPAT USA (End Child Prostitution, Child Pornography, and Trafficking of Children for Sexual Purposes). General Strain Theory. Journal of Loss and Trauma, 12, 383 418. Fuller, T., & Nieto, M. (2009). Substantiation and maltreatment rereporting. Child Maltreatment, 14 27 37.

PAGE 219

210 Gaardner, E., & Belknap, J. (2002). Tenuous borders: Girls transferred to adult cou rt. Criminology, 40, 481 518. Garson, G. D. (2009). Structural equation modeling. Retrieved January 16, 2010, from http://faculty.chass.ncsu.edu/garson/PA765/structur.htm Gavazzi, S. M., Yarcheck, C. M., & Chesney Lind, M. (2006). Global risk indicators an d the role of gender in a juvenile detention sample. Criminal Justice and Behavior, 33, 597 612. General Accounting Office (2007). Human trafficking: A strategic framework could help enhance the interagency collaboration needed to effectively combat traffi cking crimes. Washington, DC: author. Retrieved July 4, 2009, from http://www.gao.gov/new.items/d07915.pdf Gershoff, E. T. (2002). Corporal punishment by parents and associated child behaviors and experiences: A meta analytic and theoretical review. Ps ychological Bulletin, 128 539 579. Giordano, P. C., Deines, J. A., & Cernkovich, S. A. (2006). In and out of crime: A life Gender and crime: Patterns in victimization and of fending (pp. 17 40). New York: New York University Press. immersion in street crime. Women and Criminal Justice, 4, 63 89. Goddard, C., De Bortoli, L., Saunders, B. J., Child Abuse Review, 14, 275 291. Goodkind, S., Ng, I., & Sarri, R. C. (2006). The impact of sexual abuse in the lives of young women involved or at risk of involvement with the juvenile ju stice system. Violence against Women, 12, 456 477. Gottfredson, M. R. (2006). The empirical status of control theory in criminology. In F. T. Cullen, J. P. Wight, & K. R. Blevins (Eds.), Advances in criminological theory, vol. 15: Taking stock: The st atus of criminological theory (pp. 205 230) New Brunswick, NJ: Transaction. Gottfredson, M. R. (2007). Self control theory and criminal violence. In D. J. Flannery, A. T. Vazsonyi, & I. D. Waldman (Eds.), The Cambridge handbook of violent behavior and aggression (pp. 533 544). New York: Cambridge University Press. Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press.

PAGE 220

211 Bi bliography of research based literature (NIJ Grant No. 2007 VT BX K002). Washington, DC: Institute for the Study of International Migration, Walsh School of Foreign Service, Georgetown University. Research on child survivors of sex trafficking for sexual and labor exploitation (NIJ Grant No. 2005 IJ CX 0051). Washington, DC: Institute for the Study of International Migration, Walsh School of Foreign Service, Georgetown University. Collett, E. A. (2005). Research on human trafficking in North America: A review of literature. International Migration 43, 99 128. Graham, D. L., Rawlings, E. L., Ihms, K, Latimar, D., Foliano, J., Thompson, A., et al. (1995). A scale for identifying Stoc kholm syndrome reactions in young dating women: Factor structure, reliability and validity. Violence and Victims, 10, 3 22. Grant, K. E., Compas, B. E., Thurm, A. E., McMahon, S. D., Gipson, P. Y., Campbell, A. J. et al. (2009). Stressors and child and ad olescent psychopathology: Evidence of moderating and mediating effects. Clinical Psychology Review, 26, 257 283. Grant, N. (2005). Experiential youth: Canada. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 179 192). St. Louis, MO: GW Medical Publishing. Grasmick, H. G., & Bursik, R. J. (1990). Consc ience, significant others, and rational choice: Extending the deterrence model Law & Society Review, 24, 837 861. W. Cooper, R. J. Estes, A. P. Giardino, N. D. Kell ogg & V. I. Vieth (Eds.), Medical, legal, & social science aspects of child sexual exploitation: A comprehensive review of pornography, prostitution, and internet crimes (pp. 711 714). St. Louis, MO: G.W. Medical Publishing. Graycar, A., & McCusker, R. (20 07). Transnational crime and trafficking in persons: Quantifying the nature, extent and facilitation of a growing phenomenon. International Journal of Comparative and Applied Criminal Justice, 31 147 166. Greene, J. M., Ennett, S. T., & Ringwalt, C. L. (1 999). Prevalence and correlates of survival sex among runaway and homeless youth. American Journal of Public Health, 89, 1406 1409.

PAGE 221

212 Greenblatt, M. & Robertson, M. (1993). Life styles, adaptive strategies, and sexual behaviors of homeless adolescents. Hospi tal and Community Psychiatry, 44, 1177 1180. Haddal, C. C. (2007). Unaccompanied alien children: Policies and issues. Washington, DC: Congressional Research Service. Retrieved on February 6, 2010, from http://assets.opencrs.com/rpts/RL33896_20070301.pdf H delinquency and gendered pathways to disadvantage in emerging adulthood. Social Forces, 82, 53 86. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homeless ness Cambridge, MA: Cambridge University Press. Halcon, L., & Lifson, A. (2004) Prevalence and predictors of sexual risk among homeless youth. Journal of Youth and Adolescence, 33, 71 80. Halter, S. (in press). Factors that influence police conceptualizat ion of youth involved in prostitution in six U.S. cities: Child sexual exploitation victims or delinquents? Child Maltreatment sex industry and the power of love. Will iam and Mary Journal of Women and Law, 9 1 29. Harris, N. (2003). Reassessing the dimensionality of moral emotions. British Journal of Psychology, 94 457 473. Harris, N., & Maruna, S. (2006). Shame, shaming and restorative justice: A critical appraisal. In D. Sullivan, & L. Tiff (Eds.), Handbook of restorative justice. New York: Routledge. Harrison, P. A., Fulkerson, J. A., & Beebe, T. J. (1997). Multiple substance use among adolescent physical and sexual abuse victims. Child Abuse and Neglect, 21 529 5 39. Hartman, J. L., Turner, M. G., Daigle, L. E., Exum, M. L., & Cullen, F. T. (2009). Exploring the gender differences in protective factors: Implications for understanding resiliency. International Journal of Offender Therapy and Comparative Criminology, 53 249 277. Hay, C. (2003). Family strain, gender, and delinquency. Sociological Perspectives, 46, 107 135.

PAGE 222

213 Hay, C., & Evans, M. M. (2006). Violent victimization and involvement in delinquency: Examining predictions from general strain theory. Journal of Criminal Justice, 34, 261 274. Haynie, D. L., Steffensmeier, D., & Bell, K. E. (2007). Gender and serious violence: Untangling the role of friendship sex composition and peer violence. Youth Violence and Juvenile Justice, 5 235 253. Heckman, J. J. (1981) Statistical models for discrete panel data. In C. F. Manski, & D. McFadden (Eds.), Structural analysis of discrete data with econometric applications (pp. 114 178). Cambridge, MA: MIT Press. Heckman, J. J. (2000, December 8). Microdata, heterogeneity, and the evaluation of public policy. Paper presented at the Bank of Sweden Nobel Memorial Lecture in Economic Sciences Retrieved September 11, 2009 from http://athens.src.uchicago.edu/jenni/econ311_2002/nobel_text.pdf Heide, K. M., & Solomon, E. P. (2006) Biology, childhood trauma, and murder: Rethinking justice. International Journal of Law and Psychiatry, 29 220 233. Heimer, K., & De Coster, S. (1999). The gendering of violent delinquency. Criminology, 37, 277 318. Heimer, K., & Kruttschnitt, C. (Eds .). (2006). Gender and crime: Patterns in victimization and offending. New York: New York University Press. Heimer, K., Wittrock, S., Unal, H. (2006). The crimes of proverty: Economic marginalization and the gender gap in crime. In K. Heimer, & C. Kruttsch nitt (Eds.), Gender and crime: Patterns in victimization and offending (pp. 115 137). New York: New York University Press. Heineman, E., Shelton, & Anton, M. (2006). Overview of youth and young adult homelessness in Minnesota. Amherst A. Wilder Foundation, Wilder Research. St. Paul, MN. Herman, J. L. (1992). Trauma and recovery New York: Basic Books. Herman, J. L. (2003). Introduction: Hidden in plain sight: Clinical observations on prostitution. In M. Farley (Ed.), Prostitution, trafficking and traumat ic stress (pp. 17 32). Binghamton, NY: Haworth. Herrenkohl, T. I., Sousa, C., Tajima, E. A., Herrenkohl, R. C., & Moylan, C. A. (2008). Trauma, Violence, and Abuse, 9, 84 99.

PAGE 223

214 Heyman, R. E., & Slep, A. M. S. (2007). Therapeutic treatment approaches to violent behavior. In D. J. Flannery, A. T. Vazsonyi, & I. D. Waldman (Eds.), The Cambridge handbook of violent behavior and aggression (pp. 602 635). New York: Cambridge University Press. Higgins, G. E., Jennings, W. G., Tewksbury, R., & Gibson, C. L. (2009). Exploring the link between low self control and violent victimization trajectories in adolescents. Criminal Justice and Behavior, 36, 1070 1084. Hilyard, K. L., & Wolfe, D. A. (2002). Child neglect: Developmental issues and outcome. Child Abuse and Neglect, 26, 679 695. Hipwell, A., Keenan, K., Kasza, K., Loeber, R., Stouthamer Loeber, M., Bean, T. (2008). parent al punishment and warmth: A six year prospective analysis Journal of Abnormal Child Psychology, 36, 663 677. Hollist, D. R., Hughes, L. A., & Schaible, L. M. (2009). Adolescent maltreatment, negative emotion, and delinquency: An assessment of general stra in theory and family based strain. Journal of Criminal Justice, 37, 379 387. Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child clinical and pediatric psychol ogy literatures. Journal of Consulting and Clinical Psychology, 65, 599 610. Holmbeck, G. N., (2002). Post hoc probing of significant moderational and mediational effects in studies of pediatric populations. Journal of Pediatric Psychology, Special Issue o n Methodology and Design 27 87 96. Holmbeck, G. N. (2006). Testing for mediation and moderation. Retrieved on February 22, 2010 from http://www.4researchers.org/articles/370 Holtfreter, K., Reising, M. D., Piquero, N. L., & Piquero, A. R. (2010). Low self control and fraud: Offending, victimization, and their overlap. Criminal Justice and Behavior, 37, 188 203. Hotaling, N., Miller, K., & Trudeau, E. (2006) Symposium: Sex for sale: The commercial sexual exploitation of women and girls: A survivor service p rovider's perspective. Yale Journal of Law and Feminism, 181. Hoeve, M., Dubas, J. S., Eichelsheim, V. I., van der Laan, P. H., Smeenk, W., & Gerris, J. R. M. (2009). Journal of Abnormal Child Psychology, 37 : 749 775.

PAGE 224

215 Hu, L., & Bentler, P. M. (1999). Cutof f criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6 1 55. Hughes, D. M. (2002). The use of new communications and information technologies for sexual exploitation of w omen and children. 13 (1), 129 148. Hughes, D. M. (2005). The demand for victims of sex trafficking. Kingston: University of http://www.uri.edu/artsci/wms/hug hes/demand_for_victims.pdf Human Trafficking and Smuggling Center (HTSC), 2005. Fact sheet: Distinctions between human smuggling and human trafficking. Retrieved January 11, 2009, from http://www.usdoj.gov/crt/crim/smuggling_trafficking_facts.pdf Humphre adolescence to young adulthood. Journal of Adolescent Health, 27 419 427. Hunter, J. A., Goodwin, D. W., & Wilson, R. J. (1992). Attributions of blame in child sexual abuse vict ims: An analysis of age and gender influences. Journal of Child Sexual Abuse, 1 (3), 75 89. Hyman, J., & Steward, H. (2004). Agency and action. Cambridge: Cambridge University Press. Inciardi, J. (1993). Some considerations on the methods, dangers, and eth ics of crack house research: Appendix A. In J. Inciardi, D. Lockwood, & A. E. Pettieger (Eds.), Women and crack cocaine (pp. 147 157). New York, NY: Macmillan. International Labour Organization (ILO; n.d.). Commercial sexual exploitation of children and Geneva, Switzerland: International Programme on the Elimination of Child Labour (IPEC). Retrieved February 6, 2010 from http://www.ilo.org/ipec/areas/CSEC/lang -en/index.htm International Labour Organization (ILO; 2005). A global alliance against forced labor. Geneva, Switzerland: International Labour Office. Retrieved May 14, 2010, from http://www.ilo.org/public/english/standards/relm/ilc/ilc93/pdf/rep i b.pdf International Labour Organization (ILO; 2008). Report III: Ch ild labour statistics. Geneva: 18th International Conference of Labour Statisticians. Retrieved February 6, 2010 from http://www.ilo.org/wcmsp5/groups/public/ --dgreports/ --integration/ --stat/documents/meetingdocument/wcms_099577.pdf

PAGE 225

216 Ireland, T. O., Ri vera, C. J., & Hoffman, J. P. (2009). Developmental trajectories, stressful life events, and delinquency. In J. Savage (Ed.), The development of persistent criminality (pp. 90 114). New York: Oxford University Press. Ireland, T., Smith, C., & Thornberry, T (2002). Development issues in the impact of child maltreatment on later delinquency and drug use. Criminology, 40, 59 401. Jaccard, J. (2001). Interaction effects in logistic regression Thousand Oaks: Sage Publications. Jaffee, S. R., Caspi, A., Moffit t, T. E., & Taylor, A. (2004). Physical maltreatment victim to antisocial child: Evidence of an environmentally mediated process. Journal of Abnormal Psychology 113 44 55. James, B. (1994). Handbook for treatment of attachment trauma problems in childre n. New York: The Free Press. James, S. E., Johnson, J., Raghavan, C., & Woolis, D. (2004). "I couldn't go anywhere": social networks, violence and drug abuse. Violence Against Women, 10, 991 1014. Jang, S. J. (2007). Gender differences in strain, negative emotions, and coping behaviors: A general strain approach. Justice Quarterly, 23 523 553. Jang, S. J., & Johnson, B. R. (2003). Strain, negative emotions, and deviant coping among African Americans: A test of general strain theory. Journal of Quantitative Criminology, 19, 79 105. Jankowski, M. K., Leitenberg H., Henning, K., & Coffey, P. (2002). Parental caring as a possible buffer against sexual revictimization in young adult survivors of childhood sexual abuse. Journal of Traumatic Stress, 15 235 244. J effreys, S. (2009). The industrial vagina: The political economy of the global sex trade. New York: Routledge. Jeffreys, S. (2010). Globalizing sexual exploitation: Sex tourism and the traffic of women. In L. Territo, & G. Kirkman (Eds.), International sex trafficking of women & children: Understanding the global epidemic (pp. 367 388) Flushing, NY: Looseleaf Law Publications. Jehu, D. (1988). Beyond sexual abuse New York: John Wiley. Jennings, W. G., Piquero, N. L., Gover, A. R., & Perez, D. M. (2009). G ender and gender/strain hypothesis among a sample of southwestern Mexican American adolescents. Journal of Criminal Justice, 37, 404 417.

PAGE 226

217 Johansson, P., & Kempf Leonard, K. (2009). A gender specific pathway to serious, delinquency. Crime & Delinquency, 55, 216 240. Jones, T. (2005). Internet pedophilia. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kel logg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 789 798). St. Louis, MO: GW Medical Publishing. Kapla n, H. B., & Lin, C. (2000). Deviant identity as a moderator of the relationship between negative self feelings and deviant behavior. Journal of Early Adolescence, 20, 150 177. Karmen, A. (2010). Crime victims: An introduction to victimology (7 th ed.). Belm ont, CA: Wadsworth Cengage Learning. their victimization experiences: A developmental test of revised strain theory and the life course perspective. Violence Against W omen, 6, 633 660. Kaukinen, C., & DeMaris, A. (2005). Age at first sexual assault and current substance use and depression. Journal of Interpersonal Violence, 20 1244 1270. rch on trafficking in persons within and into Europe International Migration, 43, 235 265. Kempe, C. H. (1985). The last word. Child Abuse & Neglect, 9, 155 157. Kempf Leonard, K., & Johansson, P. (2007). Gender and runaways: Risk factors, delinquency, an d Juvenile Justice experiences. Youth and Juvenile Justice, 5, 308 327. Kennedy, M. A., Klein, C., Bristowe, J. T. K., Cooper, B. S., & Yuille, J. C. (2007). street prostitution. Journal of Aggression, Maltreatment & Trauma, 15 (2), 1 19. Kilpatrick, D. G., Ruggiero, K. J., Acierno, R., Saunders, B. E., Resnick, H. S., & Best, C. L. (2003). Violence and risk of PTSD, major depression, substance abuse/dependence, and comorbidity: Re sults from the national survey of adolescents. Journal of Consulting and Clinical Psychology, 71, Kim, K. J., Conger, R. D., Elder, G. H., Jr., & Lorenz, F. O. (2003). Reciprocal influences between stressful life events and adolescent internalizing and externalizing problems. Child Development, 74, 127 134.

PAGE 227

218 Kim, J., & Mueller, C. W. (1978). Factor analysis: Statistical methods and practical issues. Thousand Oaks, CA: Sage Publications. King, N. (1998). Template analysis. In G. Symon, & C. Cassell (Eds.) Qualitative methods and analysis in organizational research (pp.118 34). London: Sage Pu blications. Qualitative Health Research, 12 329 346. Kenzie, R. (1995). Street youth in Los Angeles: Profile of a group at high risk for Human Immunodeficiency Virus infection. Archives of Pediatric Adolescent Medicine 149 513 519. Kipke, M., & Unger, J., O'Connor S., Palmer, R. F., & LaFrance, S. R. (199 7). Street youth, their peer group affiliation and differences according to residential status, subsistence patterns, and use of Services. Journal of Adolescence 32 Klain, E.J., Davies, H.J., & Hicks, M.A., (2001, March). Child Pornography: The criminal j ustice system response, American Bar Association Center on Children and the Law for the National Center for Missing and Exploited Children. Retrieved October 11, 2009, from http://www.popcenter.org/problems/child_pornography/PDFs/Klain_etal_2001.pdf Kline, R. B. (2005). Principles and practice of structural equation modeling (2 nd ed.). New York: Guildford Press. Kolko, D. J. (2002). Child physical abuse. In J. E. B. Myers, L. Berliner, J. Briere, C. T. Hendrix, C. Jenny, & T. A. Reid (Eds.), The APASC hand book on child maltreatment (pp. 21 54). Thousand Oaks, CA: Sage Publications. Knutson, J. F., DeGarmo, D., Koeppl, G., Reid, J. B. (2005). Care neglect, supervisory replication and extension. Child Maltreatment, 10 92 107. Kreston, S. (2000, November/December). Prostituted children: Not an innocent image. The Prosecutor, 34 (6), 37 41. Kreston, S. (2005). Investigation and prosecution of the prostitution of children. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 745 788). S t. Louis, MO: GW Medical Publishing.

PAGE 228

219 behavior and its relationship to family functioning: A review Journal of Child and Family Studies, 18, 259 273. Laczko, F. (2005). Da ta and research on human trafficking. International Migration 43, 5 16. Laczko, F., & Gramegna, M. (2003). Developing better indicators of human trafficking. Brown Journal of World Affairs 10 179 194. Lansford, J. E., Miller Johnson, S., Berlin, L. J., D odge, K. A., Bates, J. E., Pettit, G. S. (2007). Early physical abuse and later violent delinquency: A prospective longitudinal study. Child Maltreatment, 12, 233 245. Lauritsen, J. L., & Laub, J. H. (2007). Understanding the link between victimization and offending: New reflections on an old idea. Crime Prevention Studies 22, 55 75. Lauritsen, J. L., Sampson, R. J., and Laub, J. H. (1991). Link between offending and victimization among adolescents. Criminology 29(2):265 292. Layden, M. A., & Smith, L. W. ( 2005). Adult survivors of the child sexual exploitation industry: Psychological profiles. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensiv e review of child pornography, child prostitution, and Internet crimes against children (pp. 155 177). St. Louis, MO: GW Medical Publishing. Leach, C., Freshwater, K., Aldridge, J., & Sunderland, J. (2001). Analysis of repertory grids in clinical practice. British Journal of Clinical Psychology, 40 225 248. Lebov, K. (2010). Human trafficking in Scotland. European Journal of Criminology, 7, 77 93. Leech, G. (2002). Combating child prostitution in Columbia. Columbia Journal. Retrieved December 17, 2009, fro m http://colombiajournal.org/colombia111.htm Lehti, M., & Aromma, K. (2007). Trafficking in women and children for sexual exploitation: The European context. International Journal of Comparative and Applied Criminal Justice, 31 123 146. Lemert, E. M. (19 72). Human deviance, social problems, and social control (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.

PAGE 229

220 Leserman, J., Li, Z., Drossman, D. A., Toomey, T. C., Nachman, G., & Glogau, L. (1997). Impact of sexual and physical abuse dimensions on health status : development of an abuse severity measure. Psychosomatic Medicine, 59 152 160. Leth, I. (2005). Child sexual exploitation from a global perspective. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 59 84). St. Louis, MO: GW Medical Publishing. Leung, P. (2003). Sex tourism: The case of Cambodia. In T. G. Bauer, & B. McKercher (Eds.), Sex and tourism: Journeys of romance, love, and lust (pp. 181 195). Binghamton, NY: Haworth Hospitality Press. Liang, B., Williams, L. M. & Siegel, J. A. (2006). Relational outcomes of childhood sexual trauma in female s urvivors. Journal of Interpersonal Violence, 21 (1), 42 57. Liao, T. F. (1994). Interpreting probability models: Logit, probit, and other generalized linear models. Thousand Oaks: Sage Publications. Lin, W., Mieczkowski, T. (2010). Subjective strains, con ditioning factors, and juvenile delinquency: General strain theory in Taiwan. Asian Criminology online July 10, 2010. Lloyd, R. (2005). Acceptable victims? Sexually exploited youth in the U.S. Encounter: Education for Meaning and Social Justice 18 (3), 6 18. Lo, C. C., Kim, Y. S., & Church, W. T. (2008). The effects of victimization on drug use: A multilevel analysis. Substance Use and Misuse, 43, 1340 1361. Loeber, R., Hipwell, A., Battista, D., Sembower, M., & Stouthamer Loeber, M. (2009). Intergeneratio nal transmission of multiple problem behaviors: Prospective relationships between mothers and daughters. Journal of Abnormal Child Psychology, 37, 1035 1048. Logan, T. K., Walker, R., & Hunt, G. (2009). Understanding human trafficking in the United States. Trauma, Violence, & Abuse, 10, 3 30. Lynch, M. (1996). Class, race, gender and criminology: Structured choices and the life course. In M. D. Schwartz, & D. Milovanovic (Eds.), Race, gender, and class in criminology (pp. 3 28). New York: Garland Publishing Macmillan, R. (2001). Violence and the life course: The consequences of victimization for personal and social development. Annual Review of Sociology, 27 1 22.

PAGE 230

221 Madonna, P. G., Van Scoyk, S., & Jones, D. P. (1991). Family interactions within incest and nonincest families. American Journal of Psychiatry, 148 p. 46 49 Maher, L. (1997). Sexed work: Gender, race and resistance in a Brooklyn drug market. Oxford: Clarendon Press. Maher, L., Dunlap, E., Johnson, B. D., & Hamid, A. (2006). Homelessness and temp orary living arrangements in the inner city crack culture. In L. F. Alarid & P. Cromwell (Eds.), victimization (pp. 169 181). Los Angeles: Roxbury Publishing. Malesky, L. A. (2005). The use of the Inter net for child sexual exploitation. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 469 487). St. Louis, MO: GW Medical Publishing. Malinosky Rummel, R., & Hansen, D. J. (1993). Long term consequences of childhood physical abuse. Psychological Bulletin, 114, 68 79. Manasse, M. E., & Ganem, N. M. (2009 ). Victimization as a cause of delinquency: The role of depression and gender. Journal of Criminal Justice, 37 371 378. Mason, W. A., & Windle, M. (2002). Gender, self control, and informal social control in adolescence: A test of three models of the cont inuity of delinquent behavior. Youth & Society, 33 479 524. Mazerolle, P. (1998). Gender, general strain, and delinquency: An empirical examination. Justice Quarterly, 15 65 91 McCahill, T., Meyer (Williams), L., & Fischman, A. S. (1979). The aftermath of rape. Lexington, MA: Lexington Books. Meiners developmental psychology and the lessons of Roper should inform daily practice. Temple Law Review, 79, 499 514. Mnard, K.S., Ander son, A. L., & Godboldt, S. M. (2009). Gender differences in intimate partner recidivism: A 5 year follow up. Criminal Justice and Behavior, 36, 61 76. Menard, S. (2002) Short and long term consequences of adolescent victimization (OJJDP/CDC Bulletin No. NCJ 191210). Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Menard, S., & Huizinga, D. (2001). Repeat victimization in a high risk neighborhood sample of adolescents. Youth and Society, 32 447 472.

PAGE 231

222 Mersky, J. P., & Reynolds, A. J. (2007). Child maltreatment and violent delinquency: Disentangling main effects and subgroup effects. Child Maltreatment, 12, 246 258. Messman Moore, T. L., & Brown, A. L. (2006). Risk perception, rape, and sexual revictimization: A prospective study of col lege women. Psychology of Women Quarterly, 30, 159 172. Messman Moore, T. L., Coates, A. A., Gaffey, K. J., & Johnson, C. F. (2008). Sexuality, substance use, and susceptibility to victimization. Journal of Interpersonal Violence, 23, 1730 1746. Messman Mo ore, T. L., & Long, P. J. (2003). The role of childhood sexual abuse sequelae in the sexual revictimization of women: An empirical review and theoretical reformulation. Clinical Psychology Review, 23 (4), 537 571. Messman Moore, T. L., Ward, R. M., & Brown, A. L. (2009). Substance use and PTSD symptoms impact the likelihood of rape and revictimization in college women. Journal of Interpersonal Violence, 24, 499 521. Messerschmidt, J. W. (1993). Masculinities and crime: Critique and reconceptualization of the ory. Lanham, MD: Rowman and Littfield. Milburn, G., Rotheram Borus, M., Rice E., Mallet, S., & Rosenthal, D. (2006). Cross National variations in behavioral profiles among homeless youth. American Journal of Community Psychology 37 63 76. Miller, J. (200 8a). Violence against urban African American girls. Journal of Contemporary Criminal Justice, 24, 148 162. Miller, J. (2008b). Getting played: African American girls, urban inequality and gendered violence. New York: New York Press. Miller, J., & Mullins, C. W. (2006). The status of feminist theories in criminology. In F. T. Cullen, J. P. Wight, & K. R. Blevins (Eds.), Advances in criminological theory, vol. 15: Taking stock: The status of criminological theory (pp. 217 249). New Brunswick, NJ: Transaction Publishers. Miller, J., & Mullins, C. W. (2009). Feminist theories of juvenile delinquency. In M. Zahn (Ed.), The delinquent girl (pp. 30 49). Philadelphia: Temple University Press. Miller Perrin, C. L., & Perrin, R. D. (1999). Child maltreatment: An intro duction. Thousand Oaks, CA: Sage Publications.

PAGE 232

223 Mitchell, K. J., Finkelhor, D., & Wolak, J. (2009). Conceptualizing juvenile prostitution as child maltreatment: Findings from the National Juvenile Prostitution Study. Child Maltreatment OnlineFirst. Moe, A. 116 138. Moffitt, T. E. (1993). Adolescence limited and life course persistent antisocial behavior: A developmental taxonomy. Psychological Rev iew, 100 674 701. Molnar, B., Shade, S., Kral, A., Booth, R., & Watters, J. (1998). Suicidal behavior and sexual / physical abuse among street youth. Child Abuse & Neglect, 22(3), 213 222. Moon, B., Hays, D., & Blurton, D. (2009). General strain theory, k ey strains, and deviance. Journal of Criminal Justice, 37 98 106. Moon, B., Morash, M., McCluskey, C. P., & Hwang, Hye Won (2009). A comprehensive test of general strain theory. Journal of Research in Crime and Delinquency, 46, 182 212. Moon, M. W., Binso n, D., Page Shafter, K., & Diaz, R. (2001). Correlates of HIV risk in random sample of street youths in San Francisco. Journal of the Association of Nurses in AIDS care, 12 (6), 18 27. Moossy, R. (2009). Sex trafficking: Identifying cases and victims. (NCJ No. 225759) NIJ Journal No. 262. Retrieved on May 8, 2009 from http://www.ojp.usdoj.gov/nij/journals/262/sex trafficking.htm Moran, P. B., Vuchinich, S., Hall, N. K. (2004). Associations between types of maltreatment and substance use during adolescence. Child Abuse & Neglect, 28, 565 574. Morash, M. (2006). Understanding gender, crime, and justice. Thousand Oaks, CA: Sage Publications. Morris, R. R. (1964). Female delinquency and relational problems. Social Forces, 43, 82 89. Morris, R. R. (1965). Attit udes toward delinquency by delinquents, non delinquents and their friends. British Journal of Criminology, 5, 249 265. Muehlenhard, C. L., Highby, B. J., Lee, R. S., Bryan, T. S., & Dodrill, W. A. (1998). The sexual revictimization of women and men sexuall y abused as children: A review of the literature. Annual Review of Sex Research, 9, 177 223.

PAGE 233

224 Musto, J. L. (2009). U.S. discourses of human trafficking. 2, 281 287. Muthn, B. O. (1993). Goodness of fit with categorical and other non normal variables. In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 205 243). Newbury Park, CA: Sage. Muthn, B. O., du Toit, S. H. C., & Spisic, D (1997). Unpublished manuscript. Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Los Angeles: University of California. Muthn, L. K., & Muthn, B. O. ( 1998 2007). (4 th ed.). Los Angeles: author. Nagin, D. S., & Paternoster, R. (1991). On the relationship of past to future delinquency. Criminology, 29, 163 189. Nagin, D. S., & Paternoster, R. (2000). Population heterogeneity and state dependence: State of the evidence and directions for future research. Journal of Quantitative Criminology, 16 117 144. National Center for Missing and Exploited Children (NCMEC). What is the prostitution of children? Sexual Exploitation Section. Retriev ed January 11, 2009, from http://www.missingkids.com/missingkids/servlet/PageServlet?LanguageCountry= en_US&PageId=1494 National Center for Missing and Exploited Children (NCMEC; 2002). Female juvenile prostitution: Problem and response. Retrieved October 11, 2009, from http://www.missingkids.com/missingkids/servlet/ResourceServlet?LanguageCou ntry=en_US&PageId=750 National Center for Missing and Exploited Children (NCMEC; 2008). 2008 Annual Report. Retrieved November 11, 2009, from http://www.missingkids.co m/en_US/publications/NC171.pdf National Runaway Switchboard (NRS; n.d.). Third party runaway statistics. Retrieved October 11, 2009, from http://www.1800runaway.org/news_events/third.html Ney, P.,G., Fung, T., & Wickett, A. R. (1994). The worst combination s of child abuse and neglect. Child Abuse and Neglect, 18 p. 705 714. Noell, J., Rohde, P., Ochs, L., Yovanoff, P., Bullard, J., & Black, C. (2000). Incidence and prevalence of Chlamydia, herpes, and viral hepatitis in the homeless adolescent population. Sexually Transmitted Diseases, 28 4 10.

PAGE 234

225 Nunnaly, J. (1978). Psychometric theory New York: McGraw Hill. Office on the United Nations High Commissioner for Human Rights (1989, November). Geneva, Switzerland: Convention on the Rights of the Child. Retrieve d on December 18, 2009 from http://www2.ohchr.org/english/law/pdf/crc.pdf Ousey, G. C., Wilcox, P., & Brummel, S. (2008). Dj vu all over again: Investigating temporal continuity of adolescent victimization. Journal of Quantitative Criminology, 24, 307 33 5. Owen, G., Decker, M., Gerrard, J., Heineman, J., Shelton, E., & Anton, M. (2006). Overview of youth and young adult homelessness in Minnesota. Amherst A. Wilder Foundation, Wilder Research. St. Paul, MN. Retrieved on December 18, 2009 from, http://www.w ilder.org/download.0.html?report=2087. Parent, M. (1996). Turning stones: My days and nights with children at risk. Orlando, FL: Harcourt Brace. Paternoster, R., Dean, C. W., Piquero, A., Mazerolle, P., & Brame, R. (1997). Generality, continuity, and chang e in offending. Journal of Quantitative Criminology, 13 231 266. Patterson, G. R. (1993). Orderly change in a stable world: The antisocial trait as chimera. Journal of Consulting and Clinical Psychology, 61 51 87. Patton, M. Q. (2002). Qualitative resear ch and evaluation methods (3 rd ed.). Thousand Oaks, CA: Sage Publications. Pawson, R., & Tilley, N. (1997). Realistic evaluation. Thousand Oaks, CA: Sage Publications. Payne, A., Gottfredson, D., & Kruttschnitt, C. (2005, November). Girls, schooling, and d elinquency. In M. Zahn (Ed.), Girls Study Group presentation at the meeting of the American Society of Criminology, Toronto, Canada. Retrieved September 13, 2009, from http://girlsstudygroup.rti.org/docs/2005_AS C_C&C.ppt.pdf Peled, E., & Cohavi, A. (2009). The meaning of running away for girl. Child Abuse & Nelglect, 33, 739 749. Perry, D. G., Hodges, E. V. E., & Egan, S. K. (2001). Determinants of chronic victimization by peers: A review and new model of family influence. In J. Juvonen, & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 73 104). New York: Guilford Press.

PAGE 235

226 Piquero, A. R., Farrington, D. P., & Blumstein A. (2003). The criminal career paradigm: Backgrou nd and recent developments. In M. Tonry (Ed.), Crime and justice: A review of research, vol. 30 (pp. 359 506). Chicago: University of Chicago Press. Piquero, A. R., MacDonald, J., Dobrin, A., Daigle, L. E., & Cullen, F. T. (2005). Self control, violent off ending, and homicide victimization: Assessing the general theory of crime. Journal of Quantitative Criminology, 21, 55 71. Pitts, C., & Waller, G. (1993). Self denigratory beliefs following sexual abuse: Association with the symptomatology of bulimic disor ders. International Journal of Eating Disorders, 13 407 410. Linking the micro and macro level sources of self control and criminal behavior over the life course. In J. Savage (Ed.), The development of persistent criminality (pp. 361 373). New York: Oxford University Press. Price, J. L., Hilsenroth, M. J., Petretic Jackson, P. A., & Bonge, D. (2001). A review of individual psychotherapy outcomes for adult survivors of ch ildhood sexual abuse. Clinical Psychology Review, 21 (7), 1095 1121. Priebe, A. S., & Suhr, C. (2005). Hidden in plain view: The commercial sexual exploitation of girls in Atlanta. Atlanta, Georgia: The Atlanta Women's Agenda. Retrieved on January 17, 2009, from http://www.womensagenda.com/Child_Prostitution.pdf. Putnam, F. W. (2006). Impact of trauma on child development. Juvenile and Family Court Journal, 57 (1), 1 11. Raphael, J. (2004). Listening to Olivia: Violence, poverty, and prostitution. Boston, MA: Northeastern University Press. Raphael, J., Reichert, J. A., & Powers, M. (2010). Pimp control and violence: Domestic sex trafficking of Chicago women and girls. Women and Criminal Justice, 20 89 104. Raymond, J. G. (2004). Prostitution on demand: Leg alizing the buyers as sexual consumers. Violence Against Women, 10, 1156 1186. Raymond, J. G., Hughes, D. M., & Gomez, C. J. (2010). Sex trafficking of women in the United States. In L. Territo, & G. Kirkman (Eds.), International sex trafficking of women & children: Understanding the global epidemic (pp. 3 14) Flushing, NY: Looseleaf Law Publications.

PAGE 236

227 Reid, J. A., & Sullivan, C. J. (2009a). A latent class typology of juvenile victims and exploration of risk factors and outcomes of victimization. Criminal J ustice and Behavior, 36, 1001 1024. Reid, J. A., & Sullivan, C. J. (2009b). A model of vulnerability for adult sexual victimization: The impact of attachment, child maltreatment, and scarred sexuality. Violence and Victims, 24, 485 501. Remafedi, G. (1987) Adolescent homosexuality: Psychosocial and medical implications. Pediatrics, 79, 331 337. Rikhye, K., Turka, A. R., Kelly, M. M., Gagne, G. G., Jr., Mello, A., F., Mello, M. F., et al. (2008). Interplay between childhood maltreatment, parental bonding, a nd gender effects: Impact on quality of life. Child Abuse and Neglect, 32, 19 34. Robst, J. (2008). Childhood sexual abuse and the gender wage gap. Economics Letters, 99, 549 551. Romeo, R. D. (2010). Adolescence: A central event in shaping stress reactiv ity. Developmental Psychobiology, 52 244 253. Romer, D. (2010). Adolescent risk taking, impulsivity, and brain development: Implications for prevention. Developmental Psychobiology, 52, 263 276. Roodman, A. A., & Clum, G. A. (2001). Revictimization rates and method variance: A meta analysis. Clinical Psychology Review, 21, 183 204. Roth, S., & Newman, E. (1993). The process of coping with incest for adult survivors. Journal of Interpersonal Violence, 8 (3), 363 377. Rotheram Borus, M., Mahler, K., Koopman, C., & Langabeer. K. (1996). Sexual abuse history and associated multiple risk behavior in adolescent runaways. American Journal of Orthopsychiatry 66 390 400. Rotheram Borus, M., Meyer Bahlburg, H., Koopman, C., Rosario, M., Exner, T., Henderson, R., et a l. (1992). Lifetime sexual behaviors among runaway males and females. The Journal of Sex Research 29, 15 29. Rotheram Borus, M., Meyer Bahlburg, H. Rosario, C. Koopman, C. Haignere, T. Exner, M. Matthieu, R. Henderson & P. Gruen, (1990). Lifetime sexual b ehaviors among predominately minority male runaways and gay/bisexual adolescents in New York. AIDS Education and Prevention, Supplement 34 42. Royal, M. (1998). The pimp game: Instructional guide. Los Angeles: Sharif Publishing.

PAGE 237

228 Rutter, P. (1989). Sex in the forbidden zone: When men in power therapists, doctors, clergy, teachers and other New York: Fawcett Columbine. Saah, T. (2005). The evolutionary origins and significance of drug addition. Harm Reduction Journal, 2(8). Online June 29. Saewyc, E. M., & Edinburgh, L. D. (2010). Restoring healthy developmental trajectories for sexually exploited young runaways girls: Fostering protective factors and reducing risk behaviors. Journal of Adolescent Health, 46, 180 188. Saewyc, E. M., think: Sexually exploited youth in British Columbia. Vancouver, BC: University of British Columbia School of Nursing. Retrieved May 28, 2009, from http://www.nursing.ubc.ca/PDFs/ItsNotWh atYouThink.pdf Safe Place (n.d.). Safe place: Where kids get help fast. Retrieved April 21, 2010 from http://www.nationalsafeplace.org/ Salisbury, E. J., & Van Voorhis, P. (2009). Gendered pathways: A quantitative of incarceration. Criminal Justice and Behavior, 36 541 566. Sampson, R. J., & Laub, J. H. (1992). Crime and deviance in the life course. Annual Review of Sociology, 18 63 84. Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turni ng points through life. Cambridge, MA: Harvard University Press. Sampson, R. J., & Laub, J. H. (2004). A life course theory of cumulative disadvantage and the stability of delinquency. In T. P. Thornberry (Ed.), Advances in criminological theory, vol. 7: Developmental theories of crime and delinquency (pp. 133 161). New Brunswick, NJ: Transaction Publishers. behaviour in the Caribbean. Sociology, 35, 749 764. Snchez Ta ylor, J. (2006). Female sex tourism: A contradiction in terms? Feminist review, 83 42 59. Sartor, C. E., Agrawal, A., McCutcheon, V. V., Duncan, A. E., & Lynskey, M. T. (2008). Disentangling the complex association between childhood sexual abuse and alcoh ol related problems: A review of methodological issues and approaches. Journal of Studies on Alcohol and Drugs, 69, 718 727.

PAGE 238

229 Saunders, B. E., Berliner, L., & Hanson, R. F. (Eds.). (2003). Child physical and sexual abuse: Guidelines for treatment. Charlesto n, SC: National Crime Victims Research and Treatment Center. Savage, J. (2009). Understanding persistent offending: Linking developmental psychology with research on the criminal career. In J. Savage (Ed.), The development of persistent criminality (pp. 3 33). New York: Oxford University Press. Savolainen, J., Sipila, P., Martikainen, P., & Anderson, A. L. (2009). Family, community, and lifestyle: Adolescent victimization in Helsinki. The Sociological Quarterly, 50, 715 783. Scaer, Robert C., M.D. (2001) The body bears the burden: Trauma, dissociation, and disease New York: Haworth Press. Schauer, E., & Wheaton, E. (2006). Sex trafficking into the United States: A literature review. Criminal Justice Review, 31 Scheeringa, M. S., & Zeanah, C H. (2001). A relational perspective on PTSD in early childhood. Journal of Traumatic Stress 14 799 815. Schloenhardt, A., Beirne, G., & Corsbie, T. (2009). Trafficking in persons in Australia: myths and realities. Global Crime, 10 224 247 Schneider, H J. (2001). Victimological developments in the world during the past three decades: A study of comparative victimology Part 2. International Journal of Offender Therapy and Comparative Criminology, 45 539 555. Schreck, C. J. (1999). Victimization and low self control: An extension and test of the general theory of crime. Justice Quarterly, 16, 633 654. Schreck, C. J., & Fisher, B. S. (2004). Specifying the influence of family and peers on violent victimization: Extending routine and lifestyles theorie s. Journal of Interpersonal Violence, 19 1021 1041. Schreck, C. J., Stewart, E. A., & Fisher, B. S. (2006). Self control, victimization, and their influence on risky lifestyles: A longitudinal analysis using panel data. Journal of Quantitative Criminology 22 319 340. Schreck, C. J., Stewart, E. A., & Osgood, D. (2008) A reappraisal of the overlap of violent offenders and victims. Criminology, 46 p. 871 906 Schreck, C. J., Wright, R. A., & Miller, J. M. (2002). A study of individual and situational ante cedents of violent victimization. Justice Quarterly, 19 159 180.

PAGE 239

230 Scott, K. L., Wolfe, D. A., & Wekerle, C. (2003). Maltreatment and trauma: Tracking the connections in adolescence. Child and Adolescent Psychiatric Clinics of North America, 12 211 230. S edlak, A. J., Finkelhor, D., Hammer, H., & Schultz, D. J. (2002). National estimates of missing children: An overview. National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children. Washington, DC: Offices of Juvenile Justice. Senn, T. E., Carey, M. P., Vanable, P. A., Coury Doniger, P., & Urban, M. (2007). Characteristics of sexual abuse in childhood and adolescence influence sexual risk behavior in adulthood. Archives of Sexual Behavior, 36 (5), 637 645. Sethi, A. (2010). Domestic sex trafficking of aboriginal girls in Canada: Issues and implications. In L. Territo, & G. Kirkman (Eds.), International sex trafficking of women & children: Understanding the global epidemic (pp. 27 50) Flushing, NY: Looseleaf Law Publications. Seligman, M. E.P. (1975). On depression, development and death. San Francisco: Freeman. Shadish, W. R., Cook, T. D., Campbell, D. T. (2002). Experimental and quasi experimental designs for generalized causal inference. Boston: Houghton Mifflin. Shakeshaft, C. (2004). E ducator sexual misconduct: A synthesis of the literature. Washington, DC: Prepared for Planning and Evaluation Service, Office of the Undersecretary, U.S. Department of Education Retrieved on April 20, 2010, from www2.ed.gov/rschstat/research/pubs/miscon ductreview/report.doc Shared Hope, International (SHI; 2008). Report on domestic minor sex trafficking: Clearwater, FL. Arlington, VA: Author. Retrieved October 11, 2009, from http://www.sharedhope.org/dmst/clearwater.asp Shelman, E. A., & Lazoritz, S. (20 05). The Mary Ellen Wilson case and the beginning of th century America. Jefferson, NC: McFarland. Sheridan, D. J., & VanPelt, D. (2005). Intimate partner violence in the lives of prostituted adolescents. In Cooper, S. W., R. J. Est es, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 423 435). St. Louis, MO: GW Medical Publishing. Shlyk, S. (2007). Criminal sex trade: System aspects and tendencies. International Journal of Comparative and Applied Criminal Justice, 31 245 268.

PAGE 240

231 Siegel, J. A., & Williams, L. M. (2001a). Risk factors for violent victimization o f women: A prospective study, final report (NCJ 189161). Washington, DC: United States Department of Justice, National Institute of Justice. Siegel, J. A., & Williams, L. M. (2001b). Risk factors for violent victimization of women in a major northwestern city, 1990 1991 and 1996 1997 [Computer File]. ISPSR03052 v1. Camden, NJ: Rutgers University/Wellesley, MA: Wellesley College [producers]. Ann Arbor, MI: Inter university Consortium for Political and Social Research [distributor]. Siegel, J. A., & Williams L. M. (2003). Risk factors for sexual victimization of women: Results from a prospective study. Violence Against Women, 9 (8), 902 930. Silbert, M. H., & Pines, A. M. (1982). Entrance into prostitution. Youth & Society, 13, 471 500. Silverman, J. G., Dec ker, M. R., Gupta, J., Maheshwari, A., Patel, V., & Raj, A. (2006). HIV prevalence and predictors among rescued sex trafficked women and girls in Mumbai, India. Journal of Acquired Immune Deficiency Syndromes, 43, 588 593. Silverman, J. G., Decker, M. R., Gupta, J., Maheshwari, A., Willis, B. M., & Raj, A. (2007). HIV prevalence and predictors of infection in sex trafficked Nepalese girls and women. JAMA, 298 536 542 Simpson, S. S., & Gibbs, C. (2006). Making sense of the intersections. In K. Heimer, & C. Kruttschnitt (Eds .), Gender and crime: Patterns in victimization and offending (pp. 269 302). New York: New York University Press. to jail: Analysing the lives of incarc erated women. The Australian and New Zealand Journal of Criminology, 41, 84 108. Simons, R. L., Chen, Y., Stewart, E. A., & Brody, G. H. (2003). Incidents of discrimination and risk for delinquency: A longitudinal test of strain theory with an African Amer ican sample. Justice Quarterly, 20, 827 854. Simons, R. L., Stewart, E., Gordon, L. C., Conger, R. D., & Elder, G. H. (2002). A test of life course explanations for stability and change in antisocial behavior from adolescence to young adulthood. Criminolog y, 40, 401 434. Simons, R. L., & Whitbeck, L. B. (1991). Sexual abuse as a precursor to prostitution and victimization among adolescents and adult homeless women. Journal of Family Issues, 12, 361 380. Skilbrei, M., & Tveit, M. (2008). Defining traffickin g through empirical work. Gender, Technology and Development, 12 9 30.

PAGE 241

232 Slocum, L. A., Simpson, S. S., & Smith, D. A. (2005). Strained lives and crime: Examining intra individual variation in strain and offending in a sample of incarcerated women. Criminol ogy, 43, 1067 1110. Snyder, H. N. (2000). Sexual assault of young children as reported to law enforcement: victim, incident, and offender characteristic (NCJ 182990). Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice. Snyder, H. N., & Sickmund, M. (2006). Juvenile offenders and victims: 2006 National Report. OJJDP National Report. Washington DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Solomon, E. P., & Heide, K. M (1999). Type III trauma: Toward a more effective conceptualization of psychological trauma. International Journal of Offender Therapy and Comparative Criminology, 43 (2), 202 210. Solomon, E. P., & Heide, K. M. (2005). The biology of trauma: Implications for treatment. Journal of Interpersonal Violence, 20 (1), 51 60. Solorio, M. R., Milburn, N., Andersen, R., Trifskin, S., & Rodriguez, M. (2006). Emotional distress and mental health service use among urban homeless adolescents. The Journal of Behavioral H ealth Services & Research 33 381 393. Spear, L. P., & Varlinskaya, E. I. (2010). Sensitivity to ethanol and other hedonic stimuli in an animal model of adolescence: Implications for prevention science? Developmental Psychobiology, 52, 236 243. Spidel, A. Greaves, C., Cooper, B. S, Herv, H. F., Hare, R. D., & Yuille, J. C. (2007). The psychopath as pimp. Canadian Journal of Police and Security Services, 4, 205 211. SPSS Inc. (2005). SPSS Base 14.0 for Windows user's guide Chicago: SPSS Inc. Stark, C., & Hodgson, C. (2003). Sister oppressions: A comparison of wife battering and prostitution. In M. Farley (Ed.), Prostitution, trafficking and traumatic stress (pp. 17 32). Binghamton, NY: Haworth. Steffensmeier, D., & Allen, E. (1996). Gender and crime: To ward a gendered theory of female offending. Annual Review of Sociology, 22, 459 487. gap: Statistical assessment of diverse sources. In M. Zahn (Ed.), The delinquent gir l (pp. 50 83). Philadelphia: Temple University Press.

PAGE 242

233 Steinberg, L. (2010). A dual system model of adolescent risk taking. Developmental Psychobiology, 52, 216 224. Steinberg, L. & Scott, E. S. (2003). Less guilty by reason of adolescence: Developmental i mmaturity, diminished responsibility, and the juvenile death penalty. American Psychologist, 58 (12), 1009 1018. Stevens, D. J. (1998). Inside the mind of a serial rapist Bethesda, MD: Austin & Winfield Publishers. Stewart, E. A., Elifson, K. W., & Sterk, C. E. (2004). Integrating the general theory of crime into an explanation of violent victimization among female offenders. Justice Quarterly, 21, 159 182. between prior sex ual abuse and reactions to sexual harassment. Psychology, Public Policy, and Law, 1 64 95. Stokes, D. J., Dritschel, B. H., & Bekerian, D. A. (2008). Semantic and episodic autobiographical memory recall for memories not directly associated with childhood sexual abuse. Journal of Family Violence, 23, 429 435. Stone, S. (2009, October 27). 8 child prostitutes rescued in Seattle. KOMO TV. Retrieved on January 8, 2010, from http://current.com/1suci4c Straus, M. A., & Kantor, G. K. (2005). Definition and measu rement of neglectful behavior: some principles and guidelines. Child Abuse & Neglect, 29 (1), 19 29. Streiner, D. L., & Norman, D. L. (1989). Health measurement scales: A practical guide to their development and use. New York: Oxford University Press. Stud nicka, A. C. S. (2010). Corruption and human trafficking in Brazil: Findings from a multi modal approach. European Journal of Criminology, 7, 29 43. Stuewig, J., & McCloskey, L. A. (2005). The relation of child maltreatment to shame and guilt among adoles cents: Psychological routes to depression and delinquency. Child Maltreatment, 10, 324 336. Sullivan, S. (2000). Child neglect: Current definitions and models: A review of child neglect research, 1993 1998. National Clearinghouse on Family Violence. Ottaw a, Ontario: Author. Sullivan, T. N., Farrell, A. D., & Kliewer, W. (2006). Peer victimization in early adolescence: Association between physical and relational victimization and drug use, aggression, and delinquent behaviors among urban middle school stude nts. Development and Psychopathology

PAGE 243

234 Summit, R. C. (1983). The child sexual abuse accommodation syndrome. Child Abuse and Neglect, 7 177 193. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4 th ed.). New York: Harpe rCollins. Taylor, M, & Quayle, E. (2003). Child pornography: An Internet crime Hove, England: Brunner Routledge. Taylor, M., & Quayle, E. (2005). Abusive images of children and the Internet: Research from the COPINE Project. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 257 275). St. Louis, MO: GW Medical Publishing. Taylor, T. J., Freng, A., Esbensen, F., & Peterson, D. (2008). Youth gang membership and serious violent victimization. Journal of Interpersonal Violence, 23, 1441 1464. Terry, K. J., & Tallon, J. (2004). Child sexual abuse: A review o f the literature New York: John Jay College Research Team. Retrieved April 19, 2010, from http://www.usccb.org/nrb/johnjaystudy/litreview.pdf Tittle, C. R., Broidy, L., Gertz, M. G. (2008). Strain, crime, and contingencies. Justice Quarterly, 25 283 312. Thornberry, T. (1987). Toward an interactional theory of delinquency. Criminology, 25, 863 891. Trafficking Victims Protection Act of 2000 (TVPA), 22 U.S.C. §7105(b)(1)(E)(u)[2000]) Trafficking Victims Protection Reauthorization Act of 2003 (TVPRA), Pub. L. No. 108 193, 117 Stat. 2875 (2003), (current version at 22 U.S.C. §§7101 7106 [2008]). Trafficking Victims Protection Reauthorization Act of 2005 (TVPRA), Pub. L. No. 109 164, 119 Stat. 3558 (2006), (current version at 22 U.S.C. §§7101 7106 [2008]). Tra fficking Victims Protection Reauthorization Act of 2008 (TVPRA), Pub. L. No. 110 457, 122 Stat. 5055 (2008), (current version at 22 U.S.C. §§7101 7106 [2008]). Trifiletti, C. D. (2005). Investigating Internet child exploitation cases. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 609 634). St. Loui s, MO: GW Medical Publishing.

PAGE 244

235 Turner, H. A, & Butler, M. (2003). Direct and indirect effects of childhood adversity on depressive symptoms in young adults. Journal of Youth and Adolescence, 32 89 103. Turner, H. A., Finkelhor, D., & Ormrod, R. (2006). The effect of lifetime victimization on the mental health of children and adolescents. Social Science & Medicine, 62, Turner, H. A., Finkelhor, D., & Ormrod, R. (2007). Family structure variation in patterns and predictors of child victimization. American Journal of Orthopsychiatry, 77 282 295. Tyldum, G., & Brunovskis, A. (2005). Describing the unobserved: Method ological challenges in empirical studies on human trafficking. International Migration, 43 Tyler, K. A., & Beal, M. R. (2010). The high risk environment of homeless young adults: Consequences for physical and sexual victimization. Violence and Victims, 25, 101 115. Tyler, K. A., Hoyt, D. R., Whitbeck, L. B., & Cauce, A. M. (2001). The effects of high risk environments on the sexual victimization of homeless and runaway youth. Violence and Victims, 16, 441 455. Tyler, K. A., & Johnson, K. (2006a ). Trading Sex: Voluntary or coerced? The experience of homeless youth. The Journal of Sex Research, 43 208 217. Tyler, K. A., & Johnson, K. (2006b). A longitudinal study of the effects of early abuse on later victimization among high risk adolescents. Vi olence & Victims, 21, 287 306. Tyler, K. A., Whitbeck, L. B., Chen, X., & Johnson, K. (2007). Sexual health of homeless youth: Prevalence and correlates of sexually transmissible infections. Sexual Health, 4, 57 61. Unger, J., Simon, T., Newman, T., Montg omery, S., Kipke, M., & Albornoz, M. (1998). Early adolescent street youth: An overlooked population with unique problems and service needs. Journal of Early Adolescents, 18, 325 348. United Nations (1995, September 20). Sale of children, child prostitutio n and child pornography. Note by the Secretary General. A/50/456. Retrieved December 17, 2009 from http://daccess dds ny.un.org/doc/UNDOC/GEN/N95/285/34/PDF/N9528534.pdf?OpenElement

PAGE 245

236 U.S. Bureau of the Census (1977). Money income in 1975 of families and per sons in the United States Current Population Reports, Consumer Income, Series P 60, No. 105. Retrieved May 11, 2010 from http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/000001 9b/80/31/87/ea.pdf U.S. Bureau of the Census (2009). Income poverty, and health insurance coverage in the United State: 2008. Retrieved May 11, 2010, from http://www.census.gov/prod/2009pubs/p60 236.pdf U.S. Department of Health and Human Services (HHS; n.d.a). Anti trafficking in persons. Office of Refuge Resett lement. Retrieved December 17, 2009, from http://www.acf.hhs.gov/programs/orr/programs/anti_trafficking.htm U.S. Department of Health and Human Services (HHS; n.d.b). Rescue and Restore Section. Fact sheet on sex trafficking Retrieved January 11, 2009, fr om http://www.acf.hhs.gov/trafficking/about/fact_sex.html U.S. Department of Health and Human Services (HHS; 2008). Efforts to Serve Children Retrieved on February 6, 2010 from http:/ /oig.hhs.gov/oei/reports/oei 07 06 00290.pdf U.S. Department of Health and Human Services (HHS; 2010). Child Maltreatment 2008. Washington, DC: Administration on Children, Youth and Families. U.S. Department of Justice (DOJ; n.d.). Model State Anti Traff icking Criminal Statute Retrieved December 22, 2009 from http://www.usdoj.gov/crt/crim/model_state_law.pdf U.S. Department of State (DOS; 2008). Major forms of trafficking in persons. Trafficking in Persons Report 2008 Retrieved November 29, 2009 from h ttp://www.state.gov/g/tip/rls/tiprpt/2008/105377.htm U.S. Department of State (DOS; 2009). Trafficking in Persons Report 2009 Retrieved July 4, 2009, from http://www.state.gov/g/tip/rls/tiprpt/2009/index.htm Valentine, J. C., & Cooper, H. (2003). Effect s ize substantive interpretation guidelines: Issues in the interpretation of effect sizes. Washington, DC: What Works Clearinghouse. Van Bruggen, L. K., Runtz, M. G., & Kadlec, H. (2006). Sexual revictimization: The role of sexual self esteem & dysfunctional sexual behaviors. Child Maltreatment, 11 (2), 131 145.

PAGE 246

237 van der Kolk, B. A., McFarlene, C., & Weisaeth, L. (Eds.). (2007). Traumatic stress: The effects of overwhelming experience on mind, body, and society. New York: The Guilford Press. Vieth, V. I. (2005 ). Unto the third generation: A call to end child abuse in the United States within 120 years. Journal of Aggression, Maltreatment, and Trauma, 12 5 54. Vieth, V. I., & Ragland, E. R. (2005). Shadow children: Addressing the commercial exploitation of chil dren in rural America. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I. Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet cri mes against children (pp. 1027 1039). St. Louis, MO: GW Medical Publishing. Venkatraman, B. A., Jacob, E. R., & Henley, D. B. (2005). The hidden truth of involuntary servitude and slavery. In Cooper, S. W., R. J. Estes, A. P. Giardino, N. D. Kellogg & V. I Vieth (Eds.) Medical, legal & social science aspects of child sexual exploitation: A comprehensive review of child pornography, child prostitution, and Internet crimes against children (pp. 927 946). St. Louis, MO: GW Medical Publishing. Vermeersch, H., control and adolescent risk taking. Deviant Behavior, 29, 690 725. Wagner, L., Carlin, L., Cauce, A., & Tenner, A. (2001). A snapshot of homeless youth in Seattle: Their characteristics, be haviors and beliefs About HIV protective strategies. Journal of Community Health, 26, 219 232. Walsh, C., MacMillan, H., & Jamieson, E. (2002). The relationship between parental psychiatric disorder and child physical and sexual abuse: Findings from the On tario Health Supplement. Child Abuse and Neglect, 26, 11 22. Waller, G., Ruddock, A., & Pitts, C. (1993). When is sexual abuse relevant to bulimic disorders? The validity of clinical judgments. European Eating Disorders Review, 1 (3), 143 151. Whiffen, V. E., & MacIntosh, H. B. (2005). Mediator of the link between childhood sexual abuse and emotional distress. Trauma, Violence, & Abuse, 6, 24 39. Whitbeck, L, Hoyt, D., Yoder, K., Cauce, A., & Paradise, M. (2001). Deviant behavior and victimization among hom eless and runaway adolescents. Journal of Interpersonal Violence, 16, 1175 1204. White, J. W. (2009). A gendered approach to adolescent dating violence: conceptual and methodological issues. Psychology of women quarterly, 33 1 15

PAGE 247

238 Whitehead, J. W. (2008, dirty little secret. Retrieved October 14, 2009, from http://www.huffingtonpost.com/john w whitehead/children of the night chi_b_115348.html Widom, C. S. (2000). Childhood victimization and t he derailment of girls and women to the criminal justice system. In B. Richie, K. Tsenin, & C. Widom (Eds.), Research on women and girls in the criminal justice system (pp. 27 36). Washington, DC: National Institute of Justice. Widom, C. S. & Kuhns, J. B. (1996). Childhood victimization and subsequent risk for promiscuity, prostitution, and teenage pregnancy: A prospective study. American Journal of Public Health, 86 1607 1610. Wikstrm, P. H. (2008). In search of causes and explanations of crime. In R. D King, & E. Wincup (Eds.), Doing research on crime and justice (2 nd ed.) (pp. 117 140). New York: Oxford University Press. Wikstrm, P. H., & Sampson, R. J. (2006). The explanation of crime: Context, mechanisms, and development. New York: Cambridge Univ ersity Press. memories of child sexual abuse. Journal of Consulting and Clinical Psychology 62, 1167 1176. Williams, L. M. (1995). Recovered memories of abuse in women with documented child sexual victimization histories. Journal of Traumatic Stress 8, 649 673. Williams, L. M., & Frederick, M. E. (2009). Pathways into and out of commercial sexual exploitation: Understanding and responding to sexually exploited teens. Lowel l, MA: University of Massachusetts Lowell. Williams, P. (2008). Trafficking in women: The role of transnational organized crime. In S. Cameron & E. Newman (Eds.), Trafficking in Humans: Social, cultural and political dimensions. New York: United Nations Un iversity Press. Williamson, C., & Cluse Tolar, T. (2002). Pimp controlled prostitution: Still an integral part of street life. Violence Against Women, 8, 1074 1092. Wilson, H. W., & Widom, C. S. (2010). The role of youth problem behaviors in the path from child abuse and neglect to prostitution: A prospective examination. Journal of Research on Adolescents, 20, 201 236. Wilson, J. M., & Dalton, E. (2008). Human trafficking in the Heartland. Journal of Contemporary Criminal Justice, 24, 296 313.

PAGE 248

239 Wilson, J.Q., & Herrnstein R. J, (1985). Crime and human nature. New York: Touchstone. Winterdyk, J., & Reichel, P. (2010). Introduction to special issue Human trafficking: Issues and perspective. European Journal of Criminology, 7, 5 10. Wittebrood K., & Nieuw the effects of previous victimization and patterns of routine activities. Journal of Research on Crime and Delinquency, 37 91 122. Yackoub, E., & Moss, W. (2007). Technical Report 2: Case loads in Child and Family Services. Retrieved on May 16, 2010 from http://www.community.nsw.gov.au/docswr/_assets/main/documents/caseloads_r eport.pdf Yates, G., MacKenzie, R., Pennbridge, J., & Cohen, E. (1988). At risk profile comparison of runaway and no n runaway youth. American Journal of Public Health, 78 820 821. Yoder, K., Whitbeck, L., & Hoyt, D. (2001). Event history analysis of antecedents to running away from home and being on the street. American Behavioral Scientist, 45, 51 65. Yu, C. Y. (2002 ). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes (Doctoral dissertation, University of California, Los Angeles, 2002). Retrieved March 3, 2010 from http://www.statmodel.com/download/Yudissert ation.pdf. Zhang, S. X. (2009). Beyond the 'Natasha' story a review and critique of current research on sex trafficking. Global Crime, 10 178 190.

PAGE 249

About the Author Joan A. Reid earned her Bachelor of Arts in Psychology and Master of Arts in Rehab ilitation and Mental Health Counseling from the University of South Florida. J oan is a licensed mental health counselor and certified rehabilitation counselor with professional experience counseling individuals recovering from the effects sexual abuse or assault. Her research concerns are human trafficking, child abuse, sexual violence, and rehabilitation of crime victims. As a member of the coalition against human trafficking, Joan has been active in local efforts to combat trafficking. She is engaged i n researching and reporting on the local issues facing minors entrapped in sex trafficking.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004567
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Reid, Joan.
0 245
A pathway to child sex trafficking in prostitution :
b the impact of strain and risk-inflating responses
h [electronic resource] /
by Joan Reid.
260
[Tampa, Fla] :
University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Dissertation (PHD)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Victims of child sex trafficking in prostitution in the United States are often overlooked, misidentified, and among the most underserved type of child victim of crime. The majority of previous research on child sex trafficking has been conducted without a theoretical framework or reliable sampling methods. In this study, a schematic composed of a series of stepping-stones from childhood abuse to prostitution, which has been described by gendered pathways researchers, served as a sensitizing template for the study's development of a strain-reactive pathway into child sex trafficking. Agnew's general strain theory provided the primary theoretical basis for the proposed pathway, supplying both explanations of the generative factors of the pathway and the mechanisms operating within the life trajectory terminating in child sex trafficking in prostitution. Based on this theoretical framework, this study utilized structural equation modeling to examine the pathway by investigating the effects of caregiver strain, child maltreatment, and risk-inflating responses to strain on vulnerability to victimization in child sex trafficking in prostitution. Four structural equation models, incorporating different forms of child maltreatment, were assessed using data from a matched sample of 174 minority females who were residents of one U.S. city and participated in a longitudinal study on the effects of child sexual abuse. Findings show that the occurrence of child maltreatment including child neglect, child physical abuse, and juvenile sexual victimization increased with caregiver strain. Consequentially, neglected and abused children were more likely to have engaged in the risk-inflating responses of running away and earlier initiation of drug or alcohol use, and they also reported higher levels of relational shame. Both running away and early initiation of substance use impacted vulnerability to victimization in child sex trafficking in prostitution. Lastly, implications of the findings related to protection and intervention strategies that are projected to obstruct the progression of minors along the analytically identified pathway into child sex trafficking in prostitution are presented for criminal justice professionals, child protection investigators, and social service providers.
590
Advisor: Christine Sellers, Ph.D.
653
Human trafficking
Commercial sexual exploitation of children
Child abuse
Runaways
Family strain
690
Dissertations, Academic
z USF
x Criminology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.4567