Wolak, Janis --- "Technology-facilitated Organized Abuse: An Examination of Law Enforcement Arrest Cases" [2015] IntJlCrimJustSocDem 15; (2015) 4(2) International Journal for Crime, Justice and Social Democracy 18

	Home \| Databases \| WorldLII \| Search \| Feedback International Journal for Crime, Justice and Social Democracy

Abstract

This paper looks at cases of organized abuse (that is, two or more offenders working in concert and having two or more victims, not solely familial) reported by law enforcement respondents during the three waves of the National Juvenile Online Victimization (NJOV) Study (n=29). The NJOV Study collected data from a national US sample of law enforcement agencies about technology-facilitated crimes ending in arrest at three time points: mid-2000 to mid-2001, 2005 and 2009. The paper reports on the prevalence of technology-facilitated organized abuse ending in arrest, contexts of cases and characteristics of offenders and victims.

Keywords

Organized abuse; Internet-related sex crime; child sexual abuse.

Introduction

This paper undertakes an empirical examination of a set of cases of organized sexual abuse collected during research with US law enforcement agencies about arrests for technology-facilitated crimes. Organized sexual abuse has been defined as two or more offenders acting in a coordinated way to sexually abuse multiple children, excluding cases involving single households or offenders who are unaware of each other (Salter and Richters 2012). The topic of multiple perpetrator/multiple victim sexual abuse has been controversial. It is contrary to the conventional conceptualization of child sexual abuse as a one-on-one crime by a single furtive offender. Confirmed cases have been systematically documented by researchers only occasionally (see, for example, Wild 1989).

Although there is a considerable body of research about perpetrators of sexual offenses against children, very little of it addresses the possible motivations of sexual offenders that collaborate with other perpetrators to commit sex crimes in pairs or larger groups (Harkins and Dixon 2010). There is a body of research based on case studies or small clinical samples of victims of organized abuse (Salter 2010; Snow and Sorensen 1990). These, of course, rely on victim accounts to describe case characteristics and dynamics. Some researchers have surveyed mental health providers and child protective service workers about encountering such victims, finding that small but significant numbers of practitioners have worked with victims of organized abuse.

However, even as the existence of organized abuse was debated in the media and among psychologists, law enforcement agencies documented cases of ‘sex rings’ that committed organized abuse (Lanning 1992; Wild 1989), and such cases continue to be documented (Child Exploitation and Online Protection Centre, 2013; Gallagher, 2007). In recent years, media stories have described cases of large Internet-related sex rings with numerous victims that appear to involve organized abuse (for example, Davidson 2014; Mullen 2014; Romero 2010). These cases have been uncovered in law enforcement investigations of technology-facilitated sex crimes (for example, digital photography used to produce child pornography, online circulation of child pornography, and sex offenders using the Internet to meet and seduce young adolescents). Technology-facilitated sex crimes often include extensive documentation of the abuse perpetrated by offenders, including pictures, videos and logs of online conversations. This type of electronic evidence provides powerful evidence of what occurred between perpetrators and victims and makes cases easier to prosecute and possibly less stressful for victims by reducing the need for victim testimony. However, although law enforcement cases can provide details about incidents of organized abuse, little research has systematically examined such cases.

The purpose of this paper is to describe cases of organized abuse found during three waves of a longitudinal study of technology-facilitated child sexual exploitation crimes conducted between 2000 and 2009. We know of no other research that systematically examines characteristics of law enforcement cases that include electronic evidence. We examine (1) the number of organized abuse cases and whether they appeared to increase or decrease across the three waves of the study; (2) the types of cases, particularly whether they were familial or extra-familial; (3) the dynamics of cases and characteristics of offenders and victims; and (4) comparison of characteristics of cases of organized abuse to other cases of technology-facilitated child sexual abuse. We also discuss the limitations imposed by our methodology, including incomplete information about the characteristics of offenders and victims. The data were collected as part of the National Juvenile Online Victimization Study, which gathered information from a national sample of US federal, state and local law enforcement agencies about technology-facilitated child sexual exploitation crimes that ended in arrest.

Methods

The National Juvenile Online Victimization (NJOV) Study is a longitudinal study of a national sample of US law enforcement agencies that has collected three waves of data about the number and characteristics of arrests for technology-facilitated sex crimes against minors. In Wave 1, we surveyed a national sample of 2,574 state, county, and local law enforcement agencies by mail asking if they had made arrests in technology-facilitated child pornography or sexual exploitation cases between 1 July 2000 and 30 June 2001. Then detailed telephone interviews were conducted with investigators about specific cases (unweighted n=612). Using the same procedures, we collected data for Wave 2 from a national sample of 2,598 agencies about cases ending in arrest in calendar year 2006 (unweighted n for completed interviews = 1,051) and data for Wave 3 from a national sample of 2,653 agencies about arrests in 2009 (unweighted n for completed interviews = 1,299).

Sample

The NJOV Study sample of agencies was designed to yield a nationally representative sample of technology-facilitated child sexual exploitation cases that ended in arrest. We used a stratified sample of agencies because such cases do not occur with equal probability among the more than 15,000 US law enforcement agencies. The sample was divided into three frames. The first frame consisted of agencies mandated to investigate technology-facilitated child sexual exploitation crimes, including federal agencies and federally-funded Internet Crimes Against Children (ICAC) Task Forces (Wave 1, 1st frame, n = 75; Wave 2, 1st frame, n=101, Wave 3, 1st frame, n=176). We did not sample this frame but surveyed all agencies. The second frame consisted of law enforcement agencies with staff that had received training in investigating technology-facilitated child sexual exploitation cases prior to Wave 1. These were identified through lists provided by the training agencies. About half of second frame agencies were randomly selected to participate in the study (Wave 1, 2nd frame, n=833; Wave 2, 2nd frame, n=832, Wave 3, 2nd frame, n=815). The third frame consisted of all other local, county and state law enforcement agencies in the US, approximately 13,586. About 12 per cent of 3rd frame agencies were randomly selected for the sample (Wave 1, 3rd frame, n=1,666; Wave 2, 3rd frame, n=1,665, Wave 3, 3rd frame, n=1,662). (Differences in the numbers of agencies in specific frames between Waves 1, 2 and 3 reflect changes in status among agencies, for example, as new ICAC Task Forces were funded.)

We designed a sampling procedure for case-level telephone interviews based on the number of cases reported by an agency, so that we would not unduly burden respondents in agencies with many cases. If an agency reported between one and three cases, we conducted follow-up interviews for every case. For agencies that reported more than three cases, we conducted interviews for all cases that involved identified victims and sampled other cases. (By ‘identified victims’ we mean those who were identified and contacted by law enforcement in the course of investigations. We distinguish such victims from many victims pictured in child pornography whose identities are unknown.) For agencies with between four and fifteen cases, approximately half of the cases that did not have identified victims were randomly selected for telephone interviews. In agencies that reported more than fifteen cases, approximately one-quarter of the cases with no identified victims were randomly selected. In some agencies, we could not find out which cases had identified victims, so we sampled from all cases, using the procedures described above.

The data in this paper are based on an examination of a subsample of 1,136 cases involving arrests for technology-facilitated child sexual crimes with identified victims (Wave 1, unweighted n=261; Wave 2, unweighted n=318; Wave 3, unweighted n=557). To be eligible for the NJOV Study, cases had to (1) have victims younger than 18 years of age; (2) involve arrests between 1 July 2000 and 30 June 2001 (Wave 1), in 2006 (Wave 2) or 2009 (Wave3); and (3) be technology-facilitated (that is, involve Internet use or cell phones or other handheld devices used for texting or storage or transmission of photographs). Table 1 provides details about the dispositions of the Waves 1, 2 and 3 agency-level mail survey and case-level telephone interview samples.

Study procedures

The mail survey asked respondents whether, during the relevant time period, their agency made ‘ANY ARRESTS in cases involving the attempted or completed sexual exploitation of a minor, AND at least one the of the following occurred: a. The offender and the victim first met on the Internet, or b. The offender committed a sexual offense against the victim on the Internet, regardless of whether or not they first met online’. The second Wave of the study included an additional sub-question which asked about arrests when ‘the offender was involved in prostitution or other form of commercial sexual exploitation of a minor that involved the Internet in any way’.

When respondents had such cases, we asked them to list the case number or other reference, and the name of the key investigating officer or most knowledgeable person for each case they reported. Then trained interviewers contracted respondents to schedule telephone interviews at their convenience. Interviews were conducted using a computer-assisted telephone interviewing system. To increase the reliability of responses, we asked investigators to have and refer to case files during interviews. Study procedures were approved by the University of New Hampshire Human Subjects Review Board and complied with all Department of Justice research privacy mandates.

Table 1: Final dispositions and response rates for the 3 waves of the National Juvenile Online Victimization (NJOV) Study

		*NJOV1*		*NJOV2*		*NJOV3*
*Study sample and cases qualification*		*Number*	%	*Number*	%	*Number*	%
Number of agencies in sample	2,574			2,598		2,653
No jurisdiction	65			282		190
Eligible agencies	2,509			2,316		2,463
Responded to mail survey	2,205		88	2,028	87	2,128	86
Reported cases	383		15	458	20	590	24
Number of cases reported	1,723			3,322		4,010
Not selected for sample	646		37	1,389	42	1,522	38
Ineligible	281		16	276	8	459	11
Total number of cases in sample	796			1,657		2,029
Non-responders	101		13	446	27	471	23
Refusals	25		3	118	7	159	8
Invalid or duplicate cases	40		5	30	2	100	5
Completed Interviews	612		79	1,051	64	1,299	64
Did not involve an identified victim^a	352			733		742
Involved an identified victim	261			318		557
Involved organized abuse	10			11		8

Note: NJOV1 arrests occurred between July 1, 2000 and June 30, 2001; NJOV2 arrests in 2006; NJOV3 arrests in 2009

^aThese cases involved possession or distribution of child pornography or solicitation of undercover investigators posing online as minors.

Measures

Initial telephone interview questions confirmed whether a case involved an identified victim (that is: ‘Did this case involve a victim who was identified and contacted by the police?’) and determined whether the case included multiple victims or multiple offenders. If a case had multiple victims, questions about victim and crime characteristics referred to a primary victim chosen based on the following algorithm: first, the victim who most directly used the Internet; if more than one, the victim who was most seriously victimized; if more than one, the youngest. A similar algorithm was used in cases with multiple offenders (that is, most directly used the Internet, committed most serious crime, youngest). The algorithm defaulted to the youngest victims and offenders when technology use and victimization were equal because we were particularly interested in the characteristics of those groups. Telephone interviewers also asked respondents for an overview of what occurred in each case and wrote descriptive narratives after each interview.

To identify cases of organized abuse, first we listed the cases from all three waves of the NJOV Study that involved both multiple victims and multiple offenders. Then we examined the case narratives to ascertain which of those cases appeared to meet the criteria for organized abuse (that is, two or more offenders acting in concert to abuse two or more victims, excluding cases that solely included family members). We also excluded cases that did not involve multiple offender contact sexual abuse. These included some photography-only child pornography production cases and cases where only one offender participated in contact abuse; for example a contact abuser produced child pornography and gave it to associates who were not present during the contact abuse. However, we defined participation in contact abuse to encompass online sex rings in which offenders watched live streaming videos of victims being sexually abused because such cases often appeared to include verbal communications from distant participants.

Telephone interviews covered personal characteristics of primary victims and offenders (for example, sex, age, household status); histories of arrests and related problems (for example, substance abuse, violence); context of crimes (for example, victim-offender relationships, sexual offenses committed), including whether crimes included production of child pornography. Measures were based on questions developed for the NJOV Study through interviews, pretesting and piloting with law enforcement before Wave 1 data collection.

Weighting

Weights were constructed to reflect the complex sample design and reduce bias resulting from variations in selection probabilities, response propensity and non-response (Kish 1992). The weights corrected for the probability that an agency or a case would be selected for the sample, given that the sample was stratified (some agencies had a higher probability of selection) and based on the sampling procedure we used for cases (number of cases reported by an agency and whether the case involved an identified victim). Cases that had lower probabilities of selection were given greater weight; thus the weighted distribution of cases reflects the population of cases from which the sample was selected.

NJOV Study weights and variance estimation variables were designed to support analysis in Stata survey data software, which allows users to specify multi-stage sample design information, and account for the variation from both agency and case selection stages. Each case was given a weight to account for its probability of selection to both the mail survey and telephone interview samples. The sampling weights were adjusted for agency non-response, case-level non-response, duplication of cases among agencies, and arrests by one federal agency that did not participate in case-level interviews. Primary sampling weight units were created to account for clustering of cases within each of the three sampling frames. Stratification weights were computed to reflect the differing sampling strategies for each frame. Finally, finite population correction factors accounted for the sample being conducted without replacing ineligible cases. More detailed information is available in the NJOV Study Methodology report, posted online at http://www.unh.edu/ccrc/internet-crimes/papers.html.

Statistical analysis

We used weighted data to estimate numbers of arrest cases involving organized abuse and to examine case characteristics by showing frequencies. We used chi square cross-tabulations to compare (1) familial versus extra-familial cases of technology-facilitated organized abuse ending in arrest and (2) all arrest cases of technology-facilitated organized abuse to arrest cases involving other types of technology-facilitated sexual abuse and exploitation. Chi square comparisons were performed on weighted data. Because applying standard chi square calculations to weighted data could result in the underestimation of standard errors, we used Stata SE 11.2 statistical software, which takes into account the variations in selection probabilities inherent in stratified samples when it performs chi square tests.

Results

Number of cases of technology-facilitated organized abuse ending in arrest

Across the three waves of the study, 57 cases involved both multiple offenders and multiple victims. However, approximately half of these cases did not qualify as organized abuse (n=28) because they involved familial only offenders and victims, serial victims, production of child pornography but no contact sexual abuse, or one offender only perpetrated contact sexual abuse. Also, four cases were excluded that could not be classified accurately because they lacked clear information about the roles of additional offenders.

Overall, 29 cases were classified as technology-facilitated organized abuse. Using weighted nationally representative data, we estimated the numbers of technology-facilitated organized abuse cases that ended in arrest during each wave of the NJOV Study (NJOV1 timeframe was 1 July 2000 to 30 June 2001, NJOV2 was 2006 and NJOV3 was 2009). There were an estimated 27 such cases during Wave 1 (95% confidence interval (CI) = 7 to 47), an estimated 44 cases during Wave 2 (95% CI = 20 to 67), and an estimated 28 cases during Wave 3 (95% CI = 4 to 53) (Table 2). The overlapping 95% confidence intervals indicate the number of such cases remained stable throughout the three waves of the study. These cases accounted for a small proportion of technology-facilitated arrest cases involving identified victims (3% in Waves 1 and 2, 1% in Wave 3).

Table 2: Prevalence of technology-facilitated organized abuse arrest cases; estimated number of cases in the 3 waves of National Juvenile Online Victimization (NJOV) Study

	*Identified victims: % (n)* *Estimate (95% Confidence Interval (CI), nn¹—nn)²*
NJOV wave	*Organized abuse cases*	*All other cases*	*Total cases*
NJOV1	3 (10) 27 (CI, 7—47)	97 (251) 973 (CI, 918—1028)	100 (261) 1,000 (CI, 949—1051)
NJOV2	3 (11) 44 (CI, 20—67)	97 (307) 1,449 (CI, 1339—1559)	100 (318) 1,493 (CI, 1385—1600)
NJOV3	1 (8) 28 (CI, 4—53)	99 (549) 2,979 (CI, 2755—3203)	100 (557) 3,007 (CI, 2786—3229)

Types of cases of technology-facilitated organized abuse

The cases in the NJOV sample were divided almost equally between primarily familial crimes in which offenders also abused victims who were not family members or gave abusers outside the family access to victims (52%, n=15), and those that were wholly extra-familial (48%, n=14) (Figure 1).

Figure 1: Types of organized abuse cases in a sample of technology-facilitated arrest cases

Almost one quarter of cases (23%, n=8) involved commercial sexual exploitation of children (CSEC), mostly prostitution of minors. In these cases, offenders both sexually abused victims who were minors (aged 17 years or younger) and sold them for sex or otherwise exploited them for profit. With one exception, CSEC cases that qualified as organized abuse were extra-familial. Fifteen percent of cases involved online sex rings (n=6), defined as groups of offenders who used the Internet in perpetrating contact sexual abuse of identified victims (for example, abuse during live video sessions; victims exchanged among members). These online sex ring cases were evenly divided between primarily familial and wholly extra-familial cases. All of the cases in our sample involved the use of technology to further a child sexual exploitation crime, and for most the technology use involved the production of child pornography (87%, n=25). Below, we provide a more detailed description of the types of technology-facilitated organized abuse cases.

Primarily familial cases

Most primarily familial organized abuse cases involved both male and female offenders (Table 3). Most primary offenders were men, aged 26 to 39 years, who lived with spouses or partners and children. Compared to offenders in wholly extra-familial cases, those in primarily familial cases were more likely to employ coercion in the course of their crime, produce child pornography, and possess child pornography downloaded from the Internet. The majority of primary victims in familial cases were aged 12 years or younger.

Our information about the dynamics of the organized abuse in these households is based on the perceptions of the police investigators we interviewed. Some investigators commented in detail but others provided just outlines of what occurred. Based on the limited information provided, there appeared to be two main situations. The first involved ‘dominated families’. These families were headed by dominant, violent men who sexually controlled their female partners and children, along with other children who came or were lured into the household orbit. The second situation was ‘sex focused’ families, in which male and female spouses or partners appeared to participate much more equally in the abuse. Examples of both types follow.

Among dominated families, a man, 35, sexually abused his four children, ages 6 through 14, and at least three neighborhood teens. The offender’s wife was also arrested for sexual offenses, but the investigator stated she could be considered a victim. The offender was domineering and engaged in bondage and bestiality. His wife was described as a ‘sex slave’. Pornography was displayed throughout the house, including Internet pornography shown on computers. The offender encouraged local teens to visit, gave them alcohol, allowed them to smoke and let them spend the night. The investigator stated most teens came from homes ‘where no one cared’. The offender made his family engage in group masturbation sessions when visitors were present. He forcibly raped one teen. The family had been reported and investigated for child neglect. When questioned by police, his children did not disclose any abuse. The offender was convicted of several sexual offenses largely based on evidence provided by an extra-familial victim.

Another case of a dominated family appeared to be tied to religious beliefs, although our information is not sufficiently complete to gauge whether offenders engaged in ritual abuse. The case involved a network of 11 offenders, all of whom subscribed to a magazine that advocated strict parenting and corporal punishment. The primary offender recruited network members through ads in the magazine. Many had close ties to church communities and taught Sunday school classes. The investigator described the primary offender as a ‘very violent and very disturbing individual’. He was investigated by child protective services because his children went to school with bruises, but no action followed. He hosted a website that contained stories about spanking and sexual arousal.

Table 3: Technology-facilitated organized abuse arrest cases: Primarily familial compared to wholly extra-familial

*Characteristics of offenders/victims*	*Primarily familial (n=15)*		*Wholly extra-familial (n=14)*
*Characteristics of offenders/victims*	%	n	%	n
Any female offender***	90	13	19	2
Characteristics of primary offender:
Male	95	14	100	14
Age**
Younger than 18	-	0	-	0
18 to 25	-	0	38	3
26 to 39	85	12	37	6
40 or older	15	3	25	5
Married or living with partner***	90	13	25	5
Lived with child**	70	12	18	3
Prior arrest for sexual offense against a minor	35	4	38	4
Diagnosed mental illness	-	0	6	1
History of violent behavior	41	3	33	5
Problem with drugs or alcohol	17	4	29	4
Prior arrests for non-sexual offenses*	32	4	72	8
Registered sex offender	24	2	-	0
Had child pornography downloaded from Internet**	84	11	39	8
Characteristics of primary victim:
Female	89	12	77	9
Victim age*
0 to 5	11	3	-	0
6 to 12	55	8	19	4
13 to 17	34	4	81	10
Primary victim was:
Threatened	3	1	14	1
Physically assaulted†	-	0	18	2
Coerced*	77	10	40	6
Offered or given drugs or alcohol*	36	4	81	10
Physically injured†	6	1	17	2
Case involved production of child pornography*	97	14	76	11

Ns are unweighted and percentages are weighted. Ns and percentages may not be proportionate because results are weighted to reflect selection probabilities and some cases have more influence than others.

The primary offender abused and covertly videotaped five victims: two sons, two step-daughters and another child. The children were fondled, tied up and brutally beaten with belts and paddles. The primary victim was a son who was three when the abuse started and 11 when his father was arrested. Investigators identified 11 victims of network members. Members produced over 200 videos. They traded the videos with each other. The former wife of the primary offender was involved in the distribution of videos. She was also charged.

The second type of dynamics, sex-focused, involved families that appeared to subscribe to what one investigator called ‘all opportunity’ sexual activity. For example, over a period of about two years, the primary offender sexually abused three of his daughters and several of their friends, produced images of the abuse and allowed adults he knew to do the same. The primary victim was aged 12 to 14 when this happened. Her sisters were younger. Altogether there were five offenders, both men and women, and ten victims. Several offenders were shown in images along with the victims. Some of the victims were given alcohol and cigarettes in exchange for sex or posing for pictures. Some pictures were taken covertly by cameras hidden in the bathroom and other rooms. The primary offender was described as ‘into everything [sexual]’. He was a voyeur who often watched sexual activity by other offenders with his children and other victims. He and his wife were ‘swingers’. His wife denied knowledge of the abuse, although the investigator did not believe her. She was not arrested. In another case, a man and woman abused and photographed her two daughters and three young nieces who lived in a different household. The investigator stated they had no remorse; ‘they [thought] that it should be okay for families to use whatever methods they choose to teach their children about sex’.

The children in these sex-focused families were often abused from early childhood and appear to have been indoctrinated into the family sexual ethic. In most such cases, investigators did not describe family conflict or violence. For example, in the case described above in which parents had no remorse, the investigator noted there was very little conflict between the primary victim, a teenage girl, and her parents; the teen ‘was okay’ with her stepfather. In another case, the investigator suggested that the youngest daughter, who had not been abused, felt left out of family activities. The case narrative in a third case states: ‘... the girls were home schooled and in their home having sex with your parents was the norm. [The investigator] said it was almost eerie listening to [the primary victim, a 10 year old girl] explain the acts’.

Almost all of the primarily familial cases included extra-familial victims. In several cases, these were friends of children in the offender’s household, although we have scant details about recruitment processes. In several cases, extra-familial victims were procured via the Internet either through online contacts that offered children for sex or by sexual solicitations made directly to potential victims by offenders. One case came to the attention of police when a primary offender sent his step-niece nude pictures of himself and solicited her for sex. She reported him. In another case, the girlfriend of a primary offender procured victims for him. The victims, who knew each other or were relatives of the girlfriend, were coaxed or tricked into visiting the offender’s apartment, coerced or forcibly assaulted and then bullied and threatened to keep quiet. A visitor who saw the offender take a young girl into his bedroom and lock the door reported the situation to the police.

Wholly extra-familial cases

Offenders in wholly extra-familial organized abuse cases tended to be younger than those in primarily familial cases. Not surprisingly, they were less likely to have a spouse or partner and live with a child. They were more likely to have prior arrests for non-sexual offenses. Most primary victims in extra-familial cases were teenagers (aged 13 to 17 years) and offenders were more likely to employ drugs and alcohol in the abuse.

Most cases of extra-familial organized abuse among our sample of technology-facilitated arrest cases involved male offenders who befriended and seduced victims, often offering them benefits that were otherwise unavailable to them, such as affection, gifts, alcohol and drugs. In many cases, offenders committed non-forcible sex crimes against victims who were too young to consent to sexual activity (that is, statutory rape).

The cases with male victims tended to involve pairs of male offenders who were partners and who targeted boys that lived in poverty or lacked affection and support. Some of the boys in these cases did not view their situations as victimizing, but rather developed close bonds with offenders. For example, two men, ages 36 and 51, who lived together sexually abused at least four boys. The primary victim was a boy who, when he was 13, met one of the offenders in an online chat room for gay men. About 20 times over a period of three years, he went to their home, usually for a weekend. The trip was about 100 miles and he took the bus or got a ride from a friend until he was old enough to drive himself. According to the investigator, the boy liked hanging out with the offenders, who gave him alcohol and a place to stay. The investigator said his mother did not know where the boy was but was not worried about his whereabouts. At one point, acting on a request by the offenders, the boy recruited another victim who also was abused. A third victim was an 11 year old boy one of the men met while working at a summer camp. The offenders paid the boy for sex and photographed some of the abuse. One image showed this victim in a sexual situation with another boy who police could not identify. This third victim said the boy was from another town; he did not know his name.

The cases of extra-familial organized abuse of teen girls comprised the largest group of extra-familial cases (n=9) and almost all involved commercial sexual exploitation. The number of CSEC cases may have been somewhat inflated by our methodology during Wave 2 of the study when we specifically inquired about technology-facilitated CSEC arrests. Most cases involved prostitution of victims. For example, the offender was a pimp who recruited several runaway teen girls into a prostitution ring. He sexually abused the girls that he prostituted and had them have sex with each other. The girls were photographed and advertised online and sold to others for sex. There were several other offenders who provided services to the ring; for example, a body guard. The primary offender came from a family of pimps. His father and uncle were pimps and his father was bringing his son into the business.

Some of the cases involved less organized or attempted prostitution of victims. For example, one offender was a family friend of the 12 year old victim’s mother. The girl’s family was very troubled; the girl’s father had committed suicide in her presence. The mother frequently let her daughter stay with the offender. The offender began to photograph her, saying he would make her a model. The photographs gradually became sexual. The offender allowed her to have sex with her 14 year old boyfriend at his home and photographed this. (The boyfriend is the second victim.) The offender solicited men to have sex with the primary victim, which he photographed. He was setting up a website for the apparent purpose of selling the victim and sexual images of her when he was arrested. This happened after a person in his community became suspicious of his relationship with the victim and reported him to police.

This case narrative included some details about the victim/offender relationship. According to the investigator, the victim looked to the offender as a friend and parent figure. He ‘did creep her out sometimes. But, to some extent she thought she was pulling one over on him because he would buy her things and let her do things she couldn’t do at home’. He bought her a cell phone and told her he would pay her phone bill if he could take pictures. He promised her money for modeling, gave her alcohol and cigarettes and allowed her friends to hang out at his house. According to the investigator, the offender said he knew the victim was sexually active and solicited men to have sex with her to control her sexual activity. The victim said she had sex with these men because if she did, they would leave. The investigator said the victim attempted suicide when the offender was arrested, but is doing much better after more than a year of counseling. Her mother got involved in the counseling, became more supportive of her daughter and the family seemed to be coming together.

Another case of organized abuse against a group of teen girls illustrates a slightly different dynamic. This case involved four men in their 20s and nine underage teen victims. The offenders were providing a ‘party house’ where girls were given alcohol and drugs and videoed in sex acts with the offenders and each other. The men were selling the videos. The primary offender had been charged with a sex crime as a juvenile. The primary victim told the police what was going on. She was described as having disciplinary problems at home and school and a history of running away. She had been sexually abused prior to this incident. She lived with her mother who, according to the investigator, did not supervise her and was not particularly interested in the investigation of this case. The other victims were angry at the primary victim. They had enjoyed the parties and blamed her for ending the situation.

We found six technology-facilitated organized abuse arrest cases that involved online sex rings, three from Wave 2 of the NJOV Study (arrests in 2006) and three from Wave 3 (arrests in 2009). However, these cases did not involve all members of any online sex ring or provide full information about offenders, victims and crimes committed by sex ring members. Rather, they involved arrests of individuals who had been linked to a sex ring by law enforcement in specific jurisdictions. We classified these as primarily familial or wholly extra-familial based on the circumstances of the primary offender in each arrest case.

For example, one online sex ring involved seven or eight offenders who abused boys using live streaming video so that other ring members could watch. The offenders met online (presumably through a chat room or message board) before they formed the group. The arrest case in our sample involves an offender who abused a 13 year old boy he met through a mentoring program. The boy had developmental delays. He lived with his mother, who was unemployed. This case originated in a different jurisdiction when a victim went to the police. However, we do not have information about the other cases, some of which could have involved familial offenders.

Another case involved an international investigation of a similar online sex ring that located about 27 offenders in eight countries. Our sample included two cases tied to this ring. According to the investigators we interviewed, members were required to molest children live online to gain membership to the group. The live online abuse happened regularly and abusers were able to watch other ring members abuse their children. In one familial case, the primary offender sexually abused his infant daughter online while other offenders watched and requested what they wanted to see. This offender was separated from the child’s mother and watched the baby while the mother worked. He was trying to acquire custody. The mother had no idea that the baby was being abused. In the second case, a husband and wife traced to the group possessed large quantities of child pornography but police could not find evidence they had sexually abused their own or any other children. They were charged only with child pornography possession, so their case was not counted as involving organized abuse.

Organized abuse cases compared to other technology-facilitated child sexual exploitation cases

We compared cases of organized abuse (n=29) to arrest cases involving crimes against identified victims that were not organized abuse from all three waves of the NJOV Study (n=1,107) (Table 4). These latter cases included a variety of circumstances (for example, sex offenders who used the Internet to meet victims, family and acquaintance offenders who used the Internet to groom or solicit victims, and offenders who used technology to produce child pornography). There were several similarities between these two groups of cases. In both types, primary offenders were largely male; about one-third lived with children; small percentages had diagnosed mental illnesses; and about one-quarter had problems with drugs or alcohol (Table 3). Primary victims were largely female and most victims were teenagers.

However, we also found striking differences. Technology-facilitated organized abuse cases were more likely to involve a female offender (56% of organized abuse cases versus 4% of other cases, p<.001). Primary offenders in organized abuse cases were more likely to have prior arrests for both sexual offenses against minors (36% of organized abuse cases versus 7% of other cases, p<.001) and non-sexual offenses of any sort (51% versus 31%, p<.05); be registered sex offenders at the time of their arrest (13% versus 3%, p<.05) and have histories of violent behavior (37% versus 13% of other cases, p<.01). They were more likely to possess child pornography downloaded from the Internet (62% versus 36% of other cases, p<.05) and to produce child pornography depicting their victims (87% versus 54% of other cases, p<.001). Primary victims in organized abuse cases were more likely to be coerced, offered or given drugs or alcohol, or physically injured.

Table 4: Technology-facilitated organized abuse arrest cases compared to technology-facilitated sexual abuse or exploitation not involving organized abuse

*Characteristics of offenders/victims*	*Technology-facilitated organized abuse (n=29)*		*Other technology-facilitated sexual abuse /exploitation (n=1,107)*
*Characteristics of offenders/victims*	%	n	%	n
Any female offender***	56	15	4	36
Characteristics of primary offender:
Male	98	28	95	1061
Age†
Younger than 18	-	0	9	97
18 to 25	18	3	27	263
26 to 39	62	18	33	387
40 or older	20	8	30	357
Married or living with partner***	59	18	28	340
Lived with child**	45	15	30	347
Prior arrest for sexual offense against a minor***	36	8	7	101
Diagnosed mental illness	3	1	3	41
History of violent behavior**	37	8	13	149
Problem with drugs or alcohol	23	8	23	244
Prior arrests for non-sexual offenses*	51	12	31	330
Registered sex offender*	13	2	3	42
Had child pornography downloaded from Internet*	62	19	36	439
Characteristics of primary victim:
Female	83	21	82	893
Victim age†
0 to 5	6	3	5	69
6 to 12	38	12	22	267
13 to 17	56	14	72	769
Primary victim was:
Threatened	8	2	6	68
Physically assaulted†	8	2	4	45
Coerced***	59	16	16	192
Offered or given drugs or alcohol***	58	14	20	194
Physically injured***	11	3	2	29
Case involved production of child pornography***	87	25	54	646

ns are unweighted and percentages are weighted. Ns and percentages may not be proportionate because results are weighted to reflect selection probabilities and some cases have more influence than others.

Discussion

The findings provide details about the characteristics and dynamics of a group of organized abuse cases that involved technology-facilitated child sexual exploitation and ended in arrest. Since we drew from a national sample of arrest cases and our data were weighted to reflect the complex sample design, we were able to estimate the prevalence of such cases. We found the numbers were quite low, ranging from an estimated 27 to 44 arrest cases during the three years of the research. However, because of the methodology used, we could only estimate the number of arrest cases, not numbers of offenders or victims.

Some have voiced fears that technology is fueling an increase in the numbers of organized abuse cases, particularly those involving online sex rings (for example, Child Exploitation and Online Protection Centre 2013). However, our data do not provide evidence of this. The number of cases was small and remained relatively stable in each wave of the survey, even as arrests for technology-facilitated crimes increased (Wolak, Finkelhor and Mitchell 2012). Moreover, in the US, data from multiple sources indicate that child sexual abuse and other sexual offenses against children have declined substantially since the mid-1990s (Finkelhor and Jones 2008, 2012). Statistics from national crime surveys, results of surveys of youth and of mandated-reporters (for example, teachers, doctors, social workers), and data from child protection agencies all show such declines. Thus, our findings show that technology-facilitated organized abuse exists in a variety of forms and perpetrators are using the Internet and digital technologies to enact such crimes, but arrests involving organized abuse did not grow between 2000 and 2009. However, we have no data beyond 2009 and numbers of arrests do not reflect the actual prevalence in the general population.

A unique aspect of the cases in our sample is that the large majority involved offenders who produced child pornography; this was the ‘technology-facilitated’ aspect of many of the cases. Offenders recorded and preserved images of the abuse they perpetrated, which generally removed any doubt about whether sexual abuse at the hands of multiple offenders occurred. The high rates of child pornography production among both primarily familial and wholly extra-familial cases raise concerns about victims, who may suffer additional harms if their images are distributed online. At the same time, pictorial evidence strengthens prosecutions. In some of the cases in our sample, images depicting abuse corroborated victims’ stories. In others, they allowed for prosecution of offenders when victims did not disclose. Although the perpetrators of organized abuse were significantly more likely than other perpetrators of technology-facilitated sex crimes to produce child pornography, this finding only applies to cases that conform to those in our sample (that is, technology-facilitated and ending in arrest).

Our findings provide some insight into the dynamics of organized sexual abuse when it happens in family settings and outside the family. Although our information about victims’ experiences is limited and provided through the perceptions of law enforcement investigators, the diversity and complexity of victims’ experiences is evident. The scenarios of children abused through coercive or other authority exerted by family members or by extra-familial offenders seducing victims with affection, sex, gifts and alcohol are also commonly found in conventional cases of child sexual abuse that involve single perpetrators and single or multiple victims (Berliner 2002; Berliner and Elliot, 2002; Lanning, 2002). A main distinction of organized abuse cases is the collaboration among offenders. Unfortunately, there is not much research about such offender groups and our data do not address their psychology and motivations. However, our findings that organized abusers are more likely to have violent tendencies and prior arrests for sexual offenses compared to the other sex offenders in our sample suggest that dysregulation contributes to their apparent recklessness in collaborating with other sex offenders.

While cases of online sex rings have been widely publicized as involving large numbers of offenders and victims, these cases did not appear to account for many victims in our sample. The rings we had descriptions of were either relatively small or international, with only a small proportion of cases traced to US jurisdictions. Unfortunately, we did not have much information about how any of these online sex rings formed or details of interactions between ring members.

Some cases in our sample involved commercial sexual exploitation of children, usually by prostitution. Such cases have not been categorized as organized abuse in the past, but we found scenarios that clearly met parameters of the definition. For many years, minors caught up in prostitution have been viewed as delinquents, but there is growing consensus among policy makers and practitioners working with exploited children that CSEC is child sexual abuse and victims should be acknowledged and treated as such.

Limitations

These findings have a number of limitations that must be considered when interpreting the results. First, our data pertain only to technology-facilitated cases of organized abuse that ended in arrest in the US during the three years of NJOV Study. The estimates are for arrest cases, not number of arrested offenders or numbers of victims. Second, the NJOV Study was not designed to capture information about organized abuse. In each case we collected detailed data about one primary victim and one primary offender and only limited data about remaining offenders or victims. The algorithm we used to pick primary offenders and victims was created to emphasize offender technology use. Because of this, our data about organized abuse cases are incomplete. We relied on case narratives created by interviewers to fill in details about cases, but these narratives also sometime lacked important information. The study design also probably affected our results about CSEC cases, given that we asked specifically about those cases during Wave 2.

Findings are also limited by the small sample of organized abuse cases. Further, all of the data were gathered from law enforcement investigators. They did not always have full information about every case, and they could provide only limited data about offender behavior. Some of their answers could have been biased by training, professional attitudes, or the adversarial nature of their roles in some cases. Also, our numbers are estimates based on the sample of cases that were the subject of interviews. Although the study was designed to yield a nationally representative sample of cases, sometimes samples are randomly skewed. The margin of error could be larger than calculated. Finally, keeping up with rapidly changing technologies and police responses is a challenge for researchers. Some aspects of crimes involving organized abuse, in particular cases involving online sex rings, may have changed since the last wave of the study in 2009.

Conclusion

Cases of organized abuse are particularly disturbing. The inequality of power exerted by adults over children, which is a factor in all child sexual abuse, is surely magnified when children are faced with two or more adults making sexual demands. Similarly, the violation of trust when offenders are parents, parent-figures or mentors is so much greater. In a considerable number of cases, offenders exploited victims for profit. In addition, the organized abuse offenders were significantly more criminal, violent, and coercive than the sex offenders who acted alone. However, in many respects, the tactics used by organized abuse offenders mirror those used by others who sexually offend against children. Prevention efforts aimed specifically at organized abuse are probably not needed, given existing initiatives to educate children and the public about sexual abuse and encourage disclosure and reporting. Prevention education should also include candid discussions of extra-familial abuse, statutory rape offenses and offender tactics that include seduction, manipulation, and grooming.

Correspondence: Janis Wolak, JD, Crimes against Children Research Center, University of New Hampshire, 10 West Edge Drive, Suite 106, Durham, NH 03824, USA. Email: Janis.Wolak@unh.edu

Wolak J (2015) Technology-facilitated organized abuse: An examination of law enforcement arrest cases. International Journal for Crime, Justice and Social Democracy 4(2): 18-33. doi: 10.5204/ijcjsd.v3i2.227.

Berliner L (2002) Confronting an uncomfortable reality. American Professional Sociey of the Abuse of Children 14(2): 2-4.

Berliner L and Elliott DM (2002) Sexual abuse of children The APSAC Handbook on Child Maltreatment, 2nd edn: 55-78. Thousand Oaks, California: Sage Publications Inc.

Child Exploitation and Online Protection Centre (CEOPC) (2013) Threat Assessment of Child Sexual Exploitation and Abuse. London: CEOPC.

Finkelhor D and Jones L (2008) Good news: Child victimization has been declining. Why? In Finkelhor D (ed.) Child Victimization: Violence, Crimes, and Abuse in the Lives of Young People: 122-147. New York: Oxford University Press.

Finkelhor D and Jones L (2012) Have Sexual Abuse and Physical Abuse Declined Since the 1990s? Durham, New Hampshire: Crimes against Children Research Center.

Gallagher B (2007) Internet-initiated incitement and conspiracy to commit child sexual abuse (CSA): The typology, extent and nature of known cases. Journal of Sexual Aggression 13(2): 101-119.

Harkins L and Dixon L (2010) Sexual offending in groups: An evaluation. Aggression and Violent Behavior 15(2): 87-99.

Kish L (1992). Weighting for unequal pi. Journal of Official Statistics 8(2): 183-200.

Lanning KV (1992) Child Sex Rings: A Behavioral Analysis for Criminal Justice Professionals Handling Cases of Child Sexual Exploitation. Quantico, Virginia: National Center for Missing and Exploited Children.

Lanning KV (2002) Compliant child victims: Confronting an uncomfortable reality. APSAC Advisor 14(2): 4-9.

Salter M (2010) Organised abuse and testimonial legitimacy. Paper presented at the The Australian and New Zealand Critical Criminology Conference, 1-2 July. Sydney: Sydney Law School, The University of Sydney.

Salter M and Richters J (2012) Organised abuse: A neglected category of sexual abuse with significant lifetime mental healthcare sequelae. Journal of Mental Health 21(5): 499-508.

Snow B and Sorensen T (1990) Ritualistic child abuse in a neighborhood setting. Journal of Interpersonal Violence 5(4): 474-487.

Wolak J, Finkelhor D and Mitchell KJ (2012) Trends in Law Enforcement Responses to Technology-facilitated Child Sexual Exploitation Crimes: The Third National Juvenile Online Victimization Study (NJOV-3). Durham, New Hampshire: Crimes against Children Research Center.

^[1] This project was funded by Grant No. 2009-SNB-90001 awarded by the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, US Department of Justice. The total amount of federal funding is $825,704. Points of view or opinions in this document are those of the author and do not necessarily represent the official position or policies of the US Department of Justice.

I am grateful to the many law enforcement investigators who participated in this research and to the talents, perseverance and skills of research assistants Kristina Breton, Elisabeth Cloyd, Matthew Cutler, Laura Healey, Kaitlin Lounsbury, Marisa MacDonnell, Dianne Ramey, Lynn Russ and Samantha Senechal.