EFTA00156754.pdf

Journal of all d Sexual Abuse, 23:957-976, 2014 Copyright O Taylor & Francis Group, DA: ISSN: 1053-8712 print/1547-0679 online DOI: 10.1080110538712.2014.960632 Routledge The Construct of Grooming in Child Sexual Abuse: Conceptual and Measurement Issues NATALIE BENNETT and WILLIAM O'DONOHUE unktersay of Nevada, Reno, Reno, ,Veividn, 1114 There have been claims that some child molesters engage in a "seduction stage" prior to committing abuse. These behaviors, com- monly known as 'grooming," are understood as methods child molesters use to gain access to andprep arefiaure victims to be com- pliant with abuse. However, there is a lack of consensus regarding exactly what this process entails and how it is clearly distinguished from normal adult-child interactions. It is important to devise an accurate definition of grooming for scientific, clinical, and forensic purposes. We critically evaluate the various definitions and reveal problematic heterogeneity. Furthermore, there are no methods of known psychometrics to validly assess grooming. We review the empirical literature regarding the occurrence of grooming and propose future directions for research. KEYWORDS grooming, child sexual abuse, measurement Understanding the process of child sexual abuse (CSA) is important for both its prevention and treatment. Some clinicians and researchers (e.g., Budin & Johnson, 1989; Burgess & Holmstrom, 1980; Conte, Wolf, & Smith, 1989; Elliott, Browne, & Kilcoyne, 1995) agree that a type of seduction stage, commonly called "grooming" but also variously known as "entrapment; "engagement," or "subjection" often precedes the actual sexual abuse. Offenders have admitted that they use techniques such as identifying a par- ticularly vulnerable child, gift giving, and sexual desensitization to prepare the child for the abuse (Seto, 2008). Understanding grooming has both important clinical and legal implica- tions. First, it is passible that if professionals were able to identify grooming Received 23 May 2013; revised 28 January 2014; accepted 10 April 2014. Address correspondence to William O'Donohue, Department of Psychology, University of Nevada, Reno, Abil Stop 298, Reno, NV 89557. E-mail: moratintedu 957 3502-013 Page 1 of 21 EFTA_00001413 EFTA00156754 958 N. Bennett and W. °Donohue before abuse has actually taken place the abuse may be prevented. Second, in a forensic context, sexual abuse allegations might be partially substantiated when it is established that grooming did indeed occur. However, without a clear grooming definition and a valid way of measuring grooming, this judgment that a behavior constitutes grooming becomes problematic. For example, a recently convicted sex offender in las Vegas, Nevada, is seeking to appeal his conviction on the grounds that the testimony provided by a psychologist regarding his grooming behavior is unreliable (Mower, 2012). His defense attorney claimed that "(Grooming' is not a proven science. It's a behavioral thing. .. . How can you tell that this was in the mind of this guy?" There have been attempts to criminalize grooming in several countries. In the United States, a federal law (18 USC S 2252A(aX6)) has made it illegal and thus adds years onto a sentence for people who knowingly offer child pornography to a minor to persuade the minor to participate in an illegal activity such as adult-child sexual contact (18 USC 5 2252A, certain activities relating to material constituting or containing child pornography). In the United Kingdom, Section 15 of the Sexual Offences Act 2003 has covered "the behavior of an offender who meets, or seeks to meet, a child with the intention of committing a sexual assault, if he has met or communicated with that child on at least two earlier occasions" (McAlinden, 2006, p. 342). However, as Gillespie (2004) noted, definitional problems with the construct of grooming limit the use of this law, as grooming is "a transient feature that is difficult to capture and virtually impossible to decide when it begins and ends" (p. 586). McAlinden also described another law designed to criminalize grooming in the UK: Sections 123-9 introduce the risk of sexual harm order—a new civil preventative order which can be used to prohibit specified behaviours, including the 'grooming' of children. . . . This order effectively criminal- izes acts which may be carried out for the purposes of sexual grooming, but only after an individual had been identified as pacing a risk to children. (p. 342) O'Callaghan (2011) described that in Wales a man pled guilty and was sentenced to a year in prison for one count of meeting a child following sex- ual grooming that consisted of inappropriate communication via Facebook. In addition, Vance (2012) described a proposed law in New Zealand that provides a sentence of three years in prison for anyone who participates in online "indecent communication with anyone under 16." This law is aimed at sexual offenders who use Internet chatrooms or other social media websites to find victims. It is evident that these legal definitions of grooming are both varied and limited. The sorts of activities that these laws target do not actually 3502-013 Page 2 of 21 EFIA_00001414 EFTA00156755 Crowning in Child Sexual Abuse 959 capture the notion of grooming because these already involve illegal and abusive contract with a child. Grooming is generally regarded as prior activ- ities intended to prepare the child for abuse, not actual illegal or abusive activities themselves. Thus, legitimate questions can be raised about whether showing a child pornography ought not to be regarded as grooming because it constitutes abuse itself. Clarifying a definition of grooming can thus make these laws applicable to many more behaviors that are used by offenders intending to sexually abuse children. It is important to note that clarifying key constructs is a difficult yet important process. The prominent philosopher of science Larry Laudan (1977) suggested that science has both empirical and conceptual prob- lems and that scientific progress is made when either type of problem is addressed. Conceptual analysis is particularly difficult as it is traditionally not included as a part of the research method in the social sciences and also because it involves the inherent complexity of language (O'Donohue, 2013). Here, conceptual analysis of the grooming construct is necessary in the research process, as it is a salient example where the complexity of language contributes to definitional confusion and leads to problematic implications in clinical and forensic fields. The aim of this paper is to highlight the need for a clearer definition of the grooming construct that may be applied to both clinical and foren- sic work. The courts are currently unable to take much legal action against grooming as it is not well understood and clearly demarcated. Furthermore, psychologists are currently using clinical judgment to determine whether an alleged perpetrator's behaviors are considered grooming. The reliability and validity of these judgments are largely unknown, leaving concerns of unac- ceptable rates of false positives and false negatives. An additional aim is to review the empirical literature regarding what is known about the occurrence of grooming so that a clear definition can be constructed. With a clearer def- inition of grooming, a more scientific assessment of such behavior can he established. This article proposes future directions for research, including validation of the proposed definition and development of an assessment device. CURRENT DEFINITIONS The three tables presented here list various definitions of the construct of "grooming- currently found in the literature. Table I provides various general definitions of the term, Table 2 provides subcategories of grooming that some authors have proposed, and Table 3 provides stages of grooming that several authors have suggested. Thus there is a wide variability that exists in defining sexual grooming as well as possible subtypes or stages of grooming. Although many of the 3502-013 Page 3 of 21 EFIA_00001415 EFTA00156756 960 N. Bennett and W. ODonobue TABLE 1 General Definitions of Grooming Author(s) Definition of Grooming (taken directly from reference) Sgroi (1982) Salter (1995) Howitt (1995) Labors (1997) Gallagher (1999) Brackenridge (2001) Gillespie (2002) Berson (2003) O'Connell (2003) Spiegel (2003) Craven, Brown, and Gilchrist (2006) Knoll (2010) "How does Ithe perpetrator' get the child to participate in some type of sexual behavior? Usually in a low-key, nonforcible fashion, possibly by presenting the activity as a game or something that is 'special' and fun. This always entails misrepresentation of moral standards, either verbally or implicitly. . . . Perhaps rewards or bribes will be offered.' "The establishment (and eventual betrayal) of affection and trust occupies a central role in the child molester's interactions with children. The grooming process itself often scents similar from offender to offender, largely because it takes little to discover that emotional seduction is the most effective way to manipulate children.' "The steps taken by paedophiles to 'entrap' their victims and is in sonic ways analogous to adult courtship' "The offender plans to make the victim less likely to resist, to make others unaware of what he is doing, or even to make them likely to help him, without their knowledge, to molest a child.' "Entrapment involves the use of an amy of material, illicit and emotional 'inducements to draw children into abusive situations and increases their difficulty in disclosing." "The process by which a perpetrator isolates and prepares an intended victim.' "The process by which a child is befriended by a would-be abuser in an attempt to gain the child's confidence and trust, enabling them to get the child to acquiesce to abusive activity. It is frequently a pm-requisite for an abuser to gain access to a child.' "Grooming involves a clever process of manipulation, typically initiated through a nonsexual approach, which is designed to entice a victim into a sexual encounter (Brown, 2001). The inhibitions of a child are lowered through active engagement, desensitization, power and control. It is often characterized as a seduction, involving a slow and gradual process of learning about a child and building trust. This also contributes to the difficulty in detecting the activity. Grooming is also a deceptive process in which a child is unprepared to interpret cues which signal danger of risk. Predators arc skilled at gaining the trust of a child before luring them into interactions. The process of grooming through the formation of a close bond creates a victim who is more likely to comply with sexual advances." "A course of conduct enacted by a suspected paedophile, which would give a reasonable person cause for concern that any meeting with a child arising from the conduct would be for unlawful purposes." "Subjection is the process of predisposing a boy to sexual abuse by means of subtle or blatant interactions that lead to boundary diffusion and role confusion.' "A process by which a person prepares a child, significant adults and the environment for the abuse of this child. Specific goals include gaining access to the child, gaining the child's compliance and maintaining the child's secrecy to avoid disclosure. This process serves to strengthen the offender's abusive pattern, as it may he used as a means of justifying or denying their actions." "The process by which sex offenders carefully initiate and maintain sexually abusive relationships with children. Grooming is a conscious, deliberate, and carefully orchestrated approach used by the offender. The goal of grooming is to permit a sexual encounter and keep it a secret.' 3502-013 Page 4 of 21 EFTA_00001416 EFTA00156757 Grooming In add Sewell Abuse 961 TABLE 2 Proposed Subtypes of Grooming Author Types of Grooming Leberg (1997) I. Physically grooming the victim 2. Psychologically grooming the victim and family 3. Grooming the social environment and community Craven, Brown, and Gilchrist (2006) I. Self-grooming 2. Grooming the environment and significant others 3.Grooming the child McAlinclen (2006) I. Personal 2. Familial 3. Institutional Wyre (1987)as discussed in Howitt I. Extrafamilial (1995) 2. Intrafamilial definitions share some key similarities, many behaviors may he classified as grooming under some definitions but not under others. Some of these sim- ilarities in definition include the criterion of preparing a child for abuse (Brackenridge, 2001; Craven, Brown, & Gilchrist, 2006; Gallagher, 1999), gaining a child's trust (Berson, 2003; Craven et al., 2006; Gillespie, 2002; Salter, 1995), making it more difficult to the child to resist or disclose the abuse (Berson, 2003; Craven et al., 2006; Gallagher, 1999; Gillespie, 2002; Knoll, 2010; Leberg, 1997), and the enumeration of specific tactics used to groom the child (Berson, 2003; Gallagher, 1999; Sgroi, 1982). Furthermore, a variety of different kinds of definitional features are pro- posed. For example, one definition mentions "betrayal" (Salter, 1995) while another references "courtship" (Howitt, 1995). Some proposed definitions give concrete examples of grooming (Gallagher, 1999; Sgroi, 1982), while others try to give abstract properties to capture what the authors take to he the essential properties of grooming (O'Connell, 2003; Spiegel, 2003). Some definitions are fairly brief and more vague (Brackenridge, 2001; Howitt, 1995; Spiegel, 2003), whereas others are much longer and more detailed about what grooming looks like (Berson, 2003; Craven et al., 2006; Sgroi, 1982). Obviously this heterogeneity presents serious challenges for forensic and clinical work. Some of these definitions involve additional difficulties in that the terms used to define grooming present additional serious definitional problems themselves. For example, Salter (1995) used the phrase "emotional seduc- tion" in her grooming definition. This requires further delineation of what exactly emotional seduction entails as well as a measurement strategy to validly capture this alleged dimension. As another example, Spiegel's (2003) definition involves constructs such as "boundary diffusion" and "role con- fusion." These constructs are not part of the standard scientific lexicon and thus create further impediments to enhancing our scientific understanding of the grooming process. Finally, some definitions propose stages of grooming, 3502-013 Page 5 of 21 EFIA_00001417 EFTA00156758 962 N. Bennett and W. ODonobue TABLE 3 Proposed Stages of Grooming Author Stages of Grooming van Dam (2001) Fkackenridge (2001) Applies to grooming in sport. Wyrc (1987) as discussed in Howin (1995) Christiansen and Blake (1990) Applies to father daughter grooming. O'Connell (2003) Applies to online grooming. I. Identify vulnerable child 2. Engage that child in peedike environment 3. Desensitize the child to touch 4. Isolate the child 5. Make the child feel responsible I. Targeting a potential victim 2. Buikling trust and friendship 3. Developing isolation and control, building loyalty 4. Initiation of sexual abuse and securing secrecy Applies to extrafamilial grooming: I. The offender masturbates and fantasizes about future contacts, a boy is befriended while an effort is made to earn his parents' trust, outings are common to achieve intimacy 2. The offender finds out about the boys home/school problems—a 'counseling' role is created 3. Physical contact of a nonsexual son begins, offender's masturbation and fantasy continue, sexual touches begin and gradually increase in severity Applies to intrafamilial grooming: 1. Tickling the child 2. Bathing 3. The offender's sexual arousal and fantasy 4. The child going to the offender's bed 5. The offender knows the child likes being tickled 6. "Sex education' 7. Tickling reaches child's sexual parts 8. Offender masturbates the child's genitals 9. The child is trapped into silence 10. Sexual contact is increased II. Offenders cognitive distortions increase 12. Becomes difficult to end sexual contact 13. Offending behavior reinforced through masturbation to fantasies I. Trust 2. Favoritism 3. Alienation 4. Secrecy 5. Boundary violations I. Friendship-forming 2. Relationship-forming 3. Risk assessment 4. Exclusivity 5. Sexual 3502-013 Page 6 of 21 EFTA_00001418 EFTA00156759 Grooming In add senior Abuse 963 which makes the definitional and measurement process even more complex as each of these stages must be delineated and validly measured, and this must be done with the proposed sequencing as well. For example, Wyre's (1987) proposed stages for intrafamilial abuse involved 13 separate steps, the first 10 of which according to the author an be identified as groom- ing behaviors. As Howitt (1995) pointed out, this account of the abuse cycle makes it "appear a relatively short-term and repetitive process" (p. 85), which certainly is not representative of all cases of child sexual abuse. GROOMING AS A DEVIANT PROCESS Part of the difficulty in identifying and clarifying a useful definition of groom- ing is the fact that many behaviors used by perpetrators appear quite similar to behaviors seen in normal adult-child relationships. Buying gifts for chil- dren or taking them on private outings obviously are not always precursors to sexual abuse. Using the male sports world as an example, Hartill (2009) wrote that "in preparing for the abuse, the perpetrator is able to use such 'disinhibiting' techniques through drawing on practices and discourses that are, to varying degrees, a normative feature within many, if not all, male sports contexts" (p. 239). Obviously part of the reason for this similarity to normative behavior is that the potential abuser does not want to be detected and thus wants to disguise what he is setting out to accomplish. In addition, it must he recognized that not all sexual offenders use grooming techniques. Grail, Hobson, and Gary (1982) differentiated between "child molesters" and "child rapists." Part of this distinction for these authors stems from their observation that child molesters use a grooming process on their victims, whereas child rapists do not, as their assaults occur suddenly. These authors also pointed out that child molesters are much more common than child rapists. Alternatively, some authors seem to construe grooming behaviors as mirroring behaviors seen in dating relationships between two consenting adults. For example, Herman (1981) wrote that sexually abusive fathers make an attempt to "court" their daughters by giving them flowers or presents (e.g., expensive jewelry or lingerie). THE PREVALENCE OF GROOMING Because the definition of grooming varies from study to study, currently there is no way to know precisely how prevalent grooming is because each study employs some variant of the definition. Nevertheless, there have been several empirical studies conducted with child molesters to determine what specific methods they frequently use to choose their victims, initiate the abuse, and keep their victims from disclosing. Other researchers have chosen to focus 3502-013 Page 7 of 21 EFTA_00001419 EFTA00156760 964 N. Bennett and W. ODonobue on the victims and ask them what techniques their abusers used prior to the abuse. It is important to note that in this article, examples of grooming with different genders of perpetrator and genders of victim are not readily distinguished. This is primarily due to the fact that the grooming literature reviewed did not always provide statistics about which grooming behaviors were used on boys versus girls. In addition, most of the grooming literature reviewed discussed male offenders. Identifying Potential Victims Elliott and colleagues (1995) interviewed 91 child sex offenders about the strategies they used when committing their offenses. They found that 33% of the offenders explicitly worked on becoming welcome in the child's home and 18% offered incentives or threatened their victims to recruit other children and then gave bribes to the recruits. Conte, Wolf, and Smith (1989) interviewed 26 offenders about their crimes. They found that offenders often admitted to being able to iden- tify what they considered a vulnerable child—often one who was "needy" and seemed "quiet." For example, one offender stated that his tactic was to look for a kid who is easy to manipulate. They will go along with anything you say. I would approach them by being friendly, letting them think I was someone they could confide in and talk to" (Conte et al., 1989, p. 298). In her review of literature about sexual abuse involving teachers, Shakeshaft (2004) noted that selection of a victim is "influenced by the compliance of the student and the likelihood of secrecy" (p. 32). Teachers usually look to victimize students whom they have control over. Shakeshaft also identified factors that make a child vulnerable to educator sexual abuse, such as problems at home with parents, lack of confidence, and partici- pation in other risky behavior. However, it also must he remembei ad that nonoffending adults could see the same needs in these vulnerable children and want to help them in legitimate ways. Thus the child's vulnerability and needs cannot he a sufficient condition for defining grooming. The Use of Attention, Bribery, and Coercion Elliott and colleagues (1995) found that 53% of the offenders in their sample offered to play games, teach a sport, or teach how to play a musical instru- ment. Forty-six percent gave bribes, took the child for an outing, or drove the child home. Thirty percent admitted to using affection and love to gain the child's trust. Forty-six percent of the offenders used gifts as bribes in exchange for sexual favors. The offenders interviewed by Conte and colleagues (1989) also claimed they used bribery and coercive strategies prior to sexual contact. For example, one sex offender stated that his specific methods included "play, 3502-013 Page 8 of 21 EFTA_00001420 EFTA00156761 Grooming In add Sentra Abuse 965 talking, giving special attention, trying to get the child to initiate contact with me. Get the child to feel safe to talk with me" (p. 297). In his literature review on teacher sexual misconduct, Knoll (2010) found that educator sexual offenders tend to use bribery by giving their students special attention or rewards. According to Knoll, "the power of such rewards to affect the student should not be underestimated. Rewards from a teacher may have a crucial impact on the student's motivation and cognitions" (p. 376). Budin and Johnson (1989) interviewed 72 sex offenders about methods they used to gain access to and abuse their victims. When asked what they did to gain their victims' trust, the majority of offenders admitted to acting like the child's friend and playing games with them. Other strategies included giving money, toys, candy, cigarettes, beer, or drugs to the child. In his study of institutional sexual abuse, Gallagher (2000) looked at a sample of 65 substantiated cases of abuse. He found that grooming, or 'entrapment," which he defined as "the process by which perpetrators draw children into abusive situations and make it difficult for them to disclose" (p. 810) was reported in 35% of cases. In these cases, he found that 39% of perpetrators took the child away from the institution (thus isolating the child), 22% gave the child extra attention, 22% gave money to the child, 9% provided the child with illicit goods, and 4% provided the child with games or toys. In their interviews with 23 CSA victims, Berliner and Conte (1990) found that many children shared similar experiences with bribery and coercion prior to their abuse. Sixty-one percent of children reported that their abusers made excuses to spend time alone with them; 61% indicated that they were told that they were special, different, or the only one who understood the abuser; 61% said that their abuser treated them as an adult or he acted as a child toward them; 57% reported that their abusers gave them special privileges that made them feel obligated to be compliant in the abuse; 39% indicated that their abuser shared private information about spouses with them; 39% reported that their abuser prevented them from having friends or doing activities that other children do; and 30% reported that their abuser treated them "meaner" than other children. Shakeshaft (2004) wrote that in educator sexual abuse, teachers usually "coerce- their student victims by providing additional help (e.g., advisement on a project or taking on an outing) that not only allows for time alone with the victim but are also activities for which the victim's parents tend to be grateful to the teacher. Furthermore, she pointed out that because these acts do not yet constitute recognizable sexual abuse and because they share sim- ilarities with legitimate activities, any complaint about these activities cannot lead to much disciplinary action. Christiansen and Blake (1990) discussed that in father—daughter incest, most fathers purposely build a trusting relationship with their daughters 3502-013 Page 9 of 21 EFTA_00001421 EFTA00156762 966 N. Bennett and W. ODonobue prior to beginning any sexual abuse. However, this is a somewhat flawed analysis, as there ought to be a trusting relationship in all father-daughter relationships, at least prior to any abuse. Seventy-three percent of perpetrat- ing fathers viewed this trust as crucial to the sexual relationship to reduce the risk of the daughter disclosing the abuse (Warner-Kearney, 1987, as cited in Christiansen & Blake, 1990). Many fathers also show clear signs of favoritism toward their victimized daughter relative to their other children. Burgess and Holmstrom (1980) wrote that molesters tend to use three types of pressure to make their victims compliant: material goods, misrepresentation of moral standards, and the need for human contact. They noted that material goods are the most frequent tool that offenders use. One difference has been found between genders of the victim in this emotional coercion type of grooming. Spiegel (2003) noted that in male victims, emotional coercion can take on a negative tone. For example, per- petrators may use name-calling words such as "fag" or "whore" to put the male child down and make him feel ashamed and thus less likely to disclose the abuse. Sexual Desensitization Elliott and colleagues (1995) found that of those offenders who used babysit- ting as a strategy to gain access to their victims, 27% started talking to the child about sex, 21% misrepresented the abuse as educational or loving (which again may not he part of grooming because some of this would be postahuse), and 20% offered to bathe or clothe the child. Furthermore, these authors found that 40% of all offenders said the first move they made was sexual touching or genital kissing. Thirty-two percent of the offenders asked the child for help with undressing or lying down. Forty-four percent of the offenders used coercion and persuasion, 49% talked about sex with the child, and 47% used "accidental" touch. Sixty-one percent of the offenders would stop the abuse if the child became resistant and then persuaded the child to let them begin again. Many offenders committed the abusive acts in their own homes, where 33% used pornographic videos and magazines to desensitize the child. Conte and colleagues (1989) found that sexual desensitization was com- monly used among the offenders in their sample. For example, one offender stated, Most of the time I would start by giving them a rub down. When I got them aroused, I would take the chance and place my hand on their penis to masturbate them. If they would not object, I would take this to mean it was OK. I would isolate them. I might spend the night with them. Physical isolation, closeness, contact are more important than verbal seduction. (p. 297) 3502-013 Page 10 of 21 EFTA_00001422 EFTA00156763 Grooming in Child Sexual Abuse 967 Knoll (2010) found that while a teacher is using bribery to gain the trust of a victim, typically conversation about sexual matters with the stu- dent is also starting to emerge. Physical contact is then gradually increased. Furthermore, Gallagher (2000) found that in cases where "entrapment" behaviors were reported, 43% of perpetrators initiated physical contact with the child and 17% behaved in a sexual manner with the child. Furthermore, differences between genders of the victim have also been noted in the sexual desensitization type of grooming. Spiegel (2003) noted that the use of pornography to sexually desensitize children is more common with male victims than with female victims. Boundary Violations Berliner and Conte (1990) found that 70% of children reported that their abusers "accidentally" came into their bedroom or bathroom while they were undressing; 61% indicated that their abusers "accidentally" touched their pri- vate parts; 61% said that their abusers did not respect their privacy or let them close doors; 61% reported that their abusers "accidentally" showed their naked body to them; 57% indicated that their abusers would purposely do things with the child that involved physical contact; 48% said that their abusers made sexual comments about the child's body or clothing; 44% reported that their abusers asked them to do things that involved physical contact; 30% said that their abusers would inspect the child's body "to see how it was developing"; 30% indicated that their abusers "taught sex educa- tion" by showing pornographic pictures and touching the child's body; 26% reported that their abusers told the child about sexual things he had previ- ously done; and 22% indicated that their abusers put lotion or ointment on the child when they were alone but said he was doing nothing wrong. According to Christiansen and Blake's (1990) stages of grooming in father—daughter incest, the last step involves the father violating his daugh- ter's boundaries. In particular, Fathers may insist on bathing their daughters and do not allow other family members to do this. These baths frequently involve inappropriate sexual behavior. Fathers also insist on dressing their daughters or on watching them get dressed. In addition, fathers will tend to watch the child use the bathroom. Finally, perpetrating fathers will have sexually explicit conversations with the daughter to further desensitize them. Grooming the Child's Environment Elliott and colleagues (1995) found that 20% of the offenders in their sam- ple admitted they gained the trust of the child's family with the purpose of abusing the child. Forty-eight percent isolated their victims through babysit- ting. Furthermore, Knoll (2010) found that a teacher can also manipulate the 3502-013 Page 11 of 21 EFIA_00001423 EFTA00156764 96R N. Bennett and W. °Donohue relationship with her victim's parents to gain their approval of spending time with their child. Van Dam (2001) pointed out that many child molesters spend years gain- ing the trust of members in the community before actually sexually abusing any children. She hypothesized that these offenders use several social psy- chological techniques to groom the community effectively. As an example, they may use "foot-in-the-door technique" by showing up uninvited to a child's birthday party and spending time playing games with the children. The parents would feel uncomfortable asking this person to leave and have thus subtly cooperated with the offender. From then on it would be easier for the offender to gain cooperation from the parents on spending time with the children. Offenders can also use conformity against these parents—it would go against social norms and be rude to ask a person to leave a party when the children are enjoying spending time with an offender. In addition, cognitive dissonance can play a role as the parents will try to make their beliefs about the offender consistent with their actions of letting their chil- dren around him or her (they will believe that they think he or she is a good person). Finally, confirmation bias can also play a role as the parents will tend to only accept information that is confirming their existing beliefs about the offender. Commonalities The two major commonalities in the definitions reviewed as well as the empirical studies of grooming are (a) some sort of inappropriate behav- ior on the part of the prospective abuser (whether it is a bribe, boundary violation, invasion of privacy, misstatement of morality, mischaracterizing an interaction as a "game," isolation, emotional manipulation, etc.) and (b) the function of this inappropriate behavior is to increase the likelihood that the adult can sexually abuse the child (by, for example, gaining access to them, gaining their trust, silencing them, isolating them, desensitizing them to nudity or sex, etc.). Each component of the definition may have different topographies in individual cases (e.g., sometimes the inappropriate behavior is removing a door to the child's bedroom, or sometimes it may be buy- ing the child a bikini), but the function of the behavior is to increase the likelihood of future abusive contact. A PROPOSED DEFINITION Any definition ought to use the empirical findings reviewed previously about common strategies used by sexual molesters. In addition, we believe that the most useful definition of grooming would attempt to instantiate the following definitional meta-criteria: 3502-013 Page 12 of 21 EFTA_00001424 EFTA00156765 Grooming In Child Sexual Abuse 969 I. Minimize False negatives. Thus, we wanted the definition to he sensitive to all occurrences of grooming. 2. Minimize false positives. Thus we also wanted the definition at the same time to be specific and not overinclusive (including perfectly appropriate behaviors as invalid examples of grooming). 3. Be capable of providing the basis for a valid assessment procedure. 4. Not include constructs that in themselves bring about fiwther definitional problems. 5. Minimize judgment, although not completely avoid it as we believe that determining a behavior to be grooming essentially requires some complex judgments regarding appropriateness. 6. Show interrater reliability (have a high degree of agreement across raters). 7. Allow the rater to have multiple choices regarding final decisions given the complexity of individual cases, such as clearly grooming, probably grooming, uncertain, or not grooming. 8. Allow a third party to understand the logic of these judgments and conclusions by explicating the decision pathway for these final judgments. We propose that grooming he defined as "antecedent inappropriate behavior that functions to increase the likelihood of future sexual abuse." There are no stages of grooming as there are in some definitions as proposing stages necessitates additional definitions and demarcations of each stage. Therefore, there are two individual criteria that must he met to consider a behavior to be "grooming:" (a) the behavior being evaluated must in and of itself be inappropriate and a case for this inappropriateness must he made, and (b) a sound argument must be presented that the behavior or behaviors increases the likelihood of future sexual abuse. The definition is further elucidated by providing a number of exemplars of grooming: I. Any sexualization of the relationship such as talking about sex in a way that is not permissible given the adult's relationship with the child (e.g., it is permissible for parents to provide sex education to their children) or exposing the child to sexually explicit materials such as R rated movies (showing the child pornography would be abusive in and of itself and therefore not grooming). 2. Inappropriate gift giving (developmentally or socially inappropriate, such as bikinis or bras purchased by a neighbor or teacher). 3. Inappropriate nonsexual communication with the child (e.g., telling the child she is the only one who understands the offender, or telling her 1 love you" when the social role is not appropriate for this type of commu- nication), particularly when an adult uses these statements to manipulate the child to do something (e.g., "I love you and people who love each other touch each other"). 3502-013 Page 13 of 21 EFTA_00001425 EFTA00156766 970 N. Renner: and W. O'Donobue 4. Inappropriate touching of the child (e.g., excessive tickling, hugging, wrestling, sitting on lap). 5. Bribes for inappropriate contact (e.g., bribes for nonsexual or sexual touching or bribes to meet the adult secretly). 6. Threats related to not participating in inappropriate contact. 7. Inappropriate isolation of the child (e.g., trips where the offender and victim are alone that are not part of the normal adult—child relationship. It is permissible for a father to drive a child to school), or inappropriately discouraging the child to play with friends or be with family. For parents or other caretakers, the threshold for what is considered inappropriate behavior is higher than for other adults. 8. Favoritism directed toward the child (e.g., the child is treated much better than siblings or classmates, particularly when this is intimate or isolating). 9. Boundary violations such as inappropriately bathing the child, clothing the child, sleeping with the child, the adult being in underwear around the child, the adult acting like a child, or the adult sharing private information with the child, particularly sexual or relationship information (e.g., "my wife and I are not having sex"). Again, for parents, family members or caregivers the threshold is much higher for defining a boundary violation than for others. 10. Asking the child to keep secrets, particularly about their contact (e.g., the mother's Christmas present would not be regarded as a problematic secret, whereas asking the child to not tell that she was with the offender would he). II. Providing the child drugs or alcohol (note: although this behavior is already abusive, it is not sexually abusive; thus, it can be consid- ered a grooming behavior as it is inappropriate and serves to facilitate compliance with the intended sexual abuse). 12. Misstating moral standards regarding touching, contact, or sex, particularly when these relate to adult—child sexual contact or sexualizing a situation. 13. Repeated violations of the child's privacy (e.g., walking into bathroom when child is in there, watching her get dressed, etc.). The more of these features present, the more likelihood the individual's behavior represents grooming. To further clarify the grooming definition, it may be helpful to also look at a few specific exemplars of nongrooming behaviors that may be misinterpreted: 1. Purchasing appropriate gifts for the child (e.g., for birthdays). 2. Engaging in appropriate hand-holding (e.g., to cross the street). 3. Bathing a young child by a legitimate caregiver without any inappropriate touching. 3502-013 Page 14 of 21 EFTA_00001426 EFTA00156767 Grooming in Child Sexual Abuse 971 4. Having age-appropriate and relationship-appropriate discussions of body pans. 5. A care-giving figure saying "I love you" without the goal of manipulation (not using the phrase to get the child to do something inappropriate). THE ASSESSMENT OF GROOMING Because grooming is a set of common behaviors seen in child sexual offend- ers (as the previous reviewed literature seems to support this conclusion), it should he possible to assess behaviors to determine whether they are indicative that sexual abuse is likely to occur. We have reviewed the pub- lished literature and have not been able to find any measures that validly assess grooming behaviors (by any definition of grooming). This greatly reduces the value of any definition as the practical usefulness of a definition is seen in its ability to be operationalized in valid measurement processes. It is important to develop valid measures as it is not ideal for the detection of grooming to be an entirely post hoc process—that is, only after the abuse occurs are the gifts seen as inappropriate and thus as part of a grooming process. The grooming acts should seem at least somewhat inappropriate at the time they are occurring and thus ideally adults can intervene to stop future abuse. To resolve this problem, grooming requires a valid definition and a psychometrically adequate assessment procedure to reduce both the number of false positives and false negatives. We are currently developing an assessment device that would aid a clin- ician in coming to a valid conclusion as to whether an individual's behaviors can be considered grooming. As mentioned, an assessment of grooming in our proposed definition would involve a two-step process: (a) determining that the adults behavior is inappropriate in and of itself, such as if the tickling is excessive or the bikini gift is not justified by the nature of the relationship; and (h) reasonably arguing that the function of this inappropriate behavior is to increase the likelihood of future abusive contact. What is "inappropriate" admittedly is somewhat of a vague term that requires judgment because we need to clearly differentiate the behavior from normal adult-child relationships. However, this judgment requirement currently exists for other psychological constructs as well. For example, the diagnosis of a major depressive episode might require that a clinician judges the client's guilt as "excessive." In addition, under our criteria these judgments of inappropriateness should he explicit (there must be a clear argument as to why the behavior is inappropriate). Since the argument must be explicated, others would he able to evaluate it and decide whether it is a nonproblematic judgment of the behavior. To illustrate, a male coach buying an eight-year-old girl a bikini would generally be considered inappropriate, but the act of buying her a pair of 3502-013 Page 15 of 21 EFTA_00001427 EFTA00156768 972 N. Bennett and W. °Donohue gym shoes with her mother's consent generally would not. A criticism of this definition is that "appropriate" behaviors may be used by some perpetrators to groom (e.g., buying a poor child gym shoes may still be performed to gain the child's trust to eventually abuse her). However, because this behav- ior can he entirely unrelated to abuse and because the assessment ought to strive to minimize false positives, we have chosen to require that all groom- ing behaviors be inappropriate in and of themselves. Second, we recognize that not all inappropriate behaviors ought to be considered grooming—an adult offering cigarettes to a child may be inappropriate but in addition we narrow the class of these inappropriate behaviors to those that are related to increasing the probability of sexual contact. It is important to note that there may be instances in which the question- able behavior falls in some gray area between grooming and nongrooming. For example, a father buying his daughter a bikini may or may not be con- sidered inappropriate. Unless more details are known about the context of this purchase (perhaps it was just impossible for the mother or some other female to do this and the need for a bathing suit was urgent), one could explicate reasonable arguments that the behavior is and is not representa- tive of grooming. The most logical judgment to come to, then, is that this behavior is an indeterminate case and that it Ls unclear whether it should be considered grooming. Arguments also need to be made regarding the second criterion of the definition: whether the function of the inappropriate behavior was to increase the likelihood of abuse in the future. Interpreting behavior and the intentions of a person performing a behavior is admittedly complex. The rational appraisal of behavior involves setting up a universe of plausible interpretations and gathering evidence in the individual case to rule in or rule out each. As an example, if a male neighbor has a pool and buys a bikini for a five-year-old girl to come over to swim and has her change into the bikini at his house while they are alone (the purchase of the bikini and having her change in his house without her guardian present would be con- sidered inappropriate and thus meet the first criterion of the definition), the set of major plausible interpretations regarding the function of the behavior include: I. Buying a bikini for the girl was the only way to allow the child to engage in the appropriate activity of swimming. Changing at his house was the only way to have the child dressed appropriately for swimming. These fads do not function to set the occasion for abuse to occur. 2. Buying a bikini and having the child change alone is not appropriate as there are more appropriate, prudent alternatives. In addition, bikinis can he thought of as sexualized clothing and changing alone without a guardian present is also a boundary violation. For example, giving money to the child's guardian to buy whatever bathing suit the guardian thought 3502-013 Page 16 of 21 EFTA_00001428 EFTA00156769 Grooming in Child Sexual Abuse 973 appropriate (perhaps a one piece) is a better way to allow the child to engage in the healthy activity of swimming. In addition, the child's guardian should always be present when the child is swimming in the pool or changing into her bathing suit. This pathway does not increase the probability that abuse will occur in the future. In this case, clearly the second alternative explanation is superior due to the fact that it respects the guardian's control, enhances the guardian's ability to supervise the child, does not isolate the child, might involve a less revealing swimsuit, and allows the guardian to exercise his or her discretion regarding what is appropriate swimwear. In addition, because the first altemative con- tains false assertions and can set the occasion for abuse, while the second alternative contains true assertions and is consistent with decreasing the like- lihood of abuse, it is concluded that the behavior under question meets the second criterion of our definition of grooming as it is functioning to increase the likelihood of future abuse. Again, the advantage of this approach is that it explicates the arguments for a person's behavior as meeting or not meeting the definitional criteria. The situation is complex because often grooming is meant to be disguised or ambiguous by the would-be abuser. However, this approach does allow the generation of alternatives that would he more prudent and reasonable and thus both the inappropriateness and function of the behavior can be rationally identified. Finally, before this assessment method is accepted it must be evaluated with respect to its interrater reliability, predictive validity, sensitivity, and specificity. Currently, it is unfortunate that the field has no assessment meth- ods to properly identify grooming and thus understanding the psychometrics of this definitional approach (as well as others) ought to be a priority. Obviously the proposal of this definition is just a first step, and it gen- erates a large research agenda. Validity studies need to be run on a sample of what experts clearly identify as instances of grooming and instances of normal behavior to see the extent to which professionals trained in this definition can correctly identify these two kinds of behaviors. The rates of false positives and false negatives need to he identified in these sorts of studies and modifications in the definition, assessment, or training need to occur in attempts to minimize these. Randomly controlled studies can be used to compare the accuracy of this method as compared to other pro- posed methods and definitions. Studies need to be conducted investigating different types of abuse (e.g., familial versus nonfamilial, boys versus girls, young children versus teenagers, majority versus minority culture) to see the extent to which this definition is robust across these varying dimensions. Again, modifications would need to be made when pmblents or limitations are found. It would also be useful to conduct some longitudinal research with high-risk samples to determine the likelihood of accurate detection of 3502-013 Page 17 of 21 EFTA_00001429 EFTA00156770 974 N. Bennett and W. °Donohue grooming and the prevention of future abuse (by perhaps a comparison with a no treatment control). Another important issue is to investigate what sort of training programs or materials need to be developed so that a wide vari- ety of professionals can faithfully implement the definition and proposed assessment methods. CONCLUSIONS Currently there is no consensus regarding how to define grooming. In addi- tion, there is no valid method to assess whether grooming has occurred or is occurring. The field possesses an insufficient amount of knowledge about key issues such as the interrater reliability of these judgments or the error rates of these judgments including the frequency of false negatives or false positives. Thus currently it appears that grooming is not a construct that ought to be used in forensic settings as it does not meet some of the cri- teria in the Daubert standard. The Daubert standard indicates that in court an expert witness may only testify if (a) "the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on suffi- cient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and meth- ods to the facts of the case' (Rule 702: Testimony by expert witness). Right now it does not appear to be the case that there are "reliable principles and methods' to define and detect grooming. We propose a definition of grooming that involves two parts: (a) inap- propriate behavior on the part of the adult and (b) sound arguments that this inappropriate behavior functions to increase the probability of future sexual abuse. We then provide exemplars of this kind of inappropriate behavior. Future research on grooming would be more useful to the field if data were collected using a single, clear definition such as the one we have proposed. This would provide a basis for data to be easily aggregated and better under- stood, which could lead to the admissibility of grooming evidence in forensic settings. Furthermore, we are currently working to develop valid psycho- metric instruments with known reliability and validity to assess grooming according to this standard. REFERENCES 18 USC 2252A. Certain activities relating to material constituting or containing child pornography. (n.d.). In Legal lnforneation lnsdituie online. Retrieved from littp://www.law.cornell.edu/uscode/text/18/2252A 3502-013 Page 18 of 21 EFTA_00001430 EFTA00156771 Grooming in Child Sexual Abuse 975 Berliner, L., & Conte, J. R. (1990). The process of victimization: The victims' perspective. Child Abuse and Attlee!, 14, 29-40. Berson, I. R. (2003). Grooming cybervictims: The psychosocial effects of online exploitation for youth. Journal of School Violence, 2(1), 5-18. Brackenridge, C. H. (2001). Spoilsports: Understanding and preventing sexual exploitation in sport. London, England: Routledge. Brown, D. (2001). Developing strategies for collecting and presenting groom- ing evidence in a high tech world. National Center for Prosecution of Child Abuse Update, 14(11). Retrieved from hup://www.ndaa-apri.org/publiattions/ newslettent/update_volume_14_number_ 1 1_2001.html Budin, I. E., & Johnson, C. F. (1989). Sexual abuse prevention: Offenders attitudes about their efficacy. Child Abuse & Neglect, 13, 77-87. Burgess, A. W., & Holmstrom, L. L. (1980). Sexual (name of children and adoles- cents: Pressure, sex, secrecy. In L. G. Schultz (Ed.), The sexual victimology of youth (pp. 67-82). Springfield, IL Charles C. Thomas. Christiansen, J. R., & Blake, R. H. (1990). The grooming process in father—daughter incest. In A. L Horton (Ed.), The incest perpetrator: A family member no one aunts to treat (pp. 88-98). Thousand Oaks, CA: Sage. Conte, J. R., Wolf, S., & Smith, T. (1989). What sexual offenders tell us about prevention strategies. Child Abuse and Neglect, /3(2), 293-301. Craven, S., Brown, S., & Gilchrist, E. (2006). Sexual grooming of children: Review of literature and theoretical considerations. Journal of Sexual Aggression, 12(3), 287-299. Elliott, M., Browne, K., & Kilcoyne, J. (1995). Child sexual abuse prevention: What offenders tell us. Child Abuse and Neglect, 19(3), 579-594. Gallagher, B. (1999). The abuse of children in public care. Child Abuse Review, 8, 357-365. Gallagher, B. (2000). The extent and nature of known cases of institutional child sexual abuse. British Journal of Social Work, 30(6), 795-817. Gillespie, A. (2002). Child protection on the interne: Challenges for criminal law. Child and Family Law Quarterly, 14(4), 411-425. Gillespie, A. (2004). "Grooming": Definitions and the law. New law Journal, 154(7124), 586-587. Grab, A. N., Hobson, W. F., & Gary, T. S. (1982). The child molester: Clinical obser- vations. In J. R. Conic & D. A. Shore (Eds.), Social work and child sexual abuse (pp. 129-144). Binghamton, NY: The Haworth Press. Haiti'', M. (2009). The sexual abuse of boys in organized male sports. Men and Masculinities, 12, 225-249. Herman, J. L. (1981). Farber-daughter incest. Cambridge, MA: Harvard University Press. Howie, D. (1995). Paedophiles and sexual offences against children. Oxford, England: John Wiley and Sons. Knoll, J. (2010). Teacher sexual misconduct: Grooming patterns and female offenders. Journal of Child Sexual Abuse, 19, 371-386. Laudan, L (1977). Progress and its problems. Berkeley, CA: University of California Press. Leberg, E. (1997). Understanding child molesters: Taking charge. Thousand Oaks, CA: Sage. 3502-013 Page 19 of 21 EFTA_00001431 EFTA00156772 976 Ac Bennett and W. &Donohue McAlinden, A. (2006). -Setting 'ern up: Personal, familial and institutional grooming in the sexual abuse of children. Social & Legal Studies, /5(3), 339-362. Mower, L (2012, July 22). Sex assault conviction may be test case for testimony stan- dards. Las Vegas Reviewfournal. Retrieved from littp://www.lvd.cominews/sex- assault-conviction-may-be4ea-case-for-testimony-standards-163325306.html O'Callaghan, D. (2011, November 7). Jailed after grooming teen online: Pacebook groomer is jailed. South Wales Evening Post Retrieved from vevevelexisnexis. contihottopicsaracademic O'Connell, R. (2003). A typology of child cybersexploitation and online groom- ing practices. Retrieved September 2012 from lutp://wwwaLsc.ac.uk/uploadecl_ documentsilLs_PaperJPrice.pdf O'Donohue, W. (2013). Clinical psychology and the philosophy of science. New York, NY: Springer. Rule 702. Testimony by expert witness. (n.d.). In Legal information Institute Retrieved from http://verw.law.mmelLedukules/fre/rule_702 Salter, A. (1995). Transforming trauma: A guide to understanding and treating adult survimors of child sexual abuse. Newbury Park, CA: Sage. Sew, M. (2008). Pedophilia. In D.R. Laws & W. O'Donohue (Eds.), Sexual deviance: Theory, assessment, and treatment. New York, NY: Guilford. Sgroi, S. M. (1982). Handbook of clinical intervention in child sexual abuse. Lexington, MS: Lexington Books. Shakeshaft, C. (2004). Educator sexual misconduct: A synthesis of existing literature (U.S. Department of Education Document No. 2004-09). Washington, DC: U.S. Department of Education. Spiegel, J. (2003). Sexual abuse of males: The SAAl model of theory and practice. New York, NY: Brunner-Routledge. Van Dam, C. (2001). Identifying child molesters: Preventing child sexual abuse by recognizing the patterns of the offenders. Binghamton, NY: The Haworth Press. Vance, A. (2012, August 29). Prison for online grooming proposed. The Dominion Post (Wellington, New Zealand). Retrieved from www.lexisnexis.cozn/hottopics/ lnacademic Wyre, R. (1987). Working with sex offenders. Oxford, England: Perry Publications. AUTHOR NOTES Natalie Bennett, BS, is a graduate student in the Clinical Psychology Doctoral Program at the University of Nevada, Reno. Her current research interests focus on child sexual abuse and assessment of psychological constructs. William O'Donohue, PhD, is a professor of psychology at the University of Nevada, Reno. He is also the clinical director of the Victims of Crime Treatment Center, a treatment clinic for victims of sexual abuse or assault, in Reno, Nevada. He received his MA and PhD from the State University of New York at Stony Brook in Stony Brook, New York. 3502-013 Page 20 of 21 EFIA_00001432 EFTA00156773 Copyright of Journal of Child Sexual Abuse is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listsery without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. 3502-013 Page 21 of 21 EFIA_00001433 EFTA00156774