“ Under His Spell ” : Victims ’ Perspectives of being Groomed Online

The aim of this paper is to highlight key themes within the process of online grooming from the victim’s perspective. Eight adolescents who experienced online grooming were interviewed and data were analysed using Thematic Analysis. It was found that participants, who had been both sexually abused online and/or offline, were subjected to a range of grooming experiences. Consistent grooming themes within this sample included: manipulation; deception; regular/intense contact; secrecy; sexualisation; kindness and flattery; erratic temperament and nastiness; and simultaneous grooming of those close to the victim. These themes are similar to those identified by the literature surrounding grooming offline. Analysis demonstrated that once a participant was ‘enmeshed’ in the relationship with the offender, they were more likely to endure negative feelings associated with the grooming, than if the victim was not ‘enmeshed’. This paper supports the notion that grooming is a varied and non-linear process. Recommendations are made for practitioners, parents and carers, as well as suggestions for primary preventative education.


Introduction
Social media and portable technology has contributed to young people becoming accessible to sexual abusers on a scale never known before [1][2][3][4]. While child sexual abuse has long existed and still exists offline [5], the Internet has altered the way in which abuse can be carried out. The concept of OPEN ACCESS "Under His Spell": Victims' Perspectives of Being grooming was first addressed by UK legislation within Section 15 of the Sexual Offences Act (2003) and this legislative advancement enabled the prosecution of individuals who take preparatory steps towards abusing a child [6]. Following a review, Craven, Brown and Gilchrist ( [7], p. 297) proposed that grooming is: "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 be used as a means of justifying or denying their actions." The prevalence of online grooming is difficult to gauge particularly given the low reporting rates of sexual abuse [8]. Therefore the true figure will be far greater than prevalence statistics indicate, but 64% of 2391 reports made to the Child Exploitation and Online Protection Centre (CEOP) in the UK by members of the public, in 2009 and 2010 were related to grooming [9]. In prevalence studies, Ospina, Harstall and Dennet's [10] meta-analysis found most studies report between 13% [11] and 19% [12] of young people aged 10-17 years had received sexual solicitations online. However, in a recent study of adolescent behaviour online, Whittle, Hamilton-Giachritsis, Bishopp and Beech [13] found 33% of 354 thirteen and fourteen year olds reported having been approached sexually online. However, in most of these studies, the age of the person soliciting was unknown. In contrast, Swedish based research found that 25% of [14][15] year olds had received online sexual solicitations from an adult within the last year [14].
Manipulation techniques utilized during the grooming process can vary considerably [1] and are likely to incorporate both physical and psychological grooming to sexualize the child [7,15]. The process often involves one or several of the following tactics: flattery, blackmail, threats, sexualized games, deception or bribery [4,[16][17][18].
O'Connell [18] was one of the first researchers to identify the process of online grooming, noting the following sequential process: friendship forming, relationship forming, risk assessment (whereby the offender attempts to assess the likelihood of detection), exclusivity (through intensified conversation, a sense of mutual respect is established, with an emphasis on trust and secrecy), and sexual and fantasy enactment. More recently, research has suggested that grooming is more likely to be a cyclical process and offenders do not progress through the stages sequentially [19,20]. In particular, Williams et al. [20] outlined various themes and sub-themes, notably surrounding rapport building, sexual content and assessment of the situation.
However, most research in this area stems from the perspective of the offender (e.g., [19,21,22]), and relies on the honesty of reporting of the perpetrators. However, as part of the Risk-taking Online Behaviour Empowerment through Research and Training (ROBERT) project, Quayle, Jonsson and Lööf [23] conducted research with 27 young victims of online grooming. The research identified five thematic categories within the grooming process: (1) the feeling that something is missing from one's life; (2) the importance of being someone who is connected online; (3) getting caught in a web and making choices; (4) others' involvement; and (5) closing the box and picking up the pieces (whereby victims attempted to make sense of what has happened and move on [23]. Research from the victim's perspective remains scarce, and these themes are yet to be tested and verified by further qualitative research.
Therefore, acquiring knowledge from victims themselves is imperative to better understand the processes involved in online grooming. Thus the purpose of this research was to provide an insight into the experience of the groomed victim, contributing to the limited research base currently available. Specifically, the aim of this research was to explore how the process of grooming takes place online.

Ethics
Ethical considerations were paramount to this research and the study was approved by the University of Birmingham Ethics Committee (Reference ERN_11-0083) and the CEOP Research Panel, ensuring all possible steps were taken to protect the participants. Participants were only interviewed on completion of consent forms from the professional associated with the case, their parent/primary carer and the participant. The professionals were asked to put forward only victims who they believed would have no detrimental effects from involvement. All data was anonymized in that all personal details (including their name, names of others, place names and any other identifying features) were changed before analysis commenced. All references to the interview use the pseudonym and fictional information. Therefore, the only information given about participants is their age and gender. All pre-anonymized transcripts and voice recordings were destroyed.

Sample
Eight young people were interviewed as part of this research; (see Table 1 for participant details). All the young people in this study experienced online grooming to the point of sexual abuse, whether online or offline. The length of time between the offense and the interview varied between participants, from one to six years (M = 3.00, SD = 1.77). There was a considerable range in the abusive experiences of the participants and the offences committed by the abusers. All of the females (n = 6) in this study considered their abusers to be a "boyfriend" at some point during the contact. Both males in this study were groomed by the same offender; the boys knew their male abuser in the offline, but were unaware their online female 'friend' was this individual. The abuser targeted the boys specifically because of their offline friendship. Both males were under the impression they were talking to a female and sent 'her' several semi-naked photos of themselves. Neither of the males considered the abuser to be a girlfriend.
Three of the young people met with their abusers offline and experienced contact sexual abuse: one participant travelled out of her home town with the abuser, another met the abuser twice during a year and the third participant had sex with her abuser several times a week for three months. Of the young people who did not meet their abusers offline, abusive experiences again varied considerably. One victim's abuser was 17 years old and thus the case could be interpreted as 'sexting' (peers sharing sexually explicit photos via technology); however, the characteristics of this case were similar to the findings from others. This case and one other did not result in a conviction. The age of the offenders spanned from 17 up to approximately 50, all were male and white British. Most of the offenders had no previous convictions. The majority of the victims (including two of those who met their offender offline) were incited to send a range of naked photos to the offenders, and some received similar photos in return. One victim was regularly incited to create images and videos during the course of a year, some of which were of an extreme nature (for example, the use of objects in sexual acts).

Procedure
The sample of victims was identified via professional contacts at CEOP, including police officers and social services. An email was sent to approximately 2500 professional contacts who had attended the CEOP Academy training courses during the last five years. Furthermore, professionals who had previously worked with CEOP or who were known to have cases of online grooming were approached individually via phone and email. The professionals were given an information sheet with the following inclusion criteria: (1) the young person is between 12 and 18 years old at the time of interview; (2) the young person had been a victim of online grooming in the past; (3) the case relating to the grooming has now been closed; (4) the victim would like to participate in the research; (5) the professional associated with the case agrees that participation will have no detrimental effects for the young person; (6) the parent or guardian of the young person consents for the interview to take place. The following exclusion criteria were also outlined: (a) the young person is known to have mental health difficulties; (b) the young person has severe learning difficulties. These tight inclusion criteria led to a select sample, but ensured that all cases could contribute to the research aims.
The professionals were asked to put forward young people known through their work, who had been victimized through online grooming. The young person was required to fit the inclusion criteria and be deemed psychologically ready by the professional to participate in the research. Due to poor response rates from professionals and tight inclusion criteria, the young people who were identified as potential participants were limited. Depending on the recommendation, the professional would then either approach the victim and their family or give details to the main researcher to approach the family. The parent/primary carer and the victim were each given an information sheet to help them consider the research. If the victim and their family were happy for the young person to participate, a consent form was given to the professional, the parent and the victim to sign. The first author then corresponded with the victim and their parent to arrange a mutually convenient time and place for the interview.
The victims were involved in a semi-structured interview that took between 45 and 120 minutes, depending on the pace of the individual. The young people were given as much information as possible about the research, prior to their involvement. This was provided via the information sheet, the consent form and explained in person immediately prior to interview. Interviews took place in a neutral setting of the victim's choice; this was either a victim suite at police property or an appropriate room in a Children's Services or charity property. It was stipulated that the researcher would not attend the home of the young person, as it was not deemed a neutral space. The young person was given the option of having their parent or Guardian attend the research interview. All participants declined parental presence. To enhance rapport and informality, only one interviewer was present at and led all interviews (with the exception of the first interview, where a second interviewer was present to assess the quality of the interview structure).
To enable the interviewer to give full attention to the young person, no notes were taken; instead the interview was recorded using a Dictaphone. This was outlined on the information sheet and explained to the young person prior to interview. Before the interview commenced, it was explained that the participant was in control of the interview and could stop or take breaks at any time. Emphasis was placed on the interviewee feeling able to ask for clarification if necessary and informing the interviewer if they would prefer not to answer a question. It was explained that the interview was confidential, with the limits that confidentiality could not be guaranteed if reference was made to a crime that the police are unaware of, which would have to be reported. Information was given regarding withdrawal from the research and that participants could withdraw at any time before a specified date. The interview commenced only after the consent forms had been discussed and any questions had been answered. All interviews followed the same semi-structured format. All interviews began with a free narrative for the participant to summarize what happened. Following this, events were generally discussed chronologically with some questions used to gain information about key themes/issues. During the debriefing, the participant was given a debriefing sheet with further information about the research, a reminder about the process of withdrawing (including the contact details of the interviewer) and details of a range of agencies that could provide additional support if required.

Data Analysis
As soon as possible after each interview, the first author did an exact transcription of the Dictaphone recording. Transcribing the interviews enabled the first author to become familiar with the data. Once the interview was transcribed, all personal information mentioned during interview (including individual names, place names, etc.) were anonymized. These alterations were recorded in a document and stored on an encrypted computer at CEOP (a secure Government building). Anonymous transcripts were then imported into NVivo9 qualitative analysis software. Qualitative analytic approaches vary significantly [24] and are either entrenched within a theoretical perspective, or independent of theory. Thematic Analysis is often considered the foundational technique for qualitative work and is independent from specific theoretical perspectives [25]. Hence, Thematic Analysis seeks to identify, analyse and report themes within a set of data and has the flexibility to explore rich and detailed data without theoretical constraint [25]. For these reasons, Thematic Analysis was chosen as an appropriate methodology for this research.
A bottom-up or data driven technique was utilized for this study, whereby no pre-existing theory was overlaid and consequentially themes were identified purely based on the content of the data. The first author coded each interview in turn by generating initial nodes within NVivo9. Initial coding was considered complete when each line of the data set had been allocated to one or more nodes. Following this, each interview was recorded again to ensure that any nodes created by subsequent interviews were connected to data across all interviews where necessary. At this stage, sections of the text were analysed by the second author to provide inter-rater reliability. The codes identified by the second author were very similar to those identified by the first author and dissimilar codes were resolved through discussion.
Having established a set of single nodes (approximately 800) covering all interviews, themes were searched for across the nodes. Nodes that were very similar or shared meaning were amalgamated into one node (for example, 'family money problems' and 'claimed benefits' were combined to become 'low income family') and nodes that were related to each other were combined to create themes using tree nodes (for example, 'given presents' and 'offered money' were joined under the tree of 'bribery'). Using data from all interviews, tree nodes gradually expanded to connect over-arching, inter-related themes (e.g., offender manipulation techniques). Throughout this process, the first author worked with the second author to discuss emerging themes and analytic progress. Thus, having established a comprehensive set of tree nodes through discussion with the second author, the first author then reviewed, defined and renamed themes until confident that the tree nodes and branches accurately reflected the data.

Results
Early analysis of the interviews produced three superordinate themes based on timeframes; pre-offense, during the offense and post-offense. The results of this paper primarily focus on themes identified relating to 'during offense' analysis. There were a range of sub-themes within the 'during offense' superordinate theme, including: (1) the grooming behaviour; (2) victim feelings at the time; (3) protective factors; (4) risk factors; (5) perpetrator characteristics; and (6) relationship status. This range is demonstrative of the variety of victim experiences of grooming. The grooming varied dramatically in duration and over half of the abusers were grooming other victims at the same time. The sub-themes that generated the most information regarding the process of grooming were 'grooming behaviour' and 'victim feelings at the time'. A sub-sub-theme within 'grooming behaviour' was 'offender manipulation techniques'; this identified a range of grooming techniques with various effects on victims (see Table 2). Please note that all ages given in this paper are the age of the victim during the grooming, not their age at interview.

Conversations
Common to all but one victim was that the abusers initiated contact; in the remaining case, conversation started due to a mutual online friend. In two instances, the first contact from the abusers was sexual; this could be considered atypical to 'normal' peer interactions online. One victim reacted negatively to this but agreed to 'forgive' the offender because he seemed interesting, the other continued speaking to the abuser sexually and non-sexually, before meeting him later that day.
"One day I just logged on and I got a friend request like so I accepted it. And that's like how it all started." [Jonathan,13] "Well he offered me first £200 to sleep with him...(laughs). So he offered me [that] first and I was like I'm not a prozy [prostitute] and everything and so then he went, I'll give you £400 and I went, I told ya I'm not like that, I don't go around getting money off people for sex." [Jenna, 12].
All participants described having normal conversations with their abusers (in a similar way to how they talked to their friends), about general things such as discussing mutual interests, music, their day at school, and others known to them. However, this normal conversation was often in the context of intense levels of contact (see below) and thus may be distinct from regular online friendship formation patterns.
Six victims articulated how they confided in the offenders about problems within their life, particularly family problems, and found emotional support in the relationship. These conversations resulted in the abusers feeling familiar to the victims and built a level of trust. For 63% (n = 5) of the victims in this study, friendship became the basis of a romantic relationship with the abuser. This emotional connecting is likely to have been influenced by the vulnerabilities that the victims were experiencing in their lives and increased their reliance on the offenders. On several occasions, the victims described feeling emotionally attached to their offender relatively quickly; potentially quicker than would be expected with online relationships more generally. Half of the participants mentioned that the conversation generally focused on them (although they did not necessarily notice this at the time). This dimension of the relationship may distinguish theses dynamics from age appropriate online friendships, as non-offending friendships are likely be more balanced and reciprocal.
"I told him everything, he knew just about everything that was going on. Whenever my Mum and me had had a row, the first person I'd tell, I'd start speaking to was him." [Charlotte,12].

Deception
Over half of the abusers lied to the victims about their identities; most commonly the abusers implied they were younger than their true age. In half the cases, victims were sent a photo of an attractive teenager of the opposite sex and led to believe this was the identity of the person they were speaking to. One offender used a picture of his teenage son to groom the victim. The lies told by the abusers were generally believed by the victims at the time yet, with hindsight, all but one victim thought that the abuser had frequently lied to them. Of the victims that were deceived, no substantial suspicions were apparent and all were shocked when they found out the truth.
"He'd find things out about me first and then say yeah I like that too and I like this. So it was like, it was almost like he was using my responses to shape who he was, so he could pretend to be this person that I liked in order to get what he wanted from me." [Chloe,12].
"He made an account that wasn't him; he pretended to be someone else." [Lucas,13].
On a few occasions the victims would have benefitted from questioning and critically evaluating some of the information given by the offenders, as some aspects of who the offenders were purporting to be were inconsistent with realistic peer relationships online. It is likely that victim vulnerabilities prevented this process.
"He told me he was a model, so all his photos on Tagged are of him modelling. So I found out obviously when it all came about, that the photos were actually a model from London." [Shelley,13].
"How he could work two jobs and have loads of stuff at 18? ...The little things that would've if I'd had listened to, would have, might have given some of it away." [Joanne,14].
Of the three victims who met their abusers, two found he was roughly as expected and one was shocked by how much older the offender was. The fact that this victim stayed despite this shock, indicates the extent to which she had been groomed by this point. In three cases, the abusers told the victims they did not have a working webcam, to avoid sending any videos back. The offender using his son's identity had pre-recorded video footage on webcam and streamed this so it would play to the victim, as if it were himself.
"I was a bit shocked 'cos apparently he was meant to be really young and then when I see that he wasn't, I just like stood there and froze." [Joanne,14].
"I'd say well why won't you go on webcam and he'd always be like oh it's broken and then this one time, he, I don't I honestly don't know how he did this, but there was a video of a person, like the person I thought I was talking to." [Chloe, 12].

Regular or Intense Contact
The majority of the victims were in regular and intense contact with their abusers; the prevalence and extent to which this theme was discussed leads this to be noted as a substantial element of the grooming process. The number of methods of contact with the abusers gradually increased in the majority of cases, usually involving instant messaging, texting, webcam use and phone calls. Most victims described talking to their abusers frequently, and often through the night, which appeared to increase the young persons' dependency on the offender and enmeshed the abusers more centrally within the victims' day to day life. The word 'addiction' was used by two victims when describing their need to contact the abusers; this included the victim from the 'sexting' case. The intensity of the contact between victims and offenders is likely to be a distinguishing feature of grooming in comparison to other age appropriate online relationships. Several victims felt that the intense contact was indicative of a loving relationship and regularly spoke of love with the abusers.
"It had got to about, probably about for about 4 hours a day and that's not including texting." [Mona,14].
"By the end we were texting all the, like texting through the day, talking online in the evening and then like a phone call at night, so it was quite frequently." [Chloe,12].
"I was constantly on the phone. I could be in bed and I could be on the phone all night and I'd actually fall asleep on the phone….And like wake up in the morning and the phone would be still on. So it was like an addiction." [Joanne,14].
"I was kind of addicted to talking to him...he'd text me and I'd be online within minutes and then we'd chat, when I had to log off my Facebook, I'd then begin texting him. And that'd probably be about 9 o'clock and I'd text him until 2 o'clock in the morning." [Charlotte,13].
Of the two victims (the boys) who were not communicating with the abuser intensively, relief from boredom appeared to be a contributory factor dictating when they would engage in chat with the offender ("Just someone to talk to really." [Jonathan,13]). The lack of regular and intense contact is likely to contribute to why the boys did not become enmeshed in their relationship with the offender.

Secrecy
In all but one instance, the victims' parents did not know anything about the abusers. In the vast majority of cases, one or more of the victims' friends knew they were talking to a 'boyfriend' or a 'girl' online, but details of the relationships were shared to varying extents. A few victims were warned about the relationship by their friends who became uneasy about the extent of contact the victim was having with a stranger. One friend eventually reported the abuser to the victim's parent because she was aware that the abuser was older and was worried by this. "I told...my Dad's girlfriends' son, he's same age as me... I spoke to him one night, but that was it, I didn't say anything to anyone else." [Jonathan,13].
In some cases the abusers explicitly told the victim not to tell anyone about their relationship, but in the majority of circumstances, the victims described not wanting to tell anyone (particularly their parents) about the offenders. Again, secrecy surrounding contact with the abuser is a unique feature of grooming and is likely to differ from typical peer friendships online, particularly in the cases where the offender insisted that the relationship remain a secret. This could be considered a warning sign for young people. On a couple of occasions one of the reasons for the secrecy was that the victims felt the 'relationship' would be perceived by others as wrong, usually due to the sexual elements of the contact and the fact the offender was a stranger. Thus, a variety of techniques were used by the victims to keep the relationships secret, including: (1) lying about where they were; (2) deleting messages; (3) saving conversations under a different name; and (4) terminating the conversation if interrupted by a parent. Secrecy seemed to make the relationship feel more exclusive, special and exciting to the victim. This lack of sharing information also had the effect of distancing the victim from their family and in some instances, their friends.
"I wanted to keep to myself cos it was cos it was like that nice, you know, the stuff that he was saying, I didn't like wanna tell everybody. So….deep down, it was like my own little ya know, person that likes me." [Joanne,14].
"She (Mum) thinks she's taken my phone but she's taken my decoy phone. I had a little, rubbish little phone that I used when I was around her." [Charlotte,12].
The fact that some of the secrecy tactics employed by the victims were quite extreme highlights the importance of the relationship to the victim and thus how enmeshed they had become.

Sexualisation
All victims experienced sexualisation to the point of sexual abuse online and in some cases offline sexual abuse. This theme was extremely prevalent in the data and should be considered a dominant component of the grooming process. Sexualisation involved one or several of the following: sexual chat, sexual compliments, creating and receiving sexual photos, creating and receiving sexual videos over webcam, and sex with abuser during a meeting. Sexualisation was apparent in all cases, but the process varied considerably. Most victims talked sexually with their abusers and, in the majority of cases, the offenders initiated this.
Half of the victims felt uneasy and pressured by this aspect of the relationship; despite this, these victims and a couple of others also found the sexual elements exciting, at least initially. The majority of victims described sexualisation as a slow force that emerged and intensified over a long period of time (several months in some cases); however one victim engaged in sexual chat with the abuser immediately (and met him later that day), and another after a few days. A couple of victims were originally excited by the sexual side but became scared as it intensified, whilst others who were initially reluctant began to initiate the sexual elements more often over time. For many female victims, sexuality eventually dominated the relationship and it was described often as the principal time when they felt out of control. In contrast, while sexualisation was a component for the male victims, it did not dominate. However, both males talked less openly in interview about the sexual elements of the grooming and this finding may be a reflection of this.
"It was just started off normal chat for quite a while I think and then a bit more provocative." [Lucas,13].
"It went from him saying... you're pretty, to beautiful, to sexy, to f'ing lush, erm, I'd have you in my bed tonight and it went, just kind of through getting more and more...I would have stopped it because I would've realised that we've been, we'd known each other two weeks and we're already doing that. What would we have been like in a month? We'll be already having sex and stuff." [Charlotte, 12].
The quote above relates to the 'sexting' case and the victim explained that she continued sharing sexual images with the abuser because she enjoyed it, however the speed at which the relationship and content of the images was progressing made her uneasy.
All but one victim created sexual photos and/or videos for the abusers and all said these received positive responses, which made them feel good. Photos and videos often increased in volume and became more extreme over time. In half of the cases the abusers also sent photos and/or videos to the victim in return, these received mixed responses from the victims. In most cases the abusers instigated the idea of sharing photos and videos, but there were mixed responses regarding pressures to do so. Several abusers told victims that other young people commonly send sexual photos thus, several female victims described feelings of obligation, concerned that they may lose the abusers' interest and/or love if they withdrew participation, others felt they owed the abusers for their emotional support, but others, (including the boys), said they were not pressured. However, most victims of both genders described feelings of excitement at some point. "There wasn't really any pressure until after I sent one or two like after, like the first one, that was like that was just fun, you know that was just exciting and I had my underwear on so it wasn't like massively bad or anything, I thought oh well it's only like being in a bikini; that was my justification in my head you know....Other stuff happened as well like when he asked me to do things on webcam and you know I did because it was fun and you weren't supposed to do it, which made it even more fun." [Chloe,12].
"I was excited 'cos it was something different, something new. Um he like he apparently a load of people do it so, it was like that's what he said. So it was like an exciting feeling." [Joanne,14].
"But I was on Facebook at the time so I weren't really watching it (a video of the offender masturbating), cos it kind of freaked me out in some ways, it was just kind of argh!...For a 12 year old girl I was quite scared!" [Charlotte, 12].
Two victims had been keen to meet up, but were discouraged by the abusers (who had lied about their identity), although one of these victims was planning the meeting with the offender at the point of police involvement. In some cases the abusers were encouraging the victims to meet.
"Oh I was excited; I thought it would be really fun you know. I got to meet my boyfriend, awesome." [Chloe,12].
Three female victims met their abusers offline and had been eager to do so. All had sexual intercourse on the first meeting. But while two of these young people often spoke quite positively about sexual intercourse with the abusers, the third victim generally talked more negatively it and fought back when the offender attempted to have sexual intercourse on a second occasion. The first two both had previous sexual experience; the third had no previous sexual experience. Despite, the general reflections noted above, victims sometimes made contradictory statements about the sexually abusive part of the relationship; on occasion describing feelings of excitement and enjoyment and other times describing concerns, discomfort and fear. "The second time I just wasn't bothered honestly it's the only way of putting it!" [Mona,14]. "In, some ways it wasn't just all him doing it, it was me too...The police and everything class it as rape and groomed. I don't class it as that because like, I, I loved him." [Jenna,12].
"He was more like shouting and stuff saying like, you should do it and, and stuff like that really, just erm, but I was like standing my ground, I was like I don't, don't want to. Not, I cou, I couldn't do it again. Erm, and I think what was running through my head as I, I was like, it doesn't matter whether he was to hurt me or anything, I just couldn't put myself through that again." [Joanne,14].
Despite the fact that some online peer relationships may well involve sexual elements, the sexualisation of victim-offender interactions within this sample (often instigated and controlled by the abuser) built over time and in several cases began to dominate the relationship. This is likely to be more exclusive to grooming and therefore could be considered a warning sign.

Kindness and Flattery
All victims described the abusers with positive adjectives for the majority of the contact, such as lovely, nice and kind; however this was to a lesser extent for the boys. This may also have contributed to why they were less enmeshed in the relationship. Victims often felt the abusers were fun, good at listening to their problems, interesting and supportive and, in one case, regularly helped with homework. Such interactions made them feel good and loved. This was a considerable theme across the interviews. However, given the clear overlap with typical adolescent relationship development it is difficult to draw distinctions between the two behaviours. Two victims who met their abusers, described the abusers paying for food and taxi fares and one of these victims also received money and gifts from the abuser for her birthday, which she explained made her feel special. This kindness, on several occasions, contributed to the victims becoming enmeshed or hooked on contact with the offenders.
"Genuine, as he seemed like the nice person everyone wants ya to go out and be with, he just seemed really nice to me." [Mona,14].
"Oh no he was lovely, he was just [like]. another teenager who liked the stuff that I liked and was fun to talk to or whatever." [Chloe,12].
"He was really like kind and caring and uh he's a really nice person to talk to um, like you'd like him if you met him." [Joanne,14].
All victims felt familiar with and trusted the abusers, furthermore they all felt unconcerned at the time of the grooming, often describing it as feeling normal. This lack of concern may directly relate to the young people's vulnerabilities and is likely to have decreased the accuracy of a critical assessment of the risks involved in such an online friendship.
"I trusted him with my whole life and everything like that." [Jenna,12].
All but one victim expressed feeling a confidence boost from talking to the abusers, and in all female cases; this flattery was consistent and frequent with most describing feeling special, beautiful, enjoying the attention and becoming 'enmeshed' in the relationship. The girls described considerably more flattery than the male victims. Compliments were predominantly (but not exclusively) based on the victims looks, incorporating both traditional and sexual stances. Sexualized compliments, once established, generally became the most consistently used form of flattery. Half of the victims reported that the abusers were promising them a better future and regularly discussed staying together.
"He'd just like tell me I was pretty and beautiful and he'd tell me that he loved me you know; yeah that sort of thing just that I was sort of a good person and that he liked me." [Chloe,12].
"It was mainly like everything a girl would like to hear, like you're beautiful, um you deserve good things and I can do all that for you and stuff like that" [Joanne,14].
All the female victims in this study felt that the abusers were their boyfriend and half of all participants reported they believed they were in love with the abusers.
"He made me love him well he...yeah he made me love him." [Jenna,12].
"He was kind, made you feel special, made you fall in love with him." [Charlotte,12].
"He'd say he loved me and I'd say oh yeah I love you too because I was 12 and didn't know what that was so like yeah I love you too and yeah we'd just be like that." [Chloe,12].
In contrast, the two male victims described having a mediocre interest in the girl that the abuser purported to be: "Just nice to talk to, if you're bored someone to talk to, just catch up and stuff. Nothing really serious going on, but still friends." [Lucas,13]. This is likely to have helped them stay more distant from their abuser, potentially restricting the types of manipulation the offender felt able to engage in (e.g., intense contact) and/or the extent of their abusive experiences.

Erratic Temperament and Nastiness
Some victims referred to occasions when the abusers had been possessive, jealous, controlling, or blowing hot and cold with them. On occasion this resulted in arguments and victims feeling irritated and annoyed, but more commonly, this resulted in the victim feeling confused and out of control. Some victims mentioned disliking the abusers as a result and the only victim who experienced the abuser becoming nasty in person described feeling scared by him. While typical peer relationships may involve arguments, the one-sided nature of the nastiness and erratic temperament experienced by some of the victims indicates a possible distinction for this type of abuse. Most victims who witnessed an erratic or nasty side to the abusers were anxious to please the offenders in attempts to regain their kindness and maintain the relationship. Such offender behaviour appears to highlight the hold they had over the victims and victims who experienced this type of temperament described already feeling enmeshed in the relationship with the abusers. This form of manipulation equated to one of the few times any of victims described feeling concerned by the relationship.
"If I was in an exam or something, erm if I'd turn my phone off and he'd tried ringing, he was like, oh you're ignoring me even though he knew I was in school. So and then I had to get like, like, loads of messages saying like, "oi bitch, answer your phone" and stuff like that." [Joanne,14].
"There was quite a few times when I said something he didn't like and he's just like oh go away I wish you were dead, all that type of stuff… he occasionally tried to make me feel really bad about myself like oh you're so ugly, you're so fat, I'm surprised more people don't hate you....It was a bit of a rollercoaster, because there were times when I thought I loved him and then others when I hated him and didn't want to speak to him ever again and it went on like that for about a year." [Mona,14].
"It made me feel like he, like I'm not wanted anymore...and it made me like harder fast, because like losing somebody that you think you love it's dead hard." [Jenna,12].
The victim who travelled out of her home-town with her abuser described him getting progressively nastier while they were away, despite having been kind and loving before they met. This was the victim that spoke negatively about sexual intercourse with the abuser and fought back against the abuser on a second occasion.
"He just started going really mad and getting really angry um cos we was on the park on our own and he was just like had like really evil eyes, I just remember his eyes, erm like he grabbed hold of my wrists and then was like well you're not going, you're staying with me." [Joanne,14].
Two victims were blackmailed with threats of showing parents if they did not send more sexual photos, and this emphasized their lack of control to these victims. Blackmail might be considered atypical of normal peer interactions online and thus could be a warning sign for grooming. However, young people need to be provided with information about where to seek help if they have reached the point where they are being blackmailed.
"Because you know you'd done it once or twice they just expect it all the time and then if you try and say like oh I don't want to talk about that or whatever, he'd like threaten or black(mail), like I'll send your Dad all the chat logs if you don't." [Chloe, 12].

Grooming Others
On several occasions, victims' friends were also in contact with the abusers, either when with the victim or as a means of the offender getting in contact with the victim if they were not responding. At points, the victims were warned to be careful by a friend, but more typically, the friend found the relationship exciting, as they too had been groomed. In the one case where the parents did know their daughter was talking to the abuser, they too were regularly speaking on the phone with him and were manipulated by him. This contact with family and friends led the victim to feel increasing familiarity with the abuser and misplaced trust. The two male victims knew their abuser offline, and the families knew and liked this man offline. In the case of Jonathan, his mother trusted the offender to the extent that he regularly babysat for Jonathan and his brother, including an overnight stay on one occasion.
"They knew that we were talking to each other and like he was sending me messages saying like I love you and stuff like that. And my mum knew about that 'cause my mum had, my mum would speak to him as well, like when he's speaking to my mum, he'd say like um I really, really like your daughter, I think she's lovely, I can't wait to come and meet you all and like um, just stuff that a normal lad would say when they're seeing your daughter….So it was like nothing out the ordinary to my mum or dad." [Joanne,14].
"I knew he was a [person in authoritative role] and I knew him as like well kind of a friend I suppose, cos I know he looked after me a few times when my Mum was working." [Jonathan,13].
All of these grooming techniques and subsequent effects on the victims (see Figure 1) increased the chances of the victims becoming enmeshed in the relationship. All female victims in this study implied they were 'hooked' on their offenders, for some this even transferred to after the arrest when they still believed it had been a relationship. Such victim beliefs gave the abusers increased control over the relationship and the young person. In contrast, the male victims were subject to fewer grooming techniques or often to a lesser extent, thus neither felt enmeshed; in particular one of the males appeared relatively uninterested in the relationship.
"We were like puppets, once we were under his control, anything he said would make me, would make me do it." [Shelley,13].
"Once they've got you in that certain position they want ya, they can do anything with you because you didn't know that you loved them and you'd do anything for them." [Jenna,12].

Discussion
The results of this study illustrate the range of experiences of victims of online grooming and highlight that grooming within this sample was not a linear or homogeneous process; offenders did not move through the phases of grooming in any particular order. Instead, it may be cyclical by nature and groomers may adopt, relapse and re-adopt various manipulation strategies as necessary. Despite the heterogeneity of victim experiences, the themes above highlight commonalities in the way that the adolescent victims in this study were groomed and these techniques often had similar effects on the

Possible Grooming Techniques
Grooming Others victims. Given the age group of victims within this sample and the potential for offenders to tailor their grooming style to fit what they know of the victims; the results are discussed in the context of grooming adolescents (rather than children). This study cannot make inferences about the prevalence of online grooming, but while recent US research suggests that online sexual solicitations are in fact decreasing [26] and most forms of harassment online actually come from peers [27], the current study highlights that online grooming by adults remains an issue requiring greater understanding. Existing literature recognizes grooming others can be a phase within the grooming process [7,20,28], and this is apparent through many of the victims' experiences within this study. This study suggests that any phase within grooming (e.g., sexualisation) can occur extremely quickly and is likely to vary. This finding is consistent with existing literature which recognizes variation in grooming periods [29]. Furthermore, the fact less than half of the victims in this study met with their offenders offline, is supportive of the notion that not all offenders are contact driven [1].
The most commonly utilized online grooming strategies experienced by the victims in this sample were manipulation through conversation, deception, regular/intense contact, secrecy, sexualisation, kindness and flattery, erratic temperament and nastiness, and grooming others. These grooming techniques are similar to those identified by previous research from both an offender perspective [17,18] and a victim perspective [16], although the use of deception was more prevalent within this study than previous [30]. Interestingly, this grooming process and techniques within it are similar to those identified as being used by offenders' offline [7,15]. This study therefore provides support for the notion that grooming behaviour remains constant, irrespective of the environment and that child sexual offenders utilize technology both to facilitate access to victims and facilitate the abuse.
However, one key element of the current research is that different manipulation techniques are utilized by different offenders (sometimes differing by victim) and that, although regularly sharing similar themes, the process of grooming is a unique experience to each victim. Notably, however, the experiences described by the victims in this study appear to be consistent with (but also go beyond) the themes of rapport building and sexual content, as outlined by Williams and colleagues [20]. Furthermore, the findings correlate with the theme of 'caught in a web', as identified by Quayle et al. [23], which includes: seeming like a normal relationship, telling lies, being groomed, losing control and betrayal. The current research found that victims experienced these phases in different ways and at different points within the grooming process.
The grooming techniques endorsed by the abusers in this study generated a range of immediate positive and negative effects on the adolescent victims. The victims commonly felt immediate positive effects of the grooming such as trust, love, attention, support, excitement and confidence boosting; these tended to enmesh the victim and establish the abusers hold over them. In some cases, the grooming techniques gradually evoked more overtly negative effects on the victim such as fear, confusion, lack of control, and distancing from family. The victims who were already enmeshed with the abusers endured this and continued contact, despite experiencing more overtly nasty grooming strategies. These victims were focusing on maintaining the relationship and seeking to regain the positive effects of it. Notably, the two males in this study did not report any nastiness or erratic temperament by the abuser. It is likely that the abuser was aware the boys were not yet enmeshed in the relationship and therefore such grooming techniques would not have been tolerated at this stage.
Using grooming techniques which evoked overtly negative effects on victims who were not enmeshed in the relationship could result in the loss of the relationship entirely. This finding contributes to knowledge of grooming as it indicates there is likely to be a 'tipping point' at which the offender knows the victim is enmeshed.

Implications of This Research
The heterogeneous nature of grooming as reported by this study is a challenge for preventative strategy; however this study generates important implications for prevention. Sexualisation was a key theme within this study, evoking a range of complicated and often contradictory feelings for the victims. The internet gives young people new ways to explore their sexuality and adolescents are known to behave sexually online [31]; thus the findings from this study must be viewed within the context of sexual behaviour between peers. For some young people, however, even relationships with adults who clearly state their sexual intentions can be perceived as a romantic relationship. Some of the young people in this sample believed they were in a relationship with the offender, therefore education messages which distinguish age appropriate relationships from inappropriate sexual relationships could contribute to prevention.
Consistent educational awareness-raising is required to highlight the uncertainty about with whom you are speaking online, as well as the permanent nature and digital footprint of everything posted online (whether via text, apps, photos or images). This will help young people to make more informed decisions about how they interact online and consider the potential consequences more fully. Empowering young people to risk-assess relationships and recognize potentially abusive elements, will enable them to seek help at an earlier stage.
However, it cannot be overlooked that many of the grooming techniques identified and subsequent effects on the victims are typical of adolescent friendship/relationship development. Parallels between grooming behaviour and typical adolescent relationship forming online are problematic for preventative education, as warning signs of grooming may be discreet or in some cases, virtually non-existent. Despite this, some of the themes identified by the analysis in this study are likely to be indicative of the subtle signs of grooming. In particular, intense contact, nastiness and erratic temperament, encouraged secrecy, disproportionate focus on the young person's life during conversation and sexual elements (especially when instigated, persisted and controlled by another) should be considered warning signs, most notably when a combination of these factors are experienced. Educating young people about these indicators will contribute to protection. Furthermore, extending sex education to incorporate sexuality online and the inherent risks could further equip young people with the knowledge to assist them in protecting themselves.
A small number of victims discussed that, with hindsight, some of the offenders' lies did not make sense; therefore teaching young people to objectively evaluate online friendships and relationships, regardless of emotional attachment may help them to identify risks which would be otherwise overlooked. If the victims had had a better understanding of the intricacies of online grooming techniques, how quickly it can progress and the consequences, there is a greater possibility they would have been suspicious and better equipped to resist the abusers. Young people learning directly from the experiences of young victims could assist them in understanding these subtleties and being able to recognize the warning signs among their peers. On the occasions when the victims in this sample did feel worried about the relationship (e.g., when Chloe was being blackmailed to send more photos), they were not equipped with the knowledge of how to stop the cycle. Educating young people about how to report concerns about online relationships, at any stage, is imperative.
Within this sample, other than a few concerns regarding specific issues, most victims felt largely unconcerned by their relationship with the offenders, and on occasion this lack of concern extended beyond the point of sexual abuse. Despite current research suggesting that disclosure rates are increasing [27], all cases of grooming in this study only ended due to someone reporting on behalf of the victims or the offender getting caught as part of a wider police investigation. This lack of victim suspicion is likely to be associated with the parallels between the offenders' grooming techniques and typical adolescent peer relationships, but also demonstrates the extent to which victims had been groomed. Similarly, two out of eight victims experiencing blackmail is a relatively small proportion, which further indicates the offenders' confidence and ability to incite sexual acts without the need for blackmail. The emotional hold some offenders had over their victims mirrors aspects of interpersonal violence within relationships. Therefore, it is imperative that young people are educated to also look out for signs of grooming among their peers, who may be too enmeshed within the process to identify the warning signs.
Aftercare of victims of sexual abuse can be enhanced by better understanding the effects of sexualisation on the victims; in particular the confusion victims may feel. All the victims in this study felt an element of attraction to their offenders or who the offenders were purporting to be, and all the girls considered the offenders to be boyfriends at some stage. The emotional connection and sexual relationship (online or offline) that the victims may have with the offenders needs to be recognized and sensitively addressed by professionals who provide after-care and support.
For parents and carers, there are also important lessons to be learnt. The speed with which grooming can occur needs to be emphasized and that, for all of the young people in this sample, the grooming progressed to experiencing sexual abuse online, sometimes also offline. The additional difficulty for protection is that parents may believe their child is safe upstairs in their own home and, in many cases, may remain naïve to the potential dangers in the online world. This has become more complicated by the rise in smartphones and other technologies allowing young people easy and on-going access to the internet outside of the home (including in friends' homes) and when they are to some degree beyond parental control. Despite being a difficult issue, potential risk could also be identified by encouraging young people to communicate with their parents and carers (or other trusted adults in their lives) about their online activities. If adults can facilitate these open discussions (without insinuating blame), then it is more likely a young person will discuss online relationships, thus giving adults the opportunity assist with risk assessment and intervene if necessary. As was the case for one victim in this study, parents are also at risk of being groomed by offenders and thus awareness raising and education for parents and carers is also imperative.
For child protection professionals, teachers, parents and adolescents themselves, there is a need to balance the huge advantages to be gained through the medium of the internet with the need to keep young people safe and this can only be achieved, in part, through greater awareness of the process of online grooming and abuse.
In light of the fact that this research found that offenders were required to use different strategies with different victims to keep them engaged, additional analysis has considered why some young people are more vulnerable to online victimization. In particular, analysis has considered whether vulnerabilities are the same as or different to those for offline victimization and whether these different vulnerabilities are related to the impact of the grooming and abuse on victims (see Whittle, Hamilton-Giachritsis, & Beech [32]).

Limitations of This Research
There are several limitations of this study that must be considered when interpreting results. Only one interview session was scheduled with each victim and while all possible attempts to achieve and maintain rapport were made, it is possible that the victims may have held back information on the basis of not knowing the interviewer. This is likely to be particularly relevant with the two male participants, who generally provided less information. It is possible that the male victims felt less comfortable discussing the offense generally or that it was specifically due to having a female interviewer. As a result of this, the finding that they were less enmeshed in the relationship is tentative; as it is possible they did not wish to share the extent of their feelings in interview.
Another limitation is that the length of time between the offense and the interview varied between participants, thus victims are likely to have been in different stages of recovery. Consequentially, the memories and reflections of each participant are likely to vary in quality. In addition, the participants within this sample were selected by professionals who felt their involvement would have, "no detrimental effects for the young person." Therefore, this sample is unlikely to be representative of all victims of online grooming and sexual abuse, given that those who were displaying current psychological difficulties would not have been included. Similarly, the victims were aged between 13 and 18 at the time of interview and at different developmental stages. This is likely to impact on the way in which they recall and interpret their experiences.
Finally, perhaps the main limitation of this study is the relatively small sample size. Due to the qualitative nature of the research, the sample size of eight victims is small and therefore applicability of findings to wider populations is tentative. The authors sought additional participants but encountered difficulties in recruiting further. The diversity in the experiences of the victims may be a reflection of their individual differences (e.g., gender, perception of the relationship, age of the offender, etc.); consequentially research which explores victims who are more similar to one another may offer a more robust contribution for that specific group. Further studies could also incorporate a larger sample of victims and compare the themes with those identified in this research. Additionally, studies comparing victim experiences to a wider sample of non-abused counterparts would help explore the links between grooming behaviours and typical adolescent relationship development online.

Conclusions
This research supports the notion that online grooming of adolescents is varied and cyclical, involving the abuser adopting and re-adopting a variety of manipulation techniques throughout the process. These techniques may include manipulation through conversation, deception, sexualisation, regular/intense contact, kindness and flattery, erratic temperament/nastiness, grooming others and secrecy which are notably similar to offline offender grooming strategies. Effects these tactics are likely to have on the victims include feelings of familiarity, love trust, confidence boosting, emotional support, excitement, but also, lack of control, confusion, reliance on the offender and distancing from family. Victims who considered themselves enmeshed in the relationship were more likely to tolerate immediately negative effects of the grooming, in the hope of maintaining the relationship and regaining the positive effects. Further research is required, from the perspective of the victims, to gain a deeper knowledge of grooming, what the impact is on the young person and how professionals can better respond to and prevent this form of abuse in the future.