In Their Own Words : Young Peoples ’ Vulnerabilities to Being Groomed and Sexually Abused Online

Little is known about why some children and young people are vulnerable to being groomed online, yet this has important implications for policy, practice and prevention. Therefore, the aim of this study was to identify factors contributing to a young person’s vulnerability towards online grooming. Thematic Analysis was conducted on eight interviews with young people (six females and two males) who had experienced online grooming, resulting in sexual abuse online and/or offline. The reasons why participants engaged with the offenders varied on most levels of the Bron-fenbrenner (1979) ecological model (including individual, family, community and society). The loss of family protection was found to be central in contributing to vulnerability, as is online risk taking behavior. Three victim vulnerability scenarios emerged: 1) multiple long-term risk factors, 2) trigger events and 3) online behavioral risks. In each of these scenarios, parental and family factors were very important. It is suggested that further research is needed to investigate the applicability of these scenarios to other samples. Recommendations include parent and carer communication and involvement with their children’s internet use, as well as consistent, early onset internet safety education.


Introduction
It is necessary to assess which young people are vulnerable to online grooming in order to ensure that those most H. C. Whittle et al. 1186 at risk are afforded additional protection and preventative education. Whilst a growing body of work considers grooming from the offender perspective (e.g., European Online Grooming Project [EOGP], 2012), relatively little is known about why some young people are more vulnerable to being groomed and abused online, including the role of parental and familial variables in this vulnerability and/or prevention. However, understanding such issues is vital in ensuring that appropriate preventative approaches are developed, in terms of personal awareness, education and public policy.

Review of the Literature
Since no child exists in isolation, considering the risk and protective factors within the different ecological levels (Bronfenbrenner, 1979;Huebner, 2001) is required when assessing a child's vulnerability for any form of harm (Hamilton-Giachritsis, Peixoto, & Melo, 2011). A recent literature review by Whittle, Hamilton-Giachritsis, Beech and Collings (2013a) reviewed victim risk and protective factors found to be associated with online grooming, considering the various inter-related factors which impact on an individual child, including their family, peers, community and wider society. This highlighted that while a single risk factor may not independently lead to a negative experience, recurring/co-occurring risk factors and their accumulation over time will increase the likelihood of harm to the individual (Masten & Powell, 2003;Rolf, 1999;Sameroff, Gutman, & Peck, 2003;Werner & Smith, 1992). Risk factors identified in some recent research are briefly outlined below.
Within their review, Whittle et al. (2013a) found that key risk factors identified to date include being female (Baumgartner, Valkenburg, & Peter, 2010;Helweg-Larsen, Schütt, & Larson, 2011;Mitchell, Finkelhor, & Wolak, 2007a) and in adolescence. Although the impact of gender may be over-stated due to lower reporting rates among boys (O'Leary & Barber, 2008), adolescence is a key developmental stage requiring immense biological, personal and social change (Durkin, 1995;Soto, John, Gosling, & Potter, 2011;Spear, 2010) and the formation of identity (Meeus, 2011). This period of transition is often typified by impulsive and risky behavior in any setting (Pharo, Sim, Graham, Gross, & Hayne, 2011;Romer, 2010), as well as increased awareness of sexuality, experiencing sexual arousal (Choo, 2009;Olson, Daggs, Ellevold, & Rogers, 2007) and, for some, engaging in sexual activity (Mosher, Chandra, & Jones, 2005). Thus, Subrahmanyam, Smahel and Greenfield (2006) found sexual chat to be common place in peer relations online. However, the combination of risk taking and sexual curiosity increases the vulnerability of adolescents towards online grooming.
Indeed, engaging in risk-taking behaviors online (e.g., talking to strangers, sexual behavior, sharing personal information with strangers, meeting online contacts offline), does appear to be correlated with online victimization (Mitchell et al., 2007a;Noll et al., 2009;Whittle, Hamilton-Giachritsis, Beech, & Collings, 2013b;Ybarra, Mitchell, Finkelhor, & Wolak, 2007;Young, Young, & Fullwood, 2007). In addition, Noll, Shenk, Barnes and Haralson (2013) found that young people 1) with behavioral problems; 2) who had experienced maltreatment or 3) had low cognitive abilities were also more likely to engage in high risk behavior online. Thus, adolescents with these additional risk factors might be particularly vulnerable.
The European Online Grooming Project (2012) outlined three varying youth responses to approaches from online groomers: resilient (did not respond to the groomer), risk taking (e.g., sexual chat with strangers), or vulnerable (offline difficulties). However, these findings came from interviews with online groomers, rather than the victims themselves. Based on 27 young people, one of the only studies to interview victims themselves (the ROBERT project; Risk-taking Online Behavior Empowerment through Research and Training) noted that a young person feeling something is missing from their life was a key theme (Quayle, Jonsson, & Lööf, 2012). Both of these projects (European Online Grooming Project, 2012;ROBERT Project, 2012) and others (O'Connell, 2003;Sullivan, 2009) have suggested the need for research on the vulnerabilities of a young person contributing to why they may be approached by or respond to a groomer. Yet there is still limited knowledge regarding exactly what form these vulnerabilities take and which areas of the young person's life are considered to put them at risk (Berson, 2003;Whittle et al., 2013a). Added to that, little is known about the protective factors that might act as buffers to reduce the likelihood of harm and minimize the negative impact of traumatic events (Shoon, 2006). Therefore, we would suggest there is a gap in current research providing an understanding of the complexities of victim-offender interactions within the grooming process and victim, family and community characteristics (Kloess, Beech, & Harkins, 2014;Whittle et al., 2013a). This research aims to explore the vulnerabilities associated with the victims.

Method
The methodology of this study is outlined more fully in the following open access journal: www.hrpub.org/download/201308/ujp.2013.010206.pdf; please refer to this paper for more detailed review of the sample and procedure.
Semi-structured interviews with eight young people who had experienced online grooming leading to online and/or offline sexual abuse took place as part of this research. The sample included six females and two males from across England. The mean age at the time of the grooming was 12.88 years (SD 0.84) and the mean age at the time of interview was 15.88 years (SD 2.17).
Police and social services contacts known to CEOP provided potential participants who fitted the inclusion and exclusion criteria. If the young person and their parent or guardian was happy to be involved in the research, an interview was scheduled. The same key topics were covered in all interviews, but there was flexibility for the participant to focus on aspects that were important to them. All ethical procedures were followed and the research was approved by the University of Birmingham Ethics Committee (Reference ERN_11-0083) and the Child Exploitation and Online Protection centre (CEOP) Research Panel, ensuring all possible steps were taken to protect the participants. Consent forms were completed by the participant, their parent or carer and a professional associated with the case prior to any involvement in the research. Debrief sheets were provided to give avenues of additional support to the participants as well giving further information on the research (including information on confidentiality and withdrawal). All identifying features relating to the participants (e.g., name, names of friends and family, place names) have been changed for this paper and the participants are therefore anonymous.
The interviews were all recorded by a Dictaphone and then transcribed. Thematic Analysis was conducted on all interview transcripts to identify themes across all eight interviews. During this process it became clear that many thematic nodes fitted into elements of the Ecological Model (Bronfenbrenner, 1979). It was therefore decided to use the ecological framework to present the data. Individual journeys of each victim from birth until the end of the abuse were mapped onto paper. Risk and protective factors relating to each individual were then highlighted. Following discussions with the second author, the first author created overview documents for each victim outlining distal and proximal risk and protective factors, as well as distal and proximal consequences of the grooming, all set within the Ecological Model. These documents helped clarify the various factors contributing to each young person's vulnerability prior to the grooming. In some cases "trigger" events became apparent, while in others, the risk factors (and thus vulnerability) appeared to accumulate over the young person's life.

Results
Early analysis clearly revealed that super-ordinate themes could be divided into three timeframes: "pre-offense", "during offense" and "post-offense". The "pre-offense" time-frame relates to any aspect of the victim's life prior to contact with the offender. The "during offense" time-frame relates to events and feelings during their contact with the offender, including all stages of the grooming process and the abuse itself. The "post-offense" timeframe relates to any events and feelings occurring after the abuse and contact with the offender stopped. The results relating to vulnerabilities (predominantly from the "pre-offense" analysis) are outlined below. Due to the volume of data, the "during" and "post" offense data are reported in additional papers (Whittle, Hamilton-Giachritsis, & Beech, In Submission;Whittle, Hamilton-Giachritsis, & Beech, 2013).
Thematic nodes emerged as "risk factors", "protective factors", or "attributes and experiences" (which were deemed neutral or could contribute to either risk or protection). The nodes were consistent with the Ecological Model (Bronfenbrenner, 1979) and the domains within life satisfaction (Huebner, 2001), such as self, family, friends, school, living environment and technology. Thus, they were categorized as such to better understand the themes.
Whilst individual risk factors do not independently lead to vulnerability, accumulation and/or combination of these are considered to increase the young person's vulnerability towards online grooming. When looking across all eight interviews, the volume of risk factors was substantially larger than the volume of protective factors (65 compared to 28). Despite a range of risk factors, all of the victims had some aspects of their lives which contributed to protecting them from online grooming; however these fluctuated and varied considerably between cases. For many victims, it was only when these protective factors were lost (even temporarily) that the abuse took place. Thus, accumulation and combinations of protective factors contribute to protecting the individual, more than one factor independently influencing vulnerability. The most prominent risk and protective factors found within the data set are outlined in Table 1. Please note that all identifying features have been changed and that victim ages are given at the time of the offence.

Risk and Protective Factors Relating to Self
Of the young people interviewed, 88% (n = 7) demonstrated risk factors relating to themselves as an individual, whether temporarily or over a long period of time. Half of the victims indicated that they had hit a low point in life; 75% (n = 7) discussed low self-esteem and 63% (n = 5) expressed loneliness at the time of the grooming, which they considered contributory factors in their rationale for engaging with the offender. For example, "I don't like me bum and that cos it's big and me legs and that. I'm trying to go on a diet because I'm too fat" [Jenna,12]; "I was depressed and on my own and just needed anyone to talk to... Not very happy with the way life is but getting on with it. So it was kinda like this big spark that just come firing at me and I took the, I took it" [Joanne,14]; and "I was all upset about that as well that's why I just needed a friend more than anything" [Mona,14].
One victim described having multiple boyfriends prior to meeting the offender and a willingness to engage in impersonal sexual contact. This could potentially have contributed to her comparatively quick engagement in sexual chat with the offender, when compared to the other cases.

Risk and Protective Factors Relating to Family
By far the most risk factors reported by all the victims related to their family situation. The majority of victims (75%; n = 6) had separated parents (often acrimoniously) and/or came from a reconstituted family. They described fights at home and difficult family relations, including issues with step families. Some victims had always struggled with problems at home ("I didn't get my parents being that loving towards me because they were always too busy getting one up on the other one"; [Chloe, 12]), whereas for others the family difficulties were temporary and clustered around the time the offender approached them online. Thus, whilst nearly half of the young people (38%; n = 3) reported being close to a parent, all three of these young people indicated that this relationship was temporarily jeopardized prior to the offence (sometimes due to illness, bereavement or work); a situation that was often exacerbated by the grooming:  [Joanne,14].
Half of the young people described support from a member of their wider family ("I was dead close to my Nan; she was like the sweetest person you could ever meet" [Jenna,12]) and several of the victims (38%; n = 3) were close to siblings. However, this was not sufficient to protect from the abuse.
Over half (63%; n = 5) of the victims had never discussed internet safety with their parents, several (37%; n = 3) reported that they had no internet restrictions at home and thought their parents lacked understanding of the technology ("My Mum's clueless about the internet and stuff" [Shelley,13]). However, whilst some parents (63%; n = 5) had attempted steps towards online protection, these techniques were largely inconsistent and sporadic. It was mainly Lucas who stated that his parents monitored his internet use ("I think they gave me the usual lectures... Mum always ask me who I'm talking to" [Lucas, 13]) but it was ineffective in preventing his risk taking behavior online. Only 25% of the victims said they would tell their parents if they were worried about something online.

Risk and Protective Factors Relating to Friends
Friendship was the only domain where protective factors out-weighed risk factors, with all victims describing at least one close friend and 88% (n = 7) emphasizing the importance of these friendships ("Claire's [best friend] just everything to me, no one else is anything to me" [Jenna,12]; "I met them at Primary School so we've been good friends for about eight years" [Jonathan,13]). However, in the two cases where these friendships were jeopardized (such as during fights or when the victim experienced bullying), this was a contributory factor to why the victim engaged with or sought support from the offender ("I use to hide with the geeks in the library because I used to be scared to walk down the corridor" [Charlotte,12]; "I tell my best friends everything... I rely on them both so much... [but] at that point I'd fell out with quite a few of my friends at school and stuff and we still hadn't properly made up. The first few days I wouldn't normally have spoken to him, but it was probably just all the situations at the time, I was young, I just wanted friends" [Mona,14]). In the majority of cases (63% n = 5) however, the victim received support from at least one friend throughout the grooming.

Risk and Protective Factors Relating to School
As is the case with most young people, there were mixed opinions regarding school. Over half of the victims (75%; n = 6) in this study discussed enjoying school at times, liking some teachers or generally spoke positively about it (often because they were able to see their friends). For example, "My teachers are really nice and that, proper nice, help you with loads of stuff" [Jonathan,13]; "I really enjoyed it but I think that's because I'm a bit of a geek!" [Mona,14]. However, 38% (n = 3) of victims at times described disliking school, this included discussions of being "naughty" ("I'm naughty... I'm like the clown of the class... I just get told off and I just blame somebody else" [Jenna,12]), feeling stressed by work or truanting "I never liked school, every chance I got I was off!" [Joanne,14]). A few victims mentioned having received some form of generic internet safety education at school ("It was mainly the, 'never give out personal details on the internet' sort of thing, it wasn't the, 'what to look out for and why you shouldn't trust people'"; [Chloe,12]). Some had no internet safety lessons ("I dunno I never really got educated a lot about anything like that [internet safety]; [Jonathan,13]). Only one victim reported having lessons relating specifically to online grooming, but said he did not listen and could not recall what was covered.

Risk and Protective Factors Relating to Living Environment
There were mixed comments regarding living environment, with half of the young people (n = 4) reporting they were happy where they lived, but the others indicating various problems with their local area. Half of the victims got on well with their neighbors, which provided them with additional adult support if necessary.
• "There were riots and stuff down the street. Erm, they was like bringing police down all the time from Strude.
And it was just really hard cos you couldn't hardly go out your door without being started on, you never know if you were gonna get hit with a bat or something like that" [Joanne,14].

• "I've got a very large amount of friends were I live, so I'm always out and about. Like when I get home I'll
be going out straight away. I love where I live at the minute" [Jonathan, 13]. • "Yeah the neighbors were absolutely brilliant" [Shelley,13].
The majority of victims discussed feeling bored where they lived and this may have contributed to their engaging with the offender online. For example, "Just nice to talk to, if you're bored someone to talk to" [Lucas,13]; "I hate it; nothing to do, boring. There's nothing at all, it's just houses" [Shelley,13].

Risk and Protective Factors Relating to Internet Use
Risk factors regarding technology were prevalent in this study, with all the young people speaking to a stranger online at some stage (although for many of the young people this was not a regular occurrence). These were not perceived as anything to feel concerned about: "I didn't really get a lot of friend requests like that, but I'd normally just accept them but I didn't really think anything would come of it really" [Jonathan,15]; "I thought I was popular because people wanted to speak to me" [Lucas,13].
Basic steps to protection online were taken by all of the victims at some time, including protecting personal information online and rarely sharing photos/videos. However, these safety precautions were inconsistently applied and generally evaporated when the victim began talking to the offender, who was seen as "different". For example: "I had my age and stuff [on the profile], I never had my mobile number on it though" [Mona,14]; "It was only for him that I would go on cam and take photos for" [Shelley,13].
Internet access impacted on the vulnerability of the young person and 75% (n = 7) of victims reported spending a long time online at the point when they met the offender (usually to compensate for problems in other areas of their lives); for half of the victims this internet use could be considered excessive. Two victims only had access to the internet in a family room on a shared family computer. However, once in contact with the offender, they waited until they were alone or upstairs before speaking to him and often wiped the conversations from the computer. One victim described how she would only talk to the offender when she was at her father's house, because access was in her bedroom ("At my Dad's my computer was in my bedroom... I had to have the door open but Dad was like rarely in so it didn't make a difference anyway"; [Chloe,12]). In contrast, at her mother's house, she had to use the computer downstairs where her mum looked over her shoulder and thus did not speak to the offender (nor was abused) in this environment.
At least two victims had Smart phones and used these to contact the offender privately, whilst the remainder of the participants regularly spoke with the offender via text or phone call.  [Charlotte,13]. It might be considered that sudden, excessive internet use may be a warning sign.

Development of Victim Vulnerability Scenarios
Despite the diversity of victim experiences prior to abuse, the volume of risk factors identified through Thematic Analysis was extremely prevalent. Comparing and contrasting the overview documents for each victim revealed commonalities regarding how many of the victims were feeling at the point when they began speaking to the offender. These feelings were a direct result of combinations of risk factors. For some victims the risk factors had been building up over the course of their lives whilst, for others, recent events had triggered new risk factors. One of the male victims did not encounter long-term or temporary risk factors in this way, but perceived minimal harm in speaking with the offender online. Thus, examining the victim's circumstances prior to contact with the offender resulted in the emergence of three scenarios which led to the young people becoming vulnerable: 1) Multiple long-term risk factors: Young people who have increasing risk factors in day to day life, with few protective factors and thus take increasing risks online. These young people will be considered vulnerable offline.
2) Trigger events: Young people who have some risk factors but are initially protected until a trigger event or events result in the loss of those protective factors. These young people will be considered vulnerable offline, but only at a certain point in time.
3) Online behavioral risks: Young people who have few risk factors and many protective factors but engage in risk taking behavior despite warnings. These young people will not be considered vulnerable offline.
Participants who experienced multiple long-term risk factors demonstrate consistent and increasing vulnerability throughout childhood on several ecological levels. Of note is the large volume of risk factors relating to the family within this group and it appears that the balance of risk and protective factors acting on the individual gradually tipped in the direction of risk, eroding their resilience. Thus, it would appear that young people with these experiences are potentially vulnerable to numerous negative outcomes and, if approached by an offender online, are less likely to endorse a resilient reaction.
In contrast, participants who experienced trigger events demonstrated a childhood that was relatively balanced in terms of risk and protection, which would not indicate specific vulnerability. However, common to young people within this group was an event or combination of events that temporarily left the young person without the protective factors they were accustomed to, leaving them exposed to new risk. Loss of protective factors, particularly regarding the family, appears to be pivotal and, as a result, the young person would temporarily have lower resistance if approached by an offender online.
Finally, one participant who engaged in online behavioral risks demonstrated no apparent vulnerabilities in his life. In this scenario, the protective factors out-weighed the risk factors and the young person was protected on various ecological levels. However, a combination of curiosity, risk taking behavior online and misunderstanding of perceived consequences, resulted in this individual responding to an offender online. If this finding is confirmed, this is potentially a difficult group with which to intervene due to difficulties in prior identification and the lack of warning signs that will be apparent. However, given the qualitative nature of this study and small sample size (particularly within the third vulnerability scenario), these vulnerability scenarios are tentatively suggested and should be considered a starting point for future research.

Discussion
This research offers useful insights into understanding what can make young people vulnerable to online grooming and abuse. One important point to note, though, is that for all of the young people in this study, the grooming process led to them being victims of abuse, either on or offline. These are not the individuals who manage to either avoid or stop the grooming process before it progresses to abuse. Yet acknowledgement of such a development can seem to be lacking in the literature, which consistently refers to young people being groomed with the later abuse only implied. The term "grooming" implies preparing a child or young person to be abused and the authors therefore contend that it is important that when the grooming moves beyond the preparation, the literature reflects this appropriately and names it as such.
The over-representation of girls within this research sample is consistent with existing research identifying girls as being at the most risk of online grooming (Baumgartner et al., 2010;Helweg-Larsen et al., 2011;Mitchell et al., 2007a). However, the participation of male victims within this study promotes recognition that boys can also be vulnerable to online grooming and abuse. Thus, it is important not to overlook male perspectives and continue to investigate if their vulnerabilities are the same as or differ to those of young girls.
This study supports previous research identifying that risk taking influences a young person's susceptibility towards online grooming (Mitchell et al., 2007a;Noll et al., 2009;Whittle et al., 2013b;Ybarra et al., 2007;Young et al., 2007). Victims within all of the "victim vulnerability scenarios" took risks online to some extent, but for different reasons. The victim who was vulnerable due to online behavioral risks was specifically targeted by their offender. Such targeted grooming creates a risky circumstance where even a small amount of risk taking behavior by the victim has a detrimental effect. For the young person in this situation, despite being comparatively protected in day to day life, this individual took risks by engaging with the offender out of curiosity and in what he perceived to be a consequence free environment. These impulsive and risky actions are consistent with literature surrounding adolescent developmental trends (Pharo et al., 2011;Romer, 2010) and online disinhibition (Suler, 2004). Given research indicating greater impulsivity among males when compared to females (e.g. Chapple & Johnson, 2007), it is possible that adolescent males may take more risks online and thus be exposed to greater risk of grooming.
Similarly, the victims vulnerable due to multiple long-term risk factors also engaged in risk taking behavior. However, the accumulation of risk factors over time is likely to have influenced how these young people used the internet, increasing their risk taking, as the technology continued to provide comfort. Thus, those who were vulnerable as a result of trigger events are likely to have temporarily increased their use of the internet (usually sub-consciously) to provide comfort from the negative effect of the event or events. They were "seeking to fill a void", which was a precipitating factor to the onset of grooming and subsequent abuse. Interestingly, this mirrors the "feeling that something is missing" theme within Quayle et al.'s (2012) study based on victim interviews. Such a finding is consistent with Luring Communication Theory (Olson et al., 2007) which explains that perpetrators use deceptive trust development to approach, groom and isolate victims, creating a cycle of entrapment. Vulnerable victims are likely to get caught in the cycle of entrapment as the perpetrator can exploit their need for attention and affection (Olson et al., 2007).
When compared with the typologies outlined by the EOGP (2012) based on offender interviews, it is interesting that this research has identified some similar (but overlapping) themes based on victim interviews. The "risk taker" category within the EOGP (2012) has some similarities to both those engaging in online behavioral risks and the multiple long-term risk factors. However, the theme from EOGP is more generic, whilst this research has identified the motivations for individuals engaging in such behavior which differs between groups. Furthermore the "Vulnerable" category in EOGP seems to cover both those who experienced multiple long-term risk factors and trigger events within this research, and there are different processes at work for these two groups, which is important for interventions. Thus, it is important to make this distinction.

Implications for Professionals-The Importance of Family Protection
While risks at any ecological level contributed to the vulnerability of these victims, the level with by far the most evidence was risk factors within the family. The majority of young people in this study came from separated families (largely acrimoniously) either occurring during infancy or nearer to the point that they were groomed. Attachment theory suggests that a young person's interaction with their parents/care-givers is fundamental in developing attachment and a consistently supportive relationship from one or more care-givers assists in establishing an individual's self-worth (Bowlby, 1969/82). It is possible that some of the victims in this sample may have developed insecure attachment during family relationship difficulties, which may have increased their vulnerability towards grooming. Family arguments, illness, bereavement and isolation from the family all also contributed to increased vulnerability through loss of family protection. This is consistent with research which identifies dysfunctional family dynamics as a contributory factor that increases the likelihood of a perpetrator selecting a particular victim (Elliott et al., 2007). When these risk factors are combined with issues of adolescence generally (e.g., the development of a sense of self, progression, greater autonomy) the risk intensifies and a young person can be particularly vulnerable. Therefore, teachers, practitioners and other people in positions of responsibility (e.g., sports leaders) should view young people who are experiencing family difficulties (whether temporarily or over time) as particularly vulnerable to online grooming.

Implications for Parents and Carers
It is crucial that professionals working in this field promote the importance of parents and carers communicating with their children about what they are doing online and offering internet safety advice. This of course has some limits (e.g., better monitoring within their own home compared to at a friend's house) but the majority of victims in this study did not discuss internet safety with their parents prior to the abuse and largely kept their online life private from their family. Parental involvement and discussion of internet safety is reported to be a protective factor for young people online (Berson, 2003;Noll et al., 2013;Rogers, Wczasek, & Davies, 2011;Whittle et al., 2013a) and this finding is further supported by the current study. Prior to abuse, only 25% of the victims in this study would have told their parent if they were worried about something online. While some steps were taken by the parents of the victims to help protect their children online, lack of consistent approaches and communication limited their effectiveness. Although using a shared family computer or having internet access in a family room sometimes decreased the ease with which the victims could communicate with the offenders, their access to mobile technology and various internet enabled devices limited the effectiveness of this strategy. Therefore, parental/carer endorsement of multiple protective techniques, including communicating with young people about their online life, monitoring usage and offering supporting internet safety advice, is imperative. Wider societal and Government ratification of key issues relating to internet safety could help facilitate an environment which supports parents and carers in addressing these issues, because given modern technology, parental monitoring is more limited than it might have been in the past.

Implications for Professionals-The Importance of Internet Safety Education
A need for internet safety education for young people has been widely recognized by research (e.g., Berson, 2003;Wolak, Finkelhor, & Mitchell, 2004;Wolak, Finkelhor, Mitchell, & Ybarra, 2008) and this study further supports this notion. Within this sample, the average age of the victim at the onset of online grooming was 12.8 years. None of the victims had a comprehensive understanding of how to stay safe online prior to the abuse. Of the few who received some degree of internet safety education at school, this was reported to be disparate, rare and uninteresting. While a range of e-safety education campaigns are currently being delivered across the UK (e.g., CEOP's Thinkuknow program), e-safety education is not always rolled out consistently across the country. In the context of existing literature reporting adolescents as the high risk group, these participants are at the lower end of the spectrum. Rather than beginning with education in Secondary School (as is sometimes the case), E-safety resources have been produced for children as young as 4 years old and need to be utilized in Primary Schools. This is not currently consistently established, but is required in order to facilitate effective protection for younger children. However, action should also be taken to establish internet safety as an issue within wider society, not just the classroom. By broadcasting existing or new short films through mainstream routes such as television and advertising campaigns, key messages can be accessed by a wider audience, including parents, grandparents and carers. Furthermore, this will provide consistency and further endorse the internet safety lessons delivered within schools and other structured environments.

Limitations
There are several limitations to this study which should be recognized. The length of time between the end of the abuse and when the young person was interviewed varied between one and six years. Some victims are likely to have had weaker memories of events and feelings, given the time that had passed, and this variation in timeframe is likely to have impacted upon their opportunity for reflection. Also, the information given by the two male participants was limited; possibly due to feelings of embarrassment about discussing the experience with a female interviewer or that they did not want to discuss it with any adult. Lucas (behavioral risk participant) in particular reported that he felt minimal impact of the abuse; this may have been the case and a true reflection of those who experienced online behavioral risks or he may have been reluctant to show a more emotional reaction to the interviewer. Due to the qualitative nature of the research, the sample size is small. The numbers within each identified category are also therefore small (particularly within the online behavior risks vulnerability scenario) and thus the application of the findings onto a wider population is questionable. To help combat these limitations, further studies would be beneficial in establishing if the reported patterns are identified with other victims.

Conclusion
This research offers useful insights into understanding what can make young people vulnerable to online grooming and abuse. First, the importance of family relationships has been highlighted in contributing towards risk and protection of a young person's vulnerability towards online grooming. Second, the necessity of acknowledging the progression from grooming to abuse has been emphasized, given the implications that have for impact and recovery. Finally, many of the findings are consistent with existing research in this field; however the development of victim vulnerability scenarios offers new contributions to this area. Research surrounding the vulnerabilities of young people to online grooming and abuse is extremely rare and more research is required, particularly from the perspective of those who have experienced the effects of abuse themselves. Young people across all vulnerability scenarios can be better protected through consistent, collaborative approaches by parents, carers and other adults in their lives. For example, assessing the potential risk and protective factors which impact a young person can highlight individuals who are particularly vulnerable, whether over time or temporarily. Awareness of these changing risk and protective factors by adults interacting with the young person can highlight which young people may need extra support or education at certain times. Specifically, parental involvement and communication combined with substantial and engaging internet safety education from a young age should be considered the key to prevention.