Background

Empirical research on the phenomenon of child sexual abuse in youth-oriented organisations is critical for guiding prevention initiatives. One of the most problematic issues associated with the context of sexual offending in youth-oriented organisations is that it provides a setting in which offenders can repeatedly access children for sexual contact with opportunities for actual abuse and a decreased likelihood of being reported to authorities. Leclerc and Cale (2015) indicated that offenders reported having had worked in an organisation for an average of 16 years before being caught. Another important facet of this phenomenon as pointed out by van Dam (2001) is that many offenders can move from one organisation to another without being reported to authorities if discovered because work colleagues may be reluctant to tell on them due to personal relationships. This context, in part, increases the risk of offenders sexually abusing a high number of victims as indicated in the literature (e.g., Erooga et al. 2012; Leclerc and Cale 2015; Sullivan and Beech 2004). Under these circumstances, any form of empirical data that can facilitate and guide prevention practices is indispensable. One method to provide additional insights into how best to prevent a crime phenomenon from occurring is through asking offenders themselves—an approach coined as offender-based research (Bernasco 2010). In this tradition, the focus of this study is to provide information for the prevention of sexual offenses in youth-oriented organisations through self-reported data obtained from offenders who have sexually abused children in this particular setting.

Theoretical background

Previous empirical research on child sexual offenders in youth-oriented organisations and their offending patterns

Excluding child sexual offenses committed in the Catholic church specifically (see Terry and Ackerman 2008), the bulk of empirical evidence accumulated from sex offenders working in youth-oriented organisations and their offending patterns arises from three main bodies of work (i.e., Erooga et al. 2012; Leclerc and Cale 2015; Leclerc et al. 2005; Sullivan and Beech 2004; Sullivan et al. 2011). In the UK, Erooga et al. (2012) used a sample of 19 offenders who were incarcerated or in the community under the supervision of the National Probation Service. Leclerc and colleagues used a sample of 23 Canadian incarcerated or treated offenders and, Sullivan and colleagues examined a UK sample of 41 offenders in treatment. We briefly review four key dimensions that have emerged from these studies as follows: (1) criminal histories of offenders; (2) offender access to organisations; (3) victim selection; and, (4) offending patterns. These dimensions were selected because they offer a relevant and evidence-based foundation for the current study and have implications for situational crime prevention and the prevention of this phenomenon from a criminological perspective more broadly.

In terms of criminal histories of offenders, Sullivan and Beech (2004) found that in their sample, offenders reported sexually abusing an average of 48 victims up until they were caught. At their first sex offense, 49 % of these offenders were older than 21 years of age and 37 % had never been convicted for a prior sexual offense. In Leclerc and Cale’s (2015) study, the number of victims reported per offender was on average 21 victims. Consistent with Sullivan and Beech, the average age of offenders at their first self-reported sexual offence was 23 years old. Furthermore, of all offenders, 78 % had never been arrested for a sexual offence prior to being caught and convicted for their current offences, which is as twice as many as what was reported by Sullivan and Beech. In a follow-up study, and consistent with Leclerc and Cale’s (2015) findings, Sullivan et al. (2011) added that organisational offenders were less likely to have previous sexual convictions compared to both intrafamilial and other extrafamilial offenders (16 % as opposed to 35 and 61 %, respectively). They were also more likely to abuse a higher number of victims compared to these two groups of offenders. These findings show that a large number of offenders had not been discovered before their participation in the respective studies, which may partly also explain why the number of reported victims is high.

With respect to access to organisations, Erooga et al. (2012) reported that no offenders in their study openly admitted to gaining access to organisations for the purpose of engaging in sexual contact with children. About half of these offenders (53 %) reported that they had no awareness of a sexual interest in children prior to offending, potentially suggesting that a large proportion did. For example, Sullivan and Beech (2004) and Leclerc and Cale (2015) found that more than half of offenders in their sample (57 and 52 %, respectively) sought work in a youth-oriented organisation specifically for the purpose of gaining access to children. Sullivan and Beech (2004) added that an additional 20 % indicated they were not sure whether this was part of their motivation originally and only 25 % clearly indicated that they did not seek employment in youth-oriented organisations for gaining access to children.

Interestingly, Leclerc and Cale’s (2015) findings are consistent with those of Sullivan and Beech (2004) even though the sample used is quite different. The sample used by Sullivan and Beech (2004) was mostly composed of religious-institutional offenders (n = 27), but also teachers (n = 10) and care workers (n = 4). In the case of Leclerc and Cale (2015), the setting in which these offenders committed their offenses included sporting activities, schools, foster care and a youth centre for instance (see “Methods” section below for additional details). These findings do differ dramatically from those of Erooga et al. (2012). Nonetheless, what is clear is that a potentially large proportion of offenders targets youth-oriented organisations to gain access to potential victims. For example, the sample examined by Sullivan and Beech consisted of treated offenders, which may suggest that their findings are more reflective of the reality and as a result, that a large proportion of these offenders actually chose to work or volunteer in a youth-oriented organisation for gaining access to children.

Leclerc and Cale (2015) examined how offenders select specific children in the context of youth-oriented organisations. All of the offenders reported that they would select children whom they knew had had some sort of sexual contact in the past and/or whom they perceived to know a lot about sex. Interestingly, almost all of the offenders also indicated they were likely to select children who they knew to have had attended a class at school on sexuality (94 %). These offenders’ perceptions of children’s familiarity with sexuality are congruent with the fact that offenders also perceived vulnerability in their victims such as the need for emotional support (i.e., indicated by 84 % of offenders in their sample) (Erooga et al. 2012). Taken together, emotional vulnerability in addition to perceived familiarity with sex/sexual activities seem to represent important criteria upon which these offenders are likely to select children for abuse.

In terms of the strategies adopted by offenders to sexually abuse children (sometimes labeled as ‘grooming’) in youth-oriented organisations, manipulation emerges by far as the norm (Colton and Vanstone 1996; Erooga et al. 2012; Leclerc et al. 2005; van Dam 2001). Leclerc et al. (2005) examined the strategies adopted by these offenders to gain victims’ trust and cooperation in addition to maintaining victims’ silence following abuse incidents. In order to gain cooperation, most offenders indicated giving children attention, non-sexual touching, and saying nice things about them while gradually introducing sexual touching into the relationship (100, 96, 96, and 83 %, respectively). The underlying theme emerging from this evidence on modus operandi is that offenders are in close proximity to potential victims and as a result, have the opportunity to develop a trust-based relationship with them often without the need to ask them to maintain silence once sexual activities have been introduced into the relationship (e.g., Leclerc et al. 2005; van Dam 2001). Another interesting finding is that the majority of offenders sexually abused their victims outside of the organisational setting, which again is facilitated by the nature of the relationship they have built with their victims initially. For example, Sullivan and Beech (2004) found that 85 % of their sample took the children away overnight at some point and many offenders (68 %) reported taking the children away overnight specifically for sexual activity purposes. Similarly, Leclerc and Cale (2015) found that 78 % of offenders committed their offenses offsite of youth-oriented organisations they were employed or volunteering at. In particular, these offenders most often used their own homes (52 %), took children for a drive in their car (30 %) or used other isolated places (26 %).

Although evidence based knowledge is still quite limited, we can begin to see a profile of youth-oriented organisation offenders that can be contrasted with more conventional profiles of sexual offenders against children that have emerged from research with offenders in broader incarcerated and clinical settings. For example, they have far less extensive or non-existent criminal histories upon entering youth-oriented organisations. This means that these offenders are possibly more skilled at evading detection, develop specific motivations to offend after involvement with children through work or volunteer activities, and/or possibly a combination of both. For example, a substantial proportion of offenders indicated having sought employment in these contexts for the specific purpose of creating or exploiting opportunities to offend, but many indicated they have not. Of course there is the possibility that many offenders may not provide truthful information along these lines, but this is nonetheless an important pattern to consider. Furthermore, the use of emotional manipulation facilitated by their status and emotional proximity to children is paramount in these contexts. Through the nature of the relationship they can develop with children, they gain the opportunity of exposing them to sexual activities in a manner that eventually seems normal, take them places outside the organisation for sexual activities specifically, and neutralize, to some extent the risk of being discovered. Therefore, taking these patterns and unique contexts into consideration it is necessary to consider what potential prevention strategies may look like.

Situational crime prevention and child sexual abuse

Situational crime prevention represents the core of prevention practices under the framework of environmental criminology. If environmental criminology seeks to understand crime event patterns and criminal opportunities, situational prevention seeks to dismantle them. Broadly speaking, environmental criminology is characterized by the analysis of crime event patterns in space and time, with the key aim of understanding opportunities to offend and preventing crime events from occurring in the first place (Wortley and Mazerolle 2008). This area of research primarily originated from grouping three key theoretical approaches: (1) routine activity approach (Cohen and Felson 1979); (2) the rational choice perspective (Cornish and Clarke 2008); and, (3) crime pattern theory (Brantingham and Brantingham 1978) that taken together tap into understanding criminal opportunities (Felson and Clarke 1998).

Embedded in the rational choice approach, situational crime prevention focuses on the implementation of situational measures to reduce criminal opportunities or control crime precipitators (Cornish and Clarke 2003; Wortley 2001). The main objective of this approach is to prevent crime before it occurs by manipulating the environment into which it is likely to be committed. This objective is based on the premise that the environment plays a role in shaping the decisions and actions adopted by offenders. In other words, the environment is not a passive backdrop against which offenses simply occur, but rather influences what happens during criminal events. A classification of measures has been designed over the years to provide practitioners with a template to use for guidance surrounding the design and application of situational crime prevention measures (Cornish and Clarke 2003). The most recent classification includes the following techniques: (1) Increasing efforts, (2) Increasing risk, (3) Reducing rewards, (4) Reducing provocations and, (5) Removing excuses. With these techniques alone, a total of 25 prevention measures are listed in the classification model developed by Cornish and Clarke (2003).

Situational prevention is a relatively new concept in the field of sexual offending. As a result, studies dedicated to examining situational prevention of child sexual abuse are scarce (see Leclerc et al. 2015). Based on specific type of setting in which child sexual abuse is likely to occur (i.e., organisational, public, domestic) and using the classification of 25 situational prevention measures, Wortley and Smallbone (2006) discussed a number of ways in which situational crime prevention principles could be applied to prevent child sexual abuse. In the context of Catholic Church specifically, Terry and Ackerman (2008) made a number of suggestions for situational crime prevention techniques. Through script analysis, Leclerc et al. (2011) demonstrated the relevance and utility of identifying and understanding crime-commission processes in order to stimulate thinking about potential situational crime prevention measures. Kaufman and his colleagues applied the situational crime prevention model designed by Clarke (1995) to the opportunity structure of child sexual abuse in order to better inform prevention initiatives (Kaufman et al. 2006). More recently, Kaufman et al. (2012) specifically illustrated an approach to obtain input from staff and volunteers in organisations for prevention purposes. To our knowledge, however, very few studies have previously asked offenders about how to actually prevent child sexual abuse.

Gaining insights on prevention by talking to sexual offenders

Only in a handful studies have sexual offenders been asked about their view on how best to prevent child sexual abuse (Budin and Johnson 1989; Elliott et al. 1995). Budin and Johnson (1989) surveyed 72 incarcerated adult sexual offenders of children on the relative effectiveness of a broad range of prevention methods—what works and what does not—to prevent child sexual abuse. Offenders in this sample were almost evenly divided between incestuous offenders and non-incestuous offenders. Most prominently, offenders reported that children should be taught to report if they have been victimized, learn to say no, know about the difference between appropriate versus inappropriate touching and never get into cars with strangers. On the other hand, few of these offenders indicated that shouting, crying or not talking to strangers were effective strategies for prevention. With respect to what parents could do to prevent the sexual abuse of their children, offenders suggested that parents should be emotionally involved in their children’s lives. For example, they indicated that parents should ask their children periodically whether somebody has tried to touch them.

Elliott et al. (1995) surveyed 91 convicted adult child sexual offenders about what they could recommend to children, parents and teachers to prevent child sexual abuse. In this sample, approximately one-third of offenders did not know their victims (34 %), another third knew their victims but were not related to them (34 %), and another third were related to their victims (32 %). In terms of individual protection behaviors of children, these offenders suggested ways to prevent sexual abuse in public places. For instance, they recommended that children should avoid secluded places, never go into public toilets alone, never walk to school alone, not accept car rides/lifts by strangers, knock on the door of a house if they are being followed, always tell their parents where they are going, and tell anyone if somebody abuses them. With respect to what parents could do to prevent abuse, offenders suggested that parents should be suspicious when another adult is more interested in their child than in them, teach children about sexuality and not to keep secrets, and have family discussions about preventing child sexual abuse. Other suggestions included being aware that there is a dangerous age when girls are becoming women and knowing that some people, even family members, could ask them to do sexual things. Regarding what teachers could do to prevent sexual victimisation, offenders suggested that they should have discussions at school to prompt children to disclose abuse, make sure programs do not focus on stranger abuse, have children role play what to do if they were attacked, have advertisements in school about being safe, believe children if they report abuse, and teach sex education.

Offender-based research

The current study is situated in a situational crime prevention framework but from the angle of offender-based research in criminology. Offender-based research involves examining data collected from a sample of offenders (Bernasco 2010). Too often overlooked in social sciences, offender-based research is arguably the most effective method to understand crime events simply because offenders are positioned to provide detailed information that could not be revealed about offending otherwise. For example, police, victim and/or archival data are limited in the ability to account for detailed information about crime events. In fact, only offenders are present from start to finish during crime events, that is, from crime preparation to completion. Therefore, only with offenders is it possible to reconstruct the complete crime-commission process and understand what may have prevented them from acting in a particular way or how they may have overcame obstacles during the course of actions leading to crime. The method of offender-based research is obviously well suited for the purpose of rational choice approach from which researchers seek to put themselves in the shoes of the offender to understand how crime events occur for prevention purposes.

Offender-based research is far from new. For example, Edwin Sutherland (1937) was among the first to investigate crime through the eyes of offenders—professional thieves in particular. During the past few decades, several scholars have interviewed offenders about the crimes they commit, how they select their targets, what may have prevented them from offending in certain situations, and also what, if any information they would provide for prevention purposes. One defining study in this area was conducted by Wright and Decker (1994) on active burglars through snowball sampling. Since then, burglary is potentially the crime that has most often been examined through the lens of offenders (e.g., Homel et al. 2013; Nee and Meenaghan 2006; Rengert and Walsilchick 2000). Other common examples of studies include car theft (Copes and Cherbonneau 2006), drug dealing (Jacobs 1999; Jacques and Bernasco 2013) and armed robbery (Wright and Decker 1997).

Jacques and Bonobo (2015) presented five ways by which offender-based research can be used for informing situational crime prevention. The first way to inform situational prevention is by determining what works to prevent crime. The second way is to find out what prevention measures should be put into place to prevent crime. The third way is to learn about the reasons why a given prevention strategy is effective. The fourth way to provide evidence for situational prevention purposes is to understand how offenders manage to overcome particular prevention measures. The last way to inform situational prevention practices is to collect data on negative outcomes of prevention measures if any, such as crime displacement. The current study focuses on the second recommendation outlined by Jacques and Bonobo. The goal is to inform what prevention measures may be adopted to prevent child sexual abuse in youth-oriented organisations. If the objective of one’s research is to think through the offender’s eyes for prevention purposes, offender-based research is arguably best placed to inform situational crime prevention initiatives.

Current study

Using a Canadian sample of adult sexual offenders who committed their offenses in youth-oriented organisations (see also Leclerc and Cale 2015; Leclerc et al. 2005 for more information on this sample), the current study aims to tap into insights that can be provided by offenders for prevention practices in this particular setting. Budin and Johnson (1989) and Elliott et al. (1995) have previously asked offenders on their views toward preventing child sexual abuse. With a sample considerably larger than the one used in the current study, these studies provided insights into what offenders think in terms of prevention. However, their sample contained a large proportion of stranger and/or incestuous offenders. In this context, it is somewhat difficult to reconcile what these findings mean for preventing sexual abuse in youth-oriented organisations because their recommendations were not specific to this setting. In other words, most of these recommendations may not be directly applicable to sexual offences committed in youth-oriented organisations. Following the argument made by situational prevention theorists (e.g., Clarke 2008), we believe that the more specific the crime data available, the more effective it can be to conceptualize situational prevention measures.

The ultimate objective of the current study is to tap into situational prevention measures that could assist in preventing child sexual abuse in youth-oriented organisations from the perspective of offenders who have actually abused in this setting. First, we examine the answers provided by offenders to the following questions: (1) how to identify potential offenders during recruitment interviews; (2) what policies or regulations to implement in youth-oriented organisations to prevent child sexual abuse; and, (3) what parents could do to reduce the risk of sexual abuse victimisation. Second, we map out these suggestions onto the twenty-five situational prevention measures classification table designed by Cornish and Clarke (2003) to clarify what the data means for situational prevention (i.e., which strategies and measures are reported by offenders). Even though the sample used in this study is relatively small, it has the advantage of being crime specific to the phenomenon under study, which provides an ideal context for thinking of pertinent situational prevention measures. As we are aware of no similar work in the area of child sexual abuse in youth-oriented organisations, this study makes a much needed contribution to current evidence on offending patterns.

Methods

Sample

The sample consisted of 23 adult males who had admitted to committing a sexual offense against a person less than 18 years of age through work or volunteering activities in an institutional context in Canada. These offenders were recruited through treatment centres in the province of Quebec and the Correctional Service of Canada (CSC) in 2002. The research protocols were conducted according to the ethical guidelines stipulated by the Research Ethics Board of the Université de Montréal during the time period in which the participants were interviewed. The mean age of the participants was 49.7 years (SD = 8.6 years). Just over half had never been married (52.1 %) and an equal proportion had a university degree. With respect to sexual orientation, a total of 34.1 % reported being heterosexual, 39.1 % reported being homosexual and 26.1 % indicated a sexual attraction to both males and females. Regarding the setting in which these offenders committed their offenses, eight had gained access to sexually abuse children through sporting activities [i.e., fencing, baseball (2), hockey (2), soccer, gymnastics, softball] and five did through schools [i.e., teacher (4), school bus driver]. In addition, three were in the role of a foster carer (one for child protection services), two were involved in scouts, one offender worked for a Big Brothers association, three for a youth centre, and one for a newspaper delivery agency.

Procedure

Participants were first identified by a contact person in each treatment centre/prison and approached by a practitioner for participation in the study. Those who agreed then met with the first author who invited them to complete a modified version of the Modus Operandi Questionnaire (MOQ) (Kaufman 1991), a questionnaire that emphasizes child sexual abusers’ offending behaviors prior to, during, and following the abuse. This version of the MOQ includes three open-ended questions for which offenders were asked to provide suggestions for prevention practices—the foundation of the current study. The questions were the following: (1) how to identify potential offenders during recruitment interviews, (2) what policies or regulations to implement in youth-oriented organisations to prevent child sexual abuse and, (3) what parents could do to reduce the risk of sexual abuse victimisation. Before participation, participants were told that their involvement in this study was strictly voluntary. They were presented with an information sheet, which explained the research project, its purpose and benefits for research and its potential consequences (e.g., emotional stress) on participants. Each participant signed a consent form, which stated that the information would be used for research purposes only. No incentives for participation were provided. At the time of completion of the questionnaire, the first author was present and assisted each participant if needed in order to make sure they would understand the questions. The administration of the questionnaire was completed in a private office to preserve confidentiality and ensure anonymity.

Results

On the first question, Table 1 shows examples of suggestions made by sexual offenders working in youth-oriented organisations on how to identify potential offenders during recruitment interviews in these organisations. While conducting criminal record checks was the most common suggestion made by offenders, investigating prospective employees’ motivations for working with children was also suggested as an important facet to examine. Others suggestions involved examining candidates past in terms of sexual development, alcohol abuse and work experience. More specific questions on how candidates spend their spare time and their preference for working with children (e.g., hobbies, sports) were also reported.

Table 1 Suggestions made by sexual offenders for preventing child sexual abuse in youth-oriented organisations
Full size table

The second question asked offenders what policies or regulations should be implemented in youth-oriented organisations to prevent child sexual abuse. Interestingly, a number of the responses given by offenders tapped into the environmental design of youth-oriented organisations. For instance, offenders indicated the need to eliminate hidden areas, design windows that overlook corridors and also to install CCTV surveillance cameras. Other responses emphasised the need for more general policies/rules for increasing surveillance, such as never leaving a staff member alone with a child, prohibiting staff from using the same showers as children and, importantly, policies preventing staff from bringing children to their home. Some offenders even went as far as to suggest to prohibit physical contact between staff and children.

The final question asked offenders what parents could do to reduce the risk of sexual abuse victimisation. Many of the responses suggested the need for parents to have a close relationship with their child characterized by open communication. It was also suggested that parents actively participate in the activities of the organisation. Two suggestions focused on the need for prevention education for both parents and children (i.e., teach children to say no and talk about sexuality). Two other suggestions involved the need for children to be suspicious of adults and have more information on offenders.

In summary, the important dimensions that emerged from offenders on how best to prevent sexual abuse in the context of youth-oriented organisations involved attention toward recruitment and targeted screening practices, the physical environments of youth-oriented organisations, strict policies around contact between staff and children, and finally, open and communicative relationships between children and their parents/caregivers. While some of these may seem obvious at first, and even commonplace, others require consideration as to their relevance and how some may best be implemented in youth-oriented organisational contexts.

Discussion

The main objective of this study was to learn about how to prevent child sexual abuse in youth-oriented organisations through the lens of offenders themselves. Adult sexual offenders who were previously convicted for sexually abusing a child in youth-oriented organisations provided recommendations for prevention. We have reviewed these suggestions above. We now turn our attention to discuss how these suggestions could be translated into situational crime prevention measures. The most recent classification of situational prevention measures designed by Cornish and Clarke (2003) is used as a framework to map out ways to prevent child sexual abuse in youth-oriented organisations following these suggestions (Table 2).

Table 2 Mapping out suggestions of prevention made by sexual offenders in youth-oriented organisations onto the twenty-five situational prevention measures framework
Full size table

The first question asked to offenders focused on how to detect potential offenders during recruitment interviews, that is, before they get access to potential victims. In this case, the suggestions given by offenders focused on controlling access to facilities because the objective is to stop potential offenders from getting employed into these organisations in the first place. One such measure is to examine past criminal records, which is a prudent suggestion and likely common practice in many of such organisations. However, there are a few problems related to this measure. First, recent studies have found that substantial proportions of offenders had no criminal record before being detected in these organisations (Erooga et al. 2012; Leclerc and Cale 2015; Sullivan and Beech 2004). Second, Erooga et al. (2012) identified some cases for which these verifications had not been conducted by an organization because offenders were employed before criminal record checks were possible or because offenders sexually abused a child outside a regulated system (e.g., context where parents hired a private tutor for their child). Clearly, criminal record checks should be automatically conducted for candidates seeking employment or volunteering positions in youth-oriented organisations. However, other measures are essential to screen for potential offenders and should be used in conjunction with criminal record checks (Cleary 2012). One suggestion made by offenders is to investigate previous work experiences and ask for references of potential candidates. This suggestion again seems logical and simple to adopt and is probably common in many organisations. However, Erooga et al. (2012) emphasized the importance of seeking information from a number of historical sources. This can be resource intensive and requires substantial support from human resources. For example, some previously detected offenders may try to overcome these measures by not providing truthful or complete references.

Other suggestions made by offenders were related to the content that should characterise recruitment interviews conducted with candidates. Offenders reported asking for specific motivations for working with children and how candidates spend their spare time outside of work. To some extent, these suggestions underlie the notion of emotional congruence with children—a well-established concept in sex offending literature (e.g., Finkelhor 1984). In line with these suggestions, previous literature has also indicated that a preference for spending time with children over adults is a warning sign to identify potential sexual offenders (e.g., van Dam 2001). However, this is a difficult issue to approach from a youth-oriented organisational standpoint because one would expect that some degree of emotional congruence is necessary to work successfully with children in these contexts. In addition, it may be illegal to ask for personal information on activities of candidates in some countries. The question then becomes how should individuals be subsequently screened? Suggestions put forth must be considered in conjunction with a wider breadth of information on prospective employees, and not in isolation (Cleary 2012).

Erooga et al. (2012) also identified a much more profound problem related to interviews. Many offenders in their sample reported that the recruitment interview they had to attend before being given a position contained no discussion with regard to children and expectations of how children should be treated by staff in the organisation. None of the offenders described an organisational culture which was directly focused on the welfare of children. Induction and child protection procedures were not explained to them. Coupled with our findings, this literature suggests that a rigorous step-by-step process for screening during recruitment interviews is needed by organisations. These should focus on children in particular and how best to serve their needs. Cleary’s (2012) model on safe recruitment is consistent with this view.

Cleary (2012) proposed a model for safer recruitment of adults seeking positions in youth-oriented organisations—PICK (Plan, Identify, Check and Keep). This model encapsulates the findings above and follows a four-stage process with at its core rests the notion of a safeguarding culture. This notion refers to the values and actions of organisations and the ability of those in charge of those organisations to effectively safeguard children (Cleary 2012). The first step of the model is the need for planning by organisations. Planning involves providing a clear definition of the role of staff, safeguarding responsibilities and the values and actions of the organisation in job descriptions. Expectations from candidates have to be set clearly as well as the procedures to undertake against those who fail to protect children. Recruiters should be trained in line with the values and actions adopted by the organisation where the protection of children should prevail and tools to assess values, behaviors, motives, skills and experience for working with children developed. For the second step, that is, to identify potential offenders, the model recommends a face to face interview to assess many of the factors just discussed above (i.e., motivations, preferences). The third step where checks are conducted should include the verification of references, criminal record checks and ID checks as well as the completion of a self-disclosure form that emphasises the suitability of the candidate for working with children. Lastly, the fourth step involves keeping the focus on a safeguarding culture through a thorough induction of new recruits, ongoing training, regular supervision and management of performance issues, the use of codes of conduct and whistleblowing policies. As can be seen, the model itself also taps into a range of situational prevention strategies, such as increasing efforts, increasing risks and removing excuses. However, as mentioned by Cleary (2012), a rigorous recruitment process is necessary but again by no means sufficient in and of itself.

The second question asked offenders about policies and rules that should be implemented in youth-oriented organisations to prevent child sexual abuse. Interestingly, despite the fact that a majority of offenders did not abuse children onsite of the organisation (Leclerc and Cale 2015) many of their prevention suggestions had a focus on situations; many suggestions gave rise to a large range of potential situational prevention measures in the organisations themselves. This evidence suggest that such measures may be targeted at detecting manipulation processes in addition to potential abuse incidents. A number of these suggestions are applicable to the measure of assisting natural surveillance by eliminating hidden areas in the organisation, installing CCTV surveillance cameras and designing windows overlooking corridors in organisations such as schools. These measures require the manipulation of the immediate physical environment to facilitate supervision. On a situational level, these measures are important to ensure increased supervision of children and would limit situations in which a child could be left alone with a staff unsupervised. One consideration here is that several of such measures may not be easily introduced in certain organisations, especially in older facilities as this may also involve a considerable financial investment in terms of changes to architectural designs and physical buildings. However, a range of measures specific to the physical layout of certain facilities may be applicable and warrant consideration against financial cost. For example, eliminating hidden areas may also include simply prohibiting access to unused rooms or buildings as these are by definition unsupervised.

Other suggestions made in relation to implementing policies and rules include prohibiting staff from using the same showers (or shower at the same time) as children. This measure could be classified under reducing emotional arousal as showering with children may precipitate or increase sexual arousal of potential offenders. Posted policies/instructions, such as ‘adults only’ and ‘children only’ may assist to avoid this type of situation. These measures seem logical and even obvious, but may in fact be less common than expected especially in organisations that provide little to no guidance or rules to staff regulating interactions and how staff should spend their time with children (Erooga et al. 2012). In fact, the study of Leclerc et al. (2005) indicated that some offenders sexually abused a number of children in this specific context. Another suggestion made by offenders was to never leave a child with a staff alone. This suggestion maps onto the measure of extending guardianship. Accordingly, rules requiring the presence of two staff with a child or a group of children also aim to extend surveillance of potential victims and could be adopted onsite but also offsite for activities taking place outside the organisation (e.g., sport competition trip). Reducing anonymity through uniforms for staff and children could also assist in limiting opportunities for staff to find time alone with children unsupervised especially when traffic of people may be high in the organisation (see also Kaufman et al. 2012).

Offenders also reported that opportunities for staff to develop ‘special’ relationships with children should be limited and that physical contact between staff and children prohibited. The first suggestion makes sense in light of offender modus operandi evidence. Leclerc et al. (2005) showed that these offenders benefit from their status, which provides them with the opportunity to use manipulation to develop intimate relationships with vulnerable children for obtaining sexual contact (see also Erooga et al. 2012). In addition, Erooga et al. (2012) indicated that expectations of staff and their relationship with children were rarely made explicit to offenders in their study. How to prevent relationships leading to intimacy and sexual contact between an adult and a child might be addressed by a combination of measures, such as increased supervision by the presence of two staff with children, eliminating unsupervised areas and perhaps introducing whistleblowing policies that foster discussion of potential issues consistent with a safeguarding culture. Measures, such as improved staff supervision by managers (Erooga et al. 2012) and ethics training (Cleary 2012) may also help highlight the development of inappropriate relationships and reinforce organisational messages about the importance of the welfare of children. From a situational prevention perspective, whistleblowing maps onto assisting natural surveillance, whereas staff supervision by place managers and ethics training are best categorised under alerting conscience measures. Other measures consistent with this suggestion could be to introduce policies for prohibiting the use of mobile phone between staff and children and prohibit gift giving, which refer to controlling tools that may facilitate abuse. For instance, relaxing of rules around the usage of mobile phones has been the starting point for an offender sexually abusing a child in several cases (Colton et al. 2012).

Some offenders suggested to prohibit staff from bringing children to their own home. This suggestion is consistent with evidence on offending patterns and falls into the category of situational prevention measures aimed at setting rules. Leclerc and Cale (2015) found that most episodes of sexual contact between offenders and their victims occurred outside the organisation—most often in the offender’s home. Other studies also showed that many of these offenders take children for overnight trips outside their work for establishing relationships prior to sexual contact or for sexual contact specifically (Erooga et al. 2012; Leclerc et al. 2005; Sullivan and Beech 2004). Based on this evidence and what was reported by offenders in this study, it appears sensible to limit contact between staff and children outside the organisation again by designing policies and rules that would prohibit these situations from occurring. In addition, upon appointment in a role all staff and volunteers should be provided with a copy of the contract with the expectations of the company upon them (Cleary 2012). This document should then be signed by new recruits, returned and retained as proof that they have received a copy of their obligations and expectations (Parkinson et al. 2012; see also Wonnacott 2012). This measure indirectly taps into another suggestion made by offenders, that is, introducing a committee panel for the development and management of policies and rules in youth-oriented organisations involving parents, which falls under the measure of assisting compliance. Controlling access to facilities through procedures for signing in and signing out when arriving and leaving youth-oriented organisations and screening exists through CCTV could also be used for that purpose. Lastly, awareness sessions of recent cases of unacceptable conduct, such as bringing children home, could be used to discourage imitation or reproduction of these behaviors.

Another suggestion made by offenders is for the organisation to disclose the abuse immediately to authorities. This suggestion has to be contextualised to be understood. An issue with underreporting in youth-oriented organisations is that offenders develop personal relationships with other staff and these relationships then neutralise the potential of disclosure when sexual contact between them and children occurs (van Dam 2001). Erooga et al. (2012) discussed this issue as well and reported that staff are unlikely to question offenders, their colleagues, in this context. Underreporting may also be exacerbated by the fact that some children are perceived as troublesome kids in some organisations (Erooga et al. 2012). In this context, place managers (and other persons in a management position within the organisation) could be utilized to facilitate disclosure by staff and children when a suspicious incident arises. Specifically, managers could have the obligation to investigate and disclose any form of sexual contact between staff and children. Children could also be provided with clear communication channels for disclosing abuse to managers. It would be relevant here to not restrict the disclosure process to one person only in case the manager is the offender. However, to maximize the likelihood of disclosure, there is also a need to understand situational factors that could facilitate or obstruct disclosure in youth-oriented organisations. In fact, with a broad sample of adult sexual offenders, Leclerc and Wortley (2015) found that victim disclosure was more likely to increase with victim age for victims who did not live with the offender as opposed to victims who did live with the offender. In other words, being sexually abused by an adult within the home suppressed the increased empowerment that age provided for victims abused by someone from outside of the home. This context may be comparable, to some extent, to youth-oriented organisations. Any adults who are in a position to have direct and ongoing authority over the victim may obstruct disclosure. In their obligation of disclosing sexual contact, management could also be responsible for organising regular external audits to empower staff and children to report suspicious incidents or cases of sexual abuse. This maps onto the measure of assisting compliance.

The third question asked offenders about suggestions to provide to parents for preventing child sexual abuse. Interestingly, one suggestion that stood out was to take interest in children’s activities and participate in the activities of the organisation specifically. In situational prevention terms, the participation of parents in the activities of the organisation could be translated as a measure to extend guardianship. This measure also has the potential of leading to other changes or modifications of the physical environment and policies within youth-oriented organisations because parents would not only ensure better supervision but also likely identify additional measures that may be introduced to facilitate the protection of their own children.

Offenders also suggested that parents and children should be educated on the reality of child sexual abuse. This suggestion is consistent with a number of previous studies that have indicated the need for training or informative sessions for parents on offending patterns and offenders in youth-oriented organisations (e.g., Leclerc et al. 2005, 2011; Kaufman et al. 2006; van Dam 2001). Youth-oriented organisations represent a setting in which opportunities to offend are numerous because potential offenders have authority over, and constant access to, vulnerable children. Staff are also likely to develop personal relationships with a number of children. In this context, disclosure, and under which circumstances it works best, should be a point of focus in training sessions (see discussion above on victim disclosure).

Offenders also made a number of suggestions that focus mainly on how parents can make their children resilient to sexual abuse attempts, which maps onto the measure of target hardening in situational prevention. For instance, offenders reported that parents should have and keep an open dialogue with their children, discuss with them the reality of sexual abuse and teach them to say no. To some extent, these suggestions refer to increasing children awareness but more importantly, to building childrens’ self-confidence and resilience. Once again, resilience building has been discussed many times before in the literature (see a review by MacIntyre and Carr 2000) and represents a reasonable approach in a context where surveillance by others can fail. Fortunately, most parents also seem receptive to being taught ways in which to talk about the topic of sexuality with their child (Wurtele and Kenny 2010). Suggestions made by offenders, such as being wary of adults more generally, does not represent a relevant and sensible measure to us and may lead parents to wrongly believe that any form of relationship between adults and their children is unhealthy and could cause prejudice to children.

Like any other study, the current one has limitations too. Obviously, a number of suggestions made by offenders may already be in place or failed in the past in some organisations, which may limit the immediate relevance of some of the recommendations made by offenders. In addition, as many prevention measures focus on the recruitment process, it may be easy to forget about offenders who did not intend to sexually abuse a child when first recruited—this context requires further research. The small sample size used in this study implies that the situational prevention measures are hypothetical at best and as result, require additional evidence. There are also a number of limitations related to offender self-report, such as the possibility that offenders will lie or minimize what they have done for the very reason of avoiding detection or a longer sentence especially when they have never been convicted. However, it should be noted that limitations are also evident in victim or police-based data (Jacques and Bonobo 2015). Victims may provide inaccurate information to avoid shame, guilt or to get revenge. Some may simply have diminished recollection of past events the longer time goes by. Police reporting practices are also influenced by a myriad of factors. For example, police reports of crime events may vary in accuracy and/or depth and detail. In effect, even though there is a distinct possibility that offenders may provide inaccurate information, there are limitations to self-report data of any kind. Moreover, it is unclear whether suggestions offered by offenders in this study may differ from those that could be provided by active offenders. To our knowledge, evidence-based knowledge on offender modus operandi and situational crime prevention strategies from active offenders is currently inexistent in the literature.

Conclusion

The ultimate objective of this study was to stimulate thinking of situational prevention measures that could assist in preventing child sexual abuse in youth-oriented organisations drawing from the perspective of offenders who abused in this setting. Most suggestions were complementary or consistent with scholarly work, which perhaps provides new avenues for prevention or additional support for the relevance of a number of situational prevention measures. Despite the sample size, the innovative approach of this study makes an essential contribution to understanding a phenomenon that has been substantially overlooked in the field to date (Erooga 2012). By the same token, we see ourselves promoting offender-based research to prevent child sexual abuse in youth-oriented organisations and for examining any form of sexual offending. This approach is consistent with the framework of environmental criminology and crime analysis under which researchers and criminologists put themselves in the offender’s shoes for understanding offending patterns and guiding situational prevention initiatives (Ekblom and Tilley 2002). Too often overlooked, there is no better place to start for understanding offending patterns and thinking of how best to prevent crime. Consistent with promoters of offender-based research (Bernasco 2010; Jacques and Bonobo 2015), we believe that sexual offenders have unique insights to offer for situational prevention.