Prevention Nearby

Abstract

The main aim of this study was to examine the effect of a potential guardian on the severity of child sexual abuse. Using data obtained on crime events from adult child sexual offenders incarcerated in Queensland (Australia), mixed-effects logistic regression analyses were conducted to examine the effect of potential guardianship on the severity of abuse. Controlling for victim and situational characteristics, the analyses showed that the presence of a potential guardian reduced the duration of sexual contact and the occurrence of penetration. Presence of a potential guardian decreased the risk of sexual penetration by 86%. The study highlights the importance of the presence of a potential guardian for reducing the severity of child sexual abuse, and suggests more broadly that guardianship may be an important protective factor in sexual offending.

Keywords

sexual offending child sexual abuse crime events routine activity approach guardianship capable guardian situational crime prevention

Introduction

Three important facts render child sexual abuse difficut to prevent. First, child sexual offenders often commit their offense in their own home (Elliott, Browne, & Kilcoyne, 1995; Lang & Frenzel, 1988; Wortley & Smallbone, 2006). Even when offenders use multiple locations for repetitive sexual contact with the same child, the offender’s home is the most common location (Leclerc, Wortley, & Smallbone, 2010). Second, child sexual offenders often abuse a child who they know and with whom they already have a close relationship, allowing them to adopt a range of hard-to-detect manipulative strategies (e.g., Leclerc, Wortley, & Smallbone, 2011; see also Kaufman, Hilliker, & Daleiden, 1996). Third, initial disclosure of the abuse to someone else must initially come from the victim unless a person witnesses or otherwise detects the abuse and decides to report the crime to authorities. Moreover, to disclose the abuse, the victim must first be aware or recognize the incident as sexual abuse and then find a trusted adult to whom to report the abuse (Berliner & Conte, 1990). According to London, Bruck, Ceci and Shuman (2005), the majority of abused children do not report their abuse in childhood. In their review of studies on disclosure of child sexual abuse, London et al. (2005) report a modal self-disclosure rate of only 33%. In the light of these figures, a person who is present during the abuse has the potential to play an important role in protecting the child.

Any person who is supervising the child and is willing to confront an adult sexual offender by directly intervening has the power to dissuade or stop him from sexually offending. In this case, this person is a “capable guardian” (see Felson, 1995). However, the mere presence of another person in vicinity of the crime location—even if that person is another child—may be enough to prevent an escalation of the severity of sexual acts performed by the offender. In this context, this person could be described as a “potential guardian.” In addition, even in the event that this person is not willing, or does not have the physical capacity, to stop the offender during the crime, the offender can never be sure whether this person will report the abuse or not. While playing a supervisory role naturally increases the capability of guardians to intervene (Reynald, 2010), the availability of a potential guardian may still have a preventive effect. The aim of the present study is to examine the effect of the presence of a potential guardian on the severity of child sexual abuse as measured by the duration of sexual contact and the occurrence of penetration.

Theoretical Background

Routine Activity Approach

The routine activity approach argues that the risk of victimization is strongly influenced by the everyday life routines of individuals. According to this perspective, three minimal elements must converge in time and space for a crime to occur (Cohen & Felson, 1979)—a motivated offender, a potential victim, and the absence of a capable guardian. The concept of “capable guardian” refers to the presence of any person who has the capacity to interrupt crime commission either directly or indirectly. According to Felson (1995), a person can prevent crime simply by his/her presence, that is, by being visible and/or in close proximity to the offender. As Felson (2002) also notes,

a guardian is not usually someone who brandishes a gun or threatens an offender with quick punishment, but rather someone whose mere presence serves as a gentle reminder that someone is looking. With a guardian present, the offender avoids attempting to carry out an offense in the first place. (p. 22)

A capable guardian against crime is most likely to be someone on the spot when the crime occurs, for example, a partner, relative, coworker, or even a passerby.

Additional developments were made to the routine activity approach in the mid-1980s and 1990s. A problem analysis triangle based on the routine activity approach was developed to guide prevention efforts (Clarke & Eck, 2005; Felson, 1995). The problem analysis triangle comprises inner and outer levels. At the inner level, each of the three sides represents a component of the routine activity approach (i.e., the offender, the victim, and the place). At the outer level are three types of crime controller—a handler, a capable guardian, and a place manager (Clarke & Eck, 2005). Each crime controller is responsible for exerting influence over the corresponding crime component of the inner triangle. An intimate handler is someone who, because of a close relationship with the potential offender, can control or contain his or her propensity to offend (e.g., parents, relatives; Felson, 1986). Capable guardians are responsible for protecting potential victims or other crime targets that they themselves own or that belong to others. A place manager supervises the locations or premises in which crimes may occur (e.g., bar owners, hotel maid—see for instance Eck & Weisburd, 1995). The crime triangle shows in visual form that crime prevention can be achieved through the effective supervision of the offender, the victim, and/or the place where crime is committed.

Guardianship and Child Sexual Abuse

To our knowledge, the only study to have had focused on whether offenders sexually abuse children in the presence of others was completed by Underwood, Patch, Cappelletty, and Wolfe (1999). Underwood et al. (1999) administered a questionnaire to 113 adult child sexual offenders at two community treatment programs. The questionnaire comprised 13 items focusing on the presence of others during abuse. They found that more than half of their sample (54.9%) abused a child when another child was present. A further 23.9% abused a child when another adult was present, and 14.2% did so when another child and adult were present. These findings challenge the assumption that sexual offenders always try to avoid being observed by others, and show instead that offenders will sometimes abuse a child in the presence of others. On the other hand, Beauregard and Leclerc (2007) found that most sexual offenders do estimate the risks of apprehension before committing an offense by taking into account factors such as the presence of others. In any event, it is possible that the presence of somebody nearby may in some cases be enough to deter sexual offenders from committing an offense, or more specifically, from performing severe sexual acts such as penetration.

Finkelhor’s (1984) theoretical model highlighted that an absent or ill guardian, such as the mother of the child, and the inadequate supervision of a child are conditions that help offenders overcome external barriers that protect children from abuse. In child sexual abuse, depending on the circumstances, prevention may only be accomplished by the presence of other children, especially as this crime is likely to be committed in a private home. It is clear that children may not act as capable guardians as they do not have the capacity to supervise or physically intervene. Nonetheless, children may act as potential guardians. Indeed, a child may, merely by their presence, to some extent dissuade the offender from having sexual contact with another child or at the least from involving the child in more intrusive and extensive sexual activities. This is because, from the offender’s perspective, performing intrusive sexual acts in the presence of other children may increase his risk of detection. In addition, the offender cannot be certain that other children will not tell others about the abuse. Even if the offender succeeds in convincing the child who witnessed the abuse to maintain silence, the risk remains of being reported and apprehended at some later point in time.

The Effectiveness of Guardianship in Previous Studies

To our knowledge, the effectiveness of guardianship has never been examined in child sexual abuse. However, the potential effectiveness of guardianship has been studied previously for property crimes such as burglary (e.g., Garofalo & Clark, 1992; Lynch & Cantor, 1992; Miethe & Meier, 1990; Tseloni, Wittebrood, Farrell, & Pease, 2004; Wilcox, Madensen, & Tillyer, 2007). Even though these studies used indirect measures of guardianship (e.g., home occupancy), evidence indicated that guardianship is effective in preventing property crime. The preventive effect of a guardian has been examined less extensively for violent crimes. In one study, analyzing the offender–victim interchange in homicide, Luckenbill (1977) examined the different roles taken by bystanders (or guardians). It was found that bystanders were present in 70% of cases and that they assumed one of three roles: hostile to, neutral to, or supportive of the homicide. In a hostile role—35% of cases—bystanders tried to stop the offender, assist the victim, and immediately notify the police. These bystanders were generally persons who were supportive of the victim before the violent interchange. Bystanders were neutral in 17% of cases and supportive of the offender in 52% of cases.

An interesting study on situational contexts and guardianship in violent crimes was conducted by Hart and Miethe (2008). Data from the National Crime Victimization Survey were used to examine the situational contexts in which bystanders (or guardians) were present in violent crimes and whether a bystander helped or not during these crime events. The perceived effectiveness of bystander intervention was measured by the victim’s assessment of whether the bystander helped or worsened the incident. It was found that a bystander was present in 83% of stranger assaults committed without a weapon in public places during night. Hart and Miethe found that guardians were more likely to help in sexual assaults when the offender was not carrying a weapon. For sexual assaults and robberies, daytime and assaults committed by strangers were common elements of situations in which the bystander was likely to help the victim. This study shows that guardianship operates effectively under a set of specific circumstances for each particular type of crime. Thus, no clear conclusions can be drawn from this study (or studies on property crimes) with respect to the potential effect of guardianship on child sexual abuse.

The Present Study

This study is the first to investigate the effect of guardianship on the severity of child sexual abuse. As pointed out by Felson (1986), the structure of decisions made by others may prevent the offender from pursuing his goals, regardless of whether these other persons know they are preventing a crime from occurring. An initial step in the sexual violence and abuse field is to examine whether a person (adult or child) can serve as a potential guardian simply by being present, that is, without necessarily having the willingness or the capacity to physically intervene. The availability of a potential guardian is the key stage of guardianship intensity and thus the main foundation from which we can complete an investigation to first examine the effect that potential guardianship may have on the severity of child sexual abuse (i.e., duration of sexual contact and occurrence of penetration).

Before analyzing the effect of potential guardianship on the severity of abuse, we examine the characteristics of potential guardians according to the location in which the abuse took place. We also examine whether specific offense characteristics are more likely to be present according to the presence of a potential guardian. Next, we analyze the effect of potential guardianship on the severity of abuse, controlling for victim and situational characteristics. Specifically, we include a number of relevant offense characteristics, namely, the time when the offense was committed, the offender–victim relationship, the location of the crime, and victim characteristics. This analytic strategy is consistent with the routine activity approach (and other perspectives focusing on circumstances related to crime commission; see Birkbeck & LaFree, 1993; Clarke, 1980, 1997; Cornish & Clarke, 1986; Meier, Kennedy, & Sacco, 2001), which highlights not only the importance of guardianship but also of situational factors such as the time when the offense is committed. The inclusion of these control variables also assist in capturing as much as possible the context of abuse related to the presence of a potential guardian.

Method

Sample

Eighty-seven adult males who were incarcerated between 2007 and 2009 for having committed a sexual offense against a child (16 years old or less) in Queensland (Australia) participated in this study. On average, participants were 45.16 years old at assessment (SD = 12.61). The majority was Australian born non-Aboriginal (84.3%). Other participants were either Aboriginal or Australian Torres Strait Islanders (11.4%) or born in another country (4.3%). Most of the participants did not achieve an education level higher than secondary school (70.9%). Almost a third of participants had a previous sexual offence history (31.9%) and more than half of the participants had a previous nonsex offence history (55.6%). A total of 20.9% of participants had also spent time in a foster home or institution because of problems at home before the age of 17 years. On average, participants were serving a sentence of 104.13 months (SD = 75.46, range = 358).

Procedure and Measure

This study uses data from a larger research project examining specific child sexual abuse events. Data on these events were collected via an offender self-report questionnaire specifically designed for this purpose. Initially, offenders were identified by the Queensland Department of Corrective Services (QDCS; Australia). These offenders were then approached individually by a member of the research team and invited to complete an event-based self-report questionnaire. They were told that their involvement in this study was strictly voluntary. Each offender signed a consent form stating that the information would be used for research purposes only. They were assured that the information would be kept confidential and that records of names would be destroyed after data collection. Finally, they were also debriefed following completion of the questionnaire.

The questionnaire was developed from a previous instrument designed to investigate the modus operandi employed by offenders in child sexual abuse (Smallbone & Wortley, 2000). One of the strengths of the questionnaire used in the present study is that it focuses on the onset and progression of sexual abuse and covers in a very detailed way what happened during specific offenses (e.g., offender modus operandi, offender-victim interchange, circumstances of offending). This questionnaire allowed for the examination of a maximum of three events for each participant (the first, the second, and the most recent offense). For the 87 respondents, 23 reported a single event only, 23 indicated their involvement in two events, and 41 offenders reported the occurrence of at least three events. The average number of crime events per offender available for our analysis was 2.2. A total of 194 crime events involving sexual contact with a child were statistically analyzed.

Variables

Offenders were asked to report whether somebody else was present during abuse. If the answer to this questions was “yes” then offenders were asked to select a response from a list to identify who that person was (i.e., child or adult, and relative, acquaintance, or stranger).¹ For the purpose of this study, a dichotomous variable measuring whether the offense occurred in the presence of a potential guardian is included in multivariate analyses (0 = no, 1 = yes). In these analyses, potential guardianship is treated as the key explanatory variable.

A set of independent variables were extracted from the questionnaire and included in multivariate analyses. These variables were (a) offense occurred during evening (6 p.m. to 9 p.m.; 0 = no, 1 = yes), (b) the offender was unrelated to the victim (0 = intrafamilial, 1 = extrafamilial), (c) the offense occurred in the offender’s home (0 = no, 1 = yes), (d) victim age (M = 10.85; SD = 3.18; range = 2-16 years old), and (e) the victim was a female (0 = no, 1 = yes). Except for the age of the victim, which is a continuous variable, all variables are dichotomous as specified above. Dependent variables were whether the duration of sexual contact was more than 5 min (0 = no, 1 = yes) and whether penetration occurred (0 = no, 1 = yes). These dependent variables were used to measure the severity of abuse. The use of penetration as a measure of severity is consistent with previous literature in sexual abuse (e.g., Aylwin et al., 2000; Leclerc, Proulx, Lussier, & Allaire, 2009; Leclerc & Tremblay, 2007). The rationale behind the use of duration as a measure of severity is that an event of substantial duration should be more likely than an event of short duration to lead to intrusive sexual acts such as penetration. Descriptive statistics for these variables can be found in Table 1.

Table 1.

Descriptive Statistics of Variables in Multivariate Analyses.

Variable	Category(ies)	Distributions
Age of victim	M (SD)	10.83 (3.33)
		Base N = 193
Female victim	Percent Yes	67.9
		Base N = 193
Offender unrelated to victim	Percent Yes	48.2
		Base N = 193
Occurred inside offender’s home	Percent Yes	63.9
		Base N = 194
Occurred during evening	Percent Yes	28.1
		Base N = 192
Occurred when potential guardian is present	Percent Yes	61.3
		Base N = 194
Duration of sexual contact over 5 min	Percent Yes	51.3
		Base N = 189
Penetration occurred	Percent Yes	23.7
		Base N = 194

Analytic Strategy

In this investigation, we analyze the effect of the presence of a potential guardian on the severity of abuse (duration of sexual contact and occurrence of penetration) while taking into account a set of offense characteristics. As our event data are nested within offenders (each offender may have been involved in several events for up to a maximum of three—the average number of crime events involving the same offender is 2.2), the independence of observations assumption is not met. One could expect events involving the same offender to be more similar to each other than to events involving different offenders. If this data structure is ignored, parameter estimates and their standard errors might be biased, which can artificially create significant coefficients and lead to an incorrect interpretation of findings (Guo & Zhao, 2000; see also Carjaval, Baumler, Harrist, & Parcel, 2001). In the presence of nested data, within-group homogeneity is permitted and estimated by the inclusion of random effects in the model, which corrects for the problem of dependent observations and prevents inflation of Type I error in the analysis (Bickel, 2007). In this study, random effects are estimated by performing mixed-effects/multi-level logistic regression models using Stata (version 11.0; Rabe-Hesketh & Skrondal, 2008). As the duration of sexual contact was associated with the occurrence of penetration (p = .018), we also included duration of sexual contact as a control variable when performing the analysis on the occurrence of penetration.

Results

Before performing mixed-effects logistic regression, we examined the context of abuse related to the presence of a potential guardian. Figure 1 presents the characteristics of potential guardians according to the location in which the abuse took place. First, adult strangers are more likely to be present when the abuse occurs outside the offender’s home. On the other hand, children related to the offender are more likely to be present when the abuse occurs in the home. Second, a substantial number of events involved the presence of more than one potential guardian during abuse. A qualitative examination of these data revealed that the most common situation was when a related child and adult were present in the home in which abuse occurred. We also examined whether some offense characteristics were more likely to be found when a potential guardian is present (see Table 2). It was found that the presence of a potential guardian was more likely when the victim was a female. On the other hand, the presence of a potential guardian was less likely during events that occurred in an extrafamilial setting. Most importantly, it was found that the presence of a potential guardian was less likely when the abuse was of long duration and involved penetration of the victim.

Figure 1.

Characteristics of potential guardians who were present at the time of abuse.

Table 2.

Offense Characteristics by the Presence and Absence of a Potential Guardian in Child Sexual Abuse (n = 194 events).

Offense variables	Absence of potential guardian (n = 75)	Presence of potential guardian (n = 119)	Chi-square/t	p value
Age of victim^a	11.16 (3.10)	10.82 (3.20)	0.730	.467
Female victim	58.1%	73.9%	5.251	.022
Offender unrelated to victim	62.2%	39.5%	9.389	.002
Occurred inside offender’s home	58.7%	67.2%	1.462	.227
Occurred during evening	67.2%	32.8%	3.345	.067
Duration of sexual contact over 5 min	68.5%	48.5%	14.036	<.001
Penetration occurred	38.7%	14.3%	15.118	<.001

Note. ^aMean and standard deviation (in parentheses) are presented.

Before estimating the effect of potential guardianship on duration of sexual contact and the occurrence of penetration using mixed-effects logistic regression analysis, an unconditional model (i.e., a model with no predictors or control variables) for these dependent variables was performed (analysis not shown here). The variance component was significant (p < .05), indicating that there was significant variation in the dependent variables across offenders. This finding supports the use of mixed-effects modeling over standard regression analyses.

We then examined the effect of a potential guardian on duration of sexual contact (Table 3). The model shows that female victims were less likely than males to be involved in events of a long duration (p = .035). The presence of a potential guardian proved to be negatively associated with the duration of sexual contact. The presence of a potential guardian was less likely during events of long duration (p = .020). In Table 4, the effect of potential guardianship on the occurrence of penetration is shown. As duration of sexual contact was associated with penetration (p = .018), we also controlled for this variable. The model shows that the age of the victim was positively associated with the occurrence of penetration (p = .023). This model also reveals that events occurring during the evening were less likely to involve penetration of the victim than those happening during another time (p = .019). Finally, net of other effects, the presence of a potential guardian proved to be strongly and negatively associated with the occurrence of penetration (p = .009). This finding suggests that events involving the presence of a potential guardian were less likely to lead to penetration of the victim than those that did not involve a potential guardian. The effect of the presence of a potential guardian was powerful, decreasing the odds of penetration by 86% ([OR − 1] × 100 = .861). As can be seen in Tables 3 and 4, the offender-specific random intercept is significant (p < .05), which indicates that events involving the same offender tended to be more similar to each other than to events involving different offenders.

Table 3.

Mixed-Effects Logistic Regression Models of Offense Characteristics and Potential Guardianship on Duration of Sexual Contact in Child Sexual Abuse (n = 185 events).

Fixed effects	Coefficient	SE	Odds ratio	p value
Age of victim	0.179	.099	1.20	.069
Female victim	–1.82	.862	0.162	.035
Offender unrelated to victim	1.42	.778	4.14	.068
Occurred inside offender’s home	–0.011	.749	0.989	.988
Occurred during evening	–0.128	.618	0.880	.836
Occurred when potential guardian is present	–1.43	.614	0.240	.020
Random effect	Variance-components (SE; p value)
Offender-specific random intercept	2.22 (.619; p < .05)

Note. There were nine missing values.

Table 4.

Mixed-Effects Logistic Regression Models of Offense Characteristics and Potential Guardianship on the Occurrence of Penetration in Child Sexual Abuse (n = 185 events).

Fixed effects	Coefficient	SE	Odds ratio	p value
Age of victim	0.341	.149	1.41	.023
Female victim	1.29	1.17	3.64	.269
Offender unrelated to victim	1.23	1.06	0.342	.247
Occurred inside offender’s home	–1.11	.988	0.328	.260
Occurred during evening	–2.21	.942	0.110	.019
Duration of sexual contact over 5 min	0.362	.840	1.44	.667
Occurred when potential guardian is present	–1.98	.751	0.139	.009
Random effect	Variance-components (SE; p value)
Offender-specific random intercept	2.81 (.870; p < .05)

Note. There were nine missing values.

Discussion and Conclusion

Little theoretical or empirical attention has been given to the effect of guardians on sexual offending, despite its critical importance for crime prevention generally. In the present study, we examined the effect of potential guardians on the severity of abuse while controlling for victim and situational characteristics. We found evidence indicating that the presence of a potential guardian decreases the duration of sexual abuse and the occurrence of penetration.

Consistent with Underwood et al. (1999), we first found that a large proportion of child sexual offenses are committed when a potential guardian is present. This is true wherever the abuse takes place, whether in the offender’s home or somewhere else. Children and adults related to the offender are the most likely potential guardians when the abuse is committed in the offender’s home. Taken together, these findings reveal that offenders are willing to take risks to obtain sexual contact with a child. On the other hand, these risks may be relatively easy to manage in a context where the victim is a child and the potential guardian is most likely to be a person related to the offender. Because of their developmental level, children are easier to control and manipulate for sexual contact than are older persons. As reported by Berliner and Conte (1990), children sometimes are not fully aware that they are being sexually abused. The relationship between the offender and the victim may also be important. One would expect the child to do as told by a related adult who occupies an authority-figure role.

We found that the presence of a potential guardian can reduce the severity of abuse even if that presence did not prevent the initiation of the abuse. Leaving aside victim and situational characteristics, the presence of a potential guardian appears to reduce the duration of sexual contact and the occurrence of penetration. Indeed, the presence of a potential guardian decreases the risk of sexual penetration by 86%. For instance, an adult offender can easily make up a story if caught sexually touching a child, but going no further. The offender could simply say that he was helping the child get dressed. However, such an explanation would not be believable if the offender was surprised while having overt sexual intercourse with a child. This distinction helps us understand why the proximity of a potential guardian can decrease the severity of abuse after sexual abuse is already initiated.

The impact of the presence of potential guardianship on penetration is further illustrated with the following simulation. We begin with observed findings from the present study:

Observation A—61.3% of events involved the presence of a potential guardian, while the other 38.7% involved no potential guardian;

Observation B—14.3% of events initiated in the presence of a potential guardian led to penetration;

Observation C—38.7% of events initiated in the absence of a potential guardian led to penetration; and

Observation D— In all, 23.5% of events in this study involved penetration.

Now imagine a situation in which Observations B and C continue to hold, but A is changed arbitrarily. What would this situation imply for D if potential guardians were present 90% percent of the time, and absent the other 10% of the time? We can calculate the implied risk that victims will be penetrated as follows:

0.9 (0.14) + 0.1 (0.39) = 0.165 .

Thus, the rate of penetration would decline from 23.5% to 16.5%. Next, imagine that potential guardianship is much lower, assuming that potential guardians are present only 10% of the time and absent 90% of the time. In this case, the implied risk that victims will be penetrated is:

0.1 (0.14) + 0.9 (0.39) = 0.365 .

It can be observed that changing the amount of potential guardianship can greatly alter the risk of severe abuse of children from the observed level of 23.5%. The high potential guardianship situation reduces penetration to 16.5% of events, while the low potential guardianship situation increases penetration to 36.5% of events, more than twice the risk of the low potential guardianship situation. Starting with real data, this simulation clearly implies that the proximity of potential guardians is important for reducing the severity of child sexual abuse.

Another intriguing finding is that the occurrence of penetration was less likely when the abuse took place during the evening (6.00 p.m. to 9.00 p.m.). This finding highlights the importance of routine activities in everyday life as conceptualized by Cohen and Felson (1979). Note also that a minority of offenses were committed during the evening (28.1%). This may be the case because the evening is the time period where people living in the same house are most likely to spend together after returning from work or school. Therefore, for offenses committed in a home specifically, offenders perhaps manage the risks of apprehension by avoiding the performance of more severe or intrusive behaviors during this time, that is when potential guardians are most likely to show up. This is consistent with empirical data showing that many sexual offenders try to minimize risks of apprehension when committing crimes (Beauregard & Leclerc, 2007). In summary, assuming that sexual arousal (or other factors such as alcohol and drugs) does not completely impair decision making (e.g., Loewenstein, Nagin, & Paternoster, 1997), these offenders are unlikely to ignore the risks of apprehension. Rather, they would consider what they can do and not do sexually with a child at a particular time and place when a potential guardian is present.

Reynald (2010) argues that active monitoring by guardians increases the chances of preventing crime. Although we do not have data to verify whether potential guardians were actively supervising children at the time of abuse, we can argue that potential guardians, merely by their presence, may reduce the risk of intrusive sexual acts such as penetration. This finding reveals the critical role of potential guardians in limiting crime severity. A large proportion of offenders may be willing to have sexual contact with a child when a potential guardian is nearby, but they are nevertheless constrained by the presence of a person who might suddenly intervene. This explanation is consistent with Felson’s (1995) assertion that by “their simple presence, people can discourage crime from happening at specific times and places” (p. 53).

Theoretically, this finding also takes the concept of guardianship back to the foundation of the routine activity approach (and perhaps most fundamental level) since an effective guardian is not necessarily someone who automatically or actively supervises his/her environment. Following Reynald’s (2010) study, we could also argue that the severity of abuse may be further minimized if the potential guardian was capable of engaging in active supervision and intervening if necessary. Reducing the effort required to supervise may assist potential guardians to engage in active monitoring. Perhaps reducing the effort required to witness and monitor an offense could be reduced by facilitating natural surveillance of public spaces. This may involve modifying the physical environment such as ensuring that windows on buildings face unto public space. The effective management and maintenance of places could also be promoted to help maximize the chances that potential guardians notice the offense. This might be achieved, for example, by posting reminders to monitor. However, as the most common location for sexual abuse is the offender’s home and the victim is often related or known to the offender, it is unclear what impact active supervision would have on the severity of abuse. Educating adults about patterns of offenders could be a way to extend guardianship (e.g., Leclerc et al., 2011). The presence of an additional adult during day-to-day care activities such as bathing the child could serve an important monitoring function. As other children are often present when the abuse takes place, emphasizing disclosure from them is another possible avenue for preventing some abuse from continuing.

To our knowledge, this study is the first criminological investigation of the effect of the presence of a potential guardian on the severity of child sexual abuse. However, the present findings should, of course, be considered in light of a number of limitations. First, our findings rely on self-reported data from incarcerated offenders. Therefore, some findings may be biased by offenders’ recollections and potential distortions, and the incidents included in our analysis may not be representative of all events involving the abuse of a child. Second, because the present incidents had in fact occurred, we cannot assess whether or not the presence of a potential guardian can prevent child sexual abuse from occurring in the first place. Third, to further understand the impact of guardianship on child sexual abuse, future studies should pay more attention to the offender–guardian relationship, and to who may be the most effective guardians. Despite these limitations, we believe this study makes a useful contribution by demonstrating that the presence of a potential guardian can affect sexual abuse outcomes and by showing how critical the concept of guardianship is in the field of sexual violence and abuse. Further research along these lines will be important for understanding whether and how guardianship might prevent sexual offenses from occurring in the first place. We believe that guardianship is at the core of a situational analysis of crime events, and that understanding the context in which crime occurs is critical for crime prevention.

Footnotes

Acknowledgements

We thank Marcus Felson, Danielle Reynald, and Patrick Lussier for comments made on an earlier version of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper was supported by a grant from Griffith University. The research project was supported by a grant awarded by the Australian Research Council (#LP0668287).

Notes

References

Aylwin

A. S.

Clelland

S. R.

Kirkby

Reddon

J. R.

Studer

L. H.

Johnston

(2000). Sexual offense severity and victim gender preference: A comparison of adolescent and adult sex offenders. International Journal of Law and Psychiatry, 23, 113-124.

Beauregard

É.

Leclerc

(2007). An application of the rational choice approach to the offending process of sex offenders: A closer look at the decision-making. Sexual Abuse: A Journal of Research and Treatment, 19, 115-133.

Berliner

Conte

J. R.

(1990). The process of victimization: The victim’s perspective. Child Abuse & Neglect, 14, 29-40.

Bickel

(2007). Multilevel analysis for applied research: It’s just regression. New York, NY: Guilford.

Birkbeck

LaFree

(1993). The situational analysis of crime and deviance. Annual Review of Sociology, 19, 113-117.

Carjaval

S. C.

Baumler

Harrist

R. B.

Parcel

G. S.

(2001). Multilevel models and unbiased tests for group based interventions: Examples from the Safer Choices Study. Multivariate Behavioral Research, 36, 185-205.

Clarke

R. V.

(1980). “Situational” crime prevention: Theory and practice. British Journal of Criminology, 20, 136-147.

Clarke

R. V.

(1997). Situational crime prevention: Successful case studies. Guilderland, NY: Harrow and Heston.

Clarke

R. V.

Eck

(2005). Crime analysis for problem solvers in 60 small steps. Washington, DC: US Department of Justice Office of Community Oriented Policing Services. Available from www.popcenter.org

10.

Cohen

L. E.

Felson

(1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44, 588-608.

11.

Cornish

D. B.

Clarke

R. V.

(1986). The reasoning criminal: Rational choice perspectives on offending. New York, NY: Springer-Verlag.

12.

Eck

Weisburd

(1995). Crime places in crime theory. In Eck

Weisburd

(Eds.), Crime prevention studies: Vol. 4. Crime and place (pp. 1-33). Monsey, NY: Criminal Justice Press.

13.

Elliott

Browne

Kilcoyne

(1995). Child sexual abuse prevention: What offenders tell us. Child Abuse & Neglect, 19, 579-594.

14.

Felson

(1986). Linking criminal choices, routine activities, informal control, and criminal outcomes. In Cornish

D. B.

Clarke

R. V.

(Eds.), The reasoning criminal: Rational choice perspectives on offending (pp. 119-128). New York, NY: Springer-Verlag.

15.

Felson

(1995). Those who discourage crime. In Eck

J. E.

Weisburd

(Eds.), Crime prevention studies: Vol. 4. Crime and place (pp. 53-66.). Monsey, NY: Criminal Justice Press.

16.

Felson

(2002). Crime and everyday life (3rd ed.). Thousand Oaks, CA: Sage.

17.

Finkelhor

(1984). Child sexual abuse: New theory and research. New York, NY: The Free Press.

18.

Garofalo

Clark

(1992). Guardianship and residential burglary. Justice Quarterly, 9, 443-463.

19.

Guo

Zhao

(2000). Multilevel modeling for binary data. Annual Review of Sociology, 26, 441-462.

20.

Hart

T. C.

Miethe

T. D.

(2008). Exploring bystander presence and intervention in nonfatal violent victimization: When does helping really help? Violence and Victims, 23, 637-651.

21.

Kaufman

K. L.

Hilliker

D. R.

Daleiden

(1996). Subgroup differences in the modus operandi of adolescent sexual offenders. Child Maltreatment, 1, 17-24.

22.

Lang

R. A.

Frenzel

R. R.

(1988). How sexual offenders lure children. Annals of Sex Research, 1, 303-317.

23.

Leclerc

Proulx

Lussier

Allaire

J.-F.

(2009). Offender-victim interaction and crime event outcomes: Modus operandi and victim effects on the risk of intrusive sexual offenses against children. Criminology, 47, 595-618.

24.

Leclerc

Tremblay

(2007). Strategic behavior in adolescent sexual offenses against children: Linking modus operandi to sexual behaviors. Sexual Abuse: A Journal of Research and Treatment, 19, 23-41.

25.

Leclerc

Wortley

Smallbone

(2010). Investigating mobility patterns for repetitive sexual contact in adult child sex offending. Journal of Criminal Justice, 38, 648-656.

26.

Leclerc

Wortley

Smallbone

(2011). Getting into the script of adult child sexual offenders and mapping out situational measures. Journal of Research in Crime & Delinquency, 48, 209-237.

27.

Loewenstein

Nagin

Paternoster

(1997). The effect of sexual arousal on expectations of sexual forcefulness. Journal of Research in Crime & Delinquency, 34, 443-473.

28.

London

Bruck

Ceci

S. J.

Shuman

D. W.

(2005). Disclosure of child sexual abuse: What does the research tell us about the ways that children tell? Psychology, Public Policy, and Law, 11, 194-226.

29.

Luckenbill

D. F.

(1977). Criminal homicide as a situated transaction. Social Problems, 25, 176-186.

30.

Lynch

J. P.

Cantor

(1992). Ecological and behavioral influences on property victimization at home: Implications for opportunity theory. Journal of Research in Crime & Delinquency, 29, 335-362.

31.

Meier

R. F.

Kennedy

L. W.

Sacco

V. F.

(2001). The process and structure of crime: Criminal events and crime analysis. Advances in Criminological Theory. Vol. 9. New Brunswick, NJ: Transaction Publishers.

32.

Miethe

T. D.

Meier

R. F.

(1990). Opportunity, choice, and criminal victimization: A test of a theoretical model. Journal of Research in Crime & Delinquency, 27, 243-266.

33.

Rabe-Hesketh

Skrondal

(2008). Multilevel and longitudinal modeling using stata (2nd ed.). College Station, TX: Stata Press.

34.

Reynald

(2010). Guardians on guardianship: Factors affecting the willingness to supervise, the ability to detect potential offenders, and the willingness to intervene. Journal of Research in Crime & Delinquency, 47, 358-390.

35.

Smallbone

Wortley

(2000). Child sexual abuse in Queensland: Offender characteristics and modus operandi (Full report). Brisbane: Australia: Queensland Crime Commission.

36.

Tseloni

Wittebrood

Farrell

Pease

(2004). Burglary victimization in England and Wales, the Unites States and the Netherlands: A cross-national comparative test of routine activities and lifestyle theories. British Journal of Criminology, 44, 66-91.

37.

Underwood

R. C.

Patch

P. C.

Cappelletty

G. G.

Wolfe

R. W.

(1999). Do sexual offenders molest when other persons are present? A preliminary investigation. Sexual Abuse: A Journal of Research and Treatment, 11, 243-247.

38.

Wilcox

Madensen

T. D.

Tillyer

M. S.

(2007). Guardianship in context: Implications for burglary victimization risk and prevention. Criminology, 45, 771-803.

39.

Wortley

Smallbone

(2006). Applying situational principles to sexual offenses against children. In Wortley

Smallbone

(Eds.), Crime prevention studies: Vol. 19. Situational prevention of child sexual abuse (pp. 7-35). Monsey, NY: Criminal Justice Press.