Improving Police Officers’ Perceptions of Sexual Offending Through Intensive Training

Abstract

We examined whether specialist police training on the dynamics of sexual offending can modify officers’ victim-blaming attitudes and negative perceptions regarding likely case authorization. The sample included 77 Australian police officers specialising in sexual assault investigation. The training, delivered face to face over 4 weeks, included focus on identifying elements of grooming in offending relationships and how these elements can be elicited from victims and suspects within a narrative interviewing framework. Officers’ perceptions of cases were assessed immediately pre- and posttraining using a series of case scenarios. For each scenario, officers rated (on a 10-point Likert-type scale) their confidence that the case should be authorised to proceed to prosecution and the responsibility attributable to the victim. For each case, officers also listed up to 5 factors to justify their case authorization decision. Overall, confidence in case authorization increased from pre- to posttraining, whilst perception of victim “responsibility” decreased. The pattern of results, including the qualitative evidence to justify officers’ decisions, support that the attitude change was due to greater understanding of the dynamics of sexual offending. The implications for police trainers, and directions for future research, are discussed.

Keywords

sexual abuse investigative interviewing police training sex offender suspects whole story model

Sexual abuse is a prevalent crime, the rate of which continues to rise globally (Garcia-Moreno & Watts, 2011). Conviction rates for sexual offences are low, however, due in large part to high levels of case attrition. Indeed, the rate of attrition in sexual offences is higher than that of all other personal crimes (Stern, 2010; Victorian Law Reform Commission [VLRC], 2004). On average, only 14% of all sexual assaults are reported to police organisations, of which only 30% are referred for prosecution. As a consequence of the low reporting and prosecution rates, less than 3% of all sexual assault perpetrators are convicted of their crimes (Daly & Bouhours, 2010).

One reason that many allegations of actual sexual abuse do not proceed to trial is because there are widespread misconceptions about how sexual offences occur and how victims should behave when they are offended against (Chapleau & Oswald, 2010; Horvath & Brown, 2005; Page, 2008, 2010; Ullman, 2010). Common perceptions (i.e., “rape myths”) are that “real victims fight back,” “real victims report abuse immediately,” and “real rapes are not committed by persons who the victim knows” when in reality compliance, delayed reporting, and the existence of prior victim–offender relationships are the norm (Lievore, 2003; Page, 2010; Sleath & Bull, 2012; Ullman, 2010). Police officers are “gatekeepers” to the criminal justice system (i.e., they determine which cases proceed to prosecution), yet police misconceptions of sexual offences are prevalent irrespective of officer or victim gender, whether the officers specialise in sexual assault investigation, and officer years of service (Jamel, Bull, & Sheridan, 2008; Schwartz, 2010; Sleath & Bull, 2012).

The relevance of professionals’ misconceptions about how sexual offences occur is that when a victim’s behavior is not consistent with professionals’ prior expectations, this can influence the professionals’ perception of the veracity and strength of the evidence in a manner that is prejudicial for the victim (Statewide Steering Committee to Reduce Sexual Assault [SSCRSA], 2006). Perceptions of the veracity and strength of evidence determine whether a case should be pursued (Powell, Murfett, & Thomson, 2010), and for juries, it determines whether a suspect should be convicted (Devine, Clayton, Dunford, Seying, & Pryce, 2001). Even victims’ decisions to pursue a complaint of abuse are influenced by misconceptions about sexual offending dynamics. Attributions of self-blame such as “I should have fought back,” “I should have reported when I first had the opportunity,” and “I should have seen the warning signs” lead victims to withdraw complaints due to fear that their story will not be believed (Jamel et al., 2008; Kelly, Lovett, & Regan, 2005; Lievore, 2003; Mouzos & Makkai, 2004; van Kesteren, Mayhew, & Nieuwbeerta, 2000). Feeling supported, listened to, and not judged is of paramount importance to victims, more so than whether their case results in the offender being punished (Powell & Cauchi, 2011; VLRC, 2004).

Specialised police training targeting knowledge of sexual offending is of paramount importance for reducing the high attrition rates of sexual offences. This conclusion is stated in major government inquiries into sexual offences (e.g., Stern, 2010; VLRC, 2004) as well as independent empirical research. For instance, surveys and in-depth interviews with victims and sexual assault investigators have highlighted an association between rape myths and low prioritisation of, and support for, sexual assault victims during the justice process (Jordan, 2004; Kelly et al., 2005; van Kesteren et al., 2000). Furthermore, examination of the written justifications for police authorization of sexual abuse cases for prosecution suggests a bias that shares an uncanny resemblance to the public stereotype and myths of how rapes occur (Brown, Hamilton, & O’Neill, 2007; Jordan, 2004; Powell et al., 2010; Spohn & Tellis, 2008, 2012). Frazier and Haney (1996), for example, found that of 569 reported sexual assaults, cases had a higher likelihood of being investigated and authorised for prosecution if the suspect was a stranger, the victim was injured or threatened, and there was evidence of penetration.

The acceptance of rape myths by police officers means that cases that do not fit the stereotype are dismissed as “false reports.” False reports are estimated by officers as between 50% and 80% when the true rate is less than 10% (Lisak, Gardinier, Nicksa, & Cote, 2010; Schwartz, 2010; SSCRSA, 2006; VLRC, 2004). Furthermore, evidence is sought, remembered, and compiled by police in a way that supports initial hypotheses regarding the guilt of the suspect (Ask & Granhag, 2005; Hill, Memon, & McGeorge, 2008; Rassin, Eerland, & Kuijpers, 2010). This phenomenon, referred to as confirmation bias, is a fundamental cognitive tendency that impacts performance in almost every professional domain (Nickerson, 1998). In the context of sexual assault investigation, expectations of suspect guilt or innocence can have an effect on questioning style, which in turn can lead to a self-fulfilling prophecy effect (Hill et al., 2008).

Stereotypes and biases can be tremendously difficult to change in training. Prior studies targeting negative police attitudes have clearly demonstrated this. Patterson (2004), for example, used a pre-/posttest design to examine the effect of a training course on 81 police recruits’ knowledge and skills of child abuse, as well as the prevalence of negative affect (e.g., anger, disgust, sadness, and discomfort) when reading case vignettes. The reason for targeting these reactions is that they are known to interfere with investigators’ ability to talk objectively with, and support, abusive parents (Hazzard & Rupp, 1986). The child abuse training (part of a larger course delivered in class format over several days) was 8 hours in length and included 1 hour on the dynamics of child abuse and the characteristics of abusive parents. Although the training was associated with improvement in knowledge of child abuse, officers’ perceptions about how they would feel when interacting with an abused child or an abusive parent (as measured by the Child Abuse Survey; Hazzard & Rupp, 1986) did not change.

Another study to highlight the difficulty in shifting police attitudes under brief instruction is that by Lonsway, Welch, and Fitzgerald (2001), who tested the effectiveness of two sexual assault police training programs. As with the study by Patterson (2004), the time devoted to sexual offending dynamics was relatively brief. One program included less than 30 minutes of discussion on sexual offending dynamics, and the other included 2½ hours on the dynamics of sexual offending and the impact on victims. The longer training included a video of a fictional rape of a police officer (to invoke victim empathy) and a role-play with one participant playing a robbery victim being interviewed by the instructor, to demonstrate the impact of interviewer scepticism and blame. For both training programs, attitudes and beliefs that “deny and justify male aggression against women” were measured on Payne, Lonsway, and Fitzgerald’s (1999) rape myth acceptance scale,¹ both before and at various points throughout and after the program. The results showed that neither of the training programs had any effect on police recruits’ acceptance of rape myths, even though the training was effective in improving behavioral performance on several practical skills, including interviewing. The findings are consistent with reviews highlighting the complex array of interrelated social and cognitive processes that underpin attitude change in informational settings (e.g., Wood, 2000).

The current study extends the work of Lonsway et al. (2001) and Patterson (2004) by testing the effectiveness of a more intensive, integrated, and interactive training program on sexual offending. The mandatory 4-week course, a foundation training program designed for all police investigators who specialise in sexual assault and sexual child abuse investigation, centres around the “Whole Story” approach to investigation of sexual crime, a model reported by Tidmarsh, Powell, and Darwinkel (in press). The model proposes that understanding grooming (i.e., the process by which a perpetrator prepares a victim for abuse, by gaining access, compliance, and secrecy) is a central component of the police interview and investigation procedure in cases of sexual assault and sexual child abuse. Specifically, decision-making of investigators, jurors, and legal professionals requires an understanding not only of what occurred (i.e., the offence and physical context) but details of how the victim came to be in the situation and to respond the way he or she did. These how details are elicited by direct enquiry about the offender–victim relationship and the events that occurred leading up to the offence. The dynamics were taught within a variety of different learning modalities, including information sessions, question and answer sessions with actual victims, footage of actual offenders (in the context of therapy) explaining the grooming process, case examples, and role-playing of victims reporting sexual abuse.

Overall, we expected that our intensive training in the dynamics of sexual offending would be associated with (a) significant improvement in awareness of offending dynamics, (b) higher attributions of likely case authorization, and (c) lower attributions of victim responsibility for the offences. We also expected that the improvements would be more obvious for those case scenarios where the evidence against the suspect was ambiguous (i.e., there was no obvious physical or corroborative evidence that indicated the guilt of the suspect, beyond the victims’ report).

Method

Participants

All participants (N = 77) were sworn police officers recruited through Victoria Police Service, Australia. The training described in this article was a mandatory course that the participants needed to complete to become specialist investigators in sexual offending and child abuse investigation. The sample was heterogeneous, including members from across the entire state (regional and metropolitan) and from a range of ranks including Senior Constables, Sergeants, and Senior Sergeants.² In all, 66% of participants were male and 34% were female. Participants came from a variety of backgrounds, from general duties police officers to criminal-investigation-focused detectives and victim-focused sexual offence investigators. Prior experience in sexual offence investigation ranged from none to more than 20 years. Ethical approval was obtained from the university and police ethics committees. While the course was mandatory, participation in the research evaluation component was voluntary (all professionals agreed to be involved). To protect confidentiality, no identifying information was recorded on the assessment sheets.

Procedure

The study adopted a pre-/posttraining design. All participants engaged in the 4-week training course and completed a questionnaire that assessed their perception of cases on Day 1 of the course (prior to the commencement of any educative component) and again on the last day of the training after all the coursework was complete. The questionnaires included 12 scenarios. Immediately after reading each scenario, participants were required to answer a series of four standard questions. They were instructed that there is no right or wrong answer to the questions; judgments were subjective. The details of the training and assessment tasks are described below.

Training course

The training was delivered over 4 consecutive weeks in classroom format by specialists with backgrounds in the treatment of sex offenders, police investigation of sexual offences, and the impact of sexual abuse on victims. The content focused primarily on the dynamics of sexual offending and interview technique. Table 1 presents a description of the program including the content areas and their objectives, time allocation, and methods of delivery. The theoretical premise running through the program is that sexual offending occurs within a “relationship” driven and manipulated by offenders and that witness/suspect interviews need to better incorporate the relationship dynamics to enhance the usefulness of these interviews for professional and jury decision making (see Tidmarsh et al., in press). Specifically, the course included content related to theories and motivations of offending (i.e., how offenders overcome obstacles to offending and the processes of grooming; Craven, Brown, & Gilchrist, 2006; Howells, 1995) as well as victim behavior and responses before, during, and after sexual abuse (Bloom, 2003; Levin, 2004). The elicitation of these elements within a narrative interview process was also covered in detail. Participants were required to demonstrate their knowledge and practical interview skills in various interactive classroom exercises and role-plays.

Table 1:

Description of Training Components and Time Allocated to Each

Component	Objectives	Description of Relevant Sessions	Hours
Introduce model	To introduce the framework and its importance during investigations and trials.	Lecture and discussion on the “Whole Story” framework (Tidmarsh et al., in press) and sexual offending as a crime of relationship (with case examples). Analysis of a court transcript highlighting the role of police and how the framework can impact court proceedings.	7 2
Understanding victims	To demonstrate and explain victims behavior To develop victim empathy To understand victims’ perspectives and knowledge of grooming and the offending relationship.	Lecture and discussion sessions on: - victims’ responses to trauma - the psychology of victims and their behavior - victims’ needs (with experienced abuse counsellors) Presentation and question/answer session with adult victim/survivors of childhood sexual abuse	7 5 4 4
Understanding offenders	To understand suspects’ motivation to offend To understand offenders’ perspectives and knowledge of grooming and the offending relationship	Video footage of offenders in treatment role-playing grooming Lecture and discussion session on theories of offending Lecture and discussion on offending planning and behavior Analysis of interview transcripts with offenders	2 6 7 1
Case examples	To identify elements of grooming and offending relationships in real cases	Case example presentation with question/answer session on a case of ongoing child sexual abuse Case example presentation and question/answer session with various specialists: - experienced serious sexual offences unit detectives - experienced family violence unit detectives - internet crime experts on “online” grooming	2 4 4 2
Interview theory and practice	To review the utility of the Whole Story interview framework in practice To practice eliciting grooming evidence within a narrative framework To demonstrate the importance of the Whole Story for understanding the context of offending	Lecture and discussion sessions on: - interviewing suspects - interviewing victims (including children) Interviews with actors role-playing suspects Practice victim/witness interviews with: - real children about innocuous events - persons with a cognitive impairment about innocuous events - participants role-playing a child of abuse - participants about personal life events using cognitive interviewing skills	8 9 8 7 5 2 2

Questionnaires

The questionnaires contained 12 scenarios (approximately eight lines each) with four standard questions per scenario. The scenarios were based on actual cases; for half, the victims were aged 6 to 17 years, and for the other half, the victims were adults (18 years or over). The cases also differed according to presence of alcohol and drugs, gender of victim/suspect, victim–suspect relationship (e.g., stranger, acquaintance, relative, partner, and caregiver), elements of grooming (e.g., offending games), context of offending (e.g., public, home, party), and reporting time (immediate, delayed, and historical). Two versions of the questionnaire were created (A and B), each containing a unique set of scenarios. Half the officers’ received Questionnaire A prior to the training and Questionnaire B posttraining. For the remaining participants, the order was reversed. In each set of 12 cases, the strength of evidence to support the victim’s allegations also varied; three cases presented strong (i.e., corroborative) evidence, three presented weak evidence (i.e., evidence that contradicts the victim’s account or the suspect is unknown), and the remaining six cases presented ambiguous (little if any) evidence to support the victim’s account beyond the report. Importantly, elements of common misconceptions were included in the case scenarios, such as delayed reporting and perceived compliance with the offender, and these elements were counterbalanced across questionnaires and strength of evidence. An example of an ambiguous case scenario is as follows:

A 22-year-old girl reports being raped by a man she met at a party. Witnesses saw them “flirting” and dancing together. She states that she agreed to go upstairs with him but told him she wasn’t going to have sex with him. They went into a bedroom and lay on the bed. After some prolonged kissing, he started to take off her clothes. She says that she again told him “No,” but when he took her underwear off anyway, she gave him a condom from her handbag, saying “please use this.” The suspect stated in interview that the sex was consensual.

For each scenario, officers were asked the following questions (in standard order), using 10-point Likert-type scales: (a) “How confident are you that the case will be authorised (proceed) for prosecution?” (from not at all likely to extremely likely)³ and (b) “How much responsibility should be attributed to the victim in this scenario?” (from none to all). The final question asked officers to justify their rating about case authorization by listing up to five factors (in order of importance) that they considered during their decision-making.

Coding

The Likert-type ratings were entered directly into SPSS. The factors listed by the officers to justify their authorization decisions were coded in three mutually exclusive ways. First, the nature of the factor was noted. These included evidential factors (i.e., the presence of corroborative evidence, and the presence of circumstantial evidence), victim-related factors (i.e., quality of victim disclosure, victim behavior, victims’ credibility, and victims’ wishes for the case), suspect-related factors (i.e., suspects’ admission/denial, suspects’ credibility, and whether the suspects’ identity was known or in question), offence-related factors (i.e., issue of consent, nature of the relationship and grooming processes, and clarity of the offence), and miscellaneous factors that included costs awarded against the police and the prosecution’s public interest test. From the factors listed above, three were identified as relating to the dynamics of sexual offending. These included the presence of corroborative evidence, the victim’s behaviour, and the nature of the relationship (including grooming processes).

Second, each factor listed by the officers was coded as either a positive or negative comment. Examples of positive and negative comments for the three factors related to dynamics of offending are provided in Table 2. Third, the priority given to each factor was noted. Five points was awarded to the first factor to be listed, 4 points for the second factor, 3 points for the third factor, 2 points for the fourth, and 1 point for the fifth.

Table 2:

Examples of Positive and Negative Comments for Three Main Factors Listed in Decision Making

Factor	Positive Comments	Negative Comments
Corroborative evidence (e.g., medical, physical, or witness evidence)	Injuries corroborate victims’ version of events Video on mobile phone corroborates victims’ version of events	Lack of independent witnesses One word against another (no corroboration)
Victims behaviour	Good evidence of recent complaint Victim going about her business, not putting herself in a risky situation Clearly said “no” twice	Victim agreed to go upstairs (with suspect) to bedroom, kissed on bed, and gave condom Delay in reporting Continued to work with suspect after offence
Nature of the relationship (including grooming)	Ability to establish background of relationship from start that led to victim seeing offending as “normal” Evidence of grooming, and giving victim alcohol, and threats to tell mum Because she liked him he used this “power” to manipulate her into meeting him again	Consenting victim and offender taking into account the details of their relationship Require information regarding grooming and power/control/authority

To assess coding reliability, 25% of the questionnaires were randomly selected and coded by a researcher who was not otherwise involved in this study. Interrater reliability for the reporting of the factors at each position (from first to last listed) demonstrated high interrater reliability (kappa ranged from .94 to .96).

Results

The results are presented in three sections. First, we present the correlations between officers’ ratings of case authorization and victim responsibility. Although we are primarily interested in pre- and posttraining changes in rating, this analysis is interesting because the argument for conducting the training was based on an implied negative correlation between these factors that (to date) has not been directly tested. Second, 2 (Strength of Evidence; Strong, Ambiguous, Weak) × 2 (Time; Pre- vs. Posttraining) analyses of variance (ANOVAs) were conducted on case authorization and victim responsibility ratings, with both independent measures being manipulated within subjects. Third, a series of paired samples t tests compared the factors listed in participants’ case authorization decisions pre- and posttraining.

Relationship Between Ratings of Victim Responsibility and Case Authorization

Pearson’s product-moment correlation coefficients were computed to compare the relationship between victim responsibility and brief authorization pre- and posttraining. There was a moderate negative relationship between victim responsibility and brief authorization both before, r(77) = -.43, p < .01 (two-tailed), and after training, r(75) = -.38, p < .001 (two-tailed). This relationship suggests that as the rating of victim responsibility increases, the likelihood of authorising a case decreases. Before training, 18% of the variance in case authorization decisions could be accounted for by the attribution of victim responsibility (r² = .18), which dropped to 14% after training (r² = .14).

Effect of Training on Case Authorization and Victim Responsibility Ratings

Table 3 presents the means and standard deviations for the scales’ measures authorization and victim responsibility by time and strength of evidence. First, a 2 (Time; Pre- vs. Posttraining) × 3 (Evidential Quality; Strong, Ambiguous, and Weak) ANOVA was conducted on the case authorization ratings. There was a significant main effect of time, F(1, 76) = 12.22, p < .01, partial η² = 0.14; the likelihood of case authorization was significantly higher posttraining (M = 5.96, SD = 1.37) compared to pretraining (M = 5.19, SD = 1.71). The main effect of strength of evidence was also significant, F(2, 152) = 1,157.47, p < .001, partial η² = 0.94. Post hoc analyses (Sidak) revealed that each of the pair-wise comparisons for strength of evidence were significant; authorization was significantly higher for strong evidence cases compared to ambiguous evidence (M difference = 3.40,p < .001) and weak evidence cases (M difference = 6.06, p < .001) and for ambiguous compared to weak evidence cases (M difference = 2.66, p < .001).

Table 3:

Means (Standard Deviations) for the Scale Measures Case Authorization and Victim Responsibility by Time and Strength of Evidence

	Before Training			After Training
	Strong	Ambiguous	Weak	Strong	Ambiguous	Weak
Authorization	8.52 (1.11)	4.74 (1.51)	2.29 (1.49)	8.93 (1.75)	5.91 (1.64)	3.05 (1.74)
Responsibility	0.68 (1.22)	3.03 (2.19)	3.72 (2.51)	0.75 (1.50)	1.89 (1.84)	2.37 (2.29)

The interaction between time and evidence strength was also significant, F(2, 152) = 4.85, p < .01, partial η² = 0.06. One-way repeated measures ANOVAs compared the effect of time at each level of strength of evidence. There was a significant effect of time for ambiguous cases F(1, 76) = 23.19, p < .001, partial η² = 0.23; case authorization was significantly more likely after training (M = 5.91, SD = 1.64) compared to before training (M = 4.74, SD = 1.51). The effect of time was also significant for weak cases F(1, 76) = 6.54, p < .05, partial η² = 0.08; authorization was again significantly more likely after training (M = 3.04, SD = 1.74) compared to before training (M = 2.29, SD = 1.49). However, the effect of time was not significant for strong evidence cases, F(1, 76) = 2.77, p > .05; there was no difference in the likelihood of case authorization before (M = 8.52, SD = 1.11) compared to after training for strong cases (M = 8.93, SD = 1.75).

Next, a 2 (Time) × 3 (Evidential Quality) ANOVA was conducted on ratings of victim responsibility showing an inverse relationship compared to the authorization ratings. The main effect of time was significant, F(1, 76) = 13.87, p < .01, partial η² = 0.15; ratings of victim responsibility were significantly lower posttraining (M = 1.67, SD = 1.98) compared to pretraining (M = 2.48, SD = 1.88). There was also a significant main effect of strength of evidence, F(2, 152) = 91.02, p < .001, partial η² = 0.55. Post hoc analyses (Sidak) revealed that each of the pair-wise comparisons for strength of evidence were significant; the attribution of victim responsibility in strong evidence cases was significantly lower than ambiguous evidence cases (M difference = -1.74, p < .001) and weak evidence cases (M difference = -2.33, p < .001), and ambiguous evidence cases were also significantly lower than weak evidence cases (M difference = -0.59, p < .001). The interaction between time and evidence strength was also significant, F(2, 152) = 13.15, p < .01, partial η² = 0.15. One-way repeated measures ANOVAs compared the effect of time at each level of strength of evidence. There was a significant effect of time for ambiguous cases, F(1, 76) = 16.10, p < .001, partial η² = 0.18; victim responsibility was significantly lower after training (M = 1.89, SD = 1.84) compared to before training (M = 3.03, SD = 2.19). The effect of time was also significant for weak evidence cases, F(1, 76) = 18.42, p < .001, partial η² = 0.20; victim responsibility was significantly lower post- (M = 2.37, SD = 2.30) compared to pretraining (M = 3.72, SD = 2.52). However, the effect of time for strong evidence cases was nonsignificant, F(1, 76) = 0.11, p < .05, and there was no difference in victim responsibility before (M = 0.68, SD = 1.22) and after training for strong evidence cases (M = 0.75, SD = 1.50).

Factors Listed in Case Authorization Pre- and Posttraining

The final set of analyses determined whether change in officers’ case authorization decisions pre- to posttraining were associated with better understanding of the dynamics of sexual offending. In relation to the three key factors that related to the dynamics of sexual offending (corroborative evidence, victim behaviour, and the nature of the relationship), a series of paired samples t tests were conducted to determine whether the mean frequency and prioritisation⁴ of the factors listed changed from pre- to posttraining. Analyses were separated according to whether the factor was associated with a positive or negative comment as well as overall. Table 4 presents the means, standard deviations, and results of these analyses.

Table 4:

Means (and Standard Deviations) of the Frequency and Priority of Factors Listed in Decision Making

		Frequency			Priority
Factors	Details	Pre	Post	t(72)	Pre	Post	t(72)
Corroborative evidence	All	12.49 (3.09)	11.11 (3.07)	3.60**	3.84 (0.43)	3.84 (0.50)	–0.06
	Positive	7.67 (2.03)	7.68 (2.05)	–0.06	4.01 (0.47)	4.00 (0.51)	0.19
	Negative	4.82 (2.87)	3.42 (2.53)	3.76**	3.38 (1.12)	3.10 (1.37)	1.56
Victims behaviour	All	4.74 (2.75)	3.79 (2.06)	2.72*	3.49 (1.01)	3.55 (0.88)	–0.34
	Positive	1.79 (1.30)	2.15 (1.66)	–1.54	3.04 (1.59)	3.37 (1.39)	–1.41
	Negative	2.95 (2.15)	1.64 (1.48)	4.67**	3.18 (1.47)	2.26 (1.65)	2.08*
Nature of relationship, including grooming	All	0.26 (0.53)	1.00 (1.29)	–4.85*	0.74 (1.49)	1.87 (1.91)	–4.25**
	Positive	0.18 (0.42)	0.89 (1.97)	–5.04**	0.53 (1.30)	1.77 (1.91)	–4.95**
	Negative	0.08 (0.28)	0.11 (0.39)	–0.47	0.29 (0.99)	0.29 (1.00)	0.00

p < .05. **p < .001.

As shown in Table 4, there were significant changes in both the frequency and priority of factors listed in the officers’ reasoning processes. The frequency of comments related to corroborative evidence and victim behaviour significantly decreased after training, particularly negative comments related to the lack of corroboration and counterintuitive (or negative) victim behaviour. Whilst there was no change pre- to posttraining in the priority with which corroborative evidence and victim behaviour factors were listed overall or as positive comments, there was a significant decrease in the priority of negative comments regarding the victims’ behaviour. Further, there was a significant increase in the frequency and priority of overall and positive comments related to the nature of the relationship (including grooming) posttraining. The nature of the relationship was the only factor considered that significantly increased in importance posttraining.

Importantly, the above mentioned changes in the prioritisation of grooming details across time cannot be attributed to differences in response completeness. Paired samples t tests revealed that there was no significant difference between the mean number of factors listed per participant before (M = 42.19, SD = 8.22) and after (M = 41.27, SD = 9.62) training, t(72) = 1.13, p > .05, and no significant difference between the mean number of factors listed per scenario before (M = 3.52, SD = 0.69) and after (M = 3.44, SD = 0.80) training, t(72) = 1.13, p > .05.

Discussion

This is the first study to demonstrate that police investigators’ confidence in case authorizations and attributions of victim responsibility can be shifted as a result of intensive classroom-based training in the dynamics of sexual offending. The training was associated with changes in both attribution of victim responsibility and case authorization. Specifically, attributions of victim responsibility were lower, and perceptions of likely case authorization were higher posttraining, particularly in those cases where evidence in relation to the crime was ambiguous (i.e., it did not reflect on the veracity of the victims statement one way or another). Overall, the factors listed to support the case authorization decisions reflected a better understanding of the dynamics of sexual offending posttraining. Officers showed a greater consideration of the offending relationship and grooming processes, and they made fewer negative comments in relation to the victims’ behaviour and the lack of corroborative evidence.

Overall, the pattern of results is consistent with the conclusion that greater understanding of the dynamics of sexual offending and grooming processes (as delivered in the training) were responsible for the changes in perceptions of likely case authorization. There are three reasons for making this claim. First, victim responsibility was associated with case authorization decisions, such that the higher the attribution of responsibility to the victim, the less likely the case was to be authorised. Changes in investigators’ understanding of sexual offending dynamics and victim behaviour (due to engaging in the training course) were associated with reduced negative bias against victims. This was especially evident in ambiguous cases that did not support or refute the victims’ allegation of sexual offending.

Second, the police trainees at the time of commencing the course were well aware of the high prevalence of victim blaming attitudes among police as this had been a major finding of two state-wide enquiries (SSCRSA, 2006; VLRC, 2004), circulated among members by their supervisors prior to the commencement of the course. The course itself provided knowledge related to offending, but it did not address issues related to the high rate of attrition in sexual assault cases and the role of officers’ misconceptions in contributing to these attrition rates. Third, whilst corroborative evidence remained the most important factor considered in case authorization, posttraining, the officers made fewer negative comments regarding the lack of corroborative evidence and counterintuitive (or negative) victim behaviour, such as delayed reporting. There was no change in the frequency of positive comments related to victims’ behaviour pre- to posttraining, suggesting that the participants’ attitudes were not simply more “provictim” after training. Rather, the data support that their considerations of victims’ behaviour were more cautious and considerate of counterintuitive behaviour.

The importance of these findings cannot be underestimated. Poor attitudes among professionals, including police officers, are common (Page, 2008, 2010; Schwartz, 2010; Sleath & Bull, 2012). While the need for training to improve police officers’ awareness of sexual offending dynamics has been widely emphasised (e.g., Criminal Justice Sexual Offences Taskforce, 2005; Hunter, Bentley-Cewe, & Mills, 2000; Lonsway et al., 2001; May-Parker & Kiwala, 2000; New Jersey Division of Criminal Justice, 2003), training to date, using similar designs and assessment measures as the current study, have been unsuccessful in demonstrating a shift in attitudes of sexual offending posttraining (Lonsway et al., 2001; Patterson, 2004). In a climate where classroom-based training programs have been criticised for their lack of effectiveness in improving police officers’ interviewing skills (Powell, 2008), this research shows that intensive training can effectively improve police investigators’ attitudes and decisions toward sexual offence cases.

Now that we have demonstrated changes in the classroom setting, the next step for researchers is threefold. First, it needs to be determined whether the changes observed here actually transfer to the field and improve the attitudes, decision-making, and interviewing practice of police officers, as well as attrition rates. In particular, the potential association between knowledge of sexual offending dynamics (and grooming processes) and the ability to adhere to best-practice interview guidelines is of particular interest to us in our role as interview trainers. A large body of research has documented the difficulty of police investigators to adopt best practice interview techniques (Oxburgh, Ost, & Cherryman, 2010; Powell, Fisher, & Wright, 2005; Smith, Powell, & Lum, 2009). One possible explanation for this finding is that interview skill is mediated by rape myth acceptance, such that those police officers with lower rape myth acceptance show better rapport and open-ended questioning, which is synonymous with interviewing skill (Read, Powell, Kebbell, & Milne, 2009; Rich & Seffrin, 2012). It is possible that the acquisition of a range of skills, such as best practice investigative interviewing, brief authorization, and engagement with other service providers and victims, is facilitated by concomitant training in sexual offending dynamics.

Second, the posttraining measurement point was scheduled immediately (i.e., on the final day of) training, and thus, research is needed to determine whether the changes observed are stable over time. Ongoing intervention may be needed to maintain these results, particularly if biases are still prevalent within the police units where the investigators work. Finally, the training course evaluated in this study was very resource intensive. It was developed and rolled out amid major state-wide funding aimed to improve the justice systems’ response to sexual assault. The course relied on the coordination of multiple stakeholders (including forensic professionals, school children, and victims), extensive physical space for training and interviewing exercises, facilitators with a background in psychology, offender treatment and police investigation, as well as ongoing development of up-to-date training content. Future research should examine the effectiveness of online training to provide a more cost-effective alternative for jurisdictions with less government funding or resources in this critical area.

Footnotes

Acknowledgements

Our thanks to Superintendent Rod Jouning, Mark Barnett, Tiffany Lewis, and the Sexual Offence and Child Abuse Investigation Team Specialist Development Unit for their support, especially Detective Sergeant Tony Breen, who assisted in questionnaire design. Thanks also to Amber Darwinkel for assistance with coding.

Notes

Elli Darwinkel is a PhD candidate with Deakin University, collaborating on an evaluation of the Whole Story framework with Victoria Police.

Martine Powell is a researcher with Deakin University who specialises in investigative interviewing, with over 160 publications in this area. She is a collaborator on the Whole Story framework evaluation.

Patrick Tidmarsh has a background in therapeutic treatment of sexual offenders and is currently a Forensic Interview Advisor for the Specialist Development Unit (SDU) of the Sexual Offence and Child Abuse Investigation Team (SOCIT) Project, Victoria Police. In addition, he is conducting research evaluating the Whole Story framework.

References

Ask

Granhag

P. A.

(2005). Motivational sources of confirmation bias in criminal investigations: The need for cognitive closure. Journal of Investigative Psychology and Offender Profiling, 2, 43-63.

Bloom

S. L.

(2003). Understanding the impact of sexual assault: The nature of traumatic experience. In Giardino

Datner

Asher

(Eds.), Sexual assault: Victimization across the lifespan (pp. 405-432). St. Louis, MO: GW Medical Publishing.

Brown

J. M.

Hamilton

O’Neill

(2007). Characteristics associated with rape attrition and the role played by scepticism or legal rationality by investigators and prosecutors. Psychology, Crime and Law, 13, 355-370.

Chapleau

K. M.

Oswald

D. L.

(2010). Power, sex, and rape myth acceptance: Testing two models of rape proclivity. Journal of Sex Research, 47, 66-78.

Craven

Brown

Gilchrist

(2006). Sexual grooming of children: Review of literature and theoretical considerations. Journal of Sexual Aggression, 12, 287-299.

Criminal Justice Sexual Offences Taskforce. (2005). Responding to sexual assault: The way forward. Sydney, Australia: Oxford Publishing.

Daly

Bouhours

(2010). Rape and attrition in the legal process: A comparative analysis of five countries. Crime and Justice, 39, 565-650.

Devine

D. J.

Clayton

L. D.

Dunford

B. B.

Seying

Pryce

(2001). Jury decision making: 45 years of empirical research on deliberating groups. Psychology, Public Policy, and Law, 7, 622-727.

Frazier

P. A.

Haney

(1996). Sexual assault cases in the legal system: Police, prosecutor, and victim perspectives. Law and Human Behavior, 20, 607-628.

10.

Garcia-Moreno

Watts

(2011). Violence against women: An urgent public health priority. Bulletin of the World Health Organisation, 89, 2.

11.

Hazzard

Rupp

(1986). A note on the knowledge and attitudes of professional groups toward child abuse. Journal of Community Psychology, 14, 219-223.

12.

Hill

Memon

McGeorge

(2008). The role of confirmation bias in suspect interviews: A systematic evaluation. Legal and Criminological Psychology, 13, 357-371.

13.

Horvath

Brown

(2005). Drug-assisted rape and sexual assault: Definitional, conceptual and methodological developments. Journal of Investigative Psychology and Offender Profiling, 2, 203-210.

14.

Howells

(1995). Child sexual abuse: Finkelhor’s precondition model revisited. Psychology, Crime and Law, 1, 201-214.

15.

Hunter

S. M.

Bentley-Cewe

B. R.

Mills

J. L.

(2000). Police response to crimes of sexual assault: A training curriculum (2nd ed.). East Hartford: Connecticut Sexual Assault Crisis Services, Inc.

16.

Jamel

Bull

Sheridan

(2008). Investigation of the specialist police service provided to male rape survivors. International Journal of Police Science & Management, 10, 486-508.

17.

Jordan

(2004). Beyond belief? Police, rape and women’s credibility. Criminal Justice: International Journal of Policy and Practice, 4, 29-59.

18.

Kelly

Lovett

Regan

(2005). A gap or a chasm? Attrition in reported rape cases (Home Office Research Study 293). London: Home Office.

19.

Levin

(2004). Common responses to trauma and coping strategies. Retrieved July 5, 2012, from http://www.trauma-pages.com/s/t-facts.php

20.

Lievore

(2003). Non-reporting and hidden recording of sexual assault: An international literature review. Canberra: Australian Institue of Criminology.

21.

Lisak

Gardinier

Nicksa

S. C.

Cote

A. M.

(2010). False allegations of sexual assault: An analysis of ten years of reported cases. Violence Against Women, 16, 1318-1334.

22.

Lonsway

K. A.

Welch

Fitzgerald

L. F.

(2001). Police training in sexual assault response: Process, outcomes, and elements of change. Criminal Justice and Behavior, 28, 695-730.

23.

May-Parker

Kiwala

(2000). Guidelines for police training on violence against women and child abuse. London: Commonwealth Secretariat.

24.

Mouzos

Makkai

(2004). Women’s experiences of male violence: Findings from the Australian component of the International Violence Against Women Survey (IVAWS). Canberra: Australian Institute of Criminology.

25.

New Jersey Division of Criminal Justice. (2003). Dynamics of domestic violence: In-service training for police officers. Trenton, NJ: Author.

26.

Nickerson

R. S.

(1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2, 175-220.

27.

Oxburgh

Ost

Cherryman

(2010). Police interview with suspected child sex offenders: Does use of empathy and question type influence the amount of investigation relevant information obtained? Psychology, Crime & Law, iFirst, 1-15.

28.

Page

A. D.

(2008). Gateway to reform? Policy implications of police officers’ attitudes toward rape. American Journal of Criminal Justice, 33, 44-58.

29.

Page

A. D.

(2010). True colors: Police officers and rape myth acceptance. Feminst Criminology, 5, 315-334.

30.

Patterson

G. T.

(2004). Evaluating the effects of child abuse training on the attitudes, knowledge, and skills of police recruits. Research on Social Work Practice, 14, 273-280.

31.

Payne

D. L.

Lonsway

K. A.

Fitzgerald

L. F.

(1999). Rape myth acceptance: Exploration of its structure and its measurement using the Illinois Rape Myth Acceptance (IRMA) Scale. Journal of Research in Personality, 33, 27-68.

32.

Powell

M. B.

Fisher

R. P.

Hughes-Scholes

C. H.

(2005). Investigative interviewing. In Brewer

Williams

(Eds.), Psychology and Law: An empirical perspective (pp. 11-42). New York: Guilford Press.

33.

Powell

M. B.

(2008). Designing effective training programs for investigative interviewers of children. Current Issues in Criminal Justice, 20, 189-208.

34.

Powell

M. B.

Cauchi

(2011). Victims’ perceptions of a new model of sexual assault investigation adopted by Victoria Police. Police Practice and Research: An International Journal (online first).

35.

Powell

M. B.

Murfett

Thomson

D. M.

(2010). An analysis of police officers’ decisions about whether to refer cases of child abuse for prosecution. Psychology, Crime & Law, 16, 715-724.

36.

Rassin

Eerland

Kuijpers

(2010). Let’s find the evidence: An analogue study of confirmation bias in criminal investigations. Journal of Investigative Psychology and Offender Profiling, 7, 231-246.

37.

Read

Powell

M. B.

Kebbell

Milne

(2009). Investigative interviewing of suspected sex offenders: A review of what constitutes best practice. International Journal of Police Science and Management, 11, 442-459.

38.

Rich

Seffrin

(2012). Police interviews of sexual assault reporters: Do attitudes matter? Violence and Victims, 27, 263-279.

39.

Schwartz

M. D.

(2010). National Institute of Justice visiting fellowship: Police investigation of rape—Roadblocks and solutions. Washington, DC: National Institute of Justice.

40.

Sleath

Bull

(2012). Comparing rape victim and perpetrator blaming in a police officer sample: Differences between police officers with and without special training. Criminal Justice and Behavior, 39, 646-665.

41.

Smith

R. M.

Powell

M. B.

Lum

(2009). The relationship between job status, interviewing experience, gender, and police officers’ adherence to open-ended questions. Legal and Criminological Psychology, 14, 51-63.

42.

Spohn

Tellis

(2008). Policing and prosecuting sexual assault in Los Angeles City and County: A collaborative study in partnership with the Los Angeles Police Department, the Los Angeles County Sheriff’s Department, and the Los Angeles County District Attorney’s Office. Los Angeles: U.S. Department of Justice.

43.

Spohn

Tellis

(2012). The criminal justice system’s response to sexual violence. Violence Against Women, 18, 169-192.

44.

Statewide Steering Committee to Reduce Sexual Assault. (2006). Study of reported rapes in Victoria 2000-2003. Melbourne, Australia: Office of Women’s Policy.

45.

Stern

(2010). The Stern review of rape reporting. London: Home Office.

46.

Tidmarsh

Powell

M. B.

Darwinkel

(in press). “Whole story”: A new framework for conducting investigative interviews about sexual assault. Journal of Investigative Interviewing: Research and Practice.

47.

Ullman

S. E.

(2010). Talking about sexual assault: Society’s response to survivors. Washington, DC: American Psychological Association.

48.

van Kesteren

J. N.

Mayhew

Nieuwbeerta

(2000). Criminal victimisation in seventeen industralised countries: Key findings from the 2000 International Crime Victims Survey. The Hague, the Netherlands: Ministry of Justice, Research and Documentation Centre (WODC).

49.

Victorian Law Reform Commission. (2004). Sexual offences final report. Victoria, Australia: Victorian Law Reform Commission.

50.

Wood

(2000). Attitude change: Persuasion and social influence. Annual Review of Psychology, 51, 539-570.