Acquaintance Rape

Abstract

The aim of the current study was to enhance the assessment and predictive accuracy of risk assessments for sexual offenders by utilizing detailed crime scene analysis (CSA). CSA was conducted on a sample of 247 male acquaintance rapists from Berlin (Germany) using a nonmetric, multidimensional scaling (MDS) Behavioral Thematic Analysis (BTA) approach. The age of the offenders at the time of the index offense ranged from 14 to 64 years (M = 32.3; SD = 11.4). The BTA procedure revealed three behavioral themes of hostility, criminality, and pseudo-intimacy, consistent with previous CSA research on stranger rape. The construct validity of the three themes was demonstrated through correlational analyses with known sexual offending measures and criminal histories. The themes of hostility and pseudo-intimacy were significant predictors of sexual recidivism. In addition, the pseudo-intimacy theme led to a significant increase in the incremental validity of the Static-99 actuarial risk assessment instrument for the prediction of sexual recidivism. The results indicate the potential utility and validity of crime scene behaviors in the applied risk assessment of sexual offenders.

Keywords

risk assessment acquaintance rape crime scene analysis Static-99

Eighteen percent of women living in the United States have been the victim of completed or attempted rape at sometime in their lives (Tjaden & Thoennes, 2006). Strangers are responsible for just 3% of these cases. There is still, however, a stereotype that “real rape” is committed by strangers (Estrich, 1987; Spohn & Tellis, 2012). The influence of this stereotype on victims is important as it may lead to the underreporting of nonstranger rape. Such stereotypes decrease the likeliness that police will believe the victim, that charges will be filed, and that juries will convict an offender in acquaintance rape cases (Bryden, 2000). They may also influence researchers who have shown relatively little interest in acquaintance rape compared with stranger rape (Wilson & Leith, 2001).

Studies on the risk of recidivism of acquaintance rapists are rare. Kanin (1985) found recidivism rates of 9% for self-disclosed acquaintance rapists, which is only slightly lower than the average sex offense recidivism rate of 10% to 15% of sexual offenders after 5 years (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005). Although the recidivism rate of acquaintance rapists is slightly lower than the average sex offense recidivism rate, Lisak and Miller (2002) indicated that almost two thirds of undetected rapists (the majority of whom perpetrated acquaintance rapes) subsequently sexually offend more than once. This suggests that recidivism among acquaintance rapists is both a significant and an understudied public safety issue.

To assess the risk of sexual recidivism, evaluators often use actuarial risk assessment instruments based on static and dynamic risk factors. Although the static/dynamic distinction has considerable utility, it describes differences in the indicators, not necessarily differences in the constructs being assessed. Even though researchers and forensic practitioners have increasingly emphasized the importance of behavioral assault characteristics (i.e., crime scene) for risk assessment (Beech, Fisher, & Thornton, 2003; West, 2000), few risk assessment instruments utilize crime scene variables as indicators of risk-relevant constructs. Recently, Mann, Hanson, and Thornton (2010) proposed that risk assessments should focus on psychological meaningful risk factors, or the enduring propensities that place individuals at risk for offending behaviors (e.g., sexual recidivism). In their meta-analysis, Mann et al. (2010) identified several empirically supported and promising psychologically meaningful risk factors, for example, sexualized violence, grievance/hostility, and offense-supportive attitudes. Following Mann et al. (2010), we argue that crime scene analysis (CSA) can help identify an offender’s propensities, which may also manifest in offense behaviors. In other words, the individual propensities of an offender (e.g., sexual interest in violence) may be activated during the course of an offense and manifest in a particular pattern of crime scene behaviors (e.g., offender tortures the victim, offender humiliates the victim).

Lehmann, Goodwill, Gallasch-Nemitz, Biedermann, and Dahle (2013) were the first to address the gap between CSA and risk assessment by using Behavioral Thematic Analysis (BTA), an approach to CSA developed in the field of investigative psychology (Canter, 2004). Using a sample of 167 stranger rape offenders, Lehmann et al. (2013) replicated the behavioral thematic structure of crime scene variables found in previous studies on stranger rape (e.g., Canter, Bennell, Alison, & Reddy, 2003), namely, criminality, hostility, and sexuality. Such themes are common in the general sexual offender literature and align with the three major components of the revised structure of Knight’s (2010) Massachusetts Treatment Centre: Rapist Typology, Version 4 (MTC:R4) classification (impulsivity, violence, and sexualization). More important, Lehmann and colleagues found that a crime scene behavioral theme relating to criminality in stranger rapes added significant incremental predictive ability to the Static-99 actuarial tool.

The current study extended the BTA approach taken by Lehmann et al. (2013) to a sample of acquaintance rape cases. As advocated by Lehmann et al., the application of BTA analysis must be grounded in a valid theoretical framework. Based on the extant literature on sexual offending and the BTA approach used by Lehmann et al., three potential themes were predicted to describe acquaintance rape offending behavior: (a) Hostility, (b) Criminality, and (c) Pseudo-Intimacy.

Hostility

The clinical literature on rape in general (e.g., Cohen, Seghorn, & Calmas, 1969; Groth, 1979), stranger rape (e.g., Canter et al., 2003), and rape within intimate partnerships (e.g., Finkelhor & Yllo, 1985; Shotland, 1992) describes a type of rape for which the central feature is the intention of the offender to hurt and punish the victim. The assault is characterized by physical, nonstrategic violence (Felson & Krohn, 1990). Sex is a weapon, and the aggressive behaviors are beyond those necessary to commit the offense, including aggressive undressing of the victim (e.g., tearing clothes) and forcing degrading sexual acts. The hostility theme is closely associated with overt sadism, including overtly hostile and degrading actions (Alison & Stein, 2001). In the MTC:R4 classification system (Knight, 2010), the sadistic and vindictive rapist types share many common features. Consequently, indicators of sexual sadism are also likely to be valid indicators of the dimension of hostility. As such, the current study included 7 of the 11 items of the Severe Sexual Sadism Scale (SeSaS; Mokros, Schilling, Weiss, Nitschke, & Eher, 2013). Given the common comorbidity of criminal aggression and alcohol intoxication (Hoaken et al., 2012), we expected alcohol use to be a feature of the hostility theme.

Because acquaintance rape offenses committed by male family members are often more aggressive than those committed by other groups of acquaintance rapists (Koss, Dinero, Seibel, & Cox, 1988), the hostility theme was expected to be positively related to the closeness of the victim–offender relationship. Furthermore, offenders raping violently within sexually active couples seem to use violence in sexual and nonsexual contexts (Frieze, 1983). Accordingly, hostile offenders are most likely to have a history of prior violent offenses (Canter & Heritage, 1990) and to reoffend with a violent offense (Mann et al., 2010). Given that Mann et al. (2010) found both grievance/hostility and sexualized violence to constitute psychological meaningful risk factors for sexual recidivism, the aggressive variant of acquaintance rape could be related to future sexual offending.

Criminality

The general theme of criminality occurs in a variety of forms throughout the literature on sex offender classification (e.g., Groth, 1979). For example, Gebhard, Gagnon, Pomeroy, and Christensen (1965) described an amoral delinquent, who uses threats or weapons if the woman will not fulfill her role as a sexual object. For the criminal offender, the rape is one among many antisocial behaviors (Knight, 1999). The criminal rapist steals sexual intercourse as he would steal money or property (Kopp, 1962). Because such offenders have previous experience of legal procedures (e.g., due to the probability that they have had contact with the criminal justice system for previous offenses), they might also be aware of the evidentiary implications of leaving semen and take precautions (e.g., use a condom). The offender has no specific intent to hurt the victim and only uses strategic violence (Felson & Krohn, 1990) to reach his goal of sexual gratification. Strategic violence may include controlling the victim by verbal threats, physical restraints (e.g., binding, gagging), and weapon use. Due to the high level of control over the victim, the offender is likely to complete the offense (i.e., ejaculation). Furthermore, the offender could use the opportunities presented by the crime to steal valuables from the victim. This theme could theoretically be indicative of the psychological meaningful risk factor of externalization and should predict nonsexual recidivism more strongly than sexual recidivism (Babchishin, Hanson, & Helmus, 2012).

Pseudo-Intimacy

This theme is characterized by aspects of nonviolent sexual coercion found within acquaintance rapes (Muehlenhard & Schrag, 1991). Hall and Hirschman (1991) described a sexual offender (i.e., cognitive subtype) who seems to choose situations where there is some perceived justification for the offense (e.g., victim enjoyment). Such offenders seem to adhere to the view that rape is not “real rape” if the man says he loves the victim, or is gentle with her. By showing offense behaviors frequently present in consensual relationships, offenders convince themselves that their assaults are mutually enjoyable and possibly consensual (McCabe & Wauchope, 2005). Behaviors indicating an attempt to bond with the victim, showing affection, showing the sexual arousal of the offender, or attempts to sexually arouse the victim would be expected to be included in the pseudo-intimacy theme. After the offense, the offender may recognize his misperception of the victim’s sexual intent and may apologize and show regret.

Because this theme can be related to an offender’s misperception of the victim’s sexual intent (e.g., Marx, Van Wie, & Gross, 1996), it has been described as “early date rape” (Shotland, 1992), which occurs after a few dates but before sexual “ground rules” have been established. In this regard, a negative relationship between the pseudo-intimacy theme and the closeness of the victim–offender relationship is hypothesized. Helmus, Hanson, Babchishin, and Mann (2013) identified offense-supportive attitudes as an empirically supported psychological risk factor for sexual recidivism. Such offense-supportive attitudes in rapists could include that rape is harmless or even enjoyable for the victim (e.g., Bumby, 1996). Because this theme is expected to be indicative of such offense-supportive attitudes, it could also be related to future sexual offending.

Purpose of Study

The aim of the current article was to investigate the underlying behavioral mechanism apparent in acquaintance rape to enhance our understanding of this understudied offense. Based on a review of the literature, content relevant to the three targeted themes was identified and systematically sampled (i.e., content validity; see the appendix for a full list of variable descriptions and their hypothesized themes). The predictive ability of the BTA approach was contrasted with the predictive ability of the Static-99, a risk tool explicitly designed to predict sexual recidivism.

The present study had three main objectives. The first objective was to replicate the BTA structure of crime scene variables of previous studies on stranger rape (specifically, hostility, criminality, and pseudo-intimacy) with a sample of acquaintance rapists using multivariate dimensional analysis. Second, the construct validity of the themes was tested, based on correlations with previous criminal history and established sex crime specific measures (Static-2002). Third, the predictive validity (i.e., sexual recidivism) of the behavioral thematic model was investigated, including the extent to which variables derived from this model added incrementally to established risk tools (Static-99).

Method

Sample

The initial sample included all 955 convicted adult, male sex offenders in Berlin, Germany, between 1994 and 2001, with sufficient information about crime scene behavior and follow-up time. The offenders in the sample were convicted of sexual abuse of children or adolescents in 42% of cases; of sexual assault, rape, or similar sexual offenses toward adults in 51% of cases; and of both in 7% of cases.

The current study included all cases of acquaintance rape (N = 247) where offenders had a fixed time for sexual recidivism of 5 years from their release from prison or forensic psychiatric services. Of the 247 offenders, 24% had prior charges for sexual offenses, whereas only 6% had prior convictions. According to the central registry, 8% of the offenders were convicted for a new sexual offense, whereas 15% were charged for a new sexual offense. Approximately 64% of the sample were German citizens; the remaining 36% were foreign nationals or had a dual citizenship. “Acquaintance rape” was defined as forced sexual intercourse that occurs between persons who are acquainted or romantically involved—anyone who is not a stranger. The age of the offender was recorded at the time of the index offense and ranged from 14 to 64 years (M = 32.3; SD = 11.4).

Data

The data for the current project were initially collected for a larger, distinct project involving the Institute for Forensic Psychiatry Berlin and the State Office of Criminal Investigations in Berlin. The purpose of this larger German project was to conceptualize the sexual offending process in a comprehensive and systematic way. Data were coded from computerized police databases, official criminal records, and paper records. In the majority of the cases, the paper records included a verbatim transcript of the police interrogation of the offender, and victim interviews. On occasion, they also included witness statements, court decisions, and official expert opinion(s). Official criminal records taken from the National Conviction Registry (NCR, Germany) were evaluated to tabulate the 5-year recidivism rates.

Coding

The data collection was conducted in two waves. At the beginning of the original project, a coding manual containing approximately 300 variables was developed by the primary investigator of the original project in collaboration with three student research assistants. After these three research assistants had collectively coded approximately 500 cases, a new team of three student research assistants was trained to code the remaining cases. As part of the transition between the first and second waves of data coding, one of the original coders and one of the new coders (who was not involved in the development of the original coding manual) double scored 42 previously unscored cases for rater reliability purposes. Although the reliability was acceptable for most variables (see below), the original criteria for the global motivation ratings proved ambiguous. Consequently, new motivation rating criteria were developed at this stage and applied to the 42 rater reliability cases and the remaining uncoded cases.

Although two of the authors of the current study were involved with the coding and the development of the coding manuals, the theory and hypotheses for the current study were developed after the coding was completed. The variables used in the current study were restricted to crime scene variables included in previous studies of CSA of sexual offending (e.g., Canter et al., 2003). However, variables from the SeSaS were added to improve the assessment of sadism (Nitschke, Osterheider, & Mokros, 2009). As well, some low frequency variables (<1%) and variables specific to stranger rape (e.g., offender implies knowing the victim) were excluded. In total, 44 variables were selected from the overall database of the broader German project for this study.

For the 42 reliability cases, the proportion of overall agreement (p₀)—the proportion of cases for which both raters agree—as well as Cohen’s kappa (κ) were calculated for the total of the more than 300 crime scene variables. The percent agreement for the 44 variables included in the current project ranged from .76 to 1.00, with a median value of .94. For 5 variables (offender binds victim, offender threatens weapon present, offender uses condom, offender blindfolds victim, offender gags victim), kappas could not be computed due to a lack of variance. In addition, for 5 variables (offender makes sexual comment, weapon from crime scene, offender steals clothing, and offender steals valuables), kappas were low (<.45) although percent agreement was high (range = 83%-98%). Nonetheless, all items were retained to replicate previous research and because they were considered relevant to the current hypotheses. Kappas for the remaining crime scene variables ranged from .45 to 1.00, with a median value of .77.

To assess the interrater agreement on the expected groupings of the crime scene variables (see the appendix), the first author and a professor for forensic psychology (neither involved in the larger German project nor in the current project) made a priori assignments of the variables into one of three themes (criminality, pseudo-intimacy, hostility). The percent agreement for the assignment of crime scene variables to behavioral themes was 95% with a kappa of .92.

The variables used to score Static-99 and Static-2002 were calculated by computer syntax from existing police databases. Consequently, rater reliability for these scales was not available. The syntax was written by two of the above research assistants based on the English versions of the Static-99 and Static-2002 scoring manuals (Harris, Phenix, Hanson, & Thornton, 2003; Phenix, Doren, Helmus, Hanson, & Thornton, 2008). Although neither of these research assistants had received official training in the STATIC instruments, they had carefully studied the manuals (sufficient to identify subtle differences in scoring between the English and German versions), and the police data sets were sufficiently detailed that few extrapolations were needed.

Measures

Static-99 (Hanson & Thornton, 2000) is a 10-item actuarial scale that assesses recidivism risk of adult male sex offenders. Researchers have found a moderate degree of accuracy in predicting sexual recidivism (Hanson & Morton-Bourgon, 2004).

Static-2002 (Hanson & Thornton, 2003) is a 14-item actuarial measure that assesses recidivism risk of adult male sex offenders. Static-2002 was created with the aim of improving Static-99 by adding new items, standardizing coding rules, and organizing the items into meaningful subscales. Two subscales were used in the current study: deviant sexual interests and general criminality. The deviant sexual interests subscale contains 3 items: noncontact sexual offenses, male victims, 2+ victims less than 12 years of age. The general criminality subscale contains 5 items: any prior involvement with the criminal justice system, number of prior sentencing occasions, community supervision violation, prior nonsexual violence, and time since last involvement with the criminal justice system.

Researchers have found high interrater reliability for Static-99/R and Static-2002/R (intraclass correlation coefficient [ICC] > .75; see Anderson & Hanson, 2010; Phenix et al., 2008; Phenix, Helmus, & Hanson, 2012). The Static-99 is not recommended for young offenders (those younger than 18 years at time of release). Nevertheless, four offenders in the sample were younger than 18 years at the time of release and were included in the analysis due to the focus of the current study being on victim–offender relationships independent of offender age.

Also, offender motivation (i.e., sexual, power, anger, and sadism) was rated by the research assistants for 42 cases on a 1 to 4 rating scale (1 = not at all; 4 = completely), loosely based on previous clinical definitions (e.g., Cohen et al., 1969; Groth, 1979). The interrater reliability for the judgments of offender motivation was moderate (Fleiss, 1981) with ICCs ranging from .45 to .74. Ratings of offender motivation were available for 90 of the 247 cases because offender motivation was only included in the second wave of data collection.

Data Analysis

To identify the underlying structure of the offense variables, we used a two-dimensional, nonmetric multidimensional scaling (MDS) analysis using the PROXSCAL module from SPSS for Windows (Version 19) at an ordinal level of measurement not allowing for ties. MDS analysis provides a graphical display of data that represents measurements of similarity among pairs of objects (e.g., crime scene variables) as distances between points in an n-dimensional space (Borg & Groenen, 1997). The distances between variables in the MDS solution can be interpreted as an approximation of their similarities; the closer they are together, the more similar they are to each other. The Jaccard measure of association (Jaccard, 1908) computed from a 2 × 2 contingency table (a / [a + b + c]) was used to transform the dichotomous data into a similarity coefficient matrix. To test the internal consistency of variables per theme, the Kuder–Richardson 20 (KR-20) coefficient was calculated for each of the hypothesized themes.

In line with previous research (e.g., Lehmann et al., 2013; Salfati & Canter, 1999; Trojan & Salfati, 2008), a continuous score of crime scene behaviors from each theme called the thematic sum score (TSS) was computed. The offender TSS score for each theme was derived by summing all offense behaviors of that theme that were present (e.g., performed) for a particular offender. Absence of a behavior in the records was counted as not present and scored as zero, so there were no missing values. The TSS scores for all offenders were then used to predict the probability of sexual recidivism (e.g., yes or no) using logistic regression analysis. The incremental effect of each of the TSS over the Static-99 for predicting sexual recidivism was examined by entering the Static-99 in the first block and then entering one of the three TSS themes in a separate, second block.

For the prediction of sexual recidivism, area under the curve (AUC) was used as an effect size statistic (see Swets, Dawes, & Monahan, 2000) in which AUCs of .56, .64, and .71 were considered small, medium, and large effects, respectively (Rice & Harris, 2005).

Results

Figure 1 shows the two-dimensional MDS solution with a Stress-1 index of 0.18 in 20 iterations, indicating a fair MDS solution (Kruskal, 1964). As previously mentioned, the closer two variables are, the more likely they co-occur within an offense. Typically, high frequency variables will cluster near to the center of the plot, and lower frequency variables will be on the periphery of the plot (see Table 2 for variable frequencies).

Figure 1.

Multidimensional scaling of 44 crime scene actions in 247 acquaintance rapes.

Frequencies

Consistent with previous research (Lehmann et al., 2013), we found the variables vaginal penetration, clothing disturbed, force to control the victim, and physical resistance of the victim to occur in more than 60% of the cases. These four variables were excluded from the computation of TSS scores due to their high frequency and, consequently, limited conceptual usefulness. High frequency sexual behaviors (more than 60%) are considered to define rape in general and offer little differentiation between themes. Consequently, they are typically removed in MDS studies (Canter, 1994; Canter et al., 2003; Canter & Heritage, 1990; Häkkänen, Lindlöf, & Santtila, 2004).

Themes

As hypothesized, we replicated the three previously proposed themes of hostility, criminality, and pseudo-intimacy (e.g., Alison & Stein, 2001). Furthermore, the variables of the SeSaS were found, as expected, within the hostility region. Seven of the 44 variables did not fall into expected regions, with some item misplacements not uncommon in interpersonal research (Plutchik & Conte, 1997). The mean number of variables (e.g., behaviors) committed per offense was 8.4 (SD = 3.6, range = 1-19). The means, standard deviations, observed ranges and possible ranges, and the proportion of the sample that had scores greater than zero on the TSS dimensions are summarized in Table 1. As can be gleaned from the MDS plot, the center of the thematic partitioning is not at the exact geometric midpoint of the MDS, which is not a requirement of MDS interpretation (Guttman & Greenbaum, 1998). Instead, thematic partition lines are derived based on a combination of theory, the results of the a priori interrater agreement (see the appendix), original intent of the items, and the internal reliability of variables to themes based on KR-20 coefficients. The percent agreement between the partition of the MDS plot and the a priori hypotheses about variable groupings was 83% (kappa = .73).

Table 1.

Summary of Psychometric Properties of the Thematic Sum Scores.

	Pseudo-Intimacy	Criminality	Hostility
M (SD)	1.41 (1.32)	1.36 (1.59)	2.86 (2.02)
Observed range	0-7	0-8	0-11
Possible range	0-10	0-14	0-16
Score > 0 in %	70.0	61.1	88.3
KR-20	.44	.62	.54
Correlations
Pseudo-intimacy	—	−.05	.05
Criminality	—	—	.27**

Note. KR-20 = Kuder–Richardson 20.

p < .05. **p < .01.

The identified offender themes are not offender “types.” Offender themes are propensities that may be related to one another, such that an offender may be high (or low) simultaneously on hostility, criminality, and pseudo-intimacy. In the current data, there was a significant relationship between hostility and criminality (r = .27, p < .001).

Criminality

Of the 14 variables found to form the criminality region, 11 had been predicted and 3 had not (Table 2). The three additional variables (i.e., offender steals personal items, offender steals clothing, offender demeans victim) had all been predicted to relate more to the hostility theme (see below). This may indicate that the theft of personal items is more related to criminal, rather than sadistic “trophy keeping” intent. The KR-20 coefficient for this theme was .62.

Table 2.

Themes With According Variables.

Pseudo-Intimacy	Criminality	Hostility
Fondle (45)	Verbal violence (39)	Physical force (59)
Confidence approach (24)	Ejaculation (26)	Victim injured (49)
Kiss (21)	Threatens to kill victim (15)	Verbal resistance (46)
Offender sexual comment (14)	Offender demeans victim (14)	Clothing torn (37)
Extends time (12)	Offender is forensically aware (9)	Blitz attack (28)
Offender says he loves victim (7)	Threatens no report (9)	Fellatio (16)
Cunnilingus (5)	Weapon to crime scene (7)	Offender intoxicated (10)
Offender bites victim (5)	Offender binds victim (7)	Offender humiliates victim (9)
Apologetic (5)	Steals valuables (3)	Anal penetration (9)
Offender reassures victim (4)	Steals personal items (2)	Weapon from crime scene (7)
	Offender gags victim (2)	Force against genitals (5)
	Steals clothing (1)	Penetration with object (5)
	Condom (1)	Film/photo (2)
	Threatens weapon present (1)	Ritualistic behavior (2)
		Victim is helpless (2)
		Offender blindfolds victim (1)

Note. Current study percentage frequencies are in brackets. Italicized variables were not found in the expected regions in the MDS analysis. MDS = Multidimensional scaling.

Pseudo-Intimacy

A priori hypotheses, based on the literature, suggested finding a theme relating to offender “pseudo-intimacy” (e.g., Canter et al., 2003). All of the variables expected to be found in this region were found only in this region. One additional variable (offender bites the victim) was unexpectedly found in this region, which may indicate a sexual component to this behavior (e.g., aggressive sucking). The KR-20 value of the 10 items was .44. The relative low coherence in this theme was unexpected; however, it may not be problematic. Low internal consistency can be a result of situations in which the correlation between alternate behavioral indicators of the same underlying construct is negative. For example, the offender reassuring the victim may be inversely correlated to the offender apologizing even though both could be considered external manifestations of attempting involvement with the victim.

Hostility

The proposed hostility theme (Table 2) contained 16 variables indicating a hostile interaction between offender and victim, with three variables being unexpectedly part of this theme (i.e., weapon from crime scene, offender blindfolds victim, victim is helpless). Nevertheless, blindfolding and incapacitating the victim could be interpreted as attempts to exercise power/control/domination according to the SeSaS (Knight, Sims-Knight, & Guay, 2013; Nitschke et al., 2009). The KR-20 coefficient for this theme was .54.

Construct Validity of the Behavioral Themes

The TSSs were correlated with the number of previous criminal (burglary, robbery), sexual, and nonsexual violent charges and convictions. Furthermore, the themes were correlated with ratings of the offender’s motivation and closeness of victim–offender relationship as well as with the sexual deviancy and general criminality subscales of the Static-2002R (Table 3).

Table 3.

Construct Validity: Correlations Between Behavioral Themes and Previous Offenses, Static-2002 Subscale, Static-99, and Offender Motivation (Pearson).

	Pseudo-Intimacy	Criminality	Hostility
Prior criminal offenses (No. of charges/convictions)
Burglary	−.06/−.03	.10/.04	−.04/.11*
Robbery	−.09/.02	.28/.21	.09/.05
Prior sexual offenses	−.03/−.04	−.04/−.05	.00/−.05
Prior violent offenses	−.06/.03	.09/.05	.13*/.09
Static-2002
Sexual deviance	−.03	.09	.13*
General criminality	.07	.02	.01
Static-99 total scores	−.05	.09	.07
Motivation
Sexual	.20*	−.10	−.31**
Power	−.11	.35**	.32**
Anger	−.18*	.19*	.39**
Sadism	−.08	.33*	.60**
Prior contact	−.20**	.07	.13*

p < .05. **p < .01.

The pseudo-intimacy theme was not related to any previous criminal behavior. Nevertheless, as predicted, this theme was the only one positively related to a sexual motivation (r = .20, p = .03). In addition, the pseudo-intimacy theme was negatively related to anger motivation (i.e., divergent validity). As predicted, it was negatively correlated with the closeness of victim–offender relationship (r = −.20, p = .001).

High scores in the criminality theme were correlated with an increased number of prior charges and convictions for robbery offenses but not for previous sexual, violent, or burglary offenses. Furthermore, the criminality theme was found to be most strongly related to power motivation (r = .35, p < .001). However, the expected correlation between the criminality theme and the general criminality subscale of the Static-2002R was not found (r = .02).

As hypothesized, only the hostility theme correlated with prior violent offending (i.e., prior charges) and showed the strongest relationship to anger (r = .39, p < .001) and sadistic motivation (r = .60, p < .001). In addition, only the hostility theme was related to sexual deviance as assessed by the Static-2002 subscore. Furthermore, high scores for this theme correlated with prior convictions for burglary. The hostility theme was also positively related to closeness of victim–offender relationship (r = .13, p = .02).

Sexual Recidivism

The bivariate analyses (Table 4) revealed that the two themes of pseudo-intimacy (AUC of .69) and hostility (AUC of .66) demonstrated positive associations with sexual recidivism. For the multivariate analysis, whether the offender was convicted of another sexual offense within 5 years of his release from prison was regressed onto the three thematic scores entered simultaneously in logistic regression. The overall model was significant, χ²(3) = 10.60; p = .014 (see Table 5), with only the pseudo-intimacy theme showing an incremental effect over the other themes. The Hosmer–Lemeshow goodness-of-fit test was nonsignificant, indicating the model was an adequate fit to the data. The AUC for the overall BTA model was .72, indicating a large effect (see Table 4).

Table 4.

Predictive Accuracy of the Three Themes, Behavioral Thematic Model, Static-99, Combined Models.

	Sexual recidivism
	AUC (95% CI)
Criminality	.47 [.34, .61]
Pseudo-intimacy	.69 [.56, .82]
Hostility	.66 [.56, .76]
BTA model	.72 [.60, .84]
Static-99	.67 [.53, .81]
Static-99 plus pseudo-intimacy	.76 [.65, .87]

Note. AUC = area under the curve; CI = confidence interval; BTA = Behavioral Thematic Analysis.

Table 5.

Prediction of Sexual Recidivism for a Fixed 5-Year Follow-Up Period Based on the Behavioral Thematic Model and Static-99 Scores.

		95% CI for Exp(b)
	B (SE)	Lower	Exp(b)	Upper	Model χ²(df)	Nagelkerke pseudo R²
Model 1					10.60(3)*	.11
Constant	−3.84*** (0.60)
Pseudo-intimacy	0.41* (0.16)	1.10	1.51	2.06
Hostility	0.23 (0.12)	0.99	1.26	1.61
Criminality	−0.19 (0.19)	0.57	0.82	1.20
Model 2					14.09(2)**	.14
Constant	−4.82*** (0.77)
Pseudo-intimacy	0.52** (0.17)	1.20	1.68	2.34
Static-99	0.42** (0.15)	1.12	1.52	2.05

Note. CI = confidence interval.

p < .05. **p < .01. ***p < .001.

Incremental Validity

In this data set, the average Static-99 score was 2.7 (SD = 1.6) and showed a moderate relationship with sexual recidivism (Table 4). To test the incremental validity of the three themes in relation to the Static-99, we conducted three logistic regression analyses in which the Static-99 score was entered first, and then one of the three behavioral themes was added as a second block. The Static-99 alone was a significant predictor of sexual recidivism, χ²(1) = 5.37; p = .04, and accounted for 5.5% of the variance. The regression analysis revealed that only the pseudo-intimacy (B = .52, p = .002) theme added incremental validity to the Static-99, whereas the criminality (B = −.16, p = .39) and hostility (B = .18, p = .11) themes did not. The model including the Static-99 and the pseudo-intimacy theme explained 14% of the variance (Table 5), a significant increase from Block 1, χ²(1) = 8.72; p = .003. The AUC for the model including both Static-99 and the pseudo-intimacy theme was .76, indicating a large effect (Table 4). The predictive analyses did not meaningfully change when performed with and without the four juvenile offenders. Furthermore, the results were not significantly changed using the Static-2002, Static-2002R, or Static-99R.

Discussion

The aim of this study was to replicate the BTA structure of previous studies on stranger rape with a sample of acquaintance rape cases. Furthermore, this study extends previous research by investigating the extent to which crime scene behavioral themes constitute psychological meaningful risk factors. In accordance with previous research on stranger rape (Alison & Stein, 2001; Canter, 1994; Canter et al., 2003; Canter & Heritage, 1990; Häkkänen et al., 2004), the three themes of hostility, criminality, and pseudo-intimacy were replicated. In addition, the sadistic component of acquaintance rape was identified as part of the hostility region. Moreover, the ability of detailed CSA using BTA to predict an acquaintance rapist’s risk of sexual recidivism was assessed. The results indicate that the behavioral themes of hostility and pseudo-intimacy predicted sexual recidivism and that the pseudo-intimacy theme added incrementally to the Static-99 risk assessment tool.

In line with Lehmann et al. (2013), the four variables, vaginal penetration of the victim, having their clothing disturbed, the use of force to control the victim, and physical resistance by the victim constituted the core features of rape according to the MDS analysis. This indicates that the level of force typically required in rape cases (i.e., force control) is that necessary to overcome the victim’s nonconsent (i.e., physical resistance). Furthermore, to accomplish vaginal penetration, it is necessary to remove the victim’s clothing. Given the consistency with Lehmann and colleagues’ previous findings for stranger rape, it appears that these behaviors are key elements that identify sexual intercourse as not consensual.

The three regions in the MDS plot were analogous in theory to previous research findings (e.g., Alison & Stein, 2001), as well as the three major rape outcomes proposed by Felson and Krohn (1990): sexual intercourse (“pseudo-intimacy”), harm to the victim (“hostility”), and domination of the victim (“criminality”). In general, the MDS plot indicates a division of acquaintance rape into violent (e.g., hostility) and less violent (i.e., pseudo-intimacy) rapes.

Building on research by Lehmann et al. (2013), the results show that objective crime scene behavior can help to infer the psychological meaning of the crime for the offender. Furthermore, the findings indicate that previous patterns of crime scene variables in stranger rapes are transferable to cases of acquaintance rape and are consistent with the existing and abundant general literature on sexual offender types. In particular, the three behavioral themes of criminality, pseudo-intimacy, and hostility resemble the three major components of the revised structure of the MTC rape classification (impulsivity, sexualization, and violence) proposed by Knight (2010).

In line with Knight (2010), the theme of pseudo-intimacy describes offenders relatively disinclined to violence (no relationship to prior violent offenses) and criminal impulsivity (no relationship to prior property crimes). Consequently, offenders scoring high on this theme would most likely be described as “sexual, nonsadistic” offenders according to the revised typology (Knight, 2010). In the revised MTC:R4 model, sexual nonsadistic falls at the moderate level of sexualization. In this regard, the theme of pseudo-intimacy showed a positive relationship with a sexual motivation but was not related to the sexual deviance subscale. Therefore, it is argued that the pseudo-intimacy theme might not be capturing aspects of hypersexuality (e.g., paraphilia). McCabe and Wauchope (2005) have argued that offenders showing offense behaviors frequently present in consensual relationships (e.g., kissing, fondling) may be trying to convince themselves that their assaults are mutually enjoyable and possibly consensual. Such offenders may believe that rape is not “real rape” if the man is gentle or “polite” (Polaschek & Ward, 2002). There is a certain degree of ambiguity associated with interpersonal behavior (Marx et al., 1996), which should be particularly true for situations where individuals engage in sexual activity for the first time (Littleton & Axsom, 2003). Abbey (1987) and Koss and Oros (1982) demonstrated that about 70% of women report that their friendliness had previously been mistakenly perceived as a sexual invitation by men. The previously mentioned typology of acquaintance rapists by Shotland (1992) differentiated between types of acquaintance rape with respect to whether the couple already participated in sexual intercourse or not. Accordingly, the pseudo-intimacy theme would best be described by early date rape, which occurs after several dates but before sexual ground rules are established. The negative relationship between the pseudo-intimacy theme and closeness of relationship could indicate this state of relationship. Also, research shows that men who report both a high likelihood of sexual assault and high rape myth acceptance in self-reports were less accurate in decoding a woman’s behavioral cues (e.g., misperception of sexual intent) in early states of the relationship but not in an intimate situation (Lipton, McDonel, & McFall, 1987; McDonel & McFall, 1991).

The criminality theme also overlaps relatively well with the criminality theme identified by Lehmann et al. (2013) in their BTA analysis of stranger rape cases. This theme seems to represent offenders with lower levels of aggression (no relationship to prior violent offenses) coupled with antisocial personality traits (significant relationship to prior charges and convictions for robbery) corresponding to an “opportunistic” offender according to the MTC:R4 (Knight, 2010). Sexually motivated burglars are also likely to steal from their victims; however, recent research suggests that sexually motivated burglars seek opportunities for sexually motivated crimes and do not just happen on potential sex crime victims during the course of planned property offenses (Pedneault, Beauregard, Harris, & Knight, 2014). In general, it is argued that the criminality theme could represent antisocial rapists with previous criminal experiences, who know how to control their victims; for example, verbally (e.g., threatens to kill), or physically (e.g., binding, gagging, utilizing weapons). Similar to that found in other studies of stranger rapists (e.g., Alison & Stein, 2001), previous criminal experience (e.g., knowledge of legal procedures) is further evidenced by bringing a weapon to the crime scene, using a condom, and being forensically aware. To further investigate the relationship between the theme of criminality and antisocial lifestyle, researchers should investigate the relationship to the Factor 2 lifestyle antisociality component of psychopathy (Hare, 2003) and to feelings, attitudes, and behaviors associated with an antisocial lifestyle (see Knight, 2010).

The hostility theme seems to represent offenders high in violence (significant correlation with prior charges for nonsexual violent offenses) and high on criminal impulsivity (associated with prior convictions for burglary). Furthermore, this theme was related to anger and sadistic offender motivation as well as the sexual deviance subscale of Static-2002. The new revised MTC:R4 classification (Knight, 2010) differentiates the violent types according to the absence or presence of sexualization and impulsivity. Sexualization and violence would yield the vindictive types, impulsivity and violence would yield the pervasively angry types, and the combination of all three components would increase the probability of sadism. To this point, there seems to be a relationship to offenders high in violence and expressive aggression, but researchers should investigate the relationship between the hostility theme and the three components of the MTC:R4 in more detail.

Furthermore, the indicators of sexual sadism, according to the SeSaS, were predominantly found to be part of the hostility theme. An interesting finding was that these behaviors were found at the outer edge of the MDS plot (i.e., low frequency). In line with Mokros et al. (2013), this could indicate sexual sadism being an extreme manifestation of sexualized violence on the dimension of hostility. Given the positive relationship with closeness of relationship and previous violent behavior, this theme could represent rape with battery within sexually active couples (Finkelhor & Yllo, 1985; Shotland, 1992).

In the current study, the pseudo-intimacy and hostility themes were significantly related to the persistence of sexual offending. This contrasts with previous findings for strangers rapists in which only the criminality theme was significantly related to sexual recidivism. This variation could simply be due to sampling error associated with small effects. However, it may have implications for different contributing factors for different types of offenses. In comparison with acquaintance rape, the rape of a stranger is a more prototypical crime. Consequently, those factors that are associated with a general propensity for rule violation may have particularly strong associations with stranger rape. In contrast, certain men who force sex on acquaintances may have genuine difficulty recognizing their behavior as wrong, believing that their victims wanted or deserved what they got.

Also, the theme of pseudo-intimacy showed incremental validity and significantly improved the predictive accuracy of the Static-99. Variables can contribute incrementally by either assessing new risk-relevant constructs or by improving the measurement of constructs already included. Previous research has shown that Static-99 predicts recidivism through the three major constructs of age, sexual criminality, and general criminality (Babchishin et al., 2012; Barbaree, Langton, & Peacock, 2006; Janka, Gallasch-Nemitz, & Dahle, 2011). Although loosely connected to the construct of sexual criminality, the construct of pseudo-intimacy may be sufficiently distinct to warrant further research as a risk factor in its own right.

The recidivism rate of this sample was 8%, which is close to the recidivism rate of 9% reported by Kanin (1985) for self-disclosed acquaintance rapists. However, readers should be cautious when interpreting the extent of officially recorded recidivism in acquaintance rape samples because these offenses are rarely reported to authorities. Furthermore, La Free (1980) found that there was a low likelihood of conviction for defendants with no criminal history and those who caused little physical injury to the victim. In addition, victims who sustained less serious injuries were less likely to report to the police (Bachman, 1998). Because of the low likelihood of apprehension, Hall and Hirschman (1991) argued that the cognitive subtype, similar to the current study’s pseudo-intimacy group, may be the most common of all sexually aggressive offenders.

Future Directions

Given that this study constitutes a first attempt to identify psychologically meaningful propensities from crime scene behavior in acquaintance rape, strong claims regarding practical implications for offender treatment are not yet possible. However, researchers could investigate, for example, whether offenders scoring high on “pseudo-intimacy” may benefit from cognitive interventions targeting sexual entitlement attitudes, whereas offenders high on hostility may benefit from therapy targeting emotional dysregulation.

There are also some limitations of this study that need to be discussed. Rater bias could account for some of the observed associations because the same raters coded all of the crime scene variables for each case. Consequently, some of the patterns could have been imposed on the crime scene data. This limitation would not apply, however, to the associations with the STATIC scores and recidivism outcomes as these were generated blind to the crime scene variables.

Another limitation is that the study only included convicted acquaintance rapists and they may be a distinctive subgroup of the population of all acquaintance rapists. As well, future researchers may want to control for the differing number of behaviors constituting each theme, which may have imposed bias on TSS values. Even though the empirical results indicate moderate to good construct validity of the three themes, validation is a process and not an outcome (Strauss & Smith, 2009) and the results need replication in national samples, including rural areas, as well as in other countries.

Conclusion

Psychological evaluators, whether assessing general clinical constructs or specific forensic-related measures, routinely collect diverse information to inform their assessments. The current study has shown that CSA can yield indicators of constructs relevant for sexual recidivism. In particular, the pseudo-intimacy and the hostility themes were found to be significantly related to the persistence of sexual offending in acquaintance rapists. Perhaps more important, though, the level (e.g., score) of apparent pseudo-intimate behavior an offender engaged in was found to provide significant incremental validity and improve the predictive accuracy of the Static-99 risk assessment tool. This suggests that current actuarial risk assessment tools may not be adequately accounting for the importance of pseudo-intimate behavior by acquaintance rapists in the prediction of future sexual recidivism. Even more important, it may be that current sexual risk assessment methods, in general, may be improved by utilizing crime scene behavioral indicators suggestive of an offender’s underlying offense motive. The results reported herein are strongly suggestive of the importance of including CSA in applied risk assessment.

Footnotes

Appendix

Variables used to describe behaviors during the offense of rape. All variables were coded dichotomous with values based on the presence or absence of each behavior. Variables are sorted by theoretically generated themes, informed by prior results.

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) received no financial support for the research, authorship, and/or publication of this article.

References

Abbey

(1987). Misperceptions of friendly behavior as sexual interest: A Survey of Naturally Occurring Incidents. Psychology of Women Quarterly, 11, 173-194.

Alison

L. J.

Stein

K. L.

(2001). Vicious circles: Accounts of stranger sexual assault reflect abusive variants of conventional interactions. The Journal of Forensic Psychiatry, 12, 515-538. doi:10.1080/09585180127391

Anderson

Hanson

R. K.

(2010). Static-99: An actuarial tool to assess risk of sexual and violent recidivism among sexual offenders. In Otto

R. K.

Douglas

K. S.

(Eds.), Handbook of Violence Risk Assessment (pp. 251-267). New York, NY: Taylor & Francis.

Babchishin

K. M.

Hanson

R. K.

Helmus

(2012). Even highly correlated measures can add incrementally to predicting recidivism among sex offenders. Assessment, 19, 442-461. doi:10.1177/1073191112458312

Bachman

(1998). The factors related to rape reporting behavior and arrest: New evidence from the National Crime Victimization Survey. Criminal Justice and Behavior, 25, 8-29. doi:10.1177/0093854898025001002

Barbaree

H. E.

Langton

C. M.

Peacock

E. J.

(2006). The factor structure of static actuarial items: Its relation to prediction. Sexual Abuse: A Journal of Research and Treatment, 18, 207-226. doi:10.1007/s11194-006-9011-6

Beech

A. R.

Fisher

D. D.

Thornton

(2003). Risk assessment of sex offenders. Professional Psychology: Research and Practice, 34, 339-352. doi:10.1037/0735-7028.34.4.339

Borg

Groenen

(1997). Modern multidimensional scaling: theory and applications. New York, NY: Springer.

Bryden

D. P.

(2000). Redefining rape. Buffalo Criminal Law Review, 3, 317-512.

10.

Bumby

(1996). Assessing the cognitive distortions of child molesters and rapists: Development and validation of the MOLEST and RAPE scales. Sexual Abuse: A Journal of Research and Treatment, 8, 37-54. doi:10.1007/BF02258015

11.

Canter

D. V.

(1994). Criminal shadows. London, England: HarperCollins.

12.

Canter

D. V.

(2004). Offender profiling and investigative psychology. Journal of Investigative Psychology and Offender Profiling, 1, 1-15. doi:10.1002/jip.7

13.

Canter

D. V.

Bennell

Alison

L. J.

Reddy

(2003). Differentiating sex offences: A behaviorally based thematic classification of stranger rapes. Behavioral Sciences & the Law, 21, 157-174. doi:10.1002/bsl.526

14.

Canter

D. V.

Heritage

(1990). A multivariate model of sexual offence behaviour: Developments in “offender profiling.” The Journal of Forensic Psychiatry, 1, 185-212. doi:10.1080/09585189008408469

15.

Cohen

Seghorn

Calmas

(1969). Sociometric study of the sex offender. Journal of Abnormal Psychology, 74, 249-255. doi:10.1037/h0027185

16.

Estrich

(1987). Real rape. Cambridge, MA: Harvard University Press.

17.

Felson

R. B.

Krohn

(1990). Motives for rape. Journal of Research in Crime & Delinquency, 27, 222-242. doi:10.1177/0022427890027003002

18.

Finkelhor

Yllo

(1985). License to rape: Sexual abuse of wives. New York, NY: Holt, Rinehart & Winston.

19.

Fleiss

J. L.

(1981). Statistical methods for rates and proportions (2 ed.). New York, NY: John Wiley & Sons.

20.

Frieze

I. H.

(1983). Investigating the causes and consequences of marital rape. Signs, 8, 532-553.

21.

Gebhard

P. H.

Gagnon

J. H.

Pomeroy

W. B.

Christensen

C. V.

(1965). Sex offenders: An analysis of types. New York, NY: Harper and Row.

22.

Groth

A. N.

(1979). Men who rape: The psychology of the offender. New York, NY: Plenum Press.

23.

Guttman

Greenbaum

C. W.

(1998). Facet theory: Its development and current status. European Psychologist, 3(1), 13-36. doi:10.1027//1016-9040.3.1.13

24.

Häkkänen

Lindlöf

Santtila

(2004). Crime scene actions and offender characteristics in a sample of Finnish stranger rapes. Journal of Investigative Psychology and Offender Profiling, 1, 17-32. doi:10.1002/jip.1

25.

Hall

G. C. N.

Hirschman

(1991). Toward a theory of sexual aggression: A quadripartite model. Journal of Consulting and Clinical Psychology, 59, 662-669. doi:10.1037/0022-006X.59.5.662

26.

Hanson

R. K.

Bussière

M. T.

(1998). Predicting relapse: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348-362. doi:10.1037/0022-006X.66.2.348

27.

Hanson

R. K.

Morton-Bourgon

(2004). Predictors of sexual recidivism: An updated meta-analysis. Corrections Research User Report No. 2004-02. Retrieved from http://www.static99.org/pdfdocs/hansonandmortonbourgon2004.pdf

28.

Hanson

R. K.

Morton-Bourgon

K. E.

(2005). The characteristics of persistent sexual offenders: A meta-analysis of recidivism studies. Journal of Consulting and Clinical Psychology, 73, 1154-1163. doi:10.1037/0022-006X.73.6.1154

29.

Hanson

R. K.

Thornton

(2000). Improving risk assessments for sex offenders: A comparison of three actuarial scales. Law and Human Behavior, 24, 119-136. doi:10.1023/A:1005482921333

30.

Hanson

R. K.

Thornton

(2003). Notes on the development of Static-2002 (Corrections Research User Report No. 2003-01). Retrieved from http://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/nts-dvlpmnt-sttc/index-eng.aspx

31.

Hare

R. D.

(2003). The Hare Psychopathy Checklist–Revised technical manual (2nd ed.). Toronto, Ontario, Canada: Multi-Health Systems.

32.

Harris

Phenix

Hanson

R. K.

Thornton

(2003). Static-99: Coding rules revised 2003. Ottawa, Ontario: Solicitor General Canada.

33.

Helmus

Hanson

R. K.

Babchishin

K. M.

Mann

R. E.

(2013). Attitudes supportive of sexual offending predict recidivism: A meta-analysis. Trauma, Violence, & Abuse, 14, 34-53. doi:10.1177/1524838012462244

34.

Hoaken

P. S.

Hamill

Ross

Hancock

Lau

Tapscott

(2012). Drug use and abuse and human aggressive behavior. In Verster

J. C.

Brady

Galanter

Conrod

(Eds.), Drug Abuse and Addiction in Medical Illness (pp. 467-477). New York, NY: Springer.

35.

Jaccard

(1908). Nouvelles recherches sur la distribution florale [New research on floral distribution]. Bulletin de la Societé des Sciences Naturale, 44, 223-270.

36.

Janka

Gallasch-Nemitz

Dahle

K.-P.

(2011). Zur Altersabhängigkeit von Risikovariablen bei Sexualdelinquenz [The effect of age on risk factors among sexual offenders]. Forensische Psychiatrie, Psychologie, Kriminologie, 5(1), 37-44. doi:10.1007/s11757-010-0087-y

37.

Kanin

E. J.

(1985). Date rapists: Differential sexual socialization and relative deprivation. Archives of Sexual Behavior, 14, 219-231. doi:10.1007/BF01542105

38.

Knight

R. A.

(1999). Validation of a Typology for Rapists. Journal of Interpersonal Violence, 14(3), 303-330. doi: 10.1177/088626099014003006

39.

Knight

R. A.

(2010). Typologies for rapists: The generation of a new structural model. In Schlank

(Ed.), The sexual predator (Vol. 5, pp. 1-28). Kingston, NJ: Civic Research Institute.

40.

Knight

R. A.

Sims-Knight

Guay

J.-P.

(2013). Is a separate diagnostic category defensible for paraphilic coercion? Journal of Criminal Justice, 41, 90-99.

41.

Kopp

S. B.

(1962). The character structure of sex offenders. American Journal of Psychotherapy, 16(1), 64-70.

42.

Koss

M. P.

Dinero

T. E.

Seibel

C. A.

Cox

S. L.

(1988). Stranger and acquaintance rape: Are there differences in the victim’s experience? Psychology of Women Quarterly, 12, 1-24. doi:10.1111/j.1471-6402.1988.tb00924.x

43.

Koss

M. P.

Oros

C. J.

(1982). Sexual Experiences Survey: A research instrument investigating sexual aggression and victimization. Journal of Consulting and Clinical Psychology, 50, 455-457. doi:10.1037/0022-006X.50.3.455

44.

Kruskal

(1964). Nonmetric multidimensional scaling: A numerical method. Psychometrika, 29, 115-129. doi:10.1007/BF02289694

45.

La Free

G. D.

(1980). Variables affecting guilty pleas and convictions in rape cases: Toward a social theory of rape processing. Social Forces, 58, 833-850. doi:10.1093/sf/58.3.833

46.

Lehmann

R. J. B.

Goodwill

A. M.

Gallasch-Nemitz

Biedermann

Dahle

K.-P.

(2013). Applying crime scene analysis to the prediction of sexual recidivism in stranger rapes. Law and Human Behavior, 37, 241-254. doi:10.1037/lhb0000015

47.

Lipton

D. N.

McDonel

E. C.

McFall

R. M.

(1987). Heterosocial perception in rapists. Journal of Consulting and Clinical Psychology, 55, 17-21. doi:10.1037/0022-006X.55.1.17

48.

Lisak

Miller

P. M.

(2002). Repeat rape and multiple offending among undetected rapists. Violence and Victims, 17, 73-84. doi:10.1891/vivi.17.1.73.33638

49.

Littleton

H. L.

Axsom

(2003). Rape and seduction scripts of university students: Implications for rape attributions and unacknowledged rape. Sex Roles, 49, 465-475. doi:10.1023/A:1025824505185

50.

Mann

R. E.

Hanson

R. K.

Thornton

(2010). Assessing risk for sexual recidivism: Some proposals on the nature of psychologically meaningful risk factors. Sexual Abuse: A Journal of Research and Treatment, 22, 191-217. doi:10.1177/1079063210366039

51.

Marx

B. P.

Van Wie

Gross

A. M.

(1996). Date rape risk factors: A review and methodological critique of the literature. Aggression and Violent Behavior, 1, 27-45. doi:10.1016/1359-1789(95)00003-8

52.

McCabe

M. P.

Wauchope

(2005). Behavioural characteristics of rapists. Journal of Sexual Aggression, 11, 235-247. doi:10.1080/13552600500272820

53.

McDonel

E. C.

McFall

R. M.

(1991). Construct validity of two heterosocial perception skill measures for assessing rape proclivity. Violence and Victims, 6, 17-30.

54.

Mokros

Schilling

Weiss

Nitschke

Eher

(2013). Sadism in sexual offenders: Evidence for dimensionality. Psychological Assessment, 26, 138-147. doi:10.1037/a0034861

55.

Muehlenhard

C. L.

Schrag

J. L.

(1991). Nonviolent sexual coercion. In Parrot

Bechhofer

(Eds.), Acquaintance rape: The hidden crime (pp. 115-129). New York, NY: Wiley.

56.

Nitschke

Osterheider

Mokros

(2009). A cumulative scale of severe sexual sadism. Sexual Abuse: A Journal of Research and Treatment, 21, 262-278. doi:10.1177/1079063209342074

57.

Pedneault

Beauregard

Harris

D. A.

Knight

R. A.

(2014). Rationally Irrational: The Case of Sexual Burglary. Sexual Abuse: A Journal of Research and Treatment. Advance online publication. doi: 10.1177/1079063213511669

58.

Phenix

Doren

Helmus

Hanson

R. K.

Thornton

(2008). Coding rules for Static-2002. Retrieved from http://www.static99.org/pdfdocs/static2002codingrules.pdf

59.

Phenix

Helmus

Hanson

R. K.

(2012). Static-99R & Static-2002R evaluator’s workbook. Retrieved from http://www.static99.org/pdfdocs/Static-99RandStatic-2002R_EvaluatorsWorkbook2012-07-26.pdf

60.

Plutchik

Conte

(1997). Circumplex models of personality and emotions. Washington, DC: American Psychological Association.

61.

Polaschek

D. L. L.

Ward

(2002). The implicit theories of potential rapists: What our questionnaires tell us. Aggression and Violent Behavior, 7, 385-406. doi:10.1016/S1359-1789(01)00063-5

62.

Rice

M. E.

Harris

G. T.

(2005). Comparing effect sizes in follow-up studies: ROC Area, Cohen’s d, and r. Law and Human Behavior, 29, 615-620. doi:10.1007/s10979-005-6832-7

63.

Salfati

C. G.

Canter

D. V.

(1999). Differentiating stranger murders: Profiling offender characteristics from behavioral styles. Behavioral Sciences & the Law, 17, 391-406. doi:10.1002/(SICI)1099-0798(199907/09)17:3<391::AID-BSL352>3.0.CO;2-Z

64.

Shotland

R. L.

(1992). A theory of the causes of courtship rape: Part 2. Journal of Social Issues, 48(1), 127-143. doi:10.1111/j.1540-4560.1992.tb01161.x

65.

Spohn

Tellis

(2012). The criminal justice system’s response to sexual violence. Violence Against Women, 18, 169-192. doi:10.1177/1077801212440020

66.

Strauss

M. E.

Smith

G. T.

(2009). Construct validity: Advances in theory and methodology. Annual Review of Clinical Psychology, 5, 1-25. doi:10.1146/annurev.clinpsy.032408.153639

67.

Swets

J. A.

Dawes

R. M.

Monahan

(2000). Psychological science can improve diagnostic decisions. Psychological Science in the Public Interest, 1, 1-26. doi:10.1111/1529-1006.001

68.

Tjaden

Thoennes

(2006). Extent, nature, and consequences of rape victimization: Findings from the National Violence Against Women Survey. Washington, DC: National Institute of Justice.

69.

Trojan

Salfati

C. G.

(2008). Methodological considerations of determining dominance in multidimensional analyses of crime scene behaviours and offender characteristics. Journal of Investigative Psychology and Offender Profiling, 5, 125-145.

70.

West

(2000). Clinical assessment of homicide offenders. Homicide Studies, 4, 219-233. doi:10.1177/1088767900004003002

71.

Wilson

M. A.

Leith

(2001). Acquaintances, lovers, and friends: Rape within relationships. Journal of Applied Social Psychology, 31, 1709-1726. doi:10.1111/j.1559-1816.2001.tb02747.x