Impression management and self-report among violent offenders

Offenders are assumed by many to employ socially desirable responding (SDR) response styles when completing self-report measures. Contrary to expectations, prior research has shown that accounting for SDR in self-report measures of antisocial constructs does not improve the relationship with outcome. Despite this, many self-report measures reliably predict future criminal outcome criteria. The present research examines the relationship of SDR (self-deception and impression management) with self-reported antisocial attitudes and the outcome of criminal recidivism in a sample of violent offenders. Offenders high on impression management reported lower antisocial attitudes. However, when impression management was statistically partialed from antisocial attitudes, the relationship with recidivism tended to diminish, though not to a statistically significant degree. This finding, though hypothesized based on previous empirical findings, is contrary to the theoretical assumption that controlling for SDR should improve the relationship of self-report with outcome. The discussion centers on the implications of routinely removing impression management from self-report.     

Socially desirable responding and psychometric assessment of dynamic risk in sexual offenders against children

Does socially desirable responding (SDR) represent a threat to the validity of dynamic risk assessment in sex offenders’ self-reports? We studied a sample of men (N?=?218) who completed a psychometric assessment battery while enrolled at Kia Marama, a prison-based treatment program for sexual offenders against children. SDR, as measured by the Marlowe-Crowne scale, was elevated compared to non-offender samples, increased from pre- to post-treatment, and was negatively correlated with dynamic risk (rs?=??.30 to ?.40), consistent with previous research. Measures of dynamic risk factors derived from the psychometric battery were correlated with sexual recidivism, but correlations were little changed when variance associated with SDR was removed using a regression procedure [Mills, J. F., &; Kroner, D. G. (2006). Impression management and self-report among violent offenders. Journal of Interpersonal Violence, 21(2), 178–192.]. These results suggest that there is a substantial component of the variance in psychometric self-reports of sexual offenders that is associated with SDR (approximately 10%), but that it does not compromise the predictive or construct validity of dynamic risk measures derived from these reports.     

A modified Stroop task with sexual offenders: Replication of a study

Abstract

Cognitive–behavioural treatment of sexual offenders assumes that sexual offenders are motivated by deviant attitudes, perceptions and values. Although aspects of deviant schema can be assessed by questionnaires, self-report measures are limited by the respondent's willingness to be forthright and by the fact that, typically, these cognitive processes occur quickly, revealing signs of automaticity. Recent research by Smith and Waterman has suggested that the deviant schema of sexual offenders could be assessed using a version of the Stroop colour-naming task. Long latency periods to sexual colour words imply a longer information-processing route and evidence of pre-established (deviant) sexual cognitive schema. Stroop techniques may offer the advantage of eliminating limitations that arise when using self-report techniques, such as fakeability and social desirability concerns. The current study replicates and extends Smith and Waterman's results using samples of sexual offenders, non-sexual violent offenders and non-violent offenders. The cumulative results of the two studies suggests that Stroop techniques have promise, but that further work is required before measures are available that have sufficient reliability and validity to be used in applied contexts.     

Social anxiety and sexual offending against children: A cumulative meta-analysis

Abstract

We examined the extent to which sexual offenders against children (SOC) differ from men who do not offend against children (non-SOC) on social anxiety. In study 1, 30 SOC and 31 non-sex offenders (NSO) were compared on a self-report measure of social anxiety (Social Avoidance and Distress Scale; SADS). SOC and NSO had virtually identical levels of social anxiety. In study 2, we conducted a meta-analysis of existing comparisons of SOC with various non-SOC groups on the SADS. SOC were generally more socially anxious than non-SOC. These findings suggest that the average SOC is more socially anxious than the average male who does not sexually offend against children. It remains unclear, however, whether social anxiety plays a causal role in the initiation of child sexual abuse.     

Residential Relegation of Registered Sex Offenders

This study builds on existing research (Hughes and Kadleck 2008; Mustaine et al. 2006a; Tewksbury and Mustaine 2006) that shows registered sex offenders are more likely to live in undesirable and socially disorganized communities. We extend such analyses to a geographically and demographically different community and employ more sophisticated measures of social disorganization concepts to assess the validity of the claim that registered sex offenders are relegated to socially disorganized communities. Data from Orange County, Florida, including measures of violent crime and the social disorganization concepts of concentrated disadvantage, residential instability and immigrant concentration are used to examine the distribution of residential locations for registered sex offenders. Results show that higher concentrations of sex offenders are found in communities with more concentrated economic disadvantage, more residential instability and higher rates of robbery and child sexual abuse. Implications of these findings for public policy are discussed.     

The generalized instructional set of the Criminal Attribution Inventory (CRAI): socially desirable responding and faking

Accurate assessment of criminal attributions can assist in explaining crime and identifying offender treatment targets. The current study examines the relationship between a new method of assessing the cognition of criminal attributes, the Criminal Attribution Inventory (CRAI) and a measure of socially desirable responding. 300 participants including incarcerated offenders, released offenders, sex offenders, and university students participated in the current study. Results indicate minimal relationships between socially desirable responding and the CRAI. In a fake-good testing situation, the faking index had appropriate sensitivity, specificity, and positive and negative predictive power in identifying fake-good responses. The CRAI's instructional set of general observation as opposed to self-representation appears to limit socially desirable responding and also allows the CRAI to be administered to nonoffender populations.     

A psychometric comparison of internet and non-internet sex offenders from a community treatment sample

Abstract

This paper compares psychometric test assessments of 39 men convicted of internet sex offences with 39 convicted of contact offences against a specific victim. Higher rates of socially desirable responding were identified in the internet group, as well as higher rates of emotional loneliness and under assertiveness, although the latter two findings were not statistically significant. This group also achieved lower scores measuring external locus of control, which might reflect their generally higher levels of education and employment (identified from research carried out in tandem with this work). The internet group achieved lower scores on sexualized attitudes towards children, emotional congruence with children and empathy distortions with regard to victims of child abuse. Hence, although they exhibited the kinds of general personality problems exhibited by other sex offenders, they did not support attitudes that explicitly endorse or condone the sexual abuse of children. This paper presents some hypotheses about these results and identifies further research necessary to develop understanding about this new kind of sexually abusive behaviour.     

Predicting multiple victim versus single victim sexual abuse: an examination of distal factors and proximal factors associated with the first abuse incident

Abstract

Official and confidential self-report data on 83 convicted adult male sexual abusers were analysed to examine whether sexual offending progression can be better predicted from distal antecedents, or from proximal antecedents and outcomes associated with the first sexual abuse incident. Fifty-six offenders who sexually abused more than one victim (multiple victim offenders; MVOs) were compared to 27 offenders who sexually abused a single victim only (single victim offenders; SVOs). MVOs were younger at the time of their first sexual abuse incident, and were more likely to first abuse male and non-familial children. With the exception of sexual attraction to male children, no differences were found between the two groups on distal antecedents. Proximal antecedents and outcomes associated with the first sexual abuse incident were significantly related to multiple victim offending. Logistic regression analysis identified the presence of sexual difficulties in the month prior to the first abuse incident, and sexual excitement immediately preceding the first incident, as significant unique predictors. Implications for risk assessment and risk management are discussed, and future research directions proposed.     

A cognitive-based indicator of deviant sexual interest: Concurrent validation of the Stroop task

ABSTRACT

One of the strongest predictors of sexual recidivism is sexual deviance [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]. Phallometric testing, the most commonly used method of assessing sexual deviance, has elicited methodological and ethical criticisms, while self-report is vulnerable to social desirability and poor insight. To overcome these limitations, researchers have utilised cognitive measures, including a modified Stroop task, to measure deviant sexual interests among sex offenders using victim selection as a comparison measure. However, the results have been inconclusive. The current study explored the validity of the modified Stroop task as a measure of deviant sexual interest among a non-offending sample of 570 females and 223 male participants, using self-report as the comparison measure. The results indicated a significant gender difference in concurrent validity; there was a significant relationship between self-reported deviance and deviant word Stroop performance for male but not for female respondents, suggesting the Stroop is a viable option for assessing sexual deviance among males. Implications of these findings are discussed relative to sexual recidivism risk assessment.     

Victim empathy in young sex offenders in the emergent adulthood developmental phase

ABSTRACT

This article reports on a study which focused on young sex offenders’ empathy levels for sexual abuse victims in general as portrayed in a case study, as well as empathy for the offenders’ own specific victims. Beckett and Fisher’s Victim Empathy Distortion Scale (1994) was used to measure and compare 96 young sex offenders’ empathy levels. The quantitative research results indicated that research participants displayed significantly less empathy towards their own victims when compared to the empathy displayed towards a general sexual abuse victim in a case study. Following the completion of the questionnaires, in-depth, qualitative data was obtained regarding the young sex offenders’ thoughts prior to, during and directly after committing the offence. In addition, they also explained their current thoughts and feelings for the victims in their case. The divergent responses which were given by the participants is a clear indication of the heterogeneous nature of youth sex offending.     

Good practice in eliciting disclosures from sex offenders

Abstract

Eliciting useful and reliable disclosures from sex offenders is imperative for effective risk management. Responsiveness to accurate self-report by sexual offenders of their risky thinking and problematic behaviours ensures that offenders are dealt with at the correct risk level. This requires practitioners to use a range of constructive strategies and techniques for eliciting disclosures. Drawing on a recent study that investigated the nature and usefulness of disclosures, this paper explores the inherent challenges and identifies best practice in this important area of work.     

Social networks of sexual offenders

Abstract

Despite the extensive research attention directed toward criminal peer groups, there has been little attention to the peer groups of sexual offenders. Sexual offenders are often considered to be loners who offend in isolation, but cases of sex crimes involving peer support are not difficult to find, e.g., child sex rings, gang rapes. To examine the role of peers in sexual offending, sexual offenders and nonsexual offenders were asked whether they knew other people who have committed sexual crimes. In comparison to the nonoffender community comparison group, the sexual offenders reported considerably more association and identification with sexual offenders. The associations also tended to be offence specific, such that child molesters knew other child molesters and rapists knew other rapists. These findings have important implications for the assessment and treatment of sexual offenders.     

Measuring criminal attributions with a normative instructional set: Is there a difference?

1Purpose . The purpose of this study was to evaluate the utility of the Criminal Attribution Inventory's (CRAI) normative instructional set (respondent's perception of the average type of crime) in assessing criminal attributions. Methods . To determine the role of the CRAI's instructional set, a content‐equivalent, personal instructional CRAI set was created. Correlations with an established criminal attribution scale (Blame Attribution Inventory [BAI]) and a measure of socially desirable responding (Paulhus deception scales [PDS]) were calculated. Partial correlations between the CRAI and BAI, controlling for the personal instructional CRAI, were calculated. Partial correlations were also calculated between the personal instructional CRAI and the BAI, controlling for socially desirable responding. Results . The normative instructional CRAI assessed similar domains as the personal instructional CRAI and made an additional contribution to criminal attributions. Socially desirable responding was minimally related to the normative instructional CRAI and could not account for the differences between the normative and personal instructional CRAI sets. Conclusions . The CRAI's normative instructional set assesses personal criminal attributions within the external blame domain assessing unique attributional variance. Such an instructional set has utility in assessing criminal attributions among those offenders who deny their offences.     

The long term recidivism risk of young sexual offenders in England and Wales– enduring risk or redemption?

PurposeTo examine the long-term sexual recidivism risk of juvenile sex offenders in England and Wales, and to compare the risk to that of a first time sexual offense for non-convicted juveniles. Additionally, the study explores the long term sexual recidivism risk of other types of juvenile offenders, and the long term violent recidivism risk of these groups.MethodsThe England and Wales Offenders Index was used to extract birth cohort data. Life table methods were used to estimate cumulative recidivism risk, and discrete time hazard models were used to compare hazard functions.ResultsAt the five year period, 7% of juvenile sexual offenders have been reconvicted of a sexual offense; reaching 13% by the end of the 35 year follow-up. When the reconviction hazard of the juvenile sexual offenders was compared with the first sexual conviction risk of a non-convicted comparison group, the hazards converged statistically after 17 years.ConclusionsThe study has implications for the registration periods of juvenile sex offenders. Indefinite registration for some juveniles needs to be considered, and a review of registration after a conviction free period would provide more balance between the protection of the public and the rights of the offender.     

Female Sexual Homicide Offenders: A Descriptive and Comparative Study

Research on female sexual homicide has been very scarce. In Europe, it has rarely been examined, and in Scotland, it has never previously been studied. This exploratory study aims to examine the characteristics of sexual homicides involving female offenders between 1990 and 2015 in Scotland. Using data from the Scottish Homicide Database between 1990 and 2015, female sexual homicides (n = 7) were compared to nonsexual homicides committed by females (n = 106) and to sexual homicides committed by men (n = 89) using Fisher's exact tests. The findings show that although female sexual homicide offenders are similar to both female nonsexual homicide offenders and male sexual homicide offenders in certain aspects, there are important differences that distinguish sexual homicides involving female offenders from both groups. Female sexual homicide offenders can arguably be seen as a distinct group of offenders, with specific characteristics and specific needs.     

Speaking about sexual abuse in British South Asian communities: offenders,victims and the challenges of shame and reintegration

Abstract

Cultural dynamics have a significant impact on how sexual matters, including sexual abuse, are discussed in British South Asian communities. The ways in which these communities talk about sexual violence often reinforce patriarchal norms and values, especially those concerned with honour and shame. As a result, victims are either silenced or the blame for the sexual violence they have suffered is laid at their own feet. Addressing the fact that these problems are rooted in patriarchal norms and values is key to understanding how to tackle sexual offending effectively in such communities. Both retributive and restorative justice are necessary in responding to sex crimes; retributive approaches help to recognise victims’ suffering, while restorative approaches offer promising avenues for encouraging victims and offenders alike to speak about their experiences. Both approaches are essential components to reintegrating victims and offenders into their communities.     

“I didn’t do it,but if I did I had a good reason”: Minimization,Denial, and Attributions of Blame Among Male and Female Domestic Violence Offenders

Women are increasingly being arrested and prosecuted for assaulting an intimate partner. Whereas extensive research has been conducted to identify the treatment needs of male domestic violence offenders, few studies have examined females convicted of the same charges. In the present study 1,267 men and 159 women convicted of intimate partner abuse were compared on scales assessing attributions of blame for their recent offense, minimization, denial, and socially desirable responding. Research with male offenders has identified these factors as important treatment targets, as they appear to influence an offenders risk for noncompliance and recidivism. The results of the study suggest that both male and female domestic violence offenders engage in socially desirable responding during court-ordered evaluations, that both attribute greater blame for the recent offense to their spouse/partner than they acknowledge for themselves, and that significant numbers of both genders deny the recent incident and/or minimize the severity of the offense. Areas for further research are highlighted along with a discussion of the implications of these findings for practitioners.     

Male peer support and a feminist routing activities theory: Understanding sexual assault on the college campus

Routine activities theorists traditionally have assumed offenders' motivation and victims' suitability from demographic correlates, and have done little to study effective guardianship. In this paper we ask questions directly of male date rape offenders to test the proposal that male peer support provides motivation; we ask lifestyle questions directly of both female victims and male offenders; and we discuss the extent to which abusive peers eliminate guardianship. Data from the Canadian National Survey support routine activities theory, and show that men who drink two or more times a week and have male peers who support both emotional violence and physical violence are nearly 10 times as likely to admit to being sexual aggressors as men who have none of these three traits.     

Attachment styles and violence in child molesters

Abstract

It has been suggested that sexual offenders lack the capacity to engage in or to maintain intimate relationships and that their attachment styles are insecure. The present study tested some of the implications of this theory with a group of incest offenders, nonfamilial child molesters, nonsex offenders, and nonoffenders recruited from the local community. Subjects completed a self-report questionnaire detailing their characteristic relationship style, and the level of violence employed by all sexual offenders was obtained from institutional files. Nonfamilial child molesters were five times more likely to report a fearful avoidant relationship style than a secure style compared to the community group. When they rated the degree to which they considered themselves to be fearful avoidant, the ratings of the nonfamilial child molesters were significantly higher than the ratings of community subjects and incest offenders. Incest offenders did not differ from community controls or from nonsex offender controls. Of the pooled group of child molesters, dismissively avoidant offenders employed higher levels of aggression in their offences than did secure or fearful avoidant offenders. These results are discussed in the light of current theory, and suggestions for future research and treatment initiatives are offered.     

Criminal history and social disadvantage as predictors of the severity of violent offending

ABSTRACT

Lethal violence is often seen as the tip of the iceberg and homicide perpetrators are seen as manifesting the most extreme number of various risk factors. This article explores whether that is the case. Using a unique data set combining data from several administrative registers with a nationally representative sample of different types of police-reported violence committed during 2010–2011 (N = 26,303 offenders) in Finland, we compare the offenders of five different types of violence (minor assault, assault, aggravated assault, attempted homicide, and completed homicide). In addition, we examine the association between the severity of violence and prior criminal history and different types of strain. The results give partial support to the hypothesis: the more serious the violence, the more crime prone and socially disadvantaged the offender. Yet, lethal offenders do not stand out alone; the division, rather, appears to be between offenders of serious (aggravated assault, attempted homicide, completed homicide) and less serious (minor assault, assault) forms of violence.