How nonrecidivism affects predictive accuracy: evidence from a cross-validation of the Ontario Domestic Assault Risk Assessment (ODARA)

Prediction effect sizes such as ROC area are important for demonstrating a risk assessment's generalizability and utility. How a study defines recidivism might affect predictive accuracy. Nonrecidivism is problematic when predicting specialized violence (e.g., domestic violence). The present study cross-validates the ability of the Ontario Domestic Assault Risk Assessment (ODARA) to distinguish subsequent recidivists and nonrecidivists among 391 new cases with less extensive criminal records than previous cross-validation samples, base rate=27%, ROC area=.67. Excluding ambiguous nonrecidivists increases the base rate to 33%, ROC area=.74. Random samples of 50 recidivists and 50 unambiguous nonrecidivists yield ROC areas from .71 to .80. Published norms significantly underestimate official recidivism. Ambiguous nonrecidivism is prevalent and leads to underestimating base rates and predictive accuracy.     

Intimate Partner Homicide: Risk Assessment and Prospects for Prediction

Little is known about assessing the risk of intimate partner homicide (IPH). Research has shown that women killed by an intimate partner scored higher than abuse survivors in retrospectively measured risk for IPH. In this study, we examined the characteristics of 146 men who committed an actual or attempted act of IPH. Of these, 42% had prior criminal charges, 15% had a psychiatric history, and 18% had both; events which could feasibly have permitted a prior formal assessment of risk. We also identified a subsample of 30 who could be scored on the Ontario Domestic Assault Risk Assessment (ODARA; Hilton et al., Psychological Assessment, 16, 267–275, 2004). The mean ODARA score was at the 80th percentile of risk for domestic violence, although only 13 had a previously documented partner assault. We conclude that co-operation among sectors responding to domestic violence and the shared use of validated risk assessment will increase the prediction and potential prevention of IPH.     

Risk assessment for intimate partner violence: how can the police assess risk?

ABSTRACT

The use of risk assessment tools by frontline police for intimate partner violence has the potential to make a difference to policing. In this paper, the key aspects of intimate partner violence risk assessment are outlined critically with a particular emphasis on how they can be used in practice. Two, evidence-based, exemplars are reviewed. These are the Ontario Domestic Abuse Risk Assessment (ODARA), an example of the actuarial approach, and the Brief Spousal Assault Form for the Evaluation of Risk (B-SAFER), an example of the structured professional judgement approach. In addition, the victim giving his or her own appraisal of risk is discussed. All three approaches have some validity when administered properly but practical factors reduce this validity. The content of the risk assessment tools are outlined and practical concerns such as training, time to administer, reliability, validity, and the overlap of intimate partner violence with other forms of offending are discussed. A balanced overview of the strengths, weaknesses and future potential of intimate partner violence risk assessment is provided.     

Assessment and management of risk for intimate partner violence by police officers using the Spousal Assault Risk Assessment Guide

Intimate partner violence (IPV) is a crime that is present in all countries, seriously impacts victims, and demands a great deal of time and resources from the criminal justice system. The current study examined the use of the Spousal Assault Risk Assessment Guide, 2nd ed. (SARA; Kropp, Hart, Webster, & Eaves, 1995), a structured professional judgment risk assessment and management tool for IPV, by police officers in Sweden over a follow-up of 18 months. SARA risk assessments had significant predictive validity with respect to risk management recommendations made by police, as well as with recidivism as indexed by subsequent contacts with police. Risk management mediated the association between risk assessment and recidivism: High levels of intervention were associated with decreased recidivism in high risk cases, but with increased recidivism in low risk cases. The findings support the potential utility of police-based risk assessment and management of IPV, and in particular the belief that appropriately structured risk assessment and management decisions can prevent violence.     

The criminal justice response to wife assault

This paper reviews the criminal justice response to wife assault. By establishing a set of conditional probabilites for the reporting, detection, prosecution, and conviction for wife assault, the paper establishes that a “winnowing process” occurs that is not dissimilar to that reported for other crimes. The probability of wife assault being detected by the criminal justice system is about 6.5%. Given that it is detected, the probability of arrest is about 21.2% [comparable to a 20% arrest rate for a composite of 121 crimes reported by Hood and Sparks (1970)]. Subsequent conditional probabilities for conviction and punishment generate an aggregate probability that, given that an event of wife assault occurs, the perpetrator has a 0.38% chance of being punished by the courts. The policy implications of this review are that the greatest impact on wife assault recidivism reduction would be generated by police arrest rates regardless of court outcome. At present, however, it is not known whether this effect is produced by specific deterrence or by the didactic function of law. It is concluded that too little is known of the subjective states of wife assaulters to ascertain whether deterrence or some other mechanism accounts for the decreased recidivism reported after arrest.     

Risk factors for recidivism among spousal assault and spousal homicide offenders

Abstract

The Spousal Assault Risk Assessment Guide (SARA; Kropp et al., 1995) is a 20-item checklist for the structured professional assessment of risk for partner violence. This study reported on a retrospective follow-up of file-based SARA assessments of offenders convicted 1988–1990 in Sweden. A total of 88 male batterers referred for court-ordered forensic psychiatric evaluations were included. During the 7-year follow-up, twenty-five (28%) were re-convicted of spousal assault. A few SARA items were statistically significantly associated with increased risk of recidivism, namely: Items #3, Past violation of conditional release or community supervision, #10 Personality disorder with anger, impulsivity, or behavioural instability (psychopathy), and #16 Extreme minimisation or denial of spousal assault history. The severity of the index crime (#18 Severe and/or sexual assault) was negatively related to risk for recidivism during follow-up. Offenders scoring above the median on the SARA were at more than 2.5 higher the risk for recidivism than those scoring below the median. In terms of predictive validity, the SARA actuarial score exhibited a marginal but statistically significant improvement over chance in this sample.     

The Risk Assessment Program and the Court of Penal Execution in the Province of Buenos Aires, Argentina

This study describes the working of a new system that supervises the execution of the punishment phase in the Province of Buenos Aires, Argentina; namely, the Court of Penal Execution, as well as the risk assessment program. It also reports the results of the violent recidivism baseline risk assessment carried out in candidates for conditional release. The cohort studied was recruited during 18 months in the Judicial Department of La Plata, taking into consideration psychopathological and criminological variables. The Psychopathy Checklist-Revised (PCL-R), the Historical, Clinical, and Risk Management--20 (HCR-20), and the Violence Risk Appraisal Guide (VRAG), among others, were used as measures. During the study period, 65 candidates for conditional release were part of the baseline assessment, and they constitute the population of this study. The measures' mean scores were PCL-R, 20.57; HCR-20, 18.58; VRAG, 12.17.     

Psychopathy and Recidivism Among Female Inmates

Despite a plethora of studies investigating psychopathy among male offenders, little is known about the applicability of this construct to female populations. Research has shown that prevalence rate, symptom presentation, and diagnostic comorbidity differ for females as compared to males. The current study is the first to examine the relationship between psychopathy and recidivism among women. Recidivism data on a sample of 78 female inmates were examined at a 1-year interval in relation to the Psychopathy Checklist-Revised (PCL-R), inclusion criteria for the Antisocial Personality Disorder Diagnosis from the Personality Disorder Examination (PDE), and selected scales from the Personality Assessment Inventory (PAI; Antisocial and Aggression scales). The egocentricity subscale of the PAI, Factor 1 of the PCL-R, and the verbal aggression subscale of the PAI were the best predictors of future recidivism. Specific differences emerged between male and female offenders when comparing the present data with previous studies of male psychopaths.     

Psychopathy (PCL-R) Predicts Violent Recidivism Among Criminal Offenders with Personality Disorders in Sweden

Psychopathy as conceptualized with Hare's Psychopathy Checklist Revised, PCL-R, has attracted much research during the 1990s. In the Scandinavian countries, few studies that empirically support the validity of North American risk assessment techniques in our regional context have been published. The purpose of this paper is to explore the predictive power of the PCL-R in a population of personality-disordered violent offenders subjected to forensic psychiatric evaluation in Sweden. Following release from prison (n = 172), discharge from forensic psychiatric treatment (n = 129), or probation (n = 51), a total of 352 individuals were followed for up to 8 years (mean = 3.7 years) with reconviction for violent crime as endpoint variable (base rate 34%). As the estimate of predictive power, the area under the curve of a receiver operating characteristic (AUC of ROC) analysis was calculated. For PCL-R scores to predict 2-year violent recidivism, AUC of ROC was .72 (95% CI: .66–.78). In addition, the personality dimension of psychopathy (Factor 1) and the behavioral component (Factor 2) both predicted 2-year recidivism significantly better than random: AUC of ROC .64 (95% CI: .57–.70) and .71 (95% CI: .65–.77), respectively. We conclude that psychopathy is probably as valid a predictor of violent recidivism in Swedish forensic settings as seen in previous North American studies.     

Psychopathy (PCL-R) as a Predictor of Violent Recidivism Among Criminal Offenders with Schizophrenia

Hare's Psychopathy Checklist-Revised (PCL-R) was used to test the hypothesis that psychopathy predicts violent recidivism in a cohort subjected to forensic psychiatric investigation and consisting of male violent offenders with schizophrenia (N = 202). Psychopathy was assessed with retrospective file-based ratings. Mean follow-up time after detainment was 51 months. Twenty-two percent of the offenders had a PCL-R score 26 (cutoff), and the base rate for violent recidivism (reconvictions) during follow-up was 21%. Survival analysis revealed that psychopathy was strongly associated to violent recidivism (log-rank = 17.71, df = 1, p < 0.0001). The area under the curve (AUC) of the receiver operating characteristics (ROC) of PCL-R total score to predict violent recidivism varied between different time frames from .64 to .75. Cox regression analyses revealed that other potential risk factors could not equally well or better explain violent recidivism in the cohort than psychopathy as measured by PCL-R.     

Police attitudes in assigning responsibility for wife abuse

Prevailing myths concerning wife abuse generally tend to blame the victim. The present study examines the degree of responsibility assigned by police officers to the wife, the husband, as well as the couple's socioeconomic situation in cases of wife abuse. The officer's position (neutral vs supportive) adopted toward the woman is also investigated. The influence of eight characteristics related either to the spouses or to the situation of violence on both the officer's position and the attribution of responsibility is analyzed. The characteristics are: couple's socioeconomic status, type of household, drinking by husband, alleged antagonism on the part of the wife, ambivalence of wife to press charges, type of abuse, history of assault, and violence toward children. The results are based on answers to case vignettes provided by 235 municipal police officers. The data were analyzed by performing a multivariate analysis of variance. Results indicate the type of abuse to be the characteristic with the greatest impact on police attitudes. While the husband is systematically considered more responsible than the wife, police officers always consider the wife somewhat responsible, particularly when there is alleged antagonism by the woman and when threats of violence are present.     

Testing the three- and four-factor models of the Korean version PCL-R as predictors of two antisocial outcomes – recidivism and risk

Psychopathy has consistently been associated with antisocial outcomes. The three- and four-factor models have been best fitted to data relating to Korean serious offenders (N?=?451), offering construct validity of the Korean Psychopathy Checklist-Revised (PCL-R). However, no study has yet tested the predictive power of the Korean PCL-R to explain two different types of antisocial outcomes: (1) risk of inmates measured by correctional officers during incarceration and (2) recidivism after release. By exploring these two forms of antisociality, here we sought to establish the predictive and construct validities of the Korean version of PCL-R. We found that the deviant lifestyle (Factor 3) performed best for predicting both antisocial forms (risk and recidivism) and that the deficits in interpersonal (Factor 1) and affective (Factor 2) abilities also uniquely predicted one subtype of risk, which suggests the three-factor model is better than the four-factor model in predictions. These findings will be useful for criminal justice experts and practitioners in Korea.     

Predicting Juvenile Recidivism:Using the Indiana Department of Correction's Risk Assessment Instrument

The Indiana Department of Correction uses a Risk Assessment Instrument (RAI) to determine juvenile offenders' risk of recidivism and their placement in an IDOC facility. Although it is used to make important decisions, the RAI has not been well-validated. This article describes a study of convicted male juvenile offenders to determine how well the IDOC instrument, which is similar to that used in several states, predicted recidivism. We found that the RAI did not predict recidivism in our sample.     

Predicting batterer recidivism five years after community intervention

The purpose of this study was to examine batterer recidivism rates 5 years after community intervention and to determine differences that discriminate between recidivists and nonrecidivists. Of the 100 men included in the sample, 40% were identified as recidivists because they were either convicted of domestic assault, the subject of an order for protection, or a police suspect for domestic assault. A discriminant analysis was conducted using a variety of background and intervention variables. Five variables were selected that significantly discriminated between recidivists and nonrecidivists and correctly classified 60.6% of the cases. Men who had been abusive for a shorter duration prior to the program, court ordered to have a chemical dependency evaluation, in chemical dependency treatment, abused as children, and previously convicted for nonassault crimes were more likely to be recidivists. Variables relating to intervention did not significantly predict recidivism. Implications for community intervention programs are discussed.     

Predicting Institutional Adjustment and Recidivism with the Psychopathy Checklist Factor Scores: A Meta-Analysis

This study explored the validity of the PCL/PCL-R factor scores in predicting institutional adjustment and recidivism in forensic clients and prison inmates. Forty-two studies in which institutional adjustment, release outcome (recidivism), or both were assessed prospectively with the PCL/PCL-R yielded 50 effect size estimates between the PCL/PCL-R factor scores and measures of institutional adjustment/recidivism. A meta-analysis of these findings disclosed that Factor 2 (Antisocial/Unstable Lifestyle) correlated moderately well with institutional adjustment and recidivism, whereas Factor 1 (Affective/Interpersonal Traits) was less robustly associated with these outcomes. Direct comparisons of the mean effect sizes attained by Factors 1 and 2 revealed that Factor 2 was significantly more predictive of total outcomes, general recidivism, violent recidivism, and outcomes from the 12 most methodological sound studies than Factor 1. There was less differentiation between Factors 1 and 2 on measures of institutional adjustment.     

THE MEANING OF ARREST FOR WIFE ASSAULT*

The social meaning of wife assault has changed in recent years for both citizens and formal social control agents. Research on deterrence has been partly responsible for modifying police responses to domestic violence. Police are increasingly adopting pro-arrest policies for wife assault, but little is known about perceptions held by assaulters concerning the consequences of arrest for their life circumstances. Using national survey data from samples of both assaultive and nonassaultive men, the following questions are addressed: What costs do men perceive as most likely to occur if they are arrested for wife assault? Does the perceived likelihood of these costs contribute to their overall fear (i.e., perceived severity) of arrest? To what extent is the perceived likelihood of these costs related to involvement in wife assault? Perceived costs include both direct consequences seen to result from arrest and any indirect costs for the person. Indirect consequences include stigmatic costs (e.g., familial or personal humiliation), attachment costs (e.g., damage to interpersonal relationships) or commitment costs (e.g. jeopardized investments or foreclosed opportunities). The implications of the findings for an expanded version of the deterrence doctrine are discussed.     

When is an assault not an assault? The Canadian public's attitudes towards wife and stranger assault

There is no distinction in the Canadian Criminal Code between assaults committed on strangers and assaults of spouses. Traditionally, however, wife assault has been considered a private affair, and it has been argued that this attitude continues to be reflected in the police response to domestic disputes. In this study of 240 adults'written responses to assault scenarios, the victim-offender relationship produced variations in ratings of seriousness, in the relative amounts of blame attributed to the victim and offender, and in the recommended police response. In general, nondomestic recidivist offenders tended to be treated most harshly by the subjects, as did offenders who assaulted their victims inside a private house. Additionally, female subjects gave higher ratings of seriousness of the offense, and considered future violence by the offender to be more likely, than did male subjects. The findings are explained in terms of the subjects' ability to identify with the offender, although the decision-making process in judging cases of assault is evidently complex. The implications of these findings for police officers with discretionary powers are considered.

     

Validation of Spouse Violence Risk Assessment Inventory for Police Purposes

The Spouse Violence Risk Assessment Inventory (SVRA-I) is a new scale developed by the Israeli police to measure the likelihood of male perpetrators repeating violent behavior toward their partners. This article describes the objectives of the inventory, its distinction in comparison to existing inventories, and the process of its development. Our preliminary study demonstrated a reasonable level of inter-rater reliability. Two subsequent studies examined the relationships between clinical assessments and SVRA-I scores, and additional two studies tested the validity of the inventory against criteria of repeated partner abuse. The inventory was found to be a valid and efficient tool for predicting recidivism of intimate partner violence (IPV). In the discussion, we review the limitations of devices aimed at predicting repeated IPV.     

Violent recidivism in criminal psychopaths

The predictive validity of the Psychopathy Checklist-Revised (PCL-R) was compared with 3 actuarial risk scales in a sample of 81 offenders followed for a maximum of 67 months (average of 30 months). The recommittal or general recidivism rate for the entire sample was 57% (40% for nonpsychopaths, 51.2% for a mixed group, and 85% for psychopaths). The violent reoffense rate was 10% for the sample (nonpsychopaths 0%, mixed 7.3%, psychopaths 25%). All instruments were significantly correlated with general recidivism; however, the PCL-R was the best predictor of violent recidivism. Compared to the actuarial scales, the PCL-R had a higher predictive efficiency (Relative Improvement Over Chance (RIOC)) and yielded fewer decision errors. Most importantly, Factor 1 was a better predictor of violent recidivism than Factor 2, suggesting that the trait construct of psychopathy makes a unique contribution to the prediction of violent recidivism.     

Incarceration and Recidivism among Sexual Offenders

The relationship between incarceration and recidivism was investigated in a sample of 627 adult male sexual offenders. Incarceration for the index offense was unrelated to sexual or violent recidivism. This was the case whether incarceration was examined as a dichotomous variable (incarceration vs. community sentence) or as a continuous variable (length of incarceration). Risk for sexual recidivism was assessed with a modified version of the Rapid Risk Assessment for Sexual Offense Recidivism. There was no evidence that the relationship between incarceration and recidivism was confounded or moderated by risk or that length of incarceration and recidivism were non-linearly associated. Sentencing sexual offenders to terms of incarceration appears to have little, if any, impact on sexual and violent recidivism following release.