Predictors Of Recidivism Research Articles

The prediction of criminal recidivism in male juvenile delinquents

Previous studies have demonstrated a strong association of criminal behavior of juvenile delinquents with delinquents? personality traits and family interactions. However, little is known about the extent to which family interactions and personality traits are associated with criminal recidivism. The present study aimed to examine these relationships, using the Velikih pet plus dva (Big Five Plus Two) ? short version (assessing Neuroticism, Extraversion, Openness to experience, Conscientiousness, Aggressiveness, Positive Valence, Negative Valence), The Quality of Family Interaction Scale (Satisfaction with family, Mother and Father Acceptance/Rejection), and official data from criminal records. The study included 61 institutionalized delinquents and 64 non-delinquents, 15 to 18 years of age. Neuroticism, Openness to experience, Conscientiousness, Negative valence, acceptance by father and rejection by mother are statistically significant predictors of criminal recidivism in juvenile delinquents. Delinquents showed higher Neuroticism, lower Conscientiousness and acceptance by mother compared to non-delinquents.

Beyond Mental Health Disorders: Risk Assessment by Using the 4th Generation of Forensic Tools

The causal relationship between mental health disorders and (violent) recidivism is weak. Beyond mental health disorders, criminogenic factors contribute to the prediction of recidivism. In order to be effective, interventions need to be offense-oriented, i.e. focused on factors that are directly associated to the offense mechanism. To treat mental health disorders while ignoring criminogenic needs is highly ineffective. A risk/needs orientation of the criminal justice system must be emphasized. Therefore, psychiatric treatment will not be sufficient to prevent future criminal behavior. The main approach should be based on explaining the mechanism of the offense to understand the underlying offense dynamic and develop on that basis a well-fitting intervention strategy. FOTRES (Forensic Operationalized Therapy/Risk Evaluation System) is a structured professional judgment (SPJ) tool of the 4th. generation designed to assess recidivism risk and to monitor treatment progress and intervention quality. FOTRES now is already implemented within the Risk-Oriented Enforcement of Sentence (ROES) project in Switzerland. The cost-benefit analysis with the ROES is positive. The tool is widley used also in Germany and Austria.

Exploring drug-related crimes with social network analysis

In recent years, drug abuse and drug addiction have become a major burden to the society. In order to achieve the public expectation for drug crime prevention, law enforcement agencies devote considerable resources hoping to strengthen the intensity of interventions. However, with the rapid changes in social patterns, the drug criminals also look for ways to avoid law enforcement investigations by producing, transporting and selling drugs through different regions, making drug prevention more difficult. Thus, developing dominant strategies to deal with this issue is a main task for police agencies. For more effectively analyze the structural influences of drug crime, we utilize social network analysis (SNA) techniques to discover implications of drug related crime networks. The macro-level perspective of co-offender network indicates that criminals intend to set blocks between network members to prevent law enforcement interventions. The micro-level perspective of individuals provides significant social features to predict drug recidivism. The experimental results indicate superior performance when adopting both personal and social features in classification task. Applying SNA to recidivism prediction is a leading endeavor, and the approach presented in this paper offers remarkable improvement on traditional methods. The results of this paper reveals the advantages of structural implications in analyzing drug related crime, as well as its ability to facilitate the cognition of crime prevention and intervention strategies.

비행촉발요인조사서를 통한 소년범 재범 예측 연구 - 생존분석을 중심으로 -

본 연구는 비행촉발요인조사서를 바탕으로 소년범의 재범 여부를 장기 추적하여 어 떠한 위험요인이 소년범의 재범을 예측하는가를 분석하며, 비행촉발요인조사서의 예측 타당성을 규명함에 목적이 있다. 위험요인은 비행촉발요인조사서의 하위 척도인 가족 의 구조, 가족의 기능적 역할, 학교생활, 가출경험, 비행전력 및 환경, 개인적 위험요인 으로 구성하였다. 또한 예측타당성 평가를 위해 소년범 39명을 대상으로 평균 약 13년 간의 재범 여부를 추적하였다. 결과를 도출하기 위해 Cox 회귀분석을 실시하였다. 연구 결과, 가족의 구조와 가족의 기능적 역할이 소년범의 재범을 예측하는 설명력 있는 위 험요인으로 밝혀졌다. 재범위험성에 따른 생존분석 결과, 재범위험성이 높은 집단일수 록 누적 생존확률이 가장 빠르게 하강하였으며, 가장 마지막 시점에서의 생존확률을 보 았을 때 재범위험성이 낮을수록 더 많이 생존함이 발견되어 비행촉발요인조사서가 재 범을 예측함에 있어 타당함이 나타났다. 본 결과는 결손가정의 청소년이 범죄의 위험성에 더 많이 노출되어있다는 선행연구를 지지하였으며, 비행촉발요인조사서 평가시 방 임 여부에 대해 주의를 기울여야함을 시사한다. 본 연구는 10년 이상의 장기적 관찰기 간을 두고 소년범의 재범을 추적하여 유의한 재범 위험요인과 비행촉발요인조사서의 예측력을 확인하였다는 점에서 의의가 있다. 또한 이와 같은 논의를 토대로 후속 연구 제안과 소년범 재범 예방에 관한 정책적 제안을 제시하였다.

The Role of Behavioral Phenotypes on Impaired Driving Recidivism Risk and Treatment Response to Brief Intervention: A Preliminary Study.

Heterogeneity in the driving while impaired (DWI) offender population and modest outcomes from remedial programs are fueling interest in clarifying clinically significant DWI subtypes to better assess recidivism risk and target interventions. Our previous research identified 2 putative behavior phenotypes of DWI offenders with distinct behavioral, personality, cognitive, and neurobiological profiles: (i) offenders primarily engaging in DWI (pDWI); and (ii) offenders engaging in DWI and other traffic violations (MIXED). Here, we evaluate these phenotypes' clinical significance for prediction of recidivism and intervention targeting. DWI recidivists participating in a previous randomized controlled trial (N=184 comparing brief motivational interviewing (BMI) and an information and advice control condition (IA) were retrospectively classified as either pDWI (n=97) or MIXED (n=87). Secondary analyses then evaluated the effect of this phenotypic classification on self-reported 6- and 12-month alcohol misuse outcomes and documented 5-year DWI recidivism violations, and in response to either BMI or IA (i.e., pDWI-BMI, n=46; MIXED-BMI, n=45; pDWI-IA, n=51; MIXED-IA, n=42). Two hypotheses were tested: (i) MIXED classification is associated with poorer alcohol misuse outcomes and recidivism outcomes than pDWI classification; and (ii) pDWI paired with BMI is associated with better outcomes compared to MIXED paired with BMI. MIXED classification was associated with significantly greater risk of recidivism over the 5-year follow-up compared to pDWI classification. Moreover, the pDWI-BMI pairing was associated with significantly decreased recidivism risk compared to the MIXED-BMI pairing. Analyses of 6- and 12-month alcohol use outcomes produced null findings. The clinical significance of phenotypic classification for risk assessment and targeting intervention was partially supported with respect to recidivism risk. Prospective investigation of this and other behavioral phenotypes is indicated.

Predicting Offenders' Institutional Misconduct and Recidivism: The Utility of Behavioral Ratings by Prison Officers.

Measures of current behavior are rarely incorporated into risk assessment. Therefore, the current study used a behavior rating scale to assess prison officers' observations of inmates prison behavior and examined the contribution of these ratings for risk assessment. Prison officers rated 272 sexual and violent offenders in three different correctional treatment facilities in Berlin, Germany. Factor analysis revealed three psychologically meaningful factors measuring externalizing, internalizing and adaptive prison behavior. The construct validity of the three factors was established through correlational analyses with standardized risk assessment instruments. Externalizing and internalizing behaviors were significant predictors of violent recidivism after release. In addition, externalizing was a significant predictor of institutional misconduct, whereas adaptive and internalizing behavior predicted whether an inmate was granted privileges (e.g., minimum-security confinement). Logistic regression analyses indicated that externalizing behavior ratings added incrementally to the Level of Service Inventory-Revised for the prediction of institutional misconduct and violent recidivism. The results indicate that prison officers observe important prison behaviors and that behavioral ratings can improve risk assessment.

The Relationship Between Trauma, Recidivism Risk, and Reoffending in Male and Female Juvenile Offenders

Elevated rates of traumatic experience in the juvenile justice population are well established. Nevertheless, the role of trauma and its application to rehabilitation and recidivism in a criminal justice context remains hotly debated, particularly for female youth. The Risk-Need-Responsivity framework, the predominant model for risk assessment and case management in juvenile justice, does not consider trauma to be a risk factor for offending. This study examined- Posttraumatic Stress symptomology, maltreatment history, and childhood adversity - in relation to RNR risk factors for reoffending (criminogenic needs) and recidivism in a sample of female and male juvenile offenders. Rates of PTS symptomology, maltreatment, and childhood adversity were significantly higher in this sample compared to prevalence in the general population. Females were more likely to have experienced maltreatment. Several maltreatment and childhood adversity types were significantly related to criminogenic needs. PTS symptomology and adversity were not significant predictors of recidivism when entered alongside criminogenic needs; however, maltreatment was the strongest predictor of recidivism for both male and female youth in a model that included criminogenic needs. Gender did not moderate the relationship between maltreatment and recidivism. The importance of considering youths' maltreatment history in their rehabilitative care is discussed.

Assessing the Incremental Validity of Andrews and Bonta’s “Moderate Four” Predictors of Recidivism Using a Diverse Sample of Offending and Truant Youth

Bonta and Andrews suggest that there are eight factors that predict recidivism among offenders, which they call the "Central Eight." They split up the Central Eight into the "Big Four" and the "Moderate Four," with the Big Four hypothesized to be more strongly associated with recidivism than the Moderate Four. The purpose of this study was to assess the incremental validity of the Moderate Four relative to the Big Four as they are measured on the Youth Level of Service/Case Management Inventory (YLS/CMI). The sample comprised 2,436 youth between the ages of 9 and 18 ( M = 14.34) in the formal probation and truancy division from a juvenile county court in the Midwest. Recidivism was measured as a new petition filed against a youth in court within 2 years of a youth's initial offense. Results indicated that the Moderate Four predictors provided no incremental validity beyond the Big Four but did not reduce the predictive validity of the YLS/CMI.

Can nonexperts really emulate statistical learning methods? A comment on “The accuracy, fairness, and limits of predicting recidivism”

Recent research has questioned the value of statistical learning methods for producing accurate predictions in the criminal justice context. Using results from respondents on Amazon Mechanical Turk (MTurkers) who were asked to predict recidivism, Dressel and Farid (2018) argue that nonexperts can achieve predictive accuracy and fairness on par with algorithmic approaches that employ statistical learning models. Analyzing the same data from the original study, this comment employs additional techniques and compares the quality of the predicted probabilities output from statistical learning procedures versus the MTurkers’ evaluations. The metrics presented indicate that statistical approaches do, in fact, outperform the nonexperts in important ways. Based on these new analyses, it is difficult to accept the conclusion presented in Dressel and Farid (2018) that their results “cast significant doubt on the entire effort of algorithmic recidivism prediction.”

Testing the validity of criminal risk assessment tools in sexually abusive youth.

Although accurate risk appraisals are mandatory to provide effective treatment to juveniles who have sexually offended (JSOs), the current knowledge on the validity of risk assessment instruments for JSOs is inconclusive. We compared the predictive validities of the Juvenile Sex Offender Assessment Protocol II (J-SOAP II), the Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR), and the Violence Risk Appraisal Guide-Revised (VRAG-R) scores concerning sexual, nonsexual-violent, and general criminal recidivism (based on both official and nonregistered reoffenses) in a consecutive sample of 597 male JSOs (Mage = 14.47 years, SDage = 1.57 years) while accounting for different recidivism periods, offense severities, and cumulative burden of adverse childhood experiences (ACEs). Receiver Operator Characteristic (ROC) curves and Cox regression analyses indicated that the tools allowed valid predictions of recidivism according to their intended purposes: The ERASOR was best suited to predict sexual recidivism within 0.5 and 3 years, the J-SOAP II was valid for predictions of sexual and nonsexual-violent recidivism within these recidivism periods, and the VRAG-R showed potential strengths in predicting nonsexual-violent recidivism, especially when committed above age 18. Elevated offense severity and burden of ACEs impeded predictive accuracies of the J-SOAP II and the VRAG-R, particularly in case of sexual recidivism. Our findings emphasize that risk assessment for JSOs must not rely solely on scores derived from risk assessment instruments, but a comprehensive consideration of a JSOs offense severity and psychosocial adversities is additionally necessary to approach accurate risk appraisals. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Assessing the Relationship Between Religiosity and Recidivism Among Adult Probationers in Pakistan.

Although empirical evidence supports a relationship between religiosity and criminal behavior, debate continues about the theoretical mechanisms by which they are related. Moreover, the topic has been largely ignored by practicing clinicians and correctional workers. The Muslim Religiosity-Personality Inventory: Abridged was administered to low-risk Pakistani probationers and factor analyzed, after which probationers' recidivism was monitored. Five oblique factors were obtained, three of which were correlated with recidivism (Religious Practice, Religious-Moral Values, and Fundamental Religious Beliefs), as was the full scale, while two were not (Importance of Religion and Rejection of Nonbeliever). In a logistic regression, Religious-Moral Values and Religious Practices contributed to the prediction of probationer recidivism. However, when demographic characteristics were introduced, education and marital status replaced Religious Practices. This study supports the religiosity-crime link in a non-Western, Muslim culture. Implications for assessing religiosity and for practitioners in the justice system are discussed.

Cost–utility analysis of low-intensity case management to increase contact with health services among ex-prisoners in Australia

ObjectivesThe economic burden of incarceration is substantial in Australia. People released from prison are at high risk of poor health and this is an important predictor of recidivism. The ‘Passports...

Recidivism and Reentry: The Role of Individual Coping Styles

The high rate of recidivism among those reentering the community post incarceration can be partially attributed to various barriers to reentry, including unemployment, financial difficulties, and mental illness. Learning to cope with these barriers is demanding and stressful for many people. The current article examines whether individual coping styles could act as a significant predictor of recidivism. Utilizing a cross-sectional data set collected from previously incarcerated individuals, this study examines the relationship between individual coping styles and arrest or reincarceration. Results from this study indicate a relationship between coping and arrest. Implications from this study are noteworthy given that our understanding of coping can increase the ability of the correctional staff to reduce later recidivism in the community.

A Coverage-Based Utility Model for Identifying Unknown Unknowns

A classifier’s low confidence in prediction is often indicative of whether its prediction will be wrong; in this case, inputs are called known unknowns. In contrast, unknown unknowns (UUs) are inputs on which a classifier makes a high confidence mistake. Identifying UUs is especially important in safety-critical domains like medicine (diagnosis) and law (recidivism prediction). Previous work by Lakkaraju et al. (2017) on identifying unknown unknowns assumes that the utility of each revealed UU is independent of the others, rather than considering the set holistically. While this assumption yields an efficient discovery algorithm, we argue that it produces an incomplete understanding of the classifier’s limitations. In response, this paper proposes a new class of utility models that rewards how well the discovered UUs cover (or "explain") a sample distribution of expected queries. Although choosing an optimal cover is intractable, even if the UUs were known, our utility model is monotone submodular, affording a greedy discovery strategy. Experimental results on four datasets show that our method outperforms bandit-based approaches and achieves within 60.9% utility of an omniscient, tractable upper bound.

Data Mining Techniques for Recidivism Prediction: A Survey Paper

Data Mining Techniques for Recidivism Prediction: A Survey Paper

Dynamic Risk Assessment of Sexual Offenders: Validity and Dimensional Structure of the Stable-2007

In this study, the predictive and incremental validity of the Stable-2007 beyond the Static-99 was evaluated in an updated sample of N = 638 adult male sexual offenders followed-up for an average of M = 8.2 years. Data were collected at the Federal Evaluation Center for Violent and Sexual Offenders (FECVSO) in Austria within a prospective-longitudinal research design. Scores and risk categories of the Static-99 (AUC = .721; p < .001) and of the Stable-2007 (AUC = .623, p = .005) were found to be significantly related to sexual recidivism. The Stable-2007 risk categories contributed incrementally to the prediction of sexual recidivism beyond the Static-99. Analyzing the dimensional structure of the Stable-2007 yielded three factors, named Antisociality, Sexual Deviance, and Hypersexuality. Antisociality and Sexual Deviance were significant predictors for sexual recidivism. Sexual Deviance was negatively associated with non-sexual violent recidivism. Comparisons with latent dimensions of other risk assessment instruments are made and implications for applied risk assessment are discussed.

Does Treatment Quality Matter?

Research SummaryThe Florida Department of Juvenile Justice has implemented a process to evaluate the treatment quality of interventions provided statewide in all long‐term residential programs. In the current study, we examine the predictive validity of this treatment quality component in the prediction of recidivism of youth (N = 2,397) who completed juvenile justice residential programs from July 1, 2013 to June 30, 2014. We use hierarchical linear modeling to assess the effects of treatment quality (as youth are nested within programs), controlling for demographic and criminal history factors. The results indicate that higher average treatment quality scores of interventions received within a residential program decreased the likelihood of subsequent arrest, conviction, and reincarceration, whereas the highest treatment quality score of any specific intervention provided within the program decreased the odds of reincarceration only.Policy ImplicationsIn this study, we confirm the importance of fidelity and implementation quality in the provision of crime prevention treatment interventions to serious, deep‐end juvenile offenders. Additionally, our results shed light on the ability of a juvenile justice agency to measure treatment quality in a substantively meaningful way with the use of limited additional resources. The services provided to keep communities safe, prevent future offending, and rehabilitate juvenile offenders must be held accountable for producing such outcomes, and one method of measuring such compliance is to evaluate the quality of the interventions with respect to staff training, fidelity adherence, evaluation, and corrective action processes.

Age of gray matters: Neuroprediction of recidivism.

Age is one of the best predictors of antisocial behavior. Risk models of recidivism often combine chronological age with demographic, social and psychological features to aid in judicial decision-making. Here we use independent component analyses (ICA) and machine learning techniques to demonstrate the utility of using brain-based measures of cerebral aging to predict recidivism. First, we developed a brain-age model that predicts chronological age based on structural MRI data from incarcerated males (n = 1332). We then test the model's ability to predict recidivism in a new sample of offenders with longitudinal outcome data (n = 93). Consistent with hypotheses, inclusion of brain-age measures of the inferior frontal cortex and anterior-medial temporal lobes (i.e., amygdala) improved prediction models when compared with models using chronological age; and models that combined psychological, behavioral, and neuroimaging measures provided the most robust prediction of recidivism. These results verify the utility of brain measures in predicting future behavior, and suggest that brain-based data may more precisely account for important variation when compared with traditional proxy measures such as chronological age. This work also identifies new brain systems that contribute to recidivism which has clinical implications for treatment development.

Criminal characteristics of a group of primary criminals diagnosed with aspd: approach to criminal recidivis.

Antisocial personality disorder (ASPD) is commonly associated with the risk of criminal recidivism. Knowing more about the factors associated with this pattern of behaviour can help with the design of effective prevention strategies. The purpose of this article is to establish if there are differences in socio-criminogenic variables of a group of criminals sentenced for the first time and with APSD compared to another group of first-time offenders who do not present this disorder. Analytical observation study of 70 men classified into 2 groups according to the presence of ASPD TPA (n=47; age: 29.98±7.8 years) or absence of ASPD (n=23; age: 32.35±8.7 years). The inmates with ASPD showed higher frequencies of current consumption of psychoactive substances (31.9%), criminal associations and simultaneous use of psychoactive substances (70.2%), having committed the crime under the effects of a psychoactive substance (55.3%), not having the possibility of distancing themselves from criminal associations (83%) and a lack of legal resources for proceedings for defence and release (76.6%). This sample contains a group of variables called dynamic that are more commonly present amongst first time offenders with ASPD; said variables have been associated as major predictors of recidivism. Given that they are regarded as dynamic, they may well be modifiable.

Adult Recidivism in United States: A Meta-Analysis 1994–2015

In 2014, adult correctional systems supervised an estimated 6.8 million individuals in the United States with 1 in 36 adults (or 2.8%) being under some form of correctional supervision. Unfortunately, not only are the number of individuals connected to the correctional system and the outlined disparities based on minority status worrisome, there is also the persistent concern of repeat offending. Given the fact that the most recent comprehensive meta-analysis examining predictors of adult offender was published in 1996, a current systematic review and meta-analysis focusing on United States samples of all types of re-offense is necessary for identifying current predictors of adult recidivism with U.S. studies from 1994 through 2015. Specifically, the questions addressed in this meta-analysis include (a) which attributes predict general, sexual, and violent recidivism for adults in the American justice system, and (b) are some characteristics more influential than others? We determined the following domains are statistically significant predictors of recidivism: age (r = .02), antisocial personality scales (r = .13), criminogenic needs (r = .10), distress (r = .06), family criminality (r = .18), family rearing (r = .16), gender (r = .19), history of antisocial behavior (r = .12), risk scales (r = .17), social achievement (r = .05), and substance abuse (r = .07). Implications are provided.