Predicting offending within the New Zealand youth justice system : evaluating measures of risk, need, and psychopathy : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Clinical Psychology, Massey University, Wellington, New Zealand
Youth offending in New Zealand is an often touted problem. The reality is that, although many young people break the law as part of normal adolescent behaviour, the number of youth committing antisocial acts has decreased over the past decade. There is however one exception. The number of young people who exhibit patterns of persistent, chronic and violent offending behaviour is increasing. Recent theoretical approaches have attempted to conceptualise these complex young people by considering the numerous interacting causal factors associated with their offending. These models can inform appropriate assessment, treatment, and prevention strategies. To date, social learning models incorporating risk and need factors have been the best supported. However, new developmental approaches have also been applied, including the downward extension of psychopathy: an adult personality disorder associated with recidivistic offending and treatment non-compliance. Based on these theories, promising new actuarial risk assessment measures have been developed. These measures are being increasingly employed by youth justice systems internationally as a means of identifying and case managing persistent and serious offenders. However, these measures are not widely used in New Zealand, and virtually none have been empirically examined with New Zealand youth. This gap in evidence-based practice is perplexing given the international recognition and respect afforded to New Zealand’s youth justice system. The current study therefore sets a number of objectives. Firstly, it aimed to identify a profile of youth offenders across the New Zealand youth justice system by providing data on demographics, offending behaviours, education/ employment status, and mental health using the Massachusetts Youth Screening Inventory - 2 (MAYSI-2). Secondly, the study evaluated
the predictive validity of three assessment measures of youth offending. These measures were the Youth Level of Service/ Case Management Inventory (YLS/CMI), the Youth Psychopathic Traits Inventory (YPI), and the Inventory of Callous/ Unemotional Traits (ICU). Finally, the study explored the utility of these measures within a restorative justice system whereby limited resources could be matched to those most at risk of re-offending.
Using a prospective study design, two samples aged between 14 and 17 were selected. These samples represented youth offenders at two opposing ends of the youth justice system. The “Diversion” sample initially consisted of 70 youth offenders whose matters had been diverted by Police Youth Aid Officers in Counties-Manukau. All measures were administered during a 90-minute initial assessment phase. After 6-months, 63 (90%) were followed up to complete a self-report measure of offending behaviour committed since the first assessment. The “Clinical” sample initially consisted of 59 youth offenders who had been referred for a psychological assessment by a Youth Court within the Auckland region. The YLS/CMI was part of the assessment process. A total of 44 (75%) of the clinical participants were followed up after six-months. All measures, including the self-reported offending measure, were administered to this cohort.
Male gender, Māori ethnicity, and previous police contact were overrepresented within both samples. Approximately 40% of participants from both samples were either not attending school or were unemployed. Theft and dishonesty index offences were the most prevalent for both samples, however nearly 60% of the clinical sample was charged with a violent offence. The MAYSI-2 mental health screen revealed that approximately half of both samples scored on the Caution range or above for Alcohol/ Drug Use, while over 30% of both samples reported difficulties with Anger and Irritability. Over 60% of the Clinical sample received a formal clinical diagnosis, with conduct disorder and substance use disorders being the most prevalent. The two samples were merged to describe the results of the assessment measures. The YLS/CMI total produced fair internal consistency (α = .79). Total scores from the Clinical sample were significantly higher than the Diversion sample. Internal consistency was excellent for the YPI (α = .92) and fair for the ICU (α = .77). There were no significant differences in scoring between the two samples on these psychopathy measures. All three risk measures correlated with each other; while the re-test reliability of the YLS/CMI was significant (.79). Māori ethnicity was associated with higher total scores on the YLS/CMI and the YPI. Māori youth were also more likely to come into police contact during the six-month follow-up period. Medium to large associations were found between the three risk assessment measures and the seriousness of self-reported offences, contact with police, and contact with the youth court. Binary logistic regression, multiple regression, and Receiver Operator Curve (ROC) analyses confirmed the overall predictive validity of three measures, however the YLS/CMI total score was superior to the psychopathy screening measures across all analyses. Finally, results show that many participants who scored highly on the YLS/CMI received a higher level of intervention service during the follow-up period. However, a similar number of high risk youth received little or no services.
It was concluded that the YLS/CMI, the YPI, and the ICU have a high level of predictive validity over a short time frame. These findings have direct implications for assessment, prevention, and intervention practices. However, it is argued that new assessment measures relevant to both restorative justice practices and New Zealand’s youth offenders be developed that compensate for the limitations of these generic international measures. Overall, this research has been successful in adding to the accumulating literature on youth offender risk assessment, as well as the conceptualisation of psychopathic traits within youth.