An Innovative Use of Conjoint Analysis to Understand Decision-Making by Juvenile Probation Officers

IntroductionMost youth involved in the justice system are sentenced to probation and supervised by a juvenile probation officer (JPO; Sickmund & Puzzanchera, 2014). In their supervisory role, JPOs make decisions that have immediate and lasting implications for youth probationers (Griffin & Torbet, 2002; Leifker & Sample, 2011). JPOs provide input and recommendations related to several aspects of probationer status and care, including whether probationers should be returned to the community, the terms of probation supervision, and which social services probationers should be referred to. Though juvenile court judges are ultimately responsible for many of these decisions, research (Leifker & Sample, 2011) has shown that judicial decisions align with the recommendations of JPOs in the vast majority of cases. JPOs have also been described as gateway providers to behavioral health services for youth offenders, helping identify mental health treatment needs of adolescents (Wasserman et al., 2008) and facilitating youth engagement in behavioral healthcare (Holloway, Brown, Suman, & Aalsma, 2013). Despite the range of decisions to be made by JPOs, and the discretion afforded them, questions remain about how JPOs reach their decisions. Indeed, juvenile probation has been referred to as one of the "black boxes" in justice system decision-making research (Bechtold, Monahan, Wakefield, & Cauffman, 2015, p. 325).Decisions within the context of the justice system are, ostensibly, to be made in consideration of relevant legal factors, such as the extent of an offender's criminal history or the severity of a charged offense (Schwalbe, Hatcher, & Maschi, 2009). The juvenile justice system, with its additional mandate to protect the best interests of youth offenders, allows courts to also consider extralegal factors, which are less directly tied to the factual details of a charged offense. For example, judges may consider a youth's amount of parental support when making determinations about youth culpability and sentencing. In other words, judges make a risk versus needs calculation when considering juvenile offenses (Vincent, Paiva-Salisbury, Cook, Guy, & Perrault, 2012). To this end, decision-makers within the system can be aided by formal risk assessment measures or detailed legal guidelines to weigh complex and potentially conflicting information about an individual youth offender. Some jurisdictions have implemented sentencing rubrics, for example, with the goal of consistently administering punishments proportionate to the offenses committed (Vincent & Lovins, 2015). In the pre-sentencing stage of the juvenile justice system, decision-makers use validated risk and needs assessments to determine an offender's need for both supervision and services (Grisso, 2007; Vincent et al., 2012). Guidelines, however, vary widely by jurisdiction, and their use is often voluntary. Despite the availability of these determinate processes to increase fairness within the justice system, decision-makers can choose to override prescribed outcomes, diminishing the purpose of guidelines (Wang, Mears, Spohn, & Dario, 2013). This suggests a continued need to study how legal and extralegal factors differentially influence decision-making within the juvenile justice system.A wide array of interrelated variables have been implicated in decision-making at different stages of the adult and juvenile justice systems, potentially contributing to disparate outcomes among offenders. Past studies have identified many influences on court personnel, including the demographic characteristics of individual offenders (Leiber & Fox, 2005); the severity of the criminal charge (Leiber & Peck, 2015); the mental health of the offender (Cappon & Vander Laenen, 2013); the offender's family structure or involvement in the legal process (Rodriguez, Smith, & Zatz, 2009); and each decision-maker's own characteristics, professional orientation, and personal biases (Ricks & Eno Louden, 2015). …