Police decision-making with young offenders: Examining barriers to the use of diversion options

Introduction

Numerous studies demonstrate that many young people in the justice system have multiple and complex needs (Australian Institute of Health and Welfare [AIHW], 2019; Farrington, Gaffney, & Tfofi, 2017). For example, in Australia, of the 226 young people who were in detention or on remand in Victoria in 2017: 70% reported that they were victims of abuse, trauma or neglect; 65% reported that they had been previously expelled or suspended from school and 82% reportedly offended when under the influence of alcohol, drugs or both (State of Victoria, 2018). Children in out-of-home care are over-represented in the justice system (Gerard, McGrath, Colvin, & McFarlane, 2019) as are young people who have parents who are involved in the justice system (Derzon, 2010; Murray, Farrington, & Sekol, 2012). Young offenders often co-present with other complex issues such as mental distress (McClelland, Elkington, Teplin, & Abram, 2004; Fougere, Thomas, & Daffern, 2013) and acquired brain injury (Parsonage, 2016). A younger age of first involvement in the justice system is associated with poorer outcomes including a higher likelihood of continued offending into adulthood (Chen, Matruglio, Weatherburn, & Hua, 2005; Loeber & Farrington, 2011). Aboriginal and Torres Strait Islander young people (or ‘first Australians’) are 17 times more likely to spend time under supervision than non-first Australians (AIHW, 2017) and young men are over-represented in the justice system – as is the case internationally.

Recidivism rates are persistently high among young people who receive punitive justice interventions. However, 82% of young Australians who were released from supervised detention in 2016--2017 returned within 12 months and 59% returned within 6 months (AIHW, 2018). Outcomes for young people with involvement in the criminal justice system include poor psychological well-being (McGrath & Weatherburn, 2012), low involvement in training or employment and problematic substance use following incarceration (Welty et al., 2017). Justice system involvement is associated with increased likelihood of future offending (Petitclerc, Gatti, Vitaro, & Tremblay, 2013) and contact with the system can disrupt what is typically a ‘natural’ transition away from offending (McAra & McVie, 2011). Thus, incarceration seems to compound offending, rather than providing alternative trajectories.

Consistent with the research, policy and policing guidelines in Australia highlight that youth offending should be foremost treated as a welfare issue (Australian Government, 2010). For many years, there has been a strong emphasis on diversion away from the criminal justice system for low-level offences through the use of caution and warning programmes, formal diversion programmes (e.g. education programmes, restorative justice conferencing (RJC)) and the use of bail and bail support programmes (Australian Law Reform Commission, 1997). Although evaluations investigating individual programmes have had mixed findings, meta-analyses have shown that young people who have been diverted are less likely to offend than those who receive punitive interventions alone (Schwalbe, Gearing, MacKenzie, Brewer, & Ibrahim, 2012; H. A. Wilson & Hoge, 2013).

However, while the principle of diversion is well-accepted and embedded at pre-court, post-sentence and post-conviction levels of the justice system, there are many complex factors that are known to prolong young people’s involvement in the system and seemingly work against its purported welfare focus. These include delays in sentencing, the use of remand, poor communication between the justice and child protection systems and systemic complexity (Armytage & Ogloff, 2017). Moreover, while diversion programmes are widely available, it is unclear how consistently young people are offered these opportunities at a pre-court stage and how efficiently they are directed to appropriate therapeutic and rehabilitation-focussed programmes (Green et al., 2016). Recent inquiries have highlighted that more can be done to improve the diversion of young people (Armytage & Ogloff, 2017; Legislative Assembly of NSW, 2018).

This paper explores the role of police in this landscape, particularly because (a) across Australia, police play a key role in assessing eligibility of offenders for pre-court diversion options and (b) the use of diversion options is not legislated. That is, while police are not sole determinants of individual outcomes and have very limited discretion where serious (e.g. violent) crime has occurred, they are nevertheless the key entry point or ‘gatekeepers’ to the justice and diversionary systems (Prenzler & Hayes, 2000; Schulenberg & Warren, 2009; Stewart & Smith, 2004). The following section explores the literature on police discretionary decision-making. It also introduces theoretical work on naturalistic decision-making (NDM) which considers how professionals make decisions in real-world environments by referencing patterns formed through experience (Klein, 2008; Orasanu & Connolly, 1993). This theoretical work is drawn on to later facilitate the discussion of practical implications of this study.

Methods

Study interview data were collected as part of a mixed methods study exploring the pathways of young people in the justice system (Green et al., 2016). Of central concern were barriers to diversion and treatment experienced by young people with substance use issues [Reference removed for peer review]. The study was conducted in Melbourne (in the State of Victoria) and Sydney (New South Wales (NSW)) and encapsulated different legislative and policy environments.

This paper focusses on 25 semi-structured interviews conducted with staff from the NSW Police Force (n = 12) and Victoria Police (n = 13). Twenty worked in ‘frontline’ (operational) or youth specialist roles in inner-city Sydney and Melbourne. A small number of the Victorian sample (n = 5) worked in relevant policy positions and had specialist knowledge of youth diversion programmes (three had previous experience in frontline policing). Table 1 provides information about participants by state, role type and groups frontline staff by experience level. Police recruitment was facilitated through liaison with a project reference group. Nominated officers were contacted via email by research staff and times organised to conduct interviews at the relevant stations.

OPEN IN VIEWER

Table 1.

Police interviewees (n = 25) by role type, experience and state.

Role type and years of experience	NSW	Victoria
Frontline junior (under 5 years’ experience)	3	–
Frontline mid (5–15 years’ experience)	3	3
Frontline senior (over 15 years’ experience)	4	1
Youth specialist a officer	2	4
Policy role	–	5
Sub-total	12	13

NSW: New South Wales.

^aYouth Liaison Officers in NSW and Youth Specialist Officers in Victoria.

Qualitative methods are suited to explorations of meaning, particularly in relation to groups which are difficult to access and provide complex understandings of how people’s lives are mediated by environmental and social influences (McKeganey, 1995; Sifaneck & Neaigus, 2001). A theme sheet guided discussion in all interviews. Questions were asked about the role and years of experience of interviewee, general characteristics of young offenders and common types of offending, characteristics of inner-city policing, general views on diversion, description of local diversion options, factors informing use of these options, referral of young people to services, service system gaps and, support and resourcing provided to police. While the focus of the general study was on young people with substance use issues, interviews with police focussed on approaches to policing of young people generally and the use of diversion options for low-level offences including, but not limited to, those that are related to illicit drugs. Consent was gained to digitally record and transcribe the interviews. Interviews were conducted between February and September 2014. Ethical approval to conduct the study was obtained from the Eastern Health Human Research Ethics Committee. This approval was ratified by the University of New South Wales Human Research Ethics Committee. Victoria and NSW Police also provided approval for the conduct of the study and release of the final report [removed for review].

The first two authors read all interview transcripts and interview notes and analysis was managed using the qualitative software package NVivo10. Preliminary analysis was presented to the study team at a workshop, and themes were refined and associations were drawn and interpreted. Otherwise known as inductive reasoning, this approach is regarded as one of the pillars of qualitative research (Bendassolli, 2013) and involves generation of theories based on analysis of the data (see Given, 2008). The final report had a policy and service-provision focus (Green et al, 2016). New analysis of the qualitative data sets, including interviews with young people, has since been undertaken by the team. The themes explored in this paper emerged inductively through analysis of police interview data and was then examined by drawing on theoretical scholarship. A two-city comparison was prominent in the broader study but is less so in this paper. This is because, while some structural and legislative nuances were identified that potentially impact on young people’s pathways, there was strong consistency between the issues relating to discretionary decision-making discussed among interviewees.

Findings

Consistent with justice policy and best practice literature, participants regarded offending among young people as a welfare issue, rather than a criminal one. Many emphasised the importance of using incarceration as a ‘last resort’, expressing a view that custodial environments are not rehabilitative for young people. A senior professional said:

I don’t think kids should be in detention unless they’re just totally out of control and a risk to themselves and the community, like a significant risk. I know from experience that the kids who go to jail – or go to youth detention – come out better criminals than when they go in. (#14, frontline, senior, NSW)

Discussion

The literature investigating police discretionary decision-making has consistently shown that police decisions are informed by a number of inter-related situational (e.g. offence seriousness), system (e.g. legislation, departmental culture) and offender variables (e.g. age, race and demeanour) (Gaines & Kappeler, 2011). Policing research has given increased attention to the values, skills and experience police bring to their work (Schulenberg, 2010), and some recent work has investigated values such as ‘blameworthiness’ in relation to discretionary decision-making with adults (Ishoy & Dabney, 2018). However, there has been less attention to investigation of the values that inform decision-making with young people (see Richards, 2019; Richards et al., 2019). This is an important gap given that police have significant scope for the use of discretion with this group – particularly where offences are ‘low level’ and they have the opportunity to refer to diversion programmes or support services. Drawing on 25 in-depth interviews conducted with police staff in two major Australian policing organisations, this paper has examined how values, beliefs and expectations about young people may be considered along with situational, systems and offender variables to shape discretionary practice. Drawing on NDM scholarship, three linked sets of schemas (Lipshitz & Shaul, 1997) were discussed.

The first was the view held by police that many young people do not see police as a source of assistance and/or do not want their help. This appeared to be underpinned by the reportedly poor reputation that police have among young people. It has been long established that offender demeanour is important in shaping outcomes (e.g. Arcuri, Gunn, & Lester, 1979) and this study appeared to show the same. Ishoy and Dabney (2018) found that demonstrations of remorse were associated with a softer approach and this was also the case here – with remorse being a signal that the young person has learned something and of potential future compliance. NDM scholarship emphasises how schemas such as these facilitate rapid decision-making by experts (Klein, 2008). This schema appeared to assist police work in the context of scarce time and resources and was reinforced by reportedly ‘frustrating’ experiences of spending time with/or giving ‘chances’ to young people who quickly reoffended (i.e. ‘wasted’ their time). Accordingly, judgements about young people who would likely accept offers of help were reportedly developed quickly.

The schema that young people do not want their help affects pre-court diversion, which relies on cooperation between the young person and police, including the provision of consent and an admission of guilt. Furthermore, assessments made by police based on the quality of the engagement are problematic in relation to vulnerable young people (e.g. with an acquired brain injury, substance use, familial justice involvement) who, drawing on their own schemas about police, may react strongly to police based on previous encounters, or have difficulty with emotional regulation, but who may particularly benefit from diversion.

Second, participants appeared to be affected by the schema that some young people are more open to, or deserving of, lenient and therapeutic interventions than others, leading them to show leniency towards young people with good prospects – or who appeared unlikely to reoffend. Consistent with studies conducted outside Australia (McAra & Mcvie, 2005, 2011), police reported that they almost immediately differentiate who deserves ‘a chance’. This contradicts other sections of the interviews in which participants insist that their role is distinct from psychologists or those working in more therapeutic environments. NDM scholars emphasise that judgements are always, in some way, an assessment of the future (Dowie, 1993). We contend that, in the absence of feedback or education about the needs of this group and the effectiveness of different approaches, police may have limited understanding of the impact of their decisions on a young person’s journey. Moreover, there is little opportunity to reassess the schemas that frame their decisions about how to respond to young people who offend. Schemas are, to some degree, subconsciously held and making police aware of these cognitive shortcuts has the potential to disrupt them. The importance of providing police feedback about outcomes of individuals (and of police approaches generally) was a finding of Kellow et al. (2008) who investigated barriers to use of drug diversion initiatives.

The third schema was poor faith in the system to rehabilitate. This appeared to be reinforced by apparent detachment of police from the therapeutic process. Again, consistent with Kellow et al. (2008), many interviewees suggested that cautions, warnings and other such responses have very little impact on the trajectories of young people with complex needs. In the context of NDM, this view appeared to underpin a sense of futility in using diversionary options or making referrals. Knowledge about what occurs in diversion programmes, and other support services, was inconsistent. Improving knowledge about therapeutic services coupled with enhanced feedback mechanisms (as above) could diminish a widespread view that the system is ineffective. Increasing awareness about the nature of the process of change for young people with complex issues, such as recovery from mental health issues, trauma and problematic substance use could be valuable. Improving police knowledge about how therapeutic services engage and motivate young people also has the potential to increase their faith in the capacities of the service system.

Participants supported more intensive interventions, specifically mentioning RJC and work engaging family members, although time constraints prevented more common use. Supporting police to use RJC is likely to be highly worthwhile given the rigorous evidence demonstrating its efficacy (when delivered in a certain manner) for reducing certain types of offending (Sherman, Strang, & Woods, 2000; Strang, Sherman, Mayo-Wilson, Woods, & Ariel, 2013; D. B. Wilson, Olaghere, & Kimbrell, 2017). Police involvement in well-managed restorative processes are also more likely to build their belief in its efficacy, helping to offset schemas related to the futility of giving young people ‘a chance’. Although intensive work engaging families is likely beyond the remit of police, family-focussed interventions in the youth justice context have a strong evidence base (see Mackenzie & Farrington, 2015) and were supported by police. Attention to resourcing police with direct referral links to family-focussed specialist services in the community may therefore be worthwhile.

Other measures, when implemented consistently over time, may help to shift negative schemas that inform police decisions. For example, though there were locational differences, findings suggest that ongoing effort is needed to ensure that referral processes are streamlined, quick to administer and do not require follow-up by police (see also Hughes & Ritter, 2008; Kellow et al., 2008). Police in this study particularly valued youth specialist officers in the police workforce (Youth Liaison Officers in NSW and Youth Specialist Officers in Victoria). It appeared that these personnel may be important to the challenge of improving police legitimacy among young people and that increased funding support from both Victorian and NSW state governments in recent years is worthwhile and should be considered in other contexts. A specialist initiative in Victoria, Youth Referral and Independent Person Program, which supports police when they apprehend vulnerable young people, was also discussed positively. Availability of this service, however, was inconsistent and relied on volunteers.

Participants noted a wide variation in capability and confidence of police colleagues with young people. Consistent with recent work by Richards (2019), our analysis suggests that it may be appropriate to increase youth-specific core police training covering developmentally informed practice, working with young people with complex needs, trauma-informed practice and de-escalation techniques. Nevertheless, we consider that there is not likely to be one solution to creating lasting positive changes within policing workforces. Police participants in this study reported that the most substantial learning relevant to young people occurs ‘on the job’. This indicates a preference for experiential learning and police-delivered training (see also Tallon, Spadafore, & Labriola, 2016). Pro-active policing programmes, such as police participation in structured activities with young people, have been shown to shift views of police about young people, and vice versa (Goodrich, Anderson, & LaMotte, 2014). These initiatives represent a good opportunity for experiential learning and to challenge negative schemas about young people. Extended consideration of potential strategies is not possible here; however, it is important to note that police views of young people are not ‘fixed’ and immutable – and that investing in strategies to support police learning about youth issues and developmentally appropriate practice is worthwhile.

The emotional toll and traumatising effect of police work were evident throughout these interviews, particularly in accounts of participants with cynical or pessimistic schemas. Developing defensive styles of interactions with young people and hardened emotions leads to a depersonalisation of the work (Abdollahi, 2002; Caplan, 2003) and potential use of more punitive approaches. This reiterates the importance of ensuring that stress and mental health management among police professionals is prioritised.

The study has limitations. Sampling was purposive, designed to access views of police who had experience of encountering young people in inner-city areas of Melbourne and Sydney. Researchers sought to recruit police staff with knowledge and expertise in youth-focussed policy and it is likely that our participants, particularly those in specialist youth roles (n = 6), were sympathetic to young people because of this interest. It is also likely that, despite assurances of confidentiality, participants protected the reputation of their organisation throughout their responses. It should be noted that the views of participants do not represent official policy responses of Victoria Police or the NSW Police Force. Finally, we note that young Aboriginal and Torres Strait Islanders are vastly over-represented in the justice system in Australia (AIHW, 2019). We acknowledge the need to consider responses to this serious problem through a cultural and local lens and using a community-driven approach that was unfortunately beyond the scope of the current research.

Conclusion

Our findings suggest that, along with practical and systems-oriented measures, there is a need to implement strategies that help to ameliorate the cynical schemas that underpin decision-making by some police. Using NDM theory, we have argued that the schemas held by police facilitate decision-making under pressure and that some schemas evident among our research participants may not be consistent with diversionary aims of the system. Addressing these issues has the potential to enhance welfare-oriented police practice but requires a multi-faceted and sustained approach. A focus on organisational culture and leadership is crucial. Importantly, while police have a key role to play in the lives of some young people, they are a piece in a complex puzzle and have limited contact with young people or scope to intervene positively. Strong investment in systematic early identification and engagement of vulnerable children, and their families, in comprehensive services, with a focus on intergenerational and location-based disadvantage, are critical factors for effective diversion of young people away from the justice system and supporting good outcomes across the life course.