Abstract
Youth Justice Assessments of Children Subject to Statutory Court Orders in England and Wales Take, on Average, Over 15 Hours to Complete
The Youth Justice Board for England and Wales was established, by the Crime and Disorder Act 1997, as a non-departmental public body to oversee the operation of the youth justice system. One of the Board’s most significant initiatives was the development, and subsequent national roll-out from April 2000, of a standardised, mandatory, assessment tool, called ASSET, which youth offending teams were obliged to complete for all children subject to court orders or statutory pre-court interventions. ASSET was an actuarial instrument which required practitioners to ascribe scores to individual children across 12 domains of risk. While it also aimed to identify protective factors, the latter were not subject to scoring. Where risks were identified as contributory to the child’s offending behaviour, supervision planning was expected to address those factors. With the onset of, what was known as, the ‘scaled approach’, introduced by the Youth Justice Board in 2009, levels and frequently of interventions were also to be determined by the total ‘risk of reoffending score’ aggregated from the different ASSET domains.
The risk factor prevention paradigm, in which ASSET was firmly rooted, was increasingly subject to criticism. At the same time, the growing influence of desistance theory, which focused on the processes through which children give up offending rather than identifying past risks which might help to explain their criminal behaviour, led to a growing recognition that ASSET was unduly deficit focused, limited practitioner discretion and tended to ignore the perspectives of children themselves.
The Youth Justice Board accordingly developed a new assessment framework, AssetPlus, which was designed to address some of the criticisms of ASSET. The revised tool was intended to allow risk ‘to be balanced alongside consideration of a young person’s needs, goals and strengths’ and aimed to encourage practitioners to identify ‘existing and potential strengths in a young person’s life’. It was anticipated that the modified framework would provide:
a more holistic approach through an integrated assessment and planning process . . . that follows a child though their interactions with the youth justice system.
In a separate development, Charlie Taylor, in his government commissioned review of the youth justice system, published in 2016, questioned whether a dedicated youth justice framework was required in any event, noting that many children in contact with criminal justice agencies were also users of other services for vulnerable children. Such crossover, Taylor maintained, led to considerable duplication where different assessment frameworks were applied. In this context, he recommended that the Ministry of Justice consider:
whether local authorities should be able to use their own assessment systems, rather than use systems prescribed by the centre.
Taylor was subsequently appointed as Chair of the Youth Justice Board, but during his tenure, between 2017 and 2020, there were no moves to progress this proposal.
AssetPlus was implemented on a phased basis, and adopted by youth offending teams between October 2015 and August 2017. From April 2019, with the introduction of new national standards for youth justice, completion of AssetPlus became optional for children other than those subject to community-based court disposals or custodial orders for which it remains mandatory. In that same year, the Youth Justice Board published a process evaluation of AssetPlus which noted that the principles underpinning the revised framework, in particular the focus on children’s strengths and professional discretion, were broadly welcomed by youth justice practitioners, but the evaluation also highlighted concerns about the time taken to complete assessments and the non-linear nature of the tool which contributed to making it less user-friendly in developing intervention plans. In May 2022, the Board published a follow-up outcome evaluation of AssetPlus to further explore these issues and ‘assess the impact of AssetPlus against a range of outcomes’. The Board originally asked the evaluation team to consider 18 outcome measures, but a feasibility study concluded that impact could only be determined for 8 of these. Expected outcomes fell into three broad categories:
Improvements of youth offending team operational efficiency;
Promotion of changes in children’s strengths; and
Improvements in children’s behaviour.
The outcome evaluation reports on findings against each of these headlines.
The implementation of AssetPlus was anticipated, by the Board, to improve youth justice service provision since operational guidance recommends a ‘proportionate’ approach to completing the tool, with the depth of assessment, and time spent undertaking it, reflecting the seriousness of the child’s offending and the complexity of their needs. There was an expectation that this flexibility would result in a reduction, on average, in time spent on paperwork with a corresponding rise in capacity for face-to-face work. The involvement of children and their families in the assessment process was expected to result in improved relationships between youth justice professionals and those to whom they provide services. The requirement to consult with partner agencies, as part of the assessment process, was similarly thought to be conducive of improved working relationships between those agencies and youth justice.
No data were available in respect of completion time for ASSET, so the evaluation was reliant on time spent on assessments using AssetPlus as the new framework bedded in. As shown in Table 1, the time taken for an initial assessment, as reported by youth justice practitioners, was considerably higher 12 months after the implementation of AssetPlus by comparison with a baseline at the point at which the tool was introduced. At 18 months after the baseline, completion times for prevention cases and out of court disposals had fallen somewhat, but were still substantially in excess of the baseline figure (overall increases of 124% and 112%, respectively). By contrast, completion time for statutory cases continued to rise across the whole of the 18-month period: from 603 minutes (10 hours, 3 minutes) to 947 minutes (15 hours, 47 minutes), an increase of 57 per cent.
Average reported completion time for initial AssetPlus assessments, by case type, at baseline, and at 12 and 18 months post baseline.
Contrary to expectations, while the latter case type showed a higher completion time at the baseline, the rise for other case types over the first 12 months was considerably more pronounced with the result that, at the second measurement point, both had overtaken statutory cases. The continued rise for children subject to court orders to 18 months after the baseline meant that at this final measurement point, assessment times were longest for statutory cases but the gap with other case types had narrowed considerably since the baseline.
Completion time for review assessments also showed an overall rise, for all case types, albeit that, as might be expected, all times were shorter and the increases less pronounced. Moreover, in each instance, there was a reduction between 12 and 18 months from the baseline, as shown in Table 2.
Average reported completion time for review AssetPlus assessments, by case type, at baseline and at 12 and 18 months post baseline.
Concerns identified in the earlier process evaluation that AssetPlus assessments were disproportionately time-consuming for lower level cases were accordingly corroborated by the outcome evaluation. Moreover, these findings were contrary to Youth Justice Board expectations.
AssetPlus was also anticipated to facilitate the identification, and promotion, of children’s strengths as a consequence of the revised framework’s increased focus on protective factors as opposed to risk. The evaluation, however, found that follow-up assessments consistently recorded higher levels of concern than those identified at the initial assessment, and this was true for each of the 11 ‘individual factors’ measured by AssetPlus (including, for instance, attitude to offending, mental health, risk to others and substance misuse) as well as the ‘contextual factors’ assessed by the framework (including, for instance, accommodation, family behaviour, parenting and care history). The authors of the evaluation conclude that since such concerns are likely to hinder the promotion of children’s strengths, the expectations as to the positive impact of AssetPlus, in this regard, were not borne out.
More promising outcomes were found in respect of the estimated impact of AssetPlus on improving children’s behaviour. The rate of breach, for instance, for a cohort of children assessed using AssetPlus was one percentage point lower than for a control group assessed prior to the new framework’s implementation (9.6% as against 10.6%). The improvement was focused particularly on children subject to community orders, with breach rates for AssetPlus being higher than the control group for other sentences. Nonetheless, the authors point out that:
in the year to March 2015 (before the introduction of AssetPlus) there were 4,053 breaches. In each subsequent year, while AssetPlus was being introduced, the number of breaches dropped.
The authors note that the reduction in breach is ‘small but statistically significant’. There may, however, be other explanatory factors not mentioned in the evaluation. Breach of statutory orders had been declining for some time prior to the adoption of AssetPlus: in the year ending March 2007, for instance, there were 15,910 instances of breach of statutory order recorded in Youth Justice Annual Statistics; this figure had fallen to 10,197 in 2011 and 4583 in 2014. (It should be acknowledged that there are considerable deficiencies with the figures for breach in Annual Statistics, but these appear to be the data on which the evaluation relies.) The subsequent reductions might therefore plausibly be regarded as a continuation of an existing trend rather than an outcome associated with the introduction of AssetPlus.
The evaluation also notes increases in order completion, and compliance with appointments, associated with the introduction of AssetPlus. Just over three-quarters (76%) of children in a sample assessed using AssetPlus were recorded as having successfully completed their orders compared with an equivalent rate for children in a control group of 68 per cent. As shown in Table 3, children in the former sample were also significantly more likely to comply with face-to-face contacts by 3 months into their order, although during the first month of the order the AssetPlus group was recorded as having slightly higher rates of non-compliance.
Rates of non-compliance with face-to-face appointments by children assessed using AssetPlus and a control group.
Order completion, it should be noted, is likely to be inversely related to rates of breach, since where children are returned to court they frequently have their orders revoked and are unable to complete them. To the extent that data for the control group derive from an earlier timeframe, when breach rates were higher, one would therefore anticipate that order completion for this cohort would be lower. (The respective timeframes for each sample are not given in the evaluation, but since data appear to have collated during the COVID-19 lockdown, at which point all youth offending teams were using AssetPlus, it seems likely that the control group data were drawn from an earlier point.) Moreover, it is apparent that while breach and completion rates may be indicative of changes in children’s behaviour, as assumed in the evaluation, they may also be a function of changes in professional practice and recording.
The authors concede that the evaluation was constrained by the available outcome measures but nonetheless conclude that:
[t]he overall estimated impact of AssetPlus on the outcomes included in this study has been limited.
In particular, the time spent on assessments has increased but identified concerns in relation to children have also risen. The authors argue that reductions in breach rates and non-compliance with appointments are ‘consistent with the primary stage of desistance’, suggesting some impact on children’s behaviour, but it is arguable that such changes might also reflect, at least to some degree, the adoption of a less punitive approach, over time, by youth justice professionals:
An overview of ASSET is provided in Assessing the predictive validity of the Asset youth risk assessment tool using the Juvenile Cohort Study, by Edith Wilson and Samantha Hinks, published by the Ministry of Justice and available at: Assessing the predictive validity of the Asset youth risk assessment tool using the Juvenile Cohort Study (JCS) (publishing.service.gov.uk) The Youth Justice Board’s case for replacing ASSET by a new assessment framework is given in AssetPlus rationale, by Kerry Baker, published by the Board and available at: Formal document (publishing.service.gov.uk) Review of the Youth Justice System in England and Wales, by Charlie Taylor, is published by the Ministry of Justice and is available at: Review of the Youth Justice System (publishing.service.gov.uk) Process evaluation of AssetPlus, by Natallie Picken, Kerry Baker, Camilla D’Angelo, Clément Fays and Alex Sutherland, is published by Rand Europe and is available at: Process Evaluation of AssetPlus (publishing.service.gov.uk) AssetPlus outcome evaluation: final report, by Jack Cattell and Kasra Aghajani, is published by the Youth Justice Board and is available at: AssetPlus outcome evaluation - GOV.UK (www.gov.uk) Figures for breach of statutory orders are given in Youth Justice Statistics 2015/16, published by the Ministry of Justice / Youth Justice Board, and available at: Youth justice statistics - GOV.UK (www.gov.uk) A comprehensive overview of the deficiencies of published breach data is provided in Non-compliance and breach: youth justice decision-making in Wales, by Heddwen Daniel, available at: Heddwen_Daniel_FINAL_THESIS_FOR_SUBMISSION_PRINT_AND_BINDING.pdf (southwales.ac.uk)
Eighty Per Cent of Individuals Who Commit School Shootings in the United States Use Weapons Stolen from Family Members
A report, published in June 2022, provides an overview of levels of school crime and safety in the United States. It indicates that the proportion of school students in grades 9–12, who reported taking part in a physical fight on school premises in the previous 12 months, fell from 11 per cent in 2009 to 8 per cent in 2019. Over the same period, the proportion of children who self-reported weapons carrying in the previous 30 days on school premises also declined, from 6 to 3 per cent. Reported gang presence within schools and prevalence of hate-related graffiti had also reduced: in 2009, one in five students aged 12–18 years indicated that their school had a gang presence and 29 per cent said that they had seen hate-related graffiti on the premises; a decade later, the equivalent percentages were 9 and 23 per cent, respectively.
Despite these relatively promising trends, there had been no corresponding increase in students’ perceptions of safety. In 2019, 5 per cent of children aged 12–18 years reported being afraid of being attacked or harmed at school, a figure which was not measurably different from that in 2009. Moreover, the proportion of children fearing for their safety at school was two percentage points higher than those exhibiting concerns about safety away from the school environment.
The picture in relation to the most extreme forms of harm is, moreover, much less positive than the above indicators of behaviour. During the year 2020/2021, there were 93 school shootings in the United States that resulted in death (n = 43) or injury only (n = 50). As shown in Table 4, this figure represents a sharp rise since 2009/2010, when only 11 shooting episodes with casualties were recorded.
Episodes of school shootings resulting in death or injury: 2009/10-2020/2021.
A factsheet published by the US Department of Justice, in August 2022, identifies five characteristics, derived from research, that are common across school shootings. Perhaps unsurprisingly, the large majority of individual perpetrators have struggled with aspects of mental well-being: most have experienced significant childhood trauma or hardship and almost all (92%–100%) have been suicidal during the incident or prior to it. While the prevalence of mental health difficulties across society and the rarity of shootings mean that psychological profiling cannot be used as a predictor, the authors contend that such figures reinforce the importance of targeted support to improve children’s well-being.
Most people who commit school shootings are in crisis in the period prior to the incident and this is nearly always visible to others. Frequently, perpetrators convey their plans, either directly or through social media, providing opportunities for potentially dangerous situations to be reported to the authorities. Crisis intervention with such individuals, where authorities are alerted, has shown some success in averting school shootings.
The large majority of those carrying out shootings are ‘insiders’, who have some connection with the school concerned, often as current or former students. Consequently, communities:
can help prevent mass school shootings by working together to address student crises and trauma, recognizing and reporting threats of violence, and following up consistently.
This potential to prevent shootings is enhanced where schools adopt ‘threat assessment protocols’ which ensure that individuals in crisis who may pose a risk are identified and receive appropriate support, or are subject to law enforcement responses.
Finally, research confirms that four out of five of individuals who have committed a school shooting stole the weapon used from a family member. This contrasts sharply with mass killings outside of school where more than three-quarters (77%) of firearms are bought legally. The factsheet concludes that research on the psychological factors associated with school shootings, appropriate mental health support, alongside:
other factors that precipitate school violence, can help inform targeted intervention in coordination with crisis intervention, threat assessment, and improved firearm safety practices. Report on indicators of school crime and safety: 2021, by Véronique Irwin, Ke Wang, Jaishan Cui, and Alexandra Thompson, is published by National Center for Educational Statistics and is available at: Report on Indicators of School Crime and Safety: 2021 (ojp.gov) Five facts about mass shootings in K-12 schools, published by the National Institute of Justice, is available at: Five Facts About Mass Shootings in K-12 Schools | National Institute of Justice (ojp.gov)
Detected Offences Committed by Children Involving a Knife or Offensive Weapon, in England and Wales, Fall by 23 Per Cent over the Past 3 Years
There has been growing concern in England and Wales, in recent years, in relation to what has been perceived as an increase in children carrying knives and other weapons. A British Youth Council Youth Select Committee report to Parliament in 2019, for instance, referred to knife crime as ‘our generation’s epidemic’. Such concerns have led to a number of legislative initiatives designed specifically to tackle knife crime, including mandatory minimum custodial sentences, introduced by the Criminal Justice and Courts Act 2015, for a second conviction for a knife crime offence. This statutory provision requires that a child aged 16 or 17 should be sentenced to at least 4 months imprisonment in such circumstances unless it would be unjust to do so.
While any level of weapons carrying is clearly a concern, published data do not necessarily support the contention that knife crime by children is rising sharply. The latest figures for proven knife and offensive weapons offences, published by the Ministry of Justice in August 2022, indicate that while there has been some fluctuation in the number of such incidents where the child is a perpetrator, the overall trend has been relatively stable over the past decade. Indeed, as shown in Table 5, the number of knife and weapons offences committed by children in the year ending March 2022 (3373) was lower, by 6 per cent, than that in the year ending March 2012 (3586). (There was a similar decline for adults over the same period.) The data show reductions between 2012 and 2014, followed by subsequent rises in each year to 2019, reaching a peak of 4345 in that year. Thereafter, a further decline set in leading to a 23 per cent decline between 2019 and 2022.
Detected knife and offensive weapons offences committed by children: 2012–2022.
It seems unlikely, moreover, that the recent fall is a consequence of any deterrent effect associated with more punitive treatment of weapons offending since, contrary to what government intentions appear to be, the statistical data suggest that responses to such offences have not become harsher in the recent period. As shown in Table 6, the proportion of knife and offensive weapons offences committed by children resulting in a caution, as opposed to a court conviction, rose between 2012 and 2022 from 26 to 38 per cent; over the same period, the proportion which attracted a custodial disposal declined, from 12 to 6 per cent. The introduction of mandatory sentencing for a second offence in 2015 coincided with a small increase in the use of imprisonment up to 2018, but from that year onwards, custodial penalties fell sharply.
Cautions and custodial penalties as a proportion of all outcomes for knife and offensive weapons offences committed by children: 2012–2022.
Detected offences do not, it should be acknowledged, provide a direct measure of the underlying level of knife crime, but another indicator which sheds some light on the extent to which offensive weapons are used can be derived from hospital data for admissions for injuries incurred through assault by a sharp object. While such figures cannot distinguish between adult and child assailants, the age of the victim might be understood as a proxy measure since, in many cases, perpetrators and victims are likely to be of a similar age. As shown in Table 7, these data do not suggest that there has been a consistent upward trend over the recent period: the number of children 1 admitted for such injuries fell between 2011 and 2015, rose over the next 4 years and then declined in the latest 2 years, for which data are available. In the year ending 2021, the number of admissions for children was 5 per cent lower than a decade previously.
Our generation’s epidemic: knife crime, by the British Youth Council Youth Select Committee, is published by the House of Commons, and is available at: Our Generation’s Epidemic: Knife Crime (byc.org.uk) An overview of knife offences and sentencing options, including mandatory custodial sentencing, is given in Knives and offensive weapons, by Sally Lipscombe, published by the House of Commons Library and available at: Knives and offensive weapons (parliament.uk) Knife and offensive weapon sentencing statistics: year ending March 2022, published by the Ministry of Justice, is available at: Knife and offensive weapon sentencing statistics: year ending March 2022 - GOV.UK (www.gov.uk) Figures for hospital admissions are given in Knife Crime in England and Wales, by Grahame Allen and Megan Harding, published by the House of Commons Library and available at: SN04304.pdf (parliament.uk)
Completed consultant episodes for hospital admissions for injuries arising from assault by a sharp object: individuals aged 18 years or under.
Nine in 10 Girls Under Youth Justice Supervision in South Australia Report Experience of Emotional Abuse
Research, published in June 2002, which explores the prevalence of adverse childhood experiences (ACEs) among a broadly representative sample of children subject to youth justice supervision in South Australia, confirms that the large majority had experienced a combination of both maltreatment and household dysfunction. The sample comprised 184 children, aged 14 years or over, who were either under community supervision (57%) or in custody (43%). The median age of the cohort was 16 years, 16 per cent were girls and more than one-third (38%) identified as Aboriginal. Each of the participants competed a range of self-report assessments. Where consent was given – and this was forthcoming in all but four cases – self-reported information was combined with data from administrative records.
Although a minority of the sample were in custody at the time of the research, the large majority (83%) of those who consented to access to their administrative records (n = 180) had been in custody on at least one occasion prior to the research commencing. While theft accounted for the largest proportion of offences leading to children’s first contact with the youth justice system, more than half (57%) of the children had been processed for a least one violent offence by the time of the study. Levels of reoffending among the cohort were high: 78 per cent of the sample had had at least one further conviction in the 12 months following their involvement in the research.
There was a significant crossover between youth justice involvement and contact with other services for vulnerable children: 94 per cent of the sample was known to child protection services, the large majority of whom had been referred for maltreatment. Almost one-third of the children had been in care, more than 1 in 5 (21%) had experienced at least one residential care placement, and 14 per cent were on long-term care orders until the age of 18 years. In nearly all cases (97%), contact with child protection agencies pre-dated youth justice supervision with the median age of first referral to the former being 3 years.
Exposure to ACEs was extremely high, with 88 per cent of the sample reporting that they had experience of four or more. Conversely, ‘fewer than five’ children reported zero ACEs. Eighty nine per cent had been exposed to a combination of maltreatment and household dysfunction. Table 8 details the prevalence of ACEs which children said they experienced ‘often’ or ‘very often’.
Proportion of children subject to youth justice supervision reporting experiencing selected ACEs frequently (often or very often).
ACE: adverse childhood experience. Ninety-two per cent of participants have experienced at least one of these ACEs frequently.
Girls had a higher prevalence across the range of ACEs measured than their male counterparts, with the exceptions of physical neglect and neighbourhood violence. In particular, 90 per cent of girls reported being exposed to emotional abuse. Aboriginal children’s experiences of ACES were broadly similar to those of their non-Indigenous peers, but the former group were much more likely to report having had a member of their household imprisoned.
Participants were also invited to complete Trauma Symptom for Children assessments. Results indicated that 88 per cent of respondents were in the symptomatic range on at least one indicator. A high prevalence of ‘clinically significant problems’ was established across a range of trauma symptom domains, including, most commonly, dissociation, post-traumatic stress and anger. While girls were more likely to exhibit the latter, boys were more likely to score as symptomatic for post-traumatic stress. Problematic sexual distress was considerably less prevalent, but it was much more common among girls than their male counterparts. The proportion of children from Aboriginal and non-Aboriginal backgrounds scoring in the symptomatic ranges was similar for most domains, but the former were considerably more likely to display symptoms of anger.
The large majority (86%) of the total sample also scored in the problematic range for alcohol and/or other drug use. Forty-three per cent of respondents reported drinking or using drugs because they were ‘stressed, tense or full of worry and problems’. As shown in Table 9, while the prevalence of problematic substance misuse was high among all groups of children, girls and Aboriginal children were more likely to report such difficulties.
Prevalence of problematic alcohol/drugs use by sex and cultural background.
The authors of the research note that the level of ACEs established in this study ‘is up to four times higher than that reported in the international literature’. They speculate that this difference may reflect the fact that in South Australia, youth justice supervision tends to be reserved for children who ‘have committed the most serious and repeat offences’. Such findings reinforce the importance of local data collection rather than relying on results from other jurisdictions since there may be considerable variations arising from different pathways into formal arrangements for dealing with children who offend. From a policy perspective, the authors suggest that the results of the study:
add further strength to arguments to implement more holistic approaches to the assessment and treatment of young people under youth justice supervision. It also points to the importance of early intervention efforts to prevent the accumulation of ACEs. Adverse childhood experiences and trauma among young people in the youth justice system, by Catia Malvaso, Andrew Day, Jesse Cale, Louisa Hackett, Paul Delfabbro and Stuart Ross, is published by the Australian Institute for Criminology and is available at: Adverse childhood experiences and trauma among young people in the youth justice system | Australian Institute of Criminology (aic.gov.au)
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.