Working with stats makes me happy, and you’ll think “that’s clear” when you read this publication. Numbers fascination aside, it’s important we produce quality, interesting and impactful research. In relation to our legislative responsibilities, for several years now we’ve been eager to publish this meaningful research looking at the Order for Lifelong Restriction (OLR).
Little is known about individuals subject to an OLR. It’s a unique sentence within Scottish, even international, justice. Therefore, whilst the length of the report shows we got a bit carried away (and improved self-management is required), people will see a clear intention to increase understanding of the OLR population. It’s essential we do this to contribute to evidence-based effective practice in risk assessment and risk management
This publication is really a collection of studies that involved the same data collection and analysis procedures. They had two key aims; to examine offending behaviour characteristics, and to consider the MacLean Committee expectations of who the OLR might be for. The publication is formed of a series of chapters so readers can choose to look at the publication as a whole, or at specific areas of interest. It begins with the overall OLR population, followed by areas including young people, intimate partner violence (IPV), and those with no previous convictions.
Development of the Project
One challenge in developing this project was deciding on what to, or not to, include. Our eagerness has led to the RMA’s first research book (and it easily could’ve included more information…). Data collection included coding specific offences and so the dataset could be used to explore further areas (e.g. types of assault, offence aggravators, etc.).
Coding involved trying to achieve as much inter-rater consistency as possible. We developed a detailed framework with variable definitions, and held regular team meetings and discussions to agree on how to code variables.
Hopefully I can now improve those self-management skills I mentioned earlier by trying to briefly comment on some of the most interesting findings…
Firstly, regarding extent of offending, approximately a fifth of the 202 individuals evidenced all four offence types (i.e. sexual, violent, IPV, and ‘other’) and just under 10% had multiple previous convictions including all offence types. Additionally 90% of the sample had violent offences. These examples indicate the magnitude and breadth of offending within the OLR population.
Using allegation information is a particularly unique (and contentious) aspect of the OLR assessment and sentencing process. There were 8 cases where the assessor had stated that allegations influenced the risk rating (legislation means they must state whether it did or not). Nearly 90% of the sample had alleged offending. What is particularly interesting is where allegations were of a different nature to an individual’s convictions. For example, there were 20 individuals with sexual allegations but with no sexual offences, and 14 and seven individuals in relation to IPV and violence, respectively. These findings have risk management implications as they indicate that often individuals may present with risks beyond what their convicted offending may indicate. Using this effectively and defensibly in risk assessment and is something I really look forward to exploring beyond this publication.
Analysis of sexual offending behaviour also produced interesting findings. For example, there was generally a higher prevalence of this with individuals rated as ‘medium risk’. It also featured prominently in index and alleged offending of those with no previous convictions, and occupied a higher percentage in the offending behaviour of those under 21 years old. There are general indications of an observable pattern where sexual offending is potentially viewed differently, in the context of risk assessment and decision-making, to other types of serious offending. Our findings show that where individuals had fewer convictions, sexual offending tended to feature. Whilst specific research would be necessary to explore this more effectively and thoroughly, it provokes consideration around how this observation fits with the evidence of sexual offending having lower recidivism rates than, for example, violent (non-sexual) offending. It would be interesting to explore in more detail how this potential pattern may relate to; subjective interpretations of the physical and psychological harm caused by different types of offending, moral psychology, the influence of our own attitudes, and managing biases with risk assessment and management.
It’s also worth mentioning some limitations. We analysed Risk Assessment Reports (RAR) used at sentencing, which means that this publication is a snapshot in time. Furthermore we specifically focused on offences and didn’t analyse underlying motivations; we decided that future studies could focus more on the analytic sections of assessments, and this would reduce coding subjectivity in this publication.
This study may inform future iterations of Standard and Guidelines, and risk assessment and management of individuals subject to an OLR. For example, specifically for risk management, the findings regarding allegations link to what might be required to effectively manage the risk of individuals subject to an OLR. In fact, offending brevity and density across individuals indicates their complexity. Based on principles like proportionate practice this also relates to the depth of risk management and services necessary to provide adequate opportunities for rehabilitation. Our research primarily focuses on evidencing the amount of offending and harm related to this population; it should not detract attention from the level of support and services required to effectively manage the risk and meet the criminogenic needs of individuals with an OLR.
This publication allows us to revisit our previous advice to Scottish Ministers around a presumption against OLR’s for young people (although it’s reassuring this hasn’t happened since 2014) and for there to be a sentence review mechanism. Risk being considered as ‘lifelong’ is something I’m sure many working in risk assessment and management may find uncomfortable. For example, literature on sexual offending has shown that as offence-free periods increase the risk of reoffending decreases (I highly recommend looking at Karl Hansen’s work on recidivism rates over time for more detail).
I want to emphasise that this publication marks the start of a series of OLR studies. We are currently analysing data on psychopathy, and on personality disorder (notice us learning to separate studies out!). These studies will be followed by a quite unique study looking at the profile and characteristics of the victims of individuals who are subject to an OLR. From a qualitative perspective, we’ll also be publishing a study looking at individuals subject to an OLR and their experiences of release and recall.
It’s imperative we increase our understanding of both the sentence and the people subject to it. These studies will be accompanied by a recently commenced long-term project to review the implementation of the sentence with a view to seeing how it might be improved based on what is, and what is not, working.
To wrap up I want to say I’m really proud of everyone who worked on this publication. An incredible amount of work went into this and it’s really satisfying to see the outcome. I look forward to speaking with people and agencies interested in the findings and implications. We’re working on a series of videos which discuss the findings, sign up to our Newsletter to hear first when they are released.
If you have an interest or any questions on the project, I really encourage you to get in touch with us to learn more.
Head of Development
Download the Research Report