Paper 2 - Cross Justice Working Group on Race Data and Evidence – summary of data discussions with justice organisations
In the November 2020 meeting of the group, findings from the audit of datasets held across justice organisations were presented. The audit examined existing justice datasets on the extent to which they held ethnicity data and collected key metadata in relation to those datasets. Subsequently a workshop discussion took place to identify i) priority questions which the group felt ethnicity information collected by justice partners should answer and ii) the gaps in the quality and completeness of what is currently held which would need to be addressed in order to do so.
The audit highlighted that different organisations were in different places in relation to achieving the group’s aims. This was both in terms of the technical requirements and the operational environment within which the data was being collected. In light of these differences we agreed it would be appropriate to meet with the relevant personnel from each organisation individually to get a more detailed understanding of their current position and any barriers to improvement. These meetings took place in the first quarter of 2021 and have now been concluded. This paper summarises the findings from those discussions, with suggestions for next steps to be taken.
Meetings were held with the following organisations:
COPFS, SFRS, Police Scotland, SLAB, SPS, SCRA, Justice Analytical Services (covering CJSW and SCJS)
Data under consideration includes all those interacting with the justice system. Workforce data is excluded as this is being covered by a separate dedicated group.
Detailed summary of discussions
Not all organisations collected ethnicity data directly from service users and some did not hold it at all. For example, COPFS and SCTS obtain the ethnicity data they hold from Police Scotland, and Parole Board Scotland do not hold any ethnicity data. Efficiency savings may be gained by prioritising addressing barriers in ‘source’ organisations responsible for providing others with the data, in that this would provide maximum impact. It was noted that where an individual has multiple points of contact with the same organisation, consideration should be given to what the most appropriate point to ask for the information is. Where organisations hold no ethnicity data, the potential exists for data linkage to fill some of these gaps by linking with other justice datasets. This would rely on the ability to match records from one dataset to another (e.g. through S numbers).
One of the barriers to collecting the information was the view that ethnicity data was not required for operational purposes. Some questions were raised as to whether there was a legal requirement under the Equality Act/ Public Sector Equality Duty, noting that the legislation was vague in parts. This is a very real issue as in some cases it resulted in the view that it was not appropriate to be collecting this information as part of operational practices. For others it meant staff members responsible for collecting the information felt vulnerable to challenges from service users when asked why they were asking them for information on their ethnicity. There was broad support for clarifying the legislative requirement behind collecting and holding this data so that organisations could use this as a lever for change.
A number of organisations highlighted the need for staff training on collecting the data and guidance around recording practices to ensure that the staff responsible for collecting the information gave it sufficient priority and felt safe doing so. In order to provide high quality data it was important that staff understood why they were asking for ethnicity data, the importance of it and how it will be used, and that this could be shared with those who they are asking for the information from. One organisation suggested publicising the drive for improvements in the recording of this information across the justice sector would help.
It isn’t always clear when the information entered is self-identified or when it is based on the perception of the person recording it. The consensus from the workshop discussions was that collecting self-identified ethnicity data was preferable, either in addition to or in place of non-self-identified ethnicity data, as operationally appropriate. With this it will be important to make self-identified and non-self-identified ethnicity data clearly distinguishable in recording systems.
Where ethnicity information is missing or ‘unknown’, it isn’t always clear whether the question was asked. In some cases (e.g. a casualty in a fire who is unconscious or where the individual in question in too young), it will not always be possible to obtain self-identified ethnicity. The question may also be left blank or marked ‘unknown’ because the service user did not want to answer the question or the question was not asked, potentially because the justice professional responsible for asking the question did not feel comfortable doing so. Current recording practices mean it isn’t always possible to distinguish. A move towards making self-identified ethnicity a mandatory field would assist in at least partially addressing poor completion rates, but would need staff guidance and training to ensure it resulted in accurate information and sufficient response rates to allow meaningful analysis. A mandatory field would also need to account for specific circumstances where it was not possible to obtain self-identified ethnicity or the individual chose not to disclose their ethnicity.
Some organisations were still recording ethnicity information in free text, which makes it very difficult to extract/analyse. Using designated data fields with drop-down menus from an agreed classification system would allow this information to be captured in a more useful, readily analysable way. There was widespread support for adopting an agreed classification system across justice organisations, with the census categories generally proposed as the most sensible option. The importance of having an agreed approach for mapping older data forward was also noted.
For some organisations the scale of change required to collect this information would be greater than for others, both in terms of technical requirements and staff training and recording practices. The scale of the technical requirements would to some extent depend on the number of recording systems affected, how they are currently structured and the process for system changes. It was noted that it would be beneficial to identify required changes to systems and recording practices as early as possible to allow incorporation into planned development work. In most cases staff training/education would also be required to bring about the necessary changes to recording practices. There would be associated resource implications as a result of both of these factors. In light of these, data linkage was again discussed as a potential option where it may be more efficient that implementing the wide-scale changes described, for example linking justice datasets to census ethnicity data. Census legislation would permit this use of the data in relation to specific research projects, but not for operational use.
Summary of findings
The main findings of the discussions have been summarised in the bullet points below:
- prioritise addressing barriers in ‘source’ organisations
- clarify the operational need/legislative basis for collecting the information
- self-identified ethnicity to be prioritised
- support for an agreed classification system
- need for staff guidance and training
- technical barriers
- resource implications for changes required
- potential for data linkage between justice datasets
- potential for data linkage to external sources e.g. census data
Proposed next steps
Alongside these conversations, two online surveys for academics and community groups have been produced, asking for their views on priorities for data and research. It has been proposed that the findings from these surveys and the discussions summarised above will feed into a high level report on the current state of data and future data development plans across the justice system. Following consideration by the group these reports will feed into a data development plan and future research plan to be agreed by the relevant Justice bodies.
Please feel free to get in touch if you feel any of the content of the discussions has been missed above or you have anything you wish to add.
Justice Analytical Services
There is a problem
Thanks for your feedback