IJPDS International Journal of Population Data Science IJPDS 2399-4908 Swansea University 10.23889/ijpds.v10i4.3153 10:3:124 Using Sankey diagrams to visually depict the service journeys of criminally exploited children Maxwell Nina 1 Cserző Dorottya 1 Wang Ting 2 Toby Roshan 2 Evans Hywel 2 Lois Laio 1 Cardiff University, Cardiff, United Kingdom Swansea University, Swansea, United Kingdom 01 06 2025 2025 8 4 3153 This work is licenced under a Creative Commons Attribution 4.0 International License. This article is available from the IJPDS website at: https://ijpds.org/article/view/3153 Objectives

Despite growing awareness of child criminal exploitation, there is little evidence about how to identify and respond effectively. This study explored the feasibility of linking service data with routinely collected administrative datasets for children in one Welsh local authority. The aim was to map service journeys to inform service provision.

 Methods

Data was extracted from local authority service records for the 163 children who were referred to an exploitation meeting between 2021 and 2023. This data was linked to routinely collected administrative datasets from health, education and social services, five years before referral and up to two years afterwards.

Quantitative data was analysed using Sankey diagrams which are visual representations of complex information using nodes and links between nodes. Nodes included school exclusion, living arrangements, care status and service provider. Nodes were identified through qualitative analysis of the service records and consultation with professionals from health, children’s services, police and youth justice.

 Results

Descriptive statistics showed that most children referred to an exploitation meeting were male with a median age of 16. Further analysis revealed that females were more likely to be recorded as known to child protection whereas males were recorded in a wider range of service categories, including subject to child protection, known to child protection, in need of care and support, or looked after. Information regarding ethnicity, nationality, additional learning need and missing episodes was frequently missing.

The feasibility of using latent class analysis will be discussed and Sankey diagrams will be presented to illustrate referral routes, service pathways and outcomes for sub-groups.

 Conclusion

To our knowledge, this is the first time Sankey diagrams have been used to map service and referral journeys in linked social care data. These findings are contributing to evidence-informed practice about criminally exploited children and what service pathways are more effective in relation to their circumstances.