Linking survey and administrative data to enhance understandings of risky behaviours and potential protective factors in children receiving social care: A feasibility study

The Student Health and Wellbeing (SHW) survey is used to monitor adolescent health behaviours to increase understanding and inform policy in Wales. It is conducted every two years and is based on the World Health Organization’s collaborative Health Behaviour of School-aged Children Survey to allow integration of the two surveys every four years.

Provisioning responses in SAIL Databank will enable researchers to link to routine administrative data to gain additional insights into the prevalence and patterns of young people’s involvement in a range of risky behaviours. It is accepted that children looked after have poorer educational and health outcomes than the general population.

However, low numbers and tendency to treat children looked after as a homogenous group is problematic. To understand the reasons for these inequalities, a more nuanced and intersectional approach is required which focuses not just on children looked after, but all children receiving care and support from social services. In this way comparisons can also be made against those who do not reach the threshold for becoming looked after.

Aims and objectives

The study will explore the benefits that can be derived from linking survey data to routine administrative data to provide an accurate picture of the prevalence and patterns of participation in risky behaviours amongst young people receiving social care support relative to their peers, and the role played by moderating factors.

From a methodological perspective, this proposed research will:

• Determine the reliability of the self-reported measures of exclusion and truanting, and that previously used to identify children looked after in foster, residential or kinship care
• Monitor changes in the responses of those who engage in risky behaviours over time and enhance understandings of the role of played by moderating factors
• Explore the utility of applying a combination of data linkage techniques and Bayesian approaches to enhance our understandings of complex issues in social care.

Research questions

1. How can our understandings of the patterns of participation in risky behaviours amongst those receiving care and support from social services be enhanced by combining routine data with survey data? What are the strengths and limitations of this approach?
2. In terms of protective factors, are there different patterns for these groups? Which factors might be amenable to intervention?
3. Are there additional risks for children receiving care and support, and specifically children looked after, after controlling for socio-economic status and neighbourhood deprivation?

Plan of investigation

The duration of the Fellowship means that 3 waves of the SHW data will be available for research, affording the opportunity to develop measures which can then be tested using subsequent waves. The initial priority will be to utilise social services and family justice records to define the cohort of interest. Once established, responses will be linked to the individual’s education and health records, and the various risky behaviours and associated protective factors explored. The order in which these are explored will be determined through a prioritization exercise with a diverse range of young people. Undertaking both cross-sectional and longitudinal linkage will enable behaviours and moderators to be explored over time and/or developmentally to establish if there are differences amongst the various groups.