Using administrative data to understand the service interactions of people experiencing homelessness

Abstract

It is recognized that people experiencing homelessness have complex trajectories through multiple agencies, including emergency health and homelessness services. Though there is a large body of quantitative evidence from the United States and some Scandinavian countries on the service interactions of people experiencing homelessness, the evidence base in the UK is mainly limited to qualitative studies and small scale surveys. This paper therefore addresses this evidence gap, by presenting initial findings from a longitudinal linked administrative data study, linking together data on people seeking assistance with homelessness in a single local authority in Wales along with routinely collected data from public services. In addition to exploring the overall patterns of service use by people experiencing homelessness, latent class analysis was used in order to generate a typology of service experience, in acknowledgement that not all homeless people are the same. Due to the movement of people through multiple institutional systems, administrative data are an ideal data source for researching this aspect of homelessness. It is hoped that through the creation of a typology of service, this can be used to provide a greater understanding of Welsh homelessness, thereby allowing for a more effective targeting of services and improving the pathways of people experiencing homelessness through public services.