Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The Datafication of Care in Public Homelessness Services

Erina Seh-Young Moon, Devansh Saxena, Dipto Das, Shion Guha

TL;DR

This study investigates how care is datafied in public homelessness services through an ethnographic case study in Toronto, examining how frontline workers collect and use client information within an HMIS-enabled coordinated system. Using 31 interviews and 21 observations across eight provider groups, the authors identify three care-driven objectives—matching, client privacy/agency, and equity—and show how tensions among these aims, along with spatial, technological, and staffing differences, shape data practices and potential AI use. They argue for holistic, iterative client assessments over deficit-based risk scoring, highlighting information asymmetries that can lead to biased or brittle AI models if data work is ignored. The findings offer practical guidance for designing human-centered, context-aware data systems in homelessness services, emphasizing frontline workers as advocates who should understand how data informs AI-enabled decisions and be empowered to contest adverse outcomes.

Abstract

Homelessness systems in North America adopt coordinated data-driven approaches to efficiently match support services to clients based on their assessed needs and available resources. AI tools are increasingly being implemented to allocate resources, reduce costs and predict risks in this space. In this study, we conducted an ethnographic case study on the City of Toronto's homelessness system's data practices across different critical points. We show how the City's data practices offer standardized processes for client care but frontline workers also engage in heuristic decision-making in their work to navigate uncertainties, client resistance to sharing information, and resource constraints. From these findings, we show the temporality of client data which constrain the validity of predictive AI models. Additionally, we highlight how the City adopts an iterative and holistic client assessment approach which contrasts to commonly used risk assessment tools in homelessness, providing future directions to design holistic decision-making tools for homelessness.

The Datafication of Care in Public Homelessness Services

TL;DR

This study investigates how care is datafied in public homelessness services through an ethnographic case study in Toronto, examining how frontline workers collect and use client information within an HMIS-enabled coordinated system. Using 31 interviews and 21 observations across eight provider groups, the authors identify three care-driven objectives—matching, client privacy/agency, and equity—and show how tensions among these aims, along with spatial, technological, and staffing differences, shape data practices and potential AI use. They argue for holistic, iterative client assessments over deficit-based risk scoring, highlighting information asymmetries that can lead to biased or brittle AI models if data work is ignored. The findings offer practical guidance for designing human-centered, context-aware data systems in homelessness services, emphasizing frontline workers as advocates who should understand how data informs AI-enabled decisions and be empowered to contest adverse outcomes.

Abstract

Homelessness systems in North America adopt coordinated data-driven approaches to efficiently match support services to clients based on their assessed needs and available resources. AI tools are increasingly being implemented to allocate resources, reduce costs and predict risks in this space. In this study, we conducted an ethnographic case study on the City of Toronto's homelessness system's data practices across different critical points. We show how the City's data practices offer standardized processes for client care but frontline workers also engage in heuristic decision-making in their work to navigate uncertainties, client resistance to sharing information, and resource constraints. From these findings, we show the temporality of client data which constrain the validity of predictive AI models. Additionally, we highlight how the City adopts an iterative and holistic client assessment approach which contrasts to commonly used risk assessment tools in homelessness, providing future directions to design holistic decision-making tools for homelessness.

Paper Structure

This paper contains 29 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. SIGCHI research on public sector data
  4. Data and care work
  5. HCI research on homelessness
  6. Research Context
  7. Coordinated entry and access systems in the US and Canada
  8. Site of research inquiry
  9. Methods
  10. Results
  11. Care-driven objectives shape HMIS data practices
  12. The matching objective.
  13. The client’s privacy rights and agency objective.
  14. The equity objective.
  15. Tensions between the matching objective and client's privacy rights and agency objective.
  16. ...and 14 more sections

Figures (1)

  • Figure 1: Mapping of the City's shelter system and data practices. (This is a simplified diagram of the City's homelessness system used to illustrate the coordination efforts between different groups and each groups's data practices)