Welcoming Infrastructures: Designing for Accountability in the Settlement Service Work in Canada (2024)

This dissertation uncovers a dual accountability regime in Canadian settlement services, where relational, empathy‑driven practices coexist, and often clash, with rigid, number‑focused reporting demands. The work exposes a temporal bias that privileges speed and quantification, shaping data cultures, technology choices, and ultimately the quality of support offered to newcomers. By foregrounding workers’ lived experiences, the study offers concrete pathways for policy reform, organizational redesign, and technology development that could make settlement services more humane, effective, and truly accountable.

What is this research about?

Abstract (quoted)
“Immigration and settlement policies, much like organizational policies or digital technology policies, are built from the coming together of discourses, tools, and people—people standing behind service desks, using databases and making decisions. These interdependent systems, beliefs, and tools affect how decisions are shaped as well as the information and data cultures of an organization (or a whole sector). Digital technologies and commitments to evidence‑based, accountable service delivery are critical to the capacity of settlement organizations to manage data flow. The dissertation addresses these questions through a multi‑sited study with the settlement organizations that serve refugees and immigrants in Canada, a country often recognized as a global leader in immigration and settlement policies.”

Goal & Guiding Questions
The dissertation aims to uncover how information practices shape and mediate accountability mechanisms in Canadian immigrant‑settlement services. It is driven by three research questions:

1. What does accountability mean for settlement workers and their organizations?
2. Which practices and systems (e.g., databases, reporting tools, outreach methods) underpin workers’ approaches to accountability?
3. What are the implications of these practices and systems for the work of settlement organizations and for the people they serve?

What do you need to know? – Context & Significance

Point	Explanation
Policy pressure	Most settlement organizations are funded by the federal government (IRCC) and must report detailed service data to demonstrate “accountability” for public money.
Digital transformation	Recent years have seen a rapid shift to digital service delivery (e‑mail, Zoom, mobile apps) and the introduction of mandated databases such as iCARE.
Knowledge gap	While much research has examined immigrants’ information‑seeking behaviour, little has looked at the information practices of the frontline workers who collect, organize, and report that data.
Temporal bias	The author introduces the concept of temporal bias – a cultural tilt that privileges speed, clock‑time, and easily quantifiable data over slower, relational, narrative work.
Unique approach	The study blends practice theory, social theories of time, and critical data studies to analyze both the sociotechnical and temporal dimensions of accountability. It combines a large‑scale qualitative interview program (32 workers, 15 clients) with document analysis of 30+ sector reports, policy papers, and artefacts.

These points locate the work at the intersection of information science, migration studies, and organizational sociology, offering a fresh lens on how “accountability” is lived and negotiated on the ground.

What did the researchers find? – Key Highlights & Illustrative Quotes

Dual notions of accountability

• Relational accountability – “accounting for settlement experience” through empathy, lived experience, and active listening.
  Quote: “I draw on my own family’s immigration story to understand what the client is going through” – Dalal, worker.
• Formal accountability – “accounting for the numbers” to satisfy funder requirements (client counts, service outputs).

Organizational data culture

• Values of data – transactional (funding), communicative (evidence for policy), and knowledge‑building (identifying trends).
• Divergent attitudes – Managers view data as a strategic resource; frontline workers see data entry as an extra, often meaningless task.
  Quote: “We report the cases daily… but nothing comes out of it. It’s just for show” – Negar, worker.

Socio‑technical realities

• Multiple, non‑interoperable databases (iCARE, program‑specific spreadsheets) force double‑entry and “patchworking.”
• Lack of IT support; workers learn on the job, often via YouTube tutorials.

Temporal bias & its consequences

• Speed‑over‑depth: Rapid reporting pushes workers to capture only quick, quantifiable metrics, sidelining narrative data and “returning clients.”
  Quote: “If a client shows up again, the system can’t capture that ongoing support” – Laila, worker.
• Impact on service quality – Workers feel pressured to place clients in temporary jobs to meet targets, sometimes at odds with clients’ long‑term goals.
  Quote: “We were told to count placements, so we push a refugee into a factory job even if it’s not a fit” – Ahmad, client.

Outlier / surprising findings

• Consent‑form friction – Forms are only in English; workers must translate on the spot, creating legal‑risk discomfort for clients.
• Success‑story harvesting – Organizations deliberately solicit “success stories” for funder reports, even when not explicitly required.
• Silence from funders – Workers rarely receive feedback on the data they submit; the flow of information is largely one‑way.

Interesting Themes & Outlier Findings

Theme	Why it matters
Empathy as formal accountability	Shows that caring practices can be framed as “accountable” actions, challenging the purely numeric view of performance.
Temporal bias (Chronos vs. Kairos)	Highlights a structural tension: clock‑time reporting vs. the “right moment” needed for trust‑building with clients.
Data drift	Managers adjust data collection to anticipated funder expectations, even without explicit directives – a subtle form of governance.
Gatekeeping & time‑keeping	Workers act as gatekeepers of information and of time, deciding which client needs are urgent enough to be recorded.
Technology‑induced inequities	Digital tools (e.g., Zoom, WhatsApp) improve reach but also create barriers for clients lacking devices or digital literacy.
Non‑use of data	Despite massive data collection, organisations rarely analyse the data for internal improvement; it remains a reporting artefact.

How can you use this research?

For Policy Makers / Funders (IRCC, provincial ministries)

• Action: Co‑design feedback loops that return analyzed data to settlement organizations, closing the one‑way reporting gap.
• Action: Revise funding metrics to recognize qualitative outcomes (e.g., client wellbeing, long‑term employment) alongside headcounts.
• Action: Provide clear, multilingual consent templates and guidance to reduce legal risk for frontline staff.

For Settlement Organization Leaders & Managers

• Action: Develop human‑centered data dashboards that surface both quantitative and narrative indicators, enabling staff to see the impact of their relational work.
• Action: Invest in interoperable IT infrastructure (single CRM) and dedicated tech support to curb double‑entry and spreadsheet proliferation.
• Action: Offer regular training on data ethics, digital tools, and temporal management (balancing reporting deadlines with client rapport).

For Frontline Settlement Workers

• Action: Adopt just‑in‑time data collection: record only what is needed for the immediate service episode, and flag richer narrative data for later synthesis.
• Action: Use peer‑sharing circles to exchange shortcuts, tutorials, and best practices for navigating multiple databases.
• Action: Advocate for recognition of relational work in performance reviews (e.g., client trust scores, empathy metrics).

For Researchers & Academics

• Future‑research recommendation: Conduct time‑use studies (e.g., activity logs) to quantify the exact proportion of time spent on reporting vs. direct client work.
• Future‑research recommendation: Explore design interventions (e.g., mixed‑method dashboards) that surface both quantitative and qualitative accountability data.
• Future‑research recommendation: Compare the Canadian context with other high‑immigration nations to see if temporal bias is a universal phenomenon.

For Technology Designers / Vendors

• Action: Build modular, interoperable settlement platforms that integrate with iCARE but also allow custom fields for narrative notes.
• Action: Incorporate multilingual UI and offline‑first capabilities to support clients and workers with limited connectivity.

What did the researchers do? – Methodology Overview

Component	Details
Design	Qualitative, exploratory case study using practice theory and social theories of time.
Data sources	Semi‑structured interviews – 32 settlement workers (mix of front‑line staff, managers, coordinators) and 15 immigrant clients.Document & artefact analysis – 30+ sector reports, policy papers, conference proceedings, internal manuals, consent forms, database screenshots.
Sampling	Purposive + snowball sampling across Ontario, Alberta, Manitoba, Saskatchewan, British Columbia; aimed for diversity in gender, age, ethnicity, tenure, and organizational size.
Demographics (workers)	35 % male, 65 % female; ages 25‑64; experience 1‑31 years; roles ranged from settlement workers, managers, digital navigators, administrative assistants.
Demographics (clients)	12 % male, 88 % female; ages 18‑35; mix of refugees, permanent residents, recent arrivals (2019‑2022).
Analysis	Reflexive Thematic Analysis (Braun & Clarke, 2022) – six‑step process: familiarization, coding (NVivo), theme generation, review, definition, write‑up. Theoretical lens (practice theory, temporality) guided coding of concepts such as “accountability,” “data culture,” “temporal bias.”
Ethics	Approved by University of Toronto REB; informed consent obtained; pseudonyms used; data stored securely.
Limitations noted	Pandemic‑forced shift to remote interviews; non‑probability sampling limits generalizability; urban‑centric sample (few rural participants).