Mishika Khurana
Selected work

Bridging Divides Migration Data Challenge · 2025 · Best Use of External Data

Steps Away, Worlds Apart

Roots & Routes: a spatial analysis of the last-mile access gap for recent immigrants in the Toronto area

Team
Andrea Santoso-Pardi, Mishika Khurana, Gabriel Dunk-Gifford
Presented
IRCC Research Matters Seminar, February 25, 2026
Methods
Index construction · OLS · geographically weighted regression · GTFS buffers
01

The last-mile gap

Recent immigrants rely heavily on public transit, and many settle in suburban neighbourhoods far from rapid transit. The “last mile”, the distance between a transit station and the places people need to reach, can turn a short trip into a long, expensive one.

  • Context

    Transit reliance

  • Context

    Suburban settlement

  • Context

    The last-mile gap

  • Why it matters

    Disproportionate transit use

  • Why it matters

    The burden of long commutes

  • Why it matters

    Inequitable access

  1. Q1Where are the most severe last-mile access gaps for recent immigrants in Toronto?
  2. Q2How do racialized and newcomer populations explain differences in transit access to employment?
  3. Q3How does spatial analysis reveal and help us understand local access inequities?
02

Data

  • BD-MDC, from Statistics Canada

    Access measures (0–1)

    • WALK_GS: walking access to grocery stores
    • WALK_HF: walking access to healthcare
    • WALK_EMP: walking access to employment
    • PUBLIC_EMP: transit access to employment

  • BD-MDC, from Statistics Canada

    Recent immigrants

    Arrivals 2016–2021, per 10,000 people.

  • Metrolinx

    GTFS

    Geolocated stops used to calculate proximity to transit and coverage buffers.

  • Unity Health Toronto & Public Health Ontario

    Ontario Marginalization Index (2021)

    • Material resources
    • Housing & dwelling
    • Age & labour force
    • Racialized & newcomers
Fig. 1Study area: City of Toronto at the aggregated dissemination area (ADA) level, with GO Rail, subway and UP Express stops.
03

Methods

  1. 01Select ADAsAreas more than 2 km from rapid transit stations
  2. 02Last Mile Gap IndexMin–max normalized access measures, station distance and recent-immigrant rate
  3. 03OLS regressionVariable selection: PUBLIC_EMP vs ON-Marg factors and recent-immigrant rate (robust Pr, R², AICc)
  4. 04Geographically weighted regressionPUBLIC_EMP vs racialized & newcomers, housing & dwelling, material resources
  5. 05Local outputsLocal coefficients, residuals and local R²
04

Where the gap is widest

Fig. 2Last Mile Gap Index. Lighter areas are well served by transit and walkable destinations; darker areas have worse access.

Downtown and East York score low

  • Higher density means more grocery stores and healthcare facilities
  • Closer to major employers
  • More investment in an improved public transit network

The gap grows with distance from the core

  • Much more car-dependent
  • Not always optimal for recent immigrants
  • Over-reliance on GO Transit and buses
  • More investment needed to serve newer residents
05

Marginalization and access to jobs

Fig. 3GWR local coefficients for the racialized & newcomer factor. Teal: more marginalization, better transit access to jobs; purple: worse.

Negative coefficients in Scarborough, North York and northern Etobicoke mean more marginalization goes with worse transit access to employment. Downtown and in East York the relationship flips. These local patterns match the Last Mile Gap Index in the key underserved areas.

Fig. 4Standardized residuals: overall model fit is good across most of the city.
06

Implications & next steps

  • Transit access is more than infrastructure

  • Solutions must be equity-focused and place-based

  • Safeguarding against displacement

Next steps

  • Further analysis of commute times, affordability and displacement risk
  • Employment access by time of day
  • Accessibility for seniors, women and people with disabilities
  • Population density and crowding on transit