Skip to content
Back to search results

Accessibility Equity Analysis

Introduced by
Karst Geurs & Bert van Wee (accessibility evaluation framework)
Year
2004
Type
Distributional analysis of accessibility across population groups
Family
Transport equity / accessibility evaluation
Font size

What does it do?

Accessibility equity analysis asks not just how much access to opportunities a place has, but how that access is distributed across people and social groups — who can reach jobs, healthcare, and education, and who is left behind.

It pairs an accessibility measure, in the tradition formalized by Karst Geurs and Bert van Wee, with the distributional tools of inequality measurement: Lorenz curves, Gini and Palma indices, and comparisons between advantaged and disadvantaged groups. The result reframes accessibility as a question of fairness, revealing whether a transport or land-use arrangement concentrates reachable opportunity among the already privileged or spreads it equitably.

The intuition behind the method

Two cities can have the same average accessibility yet be utterly different places to live. In one, everyone enjoys decent access to jobs and services; in the other, a wealthy core can reach everything while poor peripheries can reach almost nothing. Averages hide this. Accessibility equity analysis lifts the lid by looking at the whole distribution: it lines people up from least to most accessible and draws the Lorenz curve of access, measures how unequal that curve is with a Gini coefficient, and checks whether low-income, elderly, or minority residents systematically have less access than others. The question shifts from 'how accessible is the city?' to 'accessible for whom?'

History

Concern with the distribution of accessibility grew out of transport-geography and social-exclusion research from the 1990s onward, as scholars argued that accessibility, not mobility, is the proper object of transport equity. Karst Geurs and Bert van Wee's influential 2004 review systematized accessibility components and evaluation criteria, providing the foundation on which equity assessments are built. The field absorbed distributional tools from welfare economics — Lorenz curves and the Gini coefficient — and frameworks from environmental justice and the capabilities approach, and it has since become central to transport-appraisal practice, with cumulative-opportunity job-access metrics and equity dashboards now used by planning agencies worldwide.

When to use it

Use accessibility equity analysis when the concern is fairness, not just efficiency — appraising whether a transport investment, service plan, or land-use change benefits disadvantaged groups, identifying transport-poor populations, or setting equity targets. It is well suited to comparing scenarios on distributional outcomes and to advocacy and policy that must show who gains and who loses. It is less appropriate when the underlying accessibility measure cannot be credibly calibrated, when population and socioeconomic data are too coarse or aggregated (raising the ecological fallacy), or when individual-level constraints and preferences dominate, calling for disaggregate or activity-based approaches rather than area summaries.

Limitations and considerations

  • Inherits every assumption of the accessibility measure it builds on, including its decay parameter.
  • Area-based group tags risk the ecological fallacy — neighbourhood averages are not individual experiences.
  • Gini and similar indices compress a rich distribution into one number that can hide local patterns.
  • Equity judgements depend on a normative standard (equality, sufficiency, priority) the method does not supply.

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Shifts the question from average access to its distribution, exposing who is advantaged or excluded.
  • Borrows mature, interpretable inequality tools (Lorenz, Gini, Palma) with clear policy meaning.
  • Directly supports equity appraisal of transport and land-use scenarios, not just efficiency.
  • Combines overall inequality with between-group comparisons to pinpoint targeted interventions.

Applications

  • Appraising the equity impacts of new transit lines or service changes across income groups.
  • Identifying transport-poor or service-excluded neighbourhoods for targeted investment.
  • Comparing land-use and transport scenarios on distributional, not just average, accessibility.
  • Setting and monitoring equity targets for access to jobs, healthcare, or education.
  • Supporting environmental-justice and social-inclusion assessments in planning.

How it works

How it works

1

The analysis starts from an accessibility score A for every location, computed with a standard measure — most often gravity-based accessibility that sums opportunities O weighted by a distance-decay function f of travel cost, or a cumulative-opportunity count within a time threshold. Geurs and van Wee's evaluation framework stresses that the chosen measure must be theoretically sound and behaviourally meaningful, since the entire equity assessment inherits its assumptions.

Formula
A_i = \sum_j O_j\, f(c_{ij})
2

Each location's accessibility is joined to its population and, crucially, to the socioeconomic characteristics of the people living there — income, age, car ownership, ethnicity, or disability. This produces a population-weighted distribution of accessibility in which the unit of concern is people, not places, and in which every resident is tagged with the group whose equity is in question. This join is what turns a map of access into evidence about distributional fairness.

Formula
\{(A_i, \text{pop}_i, g_i)\}_{i=1}^{n},\quad g_i = \text{group of location } i
3

The population-weighted distribution of accessibility is summarized with inequality statistics borrowed from income analysis. A Lorenz curve plots the cumulative share of total accessibility against the cumulative share of population ranked by access; the Gini coefficient measures the area between that curve and the line of perfect equality, and the Palma or percentile ratios contrast the top and bottom. These give a single, comparable read on how unequally reachable opportunity is shared.

Formula
G = \frac{\sum_i \sum_k \text{pop}_i\,\text{pop}_k\,|A_i - A_k|}{2\,\big(\sum_i \text{pop}_i\big)\sum_i \text{pop}_i A_i}
4

Beyond overall inequality, the analysis compares mean accessibility between defined groups — for example the population-weighted average access of low-income versus high-income neighbourhoods — and tests whether disadvantaged groups face systematically lower access. Mapping the gaps and decomposing inequality between and within groups identifies where and for whom interventions would most improve equity, linking the diagnosis to action.

Formula
\Delta_{gg'} = \bar{A}_g - \bar{A}_{g'},\quad \bar{A}_g = \frac{\sum_{i: g_i = g} \text{pop}_i A_i}{\sum_{i: g_i = g} \text{pop}_i}

Common pitfalls

  • Reporting only mean accessibility and inequality while never breaking results down by the groups that matter.
  • Computing a Gini on place-level scores without weighting by population, distorting the people-based picture.
  • Assuming the chosen distance-decay or threshold is neutral when it strongly shapes both access and inequality.
  • Inferring individual disadvantage from neighbourhood averages and ignoring within-area heterogeneity.

Frequently asked questions

How does accessibility equity analysis differ from plain accessibility analysis?

Plain accessibility analysis produces a score of how much opportunity is reachable from each location and typically asks how high or low that access is. Accessibility equity analysis takes those same scores but focuses on their distribution across people and groups: it weights by population, draws Lorenz curves, computes Gini or Palma indices, and compares disadvantaged with advantaged groups. The first answers 'how accessible?'; the second answers 'how fairly is accessibility shared, and for whom?'

Why use the Gini coefficient for accessibility?

The Gini coefficient is a well-understood summary of inequality, originally for income, that ranges from 0 (everyone has equal access) to 1 (all access concentrated in one place). Applied to population-weighted accessibility, it condenses the Lorenz curve into one comparable number, letting analysts rank scenarios or cities by how equally reachable opportunity is distributed. It should be paired with group comparisons and maps, because a single index can mask which groups and areas drive the inequality.

What data do I need for an equity analysis of accessibility?

You need three things: per-location accessibility scores (from a gravity, cumulative-opportunity, or floating-catchment measure built on a travel-cost matrix), population counts for the same locations, and socioeconomic attributes — income, age, car ownership, ethnicity, disability — for the groups whose equity you care about. Joining accessibility to population and group lets you weight by people and compare groups; without the socioeconomic and population layers you can describe access but not its equity.

Related methods

Sources

  1. Geurs, K. T., & van Wee, B. (2004). Accessibility evaluation of land-use and transport strategies: review and research directions. Journal of Transport Geography, 12(2), 127–140.

    Open DOI reference
    Copied.