Geary's C¶

Measure global spatial autocorrelation by comparing weighted squared differences between neighboring observations with overall variance.

Version: v1 · 2026-09-08 · History
Domain-specific #: 4675
Origin domain: spatial statistics
Subdomain: global spatial autocorrelation

Core Idea¶

Geary's C is a global spatial autocorrelation statistic formed from weighted squared differences across spatially linked observations, normalized by total variance.^[1] Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

The load-bearing residual is not the broad topic of spatial statistics. It is the exact Geary normalization and neighbor-difference sensitivity, not spatial correlation generally. That residual remains recognizable when examples, notation, scale, or implementation change, but it disappears if weights are hidden, values are not aligned to spatial units, C below one is called correlation without inference, or local Geary statistics are substituted for the global coefficient. This gives the entry an operational identity rather than merely a historical label.

A useful analysis keeps three layers separate. The constitutive layer says what must be true: C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares. The evidential layer asks what observation or proof warrants the claim: publish the weights matrix and normalization, check zero variance and islands, compute numerator and denominator consistently, and use an explicit randomization or asymptotic null for inference. The use layer asks what reasoning becomes available once the identity is established: detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic. Conflating the layers is the most common source of scope inflation.

Structural Signature¶

Carrier: values observed on spatial units plus a declared spatial weights matrix with zero diagonal
Inputs or antecedent state: N observations, their mean and variance, weights w_ij, weight normalization, S0, neighborhood convention, null model, and inference method
Constitutive operation: Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one.
Invariant: C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares
Recognition test: publish the weights matrix and normalization, check zero variance and islands, compute numerator and denominator consistently, and use an explicit randomization or asymptotic null for inference
Output or consequence: detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic
Failure boundary: weights are hidden, values are not aligned to spatial units, C below one is called correlation without inference, or local Geary statistics are substituted for the global coefficient

What It Is Not¶

It is not the whole field of spatial statistics. The field contains many questions and methods that do not instantiate Geary's C.
It is not its most familiar example. A checkerboard of alternating high and low neighboring values produces large squared neighbor differences and a C above the no-association reference. exhibits the structure, but the example is evidence for the abstraction rather than its definition.
It is not the neighboring catalog concept Correlation. Correlation is a broad association Prime; Geary's C fixes spatial weights, a global variance baseline, and squared neighbor differences.
It is not a claim that every boundary case has one uncontested classification. a qualified variant may preserve the core while changing notation, parameterization, or implementation, so the constitutive condition must decide the boundary
It is not an unrestricted metaphor for any process that seems similar. Outside spatial statistics, the vocabulary and validity conditions do not transfer literally.

Scope of Application¶

Geary's C belongs to spatial statistics and is useful where the analyst can specify values observed on spatial units plus a declared spatial weights matrix with zero diagonal, then evaluate C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares. The scope is broad within that domain but bounded by the need for C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares. The entry records a descriptive analytical identity; practical use requires the governing domain's evidence, standards, and safety obligations.^[2]

Definition and recognition. Determine whether a proposed instance satisfies the constitutive conditions rather than merely sharing terminology.
Construction or evolution. Track how N observations, their mean and variance, weights w_ij, weight normalization, S0, neighborhood convention, null model, and inference method are converted, constrained, or organized by Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one..
Comparison. Compare instances using carrier, defining parameters, convention, scale, scope, evidence, limiting cases, and implementation, without treating convenience measures as the definition.
Boundary analysis. Diagnose cases where a qualified variant may preserve the core while changing notation, parameterization, or implementation, so the constitutive condition must decide the boundary and state which convention or theorem controls the decision.
Downstream reasoning. Use the established identity to support detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic while preserving the assumptions under which the inference is valid.

Clarity¶

The abstraction clarifies a crowded vocabulary by making C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Geary's C can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated. The disciplined statement is: given N observations, their mean and variance, weights w_ij, weight normalization, S0, neighborhood convention, null model, and inference method, the structure counts as Geary's C exactly when C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares.

This format also separates identity from measurement. Empirical, computational, or documentary proxies support recognition only under declared validity and uncertainty assumptions; formal cases require proof rather than measurement. Measurements can be noisy, implementations can approximate, and proofs can use equivalent characterizations; none of those facts licenses changing the object being measured. When reports disagree, first check scope and convention, then data or proof, and only then interpret the disagreement as substantive.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Geary's C. Geary's C compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

The compression has a price. A single label can hide standard, generalized, restricted, approximate, computational, and historically variant formulations of Geary's C. Good use therefore carries a small declaration of assumptions alongside the name. The abstraction manages complexity when it reduces the state space of the question while keeping the failure boundary visible; it mismanages complexity when the label substitutes for that boundary analysis.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: values observed on spatial units plus a declared spatial weights matrix with zero diagonal. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares independently of one notation or implementation. This step prevents the canonical example from becoming the definition.
Derive consequences. From C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares, infer detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic. Record each assumption used so that a later change of setting does not silently preserve an invalid conclusion.
Test adversarial cases. Examine a qualified variant may preserve the core while changing notation, parameterization, or implementation, so the constitutive condition must decide the boundary and a visually smooth map does not establish significant Geary autocorrelation when the weights or null distribution differ. A robust identity explains why the first is convention-sensitive and why the second is outside the class.
Compare and refine. Use carrier, defining parameters, convention, scale, scope, evidence, limiting cases, and implementation to compare legitimate instances, and refine the model when discrepancies reflect hidden variation rather than failure of the abstraction itself.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of spatial statistics because they reuse values observed on spatial units plus a declared spatial weights matrix with zero diagonal, Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one., and publish the weights matrix and normalization, check zero variance and islands, compute numerator and denominator consistently, and use an explicit randomization or asymptotic null for inference. A theorem, diagnostic, or modeling warning can travel when those roles remain literal. For example, the distinction between constitutive identity and a convenient observable transfers from A checkerboard of alternating high and low neighboring values produces large squared neighbor differences and a C above the no-association reference. to An analyst computes C for regional rates under queen-contiguity weights and assesses significance by permuting values over regions..^[3]

Transfer outside the home domain is weaker. The skeletal pattern—type a carrier, apply a constitutive relation, preserve its invariant, and derive only qualified consequences—may suggest an analogy, but the domain-specific mechanisms, admissible evidence, and consequences do not come along automatically. The safe transfer procedure maps each role explicitly, checks the invariant again, and refuses the name when only a superficial resemblance remains.

Examples¶

Canonical¶

A checkerboard of alternating high and low neighboring values produces large squared neighbor differences and a C above the no-association reference. The same marginal values clustered into contiguous regions reduce neighbor differences and move C below one. This example is canonical because every role can be inspected: the carrier is values observed on spatial units plus a declared spatial weights matrix with zero diagonal; the operative rule is Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one.; the invariant is C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares; and the result supports detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic.^[1] Changing incidental notation or scale leaves the structure intact, while removing C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares destroys the classification.

Mapped back: values observed on spatial units plus a declared spatial weights matrix with zero diagonal → Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one. → C is computed from the declared weights and the ratio of weighted pairwise squared differences to centered global sum of squares → detecting global spatial association, comparing maps and weight definitions, and complementing Moran's I with a difference-sensitive statistic

Applied / In Practice¶

An analyst computes C for regional rates under queen-contiguity weights and assesses significance by permuting values over regions. The conclusion is conditional on the chosen geography, rate construction, weights, and permutation null. The applied case is not licensed merely by vocabulary. It qualifies because the same recognition test—publish the weights matrix and normalization, check zero variance and islands, compute numerator and denominator consistently, and use an explicit randomization or asymptotic null for inference—can be run and because the same failure boundary—weights are hidden, values are not aligned to spatial units, C below one is called correlation without inference, or local Geary statistics are substituted for the global coefficient—remains meaningful.^[2] The case also shows why practical outputs should report assumptions, resolution, and uncertainty instead of a naked label.

Mapped back: declared instance → recognition test → boundary check → qualified use

Structural Tensions¶

T1: Axiomatic identity vs. operational recognition. The defining conditions may be exact while empirical or computational recognition is approximate. Neither pole can be removed without changing the analytical task. Diagnostic: Can the reviewer state both the exact condition and the evidence used to infer it?
T2: Local roles vs. global consequence. The mechanism is enacted through local relations, but the abstraction is usually valued for a global classification or prediction. Neither pole can be removed without changing the analytical task. Diagnostic: Does the claimed global result actually follow from the declared local conditions?
T3: Ideal form vs. finite representation. Theory states a clean invariant while data structures, measurements, or proofs expose only finite representations. Neither pole can be removed without changing the analytical task. Diagnostic: Would increasing resolution converge toward the same classification?
T4: Canonical convention vs. legitimate variants. A standard formulation supports communication, while variants may preserve the same core under changed assumptions. Neither pole can be removed without changing the analytical task. Diagnostic: Which role is invariant across variants, and which convention-specific conclusion changes?
T5: Compression vs. hidden assumptions. The name compresses a complex argument but can conceal prerequisites. Neither pole can be removed without changing the analytical task. Diagnostic: Can each downstream inference be traced to an explicit assumption?
T6: Autonomous residual vs. reduction to catalog neighbors. The candidate uses broader structures but adds an identity-bearing residual. Neither pole can be removed without changing the analytical task. Diagnostic: After subtracting the proposed parent and named neighbors, does the constitutive residual still support independent diagnostics?

Structural–Framed Character¶

The entry is structurally mixed but domain-framed. Its portable skeleton is type a carrier, apply a constitutive relation, preserve its invariant, and derive only qualified consequences. Its identity-bearing terms—Geary's C, carrier, parameter, relation, invariant, boundary, evidence, and application—derive their meaning from spatial statistics and cannot be replaced by generic systems language without losing the tests that distinguish valid from invalid instances.

This mixed character explains why the abstraction is reusable inside the domain yet does not meet the Prime bar. The structure organizes reasoning, but its claims still depend on domain-specific objects, evidence, and intervention semantics.

Structural Core vs. Domain Accent¶

The structural core consists of a carrier, Neighbor differences contribute to the numerator, so similar adjacent values lower C and dissimilar adjacent values raise it relative to a random-spatial-association reference near one., a recognition invariant, and a consequence. That skeleton may resemble patterns elsewhere, especially type a carrier, apply a constitutive relation, preserve its invariant, and derive only qualified consequences. The domain accent is not decorative: Geary's C, carrier, parameter, relation, invariant, boundary, evidence, and application determine what counts as an admissible carrier, a valid transition, and successful evidence.

The abstraction therefore remains domain-specific. A cross-domain reuse that preserves only words such as 'balance,' 'cut,' 'sequence,' 'loss,' or 'simulation' is metaphor. Literal transfer requires the original role structure and diagnostics, which in this case remain anchored in spatial statistics.

The proposed strict upward parent is prime:correlation. The statistic literally measures association between values linked in space; its weights and difference-ratio formula supply the residual. This is a proposal-only workspace relationship: the accepted Prime supplies a genuinely instantiated structural prerequisite or superclass, while Geary's C adds domain-specific constraints.

The entry does not collapse into that parent because the exact Geary normalization and neighbor-difference sensitivity, not spatial correlation generally It also declines a nearby thematic catalog node: the neighbor does not literally subsume the constitutive identity of Geary's C. This explicit assert-and-decline pattern keeps the proposed DAG narrow and prevents a merely thematic edge.

The prospective workspace queue contains one strict upward edge to prime:correlation. No live DAG mutation is authorized.

Relationships to Other Abstractions¶

Current abstraction Geary's C Domain-specific

Parents (1) — more general patterns this builds on

Geary's C is a kind of Correlation Prime

The proposed strict upward parent is prime:correlation.

Hierarchy path (1) — routes to 1 parentless root

Geary's C → Correlation

Neighborhood in Abstraction Space¶

Geary's C sits in a moderately populated region (56^th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Multivariate & Spatial Statistics (13 abstractions)

Nearest neighbors

Spatial Analysis of Principal Components — 0.91
Moran's I — 0.89
Euclidean random matrix — 0.87
Spatial distribution — 0.87
Spatial heterogeneity — 0.87

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

Moran's I. Uses centered cross-products and has different sensitivity and scaling.
Local Geary. A local indicator used to flag particular spatial units.
Semivariance. A distance-lag function rather than this globally normalized statistic.
Ordinary autocorrelation. Uses an ordered lag without a general spatial weights matrix.
Spatial regression coefficient. Models conditional relationships rather than a single global map statistic.

References¶

[1] R. C. Geary, ‘The Contiguity Ratio and Statistical Mapping,’ The Incorporated Statistician 5(3), 115–145 (1954), DOI 10.2307/2986645. registry ↩a ↩b

[2] A. D. Cliff and J. K. Ord, Spatial Processes: Models & Applications, Pion, 1981, ISBN 0-85086-081-4. registry ↩a ↩b

[3] Luc Anselin, ‘A Local Indicator of Multivariate Spatial Association: Extending Geary's c,’ Geographical Analysis 51(2), 133–150 (2019), DOI 10.1111/gean.12164. registry ↩