Indicator Catalogue

Normalisation Methods

All indicators are normalised across the full set of Australian postcodes before aggregation. Two methods are used:

• Min-max: (value - min) / (max - min), scaled to 0–1 (BRIC) or 0–10 (INFORM). Default method.
• Percentile-rank: percentile_rank(value) / 100. Used when skewness exceeds 2.0 across all postcodes, to prevent outlier distortion.

The method applied to each indicator is determined automatically at build time based on distribution shape. For “lower is better” indicators, the normalised value is inverted so that higher normalised scores always indicate greater resilience (BRIC) or greater crisis pressure (INFORM).

Aggregation

• Within each BRIC capital:weighted arithmetic mean of normalised indicators. Range 0–1.
• BRIC composite:sum of six capital scores. Range 0–6 (Cutter et al. 2010 convention). When fewer capitals are available, the sum is proportionally scaled: BRIC = (sum / n_available) × 6.
• Within each INFORM pillar:weighted arithmetic mean of normalised indicators. Range 0–10.
• INFORM composite:geometric mean of three pillar scores. Range 0–10 (INFORM convention). Geometric mean prevents full compensation across pillars.
• Cross-layer: no aggregation. Separate BRIC and INFORM scores with quadrant classification (ADR-002).

Social Capital (REQ-001)

Economic Capital (REQ-002)

Industry diversity is weighted at 0.30 (above the 0.20 baseline) per ADR-003. Economic monoculture is a brittleness signal — the coherence/entrainment distinction applied to employment structure.

Community Capital (REQ-003)

Organisational type diversity is weighted above voter turnout per ADR-003. Diverse community infrastructure signals coherence.

Institutional Capital (REQ-004)

Institutional capital is the hardest to source at postcode level. MVP uses distance-based proxies. Confidence for this capital is lower than for census-derived capitals.

Housing & Infrastructure Capital (REQ-005)

Transport mode diversity is weighted at 0.30 per ADR-003. A community where 90% drive to work is entrained to fuel supply — fragile in a supply chain crisis.

Environmental Capital (REQ-006)

Largely deferred to v2 due to data availability. MVP may score this capital with reduced confidence or flag as “data pending.”

Exposure Pillar (REQ-008)

Sensitivity Pillar (REQ-009)

Lack of Coping Capacity Pillar (REQ-010)

All indicators in this pillar measure capacity (higher raw value = better). They are inverted during normalisation so the pillar score reflects lack of coping capacity, consistent with the INFORM convention.

Signal Confidence Metadata

Every indicator carries a SignalMetaobject recording its source, authority level, data freshness, and geographic coverage. A composite confidence score (0–1) is computed per indicator using the following weights:

Capital and pillar confidence scores are the weighted average of their constituent indicators’ confidence scores. When indicators are missing, confidence drops proportionally.

Missing Data Handling

1. If a postcode is absent from a dataset, the indicator value is NULL.
2. NULL indicators are excluded from the weighted mean for their capital or pillar.
3. Remaining indicator weights are renormalised to sum to 1.0.
4. The capital/pillar confidence score is reduced proportionally.
5. If more than 50% of indicators in a capital/pillar are missing, the component is flagged as “insufficient data” rather than computed from sparse inputs.