Data Quality Dashboard

Quality metrics across the catalogue: completeness, timeliness, and documentation

Total Datasets
95
Curated July 2024
Avg Completeness
100.0%
Metadata field coverage
Current Data
95/95
Updated within 1 year
Avg Documentation
99.7%
Field defs, provenance, limits

Quality Dimensions

Completeness: Proportion of expected metadata fields present (name, domains, licence, temporal coverage, file count, README path). Deductions applied for empty domains or missing licence text.
Timeliness: Days since last data update, inferred from temporal coverage. Datasets with data from current year or previous year are considered "current".
Documentation: Presence of README (field definitions), licence documentation (provenance), and assumed known limitations. Score: 0.33 per component.

Top 10 by Completeness

DatasetCompletenessTimelinessLast Updated
Au Alcohol Other Drug Treatment Services
100%
Current2026-12-31
Au Ambulance Services Performance
100%
Current2026-12-31
Au Births Deaths Marriages Registrations
100%
Current2026-12-31
Au Building Approvals
100%
Current2026-12-31
Au Child Protection Services
100%
Current2026-12-31
Au Childhood Immunisation Coverage Rates
100%
Current2026-12-31
Au Consumer Credit Licensing
100%
Current2026-12-31
Au Coroners Court Statistics
100%
Current2026-12-31
Au Cost Of Living Cpi
100%
Current2026-12-31
Au Criminal Courts Sentencing Outcomes
100%
Current2084-12-31

Statistical Note

Quality metrics are computed from catalogue metadata only (not the underlying data files). Completeness scores reflect metadata coverage; they do not validate data accuracy or field-level completeness within files.

Sample size: n = 95 datasets. Average completeness score: 100.00% (95% CI: 100.00–100.00%) (95% CI via bootstrap with 10,000 resamples, seed 20240710).