Automated quality control

Data Audit

The website audits canonical files at build time for schema gaps, duplicate keys, country coverage, year coverage, missingness and unexpected provenance labels.

Data pipeline readyAdd the two canonical master CSVs to data/raw/. The site will populate automatically on the next build.Data status →

Latest schema

Missing: iso3, country, asof_year, identity_z, identity_level, identity_rank, identity_20y, identity_3y, identity_source, drive_z, drive_level, drive_rank, drive_20y, drive_3y, grit_z, grit_level, grit_rank, grit_20y, grit_3y, combo_avg_z, combo_avg_level, combo_avg_rank, combo_avg_20y, combo_avg_3y, combo_mult, combo_mult_rank, combo_mult_20y, combo_mult_3y, identity_core_z, identity_core_provenance, identity_fracture_z, openness_exposure_z, openness_exposure_provenance

Latest duplicates

No duplicate ISO3 keys detected.

Historical duplicates

No duplicate country-year keys detected.

Key-field missingness

Latest fieldMissingHistorical fieldMissing
identity_zidentity
drive_zdrive
grit_zgrit
combo_avg_zcombo_avg
identity_fracture_zidentity_fracture
openness_exposure_zopenness_exposure

Identity provenance — latest

{}

Openness provenance — historical

{}
Audit scope: passing these checks means the files match expected structural rules. It does not prove the underlying source data or model is substantively correct; statistical validation remains separate.