Claude Occupation
EvidenceBy Fatih Kansoy
fatih.ai ↗
Validation and sensitivity

US measure validation

Cross-construct and cross-period comparison with March exposure

12rows
6fields
0country / area identifiers
CSVcomplete download
Download complete table · 0.00 MBAll datasets

What one row represents

US-month-measure comparison

Key: date_start,measure
Units: Spearman rank correlation

Preview

First 8 rows. The download contains all 12 rows.

date_startmeasuren_commonspearmancomparatorinterpretation
2025-08-04coverage7560.7060201000968739Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2025-08-04usage_weighted_intensity7560.716288095801201Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2025-11-13coverage7560.6990906648855474Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2025-11-13usage_weighted_intensity7560.7197021462384782Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2026-02-05coverage7560.6954608588324556Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2026-02-05usage_weighted_intensity7560.7192057572847531Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2026-04-01coverage7560.5985421621947229Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication
2026-04-01usage_weighted_intensity7560.5895515113332728Anthropic March2026 published observed exposureConstruct/period comparison, not exact same-sample exposure replication

Field dictionary

FieldUnitDefinitionMissingness
date_startdateSource observation-window start, not retrieval dateBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning
measuremeasure identifierNamed coverage/intensity specification; generic mapped tables also carry the published_task_coverage_proxy family label; consult quantity for unitsBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning
n_commoncountOccupation pairs with nonmissing values in both compared measuresBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning
spearmancorrelationPearson correlation of average ranks on the stated common sampleBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning
comparatorsee method/sourceField retained in table schema; see docs/method_country.md, docs/country_statistics.md or docs/piaac_validation.md for its source-specific definitionBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning
interpretationmetadataExplicit availability/quantity/source interpretation; do not infer missing values as zeroBlank/NA means unavailable or not applicable unless an explicit status field states otherwise; numeric zero retains its stated publication/source meaning

Provenance and reuse

Frozen source table: outputs/evidence/us_measure_validation.csv

Download SHA-256: 4c32ba5a442eae409bdaef85fb2ac1a03f3b137bec2bc899e9cb6131f4dda291

Component-specific source terms apply. Read the attribution and reuse terms. Missing values retain the source meaning; a zero does not override an explicit publication status.

Download manifest