Data sources and evidence policy

Occupation datasets can support titles, tasks, skills, work activities and context. Research can support bounded statements about exposure, adoption or skills.

Community discussions can reveal the words people use and the decisions they face, but comments are not treated as representative labour-market evidence.

Every claim should stay within the source’s actual scope.

O*NET is the primary US source for occupation records, alternate titles, tasks, skills, work activities, context and related occupations.

Imports record the source release and identifiers so a later update is reviewable.

Official database files and documentation are available from the O*NET Resource Center at onetcenter.org/database.html.

ESCO is the European Commission classification intended for future European and multilingual occupation mapping.

A US occupation mapping should not be silently presented as a local European equivalent. Market labels, source codes and release versions remain visible so geographic limits are clear.

Measured content comes directly from an identified source field. Derived content is calculated through documented rules. Provisional content awaits review or stronger evidence.

Blocked content cannot be published as a claim. A provisional assumption never becomes a fact because it appears in a polished interface.

Learn articles begin with a real career question, then use primary or authoritative sources for factual claims.

Each article separates “what people are asking” from its evidence list and explains how every source is used.

Reddit links provide question context only; individual comments are not generalized to workers or employers.

A source update is imported as a new immutable release, compared with the active version and reviewed before activation.

Corrections should identify the page, statement or task and an authoritative source.

Private result tokens, assessment answers and confidential work details should never be included in a correction request.