Separate work, author, organisation, repository, journal and trial to cite the right object and version.
AI-generated illustration
Choose the right source
Start with the object’s identifier: DOI or record ID for a work, ORCID for a person, ROR for an organisation, NCT for a trial. Then connect document and editorial status. Inclusion in an index, repository or registry does not independently validate the result.
Compare result types
Crossref and DataCite describe identified objects; PubMed and Europe PMC index literature; HAL and Zenodo host deposits. For OSF, distinguish Projects, Registrations and Preprints: the announced transition concerns Projects. ORCID and ROR help resolve identities. Multiple records can repeat the same metadata.
Reading example
A preprint and published article have similar titles. Read the relationship and versions, compare the passage used and check corrections or retractions. Do not count two versions of one work as independent studies.
Registered trial, protocol and results available separately
Document or deposit version, publication and modification dates.
Choose the check for your case
Each question links comparable sources to a record to retain and a method. Services have different scope; open their profiles before drawing conclusions.
How can an author’s identity and affiliation be distinguished?
Keep person and organisation identifiers separately. Date the affiliation and check its documentary link; an earlier affiliation does not establish current employment.
How can two records be matched without deleting a version?
Compare identifier, object type, title and version relationships. A preprint, article and dataset may be linked without being bibliographic duplicates.
Open the publication and indicated updates. Indexing, open access and scientific findings are distinct; check the version and available correction or withdrawal notices.
How can a trial be linked to a results publication?
Compare trial identifier, protocol, intervention, population and dates. Link documents without treating each record or article from one trial as an independent study.
Consult OSF projects, preregistrations and preprints, identifying resource type and version. The announced Projects transition starts on 16 November 2026; see the documented schedule in the profile.
Distinguish a project file from a Registration. Announced preservation of content and identifiers establishes neither protocol adherence nor result quality.
Discover ongoing and completed trials indexed from multiple registries in ICTRP.
Return to the source registry and trial ID. Two records from one trial are not two studies; registration does not establish published results or effectiveness.
A category includes close services, but their content can answer different questions. The useful combination covers the dimensions that the first source leaves open.
Updated on .
Priority checks
Compare territory, period, identifiers and document types. Look for a source that adds context or chronology, rather than another repetition of the same result.
Interpretation pitfalls
Two sources of the same category may have a common origin. Their complementarity must be demonstrated by the information they provide.
Working record to retain
The source map combines question, service, obtained information and remaining gap. It makes visible the role of each consultation in the reasoning.