eduKateSG · Why Science?
A DOI becomes more useful when its research object is clearly described
Read DataCite DOI records through resource type, creators, dates, rights, versions, parts and typed relationships so research objects stay findable and citable.
Science now produces more than journal articles. Datasets, software, workflows, computational notebooks, images, posters and presentations can each become citable research objects when repositories register persistent identifiers and useful metadata. DataCite provides DOI infrastructure and a metadata schema designed for research data and other outputs.
The DOI identifies a resource through a resolvable route, while the metadata explains what the resource is. DataCite Metadata Schema 4.7, released on 3 March 2026, expanded resource types and relationship metadata. That date matters because controlled vocabularies evolve and a reader should consult the schema version used by a record or workflow.
Did you know? Two DOI records can describe versions or parts of one broader output without being interchangeable. DataCite relation types such as HasVersion, IsVersionOf, IsNewVersionOf, IsPreviousVersionOf, HasPart and IsPartOf preserve those directions.
This guide treats a DataCite record as structured provenance. Resolving a DOI does not prove that a dataset is complete, that software runs, or that a conclusion is correct. Those questions still require files, documentation, methods and domain review.
Chapter index
Chapters 1–9: DOI identity, creators and resource types
Chapters 10–17: versions, parts, rights and API retrieval
A persistent identifier is useful when the metadata explains exactly what it identifies. A DOI is designed to resolve to a landing page even when the underlying web location changes. The landing page should describe and provide access to the research object. This makes citation literacy a practical part of science learning.
Try this
Evidence question: Which exact resource does this DOI identify? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Resolve one DOI and record the landing-page owner, title and resource type. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for the doi is a persistent route, its identifier or version context, and one limit that must remain visible.
A DOI route can stay stable even while the resource location changes. DataCite members and repositories register DOIs and maintain the metadata associated with them. DataCite exposes that metadata but does not author every scientific claim in each resource. It also keeps identifier resolution separate from scientific validation.
Try this
Evidence question: Which repository is responsible for this record? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Separate registration-agency, repository and creator roles. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for the repository registers the record, its identifier or version context, and one limit that must remain visible.
Relationship types turn isolated records into a navigable research graph. The schema uses a general resource type and can include a more specific free-text description. Dataset, software, preprint, report and computational notebook are different evidence objects. The record becomes more useful when creators, dates, types and rights remain visible.
Try this
Evidence question: Does the general type match the files and description? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Write the general and specific types on separate lines. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for resource type frames the object, its identifier or version context, and one limit that must remain visible.
Version and part labels keep a broad research object from swallowing its precise evidence. Creator names support citation, while contributor roles can record people or organisations involved in curation, hosting, supervision or other work. Identifiers such as ORCID can reduce name ambiguity. A careful reader can preserve identity, context and limits in one compact receipt.
Try this
Evidence question: Who created the object and who supported it? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Make a role table without turning every contributor into an author. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for creators and contributors have roles, its identifier or version context, and one limit that must remain visible.
A persistent identifier is useful when the metadata explains exactly what it identifies. A title is a searchable label rather than a full methods description. Alternative titles or translated titles may add context. This makes citation literacy a practical part of science learning.
Try this
Evidence question: Which words describe the object, and which important method is absent? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Compare title terms with the description and files. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for titles help discovery, its identifier or version context, and one limit that must remain visible.
A DOI route can stay stable even while the resource location changes. Publication year supports citation, while collected, created, updated, available or issued dates answer different questions. A year should not be moved between roles. It also keeps identifier resolution separate from scientific validation.
Try this
Evidence question: Which event does each date describe? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Build a timeline with the date type beside every value. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for dates need date types, its identifier or version context, and one limit that must remain visible.
Relationship types turn isolated records into a navigable research graph. In DataCite metadata, publisher commonly names the entity that holds, archives, publishes, prints, distributes, releases, issues or produces the resource. It may differ from a journal publisher. The record becomes more useful when creators, dates, types and rights remain visible.
Try this
Evidence question: Who makes this research object available? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Record the publisher field without importing assumptions from articles. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for publisher means the holding entity, its identifier or version context, and one limit that must remain visible.
Version and part labels keep a broad research object from swallowing its precise evidence. Abstracts, methods, technical information and other descriptions can explain scope. The description remains metadata and should be checked against documentation and files. A careful reader can preserve identity, context and limits in one compact receipt.
Try this
Evidence question: Which description type supports the current question? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Label one sentence as scope, method or interpretation. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for descriptions have purposes, its identifier or version context, and one limit that must remain visible.
A persistent identifier is useful when the metadata explains exactly what it identifies. Subject terms can use free text or controlled vocabularies. Similar keywords do not guarantee equivalent measurements or populations. This makes citation literacy a practical part of science learning.
Try this
Evidence question: Which vocabulary or scheme supplied the subject? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Trace a subject term to its scheme when one is named. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for subjects improve discovery, its identifier or version context, and one limit that must remain visible.
A DOI route can stay stable even while the resource location changes. Rights metadata can include a licence name, URI and identifier. Open access does not mean every use is unrestricted or scientifically suitable. It also keeps identifier resolution separate from scientific validation.
Try this
Evidence question: Which permission, attribution or restriction applies? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Translate the live rights statement into a short reuse note. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for rights define reuse conditions, its identifier or version context, and one limit that must remain visible.
Relationship types turn isolated records into a navigable research graph. A relatedIdentifier pairs another identifier with a directional relation type. It can connect papers, data, software, instruments, projects and other resources. The record becomes more useful when creators, dates, types and rights remain visible.
Try this
Evidence question: What are the source and target of this relation? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Read the relation aloud in both directions before recording it. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for related identifiers create a graph, its identifier or version context, and one limit that must remain visible.
Version and part labels keep a broad research object from swallowing its precise evidence. HasVersion and IsVersionOf connect a general resource with a version, while IsNewVersionOf and IsPreviousVersionOf describe succession. Direction prevents two records from being flattened into one. A careful reader can preserve identity, context and limits in one compact receipt.
Try this
Evidence question: Which version supersedes which? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Draw arrows and attach each DOI to the correct end. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for version relations are directional, its identifier or version context, and one limit that must remain visible.
A persistent identifier is useful when the metadata explains exactly what it identifies. HasPart and IsPartOf can connect a collection and its components. The collection DOI should not replace the identifier for the exact part used. This makes citation literacy a practical part of science learning.
Try this
Evidence question: Did the analysis use the whole object or one component? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Cite the narrowest object that supplied the evidence. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for part relations preserve granularity, its identifier or version context, and one limit that must remain visible.
A DOI route can stay stable even while the resource location changes. DataCite supports relation types including Cites, IsCitedBy, References and IsReferencedBy. Their meaning depends on direction. It also keeps identifier resolution separate from scientific validation.
Try this
Evidence question: Which object cites or references the other? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Rewrite an ambiguous connection as a complete sentence. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for citations and references need types, its identifier or version context, and one limit that must remain visible.
Relationship types turn isolated records into a navigable research graph. Repository staff can correct or enrich DOI metadata while keeping the DOI route. A metadata update is not automatically a new scientific version of the underlying resource. The record becomes more useful when creators, dates, types and rights remain visible.
Try this
Evidence question: Did the object change, the description change, or both? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Compare retrieval dates and record the kind of change. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for metadata can be updated, its identifier or version context, and one limit that must remain visible.
Version and part labels keep a broad research object from swallowing its precise evidence. DataCite REST API responses expose DOI attributes and relationships for programmatic use. Enriched metadata can be requested separately and should be distinguished from member-maintained original metadata. A careful reader can preserve identity, context and limits in one compact receipt.
Try this
Evidence question: Which endpoint and representation were used? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Save the DOI, request URL, access date and enrichment setting. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for the rest api supports retrieval, its identifier or version context, and one limit that must remain visible.
A persistent identifier is useful when the metadata explains exactly what it identifies. Data stewards, repository managers, librarians, software engineers and researchers maintain the links among objects. Their shared task is precise identity and provenance. This makes citation literacy a practical part of science learning.
Try this
Evidence question: Which hand-off fails when a version relation is reversed? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Design a DOI receipt for a school research dataset. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for identifier literacy supports careers, its identifier or version context, and one limit that must remain visible.
Contents · Previous chapter · Continue at the Science Learning Hub
Follow the identifier, then test the evidence
DataCite makes diverse research outputs easier to identify, cite and connect. The record becomes scientifically useful when the reader keeps resource type, version, part, rights and repository responsibility visible.
Save the DOI, landing page, exact relation types, schema context, retrieval date and filenames used. Then examine the underlying methods and files before making a scientific claim.
| Reading object | What it can support | Question to keep visible |
|---|---|---|
| DOI | Persistent route to a research object | What exactly is identified? |
| Resource type | Broad class of output | Does it match the files? |
| Relation type | Directional link to another object | Which way does the relation run? |
| Rights | Reuse conditions | What permission and attribution apply? |
Sources and next reading
Substantive resource claims in this article were checked against the primary pages below. A structured record is a route into evidence; it does not by itself prove a scientific conclusion. Recheck the live record, evidence labels and release context before using it in assessed work or research.
For a classroom enquiry, save a short research receipt containing the question, search terms, selected object, exact identifier, version, access date and the evidence boundary you observed. Add the file or page that supports each statement. This small routine turns browsing into accountable science learning and gives a teacher, classmate or future analyst a route they can independently follow.
Before sharing the conclusion, ask another reader to retrace the route without verbal hints. If they reach a different version, file or interpretation, improve the receipt and narrow the claim. Reproducibility begins with this practical check. It teaches careful reading, supports fair collaboration and keeps uncertainty visible while evidence moves from an online service into notes, analysis and a final explanation.
A useful final review separates five questions. What object was found? Which organisation or repository maintains it? What exact evidence was inspected? Which statement does that evidence support? What remains uncertain? Answer each question in one sentence, then connect every answer to a visible field, file or source page. This review catches common mistakes: citing a collection when a file supplied the value, treating the newest release as the analysed release, confusing permission with scientific suitability, or turning a discovery count into a conclusion. It also gives students a calm way to improve work. A missing detail becomes a next research step instead of an invitation to guess. When the route is complete, the final explanation can stay simple because the identifiers, dates, versions, limitations and supporting evidence are already organised underneath it.
Primary sources
- DataCite home — accessed 24 October 2026.
- DataCite Metadata Schema 4.7 — accessed 24 October 2026.
- DataCite version and format relationships — accessed 24 October 2026.
- DataCite connections to works — accessed 24 October 2026.
- DataCite REST API single DOI retrieval — accessed 24 October 2026.