Connectivity
Connectivity is the state or extent of being connected — for a research object, whether the people, organizations, funders, and other things it names carry the identifiers that link them to everything else. It is distinct from completeness: completeness asks do we have the value? (is there a creator, an affiliation, a publisher); connectivity asks is that thing identified, so it can be found and linked? A creator name is complete; a creator name with an ORCID is connected.
Identifiers are the lifeblood of these connections. ORCIDs for people, RORs for organizations, funders, or publishers, DOIs for projects, samples, instruments, and research outputs — each one turns a piece of free text into a node in the research graph. High connectivity changes how things are found: instead of searching inside one repository's portal, you arrive by traversing the graph — finding interesting things because they are connected to things you already care about. Remember, items with high connectivity get discovered.
This page explains the idea behind the Metadata Connectivity tool. The two figures below are from Connectivity (Ted Habermann, Metadata Game Changers, 2021).
Connectivity is a simple ratio — the share of things that carry the identifier that could connect them — reported on a 0–1 scale: 0% Missing, 50% Partial, 100% Complete. For people the things are authors and the identifier is an ORCID; for organizations the things are affiliations and the identifier is a ROR.
The Metadata Connectivity tool applies this idea across a whole repository. For each connector it reports two things. The percentage is occurrence-based — identified occurrences ÷ all occurrences — the repository-wide version of the ratio above. The bar counts distinct entities (a person or organization that appears in many records is one entity): green = identified in every occurrence (Complete), yellow = identified in some (Partial), and red = never identified (Missing). The yellow segment marks the quick wins: an identifier already known for an entity that just needs to be copied to the records that are missing it — no lookup required.
| Connector | Identifier | DataCite path (after data[*].attributes) |
|---|---|---|
| Creators — name identifier | ORCID | creators[*].nameIdentifiers[*].nameIdentifier |
| Creators — affiliation | name present | creators[*].affiliation[*].name |
| Creators — affiliation identifier | ROR | creators[*].affiliation[*].affiliationIdentifier |
| Contributors — name identifier | ORCID | contributors[*].nameIdentifiers[*].nameIdentifier |
| Contributors — affiliation | name present | contributors[*].affiliation[*].name |
| Contributors — affiliation identifier | ROR | contributors[*].affiliation[*].affiliationIdentifier |
| Funders — funder identifier | Crossref Funder ID / ROR | fundingReferences[*].funderIdentifier |
| Publisher — publisher identifier | ROR | publisher.publisherIdentifier |
| Rights — rights URI | license URI | rightsList[*].rightsUri |
| Rights — rights identifier | SPDX | rightsList[*].rightsIdentifier |
Connectivity makes improvement visible. As identifiers proliferate in metadata over time, a repository’s distribution shifts from red toward green — and each added ORCID or ROR is a new edge in the research graph, making the work it describes easier to find, cite, and credit. The connectivity bars turn that into a punch list: the red and yellow entities are exactly where adding an identifier pays off.