How does a FAIR principles assessment differ from the FAIR Guiding Principles themselves?

The FAIR Guiding Principles, published by Wilkinson et al. in 2016, are aspirational design objectives — Findable, Accessible, Interoperable, Reusable. A FAIR principles assessment is the measurement layer that sits on top: a defined indicator set, a scoring method, and a procedure for applying both to a specific output. The principles describe a destination; the assessment tells you how close you are.

Which FAIR assessment framework should an institution adopt first?

For most universities and research-performing organisations, the RDA FAIR Data Maturity Model is the default starting point — it is community-developed, indicator-complete across all four principles, and explicitly designed to be implemented by automated tools, self-assessors, and external evaluators alike. F-UJI is the most common automated implementation and is the practical entry point for repositories with hundreds of datasets.

Can software, vocabularies, and physical samples be FAIR-assessed?

Yes. The FAIR principles are output-type-agnostic and assessment frameworks have been extended to FAIR4RS (software), FAIR vocabularies (FAIR4Voc), and FAIR for physical samples (IGSN). Each variant adapts the indicator set — for example, FAIR4RS interprets Interoperable as API compatibility rather than data-format alignment — while preserving the F/A/I/R structure.

Is a high FAIR score equivalent to high research quality?

No. FAIR principles assess discoverability, accessibility, interoperability, and reusability of digital outputs — not the scientific validity of the underlying research. A dataset can be highly FAIR and methodologically flawed, or methodologically excellent and poorly FAIR. The two evaluations are orthogonal and should be reported separately.

How often should a FAIR assessment be re-run?

Repository-level assessments should be re-run at least annually or whenever the repository's metadata schema, deposit workflow, or PID strategy changes materially. Dataset-level assessments should be re-run when the dataset is updated, when its associated repository upgrades, or when the underlying assessment framework releases a new indicator version. Trend over time matters more than a single snapshot.

FAIR principles assessment

Operational definition

The CASRAI Dictionary records FAIR principles assessment as a distinct operational term because the underlying FAIR Guiding Principles are aspirational by design and do not, on their own, tell an institution whether a given output complies. The principles describe properties that digital outputs should exhibit; the assessment is the procedure by which an institution measures, scores, and reports how closely a specific output approaches those properties.

The distinction matters in practice. A repository can credibly claim alignment with FAIR while still publishing datasets without persistent identifiers, machine-readable licences, or schema-conformant metadata. Without an assessment, the gap between aspiration and reality is invisible. The assessment converts the principles into indicators (typically 30–50 per framework), each tied to an objective or rubric-based check, and aggregates those indicators into a score or maturity level that funders, auditors, and downstream re-users can interpret consistently.

A FAIR principles assessment is therefore the measurement instrument, not the standard being measured. Operationally it has three load-bearing parts: a defined indicator set anchored to the F/A/I/R taxonomy, a scoring method (binary pass/fail, ordinal maturity levels, or continuous score), and a documented procedure for applying both to a named scope — a single dataset, a repository, a domain vocabulary, an institutional posture, or a software package.

Use in practice

How institutions apply a FAIR assessment

Most institutional adoptions of FAIR principles assessment follow a recognisable five-step procedure, even when the team executing it does not name the steps explicitly:

Inventory of digital outputs requiring FAIR posture. The team enumerates the population of outputs in scope: every dataset in the institutional repository, every dataset cited in an active grant, every domain vocabulary the institution publishes, or every software package released by a named research group. Scope decisions cascade through every later step.
Selection of an assessment tool or framework. The team picks an indicator set proportionate to its scope and capacity — typically the RDA FAIR Data Maturity Model for institution-wide audits, F-UJI for automated repository sweeps, or FAIR-Aware for pre-deposit researcher self-checks.
Measurement. Outputs are passed through the chosen instrument, generating either an automated score (F-UJI) or a completed indicator sheet (RDA model). Measurement is dated and versioned against the indicator-set release, since indicators evolve.
Gap analysis. Sub-threshold indicators are grouped by root cause — missing metadata fields, absent machine-readable licences, no PID strategy, no schema validation at deposit. This step is where the assessment becomes useful rather than merely informative.
Remediation plan. Each root cause maps to a remediation owner and a target next-assessment date. The plan is the contractual artefact that funders and certifiers actually read.

Typical scope choices map to different operational decisions. Dataset-level assessmentanswers the per-output question; repository-level assessment answers the platform question, often as part of a CoreTrustSeal audit;domain-vocabulary assessment evaluates whether a controlled vocabulary is itself FAIR enough to underpin FAIR datasets downstream.

Assessment frameworks

Recognised frameworks and tools

The frameworks below operationalise the FAIR Guiding Principles in different ways. Selection should be driven by scope (dataset vs repository vs institution), by whether automated or human assessment is feasible, and by whether a regulator or funder already names a preferred instrument.

Framework	Originator	Scope	Scoring
RDA FAIR Data Maturity Model 2020	Research Data Alliance (RDA)	Dataset	41 indicators across F/A/I/R; maturity levels 0–5; self- or evaluator-administered
F-UJI Automated FAIR Data Assessment Tool 2020	FAIRsFAIR project / PANGAEA	Dataset (via URL / PID)	Automated tests against landing-page and metadata; score 0–100 plus per-principle subscores
FAIR-Aware 2020	DANS-KNAW (Netherlands)	Dataset (pre-deposit)	10-question self-assessment questionnaire; awareness-oriented rather than compliance-graded
CESSDA FAIR-Aware 2022	CESSDA ERIC (social-science variant)	Social-science dataset	Adapted DANS questionnaire with social-science exemplars
ARDC FAIR self-assessment 2021	Australian Research Data Commons	Dataset (national context)	Five-stage rubric per principle, plain-language criteria, no automation
FAIRsharing 2011	FAIRsharing.org / ELIXIR / GO FAIR	Databases, standards, policies	Curated registry; enables FAIR-readiness lookup rather than per-dataset scoring
GO FAIR Office Maturity Indicators 2021	GO FAIR Foundation	Repository and dataset	Rigorous indicator set extending the RDA model; intended for evaluator-administered audits
Australian National FAIR Maturity Model 2022	ARDC / national working group	Repository and institutional posture	RDA model adapted with Australian-context indicators (Indigenous data sovereignty, CARE alignment)

Worked examples

Examples and counter-examples

Is an instance
A university library running F-UJI against its institutional repository's top-100 datasets and reporting per-principle scores in its annual library report.
Is an instance
A funder requiring grantees to complete the RDA FAIR Data Maturity Model self-assessment as part of mid-grant reporting, with the completed indicator sheet attached to the next milestone deliverable.
Is an instance
A trusted-repository certification audit (CoreTrustSeal) that includes a FAIR posture review against an explicit indicator set for each ingested object type.
Not an instance
Stating "our data is FAIR" in a data-management plan without applying any assessment framework — that is an assertion, not an assessment.
Not an instance
Counting the number of datasets in a repository or measuring deposit volume — these are throughput metrics, not FAIR posture.
Not an instance
Confirming that a dataset has a DOI — necessary but insufficient; persistent identification is one indicator within Findable, not a substitute for evaluation across all four principles.

Relationships

How this term relates

Broader

FAIR Data Principles
The aspirational principles being operationalised by the assessment.

Stewardship narrower

Cross-domain

Aliases and translations

Also known as

FAIR assessment · FAIR data maturity · FAIR compliance check · FAIR readiness audit

Language	Status
EN · English	Canonical
FR · Français	Community draft
ES · Español	Community draft
DE · Deutsch	Untranslated
PT · Português	Untranslated
NL · Nederlands	Untranslated
JA · 日本語	Untranslated
ZH · 中文	Untranslated

Community translations enter via the contribute flow and are credited with CRediT.

Machine-readable encodings

Use in your systems

Each encoding below is copy-paste ready. The JSON-LD DefinedTerm is the same payload this page emits in its <head>; the JATS fragment is the recommended reference form for manuscript data-availability statements; the GraphQL example queries the live dictionary endpoint.

JATS XML — referencing the term in a data-availability statement

xml

<sec sec-type="data-availability">
  <title>Data availability statement</title>
  <p>The datasets supporting this study are deposited in Zenodo under
    DOI <ext-link ext-link-type="doi" xlink:href="https://doi.org/10.5281/zenodo.0000000">10.5281/zenodo.0000000</ext-link>.
    FAIR posture was evaluated using the
    <named-content content-type="vocab-term"
      vocab="casrai-dictionary"
      vocab-identifier="https://casrai.org/dictionary"
      vocab-term="FAIR principles assessment"
      vocab-term-identifier="https://casrai.org/dictionary/term/fair-principles-assessment">FAIR principles assessment</named-content>
    framework F-UJI v3.0, scoring 78/100 (Findable 100, Accessible 90,
    Interoperable 60, Reusable 65). Per-indicator results are bundled with
    the deposit as <italic>fair-assessment-report.json</italic>.</p>
</sec>

Schema.org DefinedTerm (JSON-LD)

json

{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/dictionary/term/fair-principles-assessment",
  "name": "FAIR principles assessment",
  "alternateName": [
    "FAIR assessment",
    "FAIR data maturity",
    "FAIR compliance check",
    "FAIR readiness audit"
  ],
  "identifier": "casrai:term/fair-principles-assessment",
  "description": "The systematic evaluation of digital research outputs (data, software, vocabularies, samples) against the FAIR Guiding Principles (Findable, Accessible, Interoperable, Reusable) using a defined maturity model, scoring framework, or self-assessment tool, to identify FAIR compliance gaps and prioritise remediation.",
  "inDefinedTermSet": "https://casrai.org/dictionary",
  "url": "https://casrai.org/dictionary/term/fair-principles-assessment",
  "sameAs": [
    "https://www.wikidata.org/wiki/Q60169879",
    "https://www.go-fair.org/fair-principles/"
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "bibliographicCitation": "CASRAI Editorial Board (2026). FAIR principles assessment (v2026.1). CASRAI, CASRAI Dictionary. https://casrai.org/dictionary/term/fair-principles-assessment",
  "dateModified": "2026-03-15",
  "version": "v2026.1",
  "inLanguage": "en",
  "audience": {
    "@type": "Audience",
    "audienceType": "Research data managers, librarians, repository operators, funders, data stewards"
  }
}

GraphQL query against the dictionary endpoint

graphql

query GetTerm {
  dictionaryTerm(id: "fair-principles-assessment", idType: SLUG) {
    title slug version status
    definitionOperational
    aliases
    dictionaryDomains { nodes { name slug casraiTrack } }
    relationships { type targetSlug targetUri note }
    definedTermJsonLd {
      context type name identifier description
      inDefinedTermSet url sameAs
    }
  }
}

Identifiers

Persistent identifiers

CASRAI page	https://casrai.org/dictionary/term/fair-principles-assessment
CASRAI PID
Wikidata	Q60169879 (FAIR Guiding Principles)
In set	https://casrai.org/dictionary

Frequently asked

Common questions

How does a FAIR principles assessment differ from the FAIR Guiding Principles themselves?: The FAIR Guiding Principles, published by Wilkinson et al. in 2016, are aspirational design objectives — Findable, Accessible, Interoperable, Reusable. A FAIR principles assessment is the measurement layer that sits on top: a defined indicator set, a scoring method, and a procedure for applying both to a specific output. The principles describe a destination; the assessment tells you how close you are.
Which FAIR assessment framework should an institution adopt first?: For most universities and research-performing organisations, the RDA FAIR Data Maturity Model is the default starting point — it is community-developed, indicator-complete across all four principles, and explicitly designed to be implemented by automated tools, self-assessors, and external evaluators alike. F-UJI is the most common automated implementation and is the practical entry point for repositories with hundreds of datasets.
Can software, vocabularies, and physical samples be FAIR-assessed?: Yes. The FAIR principles are output-type-agnostic and assessment frameworks have been extended to FAIR4RS (software), FAIR vocabularies (FAIR4Voc), and FAIR for physical samples (IGSN). Each variant adapts the indicator set — for example, FAIR4RS interprets Interoperable as API compatibility rather than data-format alignment — while preserving the F/A/I/R structure.
Is a high FAIR score equivalent to high research quality?: No. FAIR principles assess discoverability, accessibility, interoperability, and reusability of digital outputs — not the scientific validity of the underlying research. A dataset can be highly FAIR and methodologically flawed, or methodologically excellent and poorly FAIR. The two evaluations are orthogonal and should be reported separately.
How often should a FAIR assessment be re-run?: Repository-level assessments should be re-run at least annually or whenever the repository's metadata schema, deposit workflow, or PID strategy changes materially. Dataset-level assessments should be re-run when the dataset is updated, when its associated repository upgrades, or when the underlying assessment framework releases a new indicator version. Trend over time matters more than a single snapshot.

Cite this term

Cite the FAIR principles assessment entry

These citations reference this dictionary term as a versioned, CC-BY 4.0 reference work. For guidance on citing the dictionary as a whole, see How to cite the CASRAI Dictionary.

APA 7

CASRAI Editorial Board (2026). FAIR principles assessment (v2026.1). CASRAI, CASRAI Dictionary. https://casrai.org/dictionary/term/fair-principles-assessment

Vancouver

CASRAI Editorial Board. FAIR principles assessment [Internet]. CASRAI, CASRAI Dictionary; 2026 [cited 2026]. Version v2026.1. Available from: https://casrai.org/dictionary/term/fair-principles-assessment

Chicago (author-date)

CASRAI Editorial Board. 2026. "FAIR principles assessment." Version v2026.1. CASRAI, CASRAI Dictionary. https://casrai.org/dictionary/term/fair-principles-assessment.

BibTeX

@misc{casrai_dict_fair_principles_assessment_2026,
  author       = {{CASRAI Editorial Board}},
  title        = {{FAIR principles assessment}},
  year         = {2026},
  version      = {v2026.1},
  publisher    = {CASRAI},
  series       = {CASRAI Dictionary},
  howpublished = {\url{https://casrai.org/dictionary/term/fair-principles-assessment}},
  note         = {Licensed CC-BY 4.0. Aligned with the RDA FAIR Data Maturity Model.}
}