Why English? | Reading Dimensions Content Types, Citation Metrics and Linked Research Before Comparing Evidence

Dimensions research literacy matters because publications, datasets, grants, patents, clinical trials, policy documents and reports occupy different positions in a research system even when the platform connects them on one screen.

For students, parents, educators and adult learners, precise English separates a publication from the grant that supported it, a trial record from its results paper, a dataset from an analysis, and a citation metric from a judgement about truth, usefulness or educational value.

This guide uses current official Dimensions help and documentation checked on 3 November 2026. It explains content-type coverage, subscription differences, search fields, linked records, citation counts, FCR, RCR, Altmetric attention, researcher disambiguation and update timing; it does not promise complete coverage or reduce evidence appraisal to a score.

Reading routes

• Read content types correctly
• Trace linked research objects
• Interpret citation metrics
• Handle access, people and updates
• Run the Dimensions evidence check

Chapter 1 of 18: Dimensions is a linked research-information system, not one uniform library shelf

What this Dimensions object, link or metric represents

The platform brings several research object types into a connected environment. Each type records a different event, output or administrative process.

The comparison language that prevents a false conclusion

The broad noun research does not erase object differences. Replace it with publication, dataset, grant, patent, trial, policy document or report.

For Dimensions is a linked research-information system, not one uniform library shelf, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A learner follows a topic across a grant, article and dataset while keeping separate notes for each.

Coverage and available content depend on platform version and subscription.

• Read the active content type.
• Record the account context.
• Open the details page.
• Use type-appropriate questions.

Sort eight records into object families.

The reasoning in Dimensions is a linked research-information system, not one uniform library shelf transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Next chapter

Chapter 2 of 18: Publications communicate research in several forms

What this Dimensions object, link or metric represents

Dimensions publication coverage can include journal articles, preprints, books, chapters and other scholarly forms from multiple sources. Publication type and state affect how the item should be read.

The comparison language that prevents a false conclusion

The word publication is a family label, not a synonym for peer-reviewed journal article.

For Publications communicate research in several forms, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student labels a preprint and later journal article separately before comparing versions.

Source metadata can be incomplete, and publication does not prove methodological quality.

• Read type and date.
• Check peer-review status independently.
• Resolve the source link.
• Look for corrections.

Write a publication-state timeline.

The reasoning in Publications communicate research in several forms transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 3 of 18: Datasets are research objects with their own identity

What this Dimensions object, link or metric represents

Dimensions indexes stand-alone datasets and datasets associated with publications from repositories. A dataset can have creators, dates, licences and persistent identifiers separate from a paper.

The comparison language that prevents a false conclusion

A dataset contains organised observations or measurements; an article interprets selected questions using data.

For Datasets are research objects with their own identity, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A learner opens the dataset record, reads documentation and cites it separately from the paper.

Repository presence does not guarantee completeness, appropriate consent or fitness for a new analysis.

• Check repository and identifier.
• Read documentation.
• Read licence and version.
• Assess reuse limits.

Build a dataset source card.

The reasoning in Datasets are research objects with their own identity transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 4 of 18: Grants describe awarded support, not research results

What this Dimensions object, link or metric represents

Grant records describe funding awards, programmes, organisations, amounts or project periods where available. They can connect resources to later outputs.

The comparison language that prevents a false conclusion

The verb awarded does not mean completed, successful, published or endorsed beyond the funding decision.

For Grants describe awarded support, not research results, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A careers project uses grants to identify funded research areas, then checks outputs separately.

Funding metadata can be delayed, converted across currencies or incomplete.

• Record funder and programme.
• Read award dates.
• Separate award from outcome.
• Check linked outputs.

Rewrite “the grant proved” as an accurate sentence.

The reasoning in Grants describe awarded support, not research results transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 5 of 18: Patents, trials and policy documents require different evaluation

What this Dimensions object, link or metric represents

Patent records describe legal-technical disclosures, clinical-trial records describe registered study information, and policy documents describe guidance or policy communication. Their evidence roles differ from research articles.

The comparison language that prevents a false conclusion

Avoid calling all three “studies.” Use the platform’s displayed type and the issuing institution or registry.

For Patents, trials and policy documents require different evaluation, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student links a trial registration to an article and a policy document but evaluates each under its own purpose.

Links can be algorithmic, identifier-based or incomplete and do not show agreement.

• Identify the issuing system.
• Check dates and status.
• Read the linked source.
• Avoid cross-type ranking.

Explain the role of each object in one sentence.

The reasoning in Patents, trials and policy documents require different evaluation transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 6 of 18: Content availability depends on account and subscription

What this Dimensions object, link or metric represents

Official Dimensions guidance says the free version searches publications and datasets, while access to grants, patents, trials, policy documents and other fields may require subscription.

The comparison language that prevents a false conclusion

Not visible can mean not entitled in the current account, not absent from the research world.

For Content availability depends on account and subscription, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student documents that a comparison used the free version and does not claim to have searched every Dimensions content type.

Institutional configurations can differ.

• Name the version used.
• Record visible content types.
• Avoid universal claims.
• Use other official sources where needed.

Write an access-scope sentence.

The reasoning in Content availability depends on account and subscription transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 7 of 18: Linked records show documented associations, not a single causal chain

What this Dimensions object, link or metric represents

Dimensions connects objects through identifiers, acknowledgements, citations and other metadata. These links can reveal research context.

The comparison language that prevents a false conclusion

Use linked to for a documented association and reserve caused or led to for evidence that establishes the mechanism.

For Linked records show documented associations, not a single causal chain, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A grant record links to publications; the learner describes funded outputs without claiming that funding alone produced every finding.

Links can be missing or incorrect, and complex projects have multiple influences.

• Inspect the link type.
• Open both records.
• Check dates.
• Use narrow verbs.

Build an association-versus-causation ladder.

The reasoning in Linked records show documented associations, not a single causal chain transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 8 of 18: A trial record and a results publication are separate objects

What this Dimensions object, link or metric represents

Clinical-trial registries describe planned and updated study information; publications report analyses. Linking them supports comparison of planned and reported work.

The comparison language that prevents a false conclusion

The words registered, completed, results posted and published name different states.

For A trial record and a results publication are separate objects, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A health-research assignment compares registered outcomes with the journal article while avoiding personal medical advice.

Records can be updated, and not every trial produces a publication.

• Record trial ID.
• Check registration date.
• Compare outcomes.
• Use current health guidance for decisions.

Create a trial-to-publication table.

The reasoning in A trial record and a results publication are separate objects transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 9 of 18: A dataset link does not prove the article used every available file

What this Dimensions object, link or metric represents

A publication can be associated with data through repository metadata or identifiers. The relationship should be read on both records.

The comparison language that prevents a false conclusion

The phrase associated dataset is safer than “the complete data” unless the authors and repository establish completeness.

For A dataset link does not prove the article used every available file, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student checks file descriptions and version dates before saying which dataset underlies an analysis.

Embargoes, restricted files and later versions can change what is accessible.

• Open the data record.
• Read file inventory.
• Check version and licence.
• Match the article statement.

Write a bounded data-link claim.

The reasoning in A dataset link does not prove the article used every available file transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 10 of 18: Citation counts are database-scoped observations

What this Dimensions object, link or metric represents

Dimensions counts citations from reference lists in publications it has indexed. The number reflects platform coverage and processing at a point in time.

The comparison language that prevents a false conclusion

Say Dimensions citations on [date] rather than treating the number as a universal total.

For Citation counts are database-scoped observations, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A learner reports a count with platform and date, then reads citing papers instead of using the count as the conclusion.

New papers, missing references and formatting problems can change or suppress counts.

• Name the platform.
• Date the count.
• Inspect citing sources.
• Avoid quality claims.

Write a scoped citation-count sentence.

The reasoning in Citation counts are database-scoped observations transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 11 of 18: FCR compares citation performance within assigned fields and age

What this Dimensions object, link or metric represents

Dimensions defines Field Citation Ratio as citations relative to similarly aged publications in assigned Fields of Research. A value around one reflects the relevant benchmark under its calculation.

The comparison language that prevents a false conclusion

The adjective field-normalised describes the comparison group; it does not make the metric discipline-free or value-neutral.

For FCR compares citation performance within assigned fields and age, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student compares two eligible articles only after checking their years and field assignments.

FCR is generally unavailable for very recent papers and requires sufficient field data; automatic field assignment can affect the benchmark.

• Read metric definition.
• Check publication age.
• Check field codes.
• Avoid cross-object comparison.

Explain why a recent paper may have no FCR.

The reasoning in FCR compares citation performance within assigned fields and age transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 12 of 18: RCR uses a different benchmark and coverage rule

What this Dimensions object, link or metric represents

Dimensions describes Relative Citation Ratio as a publication-level metric based on a co-citation network and NIH iCite data for qualifying PubMed-linked publications.

The comparison language that prevents a false conclusion

FCR and RCR are not two spellings of the same number. Their fields, eligibility and reference benchmarks differ.

For RCR uses a different benchmark and coverage rule, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A biomedical student records both values where present and explains why neither should be compared casually with a humanities book.

RCR coverage depends on PubMed linkage, age and citation data.

• Read eligibility.
• Name the benchmark.
• Compare only suitable objects.
• Report missing values honestly.

Write one sentence contrasting FCR and RCR.

The reasoning in RCR uses a different benchmark and coverage rule transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 13 of 18: Missing metrics can mean ineligible or uncovered, not low impact

What this Dimensions object, link or metric represents

Official help explains that absent citation, FCR, RCR or attention values can reflect recency, missing references, field assignment, PubMed coverage or no monitored attention.

The comparison language that prevents a false conclusion

Use not available or not calculated when that is what the interface shows; do not silently convert it to zero.

For Missing metrics can mean ineligible or uncovered, not low impact, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student leaves an empty RCR as unavailable and explains the coverage rule.

A displayed zero and a blank value can have different meanings.

• Read the legend.
• Distinguish zero from blank.
• Check eligibility.
• Avoid imputation.

Classify six missing-value scenarios.

The reasoning in Missing metrics can mean ineligible or uncovered, not low impact transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 14 of 18: Altmetric attention is not a citation or quality score

What this Dimensions object, link or metric represents

Dimensions can display an Altmetric Attention Score, which summarises tracked online attention under Altmetric’s method. It measures a different phenomenon from scholarly citation.

The comparison language that prevents a false conclusion

The noun attention can include discussion, news or social activity; it does not mean approval, accuracy or educational importance.

For Altmetric attention is not a citation or quality score, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A controversial paper has high attention but weak methods; the learner keeps attention and appraisal separate.

Tracked sources, language and platform activity shape the score.

• Read the attention details.
• Inspect source types.
• Separate praise from criticism.
• Appraise the paper.

Rewrite “high impact” into two precise observations.

The reasoning in Altmetric attention is not a citation or quality score transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 15 of 18: Advanced search fields encode definitions that must be read

What this Dimensions object, link or metric represents

Dimensions advanced search can target fields, ranges, identifiers and content-specific attributes. Precise queries improve reproducibility.

The comparison language that prevents a false conclusion

A field query searches a named metadata field; it is not equivalent to searching all text. Range syntax also defines boundaries that should be reported.

For Advanced search fields encode definitions that must be read, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student searches publication years and funder IDs separately from title-and-abstract words.

Some fields depend on subscription or source availability.

• Read field documentation.
• Test a known record.
• Save exact syntax.
• Record account scope.

Translate a natural question into a fielded query.

The reasoning in Advanced search fields encode definitions that must be read transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 16 of 18: Researcher profiles depend on disambiguation

What this Dimensions object, link or metric represents

Dimensions groups name variants into researcher profiles using disambiguation. Official guidance says the process is conservative and can leave one person split across profiles rather than wrongly merging people.

The comparison language that prevents a false conclusion

A profile is a modelled identity record, not the person. Name similarity is evidence to check, not proof.

For Researcher profiles depend on disambiguation, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A student verifies affiliation, ORCID and publication history before attributing a portfolio.

Common names, career moves and incomplete metadata can split or combine records imperfectly.

• Check identifiers.
• Compare affiliations and dates.
• Avoid name-only claims.
• Report uncertainty.

Audit a profile without assuming completeness.

The reasoning in Researcher profiles depend on disambiguation transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 17 of 18: Update timing differs across content and sources

What this Dimensions object, link or metric represents

Dimensions says update frequency varies: some data and metrics can refresh frequently, while funders or other providers may publish updates less often.

The comparison language that prevents a false conclusion

The adjective current needs a date and data-source context. A recent search can still contain slowly updated source data.

For Update timing differs across content and sources, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A report states the search date and notes that grant information follows funder releases.

Historical records can be corrected after a project ends.

• Record search date.
• Read update guidance.
• Recheck cycle-sensitive facts.
• Archive the query.

Write a freshness note for two content types.

The reasoning in Update timing differs across content and sources transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Next chapter

Chapter 18 of 18: A fair comparison aligns object, time, field and question

What this Dimensions object, link or metric represents

Responsible comparison controls for content type, publication age, field, database coverage and metric eligibility before interpreting differences.

The comparison language that prevents a false conclusion

The words more cited, more visible, more funded and better evidence answer different questions.

For A fair comparison aligns object, time, field and question, complete the comparison note: “This is a __ record linked to __; the platform covers it through __; the metric is __ defined as __; missing or present values reflect __; I inspected __; the comparison supports only __.” Keep system metadata, attention, citations and evidence quality separate.

A connected-research example, boundary and next move

A learner compares citation performance among similar articles, then evaluates methods separately.

Even aligned metrics cannot capture every form of educational, social or practical value.

• Define the comparison question.
• Choose like objects.
• State metric and date.
• Add qualitative appraisal.

Present a comparison with a limitations paragraph.

The reasoning in A fair comparison aligns object, time, field and question transfers to Scopus, Web of Science, OpenAlex and institutional research dashboards. Compare the same object types under the same date, coverage and metric definition before using numbers in study, course or career decisions.

Back to reading routes · Previous chapter · Continue to the FAQ

A twenty-minute Dimensions evidence check

Turn a connected research record and its metrics into a bounded comparison that another reader can reproduce.

• State the question and the Dimensions account or access context.
• Name every content type rather than calling everything “research.”
• Open linked records and identify the relation instead of assuming causation.
• Read the definition, eligibility and missing-value rule for each metric.
• Record query, fields, filters, date and update limits.
• Compare like objects, then appraise methods and claims separately.

For parents and educators: ask what the number measures, what it does not measure and why a blank value appears. A learner who can answer those questions is building durable data literacy.

For students: use Dimensions to see the research system, not to replace reading. A link is a clue, a metric is a defined observation and the source still deserves appraisal.

Frequently asked questions

Does Dimensions contain only journal articles?

No. Official guidance lists publications, datasets, grants, patents, clinical trials, policy documents and other research objects, with access depending on version.

Is FCR the same as RCR?

No. They use different field definitions, benchmarks and eligibility rules.

Does no metric mean no impact?

No. It can mean the item is too recent or outside the metric’s coverage.

Does an Altmetric score measure research quality?

No. It measures tracked online attention under a specific method.

Can linked records prove causation?

No. They establish a documented association; causal claims require further evidence.

Useful next reading

• Dimensions: what information is included
• Dimensions: Field Citation Ratio
• Dimensions: why metrics may be missing
• Dimensions: publication data-field definitions
• Dimensions: advanced search
• Dimensions: update frequency
• English Learning Hub: Primary to adult English
• Google Scholar for students: source and citation checks
• Zotero for students: research, citations and bibliographies
• How literature search and research discovery work
• How scholarly indexing works
• How citation works
• How research methods and source evaluation work
• How to use a library for academic journals and research papers

English matters in Dimensions because publication, dataset, grant, patent, trial, policy document, link, citation count, FCR, RCR, attention score and missing value describe different research-system signals. Naming those signals accurately makes comparison useful without turning a dashboard into a verdict.