How can I access and understand citation data?

Answer

What are citations? 

A citation is a reference from one research output to another—typically where one article references another in its bibliography. Citations help track the visibility and influence of scholarly work and are widely used as indicators of research impact. However, citation metrics should always be interpreted carefully and used in context.

Simple citation counts for articles and authors 

Citation counts show how many times a publication—or an author’s body of work—has been cited by others. They are available from several sources, including:

  • Scopus: (Elsevier's abstract and citation database) 

  • SciVal: (Elsevier's analytics platform, built on Scopus data)

  • Web of Science: (Clarivate's abstract and citation database)

  • InCites: (Clarivate's analytics platform, built on Web of Science data)

  • OpenAlex: (An open, community-driven scholarly metadata source)

  • Scite: (platform combining citation data with qualitative citation context)

Example of article citation count within Scopus

Please note: citation counts can vary between platforms due to differences in coverage, such as:

  • The journals and sources indexed.

  • The types of publications included (e.g., journal articles, conference papers, books).

  • The timeframe covered.

Citation counts may differ between platforms due to differences in what they index. These differences may include:

  • The journals and publication sources covered

  • Types of content included (e.g., articles, books, conference papers)

  • Time periods indexed and update frequency

For a fuller picture of citation impact, we recommend checking multiple sources and consulting our guidance on the responsible use of metrics.

What are normalised citation metrics (FWCI and CNCI)?

Raw citation counts can be misleading, especially when comparing across disciplines or time periods. For example, citation rates in the life sciences tend to be significantly higher than in the humanities. To address this, normalised metrics compare citation counts to a global average for similar publications.

Two common indicators are:

Both metrics compare the actual citation count of a publication with the expected average for similar outputs (same field, publication year, and document type).

How to interpret these metrics:

  • A score of 1.00 = average citation rate worldwide

  • A score above 1.00 = cited more than expected

  • A score below 1.00 = cited less than expected

Why use them?

  • They enable fairer comparisons across disciplines and timeframes

  • They add context to raw citation counts

  • They can support analysis at the individual, group, or institutional level

What are citation percentiles?

Percentile metrics offer another way to assess citation impact by showing where a publication ranks relative to others in the same field and year.

For example, in SciVal or InCites:

  • 99th percentile = top 1% of cited publications

  • 95th percentile = top 5%

  • 90th percentile = top 10%

  • 75th percentile = top 25%

  • 50th percentile = median performance

Why use percentiles:

  • They are field-normalised and suitable for cross-disciplinary comparisons

  • They are less affected by outliers than averages

  • They help identify standout publications in a portfolio

Elsevier and Clarivate both provide guidance on how these metrics are calculated:


Limitations to keep in mind

  • Recent publications may not yet have accumulated citations

  • Metrics can vary depending on the platform and source

  • Small or niche research areas may lack reliable benchmark data

  • Citation counts alone do not reflect quality, societal impact, or influence beyond academia

Want to learn more?

Our online resource My Research Essentials: Using Citation Analysis to Measure Research Impact provides an introduction to normalised metrics, explains field variation, and explores what citation data can—and cannot—tell you.

You can also explore:

Further information and support 

Citation metrics are useful tools—but they should always be used thoughtfully and in context. They are not definitive measures of research quality or impact and should be combined with qualitative evaluation and expert judgement.

If you would like to discuss these metrics in further detail, and how you can access them, please contact us 

 

Topics

  • Last Updated 05 Sep 2025
  • Views 83
  • Answered By John Hynes

FAQ Actions

Was this helpful? 0 0