Skip to main content
Subject Guides

Bibliometrics: a Practical Guide

Article-level indicators

Article citation counts

A high citation count compared to other articles in the same field published at around the same time indicates that an article has been highly influential on the progress of research in that field.

The number of times a known article has been cited by other scholarly publications.

A citation count is a measure of impact rather than quality: an article may be cited in order to refute its conclusions, or as an example of poor practice.

For more discussion of the limitations and best uses of article citation counts, see the Library's Information for Researchers.

Best uses for article-level bibliometrics

  • Identify the articles which have had the greatest influence on the direction of research in your field.
  • Find out who is citing your own article, and why. Has your research proved beneficial in an unexpected way? Or perhaps someone has challenged your findings.
  • Create an alert to new citations for a known article, in order to keep abreast of latest developments. 
  • Some providers of citation data have developed metrics which can be used to compare articles in different fields, by standardizing against the norm for the discipline.

Article usage counts

The usage count is an indicator of the level of interest in an article.  It can provide useful information about newer articles which haven't yet been cited. But be aware that the data is platform-specific: when an article is available on multiple platforms, no single measure of usage can be treated as definitive.

Finding usage counts:

  1. Web of Science has counted usage for each article record since February 2013.
  2. Try a subject-specific database: look for usage data in your results list or the full record for an article
  3. If neither of those avenues work, go to the publisher's website; usage counts are most likely to appear on the "landing page" for the article (i.e. where you find the article abstract and link to the full text.)  

Regardless of where you find the data, always check to see how the count was determined (usually found on a help page) as each database and website counts usage differently. For example, the usage count on Web of Science reflects the number of clicks on a link to the full text of the article, plus the number of times the article is saved for use in a bibliographic management tool such as EndNote or Paperpile.

Databases that include article citation counts

Whilst all sources of citation data are continually trying to extend their coverage, no single database indexes all the world's scholarly publications. Coverage of the arts and humanities may be less comprehensive, especially for books.

Generally, Web of Science and Scopus offer good coverage of sciences and social sciences, with less emphasis on humanities. Google Scholar covers all disciplines equally well, and captures 'grey literature' such as policy papers and technical reports in addition to journals.  But the criteria for inclusion in Google Scholar are not wholly transparent.

Increasingly, e-journal platforms and subject-specialist bibliographic databases incorporate citation information acquired from these sources - you will see 'Cited By', 'Times Cited' or similar terms in their article records. 

Finding the citation count for a known article

Example of times cited score from Web of ScienceTo find an article citation score on Web of Science, carry out a Basic Search for the article you are interested in. The number of times the article has been cited will be visible on the results page.

Click on the article title to view the full record and to see the full details of citing publications.


 

To be 100% confident that the citation score is correct, Web of Science enables you to run a Cited Reference Search, which will also find citations where the bibliographic details are incomplete or incorrect.

 

Citation alerts

Create citation alerts and receive notification by email whenever your article is cited in the future. A very easy way to keep up-to-date.

Metrics presented on the Scopus databaseTo find out how many times a known article has been cited,  search for the article by keyword and/or author. When you reach the document details for the required article, metrics are presented in the right-hand panel.

 

 

To create a citation alert:

Under the Cited by reference list, look for:
Scopus options for creating a citation alert

Google Scholar result with 'Cited by' circled

Many Google Scholar records include a Cited by score, which also acts as a link to the citing documents.

 

 

 

Finding highly-cited papers on a topic of interest

Web of Science 'sort by' Times CitedSearch Web of ScienceScopus or Google Scholar for keywords relating to your subject.  From your search results page:

  • in Web of Science, select Times Cited -- highest to lowest from the 'Sort by' drop-down menu 
  • in Scopus, select Cited by (highest) from the 'Sort on' drop-down menu 

It's not possible to sort Google Scholar search results by times cited.  However, the Relevance ranking appears to take account of citations, so highly cited articles are likely to appear near the top of your results list.

Comparing article citation counts

In different scholarly disciplines, you will find differences in the typical number of authors on a paper, the breadth of the audience and the frequency of citation.

If you wish to analyse the impact of articles from different fields, you should use "normalised" or "field-weighted" citation metrics, to ensure that you are comparing like-with-like.

Scopus has developed a Field-Weighted Citation Impact score for articles, and this score can be benchmarked against the best in the field:

Scopus metrics e.g. Field Weighted Citation Impact

Further reading: How are article metrics used in Scopus?

Limitations: Most field-weighted indicators rely upon a subject classification assigned by the database compilers, which inevitably has a subjective element.  See Wang & Waltman (2015): Large-scale analysis of the accuracy of the journal classification systems of Web of Science and Scopus.