Citations can be used for an individual person (total citations) or a publication.

Citation counts for authors and publications can be found in services such as Web of Science, Scopus, Google scholar or Dimensions; some journal pages also show citation counts for publications. You can establish a citation alert with some services to keep track of how many times a record or article is cited. Note that these different sources may give different numbers of citations for the same publication, as they vary in their coverage.

How to: See Module 3 for guidance on how to find citation counts, how to deal with variations, and for suggestions for more informative statistics.

Issues

Citations are a measure of usage. They are sometimes used as proxies for “impact” or “quality”, but while citations may give an indication of scientific impact, their reflection of "quality" and societal impact is debated (Aksnes et al. 2019). There are several issues to be aware of when using citation counts:

The relationship between citations and "quality" is not clear cut (e.g. Dougherty et al. 2022).
Citations counts lack context - some cite works to criticise them. Famous retracted papers, such as the Wakefield MMR study, have huge numbers of citations due to their controversial status or use as negative examples.
Citation counts will vary by data source, and finding accurate citation counts can be difficult for outputs like books, data or software. This means that citation counts (and by extension, the h-index) for someone publishing in a book-dominated field will likely be underestimates.
Citation practices vary between fields - thus citation counts will vary, regardless of quality. Citations will also increase with the age of the publication. This means that comparing raw citation counts for different fields, or universities with different profiles, is not informative. Weighted measures that try to account for field and age are available, e.g. field weighted citation ratio.
Biases, cultural or social factors may affect citation counts. Duyx et al. (2017) found that articles with positive results are more likely to be cited; King et al. 2017 found men cited their own work more than women; West & McIlwaine (2002) found works from developing countries were cited less; citation rates are associated with gender and ethnicity (e.g. Dion et al. 2018; Kozlowski et al. 2022).
Citations tend to have a self-reinforcing effect - 21% of citations went to 1% of researchers in 2015 (Nielsen & Andersen 2021; article in Khrono).

Useful to know: Nowadays, publication lists which include the h-index and citation-counts are visible for everyone online. Web of Science and Scopus use algorithms to automatically create profiles for authors in their databases. You can “claim” these profiles if you wish to control, edit and verify your own record.

More information: Read more in the Metrics toolkit