Citations are the most widely-used metric to assess the impact (or lack of impact) of a journal article.
Before judging an article based on a citation count or metric, consider the following limitations of citations as evaluation metrics:
- A citation is not necessarily a positive reference. An article can be highly cited because it is controversial, satirical or its claims are being challenged. Note that a paper that has been retracted by its publisher can continue to accrue citations;
- Review papers tend to be more well-cited than original research papers, and this can deny authors of original concepts and results their deserved credit. Because of this, citations must never be used to compare between articles of different types;
- Some studies of citation patterns suggest that gendered and racial biases exist in citation networks. This could mean that citation counts underrepresent the impact of female or Black, Asian and Minority Ethnic authors;
- Research is done across the world in all languages, yet the dominant language for research publishing is English. This can exclude non-English language research from being made discoverable and cited by other readers. And when an English-language article is cited by a non-English language article or journal, this citation might not be picked up in the citation count;
- Some commentators have asked ‘do authors cite another paper just because it is already highly-cited?’. Whilst not proven, a prestige effect that causes authors to cite a well-known paper over a lesser-known paper is possible.
Essentially, citations should not be the only way that impact is assessed!