• Bibliometrics are a range of statistical measures that analyse publication data
• Bibliometric analysis can be applied to an author, article, journal, researcher group or an institution and should always be used responsibly
• Citation databases (Scopus, Web of Science, Dimensions) aggregate citation data using them to create citation metrics
• A concept of ‘research impact’ is based on the premise the subsequent citation of research demonstrates quality
• International university rankings incorporate bibliometrics as part of their measurements

Central to using bibliometrics is understanding the scope (strengths, weaknesses, coverage and parameters) of the database and metrics being used. Importantly:

• not all research outputs are indexed in bibliographic databases so there will always be data gaps in any result.
• no database indexes all published works, and none cover all subject areas equally. Ideally, bibliometrics should be used alongside qualitative analysis and/or altmetrics.

Although bibliometrics provide quantitative data that can be used to evaluate authors and institutions, the limitations of bibliometric analysis mean that

• bibliometrics should never be used as the sole criteria for any evaluative purpose
• bibliometric data should always be used in combination with other forms of evaluation, such as peer review, to make final decisions.