Citation analysis
Studies how often works are cited and how citations are distributed over time and fields. Counts differ between databases because their coverage differs.
The h-index
A researcher has index h if h of their papers have each been cited at least h times. Limits: it depends on discipline, career length and the database used, it does not exclude self-citation, and it does not measure the quality of works.
Journal metrics
Databases provide journal-level indicators (such as the Journal Impact Factor in JCR, CiteScore in Scopus, and SJR). They describe journals, not individual articles or researchers, and are no substitute for reading the work.
Research visibility
- Use an ORCID iD and a consistent name and affiliation.
- Deposit your works in the repository and publish under open licences where possible.
- Use social media and blogs to share your work, without exaggeration.
Research collaboration
Co-authorship maps reveal collaboration networks between researchers, institutions and countries, and help identify potential partners.
Institutional research output
Analysing the University’s output depends on good metadata and consistent affiliation. Without them, many works are missing from the statistics.
Responsible metrics
The San Francisco Declaration (DORA) and the Leiden Manifesto call for not judging research by journal indicators alone, combining quantitative and qualitative assessment, and being transparent about methods.
Illustrative example
Demonstration data only – not real data. If a hypothetical researcher has five papers cited 10, 8, 5, 4 and 3 times, then h = 4, because four papers have at least four citations each.
| Paper (hypothetical) | Citations (hypothetical) |
|---|---|
| A | 10 |
| B | 8 |
| C | 5 |
| D | 4 |
| E | 3 |