Direct comparison
Impact Factor vs h-index — what is the difference?
The Journal Impact Factor measures a journal; the h-index measures an individual researcher. Both are citation-based metrics, both are widely used as research-evaluation proxies, and both are criticised by DORA and CoARA for being applied beyond what they can validly measure.
Side-by-side comparison
| Dimension | Journal Impact Factor | h-index |
|---|---|---|
| What it measures | A journal | An individual researcher (or group, journal, etc.) |
| Definition | Mean citations in a year to the journal's citable items from the previous two years | h papers each cited at least h times |
| Proposed / published | Eugene Garfield (concept, 1955); published annually by Clarivate | Jorge E. Hirsch, 2005 (PNAS) |
| Source | Clarivate Journal Citation Reports (Web of Science data) | Computed from any citation database (Web of Science, Scopus, Google Scholar) |
| Time window | Two-year (a five-year variant also published) | Whole career to date — cumulative, never decreases |
| Database-dependent | Single source (Clarivate) | Yes — value differs markedly between Scopus, Web of Science, and Google Scholar |
| Known weaknesses | Skewed by a few highly cited papers; varies by field; not a measure of any single article | Favours long careers; ignores author position; field-dependent; insensitive to highly cited outliers |
| Field normalisation | None inherent — citation rates differ greatly by discipline | None inherent — not comparable across disciplines |
| Stance of DORA / CoARA | DORA explicitly urges stopping its use in hiring, promotion, and funding decisions | CoARA cautions against over-reliance; neither measures the quality of an individual's work |
Common questions
FAQ
Can I use the Journal Impact Factor to judge a single paper?+
No. The JIF is a journal-level average and is heavily skewed by a small number of highly cited articles, so it says little about any individual paper in that journal. DORA was founded specifically to discourage this misuse.
Why do I have different h-index values in different tools?+
The h-index depends entirely on which citations the underlying database has indexed. Google Scholar typically returns a higher value than Scopus or Web of Science because it indexes more sources, including preprints, theses, and grey literature.
Are these metrics comparable across disciplines?+
Not directly. Citation cultures differ — a strong h-index or journal in mathematics will look very different from one in molecular biology. Field-normalised indicators exist, but raw JIF and h-index values should not be compared across fields.
What do DORA and CoARA recommend instead?+
Both promote responsible assessment: judging research on its own merits using qualitative expert review, supported by a transparent and appropriate use of indicators, rather than relying on journal- or citation-based proxies such as the JIF or h-index.
Going deeper
