Examples
Worked examples
- Is an instance
A bibliometric study reports h-index distributions across a field as context for further analysis.
- Is an instance
A scholar discloses their h-index with a caveat about field norms in a reflective piece.
- Is an instance
A researcher with papers cited 42, 23, 19, 8, 5, and 1 times has an h-index of 5, because five of their papers have at least five citations each, but the sixth paper's single citation doesn't clear a sixth.
Counter-examples
Looks similar, but isn't
- Not an instance
Comparing h-index across disciplines as a ranking criterion in hiring.
- Not an instance
Setting an h-index threshold for promotion eligibility.
Editorial commentary
The h-index is a single-integer author-level metric defined as the largest number h for which a researcher has h publications that have each been cited at least h times. A researcher with an h-index of 12 has 12 papers with 12 or more citations each, and every remaining paper has fewer than 12. Jorge E. Hirsch proposed it in PNAS in 2005 as a way to summarise a physicist’s output in one number that neither a raw publication count nor a raw citation total could capture on its own.
What the number actually measures
The h-index deliberately conflates two different things — productivity (how many papers) and impact (how often they are cited) — into one figure, and it only rises when both move together. Publishing many uncited papers cannot raise it, and a single spectacularly cited paper cannot raise it either. That is its design intent and also the source of most criticism of it.
Read graphically, the h-index is the point where a researcher’s descending citation curve crosses the line y = x: papers ranked by citation count on one axis, citations on the other. Everything above and to the left of that crossing point is the “h-core” of a researcher’s work.
For the ranked-list procedure and a worked numeric example, see the step-by-step guide: how to calculate the h-index.
Properties that follow from the definition
- It can never decrease. Citations only accumulate, so an h-index is monotonic over a career — which means it correlates strongly with academic age and systematically favours senior researchers over early-career ones.
- It is insensitive to citations above the threshold. Once a paper clears the rank-h bar, further citations to it do not move the h-index at all. A paper with 42 citations and a paper with 6 count identically toward an h-index of 5.
- It is field-dependent. Citation density and typical co-authorship counts differ by orders of magnitude between, say, biomedicine and pure mathematics, so h-index values are not comparable across disciplines.
- It is database-dependent. The same person has a different h-index in Google Scholar, Scopus and Web of Science, because each indexes a different corpus.
- It says nothing about authorship position or contribution. A citation counts the same whether the researcher was sole author or one of a thousand.
Why the same researcher has three different h-indexes
An h-index is not a property of a researcher; it is a property of a researcher as recorded in one database at one moment. Google Scholar indexes preprints, theses, book chapters and conference material that Scopus and Web of Science exclude, so Scholar-derived h-indexes are typically the highest of the three. Scopus and Web of Science apply selective journal-inclusion criteria and generally return lower figures. Because of this, an h-index quoted without naming its source database is not an interpretable number — and mixing sources within one calculation produces a figure that means nothing at all.
Profile completeness compounds the problem: an h-index computed over a fragmented or duplicated author profile undercounts. See Scopus Author ID and Google Scholar profiles for how author records are consolidated in each system.
What the h-index is not appropriate for
The major responsible-research-assessment frameworks — the San Francisco Declaration on Research Assessment (DORA), the Leiden Manifesto, the Hong Kong Principles, and the CoARA Agreement on Reforming Research Assessment — all caution explicitly against using author-level citation metrics such as the h-index as a proxy for research quality in hiring, promotion, tenure or funding decisions. The objection is not that the arithmetic is wrong but that the number is being asked a question it cannot answer: it measures citation accumulation, not the quality, rigour or significance of the work.
There is also no threshold at which an h-index becomes “good” in the abstract — the figure only becomes interpretable once field, career stage and source database are all fixed. That question is treated separately under good h-index.
Related and derived metrics
Several metrics exist specifically to patch a known weakness of the h-index. The m-quotient (h-index divided by years since first publication) attempts to normalise for career stage. The g-index gives extra weight to highly cited papers, addressing the threshold-insensitivity problem. The i10-index, reported only by Google Scholar, is a simpler fixed-threshold count of papers with at least ten citations — compared side by side under h-index vs. i10-index. Journal-level metrics such as the Journal Impact Factor answer a different question entirely and are not substitutes; see impact factor vs. h-index.
References
- Hirsch JE, “An index to quantify an individual’s scientific research output”, PNAS 102(46):16569-16572, 2005.
- Bornmann L & Daniel HD, “The state of h index research”, EMBO Reports, 2009.
Harzing’s Publish or Perish is the tool most commonly used to compute this figure from Google Scholar data.
Also known as
Hirsch index · h
Machine-readable encodings
Use in your systems
<role vocab="credit"
vocab-identifier="https://casrai.org/dictionary/"
vocab-term="h-index"
vocab-term-identifier="https://casrai.org/dictionary/term/h-index" />{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"@id": "https://casrai.org/dictionary/term/h-index",
"name": "h-index",
"identifier": "https://casrai.org/dictionary/term/h-index",
"description": "A researcher-level metric proposed by Jorge Hirsch in 2005, defined as the largest number h such that the researcher has published h papers each cited at least h times.",
"inDefinedTermSet": "https://casrai.org/dictionary/domain/responsible-assessment#set",
"url": "https://casrai.org/dictionary/term/h-index",
"sameAs": [
"Hirsch index",
"h"
],
"license": "https://creativecommons.org/licenses/by/4.0/",
"publisher": {
"@id": "https://casrai.org/#organization"
},
"author": {
"@id": "https://casrai.org/#editorial-team"
},
"datePublished": "2026-05-21T02:22:55",
"dateModified": "2026-08-24T00:01:45",
"inLanguage": "en-GB",
"isAccessibleForFree": true
}
