Research metrics, clearly
H-index vs. i10-index: What Each Research Metric Actually Tells You
Understand how the h-index and i10-index are calculated, what they leave out, and why the source behind each number matters.
The h-index and i10-index both summarize a publication list, but they answer different questions. Neither metric measures research quality by itself, and neither has meaning without the database that supplied the publications and citations.
What is the h-index?
A researcher has an h-index of h when at least h of their publications have each received at least h citations.
Imagine citation counts sorted from highest to lowest:
42, 18, 11, 7, 5, 2
The fourth paper has at least four citations, while the fifth paper has five citations and also clears the fifth threshold. The sixth paper has only two citations, so this example has an h-index of 5.
Web of Science uses the same definition and calculates its researcher-profile h-index from Web of Science Core Collection publications and citations. Scopus illustrates the calculation as the point where a descending citation curve meets the line on which publications equal citations.
The h-index rewards a body of work that has received sustained citation attention. One extremely cited paper cannot raise it indefinitely, because additional papers must also cross the threshold.
What is the i10-index?
The i10-index is simply the number of publications with at least ten citations. Google Scholar reports it alongside total citations and h-index in author profiles. Google’s profile documentation explains that Scholar computes all-time and recent versions of these metrics.
For the example above, three papers have ten or more citations, so the i10-index is 3.
The i10-index is easier to explain, but its fixed threshold also makes it coarse. A paper with nine citations does not count, while a paper with ten does. A paper with 500 citations still contributes only one unit.
What the metrics leave out
Both metrics compress a publication history into one integer. They do not directly tell you:
- why a work was cited;
- whether the citation was supportive or critical;
- the researcher’s role in a multi-author paper;
- differences between fields and career stages;
- the quality, openness, or societal use of the work;
- whether citations came from independent authors or self-citation;
- which publications were absent from the database.
They also respond differently to time. Early-career researchers have had fewer opportunities to accumulate papers and citations. Fields with long publication cycles can grow differently from fast-moving fields. Comparisons across disciplines or career lengths therefore need much more context than either number provides.
Why your h-index changes between services
The calculation may be identical while the inputs differ. Google Scholar, Scopus, Web of Science, ResearchGate, and OpenAlex each maintain their own publication and citation graph.
OpenAlex, for example, exposes precomputed h_index and i10_index values on author records, but its documentation recommends using underlying works for rigorous analysis. OpenAlex common attributes describe these as convenience metrics derived from its graph.
If one source finds more publications or more citing documents, it can produce a higher h-index. If another source has tighter collection boundaries, the resulting number may be lower without being incorrectly calculated.
Which metric should you use?
Use the metric requested by the context and identify its source:
- For a Google Scholar profile or personal website, report the Scholar value and observation date.
- For an assessment that specifies Scopus or Web of Science, use that source rather than substituting a broader number.
- For personal monitoring, compare the same source over time.
- When presenting several sources, keep them side by side instead of averaging or adding them.
A clear label is more valuable than false precision: “h-index 14, Web of Science, checked 24 August 2026” is interpretable. “h-index 14” alone leaves important questions unanswered.
The short answer
The h-index measures the size of a researcher’s consistently cited body of work. The i10-index counts papers that have crossed ten citations. Both depend on the publications and citations known to a specific provider, so always keep the provider and date attached.