What is the median h-index of a senior researcher?

Across all fields, a researcher 21+ years into publishing sits at a median h-index of 4 on OpenAlex data (interquartile range 1–10). Field medians at that stage range from 1 (humanities) to 6 (biology, physics). Professor-level profiles sit in the top deciles: h≈20 is the all-fields 90th percentile at 21+ years.

Why are these benchmarks lower than typical 'good h-index' numbers?

Two reasons. First, the sample measures all publishing researchers in a field — most are not professors, and most researchers publish briefly before leaving academia. Second, OpenAlex citation counts are more conservative than Google Scholar, which also indexes theses, books, preprints and grey literature. Scholar values typically run higher.

How was the sample drawn?

Reproducible random samples from the OpenAlex API (sample+seed parameters, seeds published in the methodology). 5,000–8,000 authors per field group, filtered to works_count above 5, assigned to their primary OpenAlex field by top publication topic, and bucketed by years since first publication derived from counts_by_year. Per-cell statistics use a 1% two-sided trim.

Can I reproduce this dataset?

Yes. The methodology section lists the exact API filters, sample sizes and seeds per field group. Running the same queries against OpenAlex today reproduces the sample (author metrics drift slowly over time, so exact percentiles may differ slightly on re-runs).

H-Index & i10 Benchmarks 2026: OpenAlex Sample

Name: H-Index & i10 Benchmarks 2026
Creator: Calcuja Research
Published: 2026-10-27
License: https://creativecommons.org/licenses/by/4.0/

Every "what is a good h-index" answer on the web traces back to the same handful of rules of thumb — because nobody had published a measured benchmark by field and career stage. This dataset fixes that: we drew seeded random samples of OpenAlex authors, assigned each to their primary field, bucketed them by years since first publication, and computed percentiles per cell.

// Key Findings

The median h-index across all publishing researchers is just 2–4 — even at 21+ years since first publication (median 4, IQR 1–10). Most publishing researchers are not professors; many publish briefly and leave academia
Professor-track profiles live in the top decile: h≈20 marks the 90th percentile at 21+ years across all fields — remarkably close to Hirsch's (2005) "successful scientist" threshold
Field gaps are real and large: senior medians run from h=1 (humanities) to h=6 (biology, physics); i10 senior medians from 0 (social sciences, humanities) to 6 (biology)
i10 stays at zero far longer than intuition suggests: the median researcher has 0 papers at 10+ citations until roughly 6–10 years in — papers need years to accumulate citations
OpenAlex counts run lower than Google Scholar — Scholar also indexes theses, books, preprints and grey literature. Do not compare your Scholar numbers directly to these tables (see Limitations)

// Short answer

A good h-index depends on field and academic age. Measured medians (all fields): 2 at 0–5 yrs, 3 at 6–20 yrs, 4 at 21+ yrs — top-10% marks: 5 / 8 / 12 / 20. For i10: medians 0–2 across stages, top decile reaching 3–30. "Good" is best read as the top quartile of your field × stage cell below.

// Dataset — h-index

H-Index Percentiles by Field × Years Since First Publication

Median (p50) with interquartile range (p25–p75) and the 90th percentile. n = authors in cell after 1% trim. Field = the author's primary OpenAlex field.

Field	Stage	n	p25	p50	p75	p90
All fields	0–5 yrs	5,823	1	2	3	5
	6–10 yrs	8,508	1	3	5	8
	11–20 yrs	18,232	1	3	7	12
	21+ yrs	31,102	1	4	10	20
Medicine & Biology OA 27, 24, 30, 36, 29, 35, 34	0–5 yrs	945	1	2	3	5
	6–10 yrs	1,308	2	4	6	9
	11–20 yrs	2,626	2	5	9	14
	21+ yrs	4,407	2	5	12	24
Life Sciences OA 13, 11, 28	0–5 yrs	517	1	2	4	5
	6–10 yrs	806	2	4	6	9
	11–20 yrs	1,763	2	5	9	15
	21+ yrs	2,944	3	6	14	26
Chemistry OA 16, 15	0–5 yrs	236	1	2	4.2	6
	6–10 yrs	320	3	5	7	11
	11–20 yrs	735	3	5	9	15
	21+ yrs	1,786	3	6	13	24
Physics & Astronomy OA 31	0–5 yrs	380	0	2	3	5
	6–10 yrs	442	2	4	6	11
	11–20 yrs	999	2	5	10	18
	21+ yrs	2,277	2	6	14	29
Mathematics OA 26	0–5 yrs	294	0	1	2	3
	6–10 yrs	449	1	3	4	7
	11–20 yrs	912	2	4	7	10
	21+ yrs	1,703	2	5	11	19
Computer Science OA 17	0–5 yrs	631	1	2	3	5
	6–10 yrs	741	2	3	5	7
	11–20 yrs	1,327	2	4	7	11
	21+ yrs	1,698	2	4	11	18
Engineering OA 22, 25, 21	0–5 yrs	832	1	3	4	6
	6–10 yrs	1,311	2	4	6	10
	11–20 yrs	2,724	2	4	8	14
	21+ yrs	4,370	2	5	10	20
Economics & Finance OA 20, 14	0–5 yrs	352	0	1	2	3
	6–10 yrs	574	1	2	4	6
	11–20 yrs	1,292	1	2	5	8
	21+ yrs	1,567	1	3	7	15
Social Sciences OA 33, 32, 18	0–5 yrs	896	0	1	2	3
	6–10 yrs	1,333	1	2	3	5
	11–20 yrs	3,001	1	2	4	6
	21+ yrs	4,549	0	2	5	11
Humanities OA 12	0–5 yrs	284	0	0	1	2
	6–10 yrs	593	0	1	2	3
	11–20 yrs	1,476	0	1	2	4
	21+ yrs	3,328	0	1	3	6
Environmental & Earth OA 23, 19	0–5 yrs	456	1	2	3	5
	6–10 yrs	631	2	4	6	8
	11–20 yrs	1,377	2	4	8	14
	21+ yrs	2,473	2	5	12	23.6

Field-level rows for all 26 OpenAlex fields (including Psychology, Neuroscience, Business, Nursing and more) are in the CSV download below.

// Dataset — i10-index

i10-Index Percentiles by Field × Years Since First Publication

The i10-index counts works with ≥10 citations. Median values sit at or near zero in many cells — that is a real property of the metric, not missing data.

Field	Stage	n	p25	p50	p75	p90
All fields	0–5 yrs	5,823	0	0	1	3
	6–10 yrs	8,508	0	1	4	7
	11–20 yrs	18,232	0	2	6	14
	21+ yrs	31,102	0	2	10	30
Medicine & Biology	0–5 yrs	945	0	0	2	4
	6–10 yrs	1,308	0	2	4	8
	11–20 yrs	2,626	1	3	8	16
	21+ yrs	4,407	0	4	13	39
Life Sciences	0–5 yrs	517	0	1	2	4
	6–10 yrs	806	0	2	5	9
	11–20 yrs	1,763	1	4	9	19
	21+ yrs	2,944	1	6	17.8	43
Chemistry	0–5 yrs	236	0	1	3	5
	6–10 yrs	320	1	3	7	12
	11–20 yrs	735	1	4	9	19
	21+ yrs	1,786	1	5	15	46
Physics & Astronomy	0–5 yrs	380	0	0	2	4
	6–10 yrs	442	0	2	5	13
	11–20 yrs	999	0	3	11	24
	21+ yrs	2,277	1	5	18	56.8
Mathematics	0–5 yrs	294	0	0	0	1.1
	6–10 yrs	449	0	1	2	5
	11–20 yrs	912	0	1	5	11
	21+ yrs	1,703	0	3	12	29.2
Computer Science	0–5 yrs	631	0	0	1	3
	6–10 yrs	741	0	1	4	7
	11–20 yrs	1,327	0	2	6	12
	21+ yrs	1,698	0	3	11	28
Engineering	0–5 yrs	832	0	1	3	5
	6–10 yrs	1,311	0	2	5	10
	11–20 yrs	2,724	0	3	8	18
	21+ yrs	4,370	0	3	11	31
Economics & Finance	0–5 yrs	352	0	0	0	2
	6–10 yrs	574	0	0	2	4
	11–20 yrs	1,292	0	1	3	8
	21+ yrs	1,567	0	1	6	20
Social Sciences	0–5 yrs	896	0	0	0	1
	6–10 yrs	1,333	0	0	1	3
	11–20 yrs	3,001	0	0	2	5
	21+ yrs	4,549	0	0	3	12.2
Humanities	0–5 yrs	284	0	0	0	0
	6–10 yrs	593	0	0	0	1
	11–20 yrs	1,476	0	0	0	2
	21+ yrs	3,328	0	0	1	4
Environmental & Earth	0–5 yrs	456	0	0	2	3.3
	6–10 yrs	631	0	2	4	7
	11–20 yrs	1,377	0	2	7	18
	21+ yrs	2,473	0	3	13	37

⬇ Download CSV (all 26 fields)

// Interpretation

How to Read These Numbers

Population, not professor-track

These percentiles describe every researcher publishing in a field — including people who left academia, adjuncts, industry researchers and single-paper authors. Medians are therefore far below what hiring committees expect of professors. If you are on a tenure track, compare against the p75–p90 band of your cell, not the median.

Academic age ≠ job title

Stages are years since first publication, not "Assistant Professor" or "Postdoc". A 12-year industry researcher and a 12-year associate professor land in the same bucket. As a rough mapping: 0–5 ≈ PhD student, 6–10 ≈ postdoc/early faculty, 11–20 ≈ mid-career, 21+ ≈ senior.

OpenAlex < Google Scholar

OpenAlex indexes peer-reviewed venues and misses much of what Scholar counts — theses, books, conference papers in many fields, preprints and grey literature. Your Scholar h-index and i10 will typically read higher. These tables benchmark OpenAlex values against OpenAlex values.

Where the professor thresholds sit

Hirsch's (2005) anchors — h≈12 for tenure, h≈20 for a successful scientist after ~20 years — map to the 75th–90th percentile of our senior cells across fields. That consistency is a useful cross-check: the rules of thumb describe the top of the distribution, and now the distribution itself is measured.

// Methodology

Data & Methodology

Population. Authors in the OpenAlex index with works_count > 5. OpenAlex computes h-index and i10-index (works with ≥10 citations) per author from its own citation graph.

Sampling. Reproducible random draws via the OpenAlex API sample parameter with fixed seeds (listed below). 5,000–8,000 authors per field group; 63,907 unique authors drawn on 2026-10-27, of which 63,665 had a determinable first-publication year. Snapshot: October 2026.

Field assignment. Each author is assigned to the OpenAlex field of their highest-count topic (topics[0].field). Authors whose primary field falls outside a group do not count toward that group — a social scientist who occasionally publishes humanities work is a social scientist. Field-group composition is noted under each table row (OA field IDs).

Career stage. Academic age = 2026 − year of the author's earliest indexed work, taken from counts_by_year. Buckets: 0–5, 6–10, 11–20, 21+ years. This is publication age, not years since PhD and not job title.

Statistics. Per field × stage cell: two-sided 1% trim, then p10/p25/p50/p75/p90 (linear interpolation). Cell sizes range from 236 (Chemistry, 0–5 yrs) to 31,102 (All fields, 21+); every cell reports its n.

Reproducibility. Same filters + same seeds = same sample (author metrics drift slowly over time). Per-group seeds and sample sizes:

Medicine & Biology — fields 27|24|30|36|29|35|34, n=5,000, seed 20261001
Life Sciences — fields 13|11|28, n=5,000, seed 20261002
Chemistry — fields 16|15, n=5,000, seed 20261003
Physics & Astronomy — field 31, n=5,000, seed 20261004
Mathematics — field 26, n=8,000, seed 20261005
Computer Science — field 17, n=5,000, seed 20261006
Engineering — fields 22|25|21, n=5,000, seed 20261007
Economics & Finance — fields 20|14, n=6,000, seed 20261008
Social Sciences — fields 33|32|18, n=6,000, seed 20261009
Humanities — field 12, n=8,000, seed 20261010
Environmental & Earth — fields 23|19, n=6,000, seed 20261011

Limitations. OpenAlex coverage is strong in peer-reviewed journals but thin for books, theses and much humanities output — humanities numbers understate those fields' real profiles. Automatic author disambiguation occasionally merges or splits profiles; the 1% trim reduces its influence on percentiles. The sample describes the full publishing population, not faculty-track subpopulations. 56% of sampled authors had at least one indexed work in 2023 or later; the rest have frozen metrics — cumulative indices remain valid career-stage values but underrepresent currently active output.

// Sources & References

Priem, J., Piwowar, H., & Orr, R. (2022). OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts. 26th International Conference on Science, Technology and Innovation Indicators (STI 2022). arXiv:2205.01833
Hirsch, J. E. (2005). An index to quantify an individual's scientific research output. PNAS, 102(46), 16569–16572. doi:10.1073/pnas.0507655102
OpenAlex API documentation. Author entity fields (summary_stats.h_index, summary_stats.i10_index, counts_by_year, topics), sample/seed parameters. docs.openalex.org

Cite this dataset:
Calcuja Research (2026). H-Index & i10 Benchmarks 2026: Measured OpenAlex Sample by Field and Career Stage. Calcuja.com. https://calcuja.com/research/openalex-benchmarks-2026/

// Use the data