Bibliometrics

Bibliometrics is the quantitative study of documents and the patterns they form: who publishes, who cites whom, which journals attract which papers, and how ideas spread through a literature.

11 frameworks, from 1910 to today · Information Science · Social Sciences

still a working view ended · each leads to its entry below

Overview

Bibliometrics is the quantitative study of documents and the patterns they form: who publishes, who cites whom, which journals attract which papers, and how ideas spread through a literature1. Its practitioners count things—papers, citations, patents, words, authors—and then look for regularities in the counts. The stakes are practical. A librarian deciding which journals to keep, a funder judging a grant program, a historian tracing how a discovery was received, and a university administrator ranking departments all draw on bibliometric evidence. The field matters because it turns the vast written record of scholarship into something measurable, and because those measurements now shape careers and institutions. That power has made the field's own history a story of expanding ambition and growing caution.

Precursors: Statistical Bibliography Before the Name

Counting books and citations is older than the discipline. In the early twentieth century, librarians and documentalists began compiling statistics on publication output. The term statistical bibliography—the quantitative description of books and journals—was used for such work well before "bibliometrics" existed1. In the 1920s, E. Wyndham Hulme, a librarian at the British Patent Office, analyzed the growth of the International Catalogue of Scientific Literature and the English Catalogue of Books, treating publication counts as evidence about the state of science. His approach was descriptive: he tabulated titles by subject and year and looked for trends.

The more consequential early work was citation counting. In 1927, Gross and Gross published a study in Science counting which journals were most cited in The Journal of the American Chemical Society, proposing that citation frequency could guide library subscription decisions. This was a practical, local use of citation data—not yet a general theory of science.

The Informetric Laws: Regularities in Skewed Distributions

The field's first theoretical core came from noticing that bibliometric quantities are almost never evenly spread. A few journals publish most of the important papers; a few authors write most of the articles; a few words account for most of the text. These skewed distributions became the field's signature finding.

In 1926, Alfred J. Lotka, an actuary at the Metropolitan Life Insurance Company, examined author names in Chemical Abstracts and the Geschichtstafeln der Physik and found that the number of authors producing n papers falls roughly as 1/n². This became known as Lotka's law: a small elite of prolific authors and a long tail of one-time contributors1. In 1934, Samuel C. Bradford, a librarian at the Science Museum in London, studied applied geophysics and lubrication literature and found that articles on a subject cluster in a small core of journals, with diminishing returns as one moves outward through more peripheral journals. Bradford's law of scattering described this core-and-zone structure2. In 1949, S. R. Ranganathan, the Indian librarian and classification theorist, articulated library science principles that influenced how collections were understood, though his direct contribution to bibliometric law is less central than Lotka's and Bradford's.

These laws were empirical regularities, not derivations. Their appeal was that they held across very different literatures, suggesting something deep about how scientific communication is organized. Later work—notably by Derek de Solla Price and by Benoit Mandelbrot on power laws—connected them to broader statistical patterns, but the original formulations were inductive and descriptive.

Citation Indexing: Making the Web of References Visible

The single most transformative development was the citation index. The idea is simple: instead of indexing only the content of a paper, index the references it makes, so that a reader can move forward in time from an older paper to everything that cites it. Eugene Garfield, trained in chemistry and library science, founded the Institute for Scientific Information (ISI) in Philadelphia in 1960 and launched the Science Citation Index in 1963. The index listed, for each cited article, the articles that cited it—a "forward" search tool that complemented traditional subject indexes.

Garfield's motivation was partly practical: citation indexing would let researchers find relevant work without relying on the vocabulary of subject headings, which often failed to match the terms authors actually used. The Science Citation Index made citation data a commercial product and, unintentionally, a research resource. Once the data existed in machine-readable form, bibliometricians could analyze it at scale.

The Journal Impact Factor, introduced by Garfield and Irving Sher in the early 1960s and published in the Journal Citation Reports, was originally a tool for selecting journals for the citation index. It measured the average number of citations in a given year to articles published in a journal in the preceding two years134. Its later career as an evaluation metric was not its original purpose, a point Garfield himself repeatedly emphasized.

Co-citation and Bibliographic Coupling: Mapping Intellectual Structure

Citation data also made it possible to map how literatures relate. In 1963, the Russian information scientist M. M. Kessler described bibliographic coupling: two papers are coupled if they share references, and the more references they share, the more closely related they are. In 1973, Henry Small, working at ISI, introduced co-citation analysis: two papers are co-cited when a later paper cites both, and the frequency of co-citation measures how often the pair is treated as belonging together. Small used co-citation to cluster papers into "specialties" and to trace how research fronts form and dissolve.

These techniques turned citation data into a structural map of science. They were among the first computational approaches to the sociology of knowledge, and they made visible the invisible colleges that Derek de Solla Price had described in Little Science, Big Science (1963)—the informal networks through which scientists exchange ideas before formal publication.

Naukometriya and the Science of Science

In the Soviet Union, a parallel tradition developed under the name naukometriya (the measurement of science)5. In 1969, V. V. Nalimov and Z. M. Mulchenko published Naukometriya: The Study of the Development of Science as an Information Process, treating science as an information system whose growth could be measured statistically5. The Soviet approach was closely tied to science policy and planning, and it shared with Western bibliometrics an interest in publication counts and growth curves.

In the United States and Western Europe, the science of science movement of the 1960s and 1970s brought together historians, sociologists, and information scientists. Derek de Solla Price's Little Science, Big Science (1963) argued that scientific output grows exponentially and that this growth has structural consequences. The journal Scientometrics was founded in 1978, giving the field a dedicated venue56. The term scientometrics came to denote the quantitative study of science specifically, while bibliometrics remained the broader term for the quantitative study of documents.

Research Evaluation and Indicator Metrics

As bibliometric data became available and computers made analysis cheap, the field's center of gravity shifted toward research evaluation. Governments and funders wanted to know whether their investments produced results. Universities wanted to compare themselves. The result was a proliferation of indicators: publication counts, citation counts, the h-index (proposed by Jorge Hirsch in 2005, defined as the largest number h such that an author has h papers each cited at least h times), and journal-level metrics like the Impact Factor7.

The appeal of these indicators was their apparent objectivity and comparability. Their problems were equally apparent to careful practitioners: citation practices vary by field, so cross-field comparisons are misleading; the h-index rewards longevity and penalizes early-career researchers; the Impact Factor measures journals, not articles, and is skewed by a small number of highly cited papers. The field's own literature documented these distortions, but the indicators spread faster than the caveats.

Altmetrics: Measuring Beyond Citations

In the 2010s, the rise of social media and online platforms created new traces of scholarly attention. Altmetrics—alternative metrics—sought to capture mentions, shares, downloads, and saves that citation counts miss. The term was coined in a 2010 altmetrics manifesto by Jason Priem and colleagues. The promise was a faster, broader view of impact; the problem was that online attention is noisy, platform-dependent, and often unrelated to scholarly significance. Altmetrics became a complement to citation metrics rather than a replacement, and the field continues to debate what they actually measure.

Open Bibliometric Data Infrastructure

The commercial dominance of citation data—first ISI, then its successor Clarivate, and later Elsevier's Scopus—created a barrier: bibliometric research depended on expensive proprietary databases. The open bibliometric data movement sought to change this. OpenAlex, launched in 2022 by the nonprofit OurResearch, offered a free, open catalogue of scholarly works, authors, and institutions8. Crossref, the DOI registration agency, provided open metadata on publications9. ORCID gave researchers persistent identifiers10. These infrastructures made large-scale bibliometric analysis possible without a subscription, and they shifted the field's center of gravity toward reproducibility and transparency.

Computational and Network Bibliometrics

The availability of large datasets and cheap computing transformed the field's methods. Network analysis became central: citation networks, co-authorship networks, and co-citation networks were analyzed with tools from graph theory and physics. The science of science was revived as a computational discipline, with researchers like Albert-László Barabási and colleagues applying network science to collaboration and citation patterns. Topic modeling and natural language processing allowed bibliometricians to map thematic structures at scale. The field moved from counting to modeling, and from description to prediction—for example, predicting which papers will be highly cited.

Responsible Metrics: The Critique Turns Inward

The most recent turning point is a critical one. As bibliometric indicators became embedded in hiring, promotion, and funding decisions, researchers began to document their unintended consequences: gaming, field bias, the discouragement of risky or interdisciplinary work, and the reduction of scholarly value to a single number. The San Francisco Declaration on Research Assessment (DORA) in 2012 and the Leiden Manifesto in 2015 articulated principles for the responsible use of metrics11112. The field's own professional organizations, notably the International Society for Scientometrics and Informetrics, took up the question of how bibliometricians should advise against misuse of their tools. Responsible metrics became a recognized area of research and advocacy, arguing that indicators should inform, not replace, expert judgment.

The Present: Plural Methods and Unresolved Questions

Bibliometrics today is a plural field. It includes descriptive studies of publication and citation patterns, evaluative work for institutions and funders, computational modeling of science as a complex system, and critical scholarship on the politics of measurement. The informetric laws remain foundational but are now understood as approximate regularities rather than universal laws1. Citation indexing, once a commercial monopoly, is increasingly open. Altmetrics and responsible metrics represent two responses to the same problem: the gap between what can be counted and what matters.

Open questions persist. How should metrics account for interdisciplinary research, which citation patterns poorly capture? How can bibliometric data be made truly global, when coverage of scholarship from the Global South remains uneven? And how should the field respond when its indicators are used in ways its own research warns against? These questions are not new—they echo the concerns of Hulme and Bradford about what counting can and cannot tell us—but they are now urgent, because the counts shape who gets to do science.

References

  1. Bibliometrics (introduction, Definition, Development of bibliography statistics (1910–1945), Bibliometric reductionism, metrics, and structuration of a research field (1960–1990), Infrastructure for open citation data), Wikipedia
  2. Bradford's law, Wikipedia
  3. Citation analysis (History), Wikipedia
  4. Impact factor, Wikipedia
  5. Scientometrics (Historical development), Wikipedia
  6. Scientometrics (journal), Wikipedia
  7. H-index, Wikipedia
  8. OpenAlex (introduction, History), Wikipedia
  9. Crossref, Wikipedia
  10. ORCID, Wikipedia
  11. Leiden Manifesto for Research Metrics, Wikipedia
  12. San Francisco Declaration on Research Assessment, Wikipedia

The frameworks of Bibliometrics

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1910-1955

    Statistical Bibliography

    How many books and journals are published, by subject and year?

    Fly to its tower in Nousfarer

  2. 1926-Present

    Informetric Laws

    What regularities govern how papers and citations are distributed?

    Fly to its tower in Nousfarer

  3. 1963-Present

    Citation Indexing

    How can a reader find every work that cites an older paper?

    Fly to its tower in Nousfarer

  4. 1963-Present

    Citation-Based Mapping of Intellectual Structure

    How are research specialties and intellectual structures mapped?

    Fly to its tower in Nousfarer

  5. 1963-Present

    Science of Science

    How does science grow and organize as a measurable system?

    Fly to its tower in Nousfarer

  6. 1969-1990

    Naukometriya

    How can the development of science be measured and planned?

    Fly to its tower in Nousfarer

  7. 1975-Present

    Research Evaluation and Indicator Metrics

    How should research performance be measured and compared?

    Fly to its tower in Nousfarer

  8. 1998-Present

    Computational and Network Bibliometrics

    How do citation and collaboration networks reveal scholarly structure?

    Fly to its tower in Nousfarer

  9. 2010-Present

    Altmetrics

    What traces beyond citations show scholarly attention?

    Fly to its tower in Nousfarer

  10. 2012-Present

    Open Bibliometric Data Infrastructure

    How can bibliographic and citation data be made openly available?

    Fly to its tower in Nousfarer

  11. 2012-Present

    Responsible Metrics

    How should research metrics be used fairly and carefully?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Statistical BibliographyInformetric LawsCitation IndexingCitation-Based Mapping of Intellectua…Science of ScienceNaukometriyaResearch Evaluation and Indicator Met…Computational and Network Bibliometri…AltmetricsOpen Bibliometric Data InfrastructureResponsible Metrics

Beyond the field

Where Bibliometrics's ideas came from in other fields, and what they led to there: the strongest 12 of 29 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Bibliometrics is a planet in Nousfarer

Its city has a tower for each of these 11 frameworks. Land, walk in, and climb one concept at a time.

Land on Bibliometrics

Related fields

Fields its ideas are linked to

More in Information Science

See where Bibliometrics sits on the Map of Human Knowledge