Tracing the Evolution of Informetrics: From Bibliography to Modern Data Analysis

Every time a researcher counts how many papers a university published, ranks journals by influence, or measures which scientists are the most cited, they are using a discipline that has been quietly growing for over a century. That discipline is informetrics, and its story stretches from dusty 19th-century bibliographies to today’s data dashboards. Understanding how it evolved helps explain why we measure knowledge the way we do, and why these methods matter so much in research evaluation, library management, and science policy.

What informetrics actually means 🔗

Before tracing the history, it helps to fix the vocabulary, because three terms are often used together and sometimes interchangeably. Bibliometrics is the application of mathematical and statistical methods to books and other forms of communication. Scientometrics studies the quantitative aspects of science as a communication system, including citations, research performance, and science policy. Informetrics is the broadest of the three. It is defined as the study of the quantitative aspects of information in any form, not just records or bibliographies, and in any social group, not just scientists.

Informetrics is best understood as an extension and evolution of traditional bibliometrics and scientometrics. Leo Egghe described it as the umbrella term covering all the “metrics” related to information science, including bibliometrics, scientometrics, and the newer field of webometrics. So when we trace the evolution of informetrics, we are really following a single intellectual current that gathered these streams together over time.

Historical context: the early origins 🔗

The roots of measuring literature run deeper than most people expect. Bibliometric methods have a history extending over a century, with the study of how knowledge accumulates as a measurable, countable thing.

The first statistical studies 🔗

A 1917 study by Cole and Eales, titled The History of Comparative Anatomy, is widely regarded as the first bibliometric study. It analysed publications statistically rather than just describing their content. A few years later, in 1923, E. Wyndham Hulme introduced the expression “statistical bibliography,” giving the activity its first proper name. This term would survive for decades until a better one came along.

The Indian contribution: librametry 🔗

An important early milestone came from S.R. Ranganathan, often called the father of library science in this country. In 1948, at the Aslib conference in Leamington Spa, he proposed the term “librametry” (also written librametrics). He suggested developing it along the lines of biometry, econometry, and psychometry. The word joins “libra,” meaning library, with “metry,” meaning measurement.

Ranganathan had actually been practising librametric techniques as far back as 1925 as librarian of the Madras University Library, using counting and measurement to solve everyday problems of book selection, staffing, and reader services. His idea focused on the quantitative study of three elements: books, readers, and staff. Although librametry never gained wide adoption outside India, it was historically one of the first attempts to put a name to this kind of measurement, and it placed an Indian scholar among the pioneers of the field.

From statistical bibliography to bibliometrics 🔗

The decisive renaming happened in 1969, when Alan Pritchard proposed replacing the clumsy phrase “statistical bibliography” with bibliometrics. He defined it as studies that seek to quantify the process of written communication. This single change of vocabulary helped the field gain a clearer identity and made it easier for researchers to talk about what they were doing.

Post-war developments: information becomes a science 🔗

The years after World War II transformed how scholars thought about information itself. The war had pushed mathematics, engineering, and communication into rapid development, and two figures in particular gave information a scientific foundation.

Claude Shannon and information theory 🔗

In 1948, Claude Shannon published A Mathematical Theory of Communication in the Bell System Technical Journal. This paper established information theory as a new scientific discipline. Shannon treated information as something that could be measured in precise units, broken into bits, and transmitted through a system. His ideas about entropy, data compression, and coding gave researchers a rigorous way to think about information as a quantity rather than a vague concept.

Shannon’s work became widely known partly thanks to Warren Weaver, whose accessible interpretation reached a broad scientific audience when their writings were published together as The Mathematical Theory of Communication. This idea that information has measurable value laid groundwork that informetrics would later build on.

Norbert Wiener and cybernetics 🔗

The same year, 1948, Norbert Wiener published Cybernetics: Or Control and Communication in the Animal and the Machine. Working independently of Shannon, Wiener treated communication engineering as a branch of statistical physics and applied this thinking to the concept of information, arriving at a formula remarkably close to Shannon’s.

Wiener’s vision was deliberately broad. He saw information, feedback, and control as connecting ideas across biology, machines, and society. His work influenced a generation of scientists and helped seed entire fields, including information theory, computer learning, and artificial intelligence. Together, the contributions of Shannon and Wiener mark the beginning of the information age and reshaped how disciplines from psychology to library science approached the study of information.

The bibliometric laws: the mathematical backbone 🔗

While information theory was developing, three empirical laws were quietly becoming the mathematical foundation of bibliometrics and informetrics. They are the laws used most often to model the distribution of information phenomena. What unites them is a striking pattern: in information systems, a small number of sources account for a large share of output, while a long tail of sources contributes very little.

Lotka’s law: author productivity 🔗

Formulated by Alfred J. Lotka in 1926, this law describes the frequency of publication by authors in a field. It states that a few authors produce most of the papers while the majority of authors publish only one or two. In its classic form, the number of authors producing a given number of articles falls off in a predictable inverse-square pattern, so that for every author who writes many papers, there are far more who write just one.

Bradford’s law: journal scattering 🔗

Samuel C. Bradford described how articles on a particular subject are scattered across journals. Bradford’s law addresses the decay in the yield of journals on a given subject, showing that journals can be divided into zones of decreasing productivity. A small “core” of journals carries a large share of the relevant articles, while progressively larger groups of journals carry the same number of articles each. This insight has profound practical value: it helps libraries decide which journals are essential to subscribe to, and it explains the pressure on researchers to publish in a small set of core journals. Of the three laws, Bradford’s has found the widest application.

Zipf’s law: word frequency 🔗

Named after the linguist George Kingsley Zipf, this law states that in a large sample of text, the frequency of any word is inversely proportional to its rank in the frequency table. A handful of words appear very often, while most words are used rarely. Zipf famously illustrated the pattern using James Joyce’s Ulysses, and he explained it through what he called the Principle of Least Effort. In practice, studies using Zipf’s law are more time-consuming because they involve counting thousands of words, which is why computers have become essential for this kind of analysis.

The differences between the three laws lie in the data each examines: Lotka counts authors, Bradford counts journals, and Zipf counts words. Yet all three produce similar probability distributions, and researchers have shown they share a common underlying mathematical relationship. Even though they were developed in the era of print journals, these classical laws continue to inform modern analysis in the age of digital publishing.

Key milestones in the journey 🔗

From these foundations, informetrics matured through a series of developments that turned scattered ideas into an organised discipline.

Citation indexing and the Science Citation Index 🔗

One of the most transformative milestones was Eugene Garfield’s work on citation indexing. In a 1955 article in Science titled “Citation Indexes for Science,” Garfield set out the case for indexing the scientific literature through the citations that link papers together. He founded the Institute for Scientific Information (ISI), and in 1964 ISI published the first Science Citation Index, covering 613 journals and 1.4 million citations.

This was revolutionary. By recording and linking the references that authors attached to their papers, the index created what Garfield called an “association of ideas index.” It also provided the raw data needed for large-scale citation analysis. Garfield later developed the journal impact factor, and the Science Citation Index eventually evolved into today’s Web of Science. In a remarkable detail, Garfield’s idea of organising information through a network of citation links anticipated web hyperlinking and search-engine ranking by three decades.

The birth of scientometrics 🔗

The term scientometrics was coined by the Russian scholar Vasily Nalimov in the 1960s, derived from the Russian word “naukometriya.” It described the quantitative study of science as a process: its growth, structure, interrelationships, and productivity. The field drew heavily on the work of historian of science Derek de Solla Price, often called the father of scientometrics, whose studies of how science grows ran in parallel with Garfield’s citation work. In 1978, the first dedicated journal, Scientometrics, began publication, giving the field a permanent home.

The arrival of informetrics 🔗

The broadest term arrived in 1979, when researchers independently proposed “informetrics” to capture the quantitative study of information in all its forms. The information and communication technology revolution, including the rise of the internet, was changing how scientists worked and communicated, and a wider term was needed. Informetrics gained ground through a series of biennial international conferences, the first held in Belgium in 1987.

The establishment of international committees 🔗

The field reached organisational maturity in 1993 with the founding of the International Society for Scientometrics and Informetrics (ISSI) in Berlin, at the International Conference on Bibliometrics, Informetrics and Scientometrics. ISSI brought together professionals working across bibliometrics, scientometrics, technometrics, and webometrics, with members from over 30 countries on all five continents. Notably, one of the early conferences in this series was held in Bangalore in 1991, again placing India on the map of this global field.

Why this evolution still matters 🔗

The journey from statistical bibliography to modern informetrics is not just a story of changing names. Each stage added a layer of capability. The early statistical studies showed that literature could be counted. The bibliometric laws revealed hidden patterns in how knowledge is produced and distributed. Shannon and Wiener gave information a scientific basis. Garfield’s citation index supplied the data, and international societies gave the field structure and standards.

Today these foundations support the indicators that shape real decisions: the h-index used to evaluate individual researchers, journal impact factors that influence where scientists publish, and the data governments use to allocate research funding. The growth of databases turned citation analysis from a painstaking manual task into something that can be done at massive scale, and the field has now extended onto the web through webometrics and altmetrics. What began as a way to manage library collections has become a core tool for understanding the entire system of scholarly knowledge.

What do you think? If a small core of authors, journals, and words consistently dominates the output across so many fields, what does that pattern reveal about how human knowledge actually grows? And as research increasingly happens online, do you think the classical laws of Bradford, Lotka, and Zipf will still hold, or will the digital age demand entirely new measures?

References

1. https://www.sciencedirect.com/science/article/abs/pii/S1751157707000740
2. https://link.springer.com/article/10.1023/A:1017919924342
3. https://ebooks.inflibnet.ac.in/liscp10/chapter/librametry-bibliometrics-scientometrics-informetrics-and-webometrics-historical-development/
4. https://www.sciencedirect.com/topics/social-sciences/informetrics
5. https://www.historyofinformation.com/detail.php?id=664
6. https://en.wikipedia.org/wiki/Norbert_Wiener
7. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6092626/
8. https://tefkos.comminfo.rutgers.edu/Courses/e530/Readings/Jayroe%20Bibliometrics%20for%20Dummies%202008.pdf
9. https://arxiv.org/pdf/2102.09182
10. https://pmc.ncbi.nlm.nih.gov/articles/PMC12476466/
11. https://ebooks.inflibnet.ac.in/liscp10/chapter/classical-law-of-bibliometrics/
12. https://www.historyofinformation.com/detail.php?id=817
13. https://clarivate.com/academia-government/the-institute-for-scientific-information/history/
14. https://en.wikipedia.org/wiki/International_Society_for_Scientometrics_and_Informetrics