Revisiting Information: Understanding its Definition and Measurement

Information surrounds us constantly. It flows through our phones, shapes the decisions we make, and forms the foundation of how modern society works. Yet here is a curious problem: despite using the word “information” every single day, scholars have struggled for decades to agree on what it actually is. Much like energy, we can clearly see what information does, but pinning down a precise definition turns out to be surprisingly difficult. This challenge becomes especially important when we try to measure information, which is exactly what the field of informetrics attempts to do. This post revisits the concept of information, looking at why it resists easy definition, the different ways scholars approach it, and how researchers try to count something so slippery.

The peculiar status of information 🔗

Information has a strange status in the world of ideas. It is everywhere, yet almost impossible to define in a way that satisfies everyone. Think about how we treat energy in physics. We never see energy directly. We only observe its effects, such as heat, movement, or light. Energy is defined functionally, by what it does rather than what it is. Information behaves the same way. We recognize its presence through its effects on people, systems, and machines, but the thing itself stays out of reach.

This is not a minor academic puzzle. The word “information” gets used across dozens of disciplines, each with its own meaning. A telecommunications engineer, a biologist studying DNA, a librarian organizing a collection, and a journalist writing a report all use the term, but they rarely mean the same thing. In communication theory, for example, Claude Shannon defined information in a purely mathematical and abstract way, treating it as a measure of uncertainty in a transmitted message rather than as anything meaningful. This created lasting confusion, because most people associate information with a message that actually informs someone.

The deeper problem is finding a single, workable general definition that holds across all these uses. Some scholars argue this may be impossible, and that we should instead accept multiple definitions suited to different purposes. This is why understanding the various approaches to information matters so much. Rather than searching for one perfect definition, it helps to understand the major attitudes scholars take when they think about what information is.

Three attitudes toward understanding information 🔗

Scholars in information science generally adopt one of three broad attitudes when defining information. Each captures a different facet of the concept, and together they show why no single definition can do all the work.

The anthropocentric approach 🔗

The anthropocentric approach places the human being at the center. From this viewpoint, information only exists in relation to a person who interprets it. A signal, a document, or a data point is not information on its own. It becomes information only when a human mind receives it, understands it, and is changed by it.

Michael Buckland’s influential framework captures this human-centred thinking well. He distinguished three meanings of information: information-as-process, information-as-knowledge, and information-as-thing. Information-as-process refers to the act of being informed, the moment when what someone knows is changed. Information-as-knowledge is the intangible result of that change, the new understanding a person holds. Both of these are deeply tied to a human mind. Without a person to be informed and to hold knowledge, neither exists. This approach reflects how we usually talk about information in everyday life, as something that affects what people know and believe.

The organization-based approach 🔗

The organization-based approach shifts the focus away from individual minds and toward structure and pattern. Here, information is understood as a property of how things are arranged. Order, structure, and reduced randomness are the markers of information. The more organized a system is, the more information it can be said to contain.

This view connects to Buckland’s third category, information-as-thing, where the word is used to describe physical objects regarded as informative, such as data, texts, documents, and even objects or events. A book on a shelf, a database record, or a fossil in a museum all carry information because they embody a certain structure that can be examined. This is the approach most relevant to libraries and information systems, because such systems can only handle physical, organized objects. A library cannot store “knowledge” directly; it can only store documents arranged in a way that lets people retrieve and use them.

The action-oriented approach 🔗

The action-oriented approach defines information by its consequences. Information is whatever produces an effect, whatever leads to a change or guides an action. This is perhaps best summed up in the well-known idea that information is “a difference that makes a difference,” a phrase associated with the anthropologist Gregory Bateson.

In this view, something is information only if it does something. If a message arrives but changes nothing about what a system does or how a person behaves, it was not really information for that recipient. This approach is especially useful in fields like decision-making and management, where the value of information is judged by whether it leads to better choices. Research on information search before decisions shows that people actively seek information to reduce different kinds of uncertainty before they act, which fits neatly with this action-centred understanding.

Defining information through scholarly eyes 🔗

Beyond these broad attitudes, individual scholars have proposed specific definitions that have shaped the field. Two stand out for their lasting influence.

Belkin and the anomalous state of knowledge 🔗

Nicholas Belkin approached information from the standpoint of the person who needs it. In 1980 he introduced the idea of the anomalous state of knowledge, often shortened to ASK. The idea is that a person seeking information does so because there is a gap or anomaly in what they currently know. They recognize that something is missing, but they often cannot even formulate exactly what they need, because the gap is precisely an absence of knowledge.

For Belkin, information is whatever can resolve this anomalous state. This is a powerful reframing. It explains why people sometimes struggle to ask the right question when they search a library catalogue or a search engine. They are working with incomplete knowledge by definition. Information, in this sense, is the thing that fills the gap and brings the recipient’s state of knowledge back into a coherent whole. This human-centred, need-based view has been enormously influential in the study of how people seek and use information.

Mackay and information as a distinction that makes a difference 🔗

Donald MacKay offered a definition that focused on distinction and difference. He proposed that information is “a distinction which makes a difference,” a formulation that predates and closely resembles Bateson’s better-known phrasing. For MacKay, information is fundamentally about telling things apart. If something allows a system to distinguish one state from another, and that distinction matters to the system, then information has been conveyed.

MacKay was particularly interested in how information works in the brain and in communication between people, rather than only in the abstract logical patterns favoured by pure mathematics. His view connects the formal idea of information to meaning, asking not just whether a signal was received but whether it made a relevant difference to the receiver. This bridges the gap between the cold mathematics of signal transmission and the human experience of actually being informed.

Information and the reduction of uncertainty 🔗

Running through many of these definitions is a single powerful idea: information reduces uncertainty. When you are uncertain about something and then receive information, your uncertainty decreases. Before checking the weather, you do not know if it will rain. After checking, you do. The information has removed some of your uncertainty.

This principle sits at the heart of the mathematical theory of communication, where entropy measures uncertainty and information is understood as what resolves it. The more uncertain a situation is, the more information is needed to clear it up. This idea also helps explain information seeking behaviour. Studies of the information search process show that uncertainty typically decreases as a person moves through a search and gradually closes the gap in their knowledge. It is worth noting, though, that this is not always simple. Sometimes receiving new information can briefly increase uncertainty before it reduces it, as a system takes in something unexpected and has to make sense of it.

Informetrics and the measurement of information 🔗

If information is so hard to define, how can anyone hope to measure it? This is the central challenge taken up by informetrics. Informetrics is the study of the quantitative aspects of information in all its forms. It is the broad umbrella field that uses mathematical and statistical methods to study how information is produced, spread, and used, regardless of the form that information takes.

Informetrics did not appear out of nowhere. It grew out of and now encompasses several related fields. Bibliometrics studies the quantitative aspects of recorded information such as books and publications. Scientometrics studies the quantitative aspects of science, including citations and research output. Webometrics studies quantitative aspects of the World Wide Web. The term “informetrics” itself was coined by Otto Nacke in 1979, and it serves as the parent field that brings these specialized areas together under one approach.

What informetrics actually does 🔗

In practice, informetrics applies quantitative techniques to measure the records of human communication. This has real value in library and information work. Managers use these methods to evaluate library resources and services more objectively, to track the productivity of researchers and institutions, and to map how knowledge develops within a discipline. Scientometrics, defined originally as the quantitative study of science as a communication process, gave rise to familiar tools such as the impact factor for journals and the h-index for individual authors.

In the Indian context, this quantitative tradition has deep roots. The statistician P. C. Mahalanobis, founder of the Indian Statistical Institute, argued in the early 1950s that statistics is a key technology essential for development and forecasting. That statistical mindset carried into library and information science, where quantitative methods are now used to assess scientific productivity and scholarly communication.

The problem of defining an “information unit” 🔗

Here is where the difficulty of defining information comes back to bite. To measure anything, you need a unit. To measure distance, we use the metre. To measure mass, we use the kilogram. But what is the unit of information?

This question has no easy answer. In the mathematical theory of communication, information is measured in bits, but this unit is formal and dimensionless. It tells us how much uncertainty a message removes, but it deliberately ignores meaning. A bit of information about a coin toss and a bit of information about a life-changing medical result count the same mathematically, even though they matter very differently to a person. This is the gap between Shannon-type information, which equals uncertainty, and meaningful information, which actually informs a human being.

Informetrics largely sidesteps this philosophical problem by measuring proxies rather than information itself. Instead of trying to count abstract “information,” it counts things we can count: numbers of publications, citations received, downloads, links, and co-authorships. These countable items stand in for information and let researchers draw useful conclusions. But it is important to remember that counting publications is not the same as measuring information or knowledge. A highly cited paper is not automatically more “informative” in any deep sense. This is the ongoing tension at the core of the field, and it is a direct consequence of the fact that information itself resists a clean, universal definition.

Why this matters 🔗

Revisiting the concept of information shows us that something we use casually every day rests on genuinely difficult foundations. The three attitudes, the anthropocentric, organization-based, and action-oriented views, each capture part of the truth without being complete on their own. Scholars like Belkin and MacKay gave us definitions centred on knowledge gaps and meaningful distinctions, and the idea of uncertainty reduction ties many of these threads together. Informetrics then takes on the brave task of measuring this elusive thing, succeeding largely by measuring what can be counted while acknowledging that the deeper question of what information truly is remains open.

Understanding these foundations is not just theoretical. It shapes how libraries are organized, how research is evaluated, and how we judge the flood of data in our digital lives. The next time you search for something, retrieve a document, or read a citation count, you are standing on top of decades of debate about a deceptively simple word.

What do you think? If information cannot be cleanly defined, can it ever be truly measured, or are we only ever measuring its shadows like citation counts and downloads? And among the three attitudes toward information, which one best matches the way you personally experience being informed in your daily life?

References

1. https://en.wikipedia.org/wiki/Uncertainty
2. https://en.wikipedia.org/wiki/Michael_Buckland
3. https://courses.washington.edu/info300/buckland_ho.pdf
4. https://asistdl.onlinelibrary.wiley.com/doi/10.1002/(SICI)1097-4571(199106)42:5%3C351::AID-ASI5%3E3.0.CO;2-3
5. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11588055/
6. https://liswiki.org/wiki/Information_behavior_theories
7. https://arxiv.org/pdf/1406.5688
8. https://www.numberanalytics.com/blog/uncertainty-information-theory-fundamentals
9. https://www.sciencedirect.com/science/article/abs/pii/S0306457310000798
10. https://en.wikipedia.org/wiki/Informetrics
11. https://arxiv.org/pdf/1501.05462
12. https://arxiv.org/pdf/0912.1357