What Is Information Science? Fields, Methods and Funding

Information science is the academic field that studies information itself: how it is created, represented, organized, found, shared, evaluated and used, by people, organizations and increasingly by software. Where a discipline such as chemistry studies a class of physical phenomena, information science studies a class of problems that cut across every domain: how do you describe a body of knowledge so that someone can find it, how do you know whether a retrieval system is working, how do people actually look for and make sense of what they find, and how do you measure the structure of the scholarly record that results? This guide explains what the field covers, its main subfields and methods, how it relates to neighboring disciplines, where training and funding come from, and why research administrators encounter its work constantly even if they never use the name.

A Working Definition

A practical definition: information science is the study of the properties, representation, organization, retrieval and use of recorded information, together with the systems and human behavior that surround it. Three features of that definition do most of the work.

• It is about recorded information, not just data or just documents. The field treats journal articles, datasets, web pages, images, clinical records and database entries as objects that must be described and found.
• It is about systems and people together. A search engine is a technical artifact, but whether it succeeds depends on how a person formulates a question and judges an answer. Information science insists on studying both halves.
• It is domain-independent by design. The same theory of indexing, relevance or citation applies to a medical literature database, a legal archive and a research data repository.

The iSchools organization, the international body of university units teaching and researching in this area, describes the scope of the field as deliberately broad and methodologically agnostic, relying heavily on the social and behavioral sciences as well as computing and linguistics. That description is a good guide to how the field behaves in practice: it borrows methods from wherever they fit.

Information Science vs. Library Science

The two are often taught together and the labels are frequently used interchangeably, but they are not the same thing. Library science is the applied, institution-centered side: running and preserving collections, cataloging, reference and instruction, archives, and the professional practice of librarianship. Information science is the more general, theory-and-systems-centered side: it asks questions about information regardless of whether a library is involved, including how a commercial search engine ranks results, how a hospital organizes clinical vocabularies, or how a research field’s citation network is structured. Many graduate programs are labelled Library and Information Science (LIS) and many sit inside iSchools, which is why the boundary looks blurry from the outside. A useful rule of thumb is that library science is defined by a type of institution and information science by a type of problem. This page concentrates on the problem side; the library-science guide covers the institutional and professional side in more depth.

The Main Subfields

Information science is not a single research program but a cluster of related ones. The following are the areas most commonly treated as its core.

Information Retrieval

Information retrieval (IR) is the study of how to find relevant material in a large collection in response to an information need. It covers indexing, query formulation, ranking models, relevance feedback, evaluation and, in recent years, the use of machine learning and language models in search. A defining habit of IR research is evaluation: a retrieval method is judged by how well it performs against a test collection and a set of relevance judgments, not by how elegant it is. IR overlaps heavily with computer science, and many IR researchers are based in computing departments, but the field’s focus on users and relevance is characteristically information-scientific.

Knowledge Organization

Knowledge organization is the study and practice of describing and arranging knowledge so that it can be found and understood. Its working objects are classification schemes, thesauri, controlled vocabularies, taxonomies, ontologies and metadata schemas such as Dublin Core. The central questions are conceptual as much as technical: what categories does a domain actually use, how do you handle terms with several meanings, how do you reconcile vocabularies built by different communities, and how do you keep a scheme usable as knowledge changes. Knowledge organization is the intellectual backbone behind library catalogs, repository metadata, biomedical vocabularies and the semantic web.

Human Information Behavior and Interaction

This area studies how people recognize that they need information, how they seek it, what sources they trust, how they judge quality and how they use what they find. Methods are largely drawn from the social sciences: interviews, surveys, observation, diary studies, think-aloud protocols and usability testing. The findings routinely contradict the assumptions built into systems, for example that users know the right search terms in advance, or that they will read past the first screen of results. Human information behavior connects directly to human-computer interaction, which supplies many of the interface-evaluation methods used in the field.

Bibliometrics and Scientometrics

Bibliometrics applies quantitative methods to publications and their citations; scientometrics extends the same approach to the study of science as a system, including productivity, collaboration, funding and impact. Typical outputs are citation counts, co-authorship and co-citation networks, topic maps and indicators such as the h-index. Related work on newer indicators is often grouped under altmetrics. This is the subfield most visible to research administrators, because its outputs feed research evaluation, rankings and funding decisions. Because this site covers it in depth, the better starting points are the guide to scientometrics, the walkthrough of bibliometric analysis methodology, and the comparison of VOSviewer, CiteSpace and Bibliometrix. Those pages are narrower and more technical than this overview.

Research Data Management and Data Curation

As research output has shifted from papers alone to papers plus data, software and protocols, information scientists have taken on the problem of managing those objects across their lifecycle: description, storage, access, sharing, citation and preservation. Data curation and research data management draw on knowledge organization (for metadata and vocabularies), on repository and systems design, and on policy (for sharing and retention requirements). In practice this is where the field meets research administration most directly: a data management plan required by a funder is an information-science artifact, and a persistent identifier such as a DOI is an information-science solution to the problem of stable reference.

Digital Libraries, Archives and Preservation

The study of digital collections, repositories and long-term preservation overlaps with library and archival science. Information scientists contribute work on repository architecture, interoperability between systems, format obsolescence and the authenticity of digital records. Open repositories and the open archive model are products of this line of work.

Health and Domain Informatics

Informatics fields apply information-science ideas to a specific domain, and the labels vary. Health informatics deals with clinical and health information, bioinformatics with biological data, and neighboring areas such as cheminformatics and social informatics do the same for their domains. The boundary with information science is a matter of emphasis: informatics tends to be defined by its domain and its systems, information science by its general theory of information. The two share vocabulary, methods and many of the same departments and journals.

How Information Science Differs From Computer Science and Data Science

Information science is frequently confused with its neighbors, so it helps to separate them by their central question.

• Computer science asks what can be computed and how to build systems that compute efficiently. Information science takes such systems as given and asks whether they organize and deliver information in ways that people can use.
• Data science concentrates on extracting patterns and predictions from data using statistics and machine learning. Information science concentrates on how data and documents are described, found and interpreted, including the metadata and curation that make analysis possible in the first place.
• Information science also overlaps with communication studies where it deals with how information moves between people, and with cognitive and social psychology where it studies information seeking.

In practice the lines are porous, and many researchers work across them. The distinction is useful mainly for knowing which literature to read and which department or program is the natural home for a given project.

Methods Used in the Field

Because the subject spans systems and people, information science uses both computational and social-science methods, and individual studies often combine them.

• Test-collection and benchmark evaluation for retrieval and classification systems, using fixed document sets, queries and relevance judgments.
• User studies: interviews, surveys, observation, think-aloud sessions, controlled experiments and usability testing.
• Bibliometric and network analysis: counting and mapping publications, citations, authors, institutions and topics.
• Content, discourse and metadata analysis: systematic examination of how documents, collections or records are described and categorized.
• Conceptual and terminological analysis: building and comparing classification schemes, ontologies and vocabularies.
• Design-science and systems building: constructing a prototype, repository or tool and evaluating it against a stated need.
• Computational text and data methods: natural-language processing, topic modelling and machine learning applied to collections.

Studies in the field are therefore judged by the standards of whichever method they use. A user study is assessed on sampling and design validity, a retrieval experiment on its test collection and metrics, a bibliometric study on its data source and its handling of known biases such as incomplete database coverage and field differences in citation habits. The guide to reading and reporting FWCI is a good example of the care required with field normalization.

A Short Note on History

Information science grew out of two older traditions: the practice of libraries and documentation, which was concerned with organizing recorded knowledge, and the mid-twentieth-century response to the rapid growth of the scientific literature and the arrival of computers, which made automated indexing and search conceivable. Those two strands, an institutional practice of organizing knowledge and a technical effort to retrieve it by machine, still define the field today. This page deliberately gives no precise dates or names for individual milestones because the field’s own accounts differ on where to draw its starting line; readers who need a dated history should consult a primary historical source rather than a summary.

Training and Career Paths

The field is taught in university schools and departments of information, many of which belong to the iSchools organization. By its own description the iSchools network includes around 130 member universities across several continents. Programs range from undergraduate degrees in information science or information studies, through professional master’s degrees (including the Master of Library and Information Science, which is the standard credential for professional librarianship in the United States), to research doctorates.

Graduates work in a wide range of roles, including:

• librarians, archivists and records managers;
• metadata, taxonomy and ontology specialists;
• research data librarians and data curators;
• user experience and information architecture designers;
• search and discovery engineers and analysts;
• bibliometric and research-intelligence analysts in universities, funders and publishers;
• health information and clinical informatics professionals;
• competitive-intelligence and knowledge-management staff in industry and government.

Doctoral graduates typically become faculty in iSchools and related departments or research staff in industry laboratories and public-sector research organizations.

Societies, Journals and Conferences

The Association for Information Science and Technology (ASIS&T) is a principal international society for the research side of the field. It describes its mission as fostering excellence and innovation in the information sciences, and it states that its members include researchers, developers, practitioners, students and professors from around 50 countries. It publishes the Journal of the Association for Information Science and Technology (JASIST), holds an annual meeting and organizes special interest groups on topics within the field. The iSchools organization, described above, is the main network of academic units, and the American Library Association is the largest professional body on the library-practice side. Beyond these, the field has specialist journals and conferences for information retrieval, knowledge organization, digital libraries, bibliometrics and scientometrics, and information behavior. Authors choosing where to publish should look at the venue’s scope for the specific subfield rather than relying on the field label alone.

Who Funds Information Science Research

Funding is spread across several kinds of sponsor because the field touches libraries, computing, health and the social sciences. Two US federal sources are especially relevant:

• Institute of Museum and Library Services (IMLS). IMLS is the federal agency that supports libraries and museums, and its grant programs include support for library and archives research, professional education and community-facing projects. Because program names, priorities and availability change, applicants should confirm the current list of programs on the agency’s own funding pages before planning a proposal.
• National Science Foundation (NSF). Information-science research that is computational or human-centered, such as retrieval, interaction, data infrastructure and cyberinfrastructure, is typically funded through NSF’s computing directorate (see NSF CISE) and related programs, and through social-science programs for work on information behavior and the science of science.

Other sponsors include health-related agencies for biomedical and health information work, humanities funders for digital collections, private foundations that support libraries and scholarly infrastructure, and national research councils outside the United States. For specific program names, deadlines and eligibility, rely on the sponsor’s current solicitation. This guide does not attempt to list award amounts or dates.

Why Information Science Matters for Research Administration

Research administrators do not usually think of themselves as users of information science, but much of their daily environment is built from it.

• Research evaluation. Citation indicators, field normalization and rankings used in promotion, assessment and funding decisions come from bibliometrics. Understanding their limits is part of using them responsibly; see the discussion of how many citations is good and the CoARA agreement on research assessment.
• Data management mandates. Funder data-sharing and data management plan requirements are applications of research data management and metadata practice.
• Identifiers and metadata. DOIs and other persistent identifiers, and the metadata attached to them, are what allow outputs to be linked to people, projects and funders in research information systems.
• Systems. Institutional repositories, current research information systems and discovery tools all rest on knowledge-organization and retrieval design choices.
• Staffing and collaboration. Research data librarians, metadata specialists and bibliometric analysts are information-science professionals, and they are often the colleagues a grants or research office most needs to work with.

Frequently Asked Questions

What does information science study?

It studies how recorded information is created, represented, organized, retrieved, shared and used, including the systems that handle it and the behavior of the people who rely on them.

Is information science the same as library science?

No. Library science is the institution-centered, professional side concerned with libraries and archives; information science is the broader study of information problems in any setting. They are usually taught together under the Library and Information Science label. See the library science guide.

Is information science the same as computer science or data science?

No. Computer science focuses on computation and system construction, data science on extracting insight from data, and information science on organizing, retrieving and using information. They overlap and share methods. See the guides to computer science and data science.

What is an iSchool?

An iSchool is a university school, college or department that teaches and researches information in its broad sense and belongs to the iSchools organization, an international network of around 130 universities.

Is health informatics part of information science?

It is closely related. Health informatics applies information-science ideas and computing to health and clinical information and has its own journals, societies and programs. See the guide to health informatics.

What degree do you need to work in information science?

It depends on the role. Professional librarian positions in the United States generally require a master’s degree in library and information science, research positions typically require a doctorate, and many metadata, UX and analyst roles accept a bachelor’s or master’s degree in information science or a related field.

Where does bibliometrics fit?

Bibliometrics and scientometrics are subfields of information science that apply quantitative methods to publications and citations. They are covered in more detail in the scientometrics guide.

Where This Fits on CASRAI

Information science is one of the disciplines profiled in the Branches of Science series. It sits next to library science, computer science and data science, and it supplies much of the conceptual machinery, metadata, identifiers, retrieval and indicators, behind the research-administration topics covered elsewhere on this site.