What Counts as Scientific Research and Development in ISIC Rev. 5?

Scientific research and development in ISIC Rev. 5, UN Statistics Division, covers systematic creative work undertaken to increase the stock of knowledge and to devise new applications of available knowledge. It includes basic research, applied research and experimental development across natural sciences, engineering, medicine, agriculture, social sciences and humanities. The work must be systematic and creative, not routine testing or routine data collection, and it is described by field of research rather than by the product that may eventually result.

New knowledge, or routine checking?

That contrast is the whole test for this scope. Scientific research and development in ISIC Rev. 5, UN Statistics Division, is systematic creative work undertaken to increase the stock of knowledge and to devise new applications of available knowledge. It includes basic research, applied research, and experimental development across natural sciences, engineering, medicine, agriculture, social sciences, and humanities. The work must be systematic and creative. Routine testing and routine data collection do not qualify, however skilled the staff or sophisticated the equipment. A laboratory can therefore sit on either side of the line, depending on its purpose. Purpose, not premises, decides.

The scope includes:

• Basic research to gain new knowledge without a specific application in view
• Applied research directed at a specific practical aim
• Experimental development to create new or improved products, processes, or services
• Research in natural, engineering, medical, agricultural, social, and human sciences

Basic research is the easiest to misunderstand, because it has no product in view. Its product is knowledge itself: a finding, a mechanism, a principle. Applied research narrows that curiosity toward a practical aim. Experimental development turns knowledge into a prototype, a process, or a service that did not exist before. All three are creative in the sense that the outcome is not known in advance. Routine checking, by contrast, starts with a fixed standard and asks only whether this batch meets it. The answer is yes or no. Nothing new is learned about the world.

Two laboratories, opposite sides of the line

A university laboratory that investigates a new material, publishes findings, and then develops a prototype moves through the research stages in order: new knowledge, then a practical direction, then a thing that embodies it. A factory laboratory that routinely tests each batch against a fixed standard does none of that, even though both laboratories wear coats, run instruments, and record results. ISIC Rev. 5, UN Statistics Division, separates the two by purpose: new knowledge versus routine checking. The same distinction separates research from teaching. Teaching in a university is education, except for research done as part of that role, where the research itself is the activity being described. And manufacturing a product that resulted from earlier research is manufacturing, grouped by the product, not research, because the research phase has ended and production has begun.

What falls outside, in plain terms

Routine testing of products against a standard sits outside. Routine data collection or market surveys sit outside, even when the data later feeds research done by someone else: collecting is not discovering. Teaching sits outside, with the research exception noted above. Manufacturing the product that resulted from earlier research sits outside. Each exclusion protects the same idea: research totals should measure the creation of new knowledge and new applications, not the repetition of known procedures. Without those exclusions, any firm with a laboratory or a survey team could claim research status, and research totals would swell with work that adds no new knowledge at all.

Why inflated research totals would mislead policy

Research statistics drive innovation policy, funding decisions, and comparisons of national scientific capacity. Governments, funders, and firms all read them as a measure of how much new knowledge an economy is creating. If routine testing or teaching were counted as research, those totals would be inflated and real research investment would be hidden inside a larger, vaguer number. Funding might flow to routine checking dressed as discovery, while genuine discovery looked smaller than it is. The systematic and creative test is therefore not pedantry. It is what makes the figures usable for the decisions they exist to inform. ISIC Rev. 5, UN Statistics Division, holds 22 sections from A to V, 87 divisions, 258 groups, and 463 classes in total, and the research entries sit inside that structure precisely so that research can be compared across countries on the same basis.

Neighbours worth reading before deciding

Technical testing activities are described separately, and they are the nearest neighbour for any laboratory case. Market research activities are separate too, and they cover studying markets rather than creating scientific knowledge. Education sits in Section Q. Manufacturing is grouped by product. Read each neighbour’s UNSD wording in full, including exclusions, and ask which entry describes the unit’s purpose most exactly. A short note of the entry consulted will make any later review much easier, because research cases are often revisited when funding or reporting depends on them.

This guide explains how the classification treats research; it does not award research status to any reader. It cannot determine a code, a principal activity, or a registration outcome. Those questions should go to the national statistical institute or business register that holds the record, with the full description of the work in hand.