Research Classifications: Three major principles

Abstract

In the discussion about research classifications, there is often the argument that these should be as universally applicable as possible - even in detail. In my presentation, however, I would like to highlight other characteristics. In short, a good subject classification satisfies the following requirements: - It is structured hierarchically whereas (1) at its finest level, it meets the needs of the respective institution (which can be very specific) and (2) on the highest levels it is compatible to international standards (esp. Frascati). - There are clear rules as to which objects are to be classified (e.g. Projects, People, Institutes) - There are defined processes that make a regular revision of the classification possible The presentation illustrates these three principles using the example of the two main classification systems used in Germany by the Federal Statistical Office (DESTATIS) and the German Research Foundation (DFG).