Sense and nonsense of S&T productivity indicators

This article presents a process for benchmarking scientific productivity fit for public policy purposes, in which quantitative indicators have the major role of initiating and focusing the discussions. It shows that the calculation of a research productivity indicator defined as the ratio of the output (publications) of basic research to its spending does not provide in itself any sensible comparison among countries. It suggests that quantitative indicators are useful as entry points into discussions, considering that their raison d'etre is to be criticised in terms of their (limited) relevance and (limited) comparability. In this interpretation, benchmarking necessarily involves direct interaction among people in an organised process. Copyright , Beech Tree Publishing.