Concentration and evenness measures as macro - level scientometric indicators - - - -

There exist different types of concentration and evenness measures. What they have in common is the fact that they all try to measure inequality. We will discuss only concentration measures that describe the scatter of items over sources, and this irrespective of the number of sources considered. Concentration measures can, e.g., be used as indicators for the convergence, or divergence of regions. Use of different types of concentration measures as scientometric indicators is illustrated for the publication output of China’s different administrative units (provinces, autonomous regions and municipalities). Three aspects of concentration, each leading to a different Lorenz curve, and resulting in a different partial order, are studied. In each case, we suggest measures respecting the corresponding partial order. Concretely, we study: • classical Lorenz curves and concentration measures • comparison with an internal or external standard • source per source comparison of items Different forms of concentration and corresponding measures are important indicators for the dynamics of an R&D system. We advance the proposition that this aspect has been neglected too much in the past.