Scientomettic& Vol. 21. No. 2 (1991) 223-244 BIBLIOMETRIC STUDIES FOR THE EVALUATION OF TRANS-NATIONAL RESEARCH G. LEWISON,* P. CUNNINGHAM** * Gesamthochschule-Universitii~ Henschelstrasse 4, D-3500 Kassei (Germany) ** Programme of Policy Research in Engineering Science and Technology (PREST), University of Manchester, UK - M 13 9PL (UK) (Received September 19, 1990) Results are given of an analysis of the scientific papers describing work carried out under two European Community research programmes, in biotechnology and environmental chemicals. They were shown to be more multinational in their authorship than other papers in the same journals, and as a consequence, more frequently cited in the 5 years following publication. The citation rates peak early for the biotechnology papers suggesting that effective measures have been taken to disseminate the results of the work to other scientists so that they have become aware of them earlier than usual. Introduction The European Community (EC) supports research in many fields through multi- annual programmes. Most of them are carried out through cost-shared research contracts (normally with 50% support) with external contractors, who are selected following a call for proposals. Each research programme is subject once during its currency to an external independent evaluation. Such an evaluation is required inter alia to determine the impact of the work, the benefits through operation at the European level and whether its objectives have been met. It is normally carried out by a panel of people with a wide range of backgrounds and experience. They are permitted to call upon the services of specialist groups in order to answer particular questions in detail. Several recent evaluation panels have used bibliometric methods in order to answer specific questions about the programmes. These studies are expected to reinforce and illustrate the panels' judgements but not to replace them. Trans-national scientific co-operation is deliberately fostered by EC research programmes. Most of these now require all project proposals to be submitted by partners in at least two Member States. A natural consequence would be to expect trans-nationally authored papers, but other forms of co-operation could occur, such as the division of a research task into complementary parts, or the provision of Scientometrics 21 (1991) Elsevier, Amsterdam-Oxford-New York- Tokyo Akad~miai Kiadd