Jointly published by Kluwer Academic Publishers, Dordrecht Scientometrics, and Akadémiai Kiadó, Budapest Vol. 51, No. 1 (2001) 335–346 Language biases in the coverage of the Science Citation Index and its consequences for international comparisons of national research performance THED N. VAN LEEUWEN, HENK F. MOED, ROBERT J. W. TIJSSEN, MARTIJN S. VISSER, ANTHONY F. J. VAN RAAN Centre for Science & Technology Studies (CWTS), Leiden University, Leiden (The Netherlands) Empirical evidence presented in this paper shows that the utmost care must be taken in interpreting bibliometric data in a comparative evaluation of national research systems. From the results of recent studies, the authors conclude that the value of impact indicators of research activities at the level of an institution or a country strongly depend upon whether one includes or excludes research publications in SCI covered journals written in other languages than in English. Additional material was gathered to show the distribution of SCI papers among publication languages. Finally, the authors make suggestions for further research on how to deal with this type of problems in future national research performance studies. Introduction Due to decreasing budgets, public accountability and the drive for efficiency in the research system, the interest in bibliometric evaluation of scientific research has strongly increased in recent years. Evaluation focusing on the comparison of national R&D systems has led to influential publications in journals like Nature and Science. In his paper of February 1997 in Science, Sir Robert May compares those national systems, thereby using various types of quantitative data. 1 Scientific publication output and impact per capita for a large number of countries were presented. Important policy relevant conclusions were drawn from his tables, which caused commotion, in circles of science policy makers as well as amongst the scientific community. The cause for concern was particularly related to pitfalls/limitations in the underlying bibliometric data. Especially for Germany and also France, the findings and conclusions of the paper were rather dramatic, both countries occupying only modest positions in the rankings. 0138–9130/2001/US $ 15.00 Copyright © 2001 Akadémiai Kiadó, Budapest All rights reserved