A comparison of the scientific performance of the U. S. and the European Union at the turn of the XXI century

In this paper, scientific performance is identified with the impact journal articles achieve through the citations they receive. The empirical exercise refers to 3.6 million articles published in 1998-2002 in 22 scientific fields, and the more than 47 million citations they receive in 1998-2007. The first finding is that a failure to exclude co-authorship among member countries within the EU (European Union) may lead to a serious upward bias in the assignment of articles to this geographical area. In the second place, standard indicators, such as normalized mean citation ratios, are silent about what takes place in different parts of the citation distribution. Consequently, this paper compares the publication shares of the U.S. and the EU at every percentile of the world citation distribution in each field. In 15 disciplines, as well as in all sciences as a whole, the EU share of total publications is greater than that of the U.S. one. But as soon as the citations received by these publications are taken into account the picture is completely reversed. The mean citation rate in the U.S. is greater than in the EU in every one of the 22 fields. In seven fields, the initial gap between the U.S. and the EU widens up as we advance towards the more cited articles, while in the remaining 15 fields –except for Agricultural Sciences– the U.S. always surpasses the EU when it counts, namely, at the upper tail of citation distributions. For all sciences as a whole, the U.S publication share becomes greater than that of the EU one for the top 50% of the most highly cited articles.