This study, through bibliometric data, adopts the concept of Revealed Comparative Advantage (RCA) to apply in publication of scientific fields for observing the feature of scientific capacity of a nation. Following that, we utilize a hierarchical clustering method to conduct national clustering according to the RCA scores of publication in 24 scientific fields of each nation. A nation can thoroughly review its relative position, identify other nations with the same characteristics, and understand different clusters' features of scientific capacity. The result also provides a reference that enables researchers to review, explore, and learn about the factors behind the innovation system in the representative clusters or nations. By leveraging the clustering result, this knowledge also allows the nation to grasp opportunities to create innovation by cooperating with other nations that have different comparative advantages. As a result, by accumulating innovation capacity in these research network collaborations, the nation can maintain its innovation momentum and then forge into the innovation lead constantly.

This paper presents a methodology for measuring the improvements in efficiency and adjustments in the scale of R&D (Research & Development) activities. For this purpose, this study decomposes academic productivity growth into components attributable to (1) world academic frontier change, (2) R&D efficiency change, (3) human capital accumulation, and (4) capital accumulation. The world academic frontier at each point in time is constructed using data envelopment analysis (DEA). This study calculates each of the above four components of academic productivity for 27 countries over 1990-2003, and finds that the components which contribute to academic productivity growth vary with the different countries' characteristics and development stages. Human capital has more weight in terms of the quantity of academic research, and capital accumulation plays a more important role in the citation impact of academic research.

No abstract

No abstract