Science Productivity, Higher Education Development and the Knowledge Society (SPHERE Project) Final Report

This project created and analyzed a new, large global dataset on scientific journal articles, published between 1900 and 2011, and a series of case studies to examine how systems of higher education developed and grew nations’ capacity for scientific research. The analysis resulted in a series of new insights about global scientific production that were only possible with a consideration of long-term trends. First, despite predictions as early as the 1960s that the growth rate of “big science” would slow, the dataset shows in fact that “big science” started a phase of exponential growth in the early 1960s that has continued unabated for decades. “Big science” has transformed into “mega-global science” and the trends of global diffusion and regional differentiation began much earlier in the 20th century than is commonly understood. Second, the analysis of rates of regional journal article production also depicts clear shifts in the competition for ascendancy in scientific production. For the first half of the 20th century, global competition for scientific impact was primarily an Atlantic battle between the top producers of Europe (Germany, France, and the U.K.) and the United States. The locus of competition shifted by the, end of the 20th century to a contest between the current research “superpower, ” the United States, and the fast-growing producer, China, along with the many less populous countries of Western Europe with their highly productive science systems. With the contributions of other East Asian, high volume producers such as Japan and South Korea in the later decades of the 20th century, and simultaneous slowing of research production in U.S. science, the center of gravity for research production has been pulled eastward for the past two decades. Third, while science may indeed be an inherently global and collaborative enterprise, the trend toward global collaboration of authors is a relatively recent one. Historically, one-third of all research articles worldwide result from international collaboration, and less than 26 percent are the product of one researcher alone. In 1980 however only about 2 percent of all SCIE publications involved a collaboration across international lines. Three decades later this proportion is eleven times what it was in 1980. Finally, the study also concluded that overall volume of production is not a sufficient measure of scientific capacity by itself. When adjusting for the size of population and the economy the proportion of GDP spent on R&D or the number of researchers some smaller countries (especially in Europe) are more productive on a per capita basis than mid-sized or even larger ones. Similarly the ratio of investment in science to scientific production is much higher in the high volume producers than it is in some small states. While output is smaller in these states, they have maximized R&D investments more efficiently than their larger competitors.