NSF examines plateau in US publications

The Science Resources Statistics Division of the National Science Foundation held a workshop last week to explore why the number of US science publications in high-impact journals has plateaued. The meeting responded to the findings of a January report from the NSF's National Science Board that showed the number of US publications remained essentially flat from 1992 to 2002, leading to a drop in the US share of publications from 38% to 30%.About 35 bibliometrics researchers, economists, sociologists, information scientists, and government and university administrators met on Oct. 7 in Washington, D.C. to explore the issue. "As far as we could tell, inputs were going up, but the number of science publications was not going up correspondingly," Robert Bell, a senior analyst in the Science Resources and Statistics Division, told The Scientist. The NSF shut the meeting's doors to the media because, according to Bell, data given in several of the workshop's presentations have not completed internal review by the NSF. Presentations at the meeting included a report describing trends in the numbers of US publications broken down by elements like subject area and journal impact, as well as a working paper analyzing research investments and publication outcomes at 200 leading US universities, and a working paper describing interviews with senior researchers at nine US universities who were asked how the research environment has changed during their tenure.Bell said he expects the report and papers to be published in the next few months.The January report, "Science and Engineering Indicators 2006," found that American scientists still lead the pack in terms of output, with over 200,000 publications in 2003, according to output captured in the Science Citation Index and the Social Sciences Citation Index. But two issues are causing concern: First, the percentage of US publications is declining as other countries increase their output, and second, the US figure, while large, is stagnating, with only one percent growth between 1993 and 2003, according to Bell. An increase in global collaborations could be part of what is affecting the US' numbers, noted Kevin Boyack, a principal member of the technical staff in the Computation, Computers, Information and Mathematics Center at Sandia National Laboratories, who did not attend the workshop. According to the report, the share of internationally coauthored articles at least doubled in the US, the EU-15, and Japan from 1988 to 2003. "Is our share being diluted because we are doing more international collaboration than we used to?" Boyack wondered. Ben Martin, a professor of science and technology policy studies at the University of Sussex who attended the meeting, said that international collaborations were a major focus of the discussion. Articles that include authors from more than one country are tallied using fractions -- so, for example, if one out of three authors is from the US, the US receives 1/3 of a point. But presenters at the meeting showed a decline in US output, though to a lesser degree, even when giving each country a whole point for participating in a paper, Martin told The Scientist. Martin said he hypothesized at the meeting that the declining US share results from a growing appreciation among non-US researchers for the value of publishing in English-language journals, making it more competitive for American scientists to get their work accepted. "It doesn't mean US science is weakening," he told The Scientist, "but that the bar is being raised around the world." He speculated that, as competition has intensified, US scientists might be publishing in lower-impact journals not among the 7,000 captured by the Science Citation Index. Much of the new competition appears to be coming from China."It's unbelievable," Loet Leydesdorff, a professor at the University of Amsterdam who did not attend the workshop, told The Scientist. "We see exponential growth in all parameters: citations, publications, funding. It's been sustained for 10 years." Leydesdorff added that other Asian countries -- namely Singapore, South Korea and Taiwan -- are also markedly increasing their output. The Science and Engineering Indicators report found that the combined publication share of these four countries surged from less than 2% in 1988 to 8% in 2003.Kerry Grens kgrens@the-scientist.com Links within this article:Science Resources Statistics Division http://www.nsf.gov/statisticsScience and Engineering Indicators 2006 http://www.nsf.gov/statistics/seind06/National Science Board http://www.nsf.gov/nsb/start.htmS. Pincock, "US Leading World Science," The Scientist, July 15, 2004 http://www.the-scientist.com/article/display/22291/Science Citation Index http://scientific.thomson.com/products/sci/Computation, Computers, Information and Mathematics Center at Sandia National Laboratories http://www.cs.sandia.gov/index.htmlBen Martin http://www.sussex.ac.uk/spru/profile1716.htmlK. Hopkin, "Most Highly Cited," The Scientist, Oct. 24, 2005 http://www.the-scientist.com/article/display/15796P. Zhou and L. Leyedesdorff, "The emergence of China as a leading nation in science," Research Policy, 35(1), 2006, 83-104 http://www.sciencedirect.comLoet Leydesdorff http://users.fmg.uva.nl/lleydesdorff/J. Chan, "Stronger Support for US Research Is Needed to Meet Asian Science's Growing Challenge," The Scientist, March 2, 1992 http://www.the-scientist.com/article/display/11217

NSF examines plateau in US publications

US scientists' share of publications is declining in the face of competition from countries like China

Interested in reading more?

Become a Member of