Category Archives: Scientometrics

WTEC output on 2/15/11

Yesterday I picked up from the Baltimore office the final report for the scalable software workshop.  This is an important contribution to an important problem.  Dave Nelson, Ben Benokraitis, and Patricia Foland did a very good job in making this a big success.

Last night I heard that the paper by Shelton and Leydesdorff was accepted for presentation at the ISSI conference in Durban in July; I passed out hardcopies of this at our last staff meeting.  Yesterday I also submitted a longer version to the Journal of the American Society for Information Science and Technology.  Loet posted this one at a preprint service hosted at Cornell. 

Yesterday we also finished a draft of the brochure summary of the nanoEHS stategy, which I will bring to NNCO today.  Patricia, Matt, David, and I have been crashing on this for the last few days.  I think this looks very good, but we’ll see what the client thinks.  In any event, we will now turn our attention back to the similar brochure on Nano2.  OmniStudio is going to prettify both.

I suppose that Geoff and company are also finishing their supplement to the President’s FY2012 budget, a document with considerably higher stakes.

R. D. Shelton

Model Accurately Predicts Growth in Chinese Publication at US Expense

Our Rio paper contained quantitative predictions of how much Chinese scientific publication would continue to grow at US expense.  Data in the new Science and Engineering Indicators 2010 allow some checking of these forecasts.  The model calculates publication share from forecasts of R&D investment share.  The data used was for 2005 then available from NSF (in S&EI2008) on the number of scientific papers and from OECD on national investments in R&D (GERD).

In 2005 the publication shares based on the 35 country OECDgroup (with more than 90% of world publication) were:
US  (30.9%)
PRC (6.3%)

Our forecast for 2007, made based on 2005 data, was:
US (29.6%)
PRC (7.9%)

The actuals for 2007 in the new SEI2010 report are:
US (29.9%)
PRC (8.1%)

Not bad for government work, particularly since some don’t even know what direction these time series are heading or why. I suspect that the small difference between forecast and actuals is because China increased their GERD share a little faster than we forecast.  I can check that from the latest OECD data and put it in a comment.

So what? Well, if my model continues to forecast correctly the rapid rise of China, it predicts that China will actually pass the US in well less than ten years to lead the world.  This is despite China’s current share being far below that of the US in this key indicator of scientific supremacy.

As shown in the Rio paper, China has already passed the US in some databases of physical science publications.  The paper also has forecasts of about a dozen other indicators of national S&T performance.  In two cases, high tech manufacturing market share, and the number of science and engineering PhDs produced, the paper’s forecasts that China would soon pass the US have already occured in the data newly available.  Also, China will likely pass the US in the number of researchers when the 2008 data becomes available.  Have a nice day, particularly if you are in the PRC. (:))

The Rio paper is here.

R. D. Shelton

Call for Papers: STI 2010 in Leiden

The next Science and Technology Indicators (STI) conference will be held in Leiden, September 8-11.  Papers or posters are due April 16. 

Conference Link

There are other related conferences, but the ones that most people in the bibliometrics field try to go to are the STI and the international conference on scientometrics and informetrics (ICSI).  The pattern is that STI is held in even years, and alternates between Leiden and some other city.  I think it is to be held in Vienna in 2012.  ICSI is held in odd numbered years; it will be in Durban, S. Africa in 2011.

The STI final paper is presented in .ppt form; no text version is required unless it is later published in the proceedings or a journal.  In the past they have required an extended abstract for selection; I don’t see these specs on the current website.

R. D. Shelton 

HR is Not the Bottleneck Resource in American Science Leadership

I really admire Norm Augustine’s success in getting the attention of the White House and Congress with his Academy report, Rising Above the Gathering Storm.  It relates the now well-known problem about America losing its lead in S&T and then focuses on increasing human resources (HR) here as the remedy.  The report directly led to the America COMPETES Act and to broad, bipartisan support for sharply increasing the research budgets of NSF, NIST, and the DOE Office of Science.

However, I venture to disagree with the distinguished Dr. Augustine that increasing the supply of American research personnel is critical to making America more competitive.  The weak job market for the scientists and engineers we already have to serves as a counterexample.  This morning, I was looking at lots of applications from such scientists for a slot at our company, and I was struck by how many are now underemployed as postdocs, visiting assistant professors, and the like.

But it’s not necessary to rely on such anecdotes.  This morning I also did a multiple linear regression using recent UN worldwide data (N = 65 countries).  The two inputs to the national scientific production systems were national R&D investment (GERD) and labor (number of researchers).  The dependent variable was the number of scientific publications in the world’s leading journals, in the Science Citation Index.  The regression shows that the HR variable is not only insignificant compared to the money variable GERD, but has a negative coefficient–more researchers produce fewer papers (sort of).  That’s pretty convincing to me that money is the bottleneck resource, not HR.  This is not really surprising, since with money you can buy almost everything else, including hiring more researchers from all over the world.  I’ve done this regression with similar results  with OECD data (N = 39) in a paper published in Scientometrics in 2008.  See Relations between research investment inputs and scientific publication output: Application to an American paradox

And, if you don’t believe me, you might like to take a look at this report posted today at the ScienceInsider blog.  The title is excessively provocative; the authors of the report just call for increasing the demand for scientists rather than increasing the supply.
Study Suggests That the U.S. Could Use Fewer, Not More, Science Students

R. D. Shelton

Upcoming Conferences on S&T Indicators

1. The European Network of Indicator Designers (ENID-PRIME) has a conference scheduled March 3-5, 2010. Apparently they are not keen on yanks presenting papers, but they would probably let you come and listen. Not a bad idea, since it’s in Paris.

2. CWTS at Leiden University is organizing an S&T indicators conference September 9-11, 2010.  The posted information doesn’t say, but they usually just require an extended abstract for evaluation, and a PowerPoint for presentation.

3. The next International Conference on Scientiometrics and Informetrics is scheduled for 2011 in S. Africa, at the University of Zululand, which is near Durban.  The ISSI website says this is tentative, but I’ve heard Ronald Rousseau say it’s definite, and he’s the president of ISSI.  Since the last conference in Rio was in July, I presume it will be about then.  We can’t say that summer, though, for these countries south of the Equator.

 R. D. Shelton

China Forecast to Lead World S&T Soon

A new paper warns that the U.S. will soon be passed by the People’s Republic of China to lead the world in S&T.  The Race for World Leadership of Science and Technology: Status and Forecasts was presented at the International Conference on Scientometrics and Informetrics in Rio de Janeiro in July, 2009 by R. D. Shelton and P. Foland.  It presents a dozen indicators of inputs and outputs of the national S&T enterprises of the U.S., EU, and PRC.  Current indicators show that the U.S. and EU have comparable positions with China well below in many cases.  However, more careful analysis shows that the trends are alarmingly unfavorable to the West.  China is improving its position rapidly, and simple extrapolations show that it will soon lead in many important indicators.

The paper focuses on the number of scientific papers in the world’s leading journals for more detailed analysis.  Using a model developed by Shelton that relates world share of papers to world share of research investment, the paper forecasts that China will pass the U.S. and EU by 2017 in this key indicator.  Kostoff has shown that China already leads the world in some physical science databases of papers.  Leydesdorf and Wagner have a paper in press, partly based on Shelton’s input-output analysis, that also predicts that China will soon pass the West.

Forecast Graph

The Rio paper is posted at

R. D. Shelton

Journal of Interest: Scientometrics

Scientometrics aims at publishing original studies, short communications, preliminary reports, review papers, letters to the editor and book reviews on scientometrics. The topics covered are results of research concerned with the quantitative features and characteristics of science. Emphasis is placed on investigations in which the development and mechanism of science are studied by means of (statistical) mathematical methods.

View Citation

Journal of Interest: Cybermetrics (Spain)

Cybermetrics is both an Electronic-only Journal and a Virtual Forum (The Journal) devoted to the study of the quantitative analysis of scholarly and scientific communications in the Internet. It is open to world-wide researchers to publish and discuss their findings. Internet offers them new and increased capabilities to distribute their results to a greater audience.

Cybermetrics also maintains a series of directories of electronic resources (The Source), including secondary archives of interesting web papers in pdf format. The aim is to provide a reference tool to those researchers involved in the quantitative description and analysis of the Internet as a scholarly communication tool. It is also intended to add original data as a source or reference for larger studies, specially including figures about the distribution and evolution of R&D contents in the World Wide Web.

View Citation

Journal of Interest: Research Evaluation

Research Evaluation is a peer-reviewed, international journal. It ranges from the individual research project up to inter-country comparisons of research performance. Research projects, researchers, research centres, and the types of research output are all relevant. It includes public and private sectors, natural and social sciences. The term ‘evaluation’ applies to all stages from priorities and proposals, through the monitoring of on-going projects and programmes, to the use of the results of research.

View Citation