Measuring global research activities using geographic data of scholarly article visits
| Date | 07 August 2017 |
| DOI | https://doi.org/10.1108/EL-10-2016-0228 |
| Pages | 822-838 |
| Published date | 07 August 2017 |
| Author | Zhichao Fang,Xinhui Guo,Yang Yang,Zhongkai Yang,Qingchun Li,Zhigang Hu,Xianwen Wang |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
Measuring global research
activities using geographic data of
scholarly article visits
Zhichao Fang, Xinhui Guo, Yang Yang, Zhongkai Yang,
Qingchun Li, Zhigang Hu and Xianwen Wang
WISE Lab, Dalian University of Technology, Dalian, China
Abstract
Purpose –This study aims to analyse the geographical distribution of global research activities and to
investigate the knowledge diffusion embodied in scientic papers.
Design/methodology/approach –The geographical summary of Frontiers articles displays the number
of visits and categorizes where the visitors hail from. This study uses the records of 23,798 articles published
in 16 Frontiers journals from 2007 to 2015 to analyse the geographical distribution of article visits at both
country and city levels. The process of knowledge diffusion is investigated on the basis of the different visiting
patterns of new and old papers.
Findings –Most article visits are concentrated around major metropolitan areas and some high-tech
clusters. The top “visiting countries” include both developed countries and developing countries, and the USA
and China are two major players. Publishing cities dominate article visits for new papers; as time passes, there
is diffusion from the publishing cities to a broader area.
Research limitations/implications –The data on visiting for open access articles may be generated
from various repositories besides the publishers’ websites; these data are ignored, as they are not
signicant enough to have much inuence. There is also a lack of a basic theory in the data processing of
outliers in the data set. In addition, only static results are given in this paper, as the data were collected
on one day, for one time. A longer time period is necessary to track the dynamic diffusion process of the
observations.
Practical implications –Introduction of usage data will propose a novel way to analyse research
activities and track knowledge diffusion.
Social implications –The visiting data of articles offer a new way to investigate research activities at the
city level in a detailed and timely manner, for the geographical distribution of research activities and the
research resource allocation of a specic country to be explored.
Originality/value –This study measured the research activities of scientic papers by examining the
usage data. Compared with previous studies that focused on the geographical distribution of scientic
activities using publication data, citation data and even altmetrics data, usage data are at the forefront of this
research. Therefore, usage data offer a fresh perspective on methodology, providing more detailed and
real-time information.
Keywords Article usage, Big geographical data, Knowledge diffusion, Research activity
Paper type Research paper
Introduction
Research activities are involved in the entire process of scientic discovery and technology
innovation. Scientic knowledge can be produced and obtained through research activities.
Measuring the research activity of regions, including the geographical distribution, is
This research is supported by the “National Natural Science Foundation of China” (71673038,
61301227).
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
EL
35,4
822
Received 24 October 2016
Revised 7 May 2017
Accepted 8 May 2017
TheElectronic Library
Vol.35 No. 4, 2017
pp.822-838
©Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-10-2016-0228
important, as it helps obtain an overview of how active regions are and the potential
relationships among different areas, so that patterns of scientic communication and
diffusion can be clearly comprehended.
However, research activities are difcult to measure and trace in a timely manner owing
to the unavailability of previous related data. The outcomes of research, such as scientic
papers and patents, are the only way to investigate research activities. Currently, following
the development of digital publishing and electronic libraries, the access behaviour of
scientic publications is recorded and disclosed, through which a more comprehensive
picture of research activities can be obtained.
Literature review
The geospatial distribution of research activities
In previous studies, the publication, citation and altmetrics data were aggregated at
different scales and used to characterize the geospatial distribution of research activities
and map science clusters quantitatively in a physical space. Frenken et al. (2009)
proposed that these kinds of studies could be categorized under the general eld of
“spatial scientometrics”. Top continents, countries and cities for scholarly research have
been identied and visualized by a publication-ranking algorithm, network-based
citation analysis, or even journals (Mazloumian et al., 2013;Zhang et al., 2013). Centres of
excellence and hot regions around the world have been mapped, and the distribution
patterns and networks of relations have been drawn based on data collected from Scopus
or Web of Science (WoS) through a range of visualization programs (Bornmann and
Leydesdorff, 2011;Bornmann and Waltman, 2011;Bornmann et al., 2011;Leydesdorff
and Persson, 2010). Some altmetrics data with specic geographical information, such as
Mendeley data, have been used to investigate the geographical reading preferences of
readers from different countries, and it was proven that Mendeley readers
disproportionately select articles from their own countries in many different elds
(Thelwall and Maahi, 2015).
Matthiessen et al. (2010) found that most research activities are concentrated around
major metropolitan areas, especially capital regions, such as the Nordic capitals of
Hovedstaden, Helsinki-Uusimaa and Stockholm, or the German and Austrian capitals of
Berlin and Vienna, respectively (Kotzeva et al., 2015). These analyses may help to identify the
factors that drive successful research clusters, indicating rising stars and aiding city
planners and policymakers in building protable centres elsewhere (Van Noorden, 2010).
The process of knowledge diffusion
Knowledge diffusion has been extensively studied from various perspectives in the past few
decades. It can be traced back to studies on innovation and technology transfer or diffusion,
which have been conducted since the 1950s (Coleman et al., 1957;Crain, 1966;Winick, 1961).
A range of disciplines, including communications, economics and sociology, contributes to
the body of relevant research (Eveland, 1986;Kogut and Zander, 1992;Krugman, 1979;
Rogers, 1962). With the rapid growth of network science, knowledge diffusion through a
variety of networks is explored to explain the mechanism of how knowledge is diffused and
the evolution of network structures (Cowan and Jonard, 2004;Lin and Li, 2010;Liu et al., 2014;
Luo et al., 2014;Tang et al., 2006;Yang et al., 2015). The methods of network analysis provide
a promising research direction to investigate the dynamics of knowledge diffusion. However,
they cannot provide all of the necessary information regarding knowledge diffusion
occurring across the globe.
In the eld of scientometrics, by using classical methods, such as co-word analysis,
co-authorship analysis, citation analysis and co-citation analysis, the diffusion of knowledge
823
Global
research
activities
To continue reading
Request your trial