The exploration in the size of scientific collaboration team using kernel density estimation
| Date | 26 August 2022 |
| Pages | 821-843 |
| DOI | https://doi.org/10.1108/AJIM-04-2022-0183 |
| Published date | 26 August 2022 |
| Author | Ran An,Wei Shan |
The exploration in the size of
scientific collaboration team using
kernel density estimation
Ran An
Library, Renmin University of China, Beijing, China and
School of Information Resource Management, Renmin University of China,
Beijing, China, and
Wei Shan
School of Economics and Management, Beihang University, Beijing, China
Abstract
Purpose –Scientific collaboration is becominga common pattern in the social organizationof knowledge production.
Thepapertriestofigureouttherelationship between scientific collaboration team size and scientific output.
Design/methodology/approach –Basedon ESI database from year 2009–2019,the paper describes changes
ofcollaboration teamsize from one author to more than10 authors in 22 disciplines.Kernel density estimationand
multidimensionalkernel density estimation method are used to calculate optimal collaboration team size and
appropriatecollaboration team sizein 22 disciplines. As bandwidthis one of the major issues in construction of
kernel densityestimation, the paper usesfive different algorithmsto calculate bandwidth. Themethod with the
lowest mean absolute percentage error is chosen.Robustness test is conductedbased on different sets of data.
Findings –The results show that scientific collaboration becomes more widely and deeply. As time goes by,
collaboration team size is becoming larger and larger. Natural science disciplines have larger collaboration
team size and faster growth rate than social science disciplines. Considering both qualitative and quantitative
measures, the paper proves the universality of optimal and appropriate scientific collaboration team size
among 22 disciplines and calculates the specific number.
Originality/value –The paper tries to investigate the law of scientific collaborationteam size variation and
provide a full picture of evolution of collaboration team size among 22 disciplines in 10 years. The paper first
applies distribution method to figure out the relationship between scientific collaboration team size and
scientific output and provides optimal collaboration team size and appropriate collaboration team size.
Keywords Scientific collaboration, Kernel density estimation, Scientometrics, Essential science indicators,
Discipline, Natural science
Paper type Research paper
1. Introduction
Over past decades, the process of knowledge creation has changed. As teams are likely to
produce high impact research, scientific collaboration becomes more and more popular
across nearly all fields (Wuchty et al., 2007). Scientific collaboration is defined as that scholars
are working together with a common purpose of producing new scientific knowledge (Katz
and Martin, 1997). In addition, more and more scholars focus on the intersection of established
disciplines and thereby collaboration among scholars from different disciplines has increased
accordingly (Liu et al., 2011;Woelert and Millar, 2013;Nyhagen and Baschung, 2013).
From 1960s, the question of scientific collaboration had attracted numerous scholars to
investigate into. The father of scientometrics price noticed the increasing trend of scientific
collaboration and he verified this conclusion based on papers during 1910–1960 (Price, 1963).
Scientific
collaboration
team size
821
The authors would like to thank the editor and anonymous reviewers for their valuable comments and
helpful suggestions. Furthermore, this work was supported by the National Natural Science Foundation
of China (Grant No. 71971008), the Beijing Natural Science Foundation (Grant No. 9222020), the Major
Program of the National Fund of Philosophy and Social Science of China (Grant No. 21ZDA012).
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2050-3806.htm
Received 9 April 2022
Revised 26 June 2022
Accepted 7 August 2022
Aslib Journal of Information
Management
Vol. 75 No. 5, 2023
pp. 821-843
© Emerald Publishing Limited
2050-3806
DOI 10.1108/AJIM-04-2022-0183
Since 1990s, the quantity and size of scientific collaboration have expanded. The literature
about this issue is rich. Based on publications in WOS SCI database, international
collaboration rate had grown from 10% in 1990 to almost 20% in 2000 (Zitt and Bassecoulard,
2004). Between 1986 and 1999, the number of international collaborated papers had doubled
or even tripled (Archibugi and Coco, 2004). Between 2000 and 2010, the share of coauthored
publications grew from 69.9% to 85.8% (Ossenblok et al., 2014). Since 1950s, the larger and
diverse team is promoting across different fields (Wuchty et al., 2007).
Scholars investigate into different aspects of scientific collaboration including
collaboration networks, geographical proximity, social proximity and collaboration
pattern, etc. With the expansion of the size of scientific collaboration, the size of scientific
collaboration and its relation with scientific output attract scholars’attention. Regarding to
scientific output, the performance of scientific collaboration output is determined by multiple
factors, including degree of collaboration which are international collaboration, institutional
collaboration and individual collaboration (Schmoch and Schubert, 2008;He et al., 2009);
characteristics of collaborators such as geographical location and popularity
(Ord
o~
nezmatamoros, 2008;Abramo et al., 2009); characteristics of collaboration such as
inertia, diversity and strength (Guan et al., 2015), etc. However, different scholars cannot
reach a unified conclusion about the effect of each factor on scientific output.
The development of big science and interdisciplinary research calls for large-scale
collaboration. Scientific collaboration team size becomes one of the most important factors
when considering scientific collaboration. Even some researchers had predicted the demise of
the single-authored paper (Price, 1963). A successful team evolves toward a size that is large
enough to enable specialization and effective division of labor among teammates but small
enough to avoid overwhelming costs of group coordination (Guimer
aet al., 2005). Larivi
ere
stated that when it comes to the relationship between collaboration and scientific impact, size
matters (Larivi
ere et al., 2015). Based on 65 million papers, patents and software, Wu proved that
different team size influence scientific output differently. To be specific, smaller team is likely to
disrupt science and technology, whereas larger team is likely to develop science and technology
(Wu et al., 2019). Therefore, the size of collaboration team is crucial to the scientific output.
Theoretically, the more authors collaborate, the more resources they can bring, and the
better scientific production will be made. There is evidence that most of cutting-edge work
tends to emerge from large, well-funded collaborative teams involving many collaborators
(Simonton, 2013). Large collaboration team can bring different sets of skills and ideas
(Franceschet and Costantini, 2010), increase diversity of collaboration team (Mart
ın-Sempere
et al., 2002;Beaver, 2013), bring human and financial resources (Larivi
ere et al., 2015) and
strength social network and increase social capital (He et al., 2009). Pan and Tan have proved
that there is a weak correlation between collaboration size and scientific output or H index
based on the data of the Derek de Solla Price Memorial Medal. The author with more
collaborated partners and bigger collaboration networks has a higher possibility of better
scientific output (Pan and Tan, 2012). The conclusion that number of authors, number of
organizations and number of countries contribute positively and significantly to the number
of citations a paper received is proved in the six specialties (Wagner et al., 2017), prestigious
finance journal (Chung et al., 2009) and six journals of medical (Figg et al., 2012). Based on 165
scientific discipline sectors in Italy ORP database, Abramo found that both research
productivity and quality are positively related with the degree of international collaboration
achieved by a scientist (Abramo et al., 2011). Puuska also proved that co-publications by 10
authors or more gather significantly more citations than other publications in Finland
(Puuska and Muhonen, 2014). It can be seen that above conclusion is held at a detailed level of
a single discipline or specific author or specific journal group.
However, with the growing trend of scientific collaboration and increase in the complexity
of collaboration relationship, some researchers held that collaboration activities are largely
AJIM
75,5
822
Get this document and AI-powered insights with a free trial of vLex and Vincent AI
Get Started for FreeStart Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting