国内管理科学领域高校间学术论文合著网络的时间演化分析

Analysis on the Time Evolution of Cross-University Academic Papers Co-author Network in the Filed of Management Science within China

本文利用社会网络分析方法,以国内管理科学领域22种重要学术期刊1980年—2010年间的高校间的学术论文合著关系为研究对象,对国内管理科学领域高校间的学术论文合著网络的时间演化问题进行了分析。对该网络时间演化的研究,可以揭示出国内高校之间科研合作的发展规律及运行状态,从宏观的角度理解国内高校之间进行科研合作的本质。研究发现国内高校之间的学术论文合著网络规模不断增加,网络新进节点具有优先连接性,高校间的合作关系更为紧密,高校之间的科研合作效率更高,并且合著网络处于健康的演化发展状态。

Cross-team, cross-organizational and cross-regional research cooperation have become ubiquitous and necessary forms in the field of academic research. Domestic and foreign studies have shown that more and more direct cooperation between scientists are expanding from within the team to in-between teams. In the meantime, a similar expansion is taking place from within universities to in- between universities. Universities have subjective initiative in academic cooperation and participants, and they can be conceptualized as a social entity. For this reason, university level academic co-author network falls under the category of a social network. The study on the characteristics of China ＇ s universities＇s academic cooperation from the social network perspective has practical significance.This paper studies the time evolution of academic co-author networks, which could assist in revealing the development law and the essence of the academic cooperation. This paper used author information of papers published in important academic journals （Academic journals specified by the National Natural Science Foundation of Department of Management Sciences, 22 academic journals） in the domestic management science as data sample. The paper considered universities as nodes and co-authorships as sides, as well as used the social network analysis tool Ucinet to establish a social network model of university academic cooperation in the field of management science. This paper used time as dimensions and established a social network model by dividing the period of the universities＇ academic cooperation, and then compared the network index in different time periods. Finally, the paper investigated the evolution characteristics of domestic universities＇ academic cooperation in recent decades. The first section is a literature review of seminal works related to this study, followed by our proposed research topics. The second section is the research method. The third section discusses the time evolution analysis of scientific collaboration networks between universities and the time division basis, which could be divided into three periods （1980 to 1994, 1995 to 2003, and 2004 to 2010）. The last section is a summary of analysis results and important findings and implications. Our findings and conclusions are： （A） The development of co-author network is healthy and cooperation scope is growing. This finding has many indicators ： （ 1 ） network connectivity is becoming better, （2） the scale of the maximum cooperation sub-network is growing, （3） the power of nodes has become decentralized, （4） network stability is strengthening, and （5） network openness is gradually increased as the barrier to entry is being diminished. （B） The new nodes in the co-author network have selective priority connectivity. This finding can be reflected in the foUowing indicators： （1） the new nodes in the early co-author network have stronger priority connectivity, （2） this priority connectivity gradually decreases along with the increased openness of academic cooperation, and （3） news nodes have the characteristic of local priority of connection and high center-priority connection. （C） The competition between nodes is intensified. The research leaders in a network are always changing. We find that sub- components of academic cooperation are academic groups formed by nodes with closer academic relationship. In this situation, the smaller academic groups which fail to fit into the big academic cooperation network would be easily pushed out from the research area. （D） The differentiation between small groups is not obvious and there are no serious research factions. Research community has grown from a single research leader to many research leaders. The vulnerability of the scientific collaboration network is gradually eliminated. Important implications can be drawn from these findings. Firstly, universities could vigorously enhance their own scientific research strength, and enhance the ability to obtain external resources. Otherwise it will subject to the exclusion of other community members in the scientific collaboration network. Secondly, it is important to clarify each member＇s characteristics and advantages, their location in the network, as well as their academic tribes. In order to improve the efficiency of cooperation and ultimately achieve a win- win result, each member needs to know his/her strength in relation to other members. Thirdly, star colleges and universities should fully play leadership roles. They should actively drive the development of their cooperation within the field of academic networks. Moreover, they cannot ignore the importance of complementary research.