知识单元特征对发明者知识组合行为的影响——知识网络的视角

The Characteristics of Knowledge Components and Its Effect on Inventors' Knowledge Combination Behavior:From the Knowledge Network Perspective

发明作为技术创新活动的重要组成部分,对其过程和质量的影响因素的探究一直是创新领域的热点话题。知识组合领域的学者认为,一项发明本质上是一组彼此连接的知识单元,发明过程的微观机制表现为知识单元之间连接关系的建立和改变。现有的研究通常以个人或组织为单位,检验"特征—行为—绩效"之间的关系。但是以个人或组织为单位的分析假定一个人或一个组织在一定时间段内的发明活动是同质化的,忽略了不同发明活动之间的差异。本文结合知识组合理论和社会网络理论,以发明活动为分析单位,推导了影响发明者行为的微观机制,提出了知识多样性、知识嵌入性影响发明者创造性知识组合行为的理论框架。并以纳米技术专利为样本,通过Probit回归的方法进行实证检验。结果表明:知识多样性和知识嵌入性对发明者创造性知识组合行为分别有正向和倒U型的影响,且知识多样性会调节知识嵌入性对创造性知识组合行为的倒U型影响。本文的重要贡献在于揭示了发明过程的微观规律,对现有的创新研究是重要的补充,同时为企业、尤其是高新技术企业的技术管理提供了重要的参考。

As one of the most important research areas of technological innovation,the process and performance of inventions have been widely discussed in innovation research area.According to knowledge combination theory,an invention can be regarded as a set of reciprocally connected knowledge components and the generation of invention is the process of combining previously unconnected knowledge components or recombining connected knowledge components in different ways.Inventors,who are the actors of inventive activities,have been widely discussed in literature.Extant research has mainly discussed the impact of demographic characteristics of inventors and the external environment on their behavior or performance.For example,some studies have examined the relationship between inventors education background and their explorative innovation performance.However,one assumption is implied in these studies that different inventive activities conducted by one person were homogeneous,and the dependent variable was measured according to all inventive activities at certain time window.Furthermore,some scholars explained the difference among inventors performance from social network perspective.Social network analysis actually explores the impact of external environment in which one embedded on individual s behavior and performance.Two types of network have been discussed,that is network consisting of individuals and network consisting of collectives of individual(such as teams,organizational subunits,organizations,or even nation-states).In fact,the network in which the nodes represent the knowledge components is substantial in shaping inventors knowledge combination behavior but has been neglected.The nodes of knowledge network are knowledge components and the lines reflect the linkage or co-list relation between them.Inventors need to search necessary knowledge elements from the knowledge network,and then combine or recombine them to create inventions.If the searched knowledge components have been connected before,the inventors have no choice but to recombine them in different ways,for example,add new components into the existed combination.In other words,the characteristics of knowledge components which is shaped by the knowledge networks at time t-1 constrain the potential of further combination.Thus,it is deduced that characteristics of knowledge components shaped by knowledge network may be used to explain the varied performance of different activities conducted by the same person.Drawing upon knowledge combination theory and social network theory,two independent variables,knowledge diversity and network embeddedness of knowledge components,are identified in this research.Knowledge diversity measures the technological variety of a set of knowledge components which are to be combined.The mechanism through which knowledge diversity has impact on inventors knowledge combination behavior is that knowledge diversity provides the potential of novel combination.And network embeddedness influences the ease of further combination and realization of potentials.It is proposed that these two independent variables separately and jointly affect inventors creative knowledge combination.A single combination activity is the unit of analysis and patent data in nanotechnology filed is collected to examine the theoretical model.Probit estimator is employed and the empirical result indicates that knowledge diversity is positively related to inventors creative knowledge combination,and embeddedness have nonlinear(Inverted-U shaped)effect on inventors creative knowledge combination.Furthermore,the interaction between knowledge diversity and network embeddedness has positive but non-significant effect on the dependent variables.This research opens the black box of the process of invention at micro-level and makes great contribution to extant innovation theory.The results have important implications for inventors or organizations.Inventors can predict the value of possible combinations by analyzing the characteristics of knowledge components,which can reduce the risk and uncertainty of invention process.Organizations can also manage their research and development activities by adopting the same method.