The Clash of Social and Knowledge Spaces: The Assumed Ismorphism Under the Hood

Abstract

So far, the dominant literature that has studied knowledge creation, transfer or absorption using social networks as explanatory variable has produced conflicting results. This research line has exponentially expanded in the last years and it has been always based on the perspective of social ties as channels where knowledge flows through. Different analysis using network structures, ego networks, social positions or relational properties has always worked on an implicit idea: social proximity describes knowledge proximity. Since innovation can be conceptualized as novel combination of knowledge, social proximity should be capable of describing how social structures affect innovative outcomes. However, vague and incoherent results are found across the literature. This may be due to the generalized assumption of an isomorphism between two spaces: the knowledge space, where innovators or researchers are located according to what they know; and a social space, where people are located according to who they know. This paper attempts to explore both dimensions as independent and assess the possibility of an isomorphism among their distances. The knowledge space is built upon the manifested compatibilities between cited papers or patents as a measure of distance between their authors, while the social space is represented by the social network scholars belong to. The overlooked interaction of these two spaces determines the potentiality of innovating since it is necessary to combine knowledge and the people who have it. After a theoretical review and analysis, an empirical test is performed. The data set comes from the research activity of 3 research departments of a Spanish University: Economics, Statistics and Business. Given the particular characteristics of the research activity and the organizational structure, the email network is taken as a representation of social ties among scholars. On the other hand, published and working papers produced by these departments are used to describe the knowledge space. Social closeness ?measured by the intensity of email activity between two researchers? is contrasted with expertise proximity ?evaluated according to bibliometric techniques?. Due to the anonymity of the email database and the consequent complexity of detecting an isomorphism between two sets of different size and without a rule of identification, I designed and used a special algorithm capable of confirming the absence of the isomorphism. Afterwards, I conducted a sociometric survey for corroborating results. I found knowledge and social distances to be significantly different implying that social spaces