Individual and Organizational Effects of Knowledge Production: A Multi-Level Study

Abstract

This dissertation investigates individual and organizational effects of knowledge production in three chapters. I depart from the notion that both interpersonal collaboration between and within organizations as well as recombination of knowledge components are instrumental in the production of new knowledge. The first chapter investigates the origins of collaborative knowledge production in plastics electronics. Using survey data, I explore how individual and organizational characteristics drive collaboration preferences. I avoid looking at established collaborations because these bias collaboration determinants. I find that resource provisioning and social similarity influence preferences and that these effects are moderated by collaboration experience and organizational objectives. Additionally, the study suggests that individuals facing an aspiration gap are more favourable towards collaboration, and proposes that relational capability, in the form of networking skills, openness, and information will moderate this relationship. The second chapter looks at the recombination of knowledge components. Using patent data from US semiconductor firms, we explore how firms build on old knowledge components to create new patented inventions. The findings suggest that familiarity of components contributes to innovative success but that adding novelty to familiar combinations is on average even more successful. We also find that the effects of different types of knowledge recombinations are moderated by the time in between these collaborations as well as by the search for non-domain specific knowledge. The final chapter posits that the search for knowledge components and the search for collaboration partners should not be understood as two isolated determinants of invention. I submit that finding knowledge components (objects) or knowledgeable partners (holders) can be construed as alternative solutions to the same knowledge problem and describe a process model of invention as escalating recombinant search. The dissertation ends with a reflexive conclusion on the contributions made in these three chapters.