Advancing Knowledge Space-based Indicators of Regional Technological Change

Technological change is a fundamental driver of long-run economic development, yet it is neither smooth nor evenly distributed across space. Regions rarely diversify into entirely new activities from scratch; instead, they typically expand into activities that are related to what they already do, while periods of major technological reorganisation can disrupt estab-lished competence bases and shift opportunities across places. Knowledge Space approaches have become a prominent way of studying these dynamics by representing the relatedness between technologies or other economic activities and using this structure to discuss diver-sification and upgrading opportunities, as well as forms of lock-in. At the same time, the central concepts that underpin these analyses – radical innovation, relatedness, capabilities, and complexity – are not directly observable. Their empirical meaning depends on how they are defined, measured, and interpreted. This dissertation promotes a measurement-aware perspective on Knowledge Space research in evolutionary economic geography. It argues that concept definition and match with associated indicators is a substantive part of theory application: operational choices embed assumptions that shape what can legitimately be inferred about path dependence and structural change. The dissertation makes four explicit contributions. First, it brings greater conceptual clarity to how the literature uses labels for major innovation and shows that commonly used empirical indicators capture only part of what these labels are often taken to imply. Second, it demonstrates that empirical measures of technological relatedness are sensitive to basic features of the underlying data and that appropriate measurement strategies depend on the data context. Third, it refines the inter-pretation of relatedness by distinguishing capability similarity across analytical levels and shows that the relative importance of these levels varies with technological complexity, sharpening the understanding of mechanisms behind technological upgrading. Fourth, it clarifies the interpretive limits of Knowledge Spaces by showing that claims about paths, bottlenecks, or diffusion require additional assumptions and can be strongly affected by rep-resentational choices. Overall, the dissertation strengthens the conceptual and methodologi-cal foundations of analyses of technological change in economic geography and supports more disciplined use of latent variables.