Regional development in South Korea: accounting for research area in centrality and networks

This paper provides a first-ever look at differences of centrality scores (i.e., networks) over time and across research specializations in Korea. This is a much needed development, given the variance which is effectively ignored when Science Citation Index (SCI) publications are aggregated. Three quantitative tests are provided—OLS, two sample t-tests, and unit-root tests—to establish the patterns of centrality scores across Korea over time. The unit-root test is particularly important, as it helps identify patterns of convergence in each region’s centrality scores. For all other geographic regions besides Seoul, Gyeonggi, and Daejeon, there appears to be little promise—at least in the immediate future—of being network hubs. For these top three regions, though, there is a pattern of convergence in three-quarters of all research specializations, which we attribute in part to policies in the mid- and late-1990s.