The Characteristic Analysis of Researches Network for Journal of Korean Neuropsychiatric Association / 신경정신의학

ABSTRACT

OBJECTIVES: This study evaluated the structural characteristics of a scientific network of psychiatry and the effect of social networks on the performance of scholars. METHODS: The data were extracted from 261 articles published from 1996 to 2013 in the Journal of the Korean Neuropsychiatric Association, and were transformed into a co-author and their affiliation matrix. We used measures from network analysis (i.e., degree centrality, weighted degree centrality, eigenvector centrality, betweenness centrality) for evaluating the effect of co-authorship network on the performance of scholars (h-index). Netminer 4.1 was used for the network analysis. RESULTS: Both co-authorship and affiliation network demonstrated power law distribution. Coauthor's centralities were correlated with research achievements. Results from poisson regression analysis showed that the eigenvector centrality has a significant positive influence on the h-index and the weighted degree centrality has a significant negative influence on the h-index. CONCLUSION: This study shows that the small world phenomenon exists in the psychiatric coauthorship network, and finds collaboration patterns and effects on scientific performance. The results suggest that in order to achieve better research performance it would be helpful for scholars to work with other well-performing scholars and avoid other scholars who previously worked together.