A Comparative Analysis on Multiple Authorship Counting for Author Co-citation Analysis

摘要: As co-authorship has been prevalent within science communities, counting the credit of co-authors appropriately is an important consideration, particularly in context identifying knowledge structure fields with author-based analysis. The purpose this study to compare characteristics co-author methods by utilizing correlations, multidimensional scaling, and pathfinder networks. To achieve purpose, analyzed a dataset 2,014 journal articles 3,892 cited authors from Journal Architectural Institute Korea: Planning & Design 2003 2008 field Architecture Korea. In study, six different crediting are selected for comparative analyses. These first-author (m1), straight full (m2), fractional (m3), proportional total score 1 (m4), between 2 (m5), first-author-weighted (m6). shown data analysis, m1 m2 found as extreme opposites, since counts only first assigns all equally 1. With correlation scaling analyses, among five (from m6), group including m3, m4, m5 be relatively similar. When visualized network, networks differently presented due connections individual links. addition, internal validity shows that (m6) might considered better method author clustering. Findings demonstrate influence network results revealed