Page 18 - The lraternational Journal of the Royal Society of Thailand.indd
P. 18

The International Journal of the Royal Society of Thailand
             Volume XV-2023



             Authorship profiles of researchers affiliated with Thai organizations
                    Based on the snowball sampling, a total of 120 researchers, affiliated with Thai

             organizations from Circles 1 to 4 of IR_01, were included in subsequent principal
             component analysis. The first two PCA dimensions explained a total of 58.8% of the
             total variation among the included researchers (Figures 3 & S1).

                    The primary driver of the first PCA dimension was the maximum and average
             number of publications in the past 5 years, followed by the number of research

             communities and the proportions of co-authors with affiliations in Western Asia and
             Thailand, respectively (Figures 3A & S1). Notably, the proportion of co-authors
             with Thai affiliations exhibited a negative correlation with the other drivers of PCA

             Dimension 1 (Figures 3A & S2). At the same time, the second PCA Dimension
             primarily differentiated the authorship profiles of the included researchers based on
             the time since their first publication (academic experience), followed by the diversity
             of research domains and the proportions of co-authors with affiliations in Western Asia
             and Southern Asia, respectively. (Figures 3A & S1). While the clustering coefficient

             exhibited a negative correlation with both the maximum and average annual publication
             numbers, as well as the number of research communities (Figures 3A & S2).

                    The projections of individual researchers onto the first two PCA dimensions
             revealed that the majority of included researchers aligned from the upper-left quadrant
             (representing junior researchers with a small number of published articles) to the

             lower-right quadrant (indicating Senior Researchers (SRs) with a larger number of
             articles) on the individual PCA plane (Figure 3B). The authorship profile of Index
             Researchers  IR_01  and  IR_02  was  notably  distinct  from  that  of  most  included

             researchers, as indicated by their highest coordinates in both dimensions, placing them
             in the upper-right quadrant (Figure 3B). This quadrant represents junior researchers
             with many published articles, and we will refer to them hereafter as High-performance
             Junior Researchers (HJRs). The detailed authorship profiles of three selected High-
             performance Junior Researchers (HJR_01, HJR_02, and HJR_03) and three Senior

             Researchers (SR_01, SR_02, SR_03) are shown in Figures S3 and S4, respectively.


















                                     A Potential Use of Ego-centric Co-authorship Network Analysis for
             10                      Inspecting Research Misconduct: A Case Study of Thai Researchers
   13   14   15   16   17   18   19   20   21   22   23