229 NTU Management Review Vol. 31 No. 3 Dec. 2021 error above the mean degree centrality raises the likelihood that the firm received a GC opinion by 4.35 %. We also find a significant positive coefficient on CPA1&2_NW. These results suggest well-connected auditors not only develop knowledge through collaboration but also receive more oversight through the auditors’ signing networks which leads to higher audit quality. We conduct several additional tests to ensure the robustness of the results. First, besides degree centrality, we use other three measures of network centrality. Specifically, we calculate closeness centrality, betweenness centrality, and eigenvector centrality of the lead auditor, and then we re-estimate model (1). The empirical results show that only the eigenvector centrality of the lead auditor is significantly positively associated with GC. Second, we separate the observations into four groups by the lead auditor’s number of clients. The results indicate that the positive association between degree centrality and the possibility of the auditor issuing a going-concern audit opinion is pronounced for the groups with lower numbers of partner audit clients. 4. Limitations and Future Research We offer some suggestions for future research. This study focuses on analyzing audit assurance of financial statements. Future research could use other individual auditor characteristics, such as academic or professional qualifications and other experiences in firms, to explore the association between auditors’ signing networks and audit outcomes. Furthermore, future research could try to distinguish the information effect and the oversight effect from auditors’ signing networks and examine the influence on audit outcomes. 5. Originality/Contribution This study makes the following contributions. First, this is one of the few studies to empirically examine the strength of auditors’ signing networks as a factor in auditor quality and thus contributes to the growing literature examining how individual auditor characteristics influence audit outcomes. Furthermore, we use four measures of network connectedness—degree centrality, closeness centrality, betweenness centrality, and eigenvector centrality—that provide a more complete picture of the influence of auditors’ signing networks on audit outcomes. Finally, the results suggest that auditors’
RkJQdWJsaXNoZXIy ODg3MDU=