企業社會責任報告書之確信服務提供者、確信程度及產業專家與可信度之關聯：實驗研究

Abstract

本研究探討確信與企業社會責任報告書（CSR報告書）的關聯，採用兩個 實驗進行研究。研究一（第一個實驗）探討確信服務提供者、確信程度以及兩者的交互作用，是否影響受試者對CSR報告書的可信度。研究一採用2（會計師或顧問）× 2（合理確信或有限確信）的受試者間設計，並招募88位的EMBA學生擔任受試者，代表投資人來衡量CSR報告書的可信度。研究結果顯示，不同確信服務提供者間，對可信度的影響並無顯著差異。然而，經合理確信的CSR報告書，其可信度顯著大於經有限確信的報告書，與假說預期一致。此外，當確信服務提供者為會計師時，經合理確信的CSR報告書，其可信度顯著大於經有限確信的報告書；但當確信服提供者為顧問時，不同確信程度間則無顯著差異。 由於當會計師為確信服務提供者時，不同確信程度對可信度的影響有顯著差異，因此研究二（第二個實驗）探討當確信服務提供者皆為會計師時，該會計師是否為財務報表確信的產業專家對CSR報告書可信度的影響。研究二採用2（產業專家與非產耊專家）× 2（合理確信或有限確信）的受試者間設計，並招募66位EMBA的學生作為受試者。研究二的結果顯示產業專家對CSR報告書的可信度有顯著影響。交互作用的結果進一步顯示，當會計師非為產業專家時，經合理確信的CSR報告書，其可信度顯著大於經有限確信的報告書；當會計師為產業專家時，經合理確信與經有限確信的報告書，其可信度則無顯著差異。此結果顯示產業專家於財報確信所獲取的知識，能夠外溢至CSR報告書的確信上。此外，較高的確信程度能彌補非產業專家所不足的產業知識；但當確信服務提供者為產業專家時，較高的確信程度則非為必要。

This thesis comprises two experimental studies on the assurance of corporate social responsibility (CSR) report. The first study examines the effects on investors’ perceived credibility of CSR report of assurance providers, assurance levels, and their interaction. Adopting a 2 (auditors vs. consultants) × 2 (reasonable vs. limited assurance) between-subjects design and recruiting 88 EMBA students as a proxy for investors to conduct the experiment, the results indicate that the effect of assurance providers on the CSR report credibility is not significant. But, consistent with the hypotheses, the perceived credibility is greater when the assurance level is reasonable than when it is limited. Further, when the assurance is provided by auditors, the perceived credibility is higher when the assurance level is reasonable than when it is limited; however, there is no such an effect when the assurance providers are consultants. Given that the assurance level has an effect on the perceived credibility when auditors provide the assurance, the second study focuses on the situation where the assurance providers are auditors, and examines the effects on the perceived credibility of whether the auditors are industry specialists in financial statements assurance, the assurance level, and their interaction. Adopting a 2 (specialists vs. non-specialists) × 2 (reasonable vs. limited assurance) between-subjects design and recruiting 66 EMBA students to run the experiment, the results indicate a significant effect of industry specialization. Further, when the auditor is not an industry specialist the perceived credibility is higher under the condition of reasonable assurance than that of limited assurance; but such an effect is not significant when the auditor is an industry specialist. This finding suggests that the knowledge possessed by an industry specialist in providing financial statement assurance can have a spillover effect on CSR report assurance. It also suggests that when the auditor is not an industry specialist, a higher level of assurance can compensate for the inadequacy of industry knowledge; but when the auditor is an industry specialist, providing a higher level of assurance seems unnecessary as far as the perceived credibility of CSR report is concerned.