由於臺灣的上市櫃企業係滿足特定要件才需要主動揭露審計公費，因而造成2002年至2008年（以下簡稱選擇性樣本）間企業不普遍揭露其公費資訊的情況。這個資料不完整的問題，嚴重地限制了研究者對國內審計公費相關議題的了解。然而企業因導入國際財務報導準則（International Financial Reporting Standard，簡稱IFRS）而需要積極尋求會計師的協助，以及資訊揭露評鑑系統鼓勵企業自願性揭露會計師公費，導致2009年以及2010年二年（以下簡稱完整樣本）有近九成的企業揭露其審計公費。利用這個獨特的完整樣本，我們得以進行審計公費的相關研究並獲得以下的發現。首先，國內的審計公費模型與相關文獻非常一致。其次，針對選擇性樣本與完整樣本的跨式檢定，我們發現這二組樣本的結構性變化並不重大，換言之，爾後研究者可以忽略選擇性樣本可能存有的樣本偏誤性問題。最後，無論利用異常應計數、盈餘門檻或者繼續經營有疑慮之審計意見，本文均未發現異常審計公費對審計品質有不利影響的證據。

Most companies listed in Taiwan did not disclose audit fees from 2002 to 2008 (selected sample period hereafter) due to conditionality. The resulting data incompleteness limits audit fee research and understanding of audit pricing in Taiwan. However, owing to IDTRS and the forthcoming adoption of IFRS, around 90 percent of listed companies in Taiwan, disclosed audit fees in 2009 and 2010 (complete sample period hereafter) Using the comprehensive data set thus provided, we have conducted several investigations of auditing. First, we find that audit pricing in Taiwan follows patterns very similar to those described in prior literature. Second, a comparison between the selected sample period and the complete sample period shows that the structural change between the two periods is immaterial. This latter finding implies that using audit fee data from the full sample period alleviates concern over potential sample selection bias. Finally, using various proxies for audit quality, abnormal accruals, earnings threshold, and auditors' propensity to issue going concern opinions, we find no evidence to support the claim that abnormal audit fees impair audit quality.