Year

2001

Degree Name

Doctor of Philosophy

Department

Department of Accounting and Finance

Abstract

Most (if not all) studies on the perceptions of auditor independence were conducted using a quantitative approach. Sending questionnaires to collect data and testing hypothesis by the use of statistics have been very common. The lack of literature on auditor independence using a qualitative approach has inspired this present study, since the advantages of the qualitative approach cannot be reached by the quantitative.

This is a perceptual study of auditor independence using an ethnographic, qualitative approach. Twenty-nine Indonesian auditors have been interviewed, in order to investigate their perceptions of factors that could affect auditor independence. Two theoretical models of independence, auditors-firms conflict of interest of Goldman and Barlev (1974) and Flint's (1988) material circumstances to independence, are used to explore the issues that were investigated. These issues are not "truths" to be proven, but used rather as interview guidelines. The interview data are summarised and pattern coded as proposed by Miles and Huberman (1984) and then analysed through description, analysis and interpretation suggested by Wolcott (1994). The use of pattern coding allows the researcher to classify answers and compare responses. The analysis introduced by Wolcott allows the researcher not only to analyse what auditors say but also to interpret beyond what they say.

This study found that auditor independence would potentially depend on external and internal factors or circumstances. The external factors or circumstances are audit firms' characteristics, clients' characteristics, the nature of auditors and clients relationship and sanctions enforced. Audit firms' characteristics included the size of the audit firm and non-audit services rendered to audit clients. The clients' backgrounds, existence of a corporate audit committee and clients' contribution to audit fees are regarded as clients' circumstances that could influence auditor independence. Competition among auditors, clients' directors' roles and the length of audit engagement are consequences of their relationship that could affect auditor independence. Sanctions enforced by the professional body could also influence auditor independence. On the other hand, internal factors influencing auditor independence comes from auditors themselves. The state of professional ethics of auditors as a consequence of professionalism and the auditors' religious values affect auditor independence.

In perceiving these factors, the study found that auditors have consensus on the effect of sanctions on auditor independence. They believed that sanctions could enhance auditor independence. Whereas, for other factors the auditors have various perceptions ranging from impairment, no effect and enhancement of auditor independence. In addition, the study also identified the auditors' understandings on independence, major threats and motivations among the above circumstances, the possibility to be total or partially independent and the auditors' efforts to maintain their independence.

The study concludes that auditor independence is a continuum, and is not black and white. Some new potential factors influencing independence have been found. The study also proposes a continuum model of independence and "better" and "worse" circumstances that potentially could affect auditor independence. Practical recommendations are suggested to the Indonesian Institute of Accountants, the Indonesian Government and relevant parties using audited financial statements.

Some limitations of the study are acknowledged such as the lack of previous qualitative studies on auditor independence, the cross-sectional data collection and a limited number of interviewees. Some suggestions for future research include narrowing the research site into one big audit firm and using a longitudinal process for data collection.

Download

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.