The literature reveals that the dominant method of research was the quantitative questionnaire (Beasley, 1996; Moyes and Hasan, 1996; Porter, 1993). The vast quantitative survey-based empirical studies have established a body of knowledge about the auditors’ responsibility for detection of fraud, but failed to conduct deeper analysis of the research phenomena, particularly the question of how stakeholders’ react to fraud detection. As Saunders et al. (2003, p. 92) pointed out: “the data collected by the survey strategy may not be as wide-ranging as those collected by other research strategies”.
The use of a qualitative approach to support a quantitative survey will serve to understand fully the question of fraud detection. This research paper is very much an exploratory study into the auditors’ responsibility for detecting fraud in Barbados. Personal face-to-face interviews were held with a random sample of auditors and users, using a semi-structured interview schedule that was developed based on the issues coming out of the literature. The use of face-to-face interviews was chosen as the research method because of the likelihood of a high response rate, a high degree of accuracy and minimal non-response, and the need to discover underlying motivations, feelings, values, attitudes and perceptions about fraud detection (Alleyne, 2002; McDaniel and Gates, 2001). In addition, the project demanded that a fairly wide-ranging approach be taken to understand the issue, and the fact that interviews generate a much richer source of insights into questions under investigation (Strauss and Corbin, 1998).
Judgemental sampling was used as a basis for the selection of the size of both groups of respondents (auditors and users), since the aim was to include all those persons related to the phenomenon (Hudaib, 2003). Much emphasis was placed on quality rather than quantity. Following Arber (1993) and Oppenheim (1992), Hudaib (2003, p. 106) used a similar approach and stated that:
. . . the number of participants in each group is determined by interviewing as many participants as possible until it is felt that no new ideas are emerging from the in-depth interview.
Random telephone calls were made to persons enquiring whether they were willing to be interviewed. Some difficulties were encountered in obtaining personal interviews with some auditors and certain senior managers of some organisations, given their hectic busy schedules and their willingness (or lack thereof) to share hard and sensitive data. These interviews were obtained to assist in finding out what is happening and to ask questions (Saunders et al., 2003, p. 96).
The interviewees were split into two groups of auditors and users. The auditor group comprised 19 auditors (including two partners/managers of the four major international audit firms, two senior government auditors and nine other sole practitioners). The user group totalled 24 and comprised 16 senior managers of auditee companies (including five public limited companies, four financial institutions and seven other businesses), seven user-investors and one representative from ICAB. On average, the 24 audit respondents had 18.53 years of experience with 5.92 standard deviations. On average, the 16 senior managers within the user group had 13.94 years of experience with 4.725 standard deviations, while on average, the other eight respondents (seven user-investors and the member from ICAB) had 8.38 years of experience as investors in public limited companies. The high level of experience of the sample should provide knowledgeable views on fraud and auditing in Barbados. In addition, 24 respondents (19 auditors and five users) had professional accounting qualifications.
Respondents were asked to rate certain questions on a five-point Likert scale varying from 1 (strongly disagree) to 5 (strongly agree). The responses to these questions are shown at Table I. The questionnaire also contained in-depth questions pertaining to fraud on which interviewees were asked to comment. Interviewees were also allowed to speak at length on any issues regarding auditing and fraud. Each interview lasted approximately one hour. The full interview questionnaire schedule is shown in Appendix 2, Figure A1. Certain questions were adapted and modified from Farrell and Franco’s (1999) study. The sample size is relatively small and those interviewed are not intended, or claimed to comprise a representative sample of persons. Consequently, the results should be interpreted with caution and could serve as a springboard for further research into this important area.
Baca Juga :