Ethical evaluations and behavioural intentions of early career accountants: the impact of mentors, peers and individual attributes
The authors would like to gratefully acknowledge the Accounting and Finance Association of Australia and New Zealand (AFAANZ) for the research grant for this project as well as Barry Cooper, Steven Dellaportas, Jenny Stewart, and participants at the 2008 Annual AFAANZ Conference for their comments.
Abstract
This study examined how mentoring support, peer influence and individual attributes of early career accountants (ECA) influence their ethical evaluations and behavioural intentions. Respondents indicate that their evaluation of the seriousness of the ethical conflict is affected by the perceived standard of ethical conduct of their peers, their personal ethical orientation, the extent of ethics education at university, and gender. ECAs’ evaluation of a senior colleague's unethical behaviour is affected by mentoring support and the perceived standard of ethical conduct of peers. In terms of ECAs’ willingness to contact accounting professional bodies for ethical advice, the size of the accounting firm and the extent of their ethics education at university are significant factors. Furthermore, the likelihood of respondents choosing a more ethical decision is correlated with his or her individual ethical orientation and the extent of ethics education at university.
1. Introduction
The recent wave of corporate scandals has increased public scrutiny over the professionalism of accountants, calling into question the ethical behaviour of accounting professionals yet again. Prior research relating to accounting ethics has largely focused on two groups: accounting students and accounting practitioners. For example, the work by Cohen et al. (1998), Mintz (1996) and, more recently, Malone (2006) have focused on accounting students’ ethical attitudes, and the findings in general suggest that factors such as gender, formal ethics education and moral reasoning ability affect accounting students’ evaluations of ethical issues and their subsequent actions to such issues. Research by Buchan (2005), Dreike and Moeckel (1995), Karcher (1996), Ponemon (1999) and the early work by Aranya and Ferris (1984) have, in contrast, focused on accounting practitioners’ ethical attitudes and decision-making behaviour. In the case of accounting practitioners, personal values, seniority, gender and moral sensitivity were found to be significant factors affecting their attitudes towards situations of ethical conflicts and their intended behaviour (i.e. choosing ethical actions). However, no study to date has been undertaken in this area with respect to early career accountants (ECA) or new accounting graduates who have recently started their career in a public accounting firm. As such, the present study aims to fill this gap in the extant literature.
This paper presents a study on the impact of mentors, peers and individual attributes on ECAs’ ethical evaluations and behavioural intentions. For the purposes of this study, an ECA is defined as ‘an accounting graduate recruit in a public accounting firm who has 3 or less years of work experience’. Furthermore, ethical evaluations pertain to (i) the extent to which ECAs interpret the seriousness of a situation involving an ethical conflict, and (ii) their assessment of the actions of their senior colleagues in such a situation. Ethical behavioural intentions, on the other hand, relate to the following two dimensions: the likelihood of an ECA contacting his or her professional accounting body for guidance on an ethical conflict situation, and the willingness of such an accountant to act ethically when confronted with such a situation. The overall objective of this study is to determine the effects of mentoring support, the perceived standard of peers’ ethical conduct, and selected individual attributes of ECAs (namely, their personal ethical orientation and the extent of ethics education at university) on their ethical evaluations and related behavioural intentions. The motivations for this study are discussed in the following section.
2. Motivations for the study
The first motivation for the present study relates to the scant and limited empirical evidence on the factors affecting the ethical evaluations and behaviour of ECAs at the workplace. This shortcoming is surprising given that new accounting recruits are a vital segment of the accounting profession in that they represent an invaluable resource pool of accountants for future generations. Furthermore, given the escalating problem of accounting skills shortage, there is now even a greater need to nurture, train and retain high-quality accountants, which in turn has significant implications for competitive advantage (Cohen et al., 1998). The relevance of ethics education for new accounting graduates is also reflected by the findings of Ahadiat and Mackie (1993) whereby employment recruiters from both Big Six and non-Big Six accounting firms ranked students’ ethical propensities as the most important criteria in their recruiting decision. However, prior research has mainly focused on understanding the ethical attitudes and decision-making of tertiary accounting students and senior public accounting professionals, such as partners and managers (Zeigenfuss, 1999; O’Leary and Cotter, 2000; Guntz et al., 2002). Therefore, the present study aims to fill this gap in the extant literature by providing a better understanding of the factors that affect the ethical stance of budding professional accountants, which in turn has implications for the overall quality of the future workforce.
A second motivation for the present study is that prior studies on mentorship indicate that mentors play an important role in affecting their protégés (Scandura and Viator, 1994; Kaplan et al., 2001). In general, mentors can be defined as ‘high-ranking organisational members who possess significant experience and knowledge and are committed to providing support to a protégé's professional career’ (Kaplan et al., 2001, p. 195). Accounting professional firms are naturally designed to train young accountants through a mentoring system where audit partners and senior managers keep close watch on the professional development of young or new recruits. Furthermore, as new accounting recruits in the workforce can be highly impressionable and open to socialisation effects, mentors can have a strong effect on shaping their professional and ethical values. According to the organisational socialisation perspective (Anderson-Gough et al., 1998; Fogarty and Ravenscroft, 2000), the professional development of a new accounting recruit is open to numerous organisational-related factors and formal mentoring. Prior studies, such as Wimbush and Shephard (1994), have shown that supervisors have a significant impact on the ethical behaviour of their subordinates. However, no study has attempted to directly examine the impact of mentoring on new accounting recruits’ ethical evaluations and behaviour. Furthermore, most of the prior studies have been undertaken in large accounting firms (e.g. a Big Four or a Big Six firm). Therefore, the impact of mentoring on new accounting recruits in small to medium-sized accounting firms remain unclear, warranting further study in this area.
Besides mentors, new professional recruits also learn from their peers and might even tend to mimic them (Jones and Kavanagh, 1996). According to Kram and Isabella (1985), peer relationships that involve individuals of similar position potentially have a similar effect as mentors in terms of providing psychological and work support. In most accounting firms, there will be a number of ECAs at any one time, and such accountants from smaller firms also have opportunities to interact with other new accounting recruits when enrolled in professional development courses. Given the lack of empirical evidence of the influence of peers on ECAs’ ethical evaluations and their behavioural intentions, the third motivation for the present study relates to filling this gap in the literature.
The fourth motivation for the present study is to provide a better understanding of the relationship between ECAs’ ethics education undertaken at university and their ethical evaluations at work. Armstrong et al. suggest that ethics educators ‘set the stage for ethical behaviour by increasing moral sensitivity, moral reasoning and moral motivation’ (2003, p. 10). Furthermore, Gaa and Thorne (2004) highlight the need for ethics education at university to be given greater importance. The study of Armstrong et al. (2003), which reviewed empirical evidence linking ethics education to accounting student's moral sensitivity and development, further suggests that education is likely to be a key factor. Yet there is little empirical evidence on the relationship between ethics education at the university level and ECAs’ ethical evaluations and behaviour from an organisational context.
The fifth and final motivation for the present study is to inform on the link between accounting professional bodies and ECAs’ information seeking behaviour. A recent study by Jackling et al. (2007) involving 66 respondents from the International Federation of Accountants indicates that accounting professional bodies ought to play a more active role in prescribing the nature of ethics education. Furthermore, both CPA Australia and the Institute of Chartered Accountants in Australia, the two largest accounting bodies in Australia, are involved in the professional training of accountants and have clear policies and guidelines on the standard of ethical behaviour. In July 2006, APES 110: Code of Ethics for Professional Accountants by the Accounting Professional and Ethical Standards Board (APESB) was released and became mandatory for all members of the accounting professional bodies.1 From a professional socialisation perspective, it can be argued that professional training and support are key mechanisms for promoting particular values and beliefs among newcomers to a profession. Yet little is known as to how willing ECAs are to gain advice and support from their respective professional bodies. However, such an understanding will be useful for professional accounting bodies to identify new and better ways of supporting ECAs in their ethical development.
The remainder of this paper is organised as follows. In the next section, an overview of the conceptual framework for the study is provided with the development of several hypotheses based on organisational socialisation, mentoring and ethical development theories. This is followed by the research method, data analysis and results of the study. Finally, a discussion of the results, followed by the study's conclusions and limitations, are provided.
3. Hypotheses development
Figure 1 provides an overview of the conceptual framework for this study. A review of prior literature indicates that the ethical reasoning ability and the associated behavioural intentions of accountants can be affected by both organisational- and individual-related factors (Karcher, 1996; Guntz et al., 2002; Buchan, 2005). The present study brings together several factors from both areas, and in doing so, focuses on the different types of support available to new accounting recruits (i.e. mentor, peer and professional level support, as well as individual- or personal-level ethical orientations and prior ethics education).
Overall conceptual framework.
3.1. Mentoring support
According to organisational socialisation research, an employee's adjustment to the organisation is directly affected by the methods of socialisation adopted by the organisation (Anderson-Gough et al., 1998; Fogarty and Ravenscroft, 2000). At entry, a new recruit is generally confronted with an unfamiliar and ambiguous organisational context, and will attempt to make sense of that context (Siegel et al., 1995). Such recruits, in turn, are seen to look to individuals in senior positions, and/or to company norms and procedures as displayed by the other staff for guidance on what is acceptable or unacceptable behaviour. Mentoring is regarded as a sponsorship system in which a senior staff member will take a personal interest in and perform advisory and guidance functions to develop their protégés’ careers (Siegel et al., 1995). Early studies by Kram (1983), Noe (1988) and Scandura and Viator (1994), indicate that mentors might provide a career-related mentoring function at a more social level. Career development mentoring involves activities that guide the protégé on career advancement through the provision of challenging work assignments, coaching, sponsorship, protection, exposure and visibility. By contrast, a social support function involves providing social acceptance, confirmation and personal friendship.2 Much of the research on mentoring relationships in public accounting firms has focused on documenting the extent to which these different types of mentoring functions are undertaken (Scandura and Viator, 1994; Barker et al., 1999; Viator, 2000) and have also found significant associations with outcomes such as the protégé's intention to leave and their organisational commitment (Scandura and Viator, 1994; Stallworth, 2003). Several studies have also found that mentoring functions and their impact on the protégés might vary across different national cultures (Scandura and Viator, 1994; Barker et al., 1999; Herbohn, 2004). For example, Herbohn (2004) found the importance of social support was lower for Irish and Australian accountants compared to their US counterparts.
Mentors can positively influence a protégé's perceptions of what is acceptable ethical behaviour and what is not acceptable behaviour in various ways. First, since a career development function involves coaching and setting high-quality standards of work, mentors can impress upon their protégés the importance of ethical behaviour. Mentors can illustrate how poor ethical behaviour will not be tolerated within the firm and how such behaviour would risk jeopardising the firm's reputation as a whole. Furthermore, mentors in their social support role could foster trust and encourage frank and full discussions of issues that an ECA might feel sensitive or unclear on. For example, receiving gifts from clients, close personal relationships with clients, and a lack of clarity in client billing matters are some of the issues that potentially raise ethical conflict, and a new recruit is likely to benefit from guidance by his or her mentor.
Based on the discussion above, the first set of hypotheses for this study is as follows:
H1: The perceived mentoring support received by an early career accountant is positively related to their ethical evaluations and behavioural intentions.
Ethical evaluations are perceptions of:
(a) the seriousness of an unethical situation; and,
(b) ethical behaviour of their senior colleagues.
Ethical behavioural intentions relate to the:
(c) likelihood of calling the professional accounting bodies for advice; and,
(d) choosing a more ethical decision.
3.2. Peer influence: perceptions of peers’ ethical conduct
Jones and Kavanagh (1996) highlight two ways in which peers might influence unethical behaviour of their colleagues: through norms and through differential association. First, peers are seen to set the norms (i.e. standards of behaviour and rules on what is and what is not acceptable behaviour). According to Schein (1984), in an organisation where the organisational culture is not strongly driven by superiors, peers are likely to guide normative behaviour. Second, from a differential association theory of criminal behaviour (Sutherland, 1949, 1983), unethical behaviour is seen to be learned through association with a peer group whereby an unlawful action can be perceived as being acceptable when a group sanctions it. Empirical evidence from early studies by Hollinger and Clark (1983) and Horning (1970) reveal that employee deviant behaviour is often sanctioned or supported by a group, although the crime may be conducted by an individual. Jones and Kavanagh (1996), based on an experimental study involving 138 upper-level undergraduate students, found that peer influence was a significant factor affecting the intentions of respondents’ ethical behaviour. More recently, Westerman et al. (2007) based on a survey of 165 graduate business students from Germany, Italy and Japan found that peers exerted a much stronger influence on an individual's ethical decision-making than national culture. Similarly, based on a survey of 191 undergraduate advertising students, Keith et al. (2003) found that peer ethical behaviour has a strong effect on the ethical behavioural intentions of the respondents.
We predict peer influence to have a significant effect on the perceptions and behaviour of accountants who are early in their career in public accounting firms. This is because new accounting graduates generally go through a period of intensive professional training and tend to undertake professional qualification studies in the first few years of their career. Subsequently, they often come to rely and support each other at work, and such close interactions in turn is likely to result in the newly recruited accountants coming to a shared understanding of what are acceptable and what are not acceptable ethical behaviours.
Based on the discussion above, the second set of hypotheses for this study is as follows:
H2: The perception of the standard of ethical conduct of peers of an early career accountant is positively related to their ethical evaluations and behavioural intentions.
Ethical evaluations are perceptions of:
(a) the seriousness of an unethical situation; and,
(b) ethical behaviour of their senior colleagues.
Ethical behavioural intentions relate to the:
(c) likelihood of calling the professional accounting bodies for advice; and,
(d) choosing a more ethical decision.
3.3. Individual attribute: ethical orientation
Ethical orientation refers to an individual attribute that reflects one's personal values in relation to ethical and unethical acts. Individuals develop values from an early age. These individual values or factors establish a context in which decision-makers choose to act. Bommer et al. (1987) developed a model of ethical and unethical behaviour at the individual level with several attributes, such as individual moral level, self-concept and demographics, as predictor variables. An individual's moral level, or orientation, is contended to have a large degree of influence on an individual's ethical or unethical behaviour. Empirical evidence based on a study of 41 owners/managers of small businesses by Quiin (1997) indicates a significant link between respondents’ personal ethical values and their attitudes to ethical problems in business. Respondents who were fairly active members of religious groups expressed an overall higher concern for ethical business issues than respondents involved in at least one business-related organisation and those involved in at least one wider community activity group, suggesting that business issues are impacted by personal ethical orientation.
Furthermore, an individual's values have not only been found to play a key role in affecting one's attitude to ethical problems but also in his or her resolution of ethical dilemmas (Glover et al., 1997). Using experimental research methods, Glover et al. (1997) studied 367 university students and observed that individuals who valued achievement made a more ethical decision when the moral intensity of the decision was not extreme. Similarly, Al-Kazemi and Zajac (1999) found that employees in public organisations in Kuwait who had a high ethical orientation (sensitivity) tended to identify a greater number of unethical behaviours. On the basis of Ruch and Newstrom's (1975) research instrument, Al-Kazemi and Zajac (1999) explored 14 interpersonal ethics and illegal and organisational ethics behaviours of 170 Kuwaiti employees. These behaviours measured respondents’ ethical orientation or their personal moral ‘roadmap’ (Schott, 1991). It was found that respondents had a high ethical orientation, with a minimum of 62 per cent of respondents identifying each of the 14 behaviours as being unethical. Nevertheless, much of the extant evidence is based on managers’ ethical orientation, and no study to date has investigated the association between individual ethical orientations of public accountants and their ethical evaluations and behaviour. Based on the above discussion, it is therefore hypothesised that:
H3: The level of an early career accountant's personal ethical orientation is positively related to their ethical evaluations and behavioural intentions.
Ethical evaluations are perceptions of:
(a) the seriousness of an unethical situation; and,
(b) ethical behaviour of their senior colleagues.
Ethical behavioural intentions relate to the:
(c) likelihood of calling the professional accounting bodies for advice; and,
(d) choosing a more ethical decision.
3.4. Individual attribute: extent of ethics education
Accounting faculties at universities have started to not only integrate ethics in individual courses or subjects, but also advocate separate standalone units. Wu (2003) conducted a survey of 126 business ethics students in two universities in Taiwan and mainland China and concluded that, after receiving an education in business ethics, students from both universities showed significant improvements in the ethical weighting of their individual values, their recognition of ethical issues and their performance in ethical decision-making. Thoma (1986) and McNeel (1994) also found that a college education positively influences an individual's level of moral reasoning. In comparison, Zeigenfuss (1999), based on a survey of 84 university students, reported no significant differences between education level (sophomore, junior, senior and graduate) and personal ethical philosophy, suggesting that undergraduate programs that integrate ethics throughout a curriculum have little or no effect on a student's personal ethical philosophy.
Based on the discussion above, the fourth set of hypotheses is as follows:
H4: The extent of an early career accountant's ethics education at university is positively related to their ethical evaluations and behavioural intentions.
Ethical evaluations are perceptions of:
(a) the seriousness of an unethical situation; and,
(b) ethical behaviour of their senior colleagues.
Ethical behavioural intentions relate to the:
(c) likelihood of calling the professional accounting bodies for advice; and,
(d) choosing a more ethical decision.
4. Research method
4.1. Sampling procedures
Data were collected via a mail questionnaire survey conducted in late 2007. First, 30 accounting firms from each of the seven Australian states and territories were randomly selected from the ‘white pages’ and ‘yellow pages’ telephone websites, leading to a sample of 210 firms in total. A mail-out package was sent to each firm, which included a covering letter to the firm's partner or manager and three separate sealed envelopes – each addressed to an ECA.3 The covering letter to the partner or manager provided information about the project and requested that the sealed envelopes be forwarded to three randomly selected new accounting recruits from their firm. In each of the separate sealed envelope, a covering letter to the ECA, a questionnaire and a return self-addressed envelope were included. Eight mail-out packages were returned unopened due to incorrect address; and one facsimile was received advising that the firm did not have any ECAs employed. Hence, the final adjusted sample population involved 603 ECAs from 201 accounting firms. A number of random follow-up calls were made about 3 weeks after the initial mail-out to check if the mail-out packages were received, and if the partner or manager was willing to distribute the sealed envelopes with the questionnaire to the ECAs, and all replies were in the affirmative.
A total of 86 useable responses were received. This equates to a response rate of 14.3 per cent. Non-response bias was investigated by applying the Mann–Whitney U statistic for differences in responses provided by early and late respondents. The first and last 25 per cent of questionnaires returned were analysed and no statistically significant differences were found. On hindsight, a possible reason for the relatively low response rate might relate to the time-constraints face by new accounting recruits who are not only learning a new job, but are also likely to be enrolled in professional qualification and development programmes such as in the chartered accounting or the certified practising accountant (CPA) programmes. In fact, 92 per cent of respondents of our study report as being enrolled in a chartered accounting or CPA programme.
4.2. Survey questionnaire design
The survey questionnaire collected information on issues relating to mentoring styles, perceived ethical behaviour, ethical behavioural intentions, and individual attributes. A pilot study of the questionnaire was undertaken involving five academics who all had public accounting experience as well as four newly recruited accountants. Based on their feedback, some minor changes to wordings of the questionnaire were subsequently undertaken.
The first part of the questionnaire provided a hypothetical case scenario (adapted from the study of Brennan and Kelly, 2007), whereby an ECA discovers improper accounting treatment of research and development expenditure. This treatment is said to have significant implications to the financial statement and when the matter is brought to the attention of the audit partner, no action is undertaken with the suggestion that the client (who is also a large and important client) would prefer the matter to be treated as it is. (Please see Appendix A for the full case description.) The survey respondents were then asked to respond to various questions relating to the scenario on a seven point Likert-type scale, including their perception on the seriousness of the ethical issue involved (SERISS), the likelihood that they would make the same decision (EBI) (Patel, 2003), how ethical would they rate the Audit Partner's response (APR) to be, the likelihood they would call an accounting professional body for advice (ACCPRO) and the likelihood that they will undertake another course of action to that undertaken by the ECA in the hypothetical case.
Respondents’ ethical orientation (ETHOR) was measured using an abridged version of Al-Kazemi and Zajac's (1999) 14 item instrument, which was adapted from a longer list of ethical behaviours originally developed by Ruch and Newstrom (1975). Respondents were asked to indicate, on a seven point Likert-type scale, how ethical or unethical they considered six different behaviours. The six behaviours appraised were divulging confidential information, accepting gifts in exchange for preferential treatment, taking longer than necessary to do a job, concealing one's errors, conducting personal business in the firm's time, and misappropriating firm resources for personal use. As each of these behaviours is considered unethical, each of the items were reversed scored for data analysis. Therefore, the higher the score, the more ethical the response was considered to be. Cronbach alpha for the six items was 0.78, which suggests acceptable scale reliability. Responses to the six items were summed and the weighted average calculated to form the measure of ETHOR.
The second section of the survey questionnaire appraised ECAs’ formal and informal mentoring relationships. Respondents were asked how many formal mentors as well as informal mentors – both internal and external – they had, the gender of their mentors, and how often they had meetings with them (i.e. weekly, monthly, quarterly, twice a year or once a year). In addition, respondents were asked how often they had formal set meetings and informal ad hoc meetings with their mentors and what position each mentor holds. If a respondent did not have a mentor, they were asked to describe the reasons why.
Mentorship style was measured using the instrument developed by Scandura and Viator (1994) and also applied by Barker et al. (1999) and Herbohn (2004). Respondents were asked to rate their relationship with their formal mentor on a seven point Likert-type scale ranging from ‘1’ (strongly disagree) to ‘7’ (strongly agree), for 15 issues including ‘my mentor takes a personal interest in my career’, ‘I share personal problems with my mentor’ and ‘my mentors helps me coordinate professional goals’. A principal components analysis of the 15 items yielded two factors with eigenvalues greater than 1 (5.947 and 4.831, respectively). The final two factor solution is reported in Table 1. Nine items loaded strongly on factor 1, the career development mentoring factor (CARDEV). In the studies of Scandura and Viator (1994) and Herbohn (2004), items 8 and 9 (Herbohn) as well as items 11 and 14 (Scandura and Viator) loaded on a third factor called role modelling. In this study, role modelling is seen as a component of the career development function of formal mentors with a total of 39.6 per cent of the variance in mentoring explained by this factor. The six items that loaded strongly on factor 2 – share personal problems, coordinate professional goals, socialise after work, exchange confidences, consider a friend and go to lunch – were the same as the combined factor items from the studies of Scandura and Viator (1994) and Herbohn (2004) that loaded on the factor called social support mentoring function (SOCSUPP).4
| Mentoring item | Factor 1: Career development a | Factor 2: Social support |
|---|---|---|
| 1. My mentor takes a personal interest in my career | 0.80 | 0.25 |
| 2. My mentor has placed me in important assignments | 0.88 | 0.15 |
| 3. My mentor gives me special coaching on the job | 0.85 | 0.29 |
| 4. My mentor advises me about promotional opportunities | 0.64 | 0.39 |
| 5. I share personal problems with my mentor | 0.24 | 0.82 |
| 6. My mentor helps me coordinate professional goals | 0.46 | 0.63 |
| 7. I socialise with my mentor after work | 0.34 | 0.76 |
| 8. I try to model my behaviour after my mentor | 0.70 | 0.52 |
| 9. I admire my mentor's ability to motivate others | 0.72 | 0.50 |
| 10. I exchange confidences with my mentor | 0.48 | 0.73 |
| 11. I respect my mentor's knowledge of the accounting profession | 0.73 | 0.38 |
| 12. I consider my mentor to be a friend | 0.52 | 0.72 |
| 13. I respect my mentor's ability to teach others | 0.62 | 0.52 |
| 14. My mentor has devoted special time and consideration to my career | 0.74 | 0.44 |
| 15. I often go to lunch with my mentor | 0.16 | 0.83 |
| Percentage of total variance in mentoring explained by factor | 39.65 | 32.21 |
| Cronbach alpha | 0.945 | 0.912 |
- a A factor loading cut-off of 0.50 was used as the criterion for an item to be considered significant to a factor.
A number of ethics themes were examined in the following section of the questionnaire. Respondents were asked to respond on a seven point Likert-type scale ranging from ‘1’ (never and poor standard) to ‘7’ (often and high standard). The ethics themes section incorporated the questions ‘how often have you encountered ethical conflicts in your workplace’, ‘how often have you encountered ethical issues in relation to your work with clients, billing hours to clients and behaviour of colleagues’, and ‘how often does your mentor provide advice, career guidance and information to help you deal with ethical conflicts’. Respondents were then asked to rate, from a poor to high standard, the ethical behaviour of their peers in the workplace (PEERS).
The final section of the questionnaire gathered background information about the ECA. Respondents were asked their age, gender (GENDER), professional programme enrolled in, proportion of professional programme completed, whether their firm was a Big Four firm, middle-tier firm or a small firm with less than 10 accountants (SIZE), their position title, length of time working as an accountant, length of time working with the current firm, graduating university, year graduated, what extent was ethics taught at university (EXTETH) and whether ethics was taught as a distinct course or integrated into accounting courses.
4.3. Control variables
Both gender and firm size were included as control variables for the following reasons. Although many studies have found that women are more ethical than men, some also have found no significant difference between genders. For example, Miceli and Near (1988) found that women were less likely to be whistleblowers than men and Thorley et al. (1998) found that female accountants have higher level of moral reasoning ability. While Brennan and Kelly (2007) found no significant differences in regards to gender in the willingness of trainee auditor's to report wrongdoing externally. Furthermore, O’Leary and Cotter (2000) found that males were between two and four times more likely than females to act unethically and Haswell and Jubb (1995) noted that almost 50 per cent of male and 25 per cent of female students would accept a bribe if there was no risk of being caught. In a similar vein, Cohen et al. (1998) examined the effect of gender on ethical evaluations, ethical intentions and ethical orientation of potential public accounting recruits and found that women consider questionable actions to be more unethical and are less likely to perform these actions than men. In addition, Adkins and Radtke (2004) find that females find accounting education concepts and goals more important than males. In contrast, Jones and Kavanagh (1996), in their study of the effects of individual and situation factors on unethical behavioural intentions in the workplace, found no gender differences.
Firm size, conceptualised as Big Four firm versus non-Big Four firm, was also included as a control variable. Previous studies have shown inconsistent results in relation to the impact of firm size on ethical behavioural intentions. For example, Thorley et al. (1998) found that accounting practitioners from small firms generally have the same level of moral reasoning as Big Six practitioners and Brennan and Kelly (2007) found no significant differences based on firm size in regard to the willingness of auditor's to whistle-blow. In comparison, Sweeney and Boyle (2005) found significant differences between Big Four and non-Big Four respondents in regard to supervisory actions, job satisfaction and intentions to remain in the firm for a sample of 166 Irish trainee chartered accountants. Patten (1995) found significant differences between small firms and Big Six firms in the USA relative to working conditions and job satisfaction, but not for leadership and mentoring. While Guntz et al. (2002) reported that the larger the accounting firm, the more likely that a respondent would seek help from their colleagues, while no differences were found between accountants from large versus small firms relying on accounting professional bodies for advice.
4.4. Characteristics of sample and descriptive statistics
The characteristics of the sample include having a comparable proportion of males and females (i.e. 56 and 44 per cent, respectively), the average age being 23 years and 3 months and 76 per cent of respondents being under 25 years of age. More respondents were enrolled in the chartered accounting programme than the CPA programme (65 vs. 27 per cent) and 70 per cent of respondents having completed 20 per cent or less of their programme. Almost 60 per cent of respondents were employed in a middle-tier firm, and 31 per cent from a Big Four firm and 9 per cent from a small firm with less than 10 accountants. Just over 38 per cent of respondents had less than 1 year working experience as an accountant, 31 per cent had between 1 and 2 years, and another 31 per cent for more than 2 years experience. Approximately 80 per cent of respondents had been working for their current firm for less than 2 years. Most respondents (i.e. over 66 per cent) graduated from university in 2006.
In terms of mentoring relationships, most respondents (68 per cent) had either one or two formal mentors in their firm, 67 per cent had one or more informal internal mentors and 26 per cent had one or more informal mentors’ external to the firm.5 Twenty-seven per cent of respondents did not have a formal mentor, and all respondents who did not have a formal or informal mentor were employed at middle-tier and small accounting firms. Six respondents did not have a formal or informal mentor, and the main reasons cited for the lack of a mentor were ‘not sure’, ‘not really required’, ‘I am knowledgeable enough to do my job without any problems’, and ‘no mentoring programme at firm’.6 An overview of the sample profile in relation to meeting frequencies and formality of meetings between the mentors and mentees is provided in Table 2. In general, most respondents meet with their formal and informal mentors every week, with 40.7 per cent of respondents ‘sometimes’ having set meetings with formal mentors, 12 per cent never having set meetings with their formal mentors and 30.2 per cent ‘often’ having informal ad hoc meetings with their formal mentors.
| Type of mentor | Formal internal Number (%) | Informal internal Number (%) | Informal external Number (%) |
|---|---|---|---|
| Number of ECAs that meet with formal mentors | |||
| Weekly | 32 | 39 | 12 |
| Monthly | 16 | 7 | 9 |
| Quarterly | 10 | 3 | 2 |
| Twice a year | 8 | 0 | 3 |
| Once a year | 0 | 0 | 2 |
| Formal set meetings with mentors | |||
| Never | 12 (14.0) | 36 (41.9) | 18 (20.9) |
| Sometimes | 35 (40.7) | 13 (15.1) | 3 (3.5) |
| Often | 17 (19.8) | 4 (4.7) | 2 (2.3) |
| No answer | 22 (25.6) | 33 (38.4) | 63 (73.3) |
| Total | 100 (100) | 100 (100) | 100 (100) |
| Informal ad hoc meetings with mentors | |||
| Never | 12 (14.0) | 8 (9.3) | 4 (4.7) |
| Sometimes | 26 (30.2) | 20 (23.3) | 6 (7.0) |
| Often | 26 (30.2) | 27 (31.4) | 12 (12.0) |
| No answer | 22 (25.6) | 31 (36.0) | 64 (74.4) |
| Total | 100 (100) | 100 (100) | 100 (100) |
The descriptive statistics of the four dependent variables (i.e. ECA responses to the ethical scenario) is presented in Table 3. For the perceived seriousness of the given scenario (SERISS), most respondents viewed the matter to be fairly serious with 100 per cent of responses above the midpoint of the seven point scale (mean = 5.86). With respect to respondents’ response to the behaviour of the senior colleague (audit partner's response in the given case scenario (APR)), most of them considered the audit partner's behaviour to be unethical (i.e. mean = 1.94, where 1 = highly unethical and 7 = highly ethical).
| Mean | Minimum | Maximum | SD | SERISS | APR | ACCPRO | EBI | CARDEV | SOCSUPP | PEERS | ETHOR | EXTETH | SIZE | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SERISS | 5.85 | 4 | 7 | 0.94 | 1.00 | |||||||||
| APR | 1.96 | 1 | 5 | 0.95 | 0.63*** | 1.00 | ||||||||
| ACCPRO | 3.38 | 1 | 7 | 1.72 | 0.28** | –0.13 | 1.00 | |||||||
| EBI | 4.17 | 1 | 7 | 1.73 | 0.47*** | 0.46*** | 0.20 | 1.00 | ||||||
| CARDEV | 5.06 | 1.7 | 6.9 | 1.18 | –0.8 | 0.13 | 0.07 | –0.27** | 1.00 | |||||
| SOCSUPP | 4.49 | 1 | 6.7 | 1.42 | –0.17 | –0.36*** | –0.18 | –0.25** | 0.78*** | 1.00 | ||||
| PEERS | 5.87 | 3 | 7 | 0.89 | 0.12 | 0.08 | 0.01 | 0.15 | 0.52*** | 0.42*** | 1.00 | |||
| ETHOR | 2.53 | 1.7 | 5.7 | 0.77 | 0.25** | 0.07 | –0.06 | 0.23** | 0.14 | 0.11 | 0.02 | 1.00 | ||
| EXTETH | 4.37 | 1 | 7 | 1.64 | 0.33*** | 0.04 | 0.07 | 0.45** | –0.01 | 0.11 | –0.06 | 0.24** | 1.00 | |
| SIZE | 0.34 | 0 (53)a | 1 (27)a | 0.48 | 0.17 | 0.08 | –0.27** | 0.20 | –0.11 | –0.06 | –0.05 | 0.04 | 0.25** | 1.00 |
| GENDER | 0.46 | 0 (45)b | 1 (35)b | 0.55 | 0.36*** | 0.36*** | 0.17 | 0.22 | –0.19 | –0.33*** | –0.22 | 0.14 | 0.24** | 0.07 |
- *** and ** indicate that correlation is significant at the 0.01 and 0.05 levels (two-tailed), respectively.
- a Data coded as 0 = non-Big Four firm and 1 = Big Four firm (frequency).
- b Data coded as 0 = male and 1 = female (frequency). SERISS is seriousness of the issue; APR is audit partner's response; ACCPRO is call an accounting professional body for advice; EBI is ethical behavioural intentions; CARDEV is career developing mentoring style; SOCSUPP is social support mentoring style; PEERS is ethical behaviour of peers; ETHOR is ethical orientation of early career accountant; EXTETH is extent of university ethics education.
The average responses for the questions ‘would you call an accounting professional body for advice (ACCPRO)’ and ‘would you make that same decision (EBI)’ were generally around the midpoint of the seven point scale (means = 3.27 and 4.12, respectively, where 1 = highly improbable, 7 = highly probable). Finally, for the question on ‘would you take another course of action’, the mean score was 4.23 with 18 respondents stating that an alternate course of action for them would be ‘to discuss with another partner’, six respondents stating ‘discuss with colleagues’ and another six respondents stating ‘to resign from the firm. Two respondents stated contacting the Australian Securities Commission and one to use the firm's hotline.
5. Results
Table 3 also provides descriptive statistics of the other variables of the study and a matrix of the Spearman's rank correlation coefficients for the dependent, independent and control variables. Although a number of the correlations between the independent variables were significant (p < 0.01), multicollinearity does not represent a significant threat to the regression analyses conducted, as the VIF values presented in Table 4 are below the generally accepted critical threshold value of 10 (Hair et al., 2006).
| Expected sign | Hypothesis | SERISS (a) | APR (b) | ACCPRO (c) | EBI (d) | VIF | |
|---|---|---|---|---|---|---|---|
| Constant | 4.84*** | 2.78*** | 2.57** | 2.89*** | |||
| (5.53) | (3.01) | (1.40) | (1.76) | ||||
| CARDEV | + | 1 | –0.04 | 0.26* | 0.12 | –0.17 | 3.27 |
| (–0.23) | (1.32) | (0.57) | (–0.89) | ||||
| SOCSUPP | + | 1 | –0.25b** | –0.63b** | –0.35b** | –0.17 | 3.20 |
| (–1.31) | (–3.21) | (–1.71) | (–0.91) | ||||
| PEERS | + | 2 | 0.29*** | 0.21** | 0.10 | 0.05 | 1.43 |
| (2.38) | (1.65) | (0.76) | (0.38) | ||||
| ETHOR | + | 3 | 0.22** | 0.04 | –0.02 | 0.18** | 1.17 |
| (1.93) | (0.36) | (–0.15) | (1.61) | ||||
| EXTETH | + | 4 | 0.20** | 0.08 | 0.21* | 0.35*** | 1.44 |
| (1.67) | (0.65) | (1.58) | (2.84) | ||||
| Control variables | |||||||
| SIZE | 0.09 | 0.09 | –0.35*** | 0.05 | 1.07 | ||
| (0.86) | (0.86) | (–3.03) | (0.47) | ||||
| GENDER | 0.25** | 0.15 | 0.05 | 0.05 | 1.44 | ||
| (2.07) | (1.20) | (0.40) | (0.39) | ||||
| Adjusted R2 | 0.25 | 0.19 | 0.11 | 0.22 | |||
| F | 4.32 | 3.37 | 2.15 | 3.887 | |||
| p-value | 0.00 | 0.00 | 0.04 | 0.00 | |||
- * p < 0.10;
- ** p < 0.05;
- *** p < 0.01 (n = 71).
- a Each cell presents the standardised regression coefficient followed by the t-value in parentheses. All t-tests are one-tailed tests of significance in hypothesised direction for independent variables: CARDEV, SOCSUPP, PEERS, ETHOR and EXTETH. t-tests are two-tailed tests of significance for control variables: SIZE and GENDER.
- b Regression coefficient is significantly negative (i.e. p < 0.05) in the opposite direction as that hypothesised. SERISS is seriousness of the issue; APR is perceptions of audit partner's response; ACCPRO is call an accounting professional body for advice; EBI is ethical behavioural intentions; CARDEV is career developing mentoring style; SOCSUPP is social support mentoring style; PEERS is ethical behaviour of peers; ETHOR is ethical orientation of early career accountant; EXTETH is extent of university ethics education; SIZE is firm size (coded: 0 = non-Big Four firm and 1 = Big Four firm); GENDER is gender (coded: 0 = male and 1 = female).
Prior to performing the regression analyses, the data were screened for the existence of any influential observations through the examination of residuals, calculation of leverage points, Mahalanobis distance and Cook's distance. No influential observations were identified. Each set of hypotheses was tested by separately fitting each of the four respective dependent variables (Yi-iv) to the following equation that includes the two control variables (gender and firm size):
where:
Y(i-iv) = ECAs’ (i) evaluation of seriousness of the ethical issue (SERISS); (ii) evaluation of the ethical behaviour of a senior colleague (APR); (iii) likelihood of calling the accounting professional body for advice (ACCPRO); and (iv) intention to behave ethically (EBI)
CARDEV = career development mentoring style
SOCSUPP = social support mentoring style
PEERS = perceptions of ethical behaviour of peers
ETHOR = ethical orientation of ECA
EXTETH = extent of ECAs’ ethics education
SIZE = firm size
GENDER = gender
Results of the regression analyses are presented in Table 4. The four regression equations were statistically significant (p < 0.05 for all equations), with adjusted R2's ranging from 11 to 25 per cent.
Hypotheses 1(a–d) predicted a positive relationship between ECAs’ perceptions of mentoring support (i.e. career development mentoring style or social support mentoring style) and their evaluation of the seriousness of an unethical situation, the ethical behaviour of their senior colleagues and their ethical behavioural intentions in terms of the likelihood they would call an accounting professional body for advice and the likelihood they would make a more ethical decision. There was weak support shown for Hypothesis 1(b) with career development mentoring style exhibiting a moderately significant and positive relationship with the ethical behaviour of their senior colleague in the hypothesised direction (p < 0.10). No support was shown for Hypotheses 1(a), 1(c) or 1(d). Interestingly, social support mentoring style was significantly negatively related to the evaluation of the seriousness of the unethical situation (p < 0.05), the ethical behaviour of their senior colleague (p < 0.01) and the likelihood they would call an accounting professional body for advice (p > 0.05). This suggests that the greater the social support mentoring style perceived by a new accounting recruit, the less serious they evaluate an unethical situation, the less unethical they view the behaviour of their senior colleague and the less likely they are to call an accounting professional body for ethical advice.
Hypotheses 2(a–d) postulated a significant positive relationship between ECAs’ perceptions of the ethical conduct of their peers and their evaluation of the seriousness of an unethical situation and the ethical behaviour of their senior colleagues as well as their behavioural intentions in terms of the likelihood they would call an accounting professional body for advice and the likelihood they would make a more ethical decision. Strong support was shown for Hypotheses 2(a) and 2(b), with the perceptions of the ethical conduct of peers significantly positively related to the evaluation of the seriousness of the unethical scenario (p < 0.01) and the respondents’ evaluation of the ethical behaviour of their senior colleagues (p < 0.05). No support was shown for Hypothesis 2(c) or 2(d).
The third set of hypotheses put forth a significant positive relationship between ECAs’ ethical orientation and their evaluation of the seriousness of an unethical situation, the ethical behaviour of their senior colleagues and their behavioural intentions in terms of the likelihood they would call an accounting professional body for advice and the likelihood they would make a more ethical decision. Individual ethical orientation was significantly positively related to the perceptions of the serious of an unethical situation (p < 0.05) and a participant's behavioural intention in terms of choosing a more ethical decision (p < 0.05). Therefore, support was shown for Hypotheses 3(a) and 3(d).
Support was also shown for Hypotheses 4(a), 4(c) and 4(d) with the extent of ethics education at university significantly positively related to the evaluation of the seriousness of an unethical situation (p < 0.05), and the ECAs’ behavioural intentions in terms of the likelihood contacting an accounting professional body for advice (p < 0.10) and that a more ethical decision be chosen (p < 0.01).
In relation to the control variables, first, firm size was found to be negatively related to the likelihood of a new accounting recruit calling an accounting professional body for advice (p < 0.01), indicating that respondents from small and middle tier firms are more likely to call an accounting professional body for advice than those from Big Four firms. Second, gender was found to be positively related to respondents’ perceptions of the seriousness of an unethical situation (p < 0.05). That is, female respondents perceive an unethical scenario as being more serious than males.
6. Discussion of results
This study provides empirical evidence on the ethical evaluations and behavioural intentions of ECAs in Australian public accounting firms. The results of this study indicate that peers have a strong effect on the ethical evaluations of accountants who are still early in their career, particularly in terms of their perceptions of the seriousness of a hypothetical ethical conflict and also the extent to which they judge an unethical act by their superiors to be unethical. As accountants starting their career perceive a higher standard of peer ethical conduct in a firm, it is more likely that they will view an ethical conflict to be serious and the inappropriate actions of their superiors to be unethical. This findings support prior studies on the influence of peers on an individual's judgements (Jones and Kavanagh, 1996). However, the impact of peers on the ECAs’ behavioural intentions has not been significant. In addition, the findings of this study also identify ethics education at university to be a significant factor affecting perceptions of the seriousness of an unethical issue as well as the behavioural intentions of accountants early in their career in terms of calling the accounting professional body for advice, and choosing a more ethical action. Clearly, these results indicate that university curricula might need to pay more attention to the quality and extent of ethics content in the accounting programmes. Furthermore, these results also suggest that the professional bodies have an important role in affecting the behaviour of ECAs through providing better ethical conduct guidance. This highlights an increasing need for greater involvement and commitment by professional bodies to ensure adequate support is available for those embarking in an accounting career at both the university level and in-house such as through their website, and more importantly access to advisors/counsellors outside the firm. Furthermore, the results have implications for the design of professional accounting training programmes in terms of increasing their ethical content.
An individual's ethical orientation attribute is also another factor that appears to affect ECAs’ perceptions of the seriousness of an ethical conflict, and to a lesser extent, the choice of ethical behaviour. The greater the individual propensity or personal values towards ethical behaviour, the higher the probability that in a work situation they will likewise perceive and act ethically. These results support prior studies by Al-Kazemi and Zajac (1999) and Quiin (1997), who found evidence of a high level of personal ethical sensitivity with an increased likelihood of ethical attitudes towards ethically sensitive business issues.
Interestingly, the results relating to the impact of mentoring style on ECAs were mixed. While a career-development mentoring style appears to support such accountants’ ethical evaluation of a senior colleague, no such impact was found on their ethical behaviour orientations. Surprisingly, the social support mentoring style seems to have a negative impact on ECA's ethical evaluations and behaviour. It is possible that when accountants early in their career interact more socially with their mentors, they might develop a more emotionally based relationship and, hence, would have difficulty viewing a mentor behaving unethically, and also would consider it unnecessary to consult an external body such as an accounting professional body. Nevertheless, further study on the social bond developed by the mentor–mentee dyad, particularly from a psychosocial effect, is clearly warranted. In particular, given that prior studies (Scandura and Viator, 1994; Herbohn, 2004) indicate that national culture might affect the mentoring relationship, extending the current study to other cultural settings would be useful for comparative purposes. Another explanation for the results of the present study may relate to the sample respondents’ limited extent of interactions with their mentors. Given that most of the sample respondents (about 70 per cent) had been employed at the firm for only about a year, and almost 15 per cent report as having not met their formal mentor at all, it is also possible that the effect of mentorship has been minimal.
7. Conclusions and limitations of study
This study provides evidence on an interesting array of factors affecting the ethical orientations and intended behaviours of ECAs, and, in particular, highlights the importance of mentoring, peer support and tertiary ethics education as well as personal ethical orientation as vital factors influencing their ethical development and stance at the workplace. However, these results need to be interpreted with caution. First, the questionnaire distribution to ECAs was reliant on the principal of the practice. As such, the sample might be biased towards respondents whom the principal might already have a strong positive view and a good mentoring relationship. Second, the use of a single hypothetical scenario limits the validity of the findings. While the ethical conflict revolved around the misclassification of R&D expenditure, this might not be seen as serious in comparison to other types of ethical misdemeanours, such as falsification of documents or lying to a client. Third, the response rate was fairly low, about 14 per cent, and as such the generalisability of the data is restricted. Fourth, we also recognise that the use of a hypothetical ethical situation might entail positive response bias due to social desirability reasons where arguably respondents tend to over-report or over-state activities that are deemed to be socially desirable (Zerbe and Paulhus, 1987). For example, respondents of this study might have rated the seriousness of the ethical scenario as being more severe or reported the probability of their undertaking an ethical decision higher than their actual perceptions or their willingness to act in an ethical manner. Nevertheless, since we had used the wording of the scenario in the third person (i.e. not in direct reference to the respondent), the effects of social desirability bias is likely to be minimal (Lord and Melvin, 1997). Finally, the usual caveats of a survey research are acknowledged.
Nevertheless, gaining a better understanding of the ethical attitudes of accountants early in their career is vital for a vibrant and forward-looking profession. ECAs are in a delicate and critical juncture in their accounting careers and as such, future work in this area by studying the impact of additional sources of influence both in and out of the workplace (e.g. organisation-specific training programmes), and the quality of ethics education at the professional and tertiary levels is clearly timely and critical.
Footnotes
References
Appendix I
Early career accountant mentoring and ethics survey
Section 1: Ethical Scenario
ZX Partners is the audit firm of Collins Ltd, a partly-owned subsidiary listed on the Australian Stock Exchange, of FirstDrug (FD). FD is a very large US-based pharmaceutical company that has seven other subsidiaries globally. Collins Ltd has several manufacturing and packaging sites in Sydney and Melbourne. Jill has been appointed as the Audit Senior for Collins Ltd in this year's audit engagement. During the course of the audit work Jill becomes concerned about the company's treatment of R&D expenditure. In particular Jill has found that the company is incorrectly classifying some research expenditure as development expenditure and capitalising it on the balance sheet. This accounting treatment has a material and favourable impact on the financial statements of Collins Ltd. Jill passes on her concerns to the Audit Partner who subsequently invites her to his office for a chat. While noting Jill's concerns, he proposes that the amounts involved can also be construed not to be material from the overall group's (i.e. FD's) perspective and furthermore this is how the client wants the R&D expenditure to be treated. He also points out the scale and importance of Collins Ltd's operations in Australia and impresses upon her that this client is very important to the firm. Noting that Jill still has a year to do under the terms of her training contract, he voices his concern for her career prospects, and suggests that Jill ‘let sleeping dogs lie’. Jill decides to take no further action.
- a.
How would you rate the seriousness of the issue?
- b.
If you were responsible for making the decision, what is the probability that you would make the same decision as Jill?
- c.
How would you rate the Audit Partner's response?
- d.
If you were responsible for making the decision, what is the probability that you would:
- •
Call one of the accounting professional bodies for advice?
- •
Undertake another course of action/s?
- •
If yes, what course of action/s would that be?
