Determinants of Canadian Accounting Practitioners' Ethical Perceptions On Earnings Management Kui-Ying Lin B.B.A., Tamkang University, Taiwan, 1995 M.B.A., University of Houston, Victoria, USA, 1996 Project Submitted in Partial Fulfillment Of The Requirements for the Degree of Master of Business Administration The University of Northem British Columbia April2009 @ Kui-Ying Lin, 2009 UNIVERSITY OF NORTHERN BRITISH COLUMBIA LIBRARY Prince George, BC - ... ···:-............ .. ~ ~- · ·. MBA Project - Kui-Ying Lin ACKNOWLEDGMENTS Nwnerous people deserve acknowledgments and thanks for their contributions to this project. First, I would like to thank my supervisor, Dr. Deborah Poff, not only for her helpful directions, constructive advice, and full support throughout every phase of this project, but also for her insights, inspiration, and encouragement to help me overcome all difficulties in obtaining raw data and completing the research. She managed to take time from her busy schedule to review my project; her comments and guidance on writing were invaluable as the text was honed and shaped into this final draft. Her unfaltering commitment to my purpose was beyond my expectations. I am extremely grateful for the knowledge and experience I have learned working with her for the past year. I am deeply indebted to Dr. Han Donker who had a major influence on my development of accounting knowledge through his teaching. His suggestions and comments on the statistical analyses were essential in helping me interpret the study results and enhanced the quality of this project; his invaluable time and efforts are greatly appreciated. I would especially like to thank Mr. Matthew Haussmann, Ms. Jennifer Kerr, and Ms. Christine Smales of British Columbia Institute of Technology and Mr. Scott Sinclair and Dr. Dan Simunic ofUBC Sauder School of Business who provided me with an opportunity to get useful samples from their students. Without their support and help, the data would not have been available. A special thanks to all participants for their valuable time to fill in all questionnaires and share their ethical values on perceptions of earnings management. I wish to thank four pretesting participants: a Certified Management Accountant, Ms. Cheryl Wallace, a Chartered Accountant, Mr. Bo Gill, and two accounting students, Mr. 11 MBA Proje c t - Kui -Ying Lin Louis Yen, and Mr. Rod Yen for their insights and suggestions which enhanced the readability and meaningfulness of the survey instruments. Above all, my husband James and my sons Louis and Rod deserve my heartfelt thanks and appreciation for their indescribable support, encouragement, and assistance which enabled me with the courage and persistence to finish the race set out for me. I would like to give a special thanks to my dear friends Franca Petrucci and Cheryl Wallace who lent an ear when needed and provided timely encouragement and support through this adventure. Finally, I would like to extend thanks to my professors at the University of Northern British Columbia (UNBC) for contributing to my academic development. Thanks for the wonderful memories of working on my MBA at UNBC for the past two years. Ill MBA Proj ect - Kui-Ying Lin ABSTRACT The collapses of Enron, World Com, and Arthur Andersen have raised a growing concern about earnings management (EM) ethics. These accounting scandals have damaged stakeholders' overall confidence and trust in the accounting profession. The academic community has responded with extensive research into American EM ethics but little is known about Canadian ethical values on perceptions of EM. This project focused on Canadian perspectives and surveyed 175 accounting students from two business schools in British Columbia to determine whether ethical ideologies and individuals' characteristics influence decision-making about EM ethics. Empirical results partially supported hypothesized direct effects for idealism, relativism, Machiavellianism, and locus of control. However, an intriguing finding is that EM ethics was also inversely related to Machiavellianism; this reverse relationship contradicts both hypothesized direction and previous research. This discovery may mitigate the sole negative image of Machiavellianism and render the reverse relationship between Machiavellianism and EM ethics possible and sustainable. lV MBA Project - Kui-Ying Lin TABLE OF CONTENTS ACKNOWLEDGMENTS ............................................................................................................................ II ABSTRACT ................................................................................................................................................ IV LIST OF FIGURES .................................................................................................................................. VII LIST 0 F TABLES ................................................................................................................................... VIII INTRODUCTION ....................................................................................................................................... 1 BACKGROUND .......................................................................................................................................... 2 EARNINGS MANAGEMENT BEHAVIOUR ..................................................................................................... 2 EARNINGS MANAGEMENT AND ETHICS ..................................................................................................... 5 THE NEED TO EXAMINE EARNINGS MANAGEMENT ETHICS ....................................................................... 7 HYPOTHESES DEVELOPMENT ........................................................................................................... 10 IDEALISM AND RELATIVISM .................................................................................................................... 11 MACHIAVELLIANISM ............................................................................................................................... 15 Locus OF CONTROL ................................................................................................................................ 18 RESEARCH METHODOLOGY ............................................................................................................... 23 MEASUREMENTS ..................................................................................................................................... 23 Ethics Position Questionnaire ................................................................................................... 23 Mach IV Scale ................................................................................................................................. 24 Internal-External Locus of Control Scale................................................................................ 25 Ethical Scenarios ........................................................................................................................... 26 PRETESTING ........................................................................................................................................... 28 PARTICIPANTS ........................................................................................................................................ 29 SCALE RELIABILITY ................................................................................................................................ 31 RESULTS ................................................................................................................................................... 34 EARNINGS MANAGEMENT ETHICS .......................................................................................................... 34 PERSONAL MORAL PHILOSOPHIES .......................................................................................................... 3 7 MACHIAVELLIANISM ............................................................................................................................... 40 Locus OF CoNTROL ................................................................................................................................ 43 DISCUSSION AND IMPLICATION ....................................................................................................... 47 v MBA Proj ect - Kui -Ying Lin EARNINGS MANAGEMENT ...................................................................................................................... 4 7 IDEALISM AND RELATIVISM .................................................................................................................... 49 MACHIAVELLIANISM ............................................................................................................................... 52 Locus oF CoNTROL ................................................................................................................................ 55 LIMITATIONS AND FUTURE RESEARCH DIRECTIONS ............................................................... 57 REFERENCES ........................................................................................................................................... 62 APPENDICES ........................................................................................................................................... 70 A. THE ETHICS POSITION QUESTIONNAIRE ........................................................................................... 70 B. THE LIKER TYPE MACH SCALE (IV) ................................................................................................. 72 C. THE INTERNAL-EXTERNAL LOCUS OF CONTROL SCALE .................................................................... 7 4 D. EARNINGS MANAGEMENT SCENARIOS QUESTIONNAIRE ..................................................................... 77 E. DEMOGRAPHIC INFORMATION QUESTIONNAIRE ................................................................................ 80 F. A COVER LETTER TO BCIT STUDENTS .............................................................................................. 81 G. A COVER LETTER TO UBC STUDENT! .............................................................................................. 82 H. A COVER LETTER TO UBC STUDENT 2 ............................................................................................ 83 I. INFORMED CONSENT FORMS ............................................................................................................... 84 VI MBA Project - Kui-Ying Lin LIST OF FIGURES FIGURE 1 REST'S FOUR-COMPONENT MODEL ........................................................................................... 88 FIGURE 2 TAXONOMY OF ETHICAL IDEOLOGIES ......................................................................................... 88 FIGURE 3 ETHICAL DECISION-MAKING MODEL ......................................................................................... 89 FIGURE 4 RESPONSE DISTRIBUTION: SCHOOL-B-INSTRUCTOR-2 ............................................................. 89 Vll MBA Proj ect - Kui-Ying Lin LIST OF TABLES TABLE 1 RESPONSE RATES ....................................................................................................•.....••.•........... 90 TABLE 2 DEMOGRAPHIC INFORMATION SUMMARY .•.....•...••...•......•.•..•.•.....••.............................................• 91 TABLE 3 RELIABILITY ESTIMATES REPORTED IN A SAMPLE OF PRIOR STUDIES ..........•...•....................... 92 TABLE 4 SCALE RELIABILITY: INDEPENDENT VARIABLES ................................................•..•..................... 93 TABLE 5 DISTRIBUTIONS OF DEPENDENT VARIABLES ............................................................................... 94 TABLE 6 SCORE RANGES: DEPENDENT vARIABLES ........................................... ········································ 95 TABLE 7 STUDY COMPARISON: EARNINGS MANAGEMENT .•....•.......................................................•......... 96 TABLE 8 FOUR CONTRASTING EARNINGS MANAGEMENT ATTRIBUTES .................................................... 97 TABLE 9 DISTRIBUTIONS OF IDEALISM AND RELATIVISM .......................................................................... 98 TABLE 10 SCORE RANGES: IDEALISM AND RELATIVISM ............................................................................ 98 TABLE 11 STUDY COMPARISON: IDEALISM AND RELATIVISM ...............................••.................................. 99 TABLE 12 CORRELATION ANALYSIS·························································· ················································· 99 TABLE 13 REGRESSION ANALYSIS: IDEALISM························································································· 100 TABLE 14 REGRESSION ANALYSIS: RELATIVISM .......................................................•..••...••.................... 101 TABLE 15 AN OVA RESULTS: FOUR ETHICAL IDEOLOGIES ......................................•.••..•....................... 102 TABLE 16 DISTRIBUTIONS OF MACHIAVELLIANISM ................................................................................ 103 TABLE 17 SCORE RANGES: MACHIAVELLIANISM···················································································· 103 TABLE 18 MACHIAVELLIAN SCORES REPORTED IN A SAMPLE OF STUDIES ............................................ 104 TABLE 19 ANOVA RESULTS: THREE MACHIAVELLIAN ATTRIBUTES .................................................... 105 TABLE 20 REGRESSION RESULTS: MACHIAVELLIANISM-1 .......................................•............................. 105 TABLE 21 REGRESSION RESULTS: MACHIAVELLIANISM-2 ..................................................................•.. 106 TABLE 22 ORIENTATION ANALYSIS: LOCUS OF CONTROL ..............................••.•..•.........................•.•.•••• 106 TABLE 23 STUDY COMPARISON: LOCUS OF CONTROL ..........................•.•......•..•........................•...•...•.... 107 TABLE 24 AN OVA RESULTS: ATTRIBUTES OF LOCUS OF CONTROL .......................................•.............. 108 TABLE 2 5 REGRESSION RESULTS: LOCUS OF CONTROL •...•......•..............................•............................... 108 Vlll MBA Proj ect - Kui -Ying Lin INTRODUCTION Earnings management (EM) practices have raised some of the most important ethical issues confronting accounting practitioners in the past few decades (Merchant and Rockness 1994). The collapse of corporate giants such as Enron, WorldCom, and Arthur Andersen has further led to a growing public and professional concern about EM ethics. These wellpublicized accounting scandals prompted the U. S. Congress to pass the Sarbanes-Oxley Act of 2002 to protect investors and enhance the accuracy and reliability of corporate financial reports. The academic community has responded with intensive research into EM ethics. However, because the research has focused only on American situations, little is known about Canadian EM ethics. The purpose of this study is to investigate the ethical perceptions of selected (EM) scenarios among accounting students at two business schools in British Columbia. The possible determinants (idealism, relativism, Machiavellianism, and locus of control) of ethical perceptions are examined. Determining the reasons for Canadian accounting students' perceptions of EM can shed some light on the accounting constituents and is the first step towards a better understanding of Canadian EM ethics. The project is organized as follows. This introduction is followed by a detailed background on EM behaviour, its ethics, and reasons for this project. Hypotheses are developed according to the ethics literature. An explanation of research methodology and results is reported. The project concludes with a discussion of the implications for the general fmdings and suggestions for future research. MBA Proj ect - Ku i -Ying Lin BACKGROUND Earnings management behaviour The accounting profession has historically been perceived to have a "reputation of solid integrity" and be highly respected by the public in tenns of preparing transparent, timely, and reliable financial statements (Fischer and Rosenzweig 1995, 434). However, this faith has been shaken by the sudden demise of high profile companies, such as Enron, WorldCom, and Arthur Andersen, due to unethical, even criminal, earnings management (EM) practices. These accounting scandals have damaged stakeholders' overall confidence and trust in the accounting profession. Levitt (1998) argues that the consequence of EM has been the erosion of financial reporting quality and of accounting professionals' integrity. Analyzing particular ethical influences on EM may be necessary to help accounting practitioners rebuild their reputation and integrity. Earnings management (EM) can be defined in several distinct ways. Scott (2006, 343) defines that, from a financial reporting and a contracting perspective, EM is "the choice by a manager of accounting policies so as to achieve some specific objective." Davidson et al. (2004, 267) assert that EM is "the use of flexible accounting principles that allow managers to influence reported earnings, thereby causing reported income to be larger or smaller than it would otherwise be," an effect that creates more or less favourable pictures of a company's financial performance. Fischer and Rosenzweig (1995, 434) define EM as "referring to actions of a manager which serve to increase (decrease) current reported earnings of the unit for which the manager is responsible without generating a corresponding increase (decrease) 2 MBA Project - Kui-Ying Lin in the long-term economic profitability of the unit", a definition that identifies two necessary components (intention and consequence) of EM behaviour. Merchant (1989, 168-69) divides EM into two types : accounting and operating manipulations. Accounting-based activities involve changing the reported amount of needed reserves such as impaired assets, obsolescent inventory, or bad debts, while operating-based activities involve changing operating decisions to push sales or expenses or to recognize gains or losses, in order to increase or decrease current reported earnings (Merchant 1989, 169-171 ). Elias (2002, 42) and Merchant and Rockness (1994, 89) have found that operatingbased practices were not considered unethical within the accounting profession, while in his empirical study Kaplan (200 1b, 294-95) noted that MBA students regarded accounting-based activities as unethical. In their survey of undergraduate students, MBA students, and accountants, Fischer and Rosenzweig (1995 , 436-39) found that accounting-based activities were less acceptable ethically than operating-based actions, while Elias (2002, 43) noted that all participants viewed accounting-based earnings management as ethical infractions. Dechow, Sloan, and Sweeney (1996) found that companies with weak corporate governance structures and/or with dual leadership, a CEO-chair, were more likely to engage in EM practices. Davidson et al. (2004) also assert that a CEO-chair tends to manage earnings to show better performance because the dual-position leader operates under greater expectations and gives him/her more authority to adjust financial reports. Healy and Wahlen (1999, 370-80) claim that the incentives for EM are to affect stock market perceptions, increase management's compensation, reduce the risk of violating lending agreements, and avoid regulatory intervention. Further, Elias (2002, 34) attributes individual EM decisions to both external and internal factors . External factors that most commonly cause management to 3 MBA Project - Kui-Yin g Lin manipulate earnings include the needs to meet financial analysts' forecasts and contractual obligations and to gain access to loan markets (Duncan 2001, 34). Su (2005, 149-153) found that EM has increasingly been used to meet analysts' earnings forecasts and that companies with more accounting flexibility in using discretionary accruals as an EM tool have been more likely to manipulate earnings. Dechow, Sloan, and Sweeney (1996, 30) found that raising external fmancing at low costs and avoiding debt covenant restrictions motivated managers to engage in EM activities and that the stock market responded negatively to allegations of misstatements. Internal factors, such as stock options, personal bonuses, promotions, or job retention, can also motivate managers to manage earnings to meet their budget targets (Duncan 2001, 35-37). Meek, Rao, and Skousen (2007) found that CEO stock option compensation was positively associated with discretionary accruals, implying that EM is more likely where stock options are a larger part of CEO compensation. Tying managers' bonuses and performances to profit targets can further encourage managers to manipulate earnings (Kaplan 2001 b, 287). Scott (2006, 346) posits that, from a contracting perspective, risk-averse managers may smooth reported earnings to receive a relatively constant bonus stream. Barton (200 I) asserts that the incentives of managerial compensation and wealth prompt managers to engage in EM and that in fact the CEO's primary job is to 'smooth' the company's earnings, although in their study, Bitner and Dolan (1998) found no real value in "unnaturally smoothed earnings." Clearly, these factors and pressures on ethically questionable EM may cause managers to act in ways that damage the professional integrity of the accounting profession. 4 MBA Project - Kui-Ying Lin Earnings management and ethics The accounting literature reveals disagreement (between legitimacy and fraudulence) over the ethical nature of EM. In testing EM intentions among MBA students, Kaplan (200la) found that when EM is intended to enhance company benefits, students viewed EM as less unethical but when the intention of EM is to benefit individuals, students considered EM unethical. Similarly, Dunmore (2008, 35) posits that when management's intention is to protect the company's interest, EM actions are considered ethical but when the intention is to benefit managers personally, EM practices are viewed as unethical. With the intention of benefiting the company, Jack Welch, the former CEO of General Electric (GE), created "artful accounting" to manage earnings and contributed to a stunning run of profit growth over 101 straight quarters (Kahn 2001) . The effectiveness of these EM practices helped Welch multiple GE's value from a market capitalization of$14 billion in 1980 to more than $400 billion in 1999 and win the title of manager ofthe 20 1h century (Colvin 1999). For the intention of benefiting individuals, on the other hand, equity compensation reliant on the realization of market expectations induced Frank Dunn, the former CEO ofNortel Networks, to play EM games dishonestly and put irreversible forces in motion, an effect that spiralled the borrowing from future revenues further and further in subsequent periods and ended with a total destruction of Norte! 's core value (Fogarty et al. , 2009, 168). The "cookie jar reserves" useable to normalize accounting performance resulted in the downfall of the company and the endless lawsuits against Dunn for financial improprieties relating to earnings manipulations (Fogarty et al., 2009, 167). Greenfield (2005, 1) claims that when "the intent of an accounting decision is to mislead or defraud stakeholders", EM is considered unethical. This is especially true in the case of Enron. The top management purposely underreported 5 MBA Project - Kui-Ying Lin Enron's debts and provided opaque disclosures about its financial conditions (Ketz 2002) and led stakeholders to lose billions and billions of dollars in the demise of the company. Dye ( 1988) notes that EM benefits current/selling shareholders at the expense of future/buying stockholders. According to Dye, current shareholders prefer a compensation contract which motivates managers to smooth current reported earnings and boost the share price in the short run, an effect that maximizes shareholders' proceeds in selling their stocks to new investors. Schipper (1989, 97), acknowledges the complicated ethical issues faced by managers as she discusses how compensation contracts may motivate managers to maximize firms' values unethically, but that eliminating these contracts can cost a company money because of the manager's lack of motivation. Parfet (2000, 485) claims that "good" EM creates stable financial performance and benefits all stakeholders because capital markets reward companies achieving stable profitability growth. Nonetheless, "good" EM can only be achieved by proper disclosure and acceptable business decisions, such as selling excess assets at reasonable prices to maintain stable profitability (Parfet 2000, 485-486). Dechow, Sloan, and Sweeney ( 1996, 31) argue that providing reliable and timely financial disclosures help companies build long-term reputations for low-cost external financing. Bochner and Clark (2008) argue that if the accounting and disclosure is handled properly, EM to meet financial analysts' forecasts is not illegal. Parfet also criticizes "bad" EM for creating artificial financial reports to hide real performance, condemning such improper EM as illegal and fraudulent. The former Chairman of the Securities and Exchange Commission (SEC), Arthur Levitt, (1998) claims that any purposeful EM, because it hides the truth of financial reports from stakeholders, is unacceptable. Levitt censures accounting professionals for fraudulent EM practices, holds them responsible for the erosion of the financial reporting 6 MBA Project - Kui-Ying Lin process, and urges them to follow and improve accounting and disclosure rules. Agreeing with Levitt, Loomis (1999) regards EM activities, such as intentionally producing misstatements in financial reports, as fraud. Merchant (1989, 188-190) urges that such EM practices not be promoted, due to the high costs of managing earnings, the risk of earnings manipulations escalating over time, and the potential delay in the company's recognition of financial problems. Clearly, such concern over EM in conjunction with the extreme EM cases of Enron, World Com, and Arthur Andersen has signalled that the integrity of financialreporting systems has been seriously undermined and that stakeholders ' trust has been violated. The need to examine earnings management ethics Most literature attributes the demise of high profile corporations such as Enron and WorldCom to questionable EM. Enron established special-purpose entities to avoid recording liabilities, while WorldCom engaged in inappropriate revenue recognition (Elias 2004. 86). Elias (2004) attributes the collapse of Arthur Andersen, the audit company for Enron and WorldCom, to these two bankrupt corporations' EM practices. These well-publicized accounting scandals have raised the issue of culpability in EM. Noreen (1988, 359-60) argues that external rewards or sanctions cannot effectively prevent managers' opportunistic behaviours, but raising ethical consciousness can inspire managers to alter their behaviours so as to make more ethical decisions. Rest (1986, 1) claims that ethics provide "basic guidelines for determining how conflicts in human interests are to be settled and for optimizing mutual benefit of people living together in groups." In their discussion of ethical decision-making, Merchant and Rockness (1994, 81) draw on Bowei's and Duska's (1990) analysis of decision-making in business and personal situations and conclude that individuals' 7 MBA Project - Kui -Ying Lin beliefs in their obligations and sense of justice far outweigh self-interest. They assert that every incidence of fraudulent financial reporting begins with an ethical transgression and that EM practices raise the most important ethical issues confronting the accounting profession (82). Likewise, Kaplan et al. (2007) claim that EM raises the most relevant and important ethical problems facing today's accounting practitioners. Greenfield (2005) claims that EM practices can lead to defrauding stakeholders even though they are not contrary to the generally accepted accounting principles (GAAP). The existing American research into EM ethics has reflected this growing concern. To explore the topic of EM ethics empirically, Merchant and Rockness (1994) used a questionnaire containing 13 questionable EM practices and found a general agreement among accounting practitioners that the ethical nature of EM in a particular situation depends on its characteristics: its type (accounting versus operating), its degree of materiality, its timing (quarterly or annual), and its purpose (increase or decrease bonuses). Elias (2002) surveyed the U.S . accounting profession and found that social responsibility was positively associated with an ethical perception of EM. Elias' subsequent study (2004) indicated that accountants in companies with high (low) corporate ethics viewed EM as relatively unethical (ethical), results which provide - even before the high profile accounting scandals went public - "early warning signals" for accounting practitioners about the ethical risks they face (85). In their surveying MBA students, Kaplan et al. (2007, 160) found that their respondents' moral judgments were strongly correlated with their perceptions of the ethical quality of the described EM behaviour. However, all of these studies used American subjects and focused only on American situations, so little is known about the degree to which these findings apply to EM ethics in Canada. 8 MBA Project - Ku i-Ying L in While the lack of research into the ethical nature of EM in Canada is one important reason for this study, a second is that Canadian companies have suffered from significant EM fraud. For example, Norte! Networks, a Canadian networking equipment maker, conducted EM fraud to manipulate reserves and change the company's earnings statements (Phillips 2007). A third reason for this research is that, unlike the U.S. Securities and Exchange Commission, whose legal power emanates from the U.S. Congress, the Ontario Securities Commission, Canada's key securities regulator, has no legal power to levy fines or enforce practitioners to forfeit illegal gains; thus Thorne, Massey, and Magnon (2003 , 307-08) conclude that Canadian accounting practitioners are "less accountable for their actions to regulatory agencies." These authors also assert that U.S. accounting standards are more detailed and rule-oriented because they rely heavily on technical guidance, while Canadian accounting standards depend on a principle-based approach which greatly emphasizes professional judgment. Therefore, increasing Canadian accountants' moral awareness of what motivates EM practices may enhance their professional judgments and help them avoid managing earnings unethically. All these three reasons indicate the need of this study. This project investigates the moral perceptions of British Columbian accounting students as they respond to selected EM scenarios and the determinants of their EM ethics. These possible determinants are personal moral philosophies (idealism and relativism) and personal characteristics (Machiavellianism and locus of control), which are tested as exogenous variables that may influence the ethical nature of EM. Bass, Barnett, and Brown (1999, 197) claim that people intend to behave in a way consistent with their personal attitudes when they confront ethical dilemmas. Therefore understanding the determinants of EM ethics may be a first step in preventing further unethical EM behaviour. 9 MBA Project - Kui-Ying L in HYPOTHESES DEVELOPMENT Rest (1986, 1-27) proposes a four-component model to describe the process by which individuals make ethical or unethical decisions. The four consecutive stages consist of recognizing ethical dilemmas, making ethical judgments, establishing ethical intention, and engaging in ethical or unethical actions (Rest 1986, 1-27; Jones 1991, 368) (see Figure 1). Rest (1986, 4) claims that the four components, although they may influence each other, nevertheless are distinctively functional: the success of one step does not ensure the success of any other. Because Rest describes each component as being discrete and because Elias (2002) also investigates only the relationship between exogenous variables and Rest's fourth component (ethical/unethical decision-making), Rest 's fourth component has been selected as dependent variables to test the hypothesized causal relationship in this project. In his survey of accountants, faculty, and students, Elias (2002) used personal moral philosophies (idealism and relativism) as two of the independent variables to examine the potential determinants of EM ethics. Elias (2002) concludes that paying more attention to individual ethical ideologies can help reduce unethical EM behaviour, and suggests that future studies should examine the relationship between locus of control and EM ethics. In response to Elias' recommendations, this project has proposed to study idealism, relativism, and locus of control as three possible determinants of EM ethics. Bass, Barnett, and Brown (1999) examined how individual characteristic variables such as ethical ideology, locus of control, and Machiavellianism affect the second and third of Rest's components, the making of ethical judgments and the establishing of behaviour intentions. In addition, because 10 MBA Project - Kui-Ying Lin Machiavellianism has been extensively used in studies of marketing practitioners, I am interested in seeing whether Machiavellianism affects accounting practitioners as they make ethical or unethical decisions. Taken together, idealism, relativism, Machiavellianism, and locus of control are proposed as the four possible determinants of EM ethics in this study. Idealism and Relativism A person's moral philosophy is a "set of beliefs, attitudes, and values" providing individuals with a framework for making ethical judgments (Barnett, Bass, and Brown 1994, 469). Ethical ideologies provide individuals with "guidelines for moral judgments, solutions to moral dilemmas, and prescriptions for actions in morally toned situations" (Forsyth and Nye 1990, 399). In their experiments, Schlenker and Forsyth (1977, 382-83) identified two dimensions of ethical ideologies: idealism-pragmatism and rule-universality. The idealismpragmatism dimension focuses on idealism in individual attitudes reflecting importance of people' s welfare while the rule-universality dimension focuses on relativists who reject universal moral principles (Schlenker and Forsyth 1997, 382-83; Forsyth 1980, 175-76). Subsequent writers discuss ethical ideologies as comprising two basic factors: idealism and relativism (Forsyth 1980; Barnett, Bass, and Brown 1994; Bass, Barnett, and Brown 1999). I employ and define the ethical ideologies based on this classification in this project although the term of idealism versus relativism is different from that of absolutism versus relativism as ethics are more commonly organized by philosophers (Poff 2005, 2). Idealism describes personal attitudes toward both the outcomes of actions and how the outcomes affect others ' welfare (Elias 2002, 36), and is defined by the extent to which individuals believe that morally appropriate actions will always generate desirable outcomes (Forsyth 1980, 175-76; Bass, Barnett, and Brown 1999, 184). High idealists believe that ethical actions always 11 MBA Proj ect - Kui-Ying Lin produce positive outcomes and avoid harming others (Bass, Barnett, and Brown 1999, 187; Elias 2002, 36); low idealists see outcomes as often being the combinations of desired and undesired consequences (Forsyth 1980, 176) and believe that harmful outcomes may be needed to obtain a greater good (Barnett, Bass, and Brown 1994, 470). On the other hand, relativism, a universal-rules dimension (Schlenker and Forsyth 1977, 393), is defined as the extent to which individuals accept or reject universal moral principles or rules when making ethical decisions (Forsyth 1980, 175; Bass, Barnett, and Brown 1999, 184). High relativists believe that moral actions depend on the particular factors of a situation and not on moral absolutes (Barnett, Bass, and Brown 1994, 469 ; Elias 2002, 36), while non-relativists or moral absolutists rely on universal moral principles or rules in judging ethical actions (Barnett, Bass, and Brown 1994, 469). Wakefield (2008, 119-20) argues that high idealists acknowledge moral absolutes and assume that the best consequences come from following universal moral principles while high relativists reject universal moral rules and base their judgments on personal values and perspectives. Forsyth (1992, 462) assumes that individuals based on personal moral philosophies can range from high to low in their emphasis on principles or consequences and identifies four distinct ethical ideologies : situationism, subjectivism, absolutism, and exceptionism (see Figure 2). Forsyth (1980, 176) asserts that when individuals were classified into four ethical ideologies they adopt one of four different approaches to making ethical judgments. Situationists are both idealists who pursue the best possible consequences and relativists who believe that ethical rules cannot be applied across situations (Forsyth and Pope 1984, 136566). Situationists eschew universal moral principles and insist on producing positive results to benefit all involved (Forsyth 1992, 462). This philosophical approach is based on ethical 12 MBA Project - Kui -Ying Lin skepticism and is similar to utilitarianism in that individuals must act in a manner that produces the greatest good for the greatest number of people (463 ). As a result, situationists are more likely to make their ethical judgements based on potential benefits to others (Barnett, Bass, and Brown 1994, 472). Subjectivists are skeptical of ethical principles and believe that it is impossible to avoid negative behaviours (Forsyth and Pope 1984, 1366). These individuals consider their moral decisions to be subjective and individualistic judgments which are not based on moral absolutes or actions to benefit others; their ideologies are parallel to "an egoistic moral philosophy" (Forsyth 1992, 462). Absolutists are idealists who believe that actions generate both positive and negative consequences and feel that actions should always be consistent with absolute moral rules (Forsyth and Pope 1984, 1366). Absolutists believe that individuals should simultaneously strive to produce positive results and follow general moral principles and condemn actions that harm others and violate moral absolutes (Forsyth 1992, 463). This ideology corresponds to deontology; deontologists' actions are based on judgements which are guided from and consistent with universal moral rules. Absolutists make their ethical judgments based on the consideration of both "consequences of others and universal moral laws" and judge other ethical issues more harshly than individuals with other ethical ideologies (Barnett, Bass, and Brown 1994, 472). Exceptionists consider exceptions to moral norms permissible and believe that individuals may need to break moral principles when circumstances warrant in order to avoid generating negative consequences (Forsyth and Pope 1984, 1366). Exceptionists appreciate moral absolutes but do not believe that harm to innocent people can be avoided or that risking others' welfare is not always wrong (Forsyth 1992, 463 ). Their outlook is closely related to rule-utilitarianism which considers following moral rules useful in terms of providing a 13 MBA Project - Ku i-Ying Lin framework for making choices and acting in ways to produce the best consequences for all concerned (463). Forsyth (1992, 461) posits that the type of ethical thinking -- situationism, subjectivism, absolutism, and exceptionism -- affects individuals ' moral judgments of business practices. Since Forsyth's study, empirical studies have examined the relationships between individuals' moral philosophies and their ethical business decisions (Elias 2002). In their experiments with university students, Barnett, Bass, and Brown (1994, 474-75) found that idealism has significant correlations with ethical judgments but that relativism does not have significant correlation with such judgments. In their survey of 153 senior undergraduate and master's students from three U.S. business schools, Singhapakdi et al. (1996, 1137) found that relativism was a significant predictor of social responsibility and profitability, long-term gains, and short-term gains. In their survey of marketing practitioners, Bass, Barnett, and Brown (1999, 197-98) found that although idealism was negatively associated with judgement of unethical selling practices and high idealists considered the questionable acts more unethical, idealism was unconnected to behavioural intentions; relativism had no significant association with either ethical judgments or behavioural intentions. Further, Elias (2002, 40) surveyed American accounting practitioners, faculty, and students and found that idealism is positively correlated with ethical perceptions of EM, while relativism is negatively correlated with such perceptions; high idealists (situationists and absolutists) view EM actions as relatively unethical while high relativists consider these practices relatively ethical. Elias concludes that absolutists judged EM scenarios most harshly, followed by situationists, exceptionists, and finally subjectivists (Elias 2002, 40). Following Elias' findings , the first two hypotheses were framed . 14 MBA Project - Kui-Ying Lin HI: High idealists will consider earnings management practices more unethical than low idealists. H2: High relativists will consider earnings management practices less unethical than low relativists. Machiavellianism Published in 1532, The Prince is an influential political and philosophical work, resting on the premise that self-interest underlies human behaviour (Codevilla 1997). Its author, Niccolo Machiavelli, focuses on "the art of government," laying out details about how political leaders can organize government and armed forces and how they can build a solid foundation of political power both by reducing the threats posed by their enemies and by making troops happy, dedicated and willing to fight and win (Codevilla 1997). The art of government has been demonstrated by modern charismatic leader behaviour. House and Hower (1992, 97) vindicate that Machiavellianism is positively connected with personalized charismatic leader behaviour. Deluga (2001, 353-54) found that Machiavellians had selfconfidence, competence, and charisma, that American presidential Machiavellianism was positively connected with charismatic leadership, and that Machiavellian-oriented presidents greatly succeeded in legislations. Bass, Barnett, and Brown (1999, 188) consider Machiavelli's goals to be "self-interest or the well-being of a community or nation." These authors describe Machiavellian individuals as being "relatively unconcerned with judging the morality of ethically ambiguous actions and ... more likely to behave in ways (ethical or unethical) that would lead to desired ends." Wakefield (2008, 120) posits that individuals with high 15 MBA Proj ect - Kui-Ying Lin Machiavellianism focus on the end results and tend to depend on personal moral conscience in decision-making to ensure that the "end of the matter" is to their advantages. Conceptualizing Machiavellianism by reference to The Prince, Christie and Geis use Machiavellian thought to examine interpersonal behaviour, identifying Machiavellian individuals as persons who are more likely to reject ethical norms and manipulate others for their own purposes ( 1970, 1). Although Machiavelli created a new ethical code without departing from traditional Christian moral guidance, The Prince has been criticized for harshness and amorality (Codevilla 1997). Hargis (2006, 29-30) claims that individuals with high Machiavellianism are more willing to lie, manipulate others, and pay a bribe than those with low Machiavellianism; thus Machiavellianism is significantly and positively connected with all kinds of unethical behaviour. Leary, Knight, and Barnes (1986, 78) argue that high Machiavellian individuals disavow absolute moral rules in the belief that moral rules may result in negative outcomes and adopt a pragmatic, opportunistic approach to moral behaviour that occasionally harm to others. Nevertheless, Beu (200 1, 40) claims that high Machiavellians may not be a lack of ethics but just operate under a set of instrumentalist moral guidelines which differ from conventional ethics. Viewing Machiavellianism very negatively, Hunt and Chonko (1984, 30) assert that highly Machiavellian individuals employ aggressive, manipulative, exploitive, and devious moves to achieve personal objectives. However, Singhapakdi and Vi tell (1990, 7) regard Hunt and Chonko ' s judgment that Machiavellianism is an "amoral" way of manipulating others as inappropriate. Instead, Singhapakdi and Vitell (1990,7) accept Robinson and Shaver's (1973 , 592) definition that Machiavellian persons have "a cool detachment" making them "less emotionally involved 16 MBA Proj ect - Kui-Ying Lin with other people, with sensitive issues, or with saving face in embarrassing situations." This is why Wastell and Booth (2003, 739) argue that Machiavellian individuals live in an "emotionally impoverished world . .. [and] in a world of crowded isolation." Deluga (2001, 354) claims that the depersonalized social influence of Machiavellianism is related to the U.S. presidential effectiveness and that a "less emotionally involving cool detachment from others enable Machiavellian-oriented [American] presidents to advance their goals." It would hardly be a contentious statement to say that self-interest has seemingly coloured ethical decisions in the accounting profession. For example, the yearly consulting fees of $20 million received from Enron undoubtedly contributed in large part to the failure of Arthur Anderson to fulfill its auditing responsibility to the investing public (Ketz 2002, 7). It is, therefore, well worth asking whether Machiavellianism affects the behaviour of accounting practitioners. Wakefield (2008) found that accountants were lower Machiavellians than other professionals but Machiavellianism has not been widely considered in empirical studies of causal relationships between Machiavellianism and accounting ethics. However, Machiavellianism has been explored in studies of marketing activities. Considerable research in this field supports a connection between Machiavellianism and unethical behaviour. Singhapakdi and Vitell (1990, 13), for example, have found that Machiavellianism has a negative effect on ethical behaviour and conclude that individuals with high Machiavellianism have the tendency to perceive ethical problems as being less serious than those with low Machiavellianism. In two similar empirical studies of the relationships between marketing decisions on kickback payments to purchasing agents, Hegarty and Sims found that individuals with high Machiavellianism were less ethical, compared with low Machiavellianism (1978, 454-55) and that Machiavellianism was 17 MBA Project - Kui -Ying Lin significantly and positively related to unethical decision-making (1979, 337). In their study, Bass, Barnett, and Brown (1999, 196-98) found strong linkages observed between Machiavellianism and behavioural intentions among marketing practitioners. Hunt and Chonko (1984, 32) claim that highly Machiavellian individuals hold in contempt conventional morality and are more likely to engage in unethical behaviour when self-interest is involved. From a survey ofuniversity psychology students, Beu (2001), too, found that persons with high Machiavellianism are relatively likely to approve of unethical behaviour. Likewise, Hargis (2006, 11; 30) asserts that Machiavellianism is significantly and positively associated with various fonns of unethical behaviour. In general, these empirical studies show that individuals with high Machiavellianism are more likely to approve of unethical decisions and less likely to behave ethically than those with low Machiavellianism. Based on these findings, the third hypothesis was formulated. H3 . High Machiavellian individuals will judge earnings management activities less harshly than low Machiavellian persons. Locus of Control Rotter (1966) asserts that individuals differ in the degree to which they feel that they control (internal locus of control) or do not control (external locus of control) subsequent events. The perceived likelihood of gratification, reward, or reinforcement partly arises from individual perceptions of the reward as contingent upon their own behaviour or independent of it (Rotter 1966, 1). Rotter (1966) further posits that internal locus of control and external locus of control exist on a continuum. Therefore, locus of control (LC) refers the extent to which "individuals believe that outcomes are contingent upon their personal characteristics 18 MBA Proj ect - Kui -Ying Li n or behaviour" (Bass, Barnett, and Brown 1999, 187). Beckham, Spray, and Pietz (2007, 28687) argue that high internal LCs believe in their own ability to make differences and attribute consequences to hard work and personal efforts while high external LCs are followers as related to passive dependency. Individuals with predominantly internal LCs believe that outcomes result mostly from their own behaviour/efforts, while those with strongly external LCs believe that life events are out of their personal control and attribute them to fate, luck, or destiny and are less likely to assume personal responsibility for the consequences of ethical/unethical behaviour (Rotter 1966, 1; Hegarty and Sims 1978, 454; 1979, 333; Trevino 1986, 61 0; Jackson, 1997, 1-2; Bue 2001 , 39-40; Hargis 2006, 10). This concept oflocus of control is related to the theory of attribution which people identify the causes of outcomes as their own behaviour or as the behaviour of others; individuals perceive action outcomes internally when they believe that they are the primary cause of the events and externally when they consider outside factors the primary cause (Howard and Levison 1985). The believed behaviour-outcome link makes individuals with high internal LCs exhibit greater ethical sensitivity and be more likely to bear responsibility for their actions while the absence of links between actions and expected consequences makes individuals with high external LCs less likely to assume responsibility for their actions and respond less ethically (Hume and Smith 2006, 53). Subjects with internality believe that they have only themselves to blame when facing failure while individuals with externality can act as "an adequate defence against failure" and those with very high scores towards externality reflect in "a defensiveness related to significant maladjustment ... [and] a passivity in the face of environmental difficulties" (Rotter 1966, 16). Phares and Lamiell ( 197 5) argue that internals, as compared to externals, are less likely to provide others with understanding and sympathy 19 MBA Project - Kui-Ying Lin and view others as needing assistance or help. Individuals with internality strongly believe that they can control their own destiny, an effect that makes them be high alert to environment, take steps to improve their environmental conditions, adjust their future behaviour, consider skills or achievement reinforcement valuable, be concerned with their failure, and resist to be influenced by others (Rotter 1966, 25). Locus of control (LC) has been extensively discussed in research on children's behaviour. Children with low social skills, self-esteem, and self-concepts score higher on an external LC measure, whereas those with leadership ability, high self-esteem and self-control, and family support score higher on an internal LC measure (Jackson 1997, 2-3). In a study of 238 school children, (Kliewer and Sandler 1992, 393-408) found that internal LCs helped children buffer the linkage between negative life events and psychological symptoms and these children had effective problem-serving skills and positive coping strategies while external LCs made children vulnerable to negative life events or psychological symptoms. Skinner and Connell (1986: in Kliewer and Sandler (1992, 394-95) claim that children become increasingly internal in their LC beliefs as they age; this tendency comes from the shift in cognitive understanding between younger and older children and between children and adults. Kliewer and Sandler (1992, 410) draw on Reich and Zautra's (1990) views that a control-enhancing intervention promotes the well-being of elderly individuals with internal LCs but not that of elderly persons with external LCs. In her study of school-age children's LCs, Jackson (1997) found that children with internal LCs are more likely to perfonn selfcare and social-adaption activities than those with external LCs. In their study of 40 schoolaged males, DuCette, Wolk, and Friedman (1972) consider locus of control as infonnation 20 MBA Project- Kui-Ying Lin seeking strategies and found that children with internal LCs responded more creatively, actively, and efficiently than those with external LC. In business, Trevino and Youngblood (1990, 378-83) conducted an experiment on 94 MBA students and found that LCs directly and strongly affected ethical decision-making behaviour and indirectly influenced moral decision making through outcome expectancies; subjects with internality demonstrated more ethical behaviour than those with externality. Trevino ( 1986, 61 0) claims that a person with an internal LC is more likely to assume responsibility for the consequences of his/her actions and is less likely to depend on external forces to motivate his/her behaviour; managers with internality demonstrate more consistency between their moral judgments and actions than those with externality. The study by Bass, Barnett, and Brown (1999, 196) found that an external LC had no significant association with making ethical judgments or forming behavioural intentions. However, in laboratory experiments on graduate business students, Hegarty and Sims, (1978, 454-455) found that there were relatively strong links between LCs and ethical decision making; subjects with externality were less ethical than those with internality. In addition, in their behavioural laboratory study, Trevino and Youngblood (1990, 382) found that LCs have a strong direct effect on decision-making and that subjects with internal LCs exhibited more ethical behaviour than those with external LCs. In an experiment on university psychology students, Beu (2001) found that individuals with internal LCs are more likely to behave ethically than those with external LCs. According to the aforementioned findings, I believe, as Elias suggests (2002, 43), that LCs are possibly an important detenninant of earnings management ethics. Therefore, the fourth hypothesis was formulated . 21 MBA Project - Kui-Ying Lin H4: Individuals with internal locus of control will tend to perceive earnings management practices as more unethical than those with external locus of control. Figure 3 summarizes the relationship between the exogenous variables (relativism, idealism, Machiavellianism, and locus of control) and the endogenous variable (ethical /unethical decision-making on earnings management). These possible causal relationships may help explain unethical earnings management practices in organizations. 22 MBA Proj ect - Kui-Ying Lin RESEARCH METHODOLOGY Measurements The objective of this study is to test the possible determinants of earnings management (EM) ethics. Personal moral philosophies and individuals' characteristics had to be measured. Forsyth's (1980) Ethics Position Questionnaire (EPQ) was used to constitute the scores of idealism and relativism. Similarly, Christie's and Geis' (1970) Mach IV scale was employed to measure Machiavellianism while Rotter's (1966) Internal-External scale helped produce scores of locus of control. In addition, a set of 18 earnings management (EM) scenarios was gleaned and synthesized from previous research on EM ethics. Taken together, these questionnaires helped produce needed mean responses of independent and dependent variables for testing the four hypotheses. Ethics Position Questionnaire Forsyth (1980) developed an Ethical Position Questionnaire (EPQ) to measure personal moral philosophies. This scale has been widely used in research in business ethics. In order to test causal relationships between personal moral philosophies and business ethi cs, researchers have used the EPQ to constitute the idealism and relativism scores of respondents (Barnett, Bass, and Brown 1994; Singhapakdi et a!. 1996; Bass, Barnett, and Brown (1999); Elias 2002). The EPQ consists of two I 0-item subscales to measure idealism and relativism. The ethical idealism subscale includes items such as "A person should make certain that their actions never intentionaJJy harm another even to a small degree" and "It is never necessary to sacrifice the welfare of others." The ethical relati vism subscale includes items such as "What 23 MBA Project - Kui-Ying L in is ethical varies from one situation and society to another" and "Different types of moralities cannot be compared as to rightness." The idealism scale measures individuals' acceptance of moral absolutes while the relativism scale measures their refusal of universal ethical principles (Singhapakdi eta!. 1996, 1135). Subjects indicate their degree of disagreement or agreement using a nine-point scale ranging from one (completely disagree) to nine (completely agree) (see Appendix A). The scoring system is that the mean rating ofthe first 10 items produces the idealism score and the mean rating the second 10 items generates the relativism score. Individuals with high scores on the idealism scale are primarily concerned with the welfare of others and believe that ethically "right" actions lead to good and positive outcomes, while those with high scores on the relativism scale reject the existence of absolute ethical principles (Forsyth 1980; Forsyth 1992, 465; Vitell eta!. 2003, 155). Following previous studies (Forsyth 1985; Barnett, Bass, and Brown 1994; Elias 2002), the median scores of idealism and relativism were calculated and subjects were classified into one of the four ethical ideologies according to their scores relative to the median of each subscale. The classification was for the purpose of determining these ethical ideologies ' perceptions of EM ethics . Mach IV Scale Christie and Geis (1970) created the Mach IV scale to measure Machiavellianism. The scale has been widely used in related marketing studies. Marketing researchers have used Machiavellianism as a predictor or moderator to investigate the relationship between Machiavellianism and decision-making (Geis and Moon 1981, 767; Hunt and Chonko 1984, 40-41 ; Bass, Barnett, and Brown 1999, 191). As a result, the Mach IV scale was used in present research to measure Machiavellianism. The Mach IV consists of20 items, 10 keyed 24 MBA Project - K ui-Ying Lin to the endorsement of Machiavellian statements and 10 keyed in the opposite direction (Christie and Geis 1970, 16). The respondents were asked to indicate their level of agreement with each of the 20 statements on the given scale. Each statement was evaluated by a sevenpoint Likert-type scale ranging from one (strongly disagree) to seven (strongly agree) (see Appendix B). The Mach IV scale was scored by summing the numerical responses to all statements and adding a constant of 20. This scoring system generates a maximum possible score of 160 (7 X 20 + 20 = 160) if subjects strongly agree with every statement worded in the pro-Machiavellian direction and strongly disagree with every statement worded in the anti-Machiavellian direction, while the reverse pattern generates a minimum score of 40 (1 X 20 + 20 = 40) (Christie and Geis 1970, 26-28). With a 100-based point, a total score for the ten items keyed in the negative direction is deducted from that of the ten items worded in the positive direction; subjects will obtain a score of 100 if their agreements or disagreements balance out (27). As a result, these authors set a score of 100 as a theoretical neutral point to assess individual Machiavellian orientations; scores higher than 100 points show tendencies to indiscriminate agreement while scores lower than 100 points indicate tendencies to disagree with Machiavellianism (27). Internal-External Locus of Control Scale Rotter (1966) developed an Internal-External (I-E) scale oflocus of control (LC) with a 29-item forced-choice test including six filler items (also forced-choice pairs) designed to mask the purpose of the test (see Appendix C). This 29-itenn forced-choice inventory was used to measure LCs because the I-E scale has internal consistency (Rotter 1966) and has been widely used in the literature. The 29 items deal with individuals' beliefs about how reinforcement is controlled (Rotter 1966, 10). The I-E scale measures personal perceptions of 25 MBA Project - Kui- Ying Lin the extent to which people exert control over events in life (Trevino 1986). For each of the 29 item pairs that are scored, respondents receive a point if they choose the item reflecting high external LCs. Since the six-filler items are not counted in the scoring, participants' scores can range from zero to twenty-three. Respondents with predominantly external LCs have high scores while those with internal LCs have low scores (Trevino and Youngblood.1990. 380). Hume and Smith (2006, 54) divided their student respondents into internals if students scored less than or equal to ten and externals if students scored greater than or equal to eleven and used the LC designation as the independent variables in their analyses. Ethical Scenarios Scenarios allow researchers to present concrete, hypothetical decision-making situations which closely approximate real-life decision-making situations (Alexander and Becker 1978, 93; Bass, Barnett, and Brown 1999, 192). Scenarios provide unbiased selfreports and are accurate in measurement. Because they are appropriate for sensitive issues, they allow researchers to elicit honest responses from participants (Rallapalli, Yitell, and Barnes 1998, 161 ). They have been widely used in research in business ethics . Schultz et al. (1993) used six diverse situations to investigate the propensity of corporate managers and professionals to report questionable acts. Similarly, Barnett, Bass, and Brown (1994, 476-78) developed and utilized 26 scenarios depicting marketing dilemmas to examine the relationship between ethical ideologies and ethical judgments. Cohen and Pant (1995) used eight scenarios to examine the differences among auditors ' ethical decision-making behaviour in an international public accounting company, while Kaplan (200lb) employed three hypothetical scenarios in his experimental study of evening MBA students to investigate earnings management (EM) ethics. To determine ethical perceptions of EM, 26 MBA Project - Kui-Ying Lin Merchant (1989, 180) developed 13 EM scenarios, including six cases of operating manipulations and seven cases of accounting manipulations. These 13 EM activities were used in the research of Merchant and Rockness ( 1994), Fischer and Rosenzweig ( 1995), and Elias' (2002; 2004) into EM ethics. However, Elias (2002) advises that future research should use more current scenarios than Merchant ' s to examine ethical perceptions of EM. In response to both the demonstrated usefulness of scenarios and the Elias' advice, I developed 18 EM scenarios gleaned from previous research on EM ethics, especially from Merchant's 13 EM scenarios. These 18 potentially questionable EM scenarios consist of nine cases of operating manipulations and nine cases of accounting manipulations (see Appendix D). Respondents were asked to evaluate each scenario in a five-point Likert scale, ranging from one (completely ethical) to five (completely unethical) and indicate the extent to which they perceive it as describing ethical behaviour. These 18-EM scenarios lend support to Elias ' (2002) notion that examining ethical perceptions should employ more recent and more creative EM scenarios. Jones (1991) argues that the moral intensity (magnitude of consequences which is the total sum of harm or benefits) of ethical issues should translate into personal and situational variables for studying ethical decision-making processes. Bass, Barnett, and Brown (1999, 199) suggest that future research should examine the impact of moral intensity on the ethical decision-making process. Merchant and Rockness ( 1994, 89) found that the larger EM action was rated significantly more unacceptable than the smaller EM practice. Likewise, in a study of American accountants, Shafer (2002) hypothesized that quantitative materiality will affect ethical judgments of the acceptability of fraud. He found that quantitative materiality did not affect moral judgments but did influence the likelihood of fraud. Singhapakdi, Vi tell, and 27 MBA Project - Kui-Ying Lin Kraft (1996, 251-253) found that moral intensity was strongly and positively related to both ethical perceptions and intentions. In response to these suggestions and findings, in this study, I purposely incorporated the feature of moral intensity into the 18 EM scenarios and the statements of a and b for scenario 5 and 16 were to examine whether moral intensity (materiality) affects accounting practitioners' moral decision-making. Pretesting Cooper and Schindler (2008, 369-71) claim that pretesting is a critical practice in developing a successful survey. In order to minimize the burden on participants and maximize response, questionnaires should be field-tested by "sample participants or participant surrogates" whose characteristics and backgrounds similar to desired participants (Cooper and Schindler 2008, 269). Hunt and Vi tell (1986, 11) assert that all scenarios should be pretested for specific ethical issues involved in the scenarios. Singhapakdi and Vitell (1990, 10) pretested their questionnaires on 597 randomly selected university alumni and obtained a 20.6% response rate before testing them on American marketing practitioners. Singhapakdi, Vitell, and Kraft (1996, 248) used 153 senior and Master's level business students from three universities to pretest their research instruments as a final step in the questionnaire development; based on the pretest results, they dropped and revised some of the original scenarios. Due to time constraints, it was not feasible to pretest on a large scale as these previous researchers had done. However, the author had her questionnaires fieldtested by four participant surrogates, a chartered accountant, a certified management accountant, and two accounting students. The approach of collaborative pretests was used and participants were aware of a preliminary test of the survey instrument in the refinement process (Cooper and Schindler 2008, 369). While completing the questionnaires, these four 28 MBA Project - Kui-Ying Lin participant surrogates helped probe the detai Is of all questions and made some helpful comments on the questionnaire design. According to their feedback, the author modified all questionnaires in terms of readability, meaningfulness, clarity and appeal to minimize the burden on participants and maximize responses. Participants This project focused only on accounting students in British Columbia because prior studies have widely used students to study causal relationships between predictors and responses. For example, Beu (2001) used business and industrial/organizational psychology students in her study to test how personalities such as locus of control and Machiavellianism influence decision-making about ethical intent or behaviour. Greenfield (2005) used 145 accounting students and 230 non-accounting students from mid-Atlantic universities and colleges in his study to examine whether professional commitments and ethical positions affect earnings management behaviour. Barnett, Bass, and Brown ( 1994) also used 166 students from business classes at a medium-sized southern university in their study to examine how ethical ideologies (idealism and relativism) influence individuals ' moral decision. The questionnaire survey was conducted with four samples of students at two business schools in British Colwnbia. Based on the expectation of their understanding of earnings management (EM) activities, the vital role of auditing in the integrity of financial reporting, and the very likelihood of their future careers in accounting, senior-level accounting students enrolled in the Auditing course at these two business schools were chosen to collect raw data. The majority of these students intend to obtain one of the three accountant designations in 29 MBA Project- Kui-Ying Lin British Columbia. Some of these subjects will start working for big accounting firms such as KPMG, PricewaterhouseCoopers, Delotte & Touche, and Ernst & Young in September 2009. As a result, theses subjects well represent would-be accountants in British Columbia. Self-administered questionnaires were used to collect raw data; students were offered a small incentive, a five-dollar Tim Hortons' card, for the return of completed questionnaires. This incentive strategy is consistent with previous studies. Wertheim, Widom, and Wortzel (1978, 236), for example, provided first-year graduate students with small incentives for their return of completed questionnaires. In Mchoskey's (1999, 269) study, subjects (257 students) received extra course credits for their voluntary participation in completing questionnaires. Each student at both business schools received a survey instrument package consisting of a cover letter (see Appendices F, G, and H) (informing them about the survey and asking for their participation), an informed consent form (see Appendix I) (stating the details of the survey), and five questionnaires (see Appendices A, B, C, D, and E): the EPQ, the Mach IV Scale, the I-E Scale, the EM scenario questionnaire, and the demographic information questionnaire. It took about 20 to 30 minutes of students' time to complete the survey. Permission was received to use the survey instrument in School A and B with the support of four instructors from Auditing classes. Based on instructors' preference, the survey processes slightly differed from one instructor to another. Two instructors at School A helped manage the survey from distributing survey instruments to collecting the completed questionnaires. Students were told to be serious about the survey, take more time, and answer all questionnaires honestly. School-A-Instructor-1 sent out 57 survey packages and received 32 responses. Four of them were determined to be unusable due to missing data, resulting in a sample size of 28 respondents and representing a total response rate of 49.12% (see Table 30 MBA Project - Kui-Ying Lin 1). Schooi-A-lnstructor-2 gave out 73 survey packages and received a total of 57 responses. Two of them were unusable because of missing data; this sample consisted of 55 respondents, representing a total response rate of75.34%. With the permission ofSchooi-B-Instructor-1, the questionnaires were administered during regular class periods with the researcher present. The researcher remained present for the duration of the survey to distribute and collect the questionnaires and answer questions. Under the encouragement of the instructor, sixty-nine students participated in the survey. After disregarding questionnaires with missing data, this single request netted a total of 66 responses, an approximate 94.29% response rate. SchooiB-Instructor-2 allowed the researcher to promote the survey and distribute the questionnaires during his class time, asked his students to complete the survey outside of the class time, and returned the completed forms by mail. Twenty-seven questionnaires were returned within five weeks of the initial distribution of 48 survey instrument packages. One of them was determined unusable due to the missing data and this sample comprised of 26 respondents, representing about 54.17% response rate. Table 2 summarizes the demographic profile of these four samples. Sixty-five point nine percent of the total respondents were female. A significant number of responses were under the age of thirty (84.44%). Subjects with no working experience in the accounting field predominated in the respondents (73 .41 % ). Scale Reliability A preliminary step of testing research hypotheses was to assess the reliability of composite independent variables in the study (Singhapakdi and Vitell 1990, 12). Reliability is a primary assessment of the internal consistency of a scale and describes the degree to 31 MBA Project - Kui -Ying Lin which an instrument is free of random and unstable error and works well at different times under different conditions (Cooper and Schindler 2008, 292-93). Internal consistency is homogeneity among items. The common techniques for assessing the internal consistency are split-halves and Cronbach's alpha (Cooper and Schindler 2008, 293). The split-half technique is that item scores obtained from the administration of an instrument are separated by items into randomly selected halves; if the two halves are highly correlated, the instrument is considered as highly reliability in an internal consistency sense (294). Cronbach 's coefficient alpha determines the internal consistency of the test to estimate the mean reliability of scales through all possible ways of splitting a set of items in half (Beu 2001, 76). Cronbach's alpha is widely accepted as a formula for evaluating internal consistency for multi-item scales at interval level of measurement (Cooper and Schindler (2008, 689). The larger the overall alpha coefficient, the more likely that items contribute to a reliable scale. The Ethical Position Questionnaire, the Mach IV, and the Internal-External scale used in present research are assumed content valid and reliable on the basis of previous research and application. The gleaned Cronbach 's Alpha coefficients from previous research are presented in Table 3. The EM scenarios used in this research were complex and not homogeneous among scenarios so the alpha value could expect to be a lower bound. Since these scenarios were gleaned and synthesized from available literature by the author, there was no previous research to provide any standard data for comparison. Therefore, the reliability evaluation focused only on the four explanatory variables. Since Rotter's (1966) Internal-External scale of locus of control consisted of 29-forced-choice items which did not meet a prerequisite condition for the split-halftechnique as Cooper and Schindler (2008, 294) 32 MBA Project - Kui-Ying Lin specified because these items had no similar statements or questions. Consequently, the Cronbach's alpha formula was employed to assess the measurement scales in this study. The SAS correlation procedures were used to assess the reliability of these measurement scales. The results of correlation analyses are reported in Table 4. Nunnally (1967; in Merchant 1985, 74) argues that the lower limit of acceptability ofthe Cronbach 's alpha coefficient is around 0.5 to 0.6. In comparison with previous research and Nunnally's standard, the overall Cronbach's alpha coefficients ofthese samples (except School-Blnstructor-1) were very low (see Table 4). The low Cronbach's alphas might raise the concern of internal consistency. However, Robinson , Shaver, and Wrigtsman (1991 , 10) argue that the alpha indices depend on a researcher's preference for large or small numbers and that an absolute threshold value is unable to be designated as a benchmark; thus, criterion of internal consistency is statistically significant correlations. According to these authors' argument, all four independent variables were included in the next testing of hypotheses, leaving correlation coefficients to assess the status of these variables. 33 MBA Project - Kui-Ying Lin RESULTS Earnings Management Ethics The first test involved finding any differences among groups of respondents on earnings management (EM) ethics. Table 5 shows the mean scores, standard deviations, and mean differences for the responses of all four samples for EM scenarios. The neutral value of 3 in a 5-point Likert-type scale of the EM questionnaire was used to calculate the differences in means by deducting the neutral value of 3 from the question outcome. As shown in Table 5, all four samples had the positive value of differences in means for operating, accounting, and total manipulations. These results indicate that these students rated these three attributes of EM actions as more unethical. In addition, the percentages of respondents ' score ranges were calculated to divide responses into ethical, neutral, and unethical orientations and the results are reported in Table 6. The first three lines of Table 6 indicate that all samples had high percentages of respondents scoring over 3 (ranging from 84.85% to 96.43%); this further confirms that respondents judged EM practices more unethical. The comparison of studies in Table 7 shows, surprisingly, that the four samples of respondents in present research had the highest mean responses and were the harshest judges of EM ethics. For 18 EM scenarios, the data show that these students overall judged most of the 18 EM practices (except scenario 1, 2, 7, and 8) as more unethical because of the positive values of differences in means (see Table 5) and the higher percentages of respondents scoring excess 3 on the 5-point scale of the EM questionnaire (see Table 6). All four samples had negative values of mean differences for scenario 1 and 7, meanings that their judgements about expensing maintenance of buildings early and pulling sales from a future period into 34 MBA Project - Kui- Ying Lin the current period were less unethical. Also, the data in Table 5 show, surprisingly, that students of two samples from School B had different judgments about EM ethics. Respondents of School-B-lnstructor-1 were the most liberal in judgments, meanings that their judgments about the acceptability of scenario 1 and 7 were the most lenient while respondents of School-B-lnstructor-2 were the most conservative in their judgments. These differences were not consistent for all scenarios. For example, the fonner judged question Sa, 5b, 6, 10, 11, 12, and 15 more unethical than the latter. In addition, all respondents consistently rated question 6 as the most unethical EM activity because of the highest mean responses and differences among all18 EM scenarios (see Table 5) and the high percentages of response means excess 3 on the 5-point scale (see Table 6) . These students viewed the activity of recording sales for customers who did not order anything as a serious ethical infraction . Among four groups, School-A-lnstructor-2 of respondents was the harshest group in judging scenario 6 while School-A-lnstructor-1 of students was the most lenient group. These two groups were at the same school but had different judgments in scenario 6. Also, lack of agreement exists among questions. The large ranges of responses and the high standard deviations from some scenarios show that respondents did not agree in their judgments about EM ethics. Take scenario 10 for example: the ranges of mean scores were from 1 to 5 and the standard deviations were from 1.011 to 1.27 (see Table 5), resulted in disagreement over judgments. Further, scenarios were re-organized to create four contrasting attributes. An Analysis of Variance (ANOV A) was then conducted to analyze the differences in each pair of the four attributes of EM scenarios. Statistical significance at the (a = o.1) level was designated to test hypotheses in both ANOV A and regression analyses because the same significant level for 35 MBA Proj ect - Kui-Ying Lin testing hypotheses has been widely used by Kaplan (200 1b); Elias (2002; 2004); Davidson et al. (2004); and Su (2005). The ANOV A results are shown in Table 8. The first four lines of Table 8 compare the mean scores for the EM questions depicting operating methods (question l - 8) with those depicting accounting methods (question 9- 16b) in the four samples. Consistently, the mean scores of all four groups provide clear evidence of statistical significance associated with differences in their judgments about EM ethics. In general, accounting-based EM practices, as compared to operating-based practices, were judged much more harshly. The results also show that School-B-lnstructor-1 of respondents was the most lenient group in judging operating manipulations and the harshest group in judging accounting manipulations (see Table 8). Conversely, School-A-lnstructor-2 of students was the harshest sample in rating operating-based EM and the most lenient sample in rating accounting methods of managing earnings . This finding suggests that the types of EM practices greatly affect ethical judgments and shed the same light as previous research. Previous studies found that operating methods of EM practices were not considered unethical but accounting methods of EM activities were regarded as unethical or ethical infractions (Merchant and Rockness 1994, 89; Fischer and Rosenzweig 1995, 436-39; Kaplan 2001b, 294-95; Elias 2002, 43). The next four lines of Table 8 indicate that there was no statistically significant difference in respondents' acceptability judgments whether EM activities were consistent with the generally accepted accounting principles (GAAP) or not. This result is consistent with the finding of Merchant and Rockness (1994, 88-89). The third four lines of Table 8 show that there was significant difference in respondents' judgments on EM activities. Students of all four groups rated increasing profits as the purpose of EM practices much 36 MBA Proje ct - Kui-Ying Lin more unethical. In other words, these students considered the direction of effect on EM activities mattered. However, this tendency differs from the study result of Merchant and Rockness ( 1994, 89). They found that EM actions to increase profits made no significant difference in the respondents' ethical judgements. The last four lines ofTable 8 show that there was a statistically significant difference in the mean scores of this contrasting attribute and that respondents judged materiality more unethical than immateriality. This suggests that larger EM actions (Sa and 16a) were rated significantly more unethical than smaller EM practices (5b and 16b). This appears consistent with previous research. Merchant and Rockness ' (1994, 89) found that the larger EM activity was judged significantly more unacceptable than the smaller EM action. Shafer (2002) found that quantitative materiality did influence the likely of fraud . This is why Jones (1991) argues that the moral intensity (magnitude of consequences) of ethics issues should incorporate into personal and situational variables in the study of ethical decision-making. Personal moral philosophies Table 9 summarizes the mean scores, standard deviations, and mean differences for the responses of idealism and relativism. The neutral value of 5 in the 9-point EPQ scale was used to calculate the differences in means by subtracting the scale neutral value of 5 from the question responses. The results show that all four samples had the positive values of mean differences for idealism and relativism (see Table 9); this suggests that respondents were both high idealists and relativists. As shown in Table 10, the percentages of respondents who scored excess 5 were very high (ranging from 84.62% to 100% for idealism and from 65 .15% to 76.92% for relativism); this is convincing evidence that these students were oriented towards high idealism and relativism. The comparison of studies shows that students 37 MBA Project - Kui-Ying Lin of School-A-lnstructor-1 scored the highest in all samples while those of Schooi-BInstructor-2 scored the lowest (see Table 11 ). However, overall responses of present research for relativism were lower than Forsyth ' s (1980) findings but higher than the study results of Barnett, Bass, and Brown (1994) and Bass, Barnett, and Brown (1999). For correlation analyses, Cooper and Schindler (2008, 549-50) argue that collinearity (where two predictor variables are highly correlated) can have damaging effects on multiple regression in terms of estimating and interpreting the regression coefficients. Johnson and Wichern (1997, 636-39) assert that collinearity or multicollinearity makes explanatory variables unstable and cause the presence of redundant infonnation among the predictor variables. According to Cooper and Schindler (2008, 550), if a correlation is equal to or greater than 0.80 or if a variable inflation factor (VIF) index is equal to or greater than 10, there is a collinearity problem for the explanatory variables. In order to avoid this problem, I conducted correlation analyses of idealism, relativism, operating, accounting, and total manipulations and examined the VIF indices in all models. The results are reported in Table 12. The findings indicate that all correlation coefficients were less than 0.35 and all VIF indices were less than 1.2 and that multicollinearity was not a problem for the variables in the regressions. Therefore both idealism and relativism were included in the final testing ofHl and H2. Separate regression analyses were adopted to determine whether personal moral philosophies affect earnings management (EM) ethics. The results of the hypotheses tests involving the dependent variables (the mean response on operating, accounting, total manipulations, and each of the 18 EM scenarios from all four groups) and independent variables (idealism and relativism) are summarized in Table 13 and 14. 38 MBA Project - Kui-Ying Lin The regression analyses (see Table 13) reveals that idealism was a significant positive coefficient in operating, accounting, total manipulations, and twelve of eighteen EM scenarios. The direction is consistent with hypothesis 1 that high, as compared to low, idealists will consider EM practices more unethical. Although idealism was a significant positive determinant of EM ethics in a considerable number of EM scenarios, it was not supported in six of eighteen EM scenarios. Thus, hypothesis 1 is partially supported. This partial support may indicate that some individuals with moral absolute beliefs do not appear to directly influence EM ethics. In addition, the significant negative coefficient for relativism in accounting manipulations and eight of eighteen EM scenarios (see Table 14) suggests that high relativists regard EM practices as less unethical than low relativists. Nevertheless, H2 is partially supported because it was not supported in operating, total manipulations, and ten of eighteen EM scenarios. This result indicates that, in some situations, relativistic individuals may tend to perceive the role of ethics and judge EM ethics as more unacceptable. Following the research of Elias (2002), median scores of idealism and relativism were calculated and subjects were classified into one of the four ideologies based on their scores relative to the medians. An Analysis of Variance (AN OVA) was then adopted to analyze the differences in the respondents' mean of these four groups on operating, accounting, and total manipulations. The ANOVA results are presented in Table 15. For three of four groups (School-A-lnstructor-1 , School-A-Instructor-2, and School-B-Instructor-2), there was no significant differences among these four ideologies with respect to EM ethics. For the School-B-lnstructor-1 of respondents, the personal ethical ideologies of individuals were significantly associated with differences in their judgements about EM ethics. The mean 39 MBA Project - Kui-Ying Lin scores of each ideology were compared to determine these four ideologies of respondents judging EM practices more or less harshly. Overall, absolutists had the highest response means and were the harshest judges of EM ethics, while subjectivists had the lowest mean responses, were the most lenient judges of EM practices, and considered EM actions more ethical. The harshness of these four ideologies' judgments in EM ethics was sequenced first absolutists, situationists, then exceptionists, and lastly subjectivists. These findings are consistent with previous research. Elias (2002, 40) found that absolutists viewed EM practices as the most unethical and subjectivists considered EM activities the most ethical. Machiavellianism The mean scores, standard deviations, and mean differences for Machiavellian responses of four samples are reported in Table 16. According to Christie and Geis (1970, 27), the score of 100 on Mach IV is represented as the theoretical neutral point which is neither agreeing nor disagreeing with Machiavelli. This theoretical neutral point of 100 was used to evaluate the extent of respondents ' Machiavellian orientations. As summarized in Table 16, the positive values of mean differences for Machiavellianism in three samples (School-A-lnstructor-1, School-A-lnstructor-2, and School-B-lnstructor-1) suggest that respondents from these three samples were high in Machiavellianism while the negative value of mean difference in the sample of School-B-lnstructor-2 indicates that respondents from this sample were low in Machiavellianism. Table 17 shows that the aforementioned three samples had high percentages (ranging from 54.55% to 60.71 %) of respondents scoring over 100 but the sample ofSchool-B-lnstructor-2 had low percentage (42.31%) of responses excess 100. These data suggest that students from the aforementioned three samples were high Machiavellian while students from School-B-lnstructor-2 were low Machiavellian. 40 MBA Project - Kui-Ying Lin However, Hunt and Chronko (1984, 36) found that a slightly over 10% of all marketers in their sample scored in excess of 100 on Mach IV scale. The high percentages, as compared to Hunt's and Chronko' s finding, of responses excess 100 suggest that these students were overall oriented to high Machiavellianism according to the interpretation of Christie and Geis (1970). In addition, comparing the respondents' mean scores of four samples with those of previous research indicates that these would-be accountants in British Columbia were more Machiavellian than other members of society and had mean scores ranked third, fourth , fifth and seventh in the gleanings of previous studies (see Table 18). This is also convincing evidence that these respondents were oriented toward high Machiavellianism. However, this result contradicts to Wakefield's (2008, 123) findings that accountants and Certified Public Accountants (CPA) were lower Machiavellian than other professionals. Nevertheless, the higher Mach IV scores may be due to their major in accounting and their student status. Wertheim, Widom, and Wortzel (1978, 237-38) found that management students were higher in Machiavellianism than social work students. Wakefield (2008, 117 -18) asserts that college students have significantly higher Mach scores than adults. Christie and Geis ( 1970) found that male subjects scored higher (mean = 93 .69) on Mach IV than female respondents (mean=87.66). Likewise, Zook and Sipps (1986, 132) revealed the same trend that males were more oriented to higher scores (mean= 93.83) on Mach IV than their female counterparts (mean = 86.88) in their studies. Conversely, Hunt and Chonko ( 1984, 3 7) found that female marketers were more Machiavellian than male marketers. Similarly, Mostafa (2007, 231) found that female respondents scored higher (mean = 107.97) on Mach IV scale than their male counterparts (mean = 104. 08). In order to 41 MBA Project - Kui- Yi n g Lin examine Machiavellian attributes, students from all four samples were classified into subgroups based on their age, gender, and working status. The ANOV A procedures were used to analyze the differences in the mean score of each pair of subgroups on Mach IV scale and the results are presented in Table 19. The results indicate that there was no significant difference in the mean scores between female and male respondents. This suggests that these students reflected on significant gender differences in scores on Mach IV scale. This result contradicts to the aforementioned studies but is consistent with the finding of Wertheim, Widom, and Wortzel (1978). They found that no significant gender difference in scores on Mach IV scale was presented in the result of surveying 348 first-year graduate students. Hunt and Chonko (1984, 36-37) found that younger marketers were more Machiavellian than their older counterparts. In surveying 512 university students majoring in marketing from an Egyptian university, Mostafa (2007) found that younger respondents (under 30) scored higher than older respondents (over 30). Nevertheless, this study found that there was no significant age difference in scores on Mach IV scale (see Table 19). This finding is inconsistent with previous research that younger subjects were more Machiavellian than their older counterparts. However, this study found that working or not did affect respondents ' Machiavellian scores. For the sample ofSchool-B-lnstructor-1 , there was a significant difference in the mean score between working and not-working groups on Mach IV scale (see Table 19). This result indicates that respondents who were not working scored higher on Mach IV scale than those who were working. This suggests that working made respondents less oriented towards Machiavellianism. In order to determine whether Machiavellianism had an impact on individuals' perceptions of EM ethics, the author conducted separate regression analyses for each mean 42 MBA Project - Kui-Ying Lin score of operating, accounting, total manipulations, and 18 EM scenarios as dependent variables and the mean score of Machiavellianism as independent variables. The regressions were also performed for all four groups separately. The results of testing hypothesis 3 in regression models are reported in Table 20 and 21. The significant negative regression coefficient for Machiavellianism suggests that respondents with high Machiavellianism judged EM activities as less unethical. This result meets my expectation but lends partial support for H3 because H3 was only supported in nine of eighteen EM scenarios. It was surprising that Machiavellianism was also significant with a positive coefficient value in many EM scenarios (especially the responses from the Schooi-B-Instructor-lsample); in other words, EM actions were inversely related to Machiavellianism (see Table 21). The positive coefficient for Machiavellianism implies that some high Machiavellian individuals may regard EM ethical issues as serious and judge EM activities more harshly. This finding is inconsistent with not only my expectation that H3 postulated a negative relationship between Machiavellianism and EM ethics but also previous research that Machiavellianism is related to unethical behaviour and has a negative effect on ethical judgments. This phenomenon needs to be further explored for the shift of Machiavellians ' ethical perceptions in business practices in the future. Locus of Control A standard response score designated by Hume and Smith (2006, 54) was used to assess whether the students of these two business schools were internals or externals. The percentages of internals and externals were calculated and the results are displayed in Table 22. The results show that the percentages (ranging from 50.91% to 63.64%) of externals were higher than those of internals in all four samples. This suggests that these students were 43 MBA Proj ect - Kui-Ying Lin oriented towards external LCs. In addition, the comparison of studies indicates that these would-be B.C. accounting practitioners had the highest locus of control (LC) scores in all members of society (see Table 23) and tended towards external LCs. Taken together, these respondents were overall oriented towards externality. These students had the same orientations as Rotter' s (1966, 15-16) expectation that college students were more external overall, especially compared with Peace Corps trainees who voluntarily sent overseas on assignment were more internally oriented. Also, students from School Al and A2 classes would possibly worry over their final exams which were to be held within two weeks from the time (at the end of November 2008) of their responses to the survey while students from School Bl and B2 classes could probably worry about their future careers or further studies after graduation in May 2009 which was about four months from their responses to the survey between January and February 2009. The anxieties could make these students have higher mean scores and reflect in the orientation towards external LCs. This is consistent with Rotter' s (1966, 16) explanation for Crowne and Conn's study results in 1965. Their subjects, students who were concerned with "reactions to acceptance and rejection for college admission" reflected in higher mean scores (9.56; the highest in the 1960's studies; see Table 23) and were oriented towards externality. Ware (1964; in Rotter 1966, 15) found that male subjects more frequently had an internal locus of control (LC) than their female counterparts (see Table 23). In order to assess the attributes of LC, the students of all four samples were divided into three pairs of subgroups according to their age, gender, and working status. The ANOV A processes were conducted to analyze the differences in the mean scores of each pair of subgroups on Internal-External (1-E) scale and the results are presented in Table 24. The results indicate 44 MBA Project - Ku i-Ying Lin that, for the sample of both instructors at School-A, there were significant differences in the mean scores between male and female groups. The finding shows that male respondents scored lower 1-E scale than female students and thus males, as compared to females, were more oriented towards internal LCs. This finding is consistent with Ware's (1964) findings that male respondents had the orientation of internal LCs. Skinner and Connell (1986: in Kliewer and Sandler 1992, 394-95) claim that children become increasingly internal in their LC beliefs as they age. However, this study found that there was no significant difference in the mean responses between younger (under 30) and older (over 30) groups (see Table 24). This suggests that there was no significant age difference in scores on 1-E scale. This result contradicts to the view of Kliewer and Sandler (1992) that older individuals are oriented towards internal LCs. In addition, this study found that working experience did affect orientations towards internal or external LCs. For the sample of School-B-Instructor-2, there was a significant difference in the mean scores between employment and unemployment groups (see Table 24). This implies that there was a significant working-experience difference in scores on the I-E scale; respondents with working experience scored lower on LCs and were more oriented towards internal LCs. The last step involved determining whether locus of control (LC) was related to earnings management (EM) ethics. Regression analyses were performed with the mean score on operating, accounting, total manipulations, and each of the 18 EM scenarios as dependent variables and that on LC as an independent variable. The regressions were conducted with each sample of respondents separately. The results of testing H4 in the regression analyses are reported in Table 25. There was no significant relationship between LC and operating, accounting, and total manipulations. However, locus of control was a significant positive 45 MBA Project - Kui-Ying Lin detenninant of EM activities in EM scenario 1, 8, and 12. The direction for LC is consistent with H4 and suggests that individuals with an internal LC will tend to judge EM actions as more unethical than those with an external LC. However, H4 is partially supported because LC was not associated with operating, accounting, total manipulations, and fifteen of eighteen EM scenarios. As shown in Table 25, the results are also inversely supported in the equally prevailing number of EM scenarios (question 5a, 13, and 15). This implies that respondents with an internal, as compared to an external, LC rated EM practices as less unethical. The presence of significantly positive and negative coefficients from equal number of EM scenarios makes whether LC affects EM ethics inconclusive. These findings surprised me and seemingly, individuals with an internal LC assume responsibility of the consequences of their actions and exhibit more ethical behaviour but they, in some situations, do not consider EM activities unethical. Future research may give us a clear picture and resolve the inconclusive finding. 46 MBA Project - Kui-Ying Lin DISCUSSION AND IMPLICATION Earnings Management Assessing the prevailing morality concerning earnings management activities first indicates that respondents in this study rated earnings management (EM) actions as more unethical than previous research (see Table 7). One possible explanation of this orientation is that an enhanced awareness of the bad consequence of the highly public accounting scandals and the prevailing ethical education at these two business schools made these students the most conservative in their judgments and judged EM actions as the harshest. Take ethical education for example: Applied Ethics, a mandatory course, at one of these two business schools is described that the purpose of the course is to foster students' abilities and values in resolving moral disagreements and making ethical decision at work (British Columbia Institute of Technology n.d.). Tang and Chen (2007, 16) found that ethics training enhanced students' ethical perceptions and reduce their propensity to engage in unethical behaviour. In addition, there was significant disagreement among respondents for most of the scenarios. In Schooi-B-lnstructor-2 of respondents, for example, the distribution of responses for scenario 10, regarding the selling of undervalued assets, was on the 1 through 5 scale 1, 7, 5, 6, and7 respectively (see Figure 4). These data show that about 30.77% [(1 +7)/26] of respondents thought this action was more acceptable; approximately 50% [(6+7)/26] considered it less acceptable, and about 19.23% [5/26] were not sure. These respondents had no unanimous agreement about where to draw the line between right and wrong. Despite this imperfection, there is consistency in ethical judgements among these respondents. 47 MBA Project - Ku i-Ying Lin All groups of respondents consistently judged scenario 6 (the activity of recording sales for customers who did not order anything) as a serious ethical infraction and scenario 1 (the activity of expensing maintenance of buildings early) as a very minor ethical infraction. Respondents were also in agreement about some characteristics of EM practices and their judgments were affected by the types of EM activities (operating versus accounting), the amount of action consequences (materiality versus immateriality), and the purpose of managing earnings (increasing versus decreasing profits). ln general, respondents rated EM practices much more unethically if the EM practices were accounting manipulations, the earnings amount of manipulations was large, and the purpose of manipulations was to increase profits than if those practices were operating manipulations, the manipulative amount was small, and the objective of managing earnings was to reduce profits. This empirical evidence would provide organizations or professions with future reference; they probably should incorporate some of these specific EM characteristics into their program design ofunethical EM prevention. The surprising discovery is that these students majoring in accounting did not, on average, take the consistency with the generally accepted accounting principles (GAAP) into consideration when making their ethical judgments. The possible explanation of this phenomenon is that these students were ignorant of the details ofGAAP or did not consider the use of GAAP as a defence for manipulators ' actions imperative, a fact that renders the findings statistically insignificant. Previously, the study ofGAAP was set out in the CICA Handbook as a necessary preparation for accounting students to pursue professional accounting as a career in Canada (Hilton and Herauf2008, 1). Globalization has made rapid changes including Canadian financial reporting standards. In 2006, the Canadian Institute of 48 MBA Project - Kui-Ying Lin Chartered Accountants (ClCA) announced that it would adopt the harmonization of the CJCA Handbook with International Financial Reporting Standards (IFRSs) and that all publicly trading companies will have to use the new standards effective 1 January 20ll(Hilton and Herauf2008, 12). During this transitional period between 2006 and 2011 , education has probably focused on the understanding of new accounting standards, the preparation of consolidated financial statements, and the implications of switching to IFRSs. This new focus of business education could make these accounting students disregard their attention to GAAP as important consideration of assessing ethical or unethical EM activities. Idealism and Relativism The linkage observed between idealism and earnings management (EM) ethics was that highly idealistic students regarded EM activities as more unethical. This suggests that high idealists seek the best outcome for all and rate EM actions less ethical. This partially supported relationship is consistent with the conclusion of earlier studies that found similar relationship among students concerning marketing issues (Barnett, Bass, and Brown 1994; Bass, Barnett, and Brown 1999) and among accounting practitioners, faculty, and students concerning EM ethics (Elias 2002). In addition, the empirical evidence partially supports the premise that high relativists will judge earnings management activities less harshly than low relativists. This incompletely supported relationship is also consistent with previous studies that high relativists rated EM practices as less unethical (Elias 2002) and that relativism was a significant negative determinant of behavioural intentions in their second scenario (Vitell et a!. 2003, 159-164). Taken together, idealism and relativism are two significant antecedents of EM ethics and paying more attention to personal moral philosophies may help reduce unethical EM activities. 49 MBA Project- Kui-Ying Lin The responses from the School-A-Jnstructor-1 sample contributed no support to H1 and H2. The comparison of sample statistics among groups (see Table 9) and previous research (see Table 11) indicates that the respondents of this group were high idealists due to their high mean score and that they agreed in their judgments of EM ethics because of the lowest standard deviation for idealism and the smallest range of responses. However, it was surprising that the high idealism and the agreement in judgments of EM ethics lent no support to Hl. Seemingly some high idealists believe that ethical actions always produce positive outcomes and avoid harming others but their beliefs of moral absolutes may not directly influence EM ethics. Equally, the agreement in judgments (due to the low standard deviation for relativism and the small range of responses) on EM ethics provided no support for H2. Ostensibly, some relativists believe that moral actions depend on situations and not on moral absolutes but their relativistic ethical perspective may not affect ethical judgments concerning EM practices. Yitell et al. (2003, 164) argue that high moral intensity can dominate individuals' personal moral perspective and make idealism and relativism insignificant in determining ethical judgments. The respondents of this group regarded larger EM actions (Sa and 16a) as more unethical than the smaller EM activities (Sb and 16b) (see the last four lines in Table 8). The effect of moral intensity could have attributed no support to both H1 and H2. The equivalent unproductive group is the Schoo1-B-Jnstructor-2 of respondents. Their responses generated support for H1 only in EM scenario 13. The data show that the low Cronbach's alpha (0.34; see Table 4) of idealism created internal inconsistency which produced random and unstable error and might attribute little support to Hl. Jn addition, the comparison of their sample statistics with other groups (see Table 9) and previous studies 50 MBA Project - Kui-Ying Lin (see Table 11) shows that the respondents of this group were low idealists due to their low sample mean and that they disagree in their judgments on EM ethics because of their high standard deviation for idealism and large range of responses. The low idealism and the disagreement in rating EM ethics could also be a reason why this group provided fruitless support for H 1. For three of eighteen EM scenarios (question 2, 6, and 10), ethical judgements on EM activities were not dependent on personal moral philosophies. One possible explanation is that some of these scenarios were judged as more unethical than others. Scenario 6, for example, was rated the most unethical of all scenarios (see Table 5) but its regression results provide no statistically significant coefficient for idealism or relativism (see Table 13 and 14). This finding is inconsistent with the assertion of Barnett, Bass and Brown ( 1994, 4 76) who claim the reason that the twelve of their twenty-six scenarios were not dependent on ethical ideologies is that they were judged less unethical than others. 2 The coefficients of determination (R value; ranging from 5% to 30%; see Table 13 and 14) for the regression models used in the hypotheses tests may be considered low; thus only a small portion of the variation in the dependent variables is explained by predictors specified. Although this range seems too low, it is about the same range as previous studies. In the four multiple regression models ofSinghapakdi and Vitell (1990, 13-14), the coefficients of determination were ranging from 5.6% to 34%. In the study of Hegarty and Sims (1978, 455), only twenty-five percent of the dependent variable was explained by their nine independent variables. 51 MBA Project - Kui-Ying Lin Machiavellianism The high-Mach, as compared to low-Mach, percentages of respondents who scored over 100 points (see Table 17) and the high bound of sample means in the comparison of this research with previous studies (see Table 18) are evidence that these students were high Machiavellian. Gable and Dangello (1994, 160-61) claim that high Machs are oriented towards success in higher levels of job involvement. Kover (1975, 86), argues that the high Machiavelli has high self-esteem and great self-confidence and that many management consultants have the characteristics of high Machiavellianism; therefore, he established and ran seminars to teach the practice of Machiavellian techniques to MBA students. Jones, Nickel, and Schmidt (1979) found, in their three separate experiments, that subjects with high Mach scores consistently recognized instrumental contingency situations and responded with ways that tended to maximize gains under cooperation and competition instructions, "a rational, functional, task-oriented approach" that strives for success and advantages of high Machiavellianism in life. McHoskey (1999, 279) found that a participant with a high Mach score had a "control-oriented motivation that is manifested in aspirations for financial success", a primary goal for business students. From these findings, would-be accountants at these two business schools can expect to greatly succeed in their future careers and be fabulously wealthy in their future life. The partially supported relationship between Machiavellianism and earnings management (EM) ethics (see Table 20) appears consistent with previous studies that Machiavellianism was connected to unethical behaviour (Hegarty ad Sims 1978, 1979) and supports the expectation that individuals with high Machiavellianism will reject moral decision rules (Bass, Barnett, and Brown 1999, 200). Singhapakdi and Vitell (1990, 13) 52 MBA Proj ect - Kui-Ying Lin found that Machiavellianism had a negative effect on ethical behaviour and conclude that individuals with high Machiavellianism have the tendency to perceive ethical problems as being less serious than those with low Machiavellianism. The explanation in this regard is that individuals with high Machiavellianism are not lacking morals but may behave under a set of instrumentalist ethical guidelines which is inconsistent with conventional morality (Christie and Geis 1990; Bass, Barnett, and Brown 1999). An intriguing finding is a significantly positive coefficient for Machiavellianism in accounting, total manipulations, and ten of eighteen EM scenarios (see Table 21). This reverse linkage between Machiavellianism and EM ethics supported in numerous EM scenarios suggests that high, as compared to low, Machiavellian students considered EM practices more unethical. Implicitly, these students had the characteristics of not only high Machiavellianism but also high ethical perspectives. Initially, this may seem counter-intuitive. Why were respondents oriented towards high Machiavellianism and also highly oriented to morality? One explanation may be an enhanced awareness of public wrong doing - a consequence of the highly public accounting scandals of late. Unveiling high-profile bankruptcies of Enron, WorldCom, and Arthur Andersen have raised the public awareness of EM ethics. This keen awareness may have resulted in the incorporation of a fuller examination of ethical issues by Canadian accounting educators into their classes. Business Ethics is now a standard course for students majoring in Business at many universities. Applied Ethics, for example, is a mandatory course for students majoring in accounting at one of these two business schools; the course intends to foster students' moral values and perceptions of resolving ethical dilemmas and making ethical decisions (British Columbia Institute of Technology n.d.). Barlas et al. (2002, 24) depict that ethics education has 53 MBA Project - Kui -Ying L in integrated with accounting curriculum and with most textbooks providing all possible opportunities to incorporate ethics into accounting programs. An auditing textbook of one of these two business schools, for example, includes the topics such as professional ethics and auditor responsibilities (Smieliauskas and Bewley 2006); in other words, these students have been taught professional ethics in their Auditing classes. In addition, understanding the ethical, legal, and professional environment has been one of their course objectives. Tang and Chen (2007, 16) found that ethical training helped students significantly change their ethical perceptions and reduce their unethical practices. As a result, one possible explanation of these respondents ' unique characteristics is that the success of ethical education in both business schools had cultivated their Auditing students to be highly moral while maintaining their orientation towards Machiavellianism, an effect that renders the reverse relationship between Machiavellianism and EM ethics possible and sustainable. These high Machiavellian respondents did agree with their judgments about EM ethics because the standard deviations for Machiavellianism of all four samples were lower than those of previous studies (see Table 18). The agreement with judgments among respondents and the prevailing support from many models make this reverse relationship not be overlooked. In other words, this reverse relationship should be treated in more attentive although the finding contradicts not only H3 but also previous research. Previous studies found that individuals with high Machiavellianism were more likely to approve of unethical decisions and less likely to behave ethically than those with low Machiavellianism. The reverse relationship between Machiavellianism and EM ethics can be interpreted as evidence that these students had unique personal characteristics with the orientation of both high Machiavellianism and high morality. This discovery may mitigate the assumption that the 54 MBA Proj ect - Kui-Ying Lin Machiavellian image is solely negative and provide model designers with future reference for formulating their hypotheses. In short, the requirements for causal inference that Machiavellianism was a significant negative determinant of EM ethics were not met due to the presence of a reverse relationship between Machiavellianism and EM ethics. As a result, the direction of causal arrows in the models should be regarded as suggestive and disregarded as an absolute indication of causality. Locus of Control The empirical evidence partially supports the hypothesis that individuals with an internal locus of control (LC) will judge earnings management (EM) activities more harshly than those with an external LC. This incompletely supported result lends support to Elias' (2002) prediction that locus of control can be one of important determinants of EM ethics and is consistent with Trevino ' s and Youngblood ' s (1990, 382) findings that respondents with an internal LC showed more ethical behaviour than those with an external LC. However, the presence of a reverse relationship can weaken the conclusion of the hypothesis supported from EM scenarios. Data may provide some possible explanations of the phenomenon. The high percentages of externals (see Table 22) and the highest LC scores in all members of society (see Table 23) were convincing evidence that respondents were oriented towards externality. Phares and Lamiell (1975) argue that externals, as compared to internals, are more likely to provide others with understanding and sympathy and view others as needing assistance or help. The orientation of an external locus of control implies that these students are more philanthropic and benevolent in ethical orientation. 55 - - - - - - - ---- --~· . - -- MBA Project - Kui-Ying L in The external orientation can lead to change the path direction of the causal relationship with a possible premise that individuals with an external locus of control (LC) consider EM practices less unethical. However, the inverse supported result is so weak that it is not strong enough to revise the model path. Hume and Smith (2006, 54) designated their subjects ' scores less than or equal to ten on the Internal-External scale as internals and those greater than or equal to eleven as externals and used this LC designation as independent variables in their analyses. This LC designation can be used to classify subjects into internals and externals in the analyses and help resolve this ambiguous relationship in future research. 56 MBA Project - Kui-Ying Lin LIMITATIONS AND FUTURE RESEARCH DIRECTIONS This study was intended to explore empirically the earnings management (EM) issue, which this author considers the most important ethical issue facing the accounting profession. Although this project has provided some preliminary evidence for several determinants of EM ethics, the finding should be treated as tentative because the hypotheses were incompletely supported by all models and the study had a number of limitations. First, the short scenario descriptions had possibly caused the variation in responses as each of respondents made his/her distinctive assumption about some of the unmentioned contextual information. In addition, the scenarios assessed by respondents in this project reflect only a portion of the many fonns of EM practices. This study did not include some other EM actions such as the changes in accounting policies or the sale of assets to its own subsidiaries to boost profits as Enron did (Kim and Nofsinger 2007, 34). Including some of these highprofile EM practices or other additional scenarios may result in different cognitive patterns and in different findings . Future researchers conducting similar studies may find it beneficial to provide participants with more detailed and diversified EM scenarios and make findings more relevant to real situations. In addition, the instruments that required about 20 to 30 minutes of students' time to complete may have caused respondents not to answer carefully or attempt to finish them as quickly as possible by agreeing indiscriminately or simply selecting the same answers to many questions. This possible phenomenon could have raised cause for concern about response sets which refer to "a tendency on the part of individuals to respond to attitude statements for reasons other than the content of the statements" (Robinson, Shaver, and 57 MBA Project - Kui-Ying Lin Wrightsman ( 1991, 7). This might be one of the reasons that the data lent inadequate support to the hypotheses in the EM scenarios or generated indetenninants of the EM ethics for some scenarios. The timing of responding to the questionnaires may also have affected the responses. Students at both School-A-Instructor-! and Schooi-A-lnstructor-2 classes, for example, responded to the survey at the end of November 2008 and their final exams were scheduled on 12 December 2008 . Their intense anxieties over final exams could have affected their responses to the survey. The different manner in which the survey was conducted by the different instructors might have influenced the responses as well. The respondents of the Schooi-B-lnstructor1sample completed the questionnaires during the class while students from the rest of the groups completed the questionnaires outside of the class time. The different outcome is that School-B-lnstructor-1 of respondents contributed to the highest response rate (94.29%; see Table 1) and the highest Cronbach' s alpha coefficient (ranging from 0.64 to 0.75 ; see Table 4) as compared to the rest of the groups. The high response rate and internal consistency might be due to the less time pressure on students to respond to the survey and thus generated better data and truer responses. There might be some other response biases, such as rating-scale and question-order biases. Take the rating-scale bias for example: the Ethical Position Questionnaires (EPQ) used nine-point scale ranging from one (completely disagree) to nine (completely agree) while the earnings management (EM) scenarios questionnaire used five-point scale ranging from one (completely ethical) to five (completely unethical). These various scale points and different rating directions could make respondents confused in terms of selecting their 58 MBA Proj ect - Kui-Ying Lin answers. This type of response biases can be prevented in future research if the consistency of scale points and rating directions is incorporated into the design of survey instruments. The next limitation is that the sampling was opportunistic. The support of two business schools and four instructors provided me with a great opportunity to obtain useful samples from their students. However, these respondents could well reflect the attitude of students only at these two business schools but might not reflect the perspectives of would-be B.C. accountants in general. Students from the Chartered Accountants School of Business (CASB) or the Certified Management Accountants (CMAs) of B.C. might have different opinions on judging earnings management (EM) ethics. Future research should incorporate samples of accountants and/or accounting students from the CASB or CMAs of B.C. and include more male participants to avoid skewing toward females (one of the limitations in present research is that females accounted for 65.9% of respondents; see Table 2). It would be interesting to see whether the causal relationship between four predictors (idealism, relativism, Machiavellianism, and locus of control) and EM ethics still holds up and whether participants from different samples have unique characteristics (both high Machiavellianism and high morality) as current respondents had. The use of students as respondents might be of concern because students might have yet to fully develop their moral cognitions (Barnett, Bass, and Brown 1994, 476). Kohlberg (1983; in Barnett, Bass, and Brown 1994, 476) asserts that the maturity of individuals' values pass through different stages of moral development and while Kohl berg' s model asserts that students of this age should all be operating as mature, morally developed individuals, they may be variability here. If graduated and certified accountants were used as respondents, the findings of this study might not be sustained. Consequently, the interpretation of the results 59 MBA Project - Kui-Ying Lin should be carefully constrained. However, Barnett, Bass, and Brown (1994, 476) claim that student samples are useful in "research areas where little prior research has been done" in terms of helping refine constructs and models and fonnulate hypotheses to test with different samples later. Implicitly, these students provided preliminarily opportunistic research into EM ethics in Canada and have made great contributions to the progress of academic research into this topic. The results of this study could help researchers refine their survey instruments and formulate their hypotheses if they research into Canadian EM ethics in the future. Barnett, Bass, and Brown (1994, 476) claim that individuals ' ethical ideologies are important predictors in models of indi vidual ethical decision making. The classification of idealism and relativism into situationism, absolutism, subjectivism, and exceptionism has been widely conducted and tested the difference among these four ethical ideologies (e.g. Forsyth and Nye 1990; Forsyth 1992; Barnett, Bass, and Brown 1994; Elias 2002). One potential avenue for future research is to examine possible causal relationships between these four ethical ideologies and EM ethics. Hegarty and Sims ( 1978, 1979) and Trevino and Youngblood (1990) found that locus of control (LC) was a significant predictor of unethical behaviour. However, this research found that LC was a significant determinant of EM ethics in both positive and negative directions. This ambiguous relationship makes the conclusion of testing H4 inconclusive. This inconclusive result indicates the needs for fuller and further exploration whether LC influences EM ethics. For future research, there may be more fruitful results if a longitudinal study with a pre-test and post-test approach is used and/or if the LC designation of Hume and Smith (2006, 54) is used to divide responses into internals and externals which serve as independent variables in the analyses. 60 MBA Project - Kui-Yin g Lin In short, previous studies have only explored American EM ethics. This study extends prior research into Canada and examined four possible antecedents (idealism, relativism, Machiavellianism, and Locus of control) of B.C. would-be accountants' ethical judgments in EM practices. Although the four determinants only partially impacted EM ethics, the results shed some light on EM ethics in B. C. and provide a potential value of further efforts to better understand Canadian EM ethics. It is hoped that this project will foster an even greater interest in the study of EM ethics in Canada. 61 MBA Pr oj ect - Kui-Ying Lin REFERENCES Alexander, Cheryl S. and Henry Jay Becher. 1978. The use of vignettes in survey research. Public Opinion Quarterly 42. no. 1: 93-104. Barlas, Stephen, Bud Fennema, Dan Lacknkr, Roland L. Madison, Tim Query, Lance Thompson, Curt Yerschoor, and Kathy Willliams. 2002. Is failure to teach ethics the casual factor? Strategic Finance 84. no. 6: 24-64. Barnett, Tim, Kenneth Bass, and Gene Brown. 1994. Ethical ideology and ethical judgment regarding ethical issues in business. Journal ofBusiness Ethics 13. no. 6: 469-80. Barton, Jan 2001. Does the use of financial derivatives affect earnings management decisions? The Accounting Review 76. no. 1:1-26. Bass, Kenneth, Tim Barnett, and Gene Brown. 1999. Individual difference variables, ethical judgments and ethical behavioural intentions. Business Ethics Quarterly 9. no. 2: 183205. Beckham, Crystal M. , Beverly J. Spray, and Christina A Pietz. 2007. Jurors' locus of control and defendants' attractiveness in death penalty sentencing. Journal of Social Psychology 147. no.3: 285-298. Beu, Danielle. 2001. Accountability as it influences ethical behaviour. PhD diss., The University of Oklahoma. In ProQuest Dissertations and Theses, http//proquest.umi. corn/pqdweb?did= 7281 06321 &sid=2&Fmt=2&clientld=6993&RQT =309&VName=PQD [accessed August 15, 2008]. Bitner, Larry N. and Robert Dolan. 1998. Does smoothing earnings add value? Management Accounting 80. no.4: 44-47. Bowie, N. and R. Duska. 1990. Business Ethics. 2"d ed. Englewood Cliffs, NJ: Prentice-Hall. Bochner, Steven E. and Douglas J. Clark. 2008. When is managing to a forecast illegal "earnings management"? Insights 22. no. 5: 2-6. British Columbia Institute of Technology. (n.d). Full-Time/Part-Time Bachelor of Technology. http://www.bcit.ca/study/programs/8400btech#PGM CONTENT [accessed April 23, 2009]. Christie, Richard and Florence Geis et al. eds. 1970. Studies in Machiavellianism. New York: Academic Press. Codevilla, Angelo M. 1997. Editor's Introduction to Th e Prince by Niccolo Machiavelli. New Haven & London: Yale University Press. 62 MBA Project - Kui-Ying Lin Cohen, Jeffrey R. and Laurie W. Pant. 1995. An exploratory examination of international differences in auditors ' ethical perceptions. Behavioural Research in Accounting 7:3754. Colvin, Geoffrey. 1999. The ultimate manager. Fortune 140. no.10 : 185-87. Cooper, Donald R. and Ramela S. Schindler. 2008 . Business Research Methods. The McGraw-Hill Companies, Inc. 369-71. Davidson, Wallace N. , Pornsit Jiraporn, Young Sang Kim, and Carol Nemec. 2004. Earnings management following duality-creating succession: Ethnostatistics, impression management and agent theory. Academy of Management Journa/47. no. 2: 267-75. Deluga, Ronald J. 2001. American presidential Machiavellianism: Implications for charismatic leadership and rated performance. Leadership Quarterly 12. no.3 :339-363. Dechow, Patricia M., Sloan, and Sweeney. 1996. Causes and consequences of earnings manipulation : An analysis of firm subject to enforcement actions by the SEC . Contemporary Accounting Research 13 . no. I: 1-36. Dion, P. A. and P. M. Banting. 1988. Industrial supplier-buyer negotiations. industrial Marketing Management 17: 43-47. Dobson, John. 1999. Is shareholder wealth maximization immoral? Financial Analysts Journal 55. no. 5: 69-75. Ducette, Joseph, Stephen Wolk, and Sarah Friedman. 1972. Locus of control and creativity in black and white children. Journal of Social Psychology 88. no. 2: 297-98. Duncan, James R. 2001. Twenty pressures to manage earnings. Th e CPA Journal7l. no. 7: 32-37. Dunmore, Paul. 3008. Earnings management: Good, bad, or downright ugly? Chartered Accountants Journal87 . no. 3: 32-37. Dye, Ronald A. 1988. Earnings management in an overlapping generation ' s model. Journal of accounting Research 26. no. 2: 195-235. Elias, Rafik Z. 2002. Determinants of earnings management ethics among accountants. Journal of Business Ethics 40. no. 1: 33-45. Elias, Rafik Z. 2004. The impact of corporate ethical values on perceptions of earnings management. Managerial Auditing Journa/19 . no.1 :84-98. 63 MBA Project- Kui-Ying Li n Fischer, Marilyn and Kenneth Rosenzweig. 1995. Attitudes of students and accounting practitioners concerning the ethical acceptability of earnings management. Journal of Business Ethics 14. no.6: 433-44. Ferrell, 0. C . and Larry G. Gresham. 1985 . A contingency framework for understanding ethical decision making in Marketing. Journal of Marketing 49: 87-96. Fogarty, Timothy, MichelL Magnan, Garen Markarian, and Serge Bohdjalian. 2009. Inside agency: The rise and fall ofNortel. Journal of Business Ethics 84. no.2:165-187. Forsyth, Donelson R. 1980. A taxonomy of ethical ideologies. Journal of Personality and Social Psychology 39. no. 1: 175-184. Forsyth, Donelson R. 1985. Individual differences in information integration during moral judgment. Journal of Personality and Social Psychology 49. no.1 : 264-272. Forsyth, Donelson R. 1992. Judging the morality of business practices: The influence of personal moral philosophies. Journal ofBusiness Ethics 11. no. 5/6: 461-470. Forsyth Donelson R . and Judith L. Nye. 1990. Personal moral philosophies and moral choice. Journal of Research in Personality 24. no.4: 398-414. Forsyth, Donelson R. and William R. Pope. 1984. Ethical ideology and judgments of social psychological research: Multidimensional analysis. Journal of Personality and Social Psychology 46. no. 6: 1365-75. Franklin, R. D. 1963. Youth's expectations about internal versus external control of reinforcement related to N variables. Unpublished doctoral dissertation, Purdue University. Gable, Myron and Frank Dangello. 1994. Job Involvement, Machiavellianism and job performance. Journal ofBusiness and Psychology 9. no.2: 159-70. Geis, F. L. and Tae H. Moon. 1981. Machiavellianism and deception. Journal of Personality and Social Psychology 41. no . 4: 766-75. Greenfield, Alfred. 2005. Introduction an ethical dimension into the earnings management decision. PhD diss ., Virginia Commonwealth University. In ProQuest Dissertations and Theses, http//rsvpn .ubc.ca/http/proquest.umi .corn/pqdweb?index=O&did=921 030771 &SrchMod. [accessed August 15, 2008]. Gore, Pearl Mayo and Julian. B. Rotter. 1963. A personality correlate of social action. Journal of Personality 31. no. 1: 58-64. 64 MBA Project - Kui-Yi n g Lin Hargis, Michael B. 2006. A pxs perspective on perceptions of corporate transgressions: The influence of personality and account giving. PhD diss., Wayne State University. In Dissertations and theses, http//proquest.umi.com/pqdweb?did= 1155569931&sid= 1&Fmt =2&clientld=6993&RQT=309&VName=PQD [accessed August 15, 2008]. Healy, Paul M. and James M. Wahlen. 1999. A review of the earnings management literature and its implications for standard setting. Accounting Horizons 13. no. 4: 365-83. Hegarty, W. Harvey and Henry P. Sims. 1978. Some determinants of unethical decision behaviour: A experiment. Journal of Applied Psychology 63. no. 4: 451-57. ---. 1979. Organizational philosophy, polices, and objectives related to unethical decision behaviour: A laboratory experiment. Journal of applied Psychology 64. no. 3: 331-38. Hilton, Murray W. and Darrell Herauf. 2008. Modern advanced accounting in Canada 51h ed. Canada: McGraw-Hill Ryerson Limited. House, Robert J. and Jane M. Howell. 1992. Personality and charismatic leadership. The Leadership Quarterly 3. no.2 :81-108. Howard, Judith A. and Randy Levinson. 1985. The overdue courtship of attribution and labelling. Social Psy chology Quarterly 48. no.3: 191-202. Hume, Evelyn C. and Aileen Smith. 2006. University student ethics: The differential explanatory effect of locus of control. Academy of Educational Leadership Journal 10. no.3: 49-58. Hunt, Shelby D. and Lawrence B. Chonko. 1984. Marketing and Machiavellianism. Journal of Marketing 48. no. 3: 30-42. Hunt, Shelby D. and Scott Vitell. 1986. A general theory of marketing ethics. Journal of Macromarketing 6. no. !: 5-16. Jackson, Willie Mae D. 1997. The relationship of parenting in responsibility to a school-age child's locus of control. PhD diss., The University of Alabama at Birminghan. In ProQuest Dissertations and Theses, http//proquest.urni.com/pqdweb?did=736668981 &sid=3&Fmt=2&clientld=6993&RQT=309&VName=PQD [accessed August 15, 2008]. Johnson, Richard A. and Dean W. Wichern. 1997. Business statistics: Decision making with .data. The United States of America: John Wiley & Sons, Inc. Jones, Thomas M. 1991. Ethical decision making by individuals in organizations: An issuecontingent model. Academy of Management Review 16. no. 2: 366-95. 65 MBA Projec t - Kui-Ying Lin Jones, Warren H ., Ted W. Nickel, and Alan Schmidt. 1979. Machiavellianism and selfdisclosure. Journal of Psychology 102. no.1: 33-41. Kahn, Jeremy. 2001. Accounting in wonderland. Fortune 143. no.6: 134-37. Kaplan, Steven E.2001a. Further evidence on the ethics of managing earnings: an examination of the ethically related judgments of shareholders and non-shareholders. Journal of Accounting and Public Policy 20. 27-44. Kaplan, Steven E. 2001 b. Ethically related judgments by observers of earnings management. Journal of Business Ethics 32. no . 4: 285-298. Kaplan, Steven E., James C. McElroy, Susan P. Ravenscroft, and Charles B. Shrader. 2007 . Moral judgment and causal attributions: Consequences of engaging in earnings management. Journal of Business Ethics 74: 149-164. Ketz, J. Edward. 2002. Can we prevent future Enron's? The Journal of Corporate Accounting and Finance 13. no. 4 : 3-10. Kim, Kenneth A. and John R . Nofsinger. 2007. Corporate Governance. 2nd ed. New Jersey: Pearson Prentice Hall. Klierwer, Wendy and Irwin N. Sandler. 1992. Locus of control and self-esteem as moderators of stressor-symptom relations in children and adolescents. Journal of Abnormal Child Psychology 20. no. 4: 393-413. Kover, Arthur J. 1975 . Machiavellian tactics forB-school students. Business Week. [October 13]. industrial Edition. Ladwig, G. W. 1963 . Personal, situational and social determinants of preference for delayed reinforcement. Unpublished doctoral dissertation, Ohio State University. Leary, Mark R., Paul D. Knight, Byron D. Byron. 1986. Ethical ideologies ofthe Machiavellian. Personality And Social Psychology Bulletin 12. no.1: 75-80. Loomis, Carol J. 1999. Lies, damned lies, and managed earnings. Fortune 140. no.3:74-83. Levitt, Arthur. 1998. The 'numbers game'. CPA Journal68. no. 12:14-19. McHoskey, John W. 1999. Machiavellianism, intrinsic versus extrinsic goals, and social interest: A self-determination theory analysis. Motivation & Emotion 23. no. 4: 267-83. Meek, Gary K. Ramesh P. Rao and Christopher J. Skousen. 2007. Evidence on factors affecting the relationship between CEO stock option compensation and earnings management. Review ofAccounting & Finance 6. no. 3: 304-323. 66 MBA Project - K ui-Ying Lin Merchant, Kenneth A. 1985. Organizational controls and discretionary program decisionmaking: A field study. Accounting, Organizations and SocietylO. no.l: 67-85. Merchant, Kenneth A. 1989. Rewarding results: Motivating profit center managers. Harvard Business School Press. Boston, Massachusetts. 161-195. Merchant, Kenneth A. and Joanne Rockness. 1994. The ethics of managing earnings: An empirical investigation. Journal ofAccounting and Public Policy 13. no. 1: 79-94. Mostafa, Mohamed M. 2007. A study of Machiavellian orientation among marketing students in Egypt. College Student Journal41. no.l: 223-238 . Nelson, Mark W. , John A . Elliott, and Robin L. Tarpley. 2003. How are earnings managed? Examples from auditors. Accounting Horizons 17. Supplement. 17-35. Noreen, Eric. 1988. The economics of ethics: A new perspective on agency theory. Accounting, Organizations and Society 13 . no.4: 359-369. Nunnally, J.C. 1967. Psychometric theo1y. New York: McGraw-Hall Parfet, William U. 2000. Accounting subjectivity and earnings management: A preparer perspective. Accounting Horizons 14. no.4: 481-88. Phares, E. Jerry and James T. Lamie!!. 1975. Internal-external, interpersonal judgments of others in need, and attribution of responsibility. Journal of Personality 43. no.! :23-38. Phillips, Matt. 2007. The afternoon report: Dropped call; Online edition. Wall Street Journal Eastern Edition. New York. Poff, Deborah C. ed. 2005. Business ethics in Canada . Toronto, Ontario: Pearson Education Canada Inc. Rallapalli, Kumar C. , Scott J. Vi tell, and James H. Barnes. 1998. The influence of nonns on ethical judgments and intentions: An empirical study of marketing professionals. Journal of Business Research 43. no . 3: 157-168. Rawwas, M., S. Vitell, and J. A. Al-Khatib. 1994. Consumer ethics: The possible effects of terrorism and civil unrest on the ethical values of consumers. Journal of Business Ethics 13. 223-232. Rest, James R. 1986. Moral development advance in research and theory. New York: Praeger Robinson, John P. and Phillip R. Shaver. 1973 . Measures of social psychological attitudes. Ann Arbor, MI: Institute for Social Research. 67 M BA Proj ect - Kui-Ying L i n Robinson, John P., Phillip R . Shaver, and Lawrence S. Wrightsman. 1991. Criteria for scale selection and evaluation. Measures of Personality and Social Psychological Attitudes. Academic Press, Inc. Rotter, Julian B. 1966. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied 80. no. 1: 1-28. Schilit, Howard. 2002. Financial Shenanigans. 2nd ed. McGraw-Hill. Schipper, Katherine. 1989. Earnings management. Accounting Horizons 3. no. 4: 91-102. Schlenker, Barry R . and Donelson R. Forsyth. 1977. On the ethics of psychological research. Journal of Experimental Social Psychology 13. 369-96. Schultz, Jr, Joseph J., Douglas A. Johnson, Deigan Morris, and Sverre Dyrnes. 1993. Journal of Accounting Research 31. no . 3 Supplement: 75-103. Scott, William R. 2006. Financial Accounting Theory. 4 1h ed. Toronto: Pearson Education Canada Inc. Shafer, William E. 2002. Effects of materiality, risk, and ethical perceptions on fraudulent reporting by financial executives. Journal of Business Ethics 38. no .3: 243-262. Shafer, William E., Roselyn E. Morris, and Alice A. Ketchand. 2001. Effects of personal Values on auditors' ethical decision. Accounting, Auditing & Accountability Journal 14. no. 3: 254-77. Singhapakdi, Anusorn and Scott J. Vitell. 1990. Marketing ethics: Factors Influencing perceptions of ethical problems and alternatives. Journal of Macro marketing 10. no.1 : 4-18. Singhapakdi , Anusorn , Scott J. Vitell, and Kenneth L. Kraft. 1996. Moral intensity and ethical decision-making of marketing professionals. Journal of Business Research 36 . no.3: 245-255. Singhapakdi, Anusorn, Scott Vitell, Kumar C . Rallapalli, and Kenneth L. Kraft. 1996. The perceived role of ethics and social responsibility: A scale development. Journal of Business Ethics 15. no. 11: 1131-40. Smielliauskas, Wally J. and Bewley Kathryn. 2007. Auditing: An international approach. 4 1h ed. McGraw-Hill Ryerson . Stack, J. J. 1963. Individual differences in the reduction of cognitive dissonance: an exploratory study. Unpublished doctoral dissertation , Ohio State University. 68 MBA Project - Ku i-Ying L i n Su, Lixin . 2005. Earnings management and forecast guidance as mechanisms to meet or beat analysts ' earnings forecasts . PhD diss ., The University of Texas at Dallas. In ProQuest Dissertations and Theses, http://proguest. wni.com/ pgdweb?index= 1&did=921 025421 &SrchMode=2&sid= 1&Fmt=6& Vlnst=PROD& VType=PQD&RQT=309& VName=P QD&TS= 1240784721&clientld=6993 [accessed April 25, 2009]. Sui, W. , and K. Tam. 1995. Machiavellianism and Chinese banking executives in Hong Kong. International Journal of Bank Marketing 13 . 27-38. Tang, Thomas Li-Ping and Yuh-Jia Chen. 2007.lntelligence vs. wisdom: The love of money, Machiavellianism, and unethical behaviour across college major and gender. Journal of Business Ethics 82. no . 1:1-26. Thome, Linda, Dawn W. Massey, and Michel Magnan. 2003. Institutional context and auditors' moral reasoning: A Canada-U .S. compari son. Journal ofBusiness Ethics 43 . no.4 : 305-321. Trevino, Linda K. 1986. Ethical decision making on organizations: A Person-Situation lnteractionist Model. Academy of Management Review 11. no.3: 601-17. Trevino, Linda K. and Stuart A. Youngblood. 1990. Bad apples in bad barrels: A causal analysis of ethical decision-making. Journal ofApplied Psychology 75. no . 4: 378-85. Yitell, Scott J. , Aysen Bakir, Joseph G.P. Paolillo, Encarnacion Ramos Hidalgo, Jamal AlKhatib, and Mohammed Y. A. Rawwas. 2003. Ethical judgments and intentions: A multinational study of marketing professionals. Business Ethics: A European Review 12. no.2: 151-171. Wakefield, Robin L. 2008 . Accounting and Machiavellianism. Behavioural Research in Acounting 20. no.1: 115-29. Wastell, Colin and Alexandra Booth. 2003. Machiavellianism: An alexithymic perspective. Jouranl of Social & Clinical Psychology 22. no. 6: 730-744. Wertheim, Edward G. , Cathy Spatz Widom, and Lawrence H. Wortzel. 1978. Multivariate analysis of male and female professional career choice correlates. Journal ofApplied Psychology 63. no.2: 234-42 . Zook, Avery and Gary J. Sipps. 1986. Reliability data and sex differences with Gender-Free Mach IV. Journal of Social Psychology 126. no. 1: 131-32. 69 MBA Project - Kui-Ying Lin APPENDICES A. The Ethics Position Questionnaire Instructions: You will find a series of general statements listed below. Each represents a commonly held opinion and there are no right or wrong answers. You will probably disagree with some items and agree with others. I am interested in the extent to which you agree or disagree with such matters of opinion. Please read each statement carefully. Then using the following scale, indicate the extent to which you agree or disagree with the statement: 1 = Completely disagree 2 = Largely disagree 3 = Moderately disagree 4 = Slightly disagree 5 = Neither agree nor disagree 6 = Slightly agree 7 = Moderately agree 8 = Largely agree 9 = Completely agree Please place the number (1 , 2, 3, 4, 5, 6, 7, 8, or 9) that best describes your degree of agreement or disagreement in the blank to the left of each statement. _ _ ( 1) A person should make certain that their actions never intentionally harm another even to a small degree. _ _ (2) Risks to another should never be tolerated, irrespective of how small the risks might be. _ _ (3) The existence of potential harm to others is always wrong, irrespective of the benefits to be gained. _ _ (4) One should never psychologically of physically harm another person. _ _ (5) One should not perform an action which might in any way threaten the dignity and welfare of another individual. _ _ (6) If an action could harm an innocent other, then it should not be done. _ _ (7) Deciding whether or not to perform an act by balancing the positive consequences of the act against the negative consequences of the act is immoral. _ _ (8) The dignity and welfare of people should be the most important concern in any society. _ _ (9) It is never necessary to sacrifice the welfare of others _ _ (10) Moral actions are those which closely match ideals of the most "perfect" action. 70 MBA Project - Kui-Ying Lin _ _ (11) There are no ethical principles that are so important that they should be a part of any code of ethics _ _ (12) What is ethical varies from one situation and society to another. (13) Moral standards should be seen as being individualistic; what one person considers to be moral may be judged to be immoral by another person. _ _ (14) Different types of moralities cannot be compared as to "rightness." _ _ (15) Questions of what is ethical for everyone can never be resolved since what is moral or immoral is up to the individual. _ _ (16) Moral standards are simply personal rules which indicate how a person should behave, and are not to be applied in making judgments of others. _ _ (17) Ethical considerations in interpersonal relations are so complex that individuals should be allowed to fonnulate their own individual codes. _ _ (18) Rigidly codifying an ethical position that prevents certain types of actions could stand in the way of better hwnan relations and adjustment. _ _ (19) No rule concerning lying can be formulated ; whether a lie is permissible or not permissible totally depends upon the situation. _ _ (20) Whether a lie is judged to be moral or immoral depends upon the circumstances surrounding the action. *Statements 1 through 10 are idealism items; 11 through 20, relativism items. Source: (forsyth. 1980, 178) 71 MBA Project - Kui-Ying Lin B. The Liker Type Mach Scale (IV) You will find 20 statements listed below. Each represents a commonly held opinion and there are no right or wrong answers. You may agree with some statements but disagree with others. 1 am interested in the extent to which you agree or disagree with such matters of opinion. Please read each statement carefully. Using the following scale, indicate the extent to which you agree or disagree with each statement: I = Strongly disagree 2 = Somewhat disagree 3 = Slightly disagree 4 = Neutral 5 = Slightly agree 6 = Somewhat agree 7 = Strongly agree Please write the number (1, 2, 3, 4, 5, 6, or 7) that best represents your degree of agreement or disagreement on the line to the left of each statement. _ _ (I) The best way to handle people is to tell them what they want to hear. _ _ (2) When you ask someone to do something for you, it is best to give the real reasons for wanting it rather than giving reasons which might carry more weight. _ _ (3) Anyone who completely trusts anyone else is asking for trouble. _ _ (4) lt is hard to get ahead without cutting comers here and there. _ _ (5) Honesty is the best policy in all cases. _ _ (6) lt is safest to assume that all people have a vicious streak and it will come out when they are given a chance. _ _ (7) Never tell anyone the real reason you did something unless it is useful to do so. _ _ (8) One should take action only when sure it is morally right. _ _ (9) lt is wise to flatter important people. _ _ (I 0) All in all, it is better to be humble and honest than important and dishonest. _ _ (11) Barnum was very wrong when he said there's a sucker born every minute. _ _ (12) People suffering from incurable diseases should have the choice of being put painlessly to death. 72 MBA Project - Kui- Ying Lin _ _ (13) It is possible to be good in all respects. _ _ (14) Most people are basically good and kind. _ _ ( 15) There is no excuse for lying to someone else. _ _ (16) Most men forget more easily the death of their father than the loss of their property. _ _ (17) Most people who get ahead in the world lead clean, moral lives. _ _ (18) Generally speaking, men won't work hard unless they're forced to do so. _ _ ( 19) The biggest difference between most criminals and other people is that criminals are stupid enough to get caught. _ _ (20) Most men are brave. Source: Christie and Geis et al. eds. 1970. 17-18. 73 MBA Project - Kui-Ying Lin C. The Internal-External Locus of Control Scale This is a questionnaire to find out the way in which certain important events in our society affect different people. This is a measure of personal belief: obviously there are no right or wrong answers. Each item consists of a pair of alternatives lettered a or b. Please select only the one statement which you feel is truer. Be sure to select the one you actually believe to be truer rather than the one you think you should choose or the one you would like to be true. In some instance you may find that you believe both statements or neither one. In such cases, be sure to select the one you more strongly believe to be the case as far as you're concerned. Please try to respond to each item independently when making your choice; do not be influenced by your previous choices. Be sure to make a choice for every pair. Please indicate your choice for every pair by placing either a or b in the blank to the left of the number. _ _ (I) a. Children get into trouble because their parents punish them too much. b. The Trouble with most children nowadays is that their parents are too easy with them _ _ (2) a. Many of the unhappy things in people 's lives are partly due to bad luck. b. People 's misfortunes result from the mistakes they make. _ _ (3) a. One of the major reasons why we have wars is because people don't take enough interest in politics. b. There will always be wars, no matter how hard people try to prevent them. ___ (4) a. In the long run people get the respect they deserve in this world. b. Unfortunately, an individual 's worth often passes unrecognized no matter how hard he tries. _ _ (5) a. The idea that teachers are unfair to students is nonsense. b. Most students don ' t realize the extent to which their grades are influenced by accidental happenings. _ _ (6) a. Without the right breaks on cannot be an effective leader. b. Capable people who fail to become leaders have not taken advantage of their opportunities. _ _ (7) a. No matter how hard you try some people just don't like you. b. People who can't get others to like them don't understand how to get along with others. 74 MBA Project - Ku i-Ying Lin _ _ (8) a. Heredity plays the major role in detennining one's personality. b. It is one's experiences in life which determine what they're like. _ _ (9) a. I have often found that what is going to happen will happen. b. Trusting to fate has never turned out as well for me as making a decision to take a definite course of action. _ _ (1 0) a. In the case of the well prepared student there is rarely if even such a thing as an unfair test. b. Many times exam questions tend to be so unrelated to course work that studying is really useless . _ _ ( 11) a. Becoming a success is a matter of hard work, luck has little or nothing to do with it. b. Getting a good job depends mainly on being in the right place at the right time. _ _ (12) a. The average citizen can have an influence in government decisions. b. This world is run by the few people in power, and there is not much the little guy can do about it. _ _ (13) a. When I make plans, I am almost certain that I can make them work. b. It is not always wise to plan too far ahead because many things tum out to be a matter of good or bad fortune anyhow. _ _ (14) a. There are certain people who are just no good. b. There is some good in everybody. _ _ ( 15) a. In my case getting what I want has little or nothing to do with luck. b. Many times we might just as well decide what to do by flipping a coin. _ _ ( 16) a. Who gets to be the boss often depends on who was lucky enough to be in the right place first. b. Getting people to do the right thing depends upon ability, luck has little or nothing to do with it. _ _ ( 17) a. As far as world affairs are concerned, most of us are the victims of forces we can neither understand, nor control. b. By taking an active part in political and social affairs the people can control world events. _ _ (18) a. Most people don ' t realize the extent to which their lives are controlled by Accidental happenings. b. There really is no such thing as "luck." _ _ (19) a. One should always be willing to admit mistakes. b. It is usually best to cover up one's mistakes. 75 MBA Project - Ku i-Ying Lin _ _ (20) a. It is hard to know whether or not a person really likes you. b. How many friends you have depends upon how nice a person you are. _ _ (21) a. In the long run the bad things that happen to us are balanced by the good ones. b. Most misfortunes are the result of lack of ability, ignorance, laziness, or all three. _ _ (22) a. With enough effort we can wipe out political corruption. b. It is difficult for people to have much control over the things politicians do in office. _ _ (23) a. Sometimes I can ' t understand how teachers arrive at the grades they give. b. There is a direct connection between how hard I study and the grades I get. _ _ (24) a. A good leader expects people to decide for themselves what they should do. b. A good leader makes it clear to everyone what their jobs are. _ _ (25) a. Many times I feel that 1 have little influence over the things that happen to me. b. It is impossible for me to believe chances or luck plays an important role in my life. _ _ (26) a. people are lonely because they don't try to be friendly. b. There ' s not much use in trying too hard to please people, if they like you, they like you. _ _ (27) a. There is too much emphasis on athletics in high school. b. Team sports are an excellent way to build character. _ _ (28) a. What happens to me is my own doing. b. Sometimes I feel that I don 't have enough control over the direction my life is taking. _ _ (29) a. Most of the time I can 't understand why politicians behave the way they do. b. In the long run the people are responsible for bad government on a national as well as on a local level. Source: Rotter 1966. 11-12. 76 MBA Project - Kui-Ying Lin D. Earnings management scenarios questionnaire The questions below describe a division general manager (GM) who engaged in many practices intended to improve short-term earnings results. Some may consider earnings management unethical, while others regard the practice as ethical. I am interested in your evaluation of such practices. Please read the following scenarios carefully and, using the following scale, indicate the extent to which you think such practices are ethical or unethical: 1 = Completely ethical 2 = Ethical 3 = Neutral 4 = Unethical 5 = Completely unethical Please place the number (1 , 2, 3, 4, or 5) that best describes your view in the blank next to each scenario. Operating methods of managing earnings: _ _ (1) The GM had the company's buildings (originally scheduled to be done in 2008) painted in 2007 because the division's earnings were significantly ahead ofbudget in that year. (Merchant 1989, 242) _ _ (2) The GM had shipments of available products postponed until January 2008 Because the division's sales exceeded the 2007 budget. (Nelson, Elliott, and Tarpley 2003, 26) _ _ (3) To meet the 2007 profit target, the GM had all discretionary expenditures (travel, advertising, hiring, and maintenance expenses) postponed from November 2007 till January 2008. (Merchant 1989, 242) _ _ (4) To meet 2007 sales targets, the GM implemented a liberal sales program at the end of2007, thus borrowing sales from 2008. (Merchant 1989, 242) (5) In December 2007, the GM ordered his controller to backdate sales invoices and shipping documents from 2008 sales. (Shafer 2002, 258-60). a. The amount of backdated sales was 10% of 2007 earnings before tax. b. The amount of backdated sa les was 0.2% of2007 earnings before tax. 77 MBA Project - Kui- Yi n g Lin _ _ (6) To meet 2007 sales targets, in December 2007 the GM ordered his controller to enter sales orders for customers who had not ordered anything. (Smieliauskas and Bewley 2007, 433). _ _ (7) The GM ordered his employees to work overtime in December 2007 to ship everything possible by the end of 2007. (Merchant 1989, 243) _ _ (8) The GM had some undervalued assets sold to boost 2007 profits. (Merchant 1989, 243; Schilit 2002) Accounting methods of managing earnings: _ _ (9) Due to strong earnings in 2007, the GM ordered his controller to depreciate fixed assets at an accelerated rate. (Nelson, Elliott, and Tarpley 2003 , 23) _ _ (1 0) Because the division exceeded its 2007 profit target, the GM ordered his controller to prepay and record as 2007 expenses the hotel rooms and exhibit costs for a major 2008 trade show. (Merchant 1989, 243) _ _ (11) Since the division ' s 2006 profit exceeded its targets, the GM ordered his controller to increase the reserve for inventory obsolescence in 2006; the writtenoff inventory could be sold for full price in the future . (Merchant 1989, 243) _ _ (12) In 2007, 70 % of the written-off inventory was sold. The GM ordered his controller to reverse 2006 written-off inventory in order to meet 2007 profit targets. (Merchant 1989, 244) _ _ (13) The GM had office supplies received in December 2007 not recorded until February 2008 because of not matching the 2007 earnings target. (Merchant 1989, 242) _ _ (14) The division failed to meet 2007 budget targets, so the GM ordered his controller not to write down an impaired non-operating asset in 2007 . (Nelson, Elliott, and Rarpley 2003, 23 ; Schilit 2002) _ _ (15) To boost 2007 earnings, in December 2007 the GM ordered his controller to record as 2007 revenue the entire rebate amount associated with a five-year bulkproduct purchase agreement running from 2007 to 2012. (Smieliauskas and Bewley 2007, 432; Schilit 2002). 78 MBA Project - Kui-Ying Lin (16) To meet 2007 profit targets, the GM asked a consulting firm to delay sending an invoice for 2007 consulting fees until 2008. (Merchant 1989, 244) a. The involved amount allowed the GM to raise 2007 earnings before tax by 10 %. b. The involved amount allowed the GM to raise 2007 earnings before tax by 0.2 %. 79 MBA Project - Kui-Ying Lin E. Demographic Information Questionnaire Please answer the following questions about yourself by placing a check mark next to the answer that describes you. (1) Gender Male Female (2) Age Under 30 30-40 41-50 51-60 Over 60 (3) How many years have you worked as an accounting practitioner? None _ _ Under 5 years _ _ 5-10 years _ _ Over 10 years (4) What is your present position? None Staff member _ _ Supervisor _ _ Manager CEO (5) What is your annual income? Less than $20,000 _ _ $20,00 1-$35,000 _ _ $35,001-$60,000 _ _$60,00 1-$100,000 More than $100,000 80 MBA Project - Kui-Ying Lin F. A Cover Letter to BCIT Students Kui-Ying Lin 17005 22nd Avenue Surrey, B.C. Y3S 9Z3 14 November 2008 Re: A Questionnaire Survey Dear Student: As a student in the BCIT Auditing- I class, you are invited to take part a research study. Your participation is entirely voluntary but I hope you decide to spend 20 minutes taking part in this study. If you decide to participate in this study, please read and sign the enclosed consent form, complete all five questionnaires, and return all six forms to your instructor. I am an MBA student at the University of Northern British Columbia (UNBC). I am interested in investigating how ethical ideologies and personal characteristics influence British Columbians' decision-making about earnings management ethics. My research is under the supervision of Dr. Deborah Poffand the direction and guidance of both the UNBC Research Ethics Board and the Graduate Studies at UNBC. The study results will be reported in a written document and a dissertation defence presentation as the primary requirement for my MBA degree. The benefits of this study to you will be in knowing that you have contributed to important knowledge about Canadian perceptions of earnings management ethics. Your participation is critical to the success of this study and greatly appreciated. In showing my appreciation of your time to fill in all questionnaires, I have prepared a five-dollar Tim Hortons' card for you when you return the six complete forms. Thank you for your valuable time. I am looking forward to including your opinions about earnings management ethics in this work. Sincerely, Kui-Ying Lin MBA Student at UNBC ENC: Consent Form; Five Questionnaires; Five-dollar Tim Hortons' Card 81 MBA Project - Kui -Ying Lin G. A Cover Letter to UBC Student 1 Kui- Ying Lin 17005 22nd Avenue Surrey, B.C. V3S 9Z3 23 January 2009 Re: A Questionnaire Survey Dear Student: As a student at UBC Sauder School of Business, you are invited to take part a research study. Your participation is entirely voluntary but 1 hope you decide to spend 20 minutes taking part in this study. If you decide to participate in this study, please read and sign the enclosed consent form, complete all five questionnaires, and return all six forms to me. 1 am an MBA student at the University of Northern British Columbia (UNBC). 1 am interested in investigating how ethical ideologies and personal characteristic influence British Columbians' decision-making about earnings management ethics. My research is under the supervisions of Dr. Deborah Poff and the direction and guidance of both the UNBC Research Ethics Board and the Graduate Studies at UNBC. The study results will be reported in a written document and a dissertation defence presentation as the primary requirement for my MBA degree. The benefits of this study to you will be in knowing that you have contributed to important knowledge about Canadian perceptions of earnings management ethics. Your participation is critical to the success of this study and greatly appreciated. In showing my appreciation of your participation, I have prepared a five-dollar Tim Hortons ' card for you when you return the six completed forms . Thank you for your valuable time. I am looking forward to including your opinions about earnings management ethics in this work. Sincerely, Kui-Ying Lin MBA Student at UNBC ENC: Consent Form; Five Questionnaires; Five-dollar Tim Horton ' s Card 82 MBA Project - Ku i -Ying Lin H. A Cover Letter to UBC Student 2 Kui-Ying Lin 17005 22nd Avenue Surrey, B.C. V3S 9Z3 23 January 2009 Re : A Questionnaire Survey Dear Student: As a student at UBC Sauder School of Business, you are invited to take part a research study. Your participation is entirely voluntary but I hope you decide to spend 20 minutes taking part in this study. If you decide to participate in this study, please read and sign the enclosed consent form, complete all five questionnaires, and return all six forms in the self-addressed and postage-paid envelope to me by mail. I am an MBA student at the University of Northern British Columbia (UNBC). I am interested in investigating how ethical ideologies and personal characteristics influence British Columbians ' decision-making about earnings management ethics. My research is under the supervision of Dr. Deborah Poff and the direction and guidance of both the UNBC Research Ethics Board and the Graduate Studies at UNBC. The study results will be reported in a written document and a dissertation defence presentation as the primary requirement for my MBA degree. The benefits of this study to you will be in knowing that you have contributed to important knowledge about Canadian perceptions of earnings management ethics. Your participation is critical to the success of this study and greatly appreciated. In showing my appreciation of your time to fill in all questionnaires, 1 have prepared a five-dollar Tim Hortons' card for you. Thank you for your valuable time. I am looking forward to including your opinions about earnings management ethics in this work. Sincerely, Kui-Ying Lin MBA Student at UNBC ENC: Consent Form; Five Questionnaires; Five-dollar Tim Hortons' Card 83 MBA Project - Kui -Ying Lin I. Informed Consent Forms Informed Consent to Questionnaire Survey University ofNorthem British Columbia Master of Business Administration Determinants of Canadian Accounting Practitioners' Ethical Perceptions on Earnings Management Principal Investigator: Kui-Ying Lin Master student Master of Business Administration University ofNorthem British Columbia Supervisor: Dr. Deborah Poff Professor at UNBC 84 MBA Proj ect - Kui-Ying Lin Introduction You are being invited to take part in a research study because you are an accounting student either at British Columbia Institute of Technology or at USC Sauder School of Business. The study is designed to investigate whether ethical ideologies and personal characteristics affect decision-making about earnings management ethics. This study focuses only on situations in British Columbia and draws on students with an accounting major in Greater Vancouver to collect the needed data. Voluntary Participation Your participation is entirely voluntary. If you wish to participate, you are asked to sign at the bottom of this consent form . Please read the following infonnation carefully before making your decision. Research Conductor This study is being conducted by Kui-Ying Lin, an MBA student at the University of Northern British Columbia (UNBC), supervised by Dr. Deborah Poff, a professor at UNBC, and directed by both the UNBC Research Ethics Board and Graduate Studies at UNBC. Background The collapses of corporate giants such as Enron, WorldCom, and Arthur Andersen have led to a growing concern about earnings management ethics. These accounting scandals have damaged stakeholders' overall confidence and trust in the accounting profession. The academic community has responded with extensive research into earnings management ethics. However, the research has focused only on American situations; little is known about Canadian earnings management ethics. Nevertheless, Canadian companies have suffered from significant earnings management fraud . For example, Norte! Networks, a Canadian networking equipment maker, conducted earnings management fraud to manipulate reserves and change the company's earnings statements (Philips 2007). In addition, Thome, Massey, and Magnan (2003 , 307-08) claim that, while the U.S . Securities and Exchange Commission (SEC) has legal power emanating from the U.S . Congress to penalize practitioners for financial misstatements, the Ontario Securities Commission, for example, has no legal power to levy fines or enforce practitioners to forfeit illegal gains. These authors also assert that, 85 MBA Proje ct - Kui - Ying Lin unlike U.S. accounting standards which are more detailed and rule-oriented because they rely heavily on technical guidances, Canadian accounting standards depend only on a principlebased approach which greatly emphasizes individual , unregulated professional judgment. Purpose of the Study The above phenomena indicate the need for research into Canadian earnings management ethics . This study will investigate how idealism, relativism, Machiavellianism, and locus of control influence decision-making about earnings management ethics. The study results will be reported in a written document and a dissertation defence presentation as the primary requirement for my MBA degree. Participants The study is design to survey students with accounting major at British Columbia Institute of Technology (BCIT) and UBC Sauder School of Business to collect the raw data. Survey procedures You are asked to sign at the bottom of this form, complete five brief questionnaires, and return the consent fonn and all five questionnaires to me through your instructors or by mail. Benefits to Participants of this study The benefit of this study to you will be in knowing that you have contributed to important knowledge of Canadian perceptions about earnings management ethics. This knowledge may give Canadian accounting practitioners a better idea of how to respond to ethical dilemmas involving earnings management practices. Costs/risks to subject from participation in this research There is no cost or risk to you for participating in this research. Completing the survey will take only about 20 minutes of your time. Confidentiality You are assured of confidentiality protection. The infonnation gathered from the questionnaires will be summarized and analyzed, and hopefully able to generalize a 86 MBA Project - Kui-Ying Lin phenomenon. The signed consent form and the completed questionnaires will be kept in a safe place for 3 months and then be shredded. The responses from all participants will be stored on a DVD (digital versatile disc) kept for 5 years. Payment for participation in the research A five-dollar Tim Hortons' card is a small gift for you and represents my appreciation of your valuable time to fill in all questionnaires. Questions If you have any further questions about the research, please feel free to contact me at 604536-1525 . Legal rights and signatures You are not waiving any legal rights by signing this consent form . Your signature below indicates that you have read and understood the provided infonnation about the study and that you agree to participate in this study. Signature of participant Date 87 MBA Proj ect - K ui- Ying Lin LIST OF FIGURES Figure 1 Rest's Four-Component Model RecoQn ize Ethical D ilemma :Ylake Ethica l JudQ111e 1t f 01111 Eth ical I 1ten tion Eng age in Ethical or C neth ical Beh a,·iour Source: Rest (1986, 1-27) Figure 2 Taxonomy of Ethical Ideologies Adopted from Forsyth and Pope (1984, 1366), Barnett, Bass, and Brown (1994, 470) and Elias (2002, 36) 88 MBA Project - Kui-Yi ng Lin Figure 3 Ethical Decision-Making Model Idea lism Et hi ca l/l"nethical Deci ion-makina on [a rn ing 'lanage ment .\Iacbian' lliani 111 H4 (+) Locu of ontrol Figure 4 Response Distribution: School-B-Instructor-2 en1-a nd -lea D i sp Ia .... .~ - ;u: :i- l reiil f -.:i -. ~ iilf ? f = c 89 Q10 MBA Project - Kui -Ying Lin LIST OF TABLES Table 1 Response Rates Survey Records Samples Number of distribution Number of Usable responses Response Rates School-A lnstructor-1 57 28 49 12% School -A- lnstructor-2 73 55 75 34% School -S- lnstructor-1 70 66 94 29% School -S- lnstructor-2 48 26 54.17% Total 248 175 70.56% 90 MBA Project - Kui-Ying Lin Table 2 Demographic Information Summary (n=26) (n=66) (n=55) (n=28 ) ~ 5) # % # % # % # % # % Male 9 32 14% 19 3-l 55% 23 ~ 85% 10 38 46% 61 35 26% Female 19 67 86% 36 65 ~ ~ -13 65 15% 16 61 5-I% 11-1 65 90% under 30 17 60 71% 45 81 82% 65 98 48% 26 100 00% 153 88 44% 30-40 11 39 29% 10 1818% 22 12 2% none 18 ~ 29% 46 83 64% 43 65 15% 20 6 92% 127 73 41 % under 5 years 8 28 57% 12 73% 23 34 85% 6 23 08% 44 25 43% Gender Age 1 52% 41-50 51-60 over 60 !Working experience 0 58% 3 57% 1 5-1 0 years over 10 years 3 57% 2 3 64% 3 1 73% [Present position none 24 85 71% 49 89 09% -15 68 18% 22 84 62% 140 80 92% staff member .j ~ 29% 6 10 91% 21 31 82% 4 15 38% 35 20 23% 21 75 00% 47 85 45% 61 9242% 23 88 46% 152 87 86% 3 1154% 14 8 09% 9 5 20% supervisor manager CEO Annua l Income less than S20 000 S20 001-535 000 4 ~ 29% 9 09% 2 3 03% S35 OOH60 000 3 10 71% 5 45% 3 4 55% S60 001-5100 000 more than S100 000 Sch School lnst Instructor 91 MBA Project - Kui-Ying Lin Table 3 Reliability Estimates Reported in A Sample of Prior Studies Cronbach's Alpha from Previous Research Cronbach's Alpha/ Split-Half Coefficient References Idealism Relativism Fors)1h (1980) 0 800 Barnett Ba ss and Bro·wn (1994) 0 820 Singhapakdi et al (1996) 0 8-10 Forsy1h (1980) 0 730 Barnett Bas s and Brown (1994) 0 790 Singhapakdi et al (1996) 07 0 Christie and Geis (1970) 0 790 Hunt and Chonko (1984) 0 760 Singhapakdi and Vitell (1990) Pretest = 0 13 Dian and Banting (1988) 0 640 Rawwas Vitell and AI -Khatib (1994) Machiavellianism Locus of Control 0 630 0 570 Siu and Ta m (1995) 0 610 rv1cHoskey (1999) 0 690 Beu (2001 ) 0 710 Hargis (2006 ) 0 740 Wakefield (2008 ) 0 650 Rotter (1966) 0 730 T ~ Test = 0 740 and Yo ungblood (1990) 0 740 Singhapakdi and Vitell (1990 ) Pretest = 0 738 Test = 0 740 Bass Barnett &Brown (1999) Beu (2001 ) 0.790 0 680 92 MBA Project - Kui -Ying Lin Table 4 Scale Reliability: Independent Variables Reliability Assessments of Independent Variables Cron bach Coefficient Alph a Sc hooi -A -In structor-1 (N=28) ldealis n 0 52 Re lativis n 0 39 Machiavellianis m 0 54 Locus of Control 0 58 Schooi -A -I nstru ctor-2 (N=55) Idealism 0 47 Relati, is 0 52 Machiavelli ani sm 0 40 Locus of Control 0 56 Schooi-B-Instructor-1 (N=66) Idealism 0 64 Relativis m 0 75 Machiavellianis m 0 68 Locus of Control 0 74 Schooi -B-I nstructor-2 (N=26) Idealism 0 34 Re lativi s 0 51 Mach iavellianis m 0 29 Locus of Control 0 25 93 1 1240 0 9370 0 9320 0 9830 0 9570 1 0840 1 1750 1 0310 3 821 3 714 4 143 3 821 4 464 2 786 3 286 3 250 03 4 OS a 05b 06 07 Q8 Q9 0 9000 0 9620 0 9160 0 9940 4 071 4 036 3 893 3 393 014 01 5 016a 016b 0 6860 0 9000 3 929 4 214 0 13 1 1970 0 12 t;rp 1 1620 3 357 02 3 893 0 8580 2 071 01 0 11 0 5073 3 653 Total 0 10 0 6290 3 810 0 9990 0~ 3 496 Operat1ng Accounting 0 9660 0 4876 3 7J6 4 - 2 618 1 4643 -0 21 43 0 2857 1-5 1-4 1-5 1-5 1-5 1-5 1-5 2-5 2-5 1-5 1-5 1-5 0 3929 0 8929 1 0357 1 071 4 1 21 43 0 9286 0 8929 0 6071 0 2500 ~---~ 3 509 4 127 .1036 .1000 3 927 3 636 3 636 3 382 3 418 0 8600 .1236 0 8214 1 1429 1-5 1-5 0 9400 0 7950 0 7930 0 9430 0 9970 0 9690 0 9100 1 0340 0 9720 1 1970 0 8920 0 9640 0 9560 3 818 3 782 0 71 43 1 0120 0 9320 0 3318 0 4189 0 3783 0 82 14 3 291 3 657 3 75J 3 5596 1-5 0 3571 -0 9286 0 6528 0 8100 0 4960 2-5 1-5 1-4 1 72-4 33 1 89-4 78 1 56-4 00 Table 5 Distributions of Dependent Variables -~ 67 0 5596 1-5 2-5 2-5 2-5 1-5 1-5 2-5 1-5 1-5 1-5 1-5 2-5 2-5 2-5 1-5 94 0 509 1 1 1273 1 0364 1 0000 0 9273 0 6364 0 6364 0 5273 0 3818 0 41 82 -0 3818 1 8545 0 7455 1 2364 0 8182 0 7818 0 2909 1-5 2-5 -0 7273 0 6566 0 7535 1-5 2 89-4 39 2 2 67-4 44 ~ 3 576 .1030 4 227 .1045 3 92J .1030 3 697 3 424 2 879 2 439 .1091 4 3 758 3 682 2 955 1 0240 0 9600 0 6747 0 8850 0 8820 0 8410 0 8940 1 0110 0 9290 1 0600 0 9300 0 6339 0 8720 0 0654 0 9290 0 9950 0 9190 0 9030 0 4325 0 5736 3 64J 0 4303 3 830 Std De•. 3 458 Mean 1-5 1-5 2-5 1-5 1-5 2-5 1-5 1-5 2-5 1-5 1-J 1-5 2-5 3-5 1-5 05758 1 0303 1 2273 1 0455 09242 1 0303 0 6970 05152 0~ 3 654 .1038 3 923 4 115 3 846 3 962 3 654 3 423 3 385 1 5769 0 9615 2-5 2-5 2-5 0 8460 2-5 3-5 0 6538 1 0385 0 9231 1 115-l 0 8-!62 0 6538 2-5 0 -!231 1-5 0 -!231 0 3846 -0 0769 2-5 1-5 2-5 1-5 3-5 0 8240 1 0550 0 5880 0 6750 0 8710 1 0560 1 2700 0 9540 0 8980 1 2300 -0 1212 0 7580 2 923 1 7576 0 7692 0 13-!6 3-5 0 8-!62 1-5 3-5 0 6154 1-5 0 7100 0 0385 0 5620 -0 6923 1-5 0 6581 0 7820 0 5342 D1fference' 1-4 3 00-4 44 3 769 0 9250 1 0610 1 0760 0 8840 0 3943 2 67-4 67 2 56-4 33 Range 4 346 3 846 3 615 3 038 3 6581 0 5230 0 4355 Std De' -0 5606 1 0909 1 5455 0 7576 0 68 18 -0 OJ55 1-4 1-5 -0 9849 1-5 0 6439 0 8300 1 89-4 67 2 28-4 39 3 5342 0 4579 2 22-J 44 3 782 Mean Difference' I Range M BA Project- Kui-Ying Lin 02 03 04 05a 05b 06 07 08 09 010 11 012 013 014 015 016a 016b 25 00% 2143% 10 71% 357% 10 71'o 11m 42 86% 21 ~ 6 ~ 9 5 17 86% 5 17 86% 1 357% 1 . 3s7% I 1 357% 714% 2 2 ; 714% 12 2 7 6 3 1 3 17 86% 14 29% 1429'1 357'1 2857'1 32Wo 5 4 4 3 4 1 3 9 4 4 9 8 1 13 8 18 20 20 23 21 25 14 29%I 15 14 29% 19 10 71'1 20 3214% I 1s 357% 26 1429'1 23 1071% 23 ~~ 29%1 22 ~. 2 4 10 71'1 3 1 2 ~ 6 4 8 7 1 26 14 12 10 7 3 i 8 13 35 7857% 5357% 6786% 7143% 6.129% 92 86% 82 14'1 ~ ~ 6.129'1 71 ~ 7143'1 82 U'l 75 00'1 89 29'1 28 57'1 6 11 14 15 15 12 13 22 13 7 6 8 0 1 15 16 182'1 I 13 364% 8 36.!'' 63 ~ 23 6.1'1 14 55'1 12 73'1 5WI 12m 182% H27't, 25 45'1 2182% 1818'1 12 73'1 14 55% 727% 10 91'1 36.1'; Table 6 Score Ranges: Dependent Variables 1273% 10 91% 23 ~ 000% ~ 00% 23 6.1'1 25 45% 2727% 27 27'1 2182% 20 00% 10 91'1 23 ~ 14 55% ~~ 000'1 182% 27 27% 29 09% 45 43 41 40 29 30 33 35 28 7 54 53 52 5 26 39 41 46 35 95 50 91'' 52 73% 5-155% 60 00'1 63 64% 72 73% 7818'1 ~ 55% 8182% 70 91'1 ~ 55'1 836.!% 63 w; 9818'1 12 73'1 5 5 1 7 4 7 13 13 33 24 1 4 21 9 7 0 ~ 909% ~ 27'1 -~ ... :·"' .~~. . 5 6 1'1 72 73% 3182% 13 6-l'o 10 61'' 0OO'o 606°b 152'1 50OO'o 36 36'( 19 70% 19 70% 10 61'1 60611 758'1 758'1 152'1 1061% 758'1 9og•; 5 6 6 19 15 15 10 7 22 15 1 10 1 !1 0 0 15 23 16 16 67'' 152'1 1515'1 152'1 37 88'1 33 33'o 28 79'1 22 73'1 22 73'1 1515'' 10 61'1 909'1 909'1 758'1 ~ ~ 000'1 000'1 22 73'1 ~ 85'1 55 59 54 54 52 4J 20 34 38 8 6<1 65 52 48 41 22 60 3 61 ( 1 ~ 5t'l 15 33 33'' 10 6212'1 2 72 73'1 2 98 ~ · c 78 79'1 0 96 97'1 0 1212% 1l j 30 30°o 5152% 3 5758% 8 66 67'1 5 78 79'' 2 8182% 1 83 33'1 c j 89 39% 8182% 2 ~ L ~. . · "' ··r. 5769''' 38Wo 169'' 1E9'' 000'1 0CO'o 0(;0'1 50 OO'o 19 23°b 115-l'o 30 ll'o 19 23'1 769'1 385% 000'1 115-l'i 769'1 385'1 000'1 2 5 5 4 5 3 6 !1 8 3 4 10 4 6 6 0 1 9 3077'1 ~ 31'1 1923% 19 23'1 1538% 19 23% 11w; 23 08'1 769% ~·~ 000'1 385'1 ~ 62% 2308% 23 08'1 15 38'' 3as•; 38WI 1538'1 22 17 16 20 20 23 13 12 13 10 22 25 25 2 10 17 20 25 16 ~·· 76 92'1 88 ~ · 65 38'' M62'1 1sm. ~ 38WI 50OO'o ~ 15'1 50 00'1 ~ 769% 3aw; 65 38'o 76 92'' 9615'1 1:" 1',1; MBA Project - Kui-Ying Lin MBA Project - Kui-Ying Lin Table 7 Study Comparison: Earnings Management Sample Mean Score Comparison: Earnings Management 1: Completely Ethical; 5: Completely Unethical 96 MBA Project - Kui-Ying Lin Table 8 Four Contrasting Earnings Management Attributes Mean Acceptability for Earnings Management Ethics with Four Contrasting Attributes Operating Method (1 - 8) Accounting Method (9-16b) Samples rvleans F-ratio P-value Means F-ratio P-value Schoo/-A.-/nstrutor-1 3 496 4 27 0 044 3 81 -+2 0 0-+4 Schoo/Ainstrutor-2 3 5596 6 49 0 012 3 7535 6 49 0 012 Schoo/-8-/nst rutor-1 3 4579 177 <0 0001 3 83 1 7 <0 0001 Schoo 1-8-ln strut or·2 3 5342 3 45 0 069 3 7821 3 45 0 069 Consistent with GAAP (11-12) Inconsistent with GAAP (1 0,11 , 16a. 16b) Sa mples rvleans F-ratio P-va/ue Me ans F-ratio P-value Schooi-A-Instrutor-1 3 7768 0 43 L ~~ r 3 9107 0 43 ~.- 5 lb Schoo/Ainstrutor-2 3 6364 1 24 c ~- 1 24 :J 2P Schoo/-8-/nstrutor-1 3 8636 0 72 c~ 3 772 3 76 14 0 2 J 396 0 11 r 7·'1c 3 7404 0 11 (I Schoo/-8-/ nstrutor-2 3 807 738 Increase earnings (3-8, 12-16b) Decrease Earnings (1 ,2,9,10,11) Sa n pies Means F-ratio P-value Mean s F-ratio P-value Schoo/A/nstrutor-1 3 8132 13 67 0 001 3 235 13 67 0 001 Schoo/Ainstrutor-2 3 8238 60 57 <0 0001 3 2218 60 5 <0 0001 Sc hoo/-8-/nstrutor-1 3 845 66 86 <0 0001 3 1212 66 86 <0 0001 Schoo 1-8-/n struto r·2 3 8462 21 82 <0 0001 3 1692 21 82 <0 0001 Mate riality (5a, 16a) Immate riality (5b, 16b) San pies Mean s F-rat io P-value Me ans F-ratio P-value Schoo/A/n strutor-1 4 0179 3 65 0 061 3 607 1 3 65 0 061 Schooi-A-Inst rutor-2 4 1818 141 <0 000 1 3 6273 14 1 <0 0001 Schoo/-8-/nstrutor-1 4 2879 14 14 <0 0001 3 8333 14 14 <0 0001 Schoo 1-8-1 nstrut or-2 4 1923 9 32 0 004 3 7115 9J2 0 004 1=Completely Ethical; S=Completely Unethical 97 MBA Proj ect - Kui -Ying Lin Table 9 Distributions of Idealism and Relativism Statistics: Idealism and Relativism Relativism Idealism lvle an Std Dev Schooi-A-Instructor-1 7 079 0 8920 Schooi-A-Instructor-2 6 844 1 0940 Schooi -B-Instructor-1 6 591 Schooi -B-Instructor-2 5 942 ......... .. ~~~ -~ - --- -- Range Difference' rvlean Std Dev Range Difference' 5 5-8 S 2 0790 5 596 0 9880 3 5- 0 5960 3 8- 8 6 1 8440 5 664 1 2870 2 2-7 8 0 6640 0 910 0 4 5-8 1 1 5910 5 609 1 2260 2 5- 8 4 0 6090 1 2950 2 9- 3 9 0.9420 5 608 1 5250 1 1-8 1 0 6080 . - ........... .. .~-~~ ... ····-------··------···-·-····-········· ...... ··-· -------··- -·-· ---·-·--·-··-·-·-····················-·······................. _____________ ~~~ -~ _____________ ---------------·--------------- ·-------· --- --------· --· -------· ------------------ ---- ----------------------------------------• The difference in means was computed by deducting the scale neutral value (=3) from its question response. !=Completely Disagree; 9=Completely Agree Table 10 Score Ranges: Idealism and Relativism Idealism Score Ranges (1 "'9) X S (H"gh 10) % Co unts % Counts % 0.00% 0 0.00% 28 100.00% 9.09% 0 : 0.00% so 90.91% 3 4.55% 0 0.00% 63 95 .45% 4 15.38 % 0 O.OOS6 22 84.62% 0 5 ~ ; : Relativism Score Ranges (1"'9) XS (H"gh Rl) % Counts % Co unts % 21.43% 2 7.14% 20 71.43% 15 : : 27.27% 0 0.00% 40 72 .73% 21 31. 82 % 2 ; 3.03% 43 65 .15% 19.23% 1 ' 3.85% 20 76.92% 6 5 ' 1=Completely Disagree; 9=Completely Agree 98 MBA Project - Kui-Ying Lin Table 11 Study Comparison: Idealism and Relativism References Subjects 11980) 2-1 1 University Students Barnett Bass and Brown (199-1! 166 Business Students Bass Barnett and Brown l1999) 602 Marketing Practitioners (American Marketing Associ ation) F ~ 6 35 Sample Standard Means Deviation Range Range 11 6 18 113 1 28 5 -13 141 6 22 1-1 ~ 8 14 0"8E 0 8917 5H5 5 596.1 0 98 9 3 5,77 s:. t..::ount1ng Students Vi.' ~ 6-1-1 1 09-1 3H6 S664 128 2 2,7 8 bG k count,ng Students 61! 6 S91 0 91 4 5' 8 1 :: G09 1 226 2 5,8 4 23 ..... csountmg Students /-1 , Present Research Sample Standard Means Deviation 7 Table 12 Correlation Analysis Collinearity Statistics: Idealism vs. Relativism Correlation Coefficient Schooi-A-Instru ctor-1 (f\J=28 ) J 0-i-W -0 1210 Schooi-B-Instructor-1 (N=66) 0 2210 Schooi-B-Instructor-2 (f\J=26) 1 002 0 8220 Sc hooi -A-Instructor-2 (N=55 ) Variable Inflation Factor (VIF) Inde x 1 015 0 3780 1 051 0 0740 -0 3190 1 113 0 1120 99 MBA Project - Kui-Ying Lin Table 13 Regression Analysis: Idealism Regression Results: Earnings Management vs.ldealism Samples Schooi-A-Instructor-2 (I 1=55) School-8-lnstructor-1 (J·J=66) School-8-lnstructor-2 (N=26) Hypothesized Model Paths Path Expected Observed tP-value Signs Coefficients statistics Signs R2 F-ratio 10 + EM015 + + 0 1 08 1 85 10 + EMQ16a + + 0 1871 1 9-1 0 0580 0 07 377 10 + EMQ16b + + 0 2209 1 94 0 0580 0 07 3 75 10 +OM + + 0 2355 -141 <0 0001 0 24 9 3 10 +AM + + 0 1996 2 58 0 0120 0 10 349 10 + TM + + 0 2176 3 96 <.0 0001 0 20 29 10 + EMQ1 + + 0-1283 3 69 <0 0001 0 18 6 94 10 + EMQ3 + + 0 2786 2 06 0 0430 0 08 2 90 10 + EMQ4 + + 0 3182 2 53 0 0140 0 09 3 28 10 + EMQ7 + + 0 3943 3 25 0 0020 0 16 5 78 10 + EMQB + + 0 4316 3 15 0 0030 0 17 6 43 10 + EMQ9 + + 0 3902 3 27 0 0020 0 18 7 00 10 + EMQ11 + + 0 2196 1 83 0 0710 0 05 3 37 10 + EMQ12 + + 0 2938 2 60 0 0110 0 11 3 75 10 + EMQ16a + + 0 2661 2 08 0 0410 0 06 4 34 10 + EMQ16b + + 0 4730 3 57 0 0010 017 6 48 10 + EMQ13 + + 0 218 22 0 0330 0 18 513 ID = Idealism OM = Operating Manipulations AM = Accounting Manipulations TM =Total Manipulations EMQ = Earnings Management Scenario Questions 100 5 38 MBA Project - Kui-Ying Lin Table 14 Regression Analysis: Relativism Regression Results: Earnings Management vs. Relativism Samples School-Alnstructor-2 (11=55) School-8lnstructor-1 (N=66) School-8lnstructor-2 (11=26) Path Observed !-statistics Signs Coefficients P-value R2 F-ratio RL + AM -0924-l -214 00370 010 29 RL + EM05a -0 1842 -2 11 00400 012 364 RL + EMQ13 -01916 -1 86 00690 006 3-l5 RL + EMQ14 -0 1 10 -175 0086 005 305 RL + EMQ15 -0 1928 -2 -l6 00170 01 538 RL + EMQ7 -01533 -1 0 009-lO 016 58 RL + EMQ9 -02210 -2 50 00150 0 18 00 RL + EM05a -01233 -174 00950 0 11 303 RL + EMQ5b -02450 -281 00100 028 440 RL + EMQ12 -02 12 -2 44 00230 022 323 RL + EMQ13 -01652 -2 03 005-lO 030 496 RL + EMQ14 -0 1652 -2 03 005-lO 030 496 RL + EMQ15 -03434 -2 64 00150 02 426 Hypothesized Model Paths Expected Signs RL =Relativism AM =Accounting Manipulations EMQ =Earn1ngs Management Scenario Questions 101 MBA Project - Kui- Ying Lin Table 15 ANOV A Results: Four Ethical Ideologies Testing Results in ANOVA Operating Acco unting Total Manipul atio ns Sc hoo i-A-In stru cto r-1 F= 1 17 P=O 342 F=1 09 P=O 373 F=1 32 P=O 291 Situationists 3 7302 3 9524 3 8413 A.bsolutists 3 2381 3 4444 3 3413 Subjectivists 3 4921 3 9683 3 7302 Exceptionists 3 5238 3 8730 3 6984 Sc hooi -A-In stru cto r- 2 F=O 07 P=O 976 F=1 66 P=O 187 F=O 88 P=O458 Situationists 3 5299 3 5556 3 5427 Absolutists 3 5952 3 9048 3 7500 Su bj ectivists 3 5476 3 7857 3 6667 Except ionists 3 5635 3 7540 3 6587 F=3 3CJ P=O 024 F=' :31 P=0 014 F=S 82 P=O 004 Situationists 3 5903 4 IF64 3 8333 Absolutists 3 6528 4 or8 3 8403 Subjectivists 3 3007 3 5359 3 4183 Ex ceptionists 3 3072 3 7059 3 5065 F= 1 97 P=O 148 F=O30 P=O 827 F=O 84 P=O484 Situationists 3 5741 3 6481 3 6111 Absolutists 3 3519 3 8704 3 6111 Sujectivi sts 3 3651 3 7143 3 5397 Exceptionists 3 8254 3 8889 3 857 1 (N =28) (N=55) Sc hoo l-S-In stru cto r -1 (N =66) Sc hoo i-B-1 nstru ctor -2 (N=26) 1=Completely Ethical ; 5=Completely Unethical 102 MBA Project - Kui -Yin g Lin Table 16 Distributions of Machiavellianism Statistics: Machiavellianism Mean Std De Range Difference' Schooi-A-Instructor-1 (N=28) 102 5 9 300 8 -124 2 5700 Schooi-A-Instructor-2 (N=55) 102 82 10 4200 76 - 128 2 8200 Schooi-B-Instructor-1 (N =66) 102 1 10 4300 79 -133 2 1 00 Schooi-B-Instructor-2 (N=26) 98 85 10 7000 82- 117 -1 1500 a The difference in means was computed by deducting the scale neutral value (=100) from its question response. - Table 17 Score Ranges: Machiavellianism Mach IV Scale Score Ranges (40-160) MA-Schooi-A.-Instructor-1 (N=28) MA-Schooi-A.-Instructor-2 (N=55) MA-Schooi-B-Instructor-1 (N=66) MA-Schooi-B-Instructor-2 (N=26) X<100 (Low-Mach) X=100 (Neutral) Counts % Counts :' : % Counts ''' 10 35 1% 1 35 % 1 i 60 1% 30 '''' 54 55% : 23 25 15 41 82% 37 88% 57 69% 103 '' 2 3 0 i ' ' 3 64% 4 55% 0 00% X>100 (High-Mach) : : 38 11 % ' ' ' '' i' 57 58% ' ' i: 42 31 % MBA Project - Kui-Ying Lin Table 18 Machiavellian Scores Reported in a Sample of Studies Comparison of Machiavellian Scores Group Mean Score Std . Dev. N Reference Male Employees 108 13 12 90 105 Touhay (1973) University Students in Marketing 105 -10 11 50 -182 Mostafa (200 ) College Students 111 A.ccount,ng 102 82 1C -12 ~. Present Research College Students '1 :.cccJntl'l9 w2 ~ ( 20 Present Resea ch L n·.ers1ty Students m A.ccourtmg 102 1" 1C -13 66 Present Researcl' Purch asing Managers 99 60 12 60 122 Chunko (1982) L!n:ers1t Students m "'.ccourtmg 98 85 1C ~ 26 Present Researcl' College Students 90 70 1-l 30 1 82 Christie and Geis (1970) Hong Kong Managers 90 70 15 80 18 Ralston et al (1993) Co mmunity College Tea chers 88 70 14 50 211 Hollon (19 51 Spec1 al Store Managers 87 10 12 80 60 Gable and Topol (199 1) Sales Professionals 86 79 11 42 98 Singhapakdi and Vitell (1992) Marketers 85 70 13 20 1076 Hunt and Chonko (198-l) Audit s 84 50 Unavailable 1477 Christie and Geis (1970) US Managers 83 90 10 60 62 Ralston et al (1993) Student Teachers 82 10 1-l 10 183 Biggers (1976) Hong Kong Bankers 81 90 12 0 50 Siu and Ta m (1995) US Ban kers 81 90 11 80 19 Corzine and Buntzman (1999) CAPs 81 00 10 90 186 Wakefield (2008) Accountants 80 90 11 20 198 Wakefield (2008) Managers 80 00 15 60 149 Gemmill and Heisler {1972) 52 School Superintendents 73 30 11 00 52 Volp and Willower (1977) Sources Hunt &Chonko (1984,36); Wakefield (2008, 123) 104 ~· MBA Project - Kui-Ying Lin Table 19 ANOV A Results: Three Machiavellian Attributes Three Contrasting Attributes of Machiavellianism Females Schooi-A-Instructor-1 (f 1=28) lv1achiavellianism F=1 97 P= O 173 104 3200 Schoo i-A.-Instructor-2 (I 1=55) Machiavellianism School-8-lnstructor-1 (I 1=66) School -6 -lnstructor-2 (I 1=26) Machiavellianism ' 101 3500 98 8900 102 1600 101 6100 102 -WOO 102 2300 102 2400 105 800 F=J 99 P= J OSC 1j-l ooon 98 0000 98 '-IOC F=O31 P= O580 Unavailable 94 6000 10-1 3000 F=O87 P=O356 F=O 16 P=O69 1 102 6100 Working F=O 48 P=O494 104 4500 102 9100 F=2 7-1 P=O 111 101 5000 Not Working F=0 02 P=O890 F=0 06 P=O830 101 9300 Over 30 F=O67 P=O420 F=O 11 P=O736 103 1 00 rv1achiavellianism Under 30 Males 99 5000 96 6700 1=Strongly d1sagree, 5=Strongly agree Table 20 Regression Results: Machiavellianism-1 Regression Results: Earning Management vs. Machiavellianism Samples I Schooi-A-Instructor-1 ~ Schoo 1-A-lnst ru ctor-2 ~ School-8-lnstructor-1 (N=66) Sc hool-8-ln stru ctor-2 (N=26) Hypothesized Model Paths Expected Signs Observed Path t-statistics Signs Coefficients P-value R2 F-ratio MA + EMQ4 -0 0334 -1 89 0 0700 0 12 3 56 MA + EMQ5a -0 0396 -2 32 0 0290 01 5 37 MA + EMQ6 -0 0350 -1 85 0 0760 0 12 3 41 MA + EMQ14 -0 0337 -2 00 0 0570 0 13 3 98 MA + EMQ15 -0 0323 -1 76 0 0900 0 11 3 11 MA + EMQ12 -0 0230 -1 89 0 0646 0 06 3 561 MA + EMQ7 -0 0221 -2 00 0 0500 0 23 4 49 MA + EMQ5b -0 0251 -2 00 0 0570 0 14 4 01 MA + EMQ8 -0 0351 -2 26 0 0330 0 18 5 09 MA = Mach1avelltamsm EMQ = Earn1ngs Management Scenano QuestiOns 105 MBA Proj ect - Kui-Yin g Lin Table 21 Regression Results: Machiavellianism-2 !-statistics Schooi -A-Instructor-1 rll=28 i MA + EM07 + Schooi -A-Instructor-2 MA + EMQ1 r11 =55 ) MA + EM010 + P-va lue R2 F-ratio MA+ AM MA + TM MA + EMQ 3 + MA +E M04 School-6-lnstructor-1 (11=66 1 MA + EM012 + MA + EM013 MA + EM014 MA + EM016b School-6-lnstructor-2 MA + EM014 MA = Mach1avelliamsm AM = Accountmg Mampulat10ns TM = Total Mampulat1ons EMQ = Eammgs Management Scenano Questions Table 22 Orientation Analysis: Locus of Control LC Score Ranges (0 - 23) X<= 10 (Internal) Count Low- LC-% LC-Schooi-A-In structor-1 (N = 28) 13 46.43% 49 .09% 27 LC-Schooi-A-In structor-2 (N = 55) 36 .36% LC-Schooi-B-Instructor-1 (N = 66) 24 11 42 .31% LC-Schooi-B-Instructor-2 N = 26 106 X >= 11 (External) Hi h-LC-% Count 15 53.57% 50.91% 28 63 .64% 42 57.69% 15 MBA Project - Kui-Ying Lin Table 23 Study Comparison: Locus of Control Ware (19641 Stack 11963) Crowne and Conn 11965) 575 8 15 3 aa 605 a 42 H6 1180 a 29 3 97 45 7 1 3 8-l 68 775 3 79 113 773 3 82 Unr:ers1t y• of Connect1cut 13-l 872 3 59 169 9 62 4 07 303 9 22 3 83 116 9 05 3 66 Peace Corps Trainees 122 6 06 3 51 33 5 48 2 ?a 155 5 94 3 36 Colu mbus Oh1o 12th grade College applicants 41 8-16 3 89 32 7 31 3 64 7 96 3 80 1 000 a so 3 7-1 57 9 56 -1 10 Florida State University Gore and Rotter (1963) Fran klin 1.1963 1 Ollie State University Kansas State Umvers1ty ~ Strat1fied Sa mple Purdue opin1on poll 18-year-old SubJects (who were applying for college) 32 10 00 -1 20 25 9 00 3 90 5 5 8 15 3 88 605 8 42 4 06 MBA Students [alpha= 0 74 p380] 9-1 a a-1 3 92 Market1ng Practitioners 602 8 46 4 20 .j 00 Rotter (1966 26 ) Trevinon and Youngblood (1990) ;--.c r,.., mt1ng Students A1 Present Research 8 33 3 71 19 11 74 3 56 Accountmg Students tA21 19 9 63 2 91 36 11 50 43 ::oj 10 86 A.ccountmg Students 1B11 23 10 70 3 69 43 11 88 3 25 66 11 47 3 43 Accountmg Students (B2) 10 10 90 2 28 16 1088 258 26 10 89 2 42 Sources Rotter (1966), Gore and Rotter {1963), Ladw1g (1963), Stack {1963) Franklm {1963), Trevmon and Youngblood {1990), Bass, Barnett, and Brown {1990), Beu {2001) 107 MBA Proje ct - Kui-Ying Lin Tab le 24 ANOVA Results: Attributes of Locus of Control Three Contrasting Attributes of Locus of Control Females Sch ooi-A.-Instructor-1 (N=28) 8 333LJ Sch ooi-A-Instructor-2 ' 1 :.oCJ Locus of Control Sch ool-6-lnstructor-1 (1·1=66) Locus of Co ntrol 11 1780 F=1 83 P=O 181 10 8180 9 4000 11 47 0 F=O000 P=O93 11 0000 10 8330 10 3000 F=2 03 P= O 155 11 1960 9 1110 F=O08 P=O77 11 5580 11 3040 F=.:l.:\7 P=O 0.:!5 Una ailable 10 9000 Working F=O 12 P=O735 F=O02 P=O391 10 6960 10 8750 Not Working F=1 63 P=O207 9 6320 11 8840 Sch ool-6-lnstructor-2 (N =26) 10 5290 F=2 80 P=O 1) ~ Over 30 F=O0.:! P= 0 852 F=S .:!5 P=C 028 Locus of Control Locus of Control Under30 Males 11 .:lOOC 9 1670 Tab le 25 Regression Results: Locus of Control Regression Results: Earnings Management vs. Locus of Control Samples Hypothesized Model Paths Expected Signs School -A-Instructor-1 (N=28) LC + EM012 + School -A-Instructor-2 (N=55) LC + EMOB School -B-Instructor-1 (N=66) School -A-Instructor-2 (N=55) Path Observed !-statistics Signs Coefficients P-value R2 F-ratio + 0 0913 2 19 0 0380 0 16 4 80 + + 0 0790 1 99 0 0510 0 07 3 98 LC + EM01 + + 0 0689 2 17 0 0340 0 07 4 69 LC + EMQ5a + LC + EM013 + LC + EM015 + LC = Locus of Control EMQ= Earn1ngs Management Scenano Questions 108