University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA COLLEGE OF HUMANITIES PERSONAL VALUES, PERCEIVED IMPORTANCE OF AN ETHICAL ISSUE AND ETHICAL DECISION MAKING OF ACCOUNTING STUDENTS BY GABRIEL KORANKYE THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL ACCOUNTING DEGREE JULY 2018 University of Ghana http://ugspace.ug.edu.gh DECLARATION I do hereby declare that this work is the result of my research and, to the best of my knowledge, has not been presented for any academic award in any university. All references used in the work have been duly acknowledged. I, therefore, take sole responsibility for any shortcomings. ..………………… ……..……………….. Korankye Gabriel Date (10383589) i University of Ghana http://ugspace.ug.edu.gh CERTIFICATION I hereby certify that this thesis was supervised in accordance with the procedures laid down by the university. …………………………… …………………….. Dr Francis Aboagye-Otchere Date (Supervisor) …………………………… …………………….. Dr Ibrahim Bedi Date (Supervisor) ii University of Ghana http://ugspace.ug.edu.gh DEDICATION This work is dedicated to God Almighty, my lovely sister, Lydia Acquah, my parents, Mr Eric Oscar Korankye and Theresa Appiah, and all my supportive brothers Emmanuel, Joseph, Samuel and Solomon. Finally, I dedicate this work to my dearest friend, Pamela Sung-Bawiere Suglo (May her soul rest in peace). iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENTS Firstly, I thank the Almighty God, for His wondrous gifts and abundant grace that has brought me this far. Next, my profound gratitude goes to my supervisors, Dr Francis Aboagye-Otchere and Dr Ibrahim Bedi, for their counsel, guidance, and support even when there was no hope of completing this work. Further, much appreciation goes to all faculty members of the University of Ghana Business School, especially, Dr C. Agyenim-Boateng, Dr G. M. Y. Owusu, Dr J. M. Onumah and Dr S. N. Y. Simpson, for their consistent support, encouragement and assistance. Also, a special thank you goes to Mr Augustine Donkor, Mr Kwadwo Appiagyei and Miss Sally Mingle Yorke. Also, I give thanks to all those who assisted in my data collection; Mr Joseph Amedolase, Mallet Ameworlor, and Felix Amoah; for without you, I couldn’t have done it. To all my lovely class family, Foster, Enusah, Fr Wisdom, Felix, Suleman, Mallet, Joseph, Nana Esi and Rhoda, I am grateful. My Priests are not left out; Fr. Larweh, Fr. Apprey, Fr. Francis, Fr. John and Fr. Numekevor. For their continuous prayers, advice and encouragement, I say thank you. Finally, to all who helped in one way or the other, for the love and support, I am more than grateful and your support is more than appreciated. Thank you all and may God bless you in double fold. iv University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION ............................................................................................................................. i CERTIFICATION .......................................................................................................................... ii DEDICATION ............................................................................................................................... iii ACKNOWLEDGEMENTS ........................................................................................................... iv LIST OF TABLES ......................................................................................................................... ix LIST OF FIGURES ........................................................................................................................ x ABBREVIATIONS ....................................................................................................................... xi ABSTRACT .................................................................................................................................. xii CHAPTER ONE: INTRODUCTION ............................................................................................. 1 1.0 Research Background ............................................................................................................ 1 1.1 Problem Statement ................................................................................................................ 3 1.2 Objectives of the Study ......................................................................................................... 6 1.2.1 Main Objective ............................................................................................................... 6 1.2.2 Specific Objectives .......................................................................................................... 7 1.3 Significance of the Study ...................................................................................................... 7 1.4 Chapter Disposition ............................................................................................................... 8 CHAPTER TWO: LITERATURE REVIEW ................................................................................. 9 2.0 Introduction ........................................................................................................................... 9 2.1 Empirical Review .................................................................................................................. 9 2.1.1 Ethical Behaviour ........................................................................................................... 9 2.1.2 Personal Values ............................................................................................................ 10 2.1.3 Ethical Decision Making .............................................................................................. 15 2.1.4 Ethical Decision Making and Individual Characteristics ............................................ 15 2.1.5 Moral Intensity and Ethical Decision Making ............................................................. 17 v University of Ghana http://ugspace.ug.edu.gh 2.1.6 Perceived Importance of an Ethical Issue .................................................................... 19 2.1.7 Ethical Judgement of Accounting Students .................................................................. 21 2.1.8 Ethics in Accounting Education ................................................................................... 22 2.2 Theoretical Review ............................................................................................................. 23 2.2.1 Rest’s Four-Stage Ethical Decision-Making Model ..................................................... 24 2.2.2 Moral Intensity ............................................................................................................. 26 2.3 Theoretical Framework ....................................................................................................... 27 2.3.1 Perceived Importance of an Ethical Issue (PIE) .......................................................... 27 2.3.2 Personal Values ............................................................................................................ 28 2.4 Hypotheses Development .................................................................................................... 29 CHAPTER THREE: METHODOLOGY ..................................................................................... 31 3.0 Introduction ......................................................................................................................... 31 3.1 Research Paradigm .............................................................................................................. 31 3.2 Philosophical Stance ........................................................................................................... 31 3.3 Research Approach and Design .......................................................................................... 32 3.4 Unit of Analysis .................................................................................................................. 33 3.5 Population............................................................................................................................ 34 3.6 Sample and Sampling Strategy ........................................................................................... 34 3.7 Research Instrument ............................................................................................................ 36 3.8 Ethical Consideration .......................................................................................................... 38 3.9 Pilot ..................................................................................................................................... 39 3.10 Administration and Collection of Questionnaire .............................................................. 39 3.11 Data Analysis Techniques ................................................................................................. 40 3.12 Structural Equation Modelling (SEM) .............................................................................. 41 3.12.1 Covariance Based Structural Equation Modelling (CB-SEM)................................... 41 vi University of Ghana http://ugspace.ug.edu.gh 3.12.2 Partial Least Square Structural Equation Modelling (PLS-SEM) ............................. 42 3.12.3 Choosing between CB-SEM and PLS-SEM ................................................................ 43 3.13 Reliability .......................................................................................................................... 45 3.14 Validity .............................................................................................................................. 46 3.15 Summary of Chapter ......................................................................................................... 48 CHAPTER FOUR: PRESENTATION OF RESULTS AND DISCUSSION OF FINDINGS .... 49 4.0 Introduction ......................................................................................................................... 49 4.1 Demographic Analysis ........................................................................................................ 49 4.2 Mean Differences ................................................................................................................ 50 4.2.1 Public vs Private Tertiary Institution – Personal Values and Ethical Judgement ....... 53 4.2.2 Male vs Female Accounting Students – Personal Values and Ethical Judgement ....... 54 4.3 Measurement Model ............................................................................................................ 56 4.3.1 Indicator Reliability ...................................................................................................... 56 4.3.2 Internal Consistency Reliability ................................................................................... 56 4.3.3 Convergent Validity ...................................................................................................... 57 4.3.4 Discriminant Validity ................................................................................................... 59 4.3.5 Cross-Loading Analysis ................................................................................................ 60 4.3.6 Heterotrait-Monotrait (HTMT) Ratio ........................................................................... 60 4.4 Structural Model .................................................................................................................. 62 4.4.1 Variance Inflation Factor Results ................................................................................ 63 4.4.2 Explanatory Power of the Model (or Level of Explanation [R2]) ................................ 65 4.4.3 Path Analysis ................................................................................................................ 66 4.5 Discussion of Results .......................................................................................................... 70 4.5.1 Structural Model ........................................................................................................... 70 4.6 Conclusion ........................................................................................................................... 77 vii University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE: SUMMARY OF FINDINGS, CONCLUSIONS AND RECOMMENDATIONS .............................................................................................................. 79 5.0 Introduction ......................................................................................................................... 79 5.1 Summary of Findings .......................................................................................................... 79 5.2 Practical Implications .......................................................................................................... 81 5.3 Recommendations ............................................................................................................... 83 5.4 Limitations of the Study ...................................................................................................... 84 5.5 Directions for Future Research ........................................................................................... 85 REFERENCES ............................................................................................................................. 87 APPENDIX A – TABLES OF RESULTS ................................................................................. 102 Gender across Institutions ....................................................................................................... 102 Gender across Levels .............................................................................................................. 102 Age across Gender .................................................................................................................. 103 Age across Levels.................................................................................................................... 103 Gender across Religion ........................................................................................................... 104 APPENDIX B – RESULTS OF INDEPENDENT T-TESTS .................................................... 105 Public vs Private Tertiary Institutions and Personal Values ................................................... 105 Public vs Private Tertiary Institutions and Ethical Judgement ............................................... 107 Male vs Female Accounting Students and Personal Values ................................................... 109 Male vs Female Accounting Students and Ethical Judgement ............................................... 111 APPENDIX C – QUESTIONNAIRE ......................................................................................... 113 viii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 4.1: Construct and Measurement Item Means .................................................................... 51 Table 4.2: Item Loadings, Average Variance Extracted and Composite Reliability .................... 58 Table 4.3: Fornell-Larcker Criterion............................................................................................. 59 Table 4.4: Cross Loading Analysis ............................................................................................... 61 Table 4.5: Heterotrait-Monotrait (HTMT) Ratio .......................................................................... 62 Table 4.6: Variance Inflation Factor Results ................................................................................ 64 Table 4.7: Results of Structural Model ......................................................................................... 70 ix University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 3.1: Rest's Four Stage Ethical Decision-Making Model ................................................... 24 Figure 3.2: Conceptual Framework .............................................................................................. 28 Figure 4.3: Structural Model ......................................................................................................... 66 x University of Ghana http://ugspace.ug.edu.gh ABBREVIATIONS Abbreviation/Acronym Meaning AVE Average Variance Extracted BI Behavioural Intention CB-SEM Covariance Based Structural Equation Modelling CR Composite Reliability CU Central University College Hon Honesty HTMT Ratio Heterotrait-Monotrait Ratio IAESB International Accounting Education Standards Board ICA-G Institute of Chartered Accountants Ghana IFAC International Federation of Accountants Int Intellectualism MES Multivariate Moral Equity Scale PIE Perceived Importance of an Ethical Issue PLS-SEM Partial Least Squares Structural Equation Modelling Rel Religiousness SelfCo Self-Control UEM Univariate Ethics Measure UG University of Ghana UPSA University of Professional Studies Accra VIF Value Inflation Factor VVU Valley View University xi University of Ghana http://ugspace.ug.edu.gh ABSTRACT The work of accountants is all based on principles and ethics and hence demands that members of this profession exhibit high levels of ethical behaviour. However, some recent events of unethical behaviours have been recounted, calling for more research in this area. It is believed that to change the face of the profession and win back the trust of stakeholders, a new revolution of accountants must be raised. This study, therefore, investigates the relationship between values and the relevance attached to an ethical issue, and ethical decision-making of accounting students. To achieve this aim, the quantitative research approach is employed. A scenario-based questionnaire was employed to gather data from 444 accounting students within 4 universities, while PLS-SEM was used for the analysis of data. The results revealed that there is a significant positive relationship between honesty and self-control, and ethical judgement; and a significant negative relationship between honesty and behavioural intention. Also, the results showed a positive relationship between ethical judgement and behavioural intention. It is consequently recommended that accounting educators find a fine balance between personal values, ethical judgement and ethical behaviour to enhance the ethical decision-making of accounting students and subsequently, their behaviour. xii University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.0 Research Background The accounting profession is a principle and ethics-based profession and hence demands that members will exhibit high levels of ethical behaviour in order to adequately reflect and perform their responsibilities. It is believed that accountants are committed to exhibiting highly moral and ethical uprightness (Alleyne, Cadogan-McClean, & Harper, 2013; Aveh, Awunyo-Vitor, & Owusu-Afriyie, 2016). It is therefore important to encourage and continue to nurture high ethical behaviour and standard among accountants and more importantly, prospective accountants (Guffey & McCartney, 2008). However, some events of unethical behaviours have been recounted over the years. Examples can be made of the Enron, Arthur Anderson case of 2001, and the WorldCom scandal of 2002, which led to the introduction of the Sarbanes Oxley’s Act (2002). Also, there have been others such as the Freddie Mac scandal of 2003, the Lehman Brothers scandal of 2008 and the Satyam scandal of 2009. These cases of unethical conducts have led to huge implications on the individuals and the firms involved. Furthermore, these unfortunate events have continuously tarnished the image of the accountancy profession (Carnegie & Napier, 2007; Guffey & McCartney, 2008; Sweeney & Costello, 2009). Owing to that, the various accounting professional and regulatory bodies, both at international and country levels, have instituted codes of ethics and regulations that guide the actions and performance of all members. For instance, the International Federation of Accountants 1 University of Ghana http://ugspace.ug.edu.gh (IFAC), the international governing body of all professional accounting bodies, has issued a Code of Ethics for Professional Accountants. The Institute of Chartered Accountants-Ghana (ICA-G), as a member of the IFAC, has adopted the IFACs Code of Ethics for Professional Accountants in the country (Aveh et al., 2016). Notwithstanding the various measures put in place, there are still some incidences of ethical concerns over the years. In Ghana, both the public and private sectors are experiencing a number of these ethical issues. There are consistent reports in the media about the number of public servants (especially, accountants) who are summoned before the Public Accounts Committee of parliament year after year for one ethical misconduct or the other. Studies have attributed these to the poor or low ethics education in the country. It is opined that there are minimal elements of the ethical issues in the educational system (Amponsah, Boateng, & Onuoha, 2016; Ismail & Yussof, 2016; Nsor-ambala & Onumah, 2015). Similarly, it is argued that the best way to bring a generation of ethical accountants is to ensure that the student accountants are equipped with more ethical values and standards (Guffey & McCartney, 2008; Sweeney & Costello, 2009). Nonetheless, even though ethics has been introduced in the accounting education curriculum, it is believed that there are still rising unethical behaviour among accounting students and subsequently accountants. According to Guffey and McCartney (2008), the introduction of the ethics course (or content) is not the problem, but rather, the delivery of the course. It is also argued that knowing the ethical judgement and behaviour of accounting students today, will serve as a good basis to predict their future ethical/unethical behaviour (Boateng & Agyapong, 2017; Guffey & McCartney, 2008). Further, it is opined that this will also serve as a starting point and key ingredient in curbing the unethical behaviours and 2 University of Ghana http://ugspace.ug.edu.gh incidents among accountants (Holmes, Marriott, & Randal, 2012), beginning with designing a well-grounded ethics education curriculum to improve ethical behaviour of the next generation accountants. Several studies have also shown that personal values are key determinants of ethical behaviour (Alleyne et al., 2013; Fritzsche & Oz, 2007; Modarres & Rafiee, 2011). This is prevalent in psychological and organisational literature. According to Harris (1991), students enter the university with different personal values. These different values of students tend to influence or are deemed to have an impact on the ethical behaviour of the students since these values are central to their cognitive make-up (Shafer, Morris, & Ketchand, 2001). There is, therefore, the need to investigate further into how these values influence the ethical judgement and behavioural intention of students. 1.1 Problem Statement Research into ethics in accounting and its surrounding issues has never ceased to be an area of high research demand over the years. This is because, studies suggest that, ethical issues differ across time and persons. That is, what may be ethical for one person, may not be ethical for another or even the same person under different conditions (Guffey & McCartney, 2008; Singhapakdi, 1994; Sweeney & Costello, 2009). Following the Enron and WorldCom scandals, a lot of researchers sought to investigate the ethical perceptions of individual accountants (Alleyne et al., 2013; Antes et al., 2007; Costa et al., 2016; Ziegenfuss & Singhapakdi, 1994). Further studies revealed that there is little about ethics learnt by the students who are being trained to become accountants and hence the need to incorporate ethics into the accounting curricula (Nsor-ambala & Onumah, 2015; Stanga & Turpen, 1991). 3 University of Ghana http://ugspace.ug.edu.gh Also, the continuous incidence of unethical behaviour among accountants has prompted researchers to delve more into the determinants of ethical/unethical behaviour. This has led to a number of studies into the factors that influence the ethical decision-making process of accountants and accounting students (Guffey & McCartney, 2008; Robin, Reidenbach, & Forest, 1996; Stanga & Turpen, 1991; Sweeney & Costello, 2009). The studies in this area have shown that social consensus, the magnitude of consequences, and demographic factors, like gender and educational qualification, influence the ethical decision-making process of auditors and accounting students (Arnold, Dorminey, Neidermeyer, & Neidermeyer, 2013; Pierce & Sweeney, 2010). In addition, it has been revealed that the perceptions of auditors are shaped by the ethical codes of the profession and as such ensuring some high levels of ethical standards (Bebi & Xhindi, 2017; Singhapakdi, 1999). Further, research in ethics (especially in accounting), has also seen a number of studies focusing on developing models to explain how the characteristics of an ethical issue (moral intensity) affects the ethical decision making of individuals (Dorantes, Hewitt, & Goles, 2006; Douglas, Davidson, & Schwartz, 2001; R. Douglas & Kevin, 2002; Leitsch, 2006; Lincoln & Holmes, 2011; Sweeney & Costello, 2009). These studies produced contradicting results. For instance, studies have shown that there are two dimensions of moral intensity (“perceived corporate concern” and “perceived involvement effect”), and these dimensions have no significant impact on moral sensitivity (Leitsch, 2006). On the other hand, a study on the impact of moral intensity on ethical dilemma, ethical judgement and ethical intentions of students revealed that there is a significant relationship between the moral intensity and the three stages of the ethical decision process (Sweeney & Costello, 2009). 4 University of Ghana http://ugspace.ug.edu.gh Following the contradicting results emanating from employing the moral intensity construct as a determinant of ethical decision-making, other scholars employed Robin et al.’s (1996) perceived importance of an ethical issue (PIE) construct. However, little attention has been focused on the level of relevance that an individual attaches to an ethical issue or dilemma (PIE), and how that influences ethical decision-making of an individual (Guffey & McCartney, 2008; Haines, Street, & Haines, 2008; Robin et al., 1996). There is, therefore, the need to investigate the effect of PIE on the ethical decision-making of accounting students. Moreover, Alleyne et al (2013), posit that different individuals exhibit distinct values and at different levels. Studies have indicated that the actions of individuals are to some extent influenced by their personal values (Akaah & Lund, 1994; Finegan, 1994; Shafer et al., 2001). Some researchers have suggested that more attention should be focused on the impact of personal values (especially those which underlie the ethical standards of the accounting profession) on ethical decision-making (Costa et al., 2016; Philmore, Cadogan-McClean, & Harper, 2013). Also, most of the studies surrounding ethics, and more specifically ethics in accounting, have employed a qualitative approach to data analysis (Aveh et al., 2016; Douglas & Kevin, 2002; Flory, Phillips, Reidenbach, & Donald, 1992; Simpson, Onumah, & Oppong-Nkrumah, 2016). The others that employ quantitative approach mostly stick to the simple correlation and mean comparison methods of data analysis (Boateng & Agyapong, 2017; Eaton & Giaciomino, 2001; Ismail, 2015; Mccuddy & Peery, 1996; Modarres & Rafiee, 2011; Philmore et al., 2013). It is therefore important that other robust approaches are employed to investigate and/or examine the relationship between determinants of ethical decision-making. Furthermore, most of these studies focus on a single institution. It is opined that this may affect the generalizability of the findings (Boateng & 5 University of Ghana http://ugspace.ug.edu.gh Agyapong, 2017; Guffey & McCartney, 2008). It is therefore suggested that further studies include samples from multiple institutions and that is what this study seeks to do. In addition, literature has shown that a greater percentage of the studies on ethics in accounting have been conducted in developed economies (Guffey & McCartney, 2008; Haines et al., 2008; Robin et al., 1996; Singhapakdi, Vitell, Lee, Nisius, & Yu, 2013; Stanga & Turpen, 1991; Sweeney & Costello, 2009). Developing economies like Ghana has seen a few of such studies (Boateng & Agyapong, 2017; Simpson et al., 2016). In addition, the few studies in developing economies are also qualitative in nature, focusing on the ethical content in accounting education (Simpson et al., 2016) and gender differences in ethical decision-making (Boateng & Agyapong, 2017). Empirically testing these quantitatively is expected to advance literature from the perspective of developing economies and to further confirm the assertions and propositions made by the interpretivists. Further, literature suggests that in shaping the values upheld by individuals (students), there is a possibility of transforming their ethical behaviour (Alleyne et al., 2013; Guffey & McCartney, 2008). This proposition has not been empirically tested within the field of accounting, hence, the need for this research. 1.2 Objectives of the Study 1.2.1 Main Objective This study extends Rest’s (1965) and Jones’ (1991) ethical decision-making model by incorporating personal values (intellectualism, honesty, self-control and religiousness), and PIE into the model. Hence, the study investigates the relationship between personal values, perceived importance of an ethical issue, and ethical decision making of accounting students. 6 University of Ghana http://ugspace.ug.edu.gh 1.2.2 Specific Objectives The study seeks to achieve the following objectives: (i) To investigate the relationship between personal values and perceived importance of an ethical issue (PIE), and ethical judgement (ii) To determine the relationship between personal values and PIE, and behavioural intention (iii) To examine the relationship between ethical judgement and behavioural intention. 1.3 Significance of the Study The relevance of this study is viewed from three perspectives; literature, policy and practice. To literature, this study introduces a new face into the ethical decision-making process. That is, this study finds that personal values (honesty and self-control) affect the ethical judgement as well as the behavioural intention of accounting students. This introduces a new perspective or dimension to the ethical decision-making process of accounting students that have not yet been studied in the literature. Also, this study adds to empirical studies that test Rest’s ethical decision-making model. To policy, the findings of the study revealed that among all the personal values, only honesty has a significant impact on behavioural intention. This finding among other notable findings is necessary for policy-makers, regulators and accounting educators. This will help them to make policies, and to design and structure the ethical education of accounting students in such a way that will encourage honesty among students and hence ensure ethical decisions and behaviours among accounting students. 7 University of Ghana http://ugspace.ug.edu.gh Lastly, to practice, accounting educators and the respective institutions will benefit enormously from the findings of this study. Upon revealing the personal values that impact ethical judgement and behavioural intention, accounting institutions and educators will now know how to deliver the ethics contents of their courses in order to enhance honesty and self-control. This will, in turn, improve the ethical judgement and ethical behaviour of accounting students. 1.4 Chapter Disposition The rest of the study will be organized as follows. The next chapter will cover an extensive review of previous works that have been done in this subject area. In the third chapter, the researcher outlines the necessary procedures and steps that were taken to ensure that the objectives of the study are achieved. This includes the method, approach and design of the work, as well as the data analysis techniques employed. In chapter four, the researcher explains the results and findings from the data gathered. Finally, the fifth chapter of this study sums up the whole study. That is, it covers summary conclusions, recommendations and suggestions for future research. 8 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.0 Introduction In this chapter, some relevant previous literature are analytically reviewed and discussed. This is to establish the basis for the research and the need to undertake this study. It also explains the relevant terms and key-words in this study. This section is broadly divided into two sections: the empirical review, geared at establishing the gap in literature and backing that with findings from results of analysed data; and a discussion on the theory underlying this research and further establishes the hypotheses to be tested. 2.1 Empirical Review 2.1.1 Ethical Behaviour Ethical behaviour is defined as the conscious effort by an individual to do the right or moral thing even in the absence of rules and supervision (Alleyne et al., 2013). To curb the ethical issues, a number of strict laws and rules have been enacted at country levels by legislatures to redefine the legal boundaries of business practices. In addition, businesses and trade associations have made efforts to guard and guide themselves by formulating codes of business ethics and establishing mechanisms for their enforcement. In the latter parts of the twentieth century, many organisations set up formal training programs for their management to better create awareness of ethical issues that confront them daily (Harris, 1991). 9 University of Ghana http://ugspace.ug.edu.gh There have been numerous calls for business students to be exposed to the changing legal, political and most importantly ethical environment around them. This dates as back as the 1950s when the Carnegie and Ford Foundation commissioned studies on business education (Harris, 1991). Further, in 1974, the American Assembly of Collegiate Schools of Business modified their curricula and for the first time introduced “ethical considerations” (Harris, 1991). Since then, the need for ethics has been heightened in business and accounting practice and education. Similarly, there has been an increase in research in these areas (Apostolou, Dorminey, Hassell, & Rebele, 2016; Guffey & McCartney, 2008; Haron, Ismail, Ibrahim, & Na, 2014; Ismail & Yussof, 2016; Obalola & Adelopo, 2012; Pierce & Sweeney, 2010; Simpson et al., 2016; Stonciuviene & Naujokaitiene, 2013; Sweeney & Costello, 2009; Ziegenfuss & Singhapakdi, 1994). Investigating the differences in ethical behaviour between business and non-business students, the results showed that business students are more ethically tolerant than non-business students (Harris, 1991). 2.1.2 Personal Values Most studies have considered the effect of organisational factors and cultures as determinants of ethical behaviour (Akaah & Lund, 1994; Bebi & Xhindi, 2017; Douglas, Davidson, & Schwartz, 2001). However, researchers have not only explored moral development and ethical behaviour from a cognitive-development perspective, which is the most dominant approach in moral development, but also, from behavioural, social, and psychoanalytic perspectives (Lincoln & Holmes, 2011). That notwithstanding, individual or personal values are not overridden by these factors. The personal values and individual characteristics are vital factors that influence an individual’s ethical 10 University of Ghana http://ugspace.ug.edu.gh behaviour (Longenecker, Mckinney, & Moore, 2004b). According to social adaptation theory, values are “a type of social cognition that facilitates an individual’s adaptation to the environment” (Fritzsche & Oz, 2007, p. 336). Personal values may also be defined as “deep- seated, pervasive, core-beliefs or guiding principles that transcend specific situations to direct or propel human behaviour in decision-making” (Alleyne et al., 2013, p. 49). Researchers should not only look at moral development from a cognitive-development point of view, which is arguably the most predominant approach in moral development, but also from other relevant perspectives, especially social and behavioural points of view (Holmes et al., 2012). It is opined that values are the most abstract social cognition. However, it is believed that values serve as the basis for attitude development and subsequently, specific decision-making (Fritzsche & Oz, 2007). Studies on personal values have concentrated primarily on a broad spectrum of values. For instance, a study conducted of about 1,200 marketing professionals to determine the effect of ethical values (that is, altruistic, self-enhancement, traditional and openness) on organisational commitment revealed that there is a positive significant relationship between ethical values and organisational commitment (Hunt, Wood, & Chonko, 1989). The authors further argue that corporate ethical values are a composite of both policies (whether formal or informal) and individual values of the managers (Hunt et al., 1989). As a result, some further studies in this area investigated the influence of some personal values on ethical decision making. For instance, a study in the US investigated the effect of ethical values on the dimensions of ethical decision making. The values that were considered are altruistic values, self-enhancement values, traditional values, and openness (Fritzsche & Oz, 2007). The results revealed that there is a positive relationship between altruistic values and ethical decision-making whereas self-enhancement negatively related to ethical decision-making (Fritzsche & Oz, 2007). 11 University of Ghana http://ugspace.ug.edu.gh The authors suggest that other values should be investigated to find out how they may also influence the ethical behaviour of individuals (Fritzsche & Oz, 2007; Hunt et al., 1989). For the purposes of this study, we consider some four values of Scott’s (1965) personal values scale. These are honesty, religiousness, intellectualism and self-control. 2.1.2.1 Honesty To be honest means to shy away from deceit, being truthful and sincere, and/or morally correct and vitreous (Wells & Molina, 2017). A study conducted by Wells and Molina (2017) to investigate the truth about honesty as a value in the public sector, revealed that honesty is a very important value that is upheld by most people and professionals. According to Wells and Molina, the respondents, however, argued that due to the complex nature of the world and some situations, deceit or dishonesty may be resorted to. Another study conducted on 69 undergraduate students to examine the relationship between personal values and ethical judgement revealed that honest students were more likely to judge the actions in the scenarios employed as immoral (Finegan, 1994). 2.1.2.2 Religiousness There have been a few studies investigating the impact of the religiousness or religiosity of individuals on their ethical behaviour, but the results of these studies are inconclusive (Longenecker et al., 2004). It is believed that in studying ethics, the focus has been on the philosophical and social aspects while little or no attention is given to the religious aspect of it (Longenecker et al., 2004). The religious nature of an individual is perceived to help determine the difference between right and wrong (Modarres & Rafiee, 2011). The Christians are guided by their Bible, which shapes their decision-making, actions and behaviour. Similarly, Muslims are guided 12 University of Ghana http://ugspace.ug.edu.gh by their Qur’an, and other deities and Holy Books for some other religions (Modarres & Rafiee, 2011). Religious virtue has played a significant role in building and developing accountants’ ethical perspectives. For instance, Modarres and Rafiee (2011) posit that accountants who believe in the Supreme Being, God, strive to do what is right and holy in His eyes and hence create a moral imperative which supersedes just human institutions. A study in the U.S.A consisting of 1,234 business managers and professionals revealed that there is significant evidence that there is a connection between religion and ethics at the workplace (Longenecker et al., 2004). On the other hand, another study in the USA found no significant correlation between religiosity and ethical judgement (Pan & Sparks, 2012). Pan and Sparks (2012) therefore suggested that further studies should focus on such variables with theoretically sound relations but no empirical support. There are two dimensions of religiosity, and these are, extrinsic and intrinsic religiosity (Singhapakdi et al., 2013). They explain extrinsic religiosity as engaging in the worship of a Supreme Being for selfish needs. On the other hand, intrinsic religiosity encompasses the act of engaging in the worship of a Supreme Being based on spiritual and inherent motivations (Singhapakdi et al., 2013). Singhapakdi et al. (2013) conducted to assess the impact of religiosity on the ethical intentions of marketing managers revealed that managers who are of high intrinsic religiosity are very ethical in intentions. On the other hand, managers who exhibited higher levels of extrinsic religiosity were also unethical in their intentions (Singhapakdi et al., 2013). 2.1.2.3 Self Control Self-control can be defined as one’s ability to manage emotions, impulse and behaviour (Baumeister & Line, 2000; Martijn, Tenbült, Merckelbach, Dreezens, & de Vries, 2002). Self- 13 University of Ghana http://ugspace.ug.edu.gh control is deemed to be a very important element of human characteristics. It also believed that one key to pursuing distant goals is the ability to override and change dominant reactions (Baumeister & Line, 2000; Martijn et al., 2002). As the saying goes, “self-control is a virtue”, and it is an important feature that guides the action of individuals. Studies have shown that even though self-control is an important personal value, people find it difficult to exercise or exhibit it (Baumeister & Line, 2000; Martijn et al., 2002). This has been attributed to the belief that self-control drains the individual of so much energy, subsequently leading to low performance. Similarly, a study in the USA also indicated that people who exercise self-control are depleted of their strength or energy and hence unable to continue exercising this value. This, in turn, reduces their moral awareness (Gino, Schweitzer, Mead, & Ariely, 2011). That notwithstanding, Alleyne et al. (2013) believe that self-control has a significant relationship with ethical behaviour. These foregoing arguments reveal the inconsistencies concerning this personal value, self-control. 2.1.2.4 Intellectualism The personal value of intellectualism is also seen as a key part of the human make-up, especially students. According to Alleyne et al. (2013), the ability of an individual to understand, discern and reason is termed as intellectualism. Using correlation analysis, Alleyne et al. (2013) revealed that there is a significant relation between intellectualism and ethical behaviour. This ultimately gives the indication that there is a possible relationship between intellectualism and ethical judgement. 14 University of Ghana http://ugspace.ug.edu.gh 2.1.3 Ethical Decision Making Accountants, and effectively accounting students encounter morally charged situations and hence make ethical judgement and decisions in their daily activities and operations (Lincoln & Holmes, 2011). An ethical decision is defined as “a decision that is legally and/or morally acceptable, and an unethical decision is illegal and/or morally unacceptable to the society at large” (Guffey & McCartney, 2008, p. 329; Johnston, 2010, p. 4). Ethical decision-making is deemed as a process (Jones, 1991). The ethical decision-making process is theoretically explained by Rest’s (1986) model. This model was couched out of a number of studies and theories which sought to define or explain ethical development and behaviour from different perspectives. The model sought to solve the problem of arriving at different outcomes or results when the various different approaches were employed (Lincoln & Holmes, 2011). Rest’s model shows that the ethical decision-making process of an individual is made of four stages. That is moral sensitivity, moral judgement, moral motivation or intention and moral character or action or behaviour. The stages are seen as a decision schema and deemed to occur sequentially (Haines et al., 2008). That is an individual makes a judgement after recognising a moral issue, then after, the individual will make up his or her mind to undertake an ethical behaviour (Haines et al., 2008). 2.1.4 Ethical Decision Making and Individual Characteristics Ethical decision making and the ethical status of individuals especially accountants and also accounting students are argued to be influenced by the individual characteristics of the individual 15 University of Ghana http://ugspace.ug.edu.gh (Apostolou, Dull, & Schleifer, 2013). Some of these individual characteristics include age, gender, level of education, course major, exposure to ethics knowledge and experience, among others (Guffey & McCartney, 2008; Modarres & Rafiee, 2011; Pan & Sparks, 2012; Sweeney & Costello, 2009). For instance, with regards to gender, there have been contradicting reports from different researchers. Whiles some posit an association between gender and ethical decision-making, others purport otherwise. In a study conducted by Guffey and McCartney (2008), the authors sought to find if there are any differences in ethical decision-making of males and females. The results confirmed that male accounting students are more unethical than their female counterparts. This may be due to the edge of men to seek competitive success and as result may resort to unethical means to achieve this (Guffey & McCartney, 2008). Also, a study conducted by Sweeney and Costello (2009) suggested that gender, degree type and firm size have significant relationships with perceived ethical intensity and ethical decision-making. They added that gender, level of education, years of experience and firm size have some influence on ethical culture. However, some other researchers report some contradicting results. In a study by Modarres and Rafiee (2011), they sought to investigate if there is any significant influence of age, gender, work experience and familiarity with ethical codes, on the ethical decision making by accounting students. The results showed that the level of education and familiarity with the codes have an influence on the ethical decisions taken by the individual. Also, a study in the US sought to find out the relationship between locus of control, delay of gratification, gender and race, and ethical beliefs. The study concluded that ethical beliefs are related to locus of control, delay of gratification and race. On the other hand, gender was not related to ethical beliefs (Mccuddy & Peery, 1996). Similarly, Harris (1989) examined the gender differences in the ethical values of 16 University of Ghana http://ugspace.ug.edu.gh graduating students. The findings of this work were no different from those of Mccuddy and Peery (1996). In other related studies, Costa, Pinheiro and Ribeiro (2016) examined the ethical perceptions of accounting students and also investigated the effects of individual factors on the ethical decision- making process. The results showed that the degree of importance attached to initiative/entrepreneurship, obedience, and responsibility is influenced by gender. It was also revealed that age influences the degree of importance that students attributed to integrity (Costa et al., 2016). Regarding ethical intention and ethical judgement, a study by Cohen et al. (1998) revealed that females viewed dilemmas as less ethical compared to males. Also, females recorded a lower intention to perform in an ethically questionable manner compared to their male colleagues. Barnett et al. (1994) also found that among the genders, females have higher levels of ethical judgement than males. Similarly, Brunton and Eweje (2010), in a study on ethical perceptions of business students in New Zealand, also found that females are comparatively ethically aware than their male colleagues. 2.1.5 Moral Intensity and Ethical Decision Making According to Jones (1991), moral intensity consists of six dimensions. These dimensions of the moral intensity construct are “Magnitude of Consequences, Temporal Immediacy, Social Consensus, Proximity, Probability of Effect, and Concentration of Effect” (Dorantes et al., 2006; Douglas et al., 2001; Haines et al., 2008; Lincoln & Holmes, 2011; Sweeney & Costello, 2009). The moral intensity construct is deemed to be monotonic in nature. That is, an increase or decrease in any of the components will lead to an overall increase in the moral intensity of a situation (Haines et al., 2008). It is believed that all the above-mentioned dimensions have an influence on 17 University of Ghana http://ugspace.ug.edu.gh the various components of the moral decision-making process of individuals (Haines et al., 2008; Lincoln & Holmes, 2011; Sweeney & Costello, 2009). There are also other scopes to moral intensity. According to Leitsch (2006), moral intensity can be seen from two different perspectives. These are “perceived corporate concern” and “perceived involvement concern” (Leitsch, 2006). A number of studies surrounding ethical decision-making have sought to find out the impact of the moral intensity construct on ethical decision-making process (Dorantes et al., 2006; Douglas et al., 2001; Douglas & Kevin, 2002; Haines et al., 2008; Leitsch, 2006; Lincoln & Holmes, 2011; Sweeney & Costello, 2009). For instance, in studying the ethical decision making of service academy students, Lincoln and Holmes (2011) found that there is a significant relationship between some dimensions of moral intensity and specific parts of the ethical decision-making process. More specifically, the results of the study showed that there is a significant association between social consensus and moral intensity, moral judgement and intention to act ethically. Also, they found that Proximity has a significant effect on only moral awareness. In addition, moral judgement and intention to act are affected by the magnitude of consequence and probability of effect (Lincoln & Holmes, 2011). In analysing the impact of the two main dimensions of moral intensity on the ethical decision- making process, Leitsch (2006) surveyed about 110 accounting students in a college in the USA. It was found that the two dimensions, perceived corporate concern and perceived involvement effect, had no significant impact on moral sensitivity. The study further examined the combination of the two dimensions with moral sensitivity and the results showed that there is a significant impact on the moral judgement of accounting students. Also, moral judgment and the dimensions 18 University of Ghana http://ugspace.ug.edu.gh of moral intensity significantly influenced the moral intentions of accounting students (Leitsch, 2006). Notwithstanding the enormous subscription to “moral intensity” as one of the dominant determinants of Rest’s four-stage decision-making process, there have been arguments against it. It is believed that the moral intensity construct focuses on the facts of the matter. That is if the facts change, the judgement of the individual will change with no regard to the individual’s moral cognition (Haines et al., 2008; Robin et al., 1996). Therefore, to consider the personal judgement of the individual decision maker, proponents propose another theory or construct, the perceived importance of an ethical issue (PIE). This construct rather takes into consideration the individuals inner view or perception about an issue rather than the facts of the matter. Hence, the moral situation may change, but the ethical stance of the person may remain the same (Haines et al., 2008; Robin et al., 1996). 2.1.6 Perceived Importance of an Ethical Issue Perceived importance of an ethical issue to an individual, also known as perceived personal values, is recently taking over as a new determinant that impacts the ethical decision-making process of an individual (Guffey & McCartney, 2008). This construct was derived by Robin et al. (1996) as an extension of the moral intensity construct established by Jones (1991). PIE has been employed in recent studies by many researchers (Guffey & McCartney, 2008; Haines et al., 2008; Johnston, 2010; Robin et al., 1996; Singhapakdi, 1999). These researchers argue that PIE influences ethical decision-making in almost the same way as Jones’ moral intensity construct. In addition to this broad overlap, the PIE also takes into consideration the individuals perception of an issue (Haines et al., 2008). That is, on a whole, it considers “an individual’s values, beliefs, needs, perceptions, 19 University of Ghana http://ugspace.ug.edu.gh special characteristics of the situation, and the personal pressures existing” in an ethical decision- making situation. Unlike moral intensity, the PIE theory (or construct) has a focus different from Jones’ (1991) model. Whereas Jones’ model focuses on characteristics of the ethical issue itself, the PIE model considers the individual factors and perceptions (Guffey & McCartney, 2008; Robin et al., 1996). In examining the impact of PIE on ethical decision making, Robin et al. (1996), using the survey approach, examined the responses of 41 advertising managers in about six states in the US. It was found that an individual’s level of PIE has a significant influence on the ethical judgement and behavioural intention of the person. Further, the study showed that individuals with high PIE are less likely to engage in unethical behaviour whereas individuals with low PIE were more likely to engage in immoral behaviour. Another study involving marketing professionals from America also investigated the impact of perceived importance of ethics on ethical decision-making. It was also revealed that all the dimensions of perceived importance of ethics and social responsibility, as measured by the “perceived role of ethics and social responsibility” (PRESOR), have a significant positive relationship with ethical decision-making (Singhapakdi, 1999). In the accounting context, a further study by Guffey and McCartney (2008) sought to investigate the ethical decision-making process of accounting students, employing the PIE construct. The results revealed that PIE is significantly related to ethical judgement and behavioural intention to act ethically or unethically. Similarly, Haines et al. (2008) examined the influence of PIE on the four stages of the ethical decision-making process. The results indicated that PIE was a predictor of moral judgement. Haines et al. (2008) also tested the impact of the PIE construct on the ethical decision-making process of accounting students in the USA. The results, on one hand, confirmed 20 University of Ghana http://ugspace.ug.edu.gh the findings of Guffey and McCartney (2008) and Robinson et al. (1996). On the other hand, the study, unlike that of Guffey and McCartney (2008) and Robinson et al. (1996), found no relationship between PIE and moral obligation, and PIE and moral intent. Further investigations revealed that moral obligation happens after a judgement has been made and also moral obligation is a significant determinant in the variation in moral intent (Haines et al., 2008). 2.1.7 Ethical Judgement of Accounting Students Accounting students are the new entrants into the accounting profession. They are the future accountants who will be making certain top-level financial and accounting decisions in various industries and sectors of the economy. Therefore, in trying to ensure improved ethical decision- making among accountants, it is also important to consider accounting students. This is because, it is opined that ethical students most probably grow to become ethical professionals (Guffey & McCartney, 2008). As such, this study assesses the ethical decision making of accounting students and the factors that influence ethical decision making among accounting students. A number of studies have been conducted on the ethical decision-making of accounting students, most comparing them to other business and non-business majors (Alleyne et al., 2013; Apostolou et al., 2016; Boateng & Agyapong, 2017; Guffey & McCartney, 2008; Harris, 1989, 1991; Leitsch, 2006; Sweeney & Costello, 2009). Many of both earlier and recent studies have shown that comparatively, accounting students exhibit higher levels of ethical decision-making compared to the other groups of students (Harris, 1991). For instance, Harris (1991), in his study on the ethical values and decision-making process of business and non-business students, revealed that business students exhibit higher ethical values than their non-business counterparts. 21 University of Ghana http://ugspace.ug.edu.gh Similarly, a study by Alleyne et al. (2013) also revealed that accounting students tend to be more ethical in their decision making than their non-accounting colleagues. Notwithstanding these results, there have been calls for more studies in this regard in order to confirm the assertion. This will subsequently help determine the factors that influence ethical decision making and hence device measures to overcome them and ensure improved ethical behaviour. 2.1.8 Ethics in Accounting Education Following the International Accounting Education Standards Board’s (IAESB) proposition that all professional accounting bodies and accounting educators should include ethics in their code of ethics and curricula, most accounting educators across the world have adhered to and tried to imbibe in their curricula, ethics education. This was reinforced after the numerous scandals in the late 1990s and early 2000s. There were further numerous calls for ethics education and standards to be more stringent. Accounting institutions and professional bodies in Ghana have also tried to abide by this directive and not to be left out. A study in Ghana in 2016 revealed that about 74% of the universities in the country are all having ethics as part of their curricula for their business students. Even though the ethics may not be a stand-alone course but rather, a part of another course (embedded), it is mostly core (Simpson et al., 2016). This indicates the level of importance attached to the course. Also, studies have shown that accounting ethics educators and researchers are still in dilemma, either to make ethics a stand-alone course, a part of a course or a simple elective for students (Gandz & Hayes, 1988; Whitla, 2011). According to Gandz and Hayes (1988) and Whitla (2011), making ethics an elective will lead to the course suffering selection bias. They also posit that making the course an elective one will send a message that ethical behaviour is non-obligatory or voluntary. 22 University of Ghana http://ugspace.ug.edu.gh Other researchers also suggest otherwise. Notwithstanding that, Guffey and McCartney (2008) argue that ethics in institutions is not just about the introduction of the course but most importantly, the delivery and how it is taught. It also argued that to be able to deliver a good product for your customer or better meet the information needs of stakeholders, it is important to know the target population. Hence in applying this to accounting students, it is imperative to study the determinants of ethical decision making of accounting students. This will aid accounting educators to better deliver the course to students in order to achieve the objectives and attain the impact intended. Generally, studies surrounding ethics and more specifically, determinants of ethical decision making have widely focused on using either the demographic or individual characteristics, moral intensity attached to a particular ethical issue or perceived importance attributed to an ethical issue (Baird, Zelin, & Brennan, 2006; Boateng & Agyapong, 2017; Costa et al., 2016; Mccuddy & Peery, 1996). A few scholars may have also investigated the effect of personal values but measured its impact on ethical behaviour only. This study investigates the impact of personal values and the perceived importance of an ethical issue on the ethical decision-making process. The study further employs Rest’s (1986) ethical decision model coupled with Robin et al.’s (1996) framework is used in a conceptual framework for the study. The researcher also introduces some values construct from the Scott (1965) personal values model. 2.2 Theoretical Review Ethical decision making and studies surrounding ethics on a whole have not seen much of theoretical involvements. Nonetheless, there have been a few theories which have been employed to explain the determinants of ethical behaviour and ethical decision making by individuals. The 23 University of Ghana http://ugspace.ug.edu.gh most predominantly used theory is this area of research is Rest’s (1986) four-stage ethical decision- making model. 2.2.1 Rest’s Four-Stage Ethical Decision-Making Model This theory is the most widely accepted, tried and validated ethical decision-making model in ethics literature (Haines et al., 2008; Robin et al., 1996). As the name suggests, the model includes four stages of ethical decision-making for an individual. This begins with “recognition of a moral issue” through to “engagement in moral behaviour”. The model is shown in the figure below. Figure 3.1: Rest's Four Stage Ethical Decision-Making Model Source: Robin et al. (1996) The first stage is the recognition of a moral issue (Moral Sensitivity). Moral sensitivity also referred to as moral awareness, is explained by Rest (1986) as an individual’s ability to detect that a situation has imbibed in it an ethical issue. Later research further explains moral sensitivity as the process of the decision maker’s recognition of moral content in a situation, and as a result, a moral perspective is valid. Recognising a moral issue involves the individual coming to a 24 University of Ghana http://ugspace.ug.edu.gh realisation that his/her actions have the potential to positively or negatively affect other people. That is their actions may harm or benefit others (Lincoln & Holmes, 2011; Robin et al., 1996). Moral judgment is the second stage of the ethical decision-making process, according to Rest. It refers to formulating and evaluating which of the alternative solutions to the moral issue has more moral justification and acceptance. This step in the process calls for the individual to reason through the possible choices and their potential consequences to find out which is more or most ethically sound (Guffey & McCartney, 2008; Lincoln & Holmes, 2011; Robin et al., 1996). The third stage involves the establishment of moral intention. The intention of the individual to choose a moral decision over other possible solutions representing different values is termed as moral intention (or moral motivation). This component of the ethical decision-making process involves committing to choosing a moral value. For example, an individual may recognize two solutions to a dilemma, one that will result in personal power and one that is right morally. The individual’s intention to choose the value of morality over the value of personal power in this instance represents moral motivation (Lincoln & Holmes, 2011; Robin et al., 1996). The final stage of Rest’s (1986) model is engagement in moral behaviour (moral action). Moral action points to the final behaviour of the individual in an ethical situation. This component refers to the individual’s action in the situation at hand. This stage of the decision-making process involves the ability, courage and determination to follow through with the moral decision (Lincoln & Holmes, 2011; Robin et al., 1996). 25 University of Ghana http://ugspace.ug.edu.gh However, Jones (1991) suggests that ethical decision-making is an issue contingent phenomenon. He, therefore, believes that there is more to the ethical decision-making process of individuals. As a result, Jones (1991) extends Rest’s model by introducing the moral intensity construct. 2.2.2 Moral Intensity According to Jones (1991), moral intensity refers to “a construct that captures the extent of issue- related moral imperative in a situation”. This construct is made up of six components, which helps in describing a moral issue. Moral intensity varies with variations in the underlisted components. These components are: (i) Magnitude of consequence – refers to the level of benefit or harm that an individual’s action will have on people; (ii) Social consensus – this implies the extent to which society approves or disapproves of an action as morally right or wrong; (iii) Probability of effect – it is the likelihood that an action’s outcome will happen or not and whether or not it will bring benefits or cause harm; (iv) Temporal immediacy – refers to the time lapse between the time of the action and the consequences. The shorter the time, the higher the immediacy; (v) Proximity – is explained as how close (in cultural, psychological, physical, or social terms) the agent of the action is to the victims of the consequence; and (vi) Concentration of effect – this refers to the number of people who will be affected by an action or decision which has a given level of impact. 26 University of Ghana http://ugspace.ug.edu.gh 2.3 Theoretical Framework In this study, Rest’s (1986) model is not employed in its entirety. The model is adapted, modified and some other constructs added. Moral intensity is replaced with Robin et al.’s (1996) empirically validated construct, perceived importance of an ethical issue (PIE). Also, the researcher incorporates in the model the personal values proposed by Scott (1991) and adapted by Akaah and Lund (1994). 2.3.1 Perceived Importance of an Ethical Issue (PIE) This construct was introduced by Robin et al. (1996) and subsequently tested empirically. PIE takes into consideration the beliefs, needs, perceptions, values, special characteristics of the situation, and existing personal pressures in the ethical decision-making process. Unlike Jones (1991) model (moral intensity) which only focuses on the characteristics of the issue or situation at hand, PIE goes further to include the individual traits of the decision maker. After proposing the theory, Robin et al. (1996), tested the construct empirically. Employing a scenario-based questionnaire, they found that there is a significant relationship between PIE and ethical judgement. The study showed that people who are high in terms of PIE are less likely to engage in unethical behaviour, whereas people who are low in terms of PIE are more likely to engage in unethical actions. Further studies have also tested this model empirically and have come out with similar results. For instance, Haines et al. (2008) surveyed 235 business students and their results and findings supported that of Robin et al. (1996). Also, Guffey and McCartney (2008), surveyed 397 accounting students and came up with similar findings. Hence it can be deduced that PIE is a useful and more comprehensive measure of perceived relevance of an ethical issue. 27 University of Ghana http://ugspace.ug.edu.gh 2.3.2 Personal Values Scott’s (1965) personal values scale is adopted and incorporated in the ethical decision-making model. This is because most studies adopt the Rokeach Value Scale (RVS), Schwartz Value Scale (SVS) which are deemed to be subjective, and hence suggested that other personal value scales should be resorted to (Alleyne et al., 2013; Shafer et al., 2001). Akaah and Lund (1994) also employed Scott’s (1965) personal values scale and improved on it, when surveying some marketing professionals. This study, in answer to Shafer et al.’s (2001) recommendations, employs this personal values scale of 16 items and 4 subscales measuring intellectualism, honesty, self- control and religiousness. Figure 3.2: Conceptual Framework 28 University of Ghana http://ugspace.ug.edu.gh 2.4 Hypotheses Development According to Alleyne et al. (2013), people who are high on personal values are less likely to judge an unethical issue as immoral. Further, Finegan (1994) also hypothesised that personal values are significant predictors of ethical judgement but found that only honesty had significant influence. The PIE construct is also seen as another important predictor of ethical judgement. In testing the PIE construct, Robin et al. (1996) proposed that PIE will significantly influence ethical judgement. Similarly, Haines et al. (2008), and Guffey and McCartney (2008), also hypothesised that individuals who score high in terms of PIE are more likely to judge immoral actions as unethical. Owing to the above, we hypothesise that: Hypothesis 1: there is a significant negative relationship between personal values and ethical judgement Hypothesis 2: there is a significant negative relationship between PIE and ethical judgement Considering the impact of personal values, PIE and Ethical judgement on behavioural intention, researchers opine that those who score high in terms of personal values and PIE are less likely to engage in unethical behaviour (Alleyne et al., 2013; Guffey & McCartney, 2008; Haines et al., 2008; Harris, 1991). On the other hand, it is opined that ethical judgment is positively related to behavioural intention. That is individuals who judge unethical as such, are less probably to engage in such unethical actions or behaviours (Guffey & McCartney, 2008; Haines et al., 2008). Following these prior literature, the following hypotheses were developed: Hypothesis 3: there is a significant negative relationship between personal values and behavioural intention 29 University of Ghana http://ugspace.ug.edu.gh Hypothesis 4: there is a significant negative relationship between PIE and behavioural intention Hypothesis 5: there is a significant positive relationship between ethical judgement and behavioural intention. 30 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE METHODOLOGY 3.0 Introduction This chapter seeks to inform the reader of the process, methods and procedures employed by the researcher in order to achieve the objectives of this study. The chapter looks at the approach and design, the data collection technique, the sampling procedure, the development and administration of the survey instrument and the data analysis technique. 3.1 Research Paradigm Paradigm is defined as “the basic belief systems or worldview that guides the investigator” (Guba & Lincoln, 1994). This refers to the way in which a social phenomenon is examined for the purpose of gaining understanding and attempting explanations. The research paradigm is based on the ontology, epistemology, and axiology of the researcher (Saunders, Lewis, & Thornhill, 2009; Sobh & Perry, 2006). But more importantly, it is dependent on the nature of the research objective that the researcher seeks to achieve. The research paradigm also informs the philosophical stance of the researcher (Saunders et al., 2009). 3.2 Philosophical Stance Owing to the objectives of the research, which is to find the relationship between personal values, perceived importance of an ethical issue and ethical decision-making process, the researcher aligns himself with the positivist’s philosophical stance. This paradigm assumes that the end of a research can lead to a law-like generalization and hence only observable phenomena can provide credible 31 University of Ghana http://ugspace.ug.edu.gh information (Guba & Lincoln, 1994). This stance helps the researcher test the assumptions and hypotheses generated and allows the researcher to generalize the findings of the study as well (Saunders et al., 2009). It should be noted that the researcher believes social entities exist in reality that is outside the actors concerned (Saunders et al., 2009). Further, the researcher believes the truth does not change. That is, the researcher is convinced that truth is objective and hence employs the objectivist ontology. Epistemology is driven by the ontological beliefs of the researcher, due to the objectivist ontology of the researcher, the researcher goes ahead to employ the “etic” epistemology. That is the researcher believes that knowledge is discovered through objective ways and hence the researcher stays as far away from the research as possible. This is to prevent the researcher from influencing the results. This goes further to explain the axiology of the researcher. Axiology simply refers to the values of the researcher, and how those values may influence the findings or results of the research. As a result, the researcher in this study stays neutral and away from the results of the research. 3.3 Research Approach and Design Research design refers to the procedures of enquiry (Saunders et al., 2009). This, to some extent, influences the decision to select the appropriate research approach. The research approach refers to the plan and procedures that span the steps from broad assumptions to detailed methods of data collection, analysis and interpretation (Creswell, 2003; Saunders et al., 2009). It outlines the way in which the whole study is conducted. It considers the way data is gathered, analysed and interpreted. It also defines how the sample is selected. The choice of research approach is 32 University of Ghana http://ugspace.ug.edu.gh dependent on the objectives of the study. Also, the selection of a research approach is influenced by the philosophical assumptions guiding the thoughts of the researcher (Saunders et al., 2009). Due to the objective of the research, the researcher chooses to employ a quantitative approach to research. The quantitative approach is a research approach that allows the testing of theories by the investigator (Creswell, 2003; Saunders et al., 2009). This is done by examining the relationship among variables of a theory or framework (whether conceptual or not). The aim of this study is to find the relationship between personal values and perceived importance of an ethical issue, and ethical decision-making. Therefore, the most suitable method to employ is the quantitative approach which allows for the hypotheses to be tested and to either confirm or debunk the propositions of the theory. As such the researcher following the positivistic paradigm, adopts the quantitative research approach, using statistical procedures to test the relationships between variables. 3.4 Unit of Analysis This refers to the level at which the study will be generalized. That is the object or subject to be studied and about which conclusions will be drawn and generalizations made. It answers the question, “who or what is being analysed?” (Saunders et al., 2009). The unit of analysis is normally dependent on the level of analysis. The levels of analysis are micro, meso or macro levels, and their corresponding units of analysis are individual, firm (organization or group), country or continent. According to Yin (1994), the best of a general guide for determining the unit of analysis is by reflecting on the research objectives questions (Saunders et al., 2009). 33 University of Ghana http://ugspace.ug.edu.gh For the purpose of this study, the researcher seeks to determine the relationship between personal values and the perceived importance of an ethical issue and ethical decision-making process. In reference to the main research objective, the appropriate level of analysis is micro or individual level and the unit of analysis is university accounting students. 3.5 Population According to Saunders et al. (2009), population refers to the total number of cases from which samples can be chosen. That is, a population is a category of objects (be it human or non-human) that a researcher seeks to examine a portion of (Alvi, 2016). The outcome of the research is beneficial to the whole population. The individual objects are normally known to have some similar features or characteristics (Alvi, 2016; Saunders et al., 2009). For the purpose of this study, the target population was accounting students across all tertiary institutions in Ghana. That is all the accounting students in all the 91 tertiary institutions whether public or private. This means a very large number of students and very difficult to reach and access, hence the need to select a sample. 3.6 Sample and Sampling Strategy A sample is explained as a selected number of cases or objects that are taken out of a large pool and generalized back to the population (Alvi, 2016; Barreiro & Albandoz, 2001; Saunders et al., 2009). Sampling is, therefore, the process of choosing a set of objects from a population so that after studying the sample, a generalization of the results can be made back to the population from which the sample was taken (Alvi, 2016; Barreiro & Albandoz, 2001). 34 University of Ghana http://ugspace.ug.edu.gh To be able to know which group is being selected for the purpose of the study, there is the need to establish a basis for selecting that sample and this is known as the sampling technique. Sampling technique is defined as a reasonable approach to selecting a sample that gives a representative view of the population as a whole. (Barreiro & Albandoz, 2001; Saunders et al., 2009). For this study, the researcher chose the purposive sampling technique. The choice of this sampling technique was based on the nature of the study and the objective the researcher sought to achieve. Purposive sampling is a technique that allows the researcher to select a sample based on certain special features or characteristics of the participants or subjects, which makes them best suited for the study (Saunders et al., 2009). There about ninety-one (91) accredited tertiary institutions offering degree programs in Ghana. They are broadly divided into two, public, consisting of ten (10) institutions and private ones (81 institutions). Out of these, about 34 of them offer business (and accounting) programs (7 public and 27 private) as of 2016 (Simpson et al., 2016). To have a fair representation of students from the two categories of universities, two (2) institutions were selected from each block. Third and fourth year accounting students were then randomly selected to participate in the survey. The two public universities selected were the University of Ghana (UG) and the University of Professional Studies, Accra (UPSA). The University of Ghana was chosen because it is the premier university in the country and it also attracts a diversified student body, both domestic and foreign. Also, UPSA was selected due to the reputation it has built for itself as a university focused on training and bringing up professional accountants. Central University College (CU) and Valley View University (VVU) were the private universities included in this study. This is because, these two are the first private institutions that have been issued a Charter by the National Accreditation 35 University of Ghana http://ugspace.ug.edu.gh Board of Ghana to operate as autonomous private universities in the country, as at the time of the study. A distribution of the number of responses received from each institution is shown in table 1 of appendix A. 3.7 Research Instrument The data collection tool employed in this study was a self-administered questionnaire. The questionnaire consisted of four sections. The first section covered issues relating to the perceived importance of an ethical issue (PIE) and ethical decision-making process, using a scenario-based approach. The scenario-based approach has been successfully employed in ethical decision- making literature (Guffey & McCartney, 2008; Haines et al., 2008; Robin et al., 1996). The stages of the ethical decision-making process that were considered are ethical judgement and behavioural intention. Two scenarios were selected for the first section of the questionnaire based on Ameen et al.’s (1996) criteria, as employed also by Guffey and McCartney (2007). The criteria used were that, first, the scenarios should be statistically different from each other in terms of seriousness and secondly, they should not be extreme cases. The less unethical scenario chosen described a student who failed to report a favourable mistake with her examination score. The second scenario, which was the more unethical case, was about a student who discovered a copy of an upcoming examination paper and sent it home to help in preparation towards the exam, rather than reporting it. A very crucial element to ethical decision-making is a reliable and valid scale. This study employed a PIE scale developed by Robin et al. (1996). This scale was a development from a 1985 study by 36 University of Ghana http://ugspace.ug.edu.gh Zaichowsky. The PIE scale has been tested and approved by a few studies (Guffey & McCartney, 2008; Haines et al., 2008; Zaichkowsky, 1985). The ethical judgement of the respondents was also measured using two different scales, a multivariate moral equity scale (MES) and a univariate ethics measure (UEM). This was adopted from Robin et al. (1996) and Flory et al. (1992). The MES measures or evaluates decisions in terms of their inherent fairness, justice, goodness and rightness (Flory et al., 1992; Robin et al., 1996). The UEM was used as a validity check for measuring decisions as either ethical or unethical (Flory et al., 1992; Robin et al., 1996). In addition to the above, behavioural intention was to assess the possibility that a subject will act in the same way as the character in the scenarios. This was measured on a 7-point Likert scale (highly improbable/highly improbable) by a single question which read, “If you were responsible for making the decision in the scenario above, what is the probability you would make the same decision?”. The second part of the questionnaire assessed the ethical behaviour of the subjects. The Akaah and Lund’s (1994) modified seventeen-item scale was adopted to assess the likelihood that the respondents will engage in unethical behaviour. The seventeen (17) item scale measures six (6) unethical behaviours in a work environment, which are personal use (4 items), passing blame (3 items), bribery (2 items), falsification (3 items), padding expenses (2 items) and deception (3 items) (Akaah & Lund, 1994). These items were all measured on a seven-point Likert scale, ranging from extremely unlikely to extremely likely. Personal values of the subjects were measured in the third section of the questionnaire. Following Alleyne et al. (2013), four personal values were selected from the personal values scale established by Scott (1965) and adapted by Akaah and Lund (1994). This scale consisted of sixteen items 37 University of Ghana http://ugspace.ug.edu.gh measuring four personal values. The scale had four subscales of four items each, measuring intellectualism, honesty, religiousness and self-control (Akaah & Lund, 1994; Lau, 1988). The items were also measured on a 7-point Likert scale, ranging from “strongly dislike it” (rated as 1) to “strongly like it” (rated as 7). The final section of the questionnaire sought to elicit some demographic information from the respondents. Such information included gender, age, level, and the institution of the respondents. Also, the religious status and religion of the respondents were asked as well as the number of courses that the respondents had taken which had ethics components. The above-listed information was to help the researcher describe the respondents. 3.8 Ethical Consideration Ethical consideration is a very critical issue of every research paper. It helps to know the difference between right and wrong, acceptable and unacceptable attitudes or behaviour to research. As such, the researcher went through all the necessary ethical clearance procedures outlined by the university. First, the researcher filled all the necessary forms and prepared the relevant documents. These documents were then submitted to the Institute of Social and Statistical Research (ISSER). The Institute followed due diligence and assessed all the documents and ascertained that all ethical rules have been adhered to. The Institute finally accepted and approved for the research to be carried out. Following the above, the researcher also assured respondents of the following; (i) First of all, the respondents would not be forced under any circumstance to answer the questionnaire. 38 University of Ghana http://ugspace.ug.edu.gh (ii) The respondents could also decide, at their own free will, to opt out of the survey at any point in time. (iii) The researcher would also keep all respondents anonymous and hence questions of such nature are not asked and any personal detail of respondents would not be revealed to the readers. 3.9 Pilot After the instrument was designed, it was tested through a pilot study of about seventy-two (72) respondents. A sample of students from the University of Ghana Business School was made to answer the questionnaire. The respondents gave their feedback concerning clarity and how precise the instrument was. Generally, they suggested that the instrument was very clear and communicated the exact information that they are supposed to communicate. Their responses were also recorded and further analysed. The results showed that the measurement items well measured the constructs that were being studied. 3.10 Administration and Collection of Questionnaire For the purpose of achieving the aims and objectives of this study, a survey research method is employed. This is because this approach helps to provide standardised information to explain variables and/or to investigate the relationships between variables (Saunders et al., 2009). Hence, the researcher ascribes to survey research since it is the most suitable approach which helps the researcher to collect data from the respondents. Thereafter, the data is used in examining the relationship between personal values, PIE and ethical decision-making. According to Yin (1994), a survey research approach is the most appropriate method when it comes to studies that answer 39 University of Ghana http://ugspace.ug.edu.gh the question, “who?” or “what?”. Hence, considering the objective of this study, the researcher employed the survey approach. The data was collected directly from the respondents using a closed-ended questionnaire. That is to say that, the researcher resorted to primary data only. The researcher met the students in their classrooms, addressed them face-to-face and administered the questionnaires to them. After, the researcher waited and collected the questionnaires back. A total of 700 questionnaires were distributed to respondents, out of which 465 were returned. This indicates a response rate of 66.42%. Out of the 465 questionnaires returned, 21 were not usable. This was due to the fact that the 21 were not fully filled, that is the respondents had only filled a section out of the four sections. Finally, four hundred and forty-four (444) questionnaires were used in the analysis, which represents 63.43% of the total number of questionnaires distributed and 95.48% of the total number of questionnaires that were returned. 3.11 Data Analysis Techniques Data analysis is a process of inspecting and modelling data with the aim of obtaining or coming out with useful information that informs conclusions and also supports final decision-making. More specifically, quantitative data analysis involves critical analysis and interpretation of figures and attempts to find the reasons behind such findings. Due to the exploratory nature of the study and the approach employed, data was analysed using Partial Least Square Structural Equation Modelling (PLS-SEM). 40 University of Ghana http://ugspace.ug.edu.gh 3.12 Structural Equation Modelling (SEM) SEM is a combination of factor analysis, path analysis and multiple regression analysis. SEM is a multivariate statistical technique that is used to analyse or examine the direct as well as indirect relationships between one or multiple independent latent variables and one or more dependent variables. It provides the researcher with the opportunity to model, simultaneously estimate and test complex theories with empirical data (Sarstedt, Ringle, Smith, Reams, & Hair, 2014). In addition to evaluating the hypothesised structural relationships among variables, SEM also evaluates the relationships existing between variables and their respective measurement items (Gefen & Straub, 2005; Gefen, Straub, & Boudreau, 2000). SEM can also be seen as a flexible modelling tool for conducting numerous multivariate statistical analyses. There are primarily two approaches to SEM. These are the component-based approach, more specifically, partial least square (PLS-SEM) and the co-variance-based approach (CB-SEM) (Fornell & Bookstein, 1982; Marcoulides, Chin, & Saunders, 2009; Wetzels, Odekerken-Schroder, & van Oppen, 2009). These two approaches differ from each other in terms of underlying statistical assumptions and the nature of the fit statistic they produce (Gefen et al., 2000). 3.12.1 Covariance Based Structural Equation Modelling (CB-SEM) CB-SEM employs the maximum likelihood function to reduce the differences between the sample covariance and that which is predicted by the theoretical model. The estimated parameters in CB- SEM tries to reproduce the covariance matrix of the observed values. In applying the maximum likelihood function, two conditions have to be met. These are, first, the variables must follow a normal distribution and secondly, the observations should be independent of one another (Chin, 41 University of Ghana http://ugspace.ug.edu.gh 1998; Hair, Ringle, & Sarstedt, 2011; Urbach & Ahlemann, 2010). CB-SEM is basically used to confirm or reject a theory. This is done by finding how well an established theoretical model can estimate the covariance matrix of a sampled data (Hair, Hult, Ringle, & Sarstedt, 2014; Sarstedt et al., 2014). 3.12.2 Partial Least Square Structural Equation Modelling (PLS-SEM) On another hand, PLS-SEM is basically used in developing new theories in exploratory research. In this approach, this is done by focusing on explaining the dependent variables’ variances when examining the model. PLS has as its main objective to optimise the covariance between the predictor latent variables and the dependent latent variables (Sarstedt et al., 2014). PLS-SEM employs least square estimations for single and multi-component models and for canonical correlations (Chin, 1998). A notable characteristic of the PLS-SEM approach is that it eliminates the numerous restrictive assumptions that underlie the maximum likelihood techniques. It also ensures that improper solutions and factor indeterminacy are eliminated or reduced (Fornell & Bookstein, 1982). Basically, PLE-SEM is “a non-parametric, multivariate approach based on iterative Ordinary Least Square (OLS) regression to estimate models with latent variables and their directed relationships” (Avkiran, 2018, p. 3). Despite the unique benefits and use of PLS-SEM, it has received a number of criticisms as being less rigorous and hence low acceptance and application in examining relationships between constructs in the past. However, in recent times, PLS-SEM has been increasingly applied in marketing and other business and non-business disciplines (Henseler, Ringle, & Sarstedt, 2015). Some scholars now regard PLS-SEM method as being a more robust estimation of structural 42 University of Ghana http://ugspace.ug.edu.gh models (Henseler et al., 2015). It is also seen as an alternative method for CB-SEM when the distributional assumptions of CB-SEM cannot be met (Hair et al., 2014). 3.12.3 Choosing between CB-SEM and PLS-SEM To be able to select which of the two approaches to adopt, the researcher must focus on the objectives and characteristics of the two methods. According to Hair et al. (2014), four main lines of guidance are suggested. These characteristics (research goal, data, model properties, and model evaluation) are discussed hereafter. Firstly, the goal or aim and objective of the research is a very important element in determining the approach to use. CB-SEM is primarily designed to be used in confirming theories. In testing theory, it requires showing how well the model fits the data. This involves hard modelling with the aim of minimizing the covariance matrix. The above points are said to be the strengths of CB-SEM. On the other hand, PLS-SEM is deemed to be more appropriate for theory development. Theory prediction and development involves soft modelling, which focuses on maximizing the level of covariance between constructs or latent variables so as to improve on the model prediction. This feature is a strong characteristic of PLS-SEM. Secondly, the data characteristics in terms of size, distribution, missing values and scale are also key elements to consider when selecting which approach to use. Studies that employ smaller sample sizes and non-normally distributed data will be best analysed using PLS-SEM. This is because, PLS-SEM does not require data to be normally distributed since it employs a more robust approach, using calibration mechanisms that transforms the non-normal data. On the contrary, CB- SEM requires data to be normally distributed and also of a large sample size. If these requirements are not met, the results will be very inaccurate (Hair et al., 2014). It is noteworthy that PLS-SEM 43 University of Ghana http://ugspace.ug.edu.gh is not immune to adequate or large sample sizes (Marcoulides & Saunders, 2006; Sarstedt et al., 2014). Another key factor to be considered is the model properties. According to Hair et al (2014), PLS- SEM is able to handle complex models, that involves several constructs (above five) and also several indicators measuring a construct (normally, greater than six indicators). Also, it is designed to manage constructs with a single measurement item as well, in order to generate a comparatively higher measure of statistical power and attain convergence. In addition, PLS-SEM easily incorporates formative (indicators cause the construct) and reflective (indicators are caused by the construct) measurement models. On the other hand, CB-SEM is known to work better with non- complex and non-recursive relationships. Also, it well accommodates models that use reflective models. Lastly, to select between the SEM approaches, researchers have to be aware of the model evaluation as well. For PLS-SEM, it assesses the model using the latent variable scores in subsequent analysis. CB-SEM, on the other hand, requires global model goodness of fit criterion and also need to test for measurement model invariance (Sosik, Kahai, & Piovoso, 2009). Considering the nature and aim of this study, the researcher deemed it most appropriate to employ PLS-SEM rather than CB-SEM. This is because, first, the objective of the study to some extent, seeks to develop a new theory, by employing personal values as determinants of ethical decision- making. Also, the model employed is complex in nature with some of the constructs having as many as ten indicators. In view of the points raised, the researcher, employs PLS-SEM approach in data analysis. 44 University of Ghana http://ugspace.ug.edu.gh 3.13 Reliability For exploratory studies of this nature, two things are very important to validate the measures and the constructs employed in the study; reliability and validity. Reliability is generally concerned with how robust a questionnaire is, and more particularly, whether or not it will produce the same or consistent outcome when employed at different times and under different conditions. For instance, it is expected that a reliable questionnaire will produce similar findings even if it is used to assess different categories of samples or, in the case of an interviewer-administered questionnaire, with different interviewers (Saunders et al., 2009). According to Mitchell (1996), there are three approaches to assessing reliability, in addition to the manual comparison of data collected with other data from a number of sources (Mohamed, Mad Shah, & Jusoh, 2016). These are test re-test, internal consistency and alternative form. Even though these analyses approaches can only be done after data collection, the researcher needs to consider them as early as the questionnaire development stage. To assess the reliability of the constructs, the researcher employed the composite reliability measure. According to Hair et al. (2014), a composite reliability between 0.70 and 0.95 is acceptable. Further, the reliability of the indicator variables is also examined. The researcher employed the outer loadings of the indicators. The square of the outer loadings gives the communality which explains how much variation in the indicator is explained by the latent variable. Higher outer loadings of 0.70 and above are preferred, and that means the indicators are reliable (Aibinu & Al-Lawati, 2010; Avkiran, 2018; Hair et al., 2014; Wong, 2013). 45 University of Ghana http://ugspace.ug.edu.gh Aligning the scores of the various constructs with the benchmark of between 0.70 and 0.95, it was realised that all the constructs have a composite reliability measure that falls within this range. Hence, there exist acceptable levels of internal consistency within the various constructs employed in the study. Indicator reliability was also checked using the outer loadings. Four of the measurement items were deleted due to the fact that, the score fell below the threshold of 0.40 (PIE1= -0.100, PIE2= -0.219, PIE3= -0.247, and PIE4= -0.122). Also, all the other measurement items showed internal reliability greater than 0.70 except MES11=0.696, MES13=0.693, MES14=0.698, UEM1=0.678 and UEM11=0.612. These items were maintained because, their removal leads to a decrease in the AVE and composite reliability of the ethical judgement measure (Hair et al., 2014; Longenecker et al., 2004; Modarres & Rafiee, 2011b; Sarstedt et al., 2014). 3.14 Validity Internal validity simply refers to the ability of a questionnaire to be able to measure the item for which it has been designed. That is, the researcher is concerned that the findings or outcomes of the questionnaire do represent the reality of the situation which the researcher seeks to measure. According to Saunders et al (2009), the problem with validity is that, if the researcher knows the reality of what he/she seeks to measure, then there will be no need to design a questionnaire for the same purpose. To overcome this, the researcher must look for relevant evidence that supports the outcome of the questionnaires. The most common approaches to validity measurement are content validity, criterion validity and construct validity (Heale & Twycross, 2015; Kimberlin & Winterstein, 2008). Content validity is explained as the extent to which the questions in the questionnaire (measurement device) provides substantial coverage of the investigative questions. Deciding on what is “adequate coverage” can 46 University of Ghana http://ugspace.ug.edu.gh be done through a variety of approaches. According to Saunders et al. (2009), it can be done through a careful definition of the research by using the literature reviewed. Criterion-related validity is sometimes referred to as “predictive validity”. It is the ability of the measures (questions) to accurately predict the items it has been designed for (Saunders et al., 2009). In assessing criterion-related validity, there is a comparison between the data collected with the questionnaire (measurement instrument) and the specified criterion in some way. This analysis is mostly undertaken using statistical analysis techniques such as correlation. Construct validity is defined as the extent to which a question measures the item for which it was designed and also performs accurately, the functions it is expected to perform (Mohamed et al., 2016; Saunders et al., 2009). The term, Construct validity is normally used when referring to construct such as attitudes scale and personality tests and other scales answering the question, “How well can you generalise from your measurement questions to your construct?”. It is suggested that validating such constructs against already existing literature or data is difficult and hence, other methods are employed. In this study, convergent validity is assessed to determine whether or not there is internal consistency among the measurement items of the various latent variables. In testing the convergent validity, Aibinu and Al-Lawati, (2010) propose three measures, the Cronbach alpha, composite reliability and average variance extracted (AVE). However, Hair et al. (2014) suggest that convergent reliability is measured using the AVE. AVE is the sum of the squared loadings divided by the number of indicators. Researchers suggest that an AVE of 0.50 or higher is acceptable since it signifies that the construct explains more than 50% of the variations in that measurement item, 47 University of Ghana http://ugspace.ug.edu.gh and hence the items converge well on the construct. Also, the validity of the constructs is necessary to be verified by assessing the discriminant validity of the model. First, the convergent validity was examined using the AVE and it showed that all the constructs have AVEs greater than the desired rate of 0.50, ranging from 0.505 to 0.771. This implies that more than 50% of the variance in the items were being explained by their corresponding latent variables. Also, the discriminant validity was assessed using the cross-loadings. It was realised that all the items load highly unto the latent variables they measure more than they load unto any other variable. Also, the HTMT ratio was employed to further examine the discriminant validity among the constructs. It was realised that all the pairs of constructs were having HTMT ratios less than 0.90 for all pairs of constructs, ranging from 0.111 to 0.750. It can be concluded from the tests and their corresponding results that there is enough evidence to support the reliability and validity of the various measurement items and their corresponding latent variables. 3.15 Summary of Chapter In summary, this chapter elaborates on the philosophical stance of the researcher. Also, the method and approach employed in the conduct of this study have been explained. The researcher also discussed the analytical techniques employed for the purposes of this research. Finally, issues of reliability and validity have been well explained and treated in this chapter. In the next chapter, the researcher presents the results and discussions of the results generated from the analysis of the data. 48 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR PRESENTATION OF RESULTS AND DISCUSSION OF FINDINGS 4.0 Introduction In this chapter, the results from the analysis of the data gathered are presented. Further, the results are discussed in line with the objectives of the study. First, demographic analysis is presented. A thorough analysis of the validity and reliability of the model is also presented. Finally, the structural model is discussed, taking them objective after objective to ensure the aim of the research is achieved. The chapter is closed with a summary of the relevant outcomes of the study. 4.1 Demographic Analysis The gender distribution showed that respondents consisted of 240 males and 204 females. The males constituted about 54.05% of the sample whereas the females represented 45.95%. Out of the 240 males, only 44 were level 300s the remaining 196 remaining were final year students of the various universities. Also, only 60 out of 204 females were level 300 students with the remaining being final year students. A distribution of the samples based on gender is provided in Table 1 of Appendix A. The respondents were also asked to provide their age range. This showed that majority of the respondents, accounting students were found between the ages of 23-25 years which represents about 55.60%. This is followed by those within the 18-22 years age bracket who are 162 in total constituting 36.5% of the total sample. Only three students indicated that they were above 30 years, with about thirty of the students within the ages of 26 and 30 years. Also, it was revealed that the 49 University of Ghana http://ugspace.ug.edu.gh female population are generally younger than the males. This is shown in the distribution, where about 25.60% of the females are within 23-25years age bracket, we find the males to be about 62.00%. Out of the 30 in the 26-30 years age bracket, only 3 were females representing 10% of the total number and just 1.50% of the total number of females. However, two of the respondents, who were females and in their third year, chose not to disclose their ages (see Table 3 of Appendix A). The researcher went forward to ask whether the respondents are religious and their religious affiliation. It turned out that the majority of the students are religious with just about 27 students, representing 6.08% of the sample, not being religious. The 93.92% which are religious, 14 are Muslims and the remaining 403 are Christians. 15 students out of the 27 unreligious were found to be females whereas the other 12 males (see Table 5 of Appendix A). 4.2 Mean Differences In the summary table, Table 4.1 below, the researcher presents a summary of the mean scores of the various items and their respective constructs. To begin with, it can be observed from the table below that averagely, the students more probably will engage in the unethical actions portrayed in the scenarios (mean score of BI1=4.600; BI11=4.550). Behavioural intention had a mean score of 4.573, which implies that the accounting students are more likely to engage in unethical behaviours as exhibited in the scenarios. On the other hand, the mean score of the ethical judgement of the subjects shows results to the contrary. The subjects recorded an average mean score of 3.714 for ethical judgement. This implies that accounting students are more likely to judge immoral and wrong attitudes as unethical. This is evident in the way accounting students perceived the behaviour in the scenarios. PIE has an average mean score of 4.717 (which is greater than 4.00). 50 University of Ghana http://ugspace.ug.edu.gh This implies the students attach greater levels of relevance to the situations in the scenarios and hence their judgement. Moreover, accounting students perceive themselves to exhibit higher levels of personal values. The mean scores of the four facets of personal values are all above four for the constructs and also the measurement items (intellectualism=5.048, honesty=4.745; self-control=4.864; religiousness=5.273). It is evident from the mean scores that religiousness is the personal value that accounting students exhibit most followed by intellectualism, and honesty being the least exhibited. Table 4.1: Construct and Measurement Item Means Std. Code Item Mean Deviation Behavioural Intention 4.573 1.667 “If you were responsible for making the decision in the scenario above, what is the probability you would make the same BI1 decision?” 4.600 1.973 “If you were responsible for making the decision in the scenario above, what is the probability you would make the same BI11 decision?” 4.550 1.995 Ethical Judgement Overall 3.714 1.379 MES1 “Unfair/Fair” 3.980 2.056 MES2 “Unjust/Just” 3.800 1.946 MES3 “Not morally right/morally right” 3.620 2.015 MES4 “Not acceptable to my family/acceptable to my family” 3.970 1.982 UEM1 “Unethical/ethical” 3.590 1.937 MES11 “Unfair/Fair” 3.700 1.914 MES12 “Unjust/Just” 3.720 1.864 MES13 “Not morally right/morally right” 3.330 1.862 MES14 “Not acceptable to my family/acceptable to my family” 3.860 1.889 UEM11 “Unethical/ethical” 3.600 1.814 Continued on the next page… 51 University of Ghana http://ugspace.ug.edu.gh Std. Code Item Mean Deviation Perceived Importance of an Ethical Issue 1 4.717 1.658 PIE11 “Unimportant issue/important issue” 4.650 1.955 PIE12 “Insignificant/significant” 4.790 1.756 PIE13 “Of no concern/of high concern” 4.790 1.799 PIE14 “Trivial issue/fundamental issue” 4.750 1.821 Intellectualism 5.048 1.423 Int1 “Having an active interest in all things scholarly” 4.910 1.785 “Having a keen interest in international, national and local Int2 affairs” 5.110 1.650 “Developing an appreciation of the fine arts – music, drama, Int3 literature, and ballet” 4.970 1.715 Int4 “Having a strong intellectual curiosity” 5.200 1.639 Honesty 4.745 1.424 “Never cheating or having anything to do with cheating, even Hon1 for a friend” 4.530 1.858 “Always telling the truth even though it may hurt one’s self or Hon2 others” 4.900 1.681 “Speaking one’s mind truthfully, without regard for the Hon3 consequences” 4.980 1.654 “Volunteering information concerning wrongdoing, even if Hon4 friends are involved” 4.570 1.702 Self-Control 4.864 1.478 SelCo1 “Never losing one’s temper, no matter what the reason” 4.590 1.768 SelCo2 “Practicing self-control” 5.160 1.650 “Not expressing anger, even when one has a reason for doing SelCo3 so” 4.750 1.802 SelCo4 “Replying to anger with gentleness” 4.960 1.805 Religiousness 5.273 1.605 Rel1 “Being devout in one’s religious faith” 5.200 1.848 Rel2 “Always living one’s religion in one’s daily life” 5.190 1.788 Rel3 “Always attending religious services regularly and faithfully” 5.220 1.768 Rel4 “Having faith in a being greater than man” 5.480 1.875 52 University of Ghana http://ugspace.ug.edu.gh 4.2.1 Public vs Private Tertiary Institution – Personal Values and Ethical Judgement After evaluating the means, the researcher further examined the significant differences in means using independent t-tests. First, independent t-tests were conducted to compare the means and find out if there are any significant differences in personal values and ethical judgement across the public and private tertiary institutions. The results showed that there are significant differences in the mean scores for personal values of honesty (t-statistic=2.139, p-value=0.033) and self-control (t-statistic=3.018, p-value=0.003) between the accounting students of public and private tertiary institutions. These values imply that at a significance level of 5%, there is enough evidence to support the assertion that accounting students of public institutions and those of the private institutions exhibit different levels of honesty and self-control. The summary results showed that accounting students of public institutions are generally more honest (mean=4.873) and also exhibit higher levels of self-control (mean=5.052) than their counterparts in the private institutions (honesty: mean=4.584; self-control: mean=4.629). This is contrary to the observation that private institutions are mostly religious affiliated and more likely to include ethics courses, hence being more ethical than public institutions (Simpson et al., 2016). Further t-tests were conducted to investigate the differences in the ethical judgements of accounting students in public and private tertiary institutions. This was done on scenario basis and finally, taking both scenarios as a single whole. The mean score for public institutions were 3.938, 3.698, and 3.816 for scenario one, scenario two and overall scenarios respectively. Similarly, accounting students of private recorded mean scores of 3.601, 3.567, and 3.585 for scenarios one, two and overall scenarios. Generally, students from both types of institutions regarded the action 53 University of Ghana http://ugspace.ug.edu.gh taken in both scenarios as unethical. However, the students of the private institutions regarded the actions in both scenarios as more unethical than their colleagues in public institutions. The results of the independent t-test indicated that there is no significant difference between the mean scores of accounting students in public and private tertiary institutions with regards to their ethical judgement of the overall scenarios (t-statistic=7.765, and p-value=0.078). Upon taking the scenarios individually, scenario 2 also showed no difference in mean scores of the ethical judgement of students of the two categories of institutions (t-statistic=0.891, p-value=0.374). On the contrary, it was revealed that for scenario 1 there was a significant difference between the ethical judgement of students of the two categories of institutions (t-statistic=2.133, p- value=0.033). That is there is no or not enough evidence supporting the hypothesis in this regard. Hence, we can conclude that there is no significant difference between the ethical judgements of accounting students across private and public institutions. 4.2.2 Male vs Female Accounting Students – Personal Values and Ethical Judgement In examining the differences between the mean scores of male and female accounting students in terms of personal values, the results showed that there are no significant differences between male and female students on all facets of personal values (intellectualism: p-value=0.663; honesty: p- value=0.988; self-control: p-value=0.517; and religiousness: p-value=0.996). This is shown in the mean scores of both gender groups as shown in Table 3 of Appendix B. This result is contrary to those findings of Ismail (2015), and Eaton and Giacomino (2001) who found significant differences in the personal value scores of males and females. However, in a study by Baird, Zelin and Brennan (2006), they found that males and females significantly differ in terms of 5 out of 10 54 University of Ghana http://ugspace.ug.edu.gh personal values. Apart from power where the males were significantly higher, the females were significantly higher in all other personal values (Baird et al., 2006). The final test of means was conducted on the ethical judgement of male and female accounting students. Following the approach used in assessing the tertiary institutions, we take the cases one after the other and then both as a whole. First, when the scenarios were taken as a single whole, it was realised that there is no significant difference between the ethical judgement score of male and female accounting students (p-value=0.075). The cases were subsequently taken one after the other. Scenario 1, just like the overall scenario, revealed that there is no significant difference in means ethical judgement score of male and female accounting students (p-value=0.359). This implies that there is not enough evidence to suggest that there exists a significant difference between male and female accounting students’ ethical judgement. Conversely, in scenario 2, we find a significant difference between the mean ethical judgement score of male and female students (p-value=0.029). That is, there is enough evidence to suggest that there exists a significant difference in the mean scores of the ethical judgement of male and female students. These results generally imply that male and female accounting students’ judgements of the actions in the scenarios are no different from each other. The mean scores also suggest the above. That is both male and female accounting students judged the actions taken in the scenarios as unethical. Males had means of 3.855, 3.710 and 3.821 for scenario 1, scenario 2 and overall scenario, and the female students had means of 3.710, 3.468 and 3.588 for scenario 1, scenario 2 and overall scenario respectively. Generally, the females judged the actions as more unethical than the male accounting students. That notwithstanding, the differences in the means are not significant. 55 University of Ghana http://ugspace.ug.edu.gh 4.3 Measurement Model In presenting the empirical results from PLS-SEM, we begin by evaluating the measurement model. Here, the research assesses the reliability and validity of the constructs and their respective measurement items. Since the indicators were reflective measures, we begin the reliability assessment with indicator reliability. 4.3.1 Indicator Reliability The indicator reliability is measured using the item loadings. Considering outer loadings, higher scores are preferred. As a rule of thumb, a standardised outer loading of 0.708 and above is deemed to be significant and implies that the researcher has a reliable indicator (Avkiran, 2018; Hair et al., 2014). All the indicators of the various constructs in this survey meet the threshold of 0.708 (see Table 4.2), except five which were less (these are MES11=0.696, MES13=0.693, MES14 = 0.698, UEM1=0.678 and UEM11=0.612). It is opined that items with loadings between 0.400 and 0.700 should be assessed to ascertain that the elimination of the item will not lead to a reduction in the composite reliability and the average variance extracted of the corresponding construct (Aibinu & Al-Lawati, 2010; Avkiran, 2018; Hair et al., 2014; Sarstedt et al., 2014). Apart from the fact that the loadings of these items were all approximately 0.70, it was observed that removing them led to lower composite reliability for that construct (ethical judgement), and hence essential for the indicators to be maintained. 4.3.2 Internal Consistency Reliability The internal consistency of the constructs was evaluated using composite reliability measure. The composite reliability is used instead of the Cronbach’s alpha due to the fact that it is sensitive to 56 University of Ghana http://ugspace.ug.edu.gh population or sample size used for the study. Hence if the sample is not large (about 200), Cronbach’s alpha may estimate a low internal consistency. Aibinu and Al-Lawati (2010), also argue that composite reliability is superior to Cronbach’s alpha because the former employs the item loadings obtained within the theoretical model in computing its measure. Nunnally (1978) posits that a composite reliability of 0.70 and above is more acceptable, indicating higher levels of reliability (Aibinu & Al-Lawati, 2010; Hair et al., 2014). As reported in Table 4.2, all the variables show good internal consistency. All the constructs show composite reliability values between 0.70 and 0.95. For instance, religiousness has the highest composite reliability of 0.931 and behavioural intention has the lowest of 0.827. It can, therefore, be concluded that all the constructs exhibit acceptable levels of internal consistency. 4.3.3 Convergent Validity After evaluating the reliability of the constructs and the indicators, the researcher went further to assess the validity of same, since this is also necessary for evaluating reflective measures. The first of this is the convergent validity and this is measured using the AVE of the constructs. AVE is expected to be more than 0.50, which implies that the construct explains more than 50% of the variations in the indicators. It is also more preferred all the AVEs are within a smaller range (Avkiran, 2018). From Table 4.2, it can be observed that all the constructs of this study meet this threshold, with ethical judgement (0.505) having the lowest AVE and religiousness (0.771) having the highest. In simple terms, the constructs in this study explain more than half of the variance in the measurement items. 57 University of Ghana http://ugspace.ug.edu.gh Table 4.2: Item Loadings, Average Variance Extracted and Composite Reliability Constructs Items Loadings AVE CR Rho_A Behavioural BI1 0.852 0.705 0.827 0.583 Intention BI11 0.827 Ethical Judgement MES1 0.753 0.505 0.911 0.895 MES11 0.696* MES12 0.740 MES13 0.693* MES14 0.698* MES2 0.765 MES3 0.743 MES4 0.718 UEM1 0.678* UEM11 0.612* Honesty Hon1 0.828 0.679 0.894 0.868 Hon2 0.851 Hon3 0.842 Hon4 0.773 Intellectualism Int1 0.775 0.69 0.899 0.873 Int2 0.857 Int3 0.835 Int4 0.854 PIE PIE11 0.773 0.719 0.911 0.909 PIE12 0.877 PIE13 0.916 PIE14 0.818 Religiousness Rel1 0.881 0.771 0.931 0.977 Rel2 0.856 Rel3 0.873 Rel4 0.903 Self-Control SelCo1 0.815 0.707 0.906 0.875 SelCo2 0.833 SelCo3 0.880 SelCo4 0.835 * below the threshold of 0.70 but maintained because its elimination negatively affects the composite reliability 58 University of Ghana http://ugspace.ug.edu.gh 4.3.4 Discriminant Validity Another important measure in evaluating the validity of the constructs is discriminant validity. There are numerous approaches to this, but the researcher chooses to report the three most widely used and accepted measures. These are the Fornell-Larcker criterion, item cross-loadings analysis and Heterotrait-Monotrait Ratio (HTMT ratio). First, using the Fornell-Larcker criterion, the square root of the AVE value of each construct is compared to the constructs correlation with other constructs, and it is expected that the square root of the AVE will be greater than the greatest of the correlations. This is shown in Table 4.3 below, as the diagonal which shows the square root of the AVE is higher than all the off-diagonal scores in the corresponding rows and columns. All the constructs of this study pass this test of validity. For instance, religiousness and intellectualism have average AVEs of 0.878 and 0.831 respectively, which are higher than all the elements in their corresponding rows and columns. This implies and confirms that the constructs share more variance in their indicator variables than they share with any other construct. Table 4.3: Fornell-Larcker Criterion Behavioural Ethical Honesty Intell.a PIE Rel.b Self- Intention Judgement Control Behavioural *0.839 Intention Ethical 0.413 *0.711 Judgement Honesty 0.026 0.197 *0.824 Intellectualism 0.106 0.076 0.49 *0.831 PIE 0.077 0.01 0.185 0.248 *0.848 Religiousness 0.094 0.105 0.626 0.521 0.241 *0.878 Self-Control 0.081 0.212 0.634 0.419 0.185 0.624 *0.841 a – Intell: Intellectualism b – Rel.: Religiousness * boldened figures represent the Fornell -Larcker measures (square root of AVE) for each construct. 59 University of Ghana http://ugspace.ug.edu.gh 4.3.5 Cross-Loading Analysis In examining discriminant using cross-loading analysis, we refer to the rule of thumb, according to Sarstedt et al. (2014), who theorise that individual items should exhibit higher correlations with the latent variables that they are supposed to measure than any other latent variable in the model under review. Upon generating the cross-loading, the results show that all the measurement items load effectively unto their respective latent variables with their correlation coefficients being higher than the correlations between the items and any other construct (see Table 4.4 below). 4.3.6 Heterotrait-Monotrait (HTMT) Ratio Lastly, we evaluate the discriminant validity using HTMT ratio. The HTMT ratio is a new measure of discriminant validity and it is argued to be a more robust and superior measure to Fornell- Larcker criterion and the cross-loadings (Avkiran, 2018; Henseler et al., 2015). The rule of thumb is, an HTMT ratio of 0.90 or less is acceptable. From Table 4.5 below, all the constructs were having an HTMT ratio less than 0.90. For example, self-control and honesty recorded the highest (0.750) and the lowest HTMT ratio was recorded between honesty and behavioural intention (0.054). It can be concluded therefore that there is enough evidence to support the establishment of discriminant validity of the constructs. 60 University of Ghana http://ugspace.ug.edu.gh Table 4.4: Cross Loading Analysis Behavioural Ethical Honesty Intell.a PIE Rel.b Self- Intention Judgement Control BI1 *0.852 0.347 0.047 0.116 0.133 0.117 0.095 BI11 *0.827 0.347 -0.006 0.060 -0.008 0.037 0.039 MES1 0.341 *0.753 0.196 0.124 0.102 0.129 0.195 MES11 0.318 *0.696 0.094 0.083 -0.056 0.088 0.139 MES12 0.311 *0.740 0.063 0.000 -0.046 0.060 0.131 MES13 0.182 *0.693 0.102 -0.005 -0.038 0.042 0.135 MES14 0.356 *0.698 0.148 0.027 0.015 0.137 0.175 MES2 0.290 *0.765 0.148 0.089 0.048 0.062 0.184 MES3 0.213 *0.743 0.147 0.037 0.045 0.011 0.113 MES4 0.295 *0.718 0.182 0.046 0.053 0.083 0.175 UEM1 0.311 *0.678 0.197 0.115 -0.033 0.066 0.137 UEM11 0.245 *0.612 0.081 -0.030 -0.053 0.017 0.085 Hon1 -0.006 0.191 *0.828 0.339 0.121 0.489 0.521 Hon2 0.025 0.162 *0.851 0.422 0.118 0.569 0.544 Hon3 0.068 0.170 *0.842 0.471 0.214 0.52 0.514 Hon4 -0.017 0.094 *0.773 0.396 0.167 0.492 0.527 Int1 0.038 0.000 0.458 *0.775 0.251 0.451 0.311 Int2 0.079 0.093 0.449 *0.857 0.198 0.445 0.349 Int3 0.097 0.062 0.344 *0.835 0.220 0.382 0.354 Int4 0.106 0.053 0.433 *0.854 0.198 0.487 0.369 PIE11 0.008 0.000 0.148 0.241 *0.773 0.164 0.164 PIE12 0.055 -0.013 0.149 0.229 *0.877 0.210 0.186 PIE13 0.081 0.002 0.200 0.232 *0.916 0.245 0.168 PIE14 0.067 0.034 0.126 0.183 *0.818 0.174 0.132 Rel1 0.078 0.086 0.573 0.485 0.216 *0.881 0.562 Rel2 0.034 0.066 0.528 0.445 0.223 *0.856 0.526 Rel3 0.051 0.093 0.559 0.403 0.180 *0.873 0.553 Rel4 0.127 0.109 0.546 0.485 0.227 *0.903 0.553 SelCo1 0.009 0.175 0.512 0.291 0.099 0.406 *0.815 SelCo2 0.087 0.163 0.587 0.442 0.126 0.608 *0.833 SelCo3 0.074 0.213 0.505 0.338 0.186 0.502 *0.880 SelCo4 0.097 0.155 0.540 0.342 0.202 0.585 *0.835 a – Intell: Intellectualism b – Rel.: Religiousness * Boldened diagonal figures represent the highest Pearson Correlation Coefficients 61 University of Ghana http://ugspace.ug.edu.gh Table 4.5: Heterotrait-Monotrait (HTMT) Ratio Behavioural Ethical Honesty Intell.a PIE Rel.b Self- Intention Judgement Control Behavioural Intention Ethical 0.560 Judgement Honesty 0.054 0.209 Intellectualism 0.133 0.097 0.591 PIE 0.125 0.086 0.213 0.308 Religiousness 0.111 0.107 0.714 0.591 0.256 Self-Control 0.111 0.234 0.750 0.480 0.213 0.704 a – Intell: Intellectualism b – Rel.: Religiousness 4.4 Structural Model After confirming the reliability and validity of the model, it is necessary and now acceptable to assess the structural model. The structural model simply represents the theoretical or conceptual framework that underlies the study. In assessing the structural model, how well the data gathered supports the framework/concept is established or determined. The assessment of the structural model involves testing the predictive abilities and the relationships between the construct of the model. This helps to ascertain whether the model employed explains the phenomenon it is purported to explain. The key areas in assessing the structural model are the path coefficients’ significance, the level of R2, the effect size measure by f2, the predictive relevance (Q2) and the q2 effect size. 62 University of Ghana http://ugspace.ug.edu.gh 4.4.1 Variance Inflation Factor Results However, before the evaluation of the structural model, the researcher needs to test for collinearity. This is to check and prevent path coefficient biases which may result from including in the estimations, predictors that exhibit collinearity. In testing whether there is collinearity in the structural model the value inflation factor (VIF) is employed. According to Hair et al. (2011), for the VIF of a measurement item to be acceptable, it should not be greater than 5. From the table below, none of the predictors shows a VIF greater than 4.0. This means that there is no problem of collinearity among the predictor constructs or items. For instance, it can be seen that BI1 and BI11 recorded the lowest VIFs of 1.202 each with MES12 exhibiting the highest VIF of 3.876, which is still far less than the threshold of 5.0. 63 University of Ghana http://ugspace.ug.edu.gh Table 4.6: Variance Inflation Factor Results Constructs Item VIF Behavioural Intention BI1 1.202 BI11 1.202 Ethical Judgement MES1 3.005 MES11 3.169 MES12 3.876 MES13 2.262 MES14 2.120 MES2 3.343 MES3 2.960 MES4 2.337 UEM1 1.684 UEM11 1.751 Honesty Hon1 1.729 Hon2 2.027 Hon3 2.063 Hon4 1.943 Intellectualism Int1 2.347 Int2 2.290 Int3 1.752 Int4 2.034 PIE PIE11 2.838 PIE12 3.864 PIE13 2.912 PIE14 1.583 Religiousness Rel1 2.880 Rel2 2.972 Rel3 2.671 Rel4 2.327 Self-Control SelCo1 1.930 SelCo2 1.988 SelCo3 2.252 SelCo4 2.034 VIF = Variance Inflation Factor 64 University of Ghana http://ugspace.ug.edu.gh 4.4.2 Explanatory Power of the Model (or Level of Explanation [R2]) The r-square (R2) simply tries to indicate the level or extent to which the endogenous variable is explained by the exogenous variables. The r-square falls between 0 and 1. The closer it is to 1, the better it is and it implies that the exogenous variables explain a higher proportion of the variations in the endogenous variable. The R2 value of ethical judgement is 0.057 and that of behavioural intention is 0.197. Following Hair et al.’s (2014) rule of thumb, the r-square value of both behavioural intention and ethical judgement are weak. This implies that personal values (intellectualism, honesty, self-control and religiousness), and PIE explain just about 5.7% of the total variations in the ethical judgement of accounting students. Also, personal values, PIE and ethical judgement explain up to 19.7% of the total variations in the behavioural intention of accounting students. In addition to the above, the researcher assesses the predictive relevance of the constructs, employing the Stone-Geisser's Q² value (Geisser, 1974; Stone, 1974, 1977), which represents an evaluation criterion for the cross-validated predictive relevance of the PLS path model. The construct cross-validated redundancy (Q2) is a measure indicating the out-of-sample predictive relevance, that is, the capability of the model to predict endogenous latent variables. Q2 greater than zero for a specific endogenous construct implies an acceptable predictive relevance. The endogenous variables in this study exhibit acceptable levels of predictive accuracy. Behavioural intention has a Q2 of 0.121 and also ethical judgement show a Q2 of 0.025, which are both above zero. 65 University of Ghana http://ugspace.ug.edu.gh 4.4.3 Path Analysis The path analysis was conducted to either confirm or deny the hypothesis that has been set up from the objectives of the study and finding its basis in literature. As a result, the path analysis will be examined and discussed on the basis of the hypotheses. The figure below shows the path models and the estimates of these paths. This is discussed further below. Figure 4.3: Structural Model 66 University of Ghana http://ugspace.ug.edu.gh H1: there is a significant negative relationship between personal values and ethical judgement of accounting students. The personal values studied in this research are intellectualism, honesty, self-control and religiousness. The path analysis revealed that there is a positive relationship between two of the constructs, that is, honesty (beta = 0.149) and self-control (beta = 0.185), and ethical judgement. The other two personal values exhibit a negative relationship with ethical judgement (intellectualism; beta = -0.025; and religiousness; beta = -0.084). The results also show that there is a significant relationship between honesty and ethical judgement, and self-control and ethical judgement at a significance level of 5%. For example, honesty has a p-value of 0.028 and self- control has a p-value of 0.002. Further, the analysis revealed that even though two of the structural paths are significant, they are not supported. That is, the outcomes from the analysis were different from the proposed path. For example, it was hypothesised that honesty has a negative relationship with ethical judgement. The outcome of the analysis rather revealed that there is a positive relationship between honesty and ethical judgement. Similarly, the initial hypothesis was that there is a negative relationship between self-control and ethical judgement, but the results show that there is a positive relationship between self-control and ethical judgement. H2: there is a significant negative relationship between PIE and ethical judgement. The bootstrap results show that there exists a negative relationship between PIE and ethical judgement. The path coefficient of PIE to ethical judgement is -0.025. That notwithstanding, the 67 University of Ghana http://ugspace.ug.edu.gh relationship is not significant. This is indicated by a t-value of 0.353 and a p-value of 0.724, which is greater the 0.05 significance level. This relationship is also not supported. H3: there is a significant negative relationship between personal values and behavioural intention Furthermore, the researcher sought to investigate if personal values have significant impact on the behavioural intention of accounting students. The results showed that there exist negative relationships between honesty (beta=0.-0.169) and self-control (beta=-0.017), and behavioural intention. However, self-control did not show a significant relationship (p-value = 0.788), but there was a significant relationship between honesty and behavioural intention (t-value=2.457 and p- value=0.014, which less than alpha=0.050). It can, therefore, be concluded that the relationship between honesty and behavioural intention is supported. On the other hand, there is a positive relationship between intellectualism (beta=0.096) and religiousness (beta=0.100), and behavioural intention. However, these relationships are not significant. That is, these two personal values do not have much impact on the behavioural intention of accounting students. The p-values of intellectualism and religiousness are 0.121 and 0.141 respectively, and these are both greater than an alpha of 0.050. H4: there is a significant negative relationship between PIE and behavioural intention. From literature, it was hypothesised that when one perceives an ethical issue is of high relevance, then the person is less likely to engage in that act. Again, the results correspond with this assertion, showing a negative coefficient (beta = -0.059). However, the relation is not significant. The t-value 68 University of Ghana http://ugspace.ug.edu.gh is 0.947 and p-value is 0.344, which is more than the 0.05 significance level. Hence, we fail to accept the null hypothesis. H5: there is a significant positive relationship between ethical judgment and behavioural intention It is believed that students who rate unethical issues as morally right will likely be found engaging in such activities. Hence the researcher proposes a positive relationship between the two constructs. The results revealed a positive relationship between ethical judgement and behavioural intention (beta = 0.659). Also, it was realised that the relationship between the ethical judgement and behavioural intention is significant and hence supported (t-value = 9.292, p-value = 0.000, which less than the significance level of 0.050, and even 0.010). 69 University of Ghana http://ugspace.ug.edu.gh Table 4.7: Results of Structural Model Path Expected Path t-value Sig. Inference Sign Coefficient (B) Honesty -> Ethical - 0.149 2.201 *0.028 Supported Judgement Intellectualism -> - -0.025 0.382 0.703 Not supported Ethical Judgement Self-Control -> Ethical - 0.185 3.072 **0.002 Supported Judgement Religiousness -> Ethical - -0.084 1.349 0.178 Not supported Judgement PIE -> Ethical - -0.025 0.353 0.724 Not supported Judgement Honesty -> Behavioural - -0.169 2.457 *0.014 Supported Intention Intellectualism -> - 0.096 1.555 0.121 Not supported Behavioural Intention Self-Control -> - -0.017 0.270 0.788 Not supported Behavioural Intention Religiousness -> - 0.100 1.474 0.141 Not supported Behavioural Intention PIE -> Behavioural - 0.059 0.947 0.344 Not supported Intention Ethical Judgement -> + 0.432 9.292 **0.000 Supported Behavioural Intention * significant at 5% significance level ** significant at 1% significance level 4.5 Discussion of Results 4.5.1 Structural Model To begin with, the explanatory relevance and the predictive relevance of the study were examined. The results showed that perceived importance of an ethical issue (PIE) and personal values (intellectualism, honesty, self-control and religiousness) have an explanatory relevance of 0.057 70 University of Ghana http://ugspace.ug.edu.gh on ethical judgement. That is, these two determinants only explain 5.7% of the total variations or changes in ethical judgement. This implies the exogenous variables have a very weak predictive relevance on ethical judgement. That is, even though the exogenous variables may be significant in explaining the endogenous variable, it only explains just a small portion of ethical judgement. Also, personal values, PIE and ethical judgement only explain 19.7% of the total variations in behavioural intention of accounting students. However, it is argued that the explanatory relevance is well explained in the context of the study at hand and based on previous studies (Sarstedt et al., 2014). It is also believed that studies of exploratory nature generally exhibit low levels of explanatory relevance (Sarstedt et al., 2014). For instance, a study by Antes et al. (2007) revealed that personality traits explain less than 20% of the total variations in ethical decision-making. Similarly, Haines et al. (2007), also studies the impact of PIE on ethical decision-making process. The results showed that in all scenarios, the r-square for the relationship between PIE and ethical judgement was between 0.30 and 0.526. hence the r- square for this study can be accepted even though it is low for both endogenous variables, ethical judgement and behavioural intention. 4.5.1.1 Personal Values and Ethical Judgement The first hypothesis of this study was to investigate if there is a significant negative relationship between personal values and ethical judgement of accounting students. Firstly, we consider the impact of honesty on ethical judgement. The results show that there exists a significant positive relationship between honesty and ethical judgement (beta=0.149, p-value=0.028). This implies that accounting students who perceive themselves as honest, judge unethical situations to be morally right or ethical. This result contradicts the findings of Alleyne et al. (2013) and Finegan 71 University of Ghana http://ugspace.ug.edu.gh (1994), who postulate that honest people are less likely to assess an unethical situation to be morally right. Also, it is generally, socially and morally expected that honest people endorse right and debunk wrong. This study reveals otherwise. This may be attributed to the fact that people perceive themselves to be honest but inherently, they are not morally or ethically upright. (Longenecker et al., 2004; Modarres & Rafiee, 2011). Also, according to Sliwa (2017) moral knowledge differs from moral understanding. Again, it is argued that accounting students are not ethically matured, hence even though they may be aware of ethics, they may not understand ethics or what is or is not ethical behaviour (Gray, Bebbington, & Mcphail, 1994). However even though the results suggest a significant relationship, the effect size (f2=0.011) is weak, following Cohen (1988), who posits that an f-square of less than 0.02 represents weak form effect size. Secondly, the results revealed that there exists a negative relationship between intellectualism and ethical judgement as hypothesised. Nonetheless, the relationship is not significant (beta=-0.025, p-value=0.703) and the effect size is also very weak (f2=0.0001), almost equal to zero. In effect, this means that the more intellectualism that students exhibit, the more they perceive an unethical issue as immoral and unethical. That is, the more intellectual an accounting student is, the more likely the student will judge an issue as unethical than ethical. Management and accounting educators are therefore advised to build on the intellectual capacity of the accounting students since this leads to an increase in ethical judgement in society and business environment. However, not much attention is to be concentrated on doing this since intellectualism does not have a significant impact and has a close to zero effect. 72 University of Ghana http://ugspace.ug.edu.gh Self-control is expected to be negatively related to ethical judgement. According to Alleyne et al. (2013) people who exhibit self-control judge unethical situations as unfair and immoral. As such, the study hypothesised that students with high levels of self-control will judge the situations of the various characters in the scenarios as unethical. The results of the study revealed otherwise. There is a positive and more importantly, a significant relationship between self-control and ethical judgement (beta=0.185, p-value=0.002). This implies that accounting students who perceive themselves to exhibit high levels of self-control tend to judge unethical issues as morally right. As a result, the more accounting students can control themselves, the more ethical they judge immoral issues. This strikes to be strange since the other way around is seen to be more logical and socially acceptable. However, even though people regard these personal values as very necessary and uphold them, the moral complexities of circumstances and this world leave them no option to judge ethically, unethical and immoral actions (Wells & Molina, 2017). Also, people, especially in this jurisdiction, do show some characteristics as an outward nature, without these attitudes being inherent and hence may falter without supervision. Further investigations may be conducted to determine the course of this result. Studies have shown that religious people are more ethical than people who are not religious. For instance, Singhapakdi et al. (2013), opine that religious people tend to be more morally upright compared to non-religious students and hence establishes a relationship between religiousness and ethics. In this study, we hypothesised that there is a significant negative relationship between religiousness and ethical judgement. The results from the data analysis confirm the above by showing a negative but insignificant relationship between religiousness and ethical judgement (beta=-0.084, p-value=0.178). This is explained as accounting students who are religious, judge immoral situations as unethical. Hence, students who perceive themselves to exhibit high levels of 73 University of Ghana http://ugspace.ug.edu.gh religiousness see immoral issues as such, unethical or immoral. It is, therefore, necessary for academicians to project and encourage religiousness among accounting students in order to improve upon the ethical nature and judgement of future accountants. In doing this, attention should be paid to the fact that religiousness is not a significant factor or value and hence not much resources should be invested here. 4.5.1.2 Perceived Importance of an Ethical Issue and Ethical Judgement Generally, it is believed that people who attach high relevance to an issue or perceive the issue to be of high importance are most likely to assess an immoral issue as unethical. A study by Guffey and McCartney (2008), showed that individuals who are high in perceived importance measure were more critical in assessing unethical academic issues and individuals who are low in perceived importance measure were less critical in assessing unethical actions. Similarly, this study finds that there exists a negative relationship between perceived importance of an ethical issue and ethical judgement. That is accounting students who attach high relevance to ethical issues judge the actions in the scenarios as unethical. That is students who attach higher importance to ethical situations will not point out unethical actions when they come across it. However, the relationship between PIE and ethical judgement was not found to be significant (beta=-0.025, p-value=0.724). 4.5.1.3 Personal Values and Behavioural Intention Personal values are key determinants of the behaviour of individuals. According to Alleyne et al. (2013), personal values have a significant negative relationship with ethical behaviour among both accounting and non-accounting students. Similarly, Giacomini and Akers (1998) and Conner and Becker (2003) found that individuals exhibiting higher personal values will hardly engage in unethical behaviour. These findings are partly supported by the results of this study. The results of 74 University of Ghana http://ugspace.ug.edu.gh this study reveal that there exist negative relationships between honesty (beta=-0.169, p- value=0.014) and self-control (beta=-0.017, p-value=0.788), and behavioural intention. This implies that accounting students who exhibit high levels of honesty and self-control will refrain from or are not likely to engage in unethical behaviour. The results also showed that there is a significant relationship between honesty and ethical behaviour. On the contrary, the relationship between self-control and behavioural intention is not significant. This implies that to encourage ethical behaviour among accounting students, accounting educators should endeavour to put in efforts to encourage honesty and self-control among these students. That notwithstanding, the educators should commit more resources to improve honesty among accounting students than self-control since honesty has a significant relationship whereas self-control does not. On the other hand, intellectualism (beta=0.096, p-value=0.121) and religiousness (beta=0.100, p- value=0.141) exhibit positive relationship with behavioural intention to engage in unethical actions. This implies that accounting students who perceive themselves to exhibit high levels of intellectualism tend to have higher intentions of engaging in unethical behaviour. Also, it is interesting to note that students who perceive themselves to be more religious tend to have higher intentions of engaging in unethical behaviour. This result is supported by Singhapakdi et al. (2013), who studied two dimensions of religiosity (intrinsic and extrinsic). The study revealed that managers higher in extrinsic religiosity are less ethical in their intentions. This suggests that the religiousness that these accounting students portray may only be extrinsic and not intrinsic. 75 University of Ghana http://ugspace.ug.edu.gh 4.5.1.4 Perceived Importance of an Ethical Issue and Behavioural Intention The findings revealed that there is a positive relationship between PIE and behavioural intention, while the study hypothesised otherwise. This can be explained as accounting students who perceive high importance to ethical issues are more unlikely to engage in it but the results revealed that accounting students who have higher scores for PIE have higher intentions to engage in unethical actions hence the positive relationship between the two constructs. However, the effect of PIE on behavioural intention is insignificant. This is consistent with the findings of Haines et al. (2007) and Robin et al.’s (1996) first proposition, which provides that PIE is the precedent cause of moral judgement and hence does not have a direct significant impact on behavioural intention. Also, the above results may be due to the fact that the actions in the scenarios employed may not have any negative impact on the respondents or any other third party, which is covered by the moral intensity construct. The moral intensity construct considers the magnitude of consequence of the action and proximity of the action among others (Lincoln & Holmes, 2011; Sweeney & Costello, 2009), which PIE does not but rather focus on the beliefs, needs, perceptions and values of the individual (Guffey & McCartney, 2008; Robin et al., 1996). As such, even though the respondents may exhibit high personal values, they may engage themselves in unethical behaviours if found in such situations as in the scenarios. This may be due to the fact that their behaviour only benefits them and does not harm them or any other third party. 4.5.1.5 Ethical Judgement and Behavioural Intention The last hypothesis was established as there exists a positive significant relationship between ethical judgement and behavioural intention. The results confirmed the hypothesis, showing a 76 University of Ghana http://ugspace.ug.edu.gh significant positive relationship between ethical judgement (beta=0.432, p-value=0.000; significant even at 1% significance level) and behavioural intention, and hence accepted. This implies that accounting undergraduate students who judge the action in the scenarios as unethical have lower intentions to engage in such or similar activities, and vice versa. This finding is also consistent with the findings of Robin et al. (1996), Haines et al. (2007), Guffey and McCartney (2008) and Singhapakdi et al. (2013) who found that there exists a significant positive relationship between ethical judgement and behavioural intention. Similarly, Pan and Sparks (2012) also suggest that people who are strict in ethical judgement have lower intentions to behave in a similar manner. Accounting educators and researchers should, therefore, focus attention on and nurture the ethical judgement of accounting students since it has a strong significant impact on the behavioural intentions of these students (Singhapakdi et al., 2013). This will subsequently improve the decision-making and ethical behaviour of future accountants. 4.6 Conclusion In this chapter, we presented the results of the study as well as literature situated discussion of the results. Generally, we find that there are significant differences in the mean scores of two out of four of the personal values across public and private institutions. This result showed that accounting students in public institutions generally exhibit higher levels of personal values than those of private institutions. Also, we realised that both private and public tertiary accounting students judge immoral actions as unethical. Further, we found that there are no significant differences between males and female accounting students’ personal values and ethical judgement. It is noteworthy that the results showed that there was no significant relationship between PIE and ethical judgement, and also between PIE and behavioural intention. However, some other findings 77 University of Ghana http://ugspace.ug.edu.gh of the study include a positive significant relationship between honesty and self-control and ethical judgement. Also, we found honesty and ethical judgement to be significantly related to behavioural intention. These results bring out some interesting revelations that may require more investigation. 78 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE SUMMARY OF FINDINGS, CONCLUSIONS AND RECOMMENDATIONS 5.0 Introduction In this chapter, the researcher brings the whole study to a close. This chapter summarises the whole work. Some key findings are discussed in this chapter and the practical implications subsequently brought to bare, as well as some recommendation for accounting educators and policy-makers. Further, the researcher will identify some limitations of this study and make recommendations for further studies. 5.1 Summary of Findings The main objective of this study was to find the relationship between personal values, PIE and ethical decision-making of accounting students. In order to achieve this goal, the researcher adapted Rest’s four-stage decision-making model and modified it, by incorporating personal values and PIE. A sample of 444 accounting students from four universities (2 public and 2 private) in Ghana were surveyed. PLS-SEM is employed in analysing the data that was gathered. A summary of the results is discussed in the subsequent paragraphs. Firstly, the study revealed that personal values, honesty and self-control, have significant positive relationships with ethical judgement. That is to say, accounting students who score high in terms of honesty and self-control are more likely to see the immoral action, like in the scenario, as ethical or acceptable. Hence, the more a student is perceived to be honest, the more likely that student is to pronounce an unethical behaviour as right or morally acceptable. This finding is contrary to the 79 University of Ghana http://ugspace.ug.edu.gh position of Alleyne et al. (2013). Personal values of intellectualism and religiousness were found to have no significant relationship with ethical judgement even though these values showed negative relationships. That is, the more intellectual or religious one may be, the more likely the individual is to judge an immoral action as unethical. The results of the study also showed that there is a negative relationship between PIE and ethical judgement but not significant. The negative relationship implies that accounting students who perceive ethical issues as of high relevance are more likely to judge such actions as unethical. This outcome was insignificant at 5% significance level. This finding contradicts the findings of Robin et al. (1996), Guffey and McCartney (2008) and Haines et al. (2008). Thirdly, it was found that two personal values (intellectualism and religiousness) have a positive relationship with behavioural intention. That is the more accounting students are religious and perceive themselves as intellectuals, the more likely it is for them to engage in unethical behaviour. However, the relationship between these constructs is not significant. On the other hand, the results showed a negative and significant relationship between honesty and behavioural intention. Also, the results revealed that there is a negative, but insignificant relationship between self-control and behavioural intention. Further, the results from the path analysis showed that there is a positive but insignificant relationship between PIE and ethical behaviour of accounting students. That is, this result is not supported, and is contrary to prior literature. Robin et al. (1996) and Haines et al. (2008) opined that people who perceive an unethical issue as of high relevance are less likely to engage in such behaviour, but the findings of this study suggest otherwise. 80 University of Ghana http://ugspace.ug.edu.gh In addition, the path analysis results showed that there is a strong positive relationship between ethical judgement and ethical behaviour. This result supports the hypothesis of this study and is supported in the literature. Also, the researcher performed some independent t-test to find if there are any significant differences in the personal values and ethical judgement of public and private tertiary students and male and female accounting students. It was interesting to note that there is a significant difference between the personal values, honesty and self-control, of accounting students of private and public institutions. The results indicate that accounting students of public institutions are generally honest and exhibit higher levels of self-control than those in private institutions. Also, it was found that there is no significant difference in the intellectualism and religiousness between accounting students of public and private institutions. In terms of ethical judgement, the researcher found no significant difference in the judgement of private and public tertiary accounting students. Finally, comparing the personal values of male and female accounting students, we found no significant difference between them in the facets of personal values. Also, there was no evidence to suggest that the ethical judgements of male and female accounting students are different. 5.2 Practical Implications Following the findings of the study, some conclusions can be drawn. First, it can be concluded that personal values affect ethical judgement of accounting students. That is, the personal values of these accounting students are driven by their level of honesty and self-control. The implications for firms is that their recruitment process will have to be able to select for them, students of high reputation in terms of honesty and self-control since these values affect their ethical judgement. 81 University of Ghana http://ugspace.ug.edu.gh Secondly, it was established that honesty is also a key determinant of behavioural intention to engage in ethical behaviour. This finding is relevant for firms, in that, to ensure good ethical behaviour, they need to employ students who are honest. Also, accounting academics have a responsibility to nurture and grow the value of honesty in all accounting students and prospective accountants. Furthermore, it was revealed that ethical judgement is a strong predictor of behavioural intention to engage in ethical behaviour. This finding is important to firms in forming a firm culture and also in recruitment procedures. The firms will have to ensure that their recruitment process attracts students or prospective accountants with good ethical judgement. Also, their firm culture should be one which encourages good ethical judgement since this positively affects the behavioural intentions of accounting students. On the other hand, it was discovered that PIE has no significant relationship with ethical judgement as well as behavioural intentions of accounting students. This implies that the perceived relevance attached to a moral issue does not affect accounting students’ judgement or intention to engage in such behaviours. It therefore important for accounting educators and firms not to place much importance on the perception students have on unethical actions. Notwithstanding, prior studies have proven otherwise (Guffey & McCartney, 2008; Haines et al., 2008; Robin et al., 1996), hence, managers and accounting educators may confirm this finding before acting on it. In addition to the above, further analysis showed that male and female accounting students do not differ in personal values nor ethical judgement. There is no evidence to support the proposition that there is a significant difference between their mean scores of personal values and ethical judgement. This implies that male and female accounting students are both statistically of the same 82 University of Ghana http://ugspace.ug.edu.gh level in terms of personal values of intellectualism, honesty, self-control and religiousness. Accounting educators and managers, therefore, do not need to provide separate environments for the different genders in order to support their behaviour or thoughts. 5.3 Recommendations In consideration of the above findings and their practical implications, the following recommendations are outlined to help academia, practice and policy to manage the findings. To begin with, the researcher suggests that accounting educators should increase the use of scenario-based teaching and learning in accounting ethics courses. This will sensitise the students more on the nature of ethical situations in the academic setting as well as work setting. With more knowledge, students will be more aware of ethical issues, which improve their judgement of same and subsequently influence their intentions and behaviour. Secondly, based on the findings of the study, the determinants of ethical judgement and behavioural intentions have been made clear. Hence, it is suggested that accounting regulatory and professional bodies inculcate these values in their code of ethics and also highlight the ones which already exist. This will inform and encourage members of these bodies and accountants as well as prospective accountants to pay attention to these values. This will, in turn, have a positive impact on their ethical judgement and subsequently their behaviours. Also, educational institutions should have ethical codes, which will present to the students, basic ethical behaviours and consequences of acting unethically. The results of the study also showed that religiousness and intellectualism do not have significant influence on ethical judgement and behavioural intentions of accounting students. It is therefore 83 University of Ghana http://ugspace.ug.edu.gh recommended that families, societies and institutions should create a motivating and supporting environment that will encourage good and morally right ethical judgement and behavioural intentions among accounting students. Also, students who are high in intellectualism should be monitored and directed as to how to combine or blend their wisdom with ethics. This will aid the students to become better accountants. To firms that recruit fresh graduate accountants, it is recommended that their recruitment procedures will assess the ethical judgement and behaviour of such graduates to ensure that the human resources employed are morally upright and will not drag the firms’ names to disrepute. To add to the above, it is believed that private universities, due to their religious affiliation (especially in Ghana), are more ethical. The results of the study indicated that there is significant evidence that accounting students in public universities are higher in terms of personal values. Also, there is no significant difference in the ethical judgement of students of the two categories of institutions. The researcher suggests that moral training in such schools should be intensified. This is to improve the ethical values of the accounting students and subsequently their behaviour. 5.4 Limitations of the Study In spite of the numerous strengths of this study, there are some limitations that readers need to note. The four personal values of Scott’s (1965) personal values scale were employed in this study. As such the findings of this study should not be generalised for all personal values. That is, following Akaah and Lund (1994), the researcher only included intellectualism, honesty, self- control and religiousness in this study. It is important for readers to follow this study whiles having this in mind. 84 University of Ghana http://ugspace.ug.edu.gh Also, the sample of the study was taken from only four out of the 92 accredited tertiary institutions in the country. It is believed that more extensive work could be done with a higher number of institutions. That notwithstanding, the researcher was bound by time and access and hence needed to make a decision. Also, the four chosen institutions are a good representation of the universities in the country. Moreover, the study is believed to have suffered from social desirability bias. The student respondents answered the questionnaire to suit the expectation of their friends (since they answered the questionnaires in class) or the perceived expectation of the researcher. This, to some extent, caused a certain level of disparity in the results. It is noteworthy that most of the responses which seemed to be too genuine or good were removed to curb this situation. Also, all reliability and validity tests were made to ensure that the presented results can be relied on to make meaningful inferences. Finally, the results of the study showed a low predictive value (R-square). This simply implies that only a small portion of the variation in ethical judgement and behavioural intention (ethical decision-making) was explained by the independent variables, personal values and PIE. However, according to Haines et al. (2008) and Fritzsche and Oz (2007), the predictive value attained for a study is best explained when it is situated in the context of other studies of a similar nature. They further suggested that, for exploratory studies of this nature, the R-square attained is good. 5.5 Directions for Future Research Taking into account the limitations of this study, the following suggestions are made for future research. First, the researcher suggests that more studies should be done to investigate the impact 85 University of Ghana http://ugspace.ug.edu.gh of PIE on the ethical decision-making process of accounting students. This is to help clarify the dilemma in this area, as Robin et al. (1996), and Guffey and McCartney (2008) suggest that PIE has a significant impact on ethical judgement, the findings of this study suggest otherwise. Also, the cultural setting may be investigated, since the above-mentioned studies were all conducted in developed countries whereas this study was undertaken in the developing setting. In addition, future researchers should consider increasing the sample size and across more institutions. Also, if the institutions could be taken from all the geographical demarcations of the country, that will serve as a good basis to make more generalisable inferences and conclusions. Further studies may consider all business students rather than just accounting students. Future researchers can also consider including students who are writing the accountancy professional examinations since they are all being trained to become accountants. Thirdly, future research may also consider involving more personal values or testing other personal values. These studies may employ other personal values scales as the Rokeach value scale and Schwartz value scale. This may tend to show a higher predictive value and bring more diversity into the ethics research in Ghana and other developing settings. Lastly, a qualitative study to investigate how organisational and social culture influence the ethical beliefs of individuals may be necessary. This will help explain the contradictory findings of this study to most prior literature. 86 University of Ghana http://ugspace.ug.edu.gh REFERENCES Aibinu, A. A., & Al-Lawati, A. M. (2010). Using PLS-SEM Technique to Model Construction Organizations’ Willingness to Participate in E-Bidding. Automation in Construction, 19(6), 714–724. https://doi.org/10.1016/j.autcon.2010.02.016 Akaah, I. P., & Lund, D. (1994). The Influence of Personal and Organizational Values on Marketing Professionals’ Ethical Behavior. Journal of Business Ethics, 13(6), 417–430. Retrieved from http://libezproxy.must.edu.mo/login?url=http://search.ebscohost.com/login.aspx?direct=true &db=edsjsr&AN=edsjsr.25072545&site=eds-live Alleyne, P., Cadogan-McClean, C., & Harper, A. (2013). Examining Personal Values and Ethical Behaviour Perceptions between Accounting and Non-accounting Students in the Caribbean. The Accounting Educators’ Journal, XXIII, 47–70. Alvi, M. H. (2016). A Manual for Selecting Techniques in Research. Karachi. Retrieved from https://mpra.ub.uni-muenchen.de/70218/ Amponsah, E. B., Boateng, P. A., & Onuoha, L. N. (2016). Academics’ Non-Adherence to Accounting Ethics in Ghanaian Universities: Internal Evidence of the Antecedents. VVU Journal of Business Research, 1(1), 35–50. Amponsah, E. B., Boateng, P. A., & Onuoha, L. N. (2016). Lack of Integrity among Ghanaian Accounting Academics : Upshot on Employers ’ Operational Costs. Journal of Accounting and Finance, 16(2), 87–105. Antes, A. L., Brown, R. P., Murphy, S. T., Waples, E. P., Mumford, M. D., Connelly, S., & 87 University of Ghana http://ugspace.ug.edu.gh Devenport, L. D. (2007). Personality and Ethical Decision-Making in Research: The Role of Perceptions of Self and Others. Journal of Empirical Research on Human Research Ethics, 2(4), 15–34. Antwi, S. K., & Kasim, H. (2015). Qualitative and Quantitative Research Paradigms in Business Research : A Qualitative and Quantitative Research Paradigms in Business Research : A Philosophical Reflection. European Journal of Business and Management, 7(3), 217–226. Apostolou, B., Dorminey, J. W., Hassell, J. M., & Rebele, J. E. (2015). Accounting Education Literature Review (2013 – 2014). Journal of Accounting Education, 33(2), 69–127. https://doi.org/10.1016/j.jaccedu.2015.04.001 Apostolou, B., Dorminey, J. W., Hassell, J. M., & Rebele, J. E. (2016). Accounting Education Literature Review (2015). Journal of Accounting Education, 35(1), 20–55. https://doi.org/10.1016/j.jaccedu.2016.03.002 Apostolou, B., Dull, R. B., & Schleifer, L. L. F. (2013). A Framework for the Pedagogy of Accounting Ethics. Accounting Education: An International Journal, 22(1), 1.17. https://doi.org/10.1080/09639284.2012.698477 Arnold, D. F., Dorminey, J. W., Neidermeyer, A. A., & Neidermeyer, P. E. (2013). Internal and External Auditor Ethical Decision-Making. Managerial Auditing Journal, 28(4), 300–322. https://doi.org/10.1108/02686901311311918 Aveh, F. K., Awunyo-Vitor, D., & Owusu-Afriyie, R. (2016). Code of Ethics Compliance by Accountants in Ghana. African J. of Accounting, Auditing and Finance, 5(2), 135–154. Avkiran, N. K. (2018). Rise of the Partial Least Squares Structural Equation Modeling: An 88 University of Ghana http://ugspace.ug.edu.gh Application in Banking. In N. K. Avikran & C. M. Ringle (Eds.), Partial Least Squares Structural Equation Modeling Associate Series Editor (pp. 211–226). Cham: Springer International Publishing AG. Baird, J. E., Zelin, I., & Brennan, P. J. (2006). Academic Major, Gender, Personal Values, and Reactions to an Ethical Dilemma. Journal of Business Case Studies (JBCS), 2(2), 73–82. https://doi.org/10.19030/jbcs.v2i2.4890 Barnett, T., Brown, G., & Bass, K. (1994). The Ethical Judgements of College Students Regarding Business Issues. Journal of Education for Business, 69(1), 333-338. Barreiro, P. L., & Albandoz, J. P. (2001). Population and Sample. Sampling Techniques. In Management Mathematics for European Schools (pp. 1–12). Baumeister, R. F., & Line, J. J. E. (2000). Self-Control, Morality and Human Strength. Journal of Social and Clinical Psychology, 19(1), 29–42. Bebi, L., & Xhindi, T. (2017). The Impact of Organizational Factors on Ethical Decision Making Process of Albanian Accountants. European Journal of Multidisciplinary Studies, 6(2), 95– 101. Beltramini, R. F., Peterson, R. A., & Kozmetsky, G. (1984). Concerns of College Students Regarding Business Ethics. Journal of Business Ethics, 3(1), 195–200. Bilsky, W., Munster, W. W., & Schwartz, S. H. (1994). Values and Personality. European Journal of Personality, 8(1), 163–181. Blamey, R. K., & Braithwaite, V. A. (1997). A Social Values Segmentation of the Potential Ecotourism Market. Journal of Sustainable Tourism, 5(1), 29–45. 89 University of Ghana http://ugspace.ug.edu.gh https://doi.org/10.1080/09669589708667274 Boateng, E. A., & Agyapong, D. (2017). Gender and Ethical Behaviour of Accounting Students : An Empirical Evidence of University of Cape Coast. Research Journal of Finance and Accounting, 8(4), 49–56. Boynton, P. M., & Greenhalgh, T. (2004). Selecting, Designing, and Developing Your Questionnaire. British Medical Journal, 328(1), 1312–1315. Bradley, N. (1999). Sampling for Internet Surveys. An Examination of Respondent Selection for Internet Research. International Journal of Market Research, 41(4), 1–10. Brunton, M., & Eweje, G. (2010). The Influence of Culture on Ethical Perception Held by Business Students in a New Zealand University. Business Ethics: A European Review, 19(4), 349–362. https://doi.org/10.1111/j.1467-8608.2010.01604.x Carnegie, G. D., & Napier, C. J. (2007). The Portrayal of Accounting and Accountants Following the Enron Collapse. In The 30th Annual Congress of the European Accounting Association, Lisbon (pp. 1–26). Chin, W. W. (1998). Commentary: Issues and Opinion on Structural Equation Modeling. MIS Quarterly, 22(1), 7–16. Cohen, S., Frank, E., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney, J. M. (1998). Types of Stressors That Increase Susceptibility to the Common Cold in Healthy Adults. Health Psychology, 17(3), 214–223. Conroy, S. J., & Emerson, T. L. N. (2004). Business Ethics and Religion : Religiosity as a Predictor of Ethical Awareness among Students. Journal of Business Ethics, 50(1), 383– 90 University of Ghana http://ugspace.ug.edu.gh 396. Costa, A. J., Pinheiro, M. M., Ribeiro, M. S., Costa, A. J., Pinheiro, M. M., & Ethical, M. S. R. (2016). Ethical Perceptions of Accounting Students in a Portuguese University: The Influence of Individual Factors and Personal Traits. Accounting Education, 25(4), 327–348. https://doi.org/10.1080/09639284.2016.1191270 Creswell, J. (2003). Research Design. In Qualitative, Quantitative, and Mixed Method Approaches (Vol. 53, pp. 1689–1699). https://doi.org/10.1017/CBO9781107415324.004 Dorantes, C. A., Hewitt, B., & Goles, T. (2006). Ethical Decision-Making in an IT Context: The Roles of Personal Moral Philosophies and Moral Intensity. In 39th Hawaii International Conference on System Sciences (Vol. 0, pp. 1–10). Douglas, P. C., Davidson, R. A., & Schwartz, B. N. (2001). The Effect of Organizational Culture and Ethical Orientation on Accountants’ Ethical Judgments. Journal of Business Ethics, 34(1), 101–121. Douglas, R., & Kevin, P. (2002). The Role of Moral Intensity in Ethical Decision Making. Business & Society, 41(1), 84–117. https://doi.org/10.1177/0007650302041001006 Eaton, T. V, & Giaciomino, D. E. (2001). An Examination of Personal Values: Differences Between Accounting Students and Managers and Differences Between Genders. Teaching Business Ethics, 5(2), 213–229. Els, G. (2009). Attitudes of Accounting Students Towards Ethics, Continuous Professional Development and Lifelong Learning. African Journal of Business Ethics, 4(1), 46–53. Finegan, J. (1994). The Impact of Personal Values on Judgments of Ethical Behaviour in the 91 University of Ghana http://ugspace.ug.edu.gh Workplace. Journal of Business Ethics, 13(9), 747–755. https://doi.org/10.1007/BF00881335 Flory, S. M., Phillips, T. J. J., Reidenbach, R. E., & Donald, R. P. (1992). A Multidimensional Analysis of Selected Ethical Issues in Accounting. The Accounting Review, 67(2), 284–302. Fornell, C., & Bookstein, F. L. (1982). Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory. Journal of Marketing Research, 19(4), 440–452. Friedrich, T., Schlauderer, S., Weidinger, J., & Raab, M. (2017). On the Research Paradigms and Research Methods Employed in the BISE Journal – A Ten-Year Update. In Proceedings der 13. Internationale Tagung Wirtschaftsinformatik (WI) (pp. 1111–1125). Retrieved from http://www.wi2017.ch/de/proceedings Fritzsche, D. J., & Becker, H. (1982). Business Ethics of Future Marketing Managers. Journal of Marketing Education, 1(1), 2–7. Fritzsche, D. J., & Oz, E. (2007). Personal Values’ Influence on the Ethical Dimension of Decision Making. Journal of Business Ethics, 75(4), 335–343. Froelich, K. S., & Kottke, J. L. (1991). Measuring Individual Beliefs about Organizational Ethics. Educational and Psychological Measurement, 51(1), 377–383. Gandz, J., & Hayes, N. (1988). Teaching Business Ethics. Journal of Business Ethics, 8(3), 657– 665. https://doi.org/10.1177/105256298300800305 Gefen, D., & Straub, D. (2000). The Relative Importance of Perceived Ease of Use in IS Adoption: A Study of E-Commerce Adoption. Journal of the Association for Information Systems, 1(8), 1–30. 92 University of Ghana http://ugspace.ug.edu.gh Gefen, D., & Straub, D. (2005). A Practical Guide to Factorial Validity Using PLS-Graph: Tutorial and Annotated Example. Communications of the Association for Information Systems, 16(5), 91–109. Gefen, D., Straub, D., & Boudreau, M.-C. (2000). Structural Equation Modeling and Regression: Guidelines for Research Practice. Communications of the Association for Information Systems, 4(7), 1–77. Geisser, S. (1974). A Predictive Approach to the Random Effect Model. Biometrika, 61(1), 101– 107. Gino, F., Schweitzer, M. E., Mead, N. L., & Ariely, D. (2011). Unable to Resist Temptation: How Self-Control Depletion Promotes Unethical Behavior. Organizational Behavior and Human Decision Processes, 115(2), 191–203. https://doi.org/10.1016/j.obhdp.2011.03.001 Gray, R., Bebbington, J., & Mcphail, K. (1994). Teaching Ethics in Accounting and the Ethics of Accounting Teaching: Educating for Immorality and a Possible Case for Social and Environmental Accounting Education. Accounting Education: An International Journal, 3(1), 51–75. https://doi.org/10.1080/09639289400000005 Guba, E. G., & Lincoln, Y. S. (1994). Competing Paradigms in Qualitative Research. In Handbook of Qualitative Research (pp. 105–117). https://doi.org/http://www.uncg.edu/hdf/facultystaff/Tudge/Guba%20&%20Lincoln%20199 4.pdf Guffey, D. M., & McCartney, M. W. (2008). The Perceived Importance of an Ethical Issue as a Determinant of Ethical Decision-Making for Accounting Students in an Academic Setting. 93 University of Ghana http://ugspace.ug.edu.gh Accounting Education, 17(3), 327–348. https://doi.org/10.1080/09639280701272666 Haines, R., Street, M. D., & Haines, D. (2008). The Influence of Perceived Importance of an Ethical Issue on Moral Judgment, Moral Obligation and Moral Intent. Journal of Business Ethics, 81(2), 387–399. Hair, J. F. J., Hult, T. M. G., Ringle, C. M., & Sarstedt, M. (2014). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a Silver Bullet. Journal of Marketing Theory and Practice, 19(2), 139–152. https://doi.org/10.2753/MTP1069- 6679190202 Haron, H., Ismail, I., Ibrahim, D. N., & Na, A. Lo. (2014). Factors Influencing Ethical Judgement of Auditors in Malaysia. Malaysian Accounting Review, 13(2), 47–86. Harris, J. R. (1989). Ethical Values and Decision Processes of Male and Female Business Students. Journal of Education for Business, 64(5), 234–238. https://doi.org/10.1080/08832323.1989.10117365 Harris, J. R. (1991). Ethical Values and Decision Processes of Business and Non-Business Students: A Four-Group Study. The Journal of Legal Studies Education, 9(1), 215–232. Heale, R., & Twycross, A. (2015). Validity and Reliability in Quantitative Studies. Evidence Based Nursing, 18(3), 66–70. Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A New Criterion for Assessing Discriminant Validity in Variance-Based Structural Equation Modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8 94 University of Ghana http://ugspace.ug.edu.gh Holmes, K., Marriott, L., & Randal, J. (2012). Ethics and Experiments in Accounting: A Contribution to the Debate on Measuring Ethical Behaviour. Pacific Accounting Review, 24(1), 80–100. https://doi.org/10.1108/01140581211221560 Hunt, S. D., Wood, V. R., & Chonko, L. B. (1989). Corporate Ethical Values and Organizational Commitment in Marketing. Journal of Marketing, 53(3), 79–90. Ismail, H. (2015). Gender Differences in Personal Values: A Business Perspective. International Journal of Commerce and Management, 25(4), 687–698. https://doi.org/10.1108/IJCoMA- 10-2013-0104 Ismail, S., & Yussof, S. H. (2016). Cheating Behaviour among Accounting Students: Some Malaysian Evidence. Accounting Research Journal, 29(1), 20–33. https://doi.org/10.1108/ARJ-05-2014-0050 Jacob, L., Pedersen, T., & Heinrich, R. F. (2011). Positive Ethical Decision-Making. On the Need to Understand How the Human Mind Works in Order to Improve Business Practice. Paris. Johnston, M. M. (2010). Perceived Importance of an Ethical Situation (PIE) on Ethical Judgment and Intention. Northern Iowa. Jones, T. M. (1991). Ethical Decision Making by Individuals in Organizations: An Issue- Contingent Model. Academy of Management Review, 16(2), 366–396. Kimberlin, C. L., & Winterstein, A. G. (2008). Validity and Reliability of Measurement Instruments used in Research. American Journal of Health-System Pharmacy, 65(23), 2276–2284. https://doi.org/10.2146/ajhp070364 95 University of Ghana http://ugspace.ug.edu.gh Lau, S. (1988). The Value Orientations of Chinese University Students in Hong Kong. International Journal of Psychology, 23(1–6), 583–596. https://doi.org/10.1080/00207598808247787 Leitsch, D. L. (2006). Using Dimensions of Moral Intensity to Predict Ethical Decision-making in Accounting. Accounting Education: An International Journal, 15(2), 135–149. Lincoln, S. H., & Holmes, E. K. (2011). Articles Ethical Decision Making: A Process Influenced by Moral Intensity. Journal of Healthcare, Science and the Humanities, I(1), 55–69. Longenecker, J. G., Mckinney, J. A., & Moore, C. W. (2004). Religious Intensity, Evangelical Christianity and Business Ethics: An Empirical Study. Journal of Business Ethics, 55(1), 373–386. Marcoulides, G. A., Chin, W. W., & Saunders, C. (2009). A Critical Look at Partial Least Squares Modeling. MIS Quarterly, 33(1), 171–175. Marcoulides, & Saunders. (2006). Editor’s Comments: PLS: A Silver Bullet? MIS Quarterly, 30(2), 3–9. https://doi.org/10.2307/25148727 Martijn, C., Tenbült, P., Merckelbach, H., Dreezens, E., & de Vries, N. K. (2002). Getting a Grip on Ourselves: Challenging Expectancies about Loss of Energy after Self-Control. Social Cognition, 20(6), 441–460. https://doi.org/10.1521/soco.20.6.441.22978 Mbawuni, J., & Nimako, S. G. (2015). Modelling Job-related and Personality Predictors of Intention to Pursue Accounting Careers among Undergraduate Students in Ghana. World Journal of Education, 5(1), 65–81. https://doi.org/10.5430/wje.v5n1p65 Mccuddy, M. K., & Peery, B. L. (1996). Selected Individual Differences and Collegians’ Ethical 96 University of Ghana http://ugspace.ug.edu.gh Beliefs. Journal of Business Ethics, 15(1), 261–272. Modarres, A., & Rafiee, A. (2011). Influencing Factors on the Ethical Decision Making of Iranian Accountants. Social Responsibility Journal, 7(1), 136–144. https://doi.org/10.1108/17471111111114594 Mohamed, Z. A. K. A., Mad Shah, I., & Jusoh, A. (2016). Consideration and Methodological Approaches in Studying Transformational Leadership Impact on Work Performance Behaviors. International Journal of Advanced Research, 4(1), 889–907. Nitzl, C. (2016). The Use of Partial Least Squares Structural Equation Modelling (PLS-SEM) in Management Accounting Research: Directions for Future Theory Development. Journal of Accounting Literature, 37(1), 19–35. https://doi.org/10.1016/j.acclit.2016.09.003 Nsor-ambala, R., & Onumah, J. M. (2015). Ethics Education in Management Accounting Studies: Evidence from Business Schools in Ghana. Journal of Accounting and Taxation, 7(May), 88–96. https://doi.org/10.5897/JAT2015.0178 Obalola, M., & Adelopo, I. (2012). Measuring the Perceived Importance of Ethics and Social Responsibility in Financial Services: A Narrative-Inductive Approach. Social Responsibility Journal, 8(3), 418–432. https://doi.org/10.1108/17471111211247992 Pan, Y., & Sparks, J. R. (2012). Predictors, Consequence and Measurement of Ethical Judgments: Review and Meta-analysis. Journal of Business Research, 65(1), 84–91. https://doi.org/10.1016/j.jbusres.2011.02.002 Patterson, D. M. (2001). Causal Effects of Regulatory, Organizational and Personal Factors on Ethical Sensitivity. Journal of Business Ethics, 30(1), 123–159. 97 University of Ghana http://ugspace.ug.edu.gh Pierce, B., & Sweeney, B. (2010). The Relationship between Demographic Variables and Ethical Decision Making of. International Journal of Auditing, 14(1), 79–99. https://doi.org/10.1111/j.1099-1123.2009.00404.x Randall, D. M., & Fernandes, M. E. (1991). The Social Desirability Response Bias in Ethics Research. Journal of Business Ethics, 10(1), 805–817. Rest, J. R. (1986). Moral Development: Advances in Research & Theory. New York: Praeger. Rest, J. R. (1988). Why does College Promote Development in Moral Judgement? Journal of Moral Education, 17(3), 183–194. https://doi.org/10.1080/0305724880170303 Riemenschneider, C. K., Leonard, L. N. K., & Manly, T. S. (2011). Students’ Ethical Decision- Making in an Information Technology Context : A Theory of Planned Behavior Approach. Journal of Information Systems Education, 22(3), 203–214. Robin, D. P., Reidenbach, E. R., & Forest, P. J. (1996). The Perceived Importance of an Ethical Issue as an Influence on the Ethical Decision-making of Ad Managers. Journal of Business Research, 35(1), 17–28. Robinson, S. L., & Bennett, R. J. (2013). A Typology of Deviant Workplace Behaviors: A Multidimensional Scaling Study. Academy of Management Journal, 38(2), 555–572. Sarstedt, M., Ringle, C. M., Smith, D., Reams, R., & Hair, J. F. (2014). Partial Least Squares Structural Equation Modeling (PLS-SEM): A Useful Tool for Family Business Researchers. Journal of Family Business Strategy, 5(1), 105–115. https://doi.org/10.1016/j.jfbs.2014.01.002 Saunders, M., Lewis, P., & Thornhill, A. (2009). Research Methods for Business Students. 98 University of Ghana http://ugspace.ug.edu.gh Prentice Hall (Vol. 5). https://doi.org/10.1017/CBO9781107415324.004 Scott, W. A. (1965). Personal Values Scale, in W. A. Scott (Ed.). Values and Organisations. Chicago: Rand McNally Colleg Publishing Company. Shafer, W. E., Morris, R. E., & Ketchand, A. A. (2001). Effects of Personal Values on Auditors’ Ethical Decisions. Accounting, Auditing & Accountability Journal, 14(3), 254–277. Simpson, S. N. Y., Onumah, J. M., & Oppong-Nkrumah, A. (2016). Ethics Education and Accounting Programmes in Ghana: Does University Ownership and Affiliation Status Matter? International Journal of Ethics Education, 1(1), 43–56. https://doi.org/10.1007/s40889-015-0005-4 Singhapakdi, A. (1999). Perceived Importance of Ethics and Ethical Decisions in Marketing. Journal of Business Research, 45(1), 89–99. Singhapakdi, A., Vitell, S. J., Lee, D.-J., Nisius, A. M., & Yu, G. B. (2013). The Influence of Love of Money and Religiosity on Ethical Decision-Making in Marketing. Journal of Business Ethics, 114(1), 183–191. https://doi.org/10.1007/s Sliwa, P. (2017). Moral Understanding as Knowing Right from Wrong. Ethics, 127(3), 521–552. https://doi.org/10.1086/690011 Sobh, R., & Perry, C. (2006). Research Design and Data Analysis in Realism Research. European Journal of Marketing, 40(11–12), 1194–1209. https://doi.org/10.1108/03090560610702777 Sosik, J. J., Kahai, S. S., & Piovoso, M. J. (2009). Silver Bullet or Voodoo Statistics? A Primer for Using the Partial Least Squares Data Analytic Technique in Group and Organization 99 University of Ghana http://ugspace.ug.edu.gh Research. Group & Organization Management, 34(1), 5–36. Stanga, K. G., & Turpen, R. A. (1991). Ethical Judgments on Selected Accounting Issues: An Empirical Study. Journal of Business Ethics, 10(1), 739–747. Stonciuviene, N., & Naujokaitiene, J. (2013). Formation of the Ethics of Professional Accountants from a Moral Standpoint: Analysis of Decisive Factors and their Influence. European Scientific Journal, 1(1), 106–115. Stone, M. (1974). Cross-Validatory Choice and Assessment of Statistical Predictions. Journal of the Royal Statistical Society, 36(2), 111–147. Stone, M. (1977). An Asymptotic Equivalence of Choice of Model by Cross-Validation and Akaike’s Criterion. Journal of the Royal Statistical Society, 39(1), 44–49. Sweeney, B., & Costello, F. (2009). Moral Intensity and Ethical Decision-making: An Empirical Examination of Undergraduate Accounting and Business Students. Accounting Education: An International Journal, 18(1), 75–97. https://doi.org/10.1080/09639280802009454 Thatcher, R. W. (2010). Validity and Reliability of Quantitative Electroencephalography. Journal of Neurotherapy, 14(2), 122–152. https://doi.org/10.1080/10874201003773500 Urbach, N., & Ahlemann, F. (2010). Structural Equation Modeling in Information Systems Research Using Partial Least Squares. Journal of Information Technology Theory and Application, 11(2), 5–40. Wells, D. D., & Molina, A. D. (2017). The Truth about Honesty. Journal of Public and Nonprofit Affairs, 3(3), 292–308. https://doi.org/10.20899/jpna.3.3.292-308 100 University of Ghana http://ugspace.ug.edu.gh Wetzels, M., Odekerken-Schroder, G., & van Oppen, C. (2009). Using PLS Path Modeling for Assessing Hierarchical Construct Models: Guidelines and Empirical Illustration. MIS Quarterly, 33(1), 177–195. Whitla, P. (2011). Integrating Ethics into International Business Teaching: Challenges and Methodologies in the Greater China Context. Journal of Teaching in International Business, 22(3), 168–184. https://doi.org/10.1080/08975930.2011.653745 Wong, K. K. (2013). Partial Least Squares Structural Equation Modeling (PLS-SEM) Techniques Using SmartPLS. Marketing Bulletin, 24(1), 1–32. Zaichkowsky, J. L. (1985). Measuring the Involvement Construct. Journal of Consumer Research, 12(3), 341. https://doi.org/10.1086/208520 Ziegenfuss, D. E., & Singhapakdi, A. (1994). Professional Values and the Ethical Perceptions of Internal Auditors. Managerial Auditing Journal, 9(1), 34–44. 101 University of Ghana http://ugspace.ug.edu.gh APPENDIX A – TABLES OF RESULTS Gender across Institutions GEN_CODE FEMALE MALE Total INST Central University College 52 27 79 University of Ghana 41 65 106 University of Professional 67 76 143 Studies Accra Valley View University 44 72 116 Total 204 240 444 Gender across Levels LEV_CODE 300 400 Total GEN_CODE FEMALE Count 60 144 204 % within GEN_CODE 29.4% 70.6% 100.0% MALE Count 44 196 240 % within GEN_CODE 18.3% 81.7% 100.0% Total Count 104 340 444 % within GEN_CODE 23.4% 76.6% 100.0% 102 University of Ghana http://ugspace.ug.edu.gh Age across Gender GEN_CODE Total FEMALE MALE AGE Count 2 0 2 % within GEN_CODE 1.0% 0.0% 0.5% 18-22years Count 100 62 162 % within GEN_CODE 49.0% 25.8% 36.5% 23-25years Count 98 149 247 % within GEN_CODE 48.0% 62.1% 55.6% 26-30years Count 3 27 30 % within GEN_CODE 1.5% 11.3% 6.8% Above 30years Count 1 2 3 % within GEN_CODE 0.5% 0.8% 0.7% Total Count 204 240 444 % within GEN_CODE 100.0% 100.0% 100.0% Age across Levels LEV_CODE Total 300 400 AGE Count 2 0 2 % within LEV_CODE 1.9% 0.0% 0.5% 18-22years Count 65 97 162 % within LEV_CODE 62.5% 28.5% 36.5% 23-25years Count 31 216 247 % within LEV_CODE 29.8% 63.5% 55.6% 26-30years Count 5 25 30 % within LEV_CODE 4.8% 7.4% 6.8% Above Count 1 2 3 30years % within LEV_CODE 1.0% 0.6% 0.7% Total Count 104 340 444 % within LEV_CODE 100.0% 100.0% 100.0% 103 University of Ghana http://ugspace.ug.edu.gh Gender across Religion RELIGION Total CHRISTIAN MUSLIM NONE GEN_CODE FEMALE Count 180 9 15 204 % within RELIGION 44.7% 64.3% 55.6% 45.9% MALE Count 223 5 12 240 % within RELIGION 55.3% 35.7% 44.4% 54.1% Total Count 403 14 27 444 % within RELIGION 100.0% 100.0% 100.0% 100.0% 104 University of Ghana http://ugspace.ug.edu.gh APPENDIX B – RESULTS OF INDEPENDENT T-TESTS Public vs Private Tertiary Institutions and Personal Values Group Statistics INST_CODE N Mean Std. Deviation Std. Error Mean Intellectualism PUBLIC 246 5.0854 1.41704 .09035 PRIVATE 197 5.0013 1.43158 .10200 Honesty PUBLIC 247 4.8735 1.42064 .09039 PRIVATE 197 4.5838 1.41600 .10089 Self_control PUBLIC 247 5.0516 1.45813 .09278 PRIVATE 197 4.6294 1.47245 .10491 Religiousness PUBLIC 247 5.4049 1.61518 .10277 PRIVATE 196 5.1059 1.58052 .11289 105 University of Ghana http://ugspace.ug.edu.gh Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2- Mean Std. Error Difference F Sig. t df tailed) Difference Difference Lower Upper Intellectualism Equal variances 1.358 .244 .618 441 .537 .08410 .13610 -.18339 .35159 assumed Equal variances .617 418.247 .537 .08410 .13626 -.18374 .35193 not assumed Honesty Equal variances .142 .706 2.138 442 .033 .28973 .13551 .02340 .55605 assumed Equal variances 2.139 420.892 .033 .28973 .13546 .02347 .55598 not assumed Self_control Equal variances .035 .851 3.018 442 .003 .42218 .13989 .14724 .69712 assumed Equal variances 3.015 418.513 .003 .42218 .14005 .14689 .69746 not assumed Religiousness Equal variances .408 .523 1.954 441 .051 .29899 .15305 -.00180 .59979 assumed Equal variances 1.958 422.249 .051 .29899 .15267 -.00109 .59907 not assumed 106 University of Ghana http://ugspace.ug.edu.gh Public vs Private Tertiary Institutions and Ethical Judgement Group Statistics INST_CODE N Mean Std. Deviation Std. Error Mean Ethical Judgement PUBLIC 247 3.8165 1.41331 .08993 Overall PRIVATE 197 3.5847 1.32630 .09449 Ethical Judgement 1 PUBLIC 247 3.9385 1.69778 .10803 PRIVATE 197 3.6005 1.60829 .11459 Ethical Judgement 2 PUBLIC 247 3.6980 1.55271 .09880 PRIVATE 197 3.5674 1.51131 .10768 107 University of Ghana http://ugspace.ug.edu.gh Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2- Mean Std. Error Difference F Sig. t df tailed) Difference Difference Lower Upper Ethical Equal Judgement variances .061 .806 1.765 442 .078 .23184 .13138 -.02637 .49006 Overall assumed Equal variances not 1.777 430.467 .076 .23184 .13045 -.02454 .48823 assumed Ethical Equal Judgement variances .708 .401 2.133 442 .033 .33795 .15844 .02656 .64935 1 assumed Equal variances not 2.146 429.141 .032 .33795 .15748 .02843 .64748 assumed Ethical Equal Judgement variances .137 .711 .891 442 .374 .13055 .14658 -.15753 .41863 2 assumed Equal variances not .893 424.957 .372 .13055 .14613 -.15669 .41778 assumed 108 University of Ghana http://ugspace.ug.edu.gh Male vs Female Accounting Students and Personal Values Group Statistics GEN_CODE N Mean Std. Deviation Std. Error Mean Intellectualism MALE 240 5.0208 1.46332 .09446 FEMALE 203 5.0800 1.37564 .09655 Honesty MALE 240 4.7458 1.44385 .09320 FEMALE 204 4.7439 1.40448 .09833 Self_control MALE 240 4.9063 1.50246 .09698 FEMALE 204 4.8150 1.45046 .10155 Religiousness MALE 240 5.2729 1.57334 .10156 FEMALE 203 5.2722 1.64561 .11550 109 University of Ghana http://ugspace.ug.edu.gh Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2- Mean Std. Error Difference F Sig. T df tailed) Difference Difference Lower Upper Intellectualism Equal variances 1.314 .252 -.436 441 .663 -.05922 .13577 -.32605 .20762 assumed Equal variances -.438 436.082 .661 -.05922 .13507 -.32469 .20626 not assumed Honesty Equal variances 1.122 .290 .014 442 .988 .00196 .13579 -.26491 .26883 assumed Equal variances .014 434.036 .988 .00196 .13548 -.26432 .26825 not assumed Self_control Equal variances .187 .666 .648 442 .517 .09130 .14083 -.18547 .36807 assumed Equal variances .650 434.889 .516 .09130 .14042 -.18469 .36729 not assumed Religiousness Equal variances .573 .449 .005 441 .996 .00075 .15322 -.30039 .30189 assumed Equal variances .005 421.934 .996 .00075 .15380 -.30156 .30306 not assumed 110 University of Ghana http://ugspace.ug.edu.gh Male vs Female Accounting Students and Ethical Judgement Group Statistics GEN_CODE N Mean Std. Deviation Std. Error Mean Ethical Judgement Overall MALE 240 3.8206 1.41616 .09141 FEMALE 204 3.5879 1.32567 .09282 Ethical Judgement 1 MALE 240 3.8554 1.70742 .11021 FEMALE 204 3.7098 1.61502 .11307 Ethical Judgement 2 MALE 240 3.7867 1.55090 .10011 FEMALE 204 3.4676 1.49965 .10500 111 University of Ghana http://ugspace.ug.edu.gh Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2- Mean Std. Error Difference F Sig. t df tailed) Difference Difference Lower Upper Ethical Judgement Equal variances .579 .447 1.777 442 .076 .23270 .13097 -.02471 .49010 Overall assumed Equal variances 1.786 437.880 .075 .23270 .13027 -.02334 .48873 not assumed Ethical Judgement Equal variances 1.054 .305 .918 442 .359 .14561 .15862 -.16612 .45735 1 assumed Equal variances .922 436.958 .357 .14561 .15790 -.16473 .45595 not assumed Ethical Judgement Equal variances .040 .842 2.194 442 .029 .31910 .14547 .03320 .60500 2 assumed Equal variances 2.200 434.710 .028 .31910 .14507 .03397 .60423 not assumed 112 University of Ghana http://ugspace.ug.edu.gh APPENDIX C – QUESTIONNAIRE This questionnaire is being used to solicit information on “Personal Values and Perceived Importance of an Ethical Issue as Determinants of Ethical Decision-Making”. Kindly respond by providing the appropriate answers and comments where necessary. Information provided by respondents will be treated as confidential and used only for the purpose of the research work. Participation in this exercise is totally voluntary and one can withdraw from the exercise at his/her discretion. SECTION A: ETHICAL JUDGEMENT In this section, the researcher seeks to evaluate the respondent’s ethical judgement and decision process. Scenario 1: Take Examination Home Adwoa Yankey is pursuing a master’s degree in accounting in a major university. She received a graduate assistantship in the accounting department to help pay her bills. She works 20 hours a week and makes copies of tests and performs various other routine duties as needed. Adwoa excelled as an undergraduate major but has found graduate school to be much more difficult. She has already made two C’s in graduate school and one more grade of C will mean automatic expulsion from the school. She is currently taking a theory course that is very difficult and she barely has a B average. One day in the copy room Adwoa noticed that some smudged copies of the next examination paper for her theory class had been thrown in the trash. Out of curiosity, she briefly looked over the examination paper and realized she could not pass it. However, if she took the examination paper home and started studying it she could make an excellent grade. Adwoa knows that it is a violation of university policy to obtain a copy of an examination paper prior to its being given. Action: Adwoa decides to take the examination paper home so she can prepare for the next examination. Please evaluate this action of Adwoa Yankey. 1 2 3 4 5 6 7 Perceived Im porta nce of Ethic al Iss ues Extremely Extremely Important unimportant Issue Issue Highly insignificant Highly significant issue Issue of no concern Issue of High Concern Trivial Issue Fundamental Issue Mo ral Eq uity S cale Unfair Fair Unjust Just Not Morally Right Morally Right Not Acceptable to my Acceptable to my Family Family 113 University of Ghana http://ugspace.ug.edu.gh Univar iate E thics M easu re Unethical Ethical Behavioural Intention If you were responsible for making the decision in the scenario above, what is the probability you would make the same decision?’ Highly improbable Highly probable Scenario 2: The Grading Error Naa Akorfa Okai is a senior accounting major student at a major university who is expecting to graduate at the end of the current semester. Her instructor returned all examinations on the final day of classes so the students could prepare for the final examination. Naa noticed that the instructor had made an arithmetic error on her last regular examination in her favour. The mistake was substantial; it was 30 points in her favour. The 30 points could make a difference in Naa’s final course grade. The professor has not approved ignoring grading errors in the student’s favour but has no stated or written policy on the matter. Action: Naa decides not to inform the instructor about the grading error. Please evaluate this action by Naa Akorfa Okai. 1 2 3 4 5 6 7 Perceived Im porta nce of Ethic al Iss ues Extremely unimportant Extremely Important Issue Issue Highly insignificant Highly significant issue Issue of no concern Issue of High Concern T rivial Issue Fundamental Issue Mo ral Eq uity S cale Unfair Fair Unjust Just Not Morally Right Morally Right Not Acceptable to my Acceptable to my F amily Family Univar iate E thics M easu re Unethical Ethical Behavioural Intention If you were responsible for making the decision in the scenario above, what is the probability you would make the same decision?’ Highly improbable Highly probable 114 University of Ghana http://ugspace.ug.edu.gh SECTION B: ETHICAL BEHAVIOUR The following section seeks to evaluate your ethical behaviour perceptions. On a scale of 1 to 7, (where 1=Extremely Unlikely, 2=Moderately Unlikely, 3=Slightly Unlikely, 4=Neutral, 5=Slightly Likely, 6= Moderately Likely, and 7= Extremely Likely), indicate the extent to which you are likely to engage in the following behaviours below in an organisation. Personal Use 1 2 3 4 5 6 7 Using company services for personal use Doing personal business on company time Pilfering company materials and supplies Taking extra personal time (lunch hour, breaks, early departure) Passing Blame Concealing one’s errors Passing blame for errors to an innocent co-worker Claiming credit for someone else’s work Bribery Giving gifts/favours in exchange for preferential treatment Accepting gifts/favours in exchange for preferential treatment Falsification Falsifying time/quality/quantity reports Calling in sick to take a day off Authorizing a subordinate to violate company rules Padding Expenses Padding an expense account up to 10% Padding an expense account more than 10% Deception Taking longer than necessary to do a job Divulging confidential information Not reporting others’ violations of company policies and rules SECTION C: PERSONAL VALUES The following section seeks to evaluate you’re the personal values you exhibit and uphold. On a scale of 1 to 7, (where 1=Strongly Dislike It, 2=Moderately Dislike It, 3=Slightly Dislike It, 4=Neutral, 5= Slightly Like It, 6= Moderately Like It, and 7=Strongly Like It), indicate the extent to which you like or dislike each of the following traits. 115 University of Ghana http://ugspace.ug.edu.gh Intellectualism 1 2 3 4 5 6 7 Having an active interest in all things scholarly Having a keen interest in international, national and local affairs Developing an appreciation of the fine arts – music, drama, literature, and ballet Having a strong intellectual curiosity Honesty Never cheating or having anything to do with cheating, even for a friend Always telling the truth even though it may hurt one’s self or others Speaking one’s mind truthfully, without regard for the consequences Volunteering information concerning wrongdoing, even if friends are involved Self-control Never losing one’s temper, no matter what the reason Practising self-control Not expressing anger, even when one has a reason for doing so Replying to anger with gentleness Religiousness Being devout in one’s religious faith Always living one’s religion in one’s daily life Always attending religious services regularly and faithfully Having faith in a being greater than man SECTION D: DEMOGRAPHIC INFORMATION 1. Indicate your gender. (a) Male (b) Female 2. Indicate your age. (a) 18 – 22 years (b) 23 – 25 years (c) 26 – 30 years (d) above 30 years 3. Indicate your institution: ………………………………………………………………… 4. Indicate your level. (a) Level 300 (b) Level 400 5. Indicate your course major: ……………………………………………………………… 6. Do you have any professional qualification? (a) Yes (b) No If yes, indicate (eg. ICA, ACCA, CIMA, etc): …………………………………………… 7. Are you religious? (a) Yes (b) No 8. If yes, please state your religious affiliation: ………………………………………………. 9. State the number of courses done with an ethics component: …………………………….. 116