SEARCH

SEARCH BY CITATION

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

Since the financial crisis, the malfeasance of business leaders has been a recurring theme in the news, along with calls for increased regulation and oversight. This focus on the ethics of the business community raises a concern about the ethics of those in business or going into business. The ethics of business people and business students has been explored by a number of researchers using survey techniques. We propose and report the results of an alternative method for investigating unethical behavior by students. In a motivated economic experiment with introductory level students, we find that business students were almost twice as likely to lie for a monetary reward as students in other disciplines, demonstrating the need for effective business ethics.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

Businesses have received a remarkable amount of negative press since the financial crisis. The conviction of Bernard Madoff, the civil case against Goldman-Sachs for fraudulent misconduct, and the accusations of “robo-signing” foreclosures all contribute the public belief that those in business are untrustworthy. Incidences of fraud have not been restricted to the banking sector. The examples of Enron, WorldCom, and Parmalat are still cited as evidence of unethical behavior in business. Public reports of losses to employee theft and fraud, approximately one trillion dollars in the United States alone (Association of Certified Fraud Examiners), further harm the reputation of business.

The public perception begs the question: are those in the world of business truly more prone to dishonesty than those in other fields? A growing body of literature attempts to address this issue. Direct surveys of business people are rare, but some have found that dishonest behavior is not uncommon (Greeman and Sherman 1999). Dishonesty in the workplace has been linked to cheating in university and college (Lawson 2004; Nonis and Smith 2001; Simkin and McLeod 2010 p. 443; Sims 1993; Teixeira and Rocha 2010), making surveys of students a viable alternative. As researchers have more access to students, more work has been done exploring the ethical behavior of students in business schools and business programs; Day et al. (2010) and Brown and McInerney (2008) both provided reviews of the existing literature. In general, this research finds that academic dishonesty is extremely common. In short, students cheat.

The ubiquity of cheating by students is a concern for both academic institutions and future employers. If cheating is prominent in a program or institution, the reputation of the school falls, reducing the value of the degree to potential students. Detecting and preventing cheating is clearly in the best interest of any academic institution. From the viewpoint of a future employer, academic dishonesty is a concern as well. If a degree is obtained through fraud, the employer cannot be certain that the applicant actually has the skills the degree is supposed to indicate. This would cause employers to demand more experience of job applicants. Of course, job experience tends to be associated with demands for higher pay.

A natural starting place is to see if students differ by discipline of study. Given the current attitudes portrayed in the media, it makes particular sense to focus on business students. There is disagreement in the literature about the relationship between field of study and dishonesty. Some find that business students are significantly more likely to engage in dishonest behavior than students in other disciplines (Bernardi et al. 2004; Smyth and Davis 2004). Other research has found no relationship between dishonesty and field of study (Klein et al. 2007; Molnar et al. 2008; Smyth et al. 2009; Zopiatis and Karmbia-Kapardis 2008). If business students are in fact more prone to dishonest behavior, it will be in the best interest of business schools to take steps to improve the ethics of business students both while they are studying and later in life.

The studies cited earlier all report the results of unmotivated surveys. Given the conflicting results of this literature, an alternative method of assessing students' propensity for dishonesty is justified. We report the results of an economic experiment with monetary incentives in which subjects were given an opportunity to lie. In the experiment, business students were significantly more likely to lie than students in all other disciplines.

In the next section, we make the case for an alternative method of assessing students' behavioral tendencies. We then present the experimental design. This is followed by the results of the experiment. In the final section, we offer some discussion of the results and their implications.

THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

The existing literature on the likelihood of dishonest behavior of students makes use of two different types of unmotivated surveys. One approach is to ask students to report their own or other's dishonesty, generally in terms of academic cheating. This approach has two limitations. First, there is no general definition of cheating among students. For example, while some students may consider sharing information about an assignment cheating, others may simply view it as teamwork. A well-designed survey can address this, but it does make the survey more complex and time-consuming to complete. The larger limitation of this type of survey is that it relies on honest reporting of dishonest behavior. If business students are actually more honest than students in other disciplines, and honesty includes survey responses, business students would appear to be less honest than others because they told the truth about lying on a survey. For example, if those business students who score highly on a test of Machiavellianism are more willing to admit to dishonest behavior, they would appear to be more dishonest without actually being so. This could explain the findings of Bloodgood et al. (2010) and Tang and Chen (2008). This problem of honest reporting extends to the other survey method, which asks students about their attitudes toward dishonesty and cheating. The responses of students who are dishonest on a survey could indicate that they find dishonesty objectionable even though they have no reservations about taking such actions themselves.

Another weakness of the existing literature is that it focuses on academic dishonesty. In the minds of students, academic dishonesty occurs at the expense of the course instructor. Students' attitudes toward their instructors are potentially different from their attitudes toward employers, other students, or future clients. In existing studies of academic dishonesty, a difference between business and nonbusiness students may arise due to how different students view their instructors and their program. This may play a part in explaining why students are found to be more dishonest than practitioners (Teixeira and Rocha 2010).

Techniques developed in experimental economics offer an alternative to surveys. Instead of being asked whether or not they have been dishonest, an experimental environment gives students an opportunity to be dishonest in exchange for monetary gain. The choices of students in this environment can be directly observed, meaning the analysis is based on actual behavior rather than self-reported actions.

There is a growing body of literature in both economics and business that makes use of motivated experiments to further our understanding of dishonesty. Mazar and Ariely (2006) and Mazar et al. (2008) used a simple testing environment to explore the internal rewards, social norms or self-concept, threshold explanation of cheating. Different groups of students were asked to find pairs of numbers in matrices that summed to 10. The control group received only the matrices and a sheet on which to report the number of pairs found. Those in the treatment groups were asked to perform short tasks1 before searching for pairs of numbers and were provided with an opportunity to cheat. The groups that were provided the opportunity to cheat reported higher average scores than groups that did not have the opportunity to cheat. The increased scores are interpreted as evidence cheating by the authors. While the treatments explored are interesting, this approach is limited by the fact that dishonesty can only be inferred from aggregate behavior and not directly observed at an individual level. Observing only group behavior makes it difficult to determine the impact individual characteristics have on dishonesty.

Another, arguably simpler, approach is based on a sender–receiver environment initially used by Gneezy (2005). In this type of experiment, one subject is shown a set of two payoffs. This subject then sends a message (making them a sender) to the other subject (the receiver) about the two payoffs. The message either honestly identifies which payoff would yield the most money to the receiver or is a lie. The direct observation approach avoids the limitations of other methods.

This approach has been used to consider the relationship between a number of factors and the propensity to lie. The original work (Gneezy 2005) examined the impact of different levels of monetary incentives on lying. Dreber and Johannesson (2008), Erat and Gneezy (2011), and Childs (2012) examined lying by gender. None of the preceding work has considered the relationship between the chosen field of study and lying. We use this sender–receiver environment to examine the relationship between field of study and willingness to lie.

PROCEDURE

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

The experimental environment is almost identical to that used by Gneezy (2005).2 Individuals in separate rooms were paired with an unknown partner. Subjects in one room were assigned the role of sender and those in the other room the role of receiver. The sender sees two different payoffs: one labeled A and the other B. For half the senders, A offered a payoff of ($15, $5), $15 for the sender and $5 for the receiver, while B offered a payoff of ($5, $15). For the remaining senders, the payoffs were A ($5, $15) and B ($15, $5).

In presenting these payoffs to the senders, two forms were used. The first, the gains treatment, was identical to treatment three by Gneezy (2005) in which all payoffs are framed as gains. In the second form, the losses treatment, senders were initially endowed with a voucher for $20, and the payoffs were framed as losses. While there was a difference in framing, the net payoffs in both forms were identical.

The sender then chooses one of two message to send his or her partner; either “Option A will earn you more money than option B” or “Option B will earn you more money than option A.” A lie is defined as a message the sender knows to be inaccurate.3 Therefore, when the payoffs shown to the sender were actually A ($15, $5), B ($5, $15) sending the message “Option A will earn you more money than option B” is a lie.

After having received a message from the sender, receivers choose which payoff both will receive. The receivers must make this choice with no information other than the sender's message. After choosing their message or selecting payoffs, the subjects were asked to complete a brief survey. In this survey, the subjects were asked to identify their major and faculty of study.

The experiment was conducted with 200 students, 100 senders and 100 receivers, recruited from introductory economics classes at a Western Canadian university. The composition of the self-reported fields of study is consistent with the class records. The experiment was originally designed to consider the impact of framing on lying.4 In this article, however, we focus only on the relationship between faculty of study and behavior. All participants remained anonymous to their partners. The actions of three subjects were dropped from the analysis for poor understanding of the instructions.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

The results of the experiment are shown in Figure 1.

image

Figure 1. Frequency of Lying by Faculty of Study and Treatment.

Download figure to PowerPoint

In total, 55 of 97 (57 percent) subjects sent incorrect messages with 38 of 54 (70 percent) business students and 17 of 43 (40 percent) nonbusiness students lying.5 This difference is statistically significant (P value 0.002).6 It is possible that students in different faculties reacted to the framing differently. The frequency of lying was higher for both groups in the loss treatment, increasing from 20 of 31 (64.5 percent) to 18 of 23 (78 percent) business students (P value 0.274) and from six of 19 (31.5 percent) to 11 of 24 (46 percent) nonbusiness students (P value 0.342). The observed difference between subjects based on field of study is statistically significant regardless of treatment (P= 0.024 in the gains treatment and P= 0.022 in the losses treatment).

Given that lying is only profitable if the false information is believed, it is also important to consider trust. The actions of three receivers are not included in this analysis as they did not identify their faculty of study. From a receiver's point of view, there is no difference between the treatments so the actions of all the receivers were pooled. Twenty-four of 35 (69 percent) business students trusted the messages they received and chose the recommended action, whereas 46 of 62 (74 percent) nonbusiness students trusted their senders (P value 0.553). Though not statistically significant, a lower level of trust by business students is reasonable given their greater propensity to lie.

CONCLUSION AND DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES

Lying, particularly employee theft and fraud, is a major problem for both businesses and regulators, costing American businesses an estimated $994 billion in 2008 (Association of Certified Fraud Examiners n.d.). In addition to the costs of internal theft and fraud, the business community around the world has seen several high-profile cases of dishonesty (Goldman-Sachs) and outright fraud (Bernard Madoff) in recent years. The Sarbanes–Oxley Act illustrates the increase in regulation that can occur in response to high-profile instances of dishonesty by business people.

Given the explicit costs of dishonest dealing and the likely costs of increased regulation and oversight, businesses are rightly concerned with reducing dishonesty in the workplace. In order to reduce dishonesty, it must be understood. Many researchers have demonstrated a link between dishonest behavior as students and malfeasance in professional life (Lawson 2004; Simkin and McLeod 2010, p. 443; Sims 1993; Teixeira and Rocha 2010).

The next logical step is to understand the nature and motives of students when they engage in dishonesty. This question has generated a wealth of literature. One of the recurring themes is whether or not students in business have a higher propensity to act dishonestly than others. If business students are in fact more likely to be dishonest, much can be gained by focusing on the ways in which business students differ from students in other faculties and programs.

The existing literature on academic dishonesty reports mixed results. Some demonstrate that business students are in fact more dishonest or are more accepting of dishonest behavior (Bernardi et al. 2004; Smyth and Davis 2004), while others have found no difference between students in different faculties (Klein et al. 2007; Molnar et al. 2008; Smyth et al. 2009; Zopiatis and Karmbia-Kapardis 2008).

The method of acquiring data on dishonesty binds all these studies together. These studies use surveys of students as their data source. This relies on students to honestly tell their professors that they have in fact engaged in academic dishonesty. Surveying students in this way could lead to findings that are the reverse of actual behavior. Groups of students that were in fact more honest about cheating on the survey would be seen as more dishonest than other students. Surveys of attitudes toward dishonesty are illuminating in their own right but are subject to the same concern.

The focus on academic misconduct creates another confounding factor. Students generally believe that the victims of cheating or plagiarism are their instructors and not fellow students or society in general. Some researchers have considered exactly this sort of factor in considering the impact of teaching vignettes on students' attitudes (Day et al. 2010) and found that the characteristics of the instructor have an impact on students' attitudes toward dishonesty. In one extreme case, the instructor facilitated cheating by students (Jones and Spraakman 2011). The victim of fraud, particularly in the manipulation of financial reports, is often an anonymous public. Therefore, if the characteristics of the instructor are central to academic cheating, the individual characteristics that lead to academic misconduct may not be accurate predictors of professional misconduct despite the correlation between the types of misconduct.

Instead of surveys, economic experiments with monetary incentives can be used to explore dishonest behavior. Mazar and Ariely (2006) and Mazar et al. (2008) used a simple search task to provide subjects with an opportunity to lie to increase their monetary payoff. While individual actions were not directly observed, they did observe an increase in aggregate reported successes when the subjects had the opportunity to cheat. The authors found that a number of factors have an influence on cheating, such as signing an ethics code statement, writing down the Ten Commandments, etc. However, the inability to observe individual choices makes this environment unsuited to researching the impact of individual characteristics on dishonesty.

We use another experimental environment, originally developed by Gneezy (2005), to explore the link between chosen field of study and dishonest behavior. In this environment, individual subjects have an opportunity to increase their payoffs by lying to an anonymous partner. We discover that those subjects who identified themselves as business students are significantly more likely to lie than those in any other discipline.

There are a variety of ways in which business students could differ from those in other disciplines. For example, individuals who are more motivated by money may be more likely to pursue a business degree than others. Business students could also be more competitive by nature, leading to their being more likely to make choices that improve their payoffs at the expense of others, whether or not those choices involve lying (Hurkens and Kartick 2009). Further, business students may be more Machiavellian than those in other disciplines, and this willingness to pursue an end may explain the observed difference.

The subjects in this experiment were recruited from introductory classes in economics. The business program at the university in question requires that students take an ethics class in their third year of study, meaning that the majority of business students had not completed an ethics course. Thus, this experiment does not provide a basis for assessing the effectiveness of the required ethics course, only the need for it. By allowing researchers to directly observe dishonest behavior in response to known motivation, future work can explore the impact of these factors and others, such as taking an ethics course, on a subject's directly observed propensity to lie.

NOTES
  • 1

    The extra tasks were things such as listing 10 books they had read in high school, listing the Ten Commandments, signing an honor code statement, and so on.

  • 2

    Full details and experimental instructions are available from the author.

  • 3

    This definition approximates the definition of fraud.

  • 4

    While the framing did have an impact on the frequency of lying, the frequency of lying is not statistically significant different (two-sided Pearson chi-square P= 0.335).

  • 5

    The subjects identified six different faculties of study—business, arts, science, engineering, education, and social work. Only students in business were statistically different.

  • 6

    All P values reported are from two-sided Pearson chi-squared tests.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THE CASE FOR AN ALTERNATIVE METHOD OF MEASUREMENT
  5. PROCEDURE
  6. RESULTS
  7. CONCLUSION AND DISCUSSION
  8. REFERENCES
  • Association of Certified Fraud Examiners. (n.d). 2008 Report to the Nation on Occupational Fraud and Abuse. Retrieved February 23, 2011, from http://www.acfe.com/uploadedFiles/ACFE_Website/Content/documents/2008-rttn.pdf.
  • Bernardi, R., Metzger, A., Bruno, R., Hoogkamp, M., Reyes, L., and Barnaby, G. 2004. “Examining the decision process of students' cheating behavior: An empirical study,” Journal of Business Ethics 50(4): 397301.
  • Bloodgood, J., Turnley, W., and Mudrack, P. 2010. “Ethics instruction and the perceived acceptability of cheating,” Journal of Business Ethics 95: 2337.
  • Brown, B., and McInerney, M. 2008. “Changes in academic dishonesty among business students in the United States 1999–2006,” International Journal of Management 25(3): 621633.
  • Childs, J. 2012. “Gender differences in lying,” Economics Letters 114(2): 147149.
  • Day, N., Hudson, D., Dobies, P., and Waris, R. 2010. “Student or situation? Personality and classroom context as predictors of attitudes about business school cheating,” Social Psychology of Education 14(2): 122.
  • Dreber, A., and Johannesson, M. 2008. “Gender differences in deception,” Economics Letters 99: 197199.
  • Erat, S., and Gneezy, U. 2011. “White lies,” Management Science Articles in Advance, 111. Available at http://dx.doi.org/10.1287/mnsc.1110.1449.
  • Gneezy, U. 2005. “Deception: The role of consequences,” American Economic Review 95: 384394.
  • Greeman, F., and Sherman, J. 1999. “Business school ethics—An overlooked topic,” Business and Society Review 104(2): 171177.
    Direct Link:
  • Hurkens, S., and Kartick, N. 2009. “Would I lie to you: On social preferences and lying,” Experimental Economics 12(2): 180192.
  • Jones, J., and Spraakman, G. 2011. “A case of academic misconduct: Does self-interest rule?” Accounting Perspectives 10(1): 122.
  • Klein, H., Levenburg, N., McKendall, M., and Mothersell, W. 2007. “Cheating during the college years: How do business school students compare?” Journal of Business Ethics 72: 197206.
  • Lawson, R. 2004. “Is classroom cheating related to business students' propensity to cheat in the ‘real world,’ ” Journal of Business Ethics 49(2): 189199.
  • Mazar, N., and Ariely, D. 2006. “Dishonesty in everyday life and its policy implications,” Journal of Public Policy and Marketing 52(1): 117126.
  • Mazar, N., Amir, O., and Ariely, D. 2008. “The dishonesty of honest people: A theory of self-concept maintenance,” Journal of Marketing Research 45(6): 633644.
  • Molnar, K., Kletke, M., and Chongwatpol, J. 2008. “Ethics vs. IT ethics: Do undergraduate students perceive a difference?” Journal of Business Ethics 83(4): 657671.
  • Nonis, S., and Smith, C. 2001. “Personal value profiles and ethical business decisions,” Journal of Education for Business 76(5): 251256.
  • Simkin, M., and McLeod, A. 2010. “Why do college students cheat?” Journal of Business Ethics 94: 441453.
  • Sims, R. 1993. “The relationship between academic dishonesty and unethical business practices,” Journal of Education for Business 68(4): 207211.
  • Smyth, L., Davis, J., and Kroncke, C. 2009. “Students' perceptions of business ethics: Using cheating as a surrogate for business situations,” Journal of Education for Business 84(4): 229238.
  • Smyth, M., and Davis, J. 2004. “Perceptions of dishonesty among two-year college students: Academic versus business situations,” Journal of Business Ethics 51: 6373.
  • Tang, T., and Chen, Y. 2008. “Intelligence vs. wisdom: The love of money, Machiavellianism, and unethical behavior across college major and gender,” Journal of Business Ethics 82: 126.
  • Teixeira, A., and Rocha, M. 2010. “Cheating by economics and business undergraduate students: An exploratory assessment,” Higher Education 59: 663701.
  • Zopiatis, A., and Karmbia-Kapardis, M. 2008. “Ethical behavior of tertiary education students in Cyprus,” Journal of Business Ethics 81(3): 647663.