THE EFFECT OF COMMITMENT TO IMMORAL BEHAVIOR AND THE INFLUENCE OF A DOMAIN SWITCH
Master Thesis
University of Groningen Faculty of Economics and Business MSc Human Resource Management
Supervisor: dr. L.B. Mulder Second supervisor: R.B.L. Sijbom
Maarten Heeres Student number: 1685252
Stadsweg 8 9917PW Wirdum
Mail address: maartenjh@hotmail.com
Date: 21-06-2013
3 ABSTRACT
In this paper is suggested that initial immoral behavior can motivate subsequent immoral behavior if people are committed to their initial immoral behavior. In order to justify their previous acts people continue to behave immorally. It is argued that a domain switch will function as a justification mechanism; it will enable people to change their behavior from immoral to moral, without having to admit that the initial displayed behavior is immoral. A bug killing paradigm was used to test if commitment to bug termination motivates people to kill more bugs and whether a domain switch moderated this. Participants were asked to put bugs in the exterminator at their own pace after putting either one (low commitment) or five (high commitment) bug(s) into the grinder initially. To manipulate the domain switch, the type of bug in the second round was varied (i.e. both slaters and mealworms were used in the experiment). Results show a marginally significant effect of commitment only for slater starting conditions. Analysis points out that exterminating mealworms is considered less unethical compared to exterminating slaters. Consequently, the commitment to the immoral behavior is less strong. Because the behavior is perceived as less immoral, the need to justify the behavior is less strong for mealworms; resulting in less escalation of commitment. No evidence is found for the moderating role of a domain switch. Although a possible explanation for this can be the unsuitable operationalization of the switch (i.e. the difference in perceived immorality between the bugs), further research needs to make clear if a domain switch can reduce or diminish commitment to immoral behavior.
4
INTRODUCTION
Immoral behavior in today’s society is an issue which affects many companies (Anand, Ashforth & Joshi, 2005). Fraud is taking place within many organizations and has a profound effect on the organization as a whole. Immoral behavior can cause major conflicts and may even lead to bankruptcy, which is illustrated painfully by the Enron affair. The American energy company Enron engaged in accounting fraud by illegally enlarging their profits while at the same time concealing billions of dollars of debt (Galit, 2012). Not only in organizations immoral behavior is present, also in society in general immoral behavior plays a hands on role in the lives of many (Anand et al., 2005).
Within this paper, I will focus on the role of commitment in immoral behavior. An analysis is performed to determine if commitment to immoral behavior can facilitate subsequent immoral behavior. Furthermore, this thesis intends to make clear if commitment to immoral behavior can be diminished or reduced by a domain switch. By making people believe that the situation has changed, although in reality there is no significant change, their willingness to continue to behave immorally may be altered as well.
5 THEORY
Unethical behavior is defined by Jones (1991) as ‘any action that is either illegal or morally unacceptable to the larger community.’ Examples of this include violations of the ethical norm or standards, lying, cheating and other forms of dishonesty. Unethical and immoral behavior can be seen as similar concepts (Jones, 1991) and are used interchangeably in this paper. The behavior that a person will show, either moral or immoral, is strongly determined by the tendency that people have to obtain a high moral self-image. A high moral self-image implies that individuals have strong desires to be moral people and will exhibit behavior that aids in realizing the desired moral state. Research by Blasi (1983; 2004) confirms this reasoning. He suggested that people have a strong general desire for a high moral-self. When people engage in immoral behavior they experience stress because their moral self-image is disturbed. Due to the distance between the reality and the desired mental state people will try to adjust their behavior in a way that aids in obtaining a moral self-image (Blasi, 1983; 2004).
When people have engaged in immoral behavior they can subsequently show two possible reactions. On the one hand a person may stop or reduce the immoral behavior and try to compensate his or her immoral actions by acting more morally (Sachdeva, Iliev & Medin, 2009). This compensatory mechanism an individual uses is referred to as ‘moral cleansing’. On the other hand the person may be consistent in displaying the immoral behavior and continue behaving immorally (Sherman, Nave & Funder, 2010). Both possible reactions are explained in more detail.
Compensating Mechanisms
A number of research articles (Sachdeva et al., 2009; Zhong, Ku, Lount & Murnighan, 2010; Jordan, Mullen & Murnighan, 2011) suggest that people may behave more morally as a response to their previous shown immoral behavior. Because behaving immorally has a negative influence on the person’s perceptions of the self-worth, a person will be more willing to engage in moral behavior to regain some of that lost self-worth. The moral behavior displayed after the immoral behavior helps to balance the moral self-worth. Sachdeva et al. (2009) use the term ‘moral cleansing’ to describe the pro social behavior people will engage in when their moral self-worth has been threatened. The way moral cleansing functions is illustrated by Zhong et al. (2010) as well. They find that decision makers act based on their previous decisions. If their previous choices were immoral decision makers act significantly more ethical in their subsequent choice.
6
illustrate this with their finding that if previous choices of decision makers were moral, decision makers act significantly less ethical in subsequent choices.
Consistency in Behavior
Contradicting the compensation theory, a different stream of research suggests that people have the tendency to be consistent in the behavior they show. The need for consistency in behavior is supported by psychological research (Festinger, 1957; Heider, 1958). In psychology, the cognitive dissonance theory (Festinger, 1957) and the balance theory (Heider, 1958) point out that behaving inconsistently is psychologically distressing. The cognitive dissonance theory explains that we have an inner drive to hold all our attitudes and beliefs in harmony and avoid disharmony (or dissonance). When we experience conflicting attitudes, beliefs or behaviors this generates a feeling of discomfort. It leads towards a change in one of the attitudes, beliefs or behaviors in order to reduce the comfort and restore the desired balance. In line with this, there is evidence that people often continue with their immoral actions. To illustrate this, think of the fraudulent affairs of Professor Diederik Stapel, who falsified over 55 papers (Deira, 2012). Another example is the widespread use of unlawfully substances within cycling, which lasted for several decades (Winnen, 2007).
Continuing immoral behavior contributes to the escalation of immoral behavior. Escalation of unethical behavior occurs when an individual increases the extent to which he or she is behaving immorally. In the literature different mechanisms are mentioned that underlie the tendency that immoral behavior has to escalate.
7
rationalization to make sure that the rationalization covers the immoral behavior. Because people are uncertain if their rationalization is enough to cover up the immoral behavior they overstate the rationalization to make sure that their identity remains moral. This rationalization as a result can cover behavior that is even more immoral. For example, immoral behavior related to accounting fraud initially rationalized with ‘we do this because we are saving our great firm’ becomes ‘we do this because we are saving our great economy’. As a consequence, because the rationalization is bigger than the initial immoral behavior it opens the door to further and more extensive corruption (Zyglidopoulos et al., 2009).
Second, immoral behavior can escalate by a mechanism called the slippery slope effect (Gino & Bazerman, 2009). The slippery slope effect implies that the immoral behavior of an individual increases in minor incremental steps. When the ethical erosion is gradual, people are not fully aware of the ethical lapses that they display. Therefore they are more willing to accept the unethical behavior. If the change in unethical behavior is more abrupt, the awareness of the shift in behavior is much larger; and therefore less likely to be accepted. According to Gino & Bazerman (2009) the slippery slope effect is a psychological phenomenon. Behaving more unethically (or accepting more unethical behavior) can occur without an individuals’ awareness or intention.
A third mechanism by which immoral behavior can escalate is escalation of commitment. Escalation of commitment is described by Staw (1976) as the tendency to persist with unsuccessful initiatives or courses of action. People keep on pursuing these unsuccessful initiatives because they want to justify their prior decisions. Research in this area (Staw, 1981; Street, Robertson & Geiger, 1997; Street & Street, 2006) has shown that commitment to undesirable behavior can facilitate further undesirable behavior. According to Staw (1981) an individual faces an escalation situation if ‘costs have been incurred in pursuit of an objective that is unlikely to be accomplished, regardless of any future course of action undertaken’. If individuals continue to invest resources in this failing course of action, they are escalated in their commitment. The ‘sunk cost’ effect is one of the most well-known drivers of escalation of commitment (Staw, 1976). A sunk cost is a resource already expended in pursuit of a given course of action. Although economic theory suggests these sunk costs should not be taken into account, decision makers tend to favor making a decision in which they previously invested. The reason for this is that sunk costs trigger a pressure of self-justification. Decision makers do not want to be seen as people who waste organizational resources (Staw, 1976). Consider for example a project manager who keeps investing resources in his project although the project is highly unlikely to be profitable. To show the righteousness of his actions the manager keeps investing resources although the objective is unlikely to be accomplished. The manager is committed to his actions of spending resources.
8
immoral behavior can motivate more subsequent immoral behavior. Participants in this experiment were asked to kill bugs by dumping slaters in an exterminating machine. The first round half of the participants (the low committed condition) put only one slater in the terminating machine whereas the other half (the high committed condition) put in five bugs. Next, it was measured how many bugs participants chose to exterminate in a second round. The results of this experiment show that high committed participants exterminated a significant higher number of bugs compared to low committed participants. So high committed participants escalated in their immoral behavior. The explanation proposed for this effect is that after killing one bug initially, people show an aversion to or inhibition of subsequent killing. Initially killing one bug leads to a state of cognitive dissonance because the initial killing is perceived as immoral. Because people are motivated to avoid cognitive dissonance (Festinger, 1957; Heider, 1958), during the subsequent self-paced killing people put fewer bugs in the grinder. But when five bugs are killed initially, individuals cross a ‘‘point of no return,’’ such that continued or increased killing becomes a means of justifying the prior killing. Repeated initial killing commits people to this line of behavior. People justify the behavior through increased subsequent killing and escalate in their commitment (Staw & Fox, 1977). Drawing on these theories and research, the first hypothesis is formulated as follows:
Hypothesis 1: Commitment to initial immoral behavior will lead towards an increase in subsequent immoral behavior.
Domain Switch
The experiment by Martens et al. (2007) shows that due to commitment to immoral behavior an individual tends to display immorality in subsequent behavior. The current research also investigates if this commitment can be reduced or broken by the involvement of a domain switch. A domain switch implies a perceived change in situation.
9
Based on this, it may be expected that when there is a domain switch individuals may be more willing to change their behavior. As people aspire to be moral, they will move away from immoral behavior. But this withdrawal is restricted by the commitment to the immoral behavior (Staw & Fox, 1977). A domain switch however can reduce or diminish this commitment. The commitment does not oppose the change in behavior because the commitment is not linked to the new domain. Therefore the individual can change his or her behavior without having to admit that the initial behavior is immoral. Because the two situations are viewed as dissimilar an individual can shift from the immoral behavior and display moral behavior without threatening the moral self. From this argumentation the second hypothesis can be deducted:
Hypothesis 2: Commitment to previous immoral behavior will increase subsequent immoral behavior. However, this is less so the case when subsequent behavior is in a different domain than the previous behavior.
Conceptual Model
The two different hypotheses are visualized in figure 1, the conceptual model.
Subsequent Immoral Behavior Subsequent Immoral Behavior Domain Switch Domain Switch Initial Commitment to Immoral Behavior Initial Commitment to Immoral Behavior
-+
10
METHODOLOGY
Participants and Design
A total of 108 students at the University of Groningen and Hanze University participated to either fulfill a course requirement or to receive a small amount of money (€4). These participants were randomly assigned to the four conditions of a 2 (low commitment vs. high commitment) X 2 (no switch vs. switch) design. The youngest participant was 18 years of age; the oldest participant was 27. The mean age of the participants is 20.86 years (SD = 1.89).
Materials and Procedure
The method of this experiment is based on “the bug killing experiment” by Martens et al. (2007). The experiment was conducted individually. After arrival in the lab each participant was explicitly notified about the nature of the experiment. The participants read and signed a consent form which stated: “You will engage in a bug extermination task, in which you will actually exterminate bugs by throwing them in an extermination machine (without having to touch these bugs yourself)”. Also the goal of the experiment was clarified in the consent form. The objective was stated as: “testing how people in different roles deal with different animals.”
Hereafter, each individual participant was taken to a small room in which the participant was instructed to engage in another unrelated experiment. After this experiment the participant filled in a survey which contained personality questions. After filling in the survey the participant was led towards a different room in which the experiment took place. The experimenter welcomed the participant and led the participant to a table in the front of the room on which a bug exterminator machine was placed. The exterminating machine was a modified coffee grinder, created by attaching a plastic tube and funnel to the side of the grinder (See appendix A).
11
To familiarize the participant with the bugs and the exterminating machine the participant was asked to put either five bugs (high initial commitment) or one bug (low initial commitment) in the exterminator. The participant was explicitly instructed to drop the bugs one by one in the machine. After the bugs were dropped in the machine (either one or five) the experimenter asked the participant to push the “on” button of the machine for three seconds. By pushing the button the bugs were exterminated. After this round of initial killing the experimenter explained that the familiarization round was over and that a second round would follow. In the second round the experimenter placed a similar plate, now containing 20 white cups on the table. For the participants in the condition of no domain switch the same bugs were in the cups. For the participants in the domain switch condition, the cups did no longer consist slaters, but were replaced by mealworms.
The experimenter made clear to the participant that the second round of the experiment was about to start: “We will continue with the second round, the experiment really starts now.” Furthermore, the experimenter specifically addressed the change to the participants in the domain switch condition: “In the second round you will work with a different kind of bug compared to the familiarization round. Where slaters were used in the familiarization round we will work with mealworms in the second round. So the kind of bug you worked with in the familiarization round differs from the kind of bug you are working with in the second round”. In the non-domain switch condition the participant was provided with the following instructions: “In the second round you will work with the same kind of bug you worked with in the familiarization round. Where slaters were used in the familiarization round we will continue to work with slaters in the second round. So the kind of bug you worked with in the familiarization round is the same bug you are working with in the second round”. After these specific directions the following instructions were similar in all the conditions.
12
number of exterminated bugs; which equaled 20 minus the remaining bugs. Finally, after the participant finished the questionnaire, an extensive computerized debriefing followed. The participant was told the bugs were still alive, the true goal of the study was explained, a detailed explanation was given with regard to the variables that were analyzed and the participant was given the opportunity to express concerns about the experiment.
Manipulations
Commitment
Commitment was manipulated in the familiarization round. In the familiarization round participants were instructed to exterminate either one or five bugs. Participants in the low commitment condition were instructed to exterminate one bug; participants in the high commitment condition were instructed to exterminate five bugs.
Domain switch:
The domain switch was manipulated by changing the type of bug that needed to be exterminated during the second round. In the experiment, two kinds of bugs were used; slaters (see appendix B) and mealworms (see appendix C). In the domain switch condition the kind of bug changed after the first round. To control for the effect the different bugs themselves might have on the propensity to exterminate, the kind of bug that was used in the first round (slater or mealworm) was varied. This implied the 2 (low commitment vs. high commitment) X 2 (no switch vs. switch) was expanded with a counter balance measure. Participants were randomly assigned to one of these conditions. The counter balance measure conditions were supposed to be collapsed during the analysis of the hypotheses. I.e. the ‘slater first’ conditions were supposed to be collapsed with the ‘mealworm first’ conditions. However, the results appeared to differ between these counter balance measure conditions. Therefore, both the slater starting conditions and the mealworm starting conditions were analyzed individually as well.
Immoral behavior
Immoral behavior; the dependent variable was represented by the number of bugs participants terminated during the second round. The number of exterminated bugs was noted by counting the empty cups. The counting was performed as soon as the participant left the experiment room.
Manipulation Checks
13
items (α = .72) which were posed as statements on a seven point answering scale (1 = completely disagree, 7 = completely agree). The four statement were formulated as follows: “the bugs used during the practice round were similar to the bugs used during the 12-second round” (recoded), “the type of bug used during the 12-second round differed compared to the type of bug used during the practice round,” “I perceived the task I had to perform during the practice round as different compared to the task I had to perform during the 12-second round” and “I perceived the task I had to perform during the 12-second round as identical to the task I had to perform during the practice round ” (recoded).
Ancillary Measurements
Participants’ view on general characteristics of the two bugs was measured by assessing their opinion on six different aspects of the two bugs; i.e. the extent to which the bugs were considered to be; useful, harmful, creepy, unwanted, appealing, dirty, hygienic and the extent to which it was considered a bug. These questions were posed as statements and had to be answered for both mealworms and slaters individually on a seven point answering scale (1 = completely disagree, 7 = completely agree).
The differences in moral views on extermination of the different bugs were measured by eight items; the extent to which the killing is viewed as; acceptable, ethical, justifiable, morally wrong (recoded), consistent with my norms, unfair (recoded), rightful, cruel (recoded) and necessary. The items were posed as statements and had to be answered for slaters (α = .86) and for mealworms (α = .84) individually. The participants had to express their cognitions and feelings with regard to this statement on a seven point answering scale (1 = completely disagree, 7 = completely agree).
14 RESULTS
Participants
Of the participants, two discontinued the study after reading the consent form and two participants refused to complete the extermination task during the experiment. Furthermore, 12 were excluded from analyses. Out of these 12 participants two participants initially resisted to complete the experiment; they expressed severe feelings of not wanting to complete the experiment. When discussing the signed consent form they made clear they had not read it sufficiently and were unaware that actual bugs had to be killed. Although they did continue with the experiment they were left out. Further, seven participants were excluded because they expressed severe suspicion they were not killing the bugs. They expressed this during the debriefing and at the end of the computerized survey, where a possibility was presented to give remarks on the experiment in general. One participant was left out because he did not speak sufficiently Dutch and one participant was excluded because he had perceived the bugs to be dead (indicated in the computerized survey as a concluding remark). Lastly, one participant was detected as an outlier as she scored more than three times the standard deviation above the mean, and was removed. This left 92 participants (65 male, 27 female).
Manipulation Checks
A 2 (commitment) X 2 (switch) ANOVA on perceived commitment showed that the commitment manipulation was successful. Only a main effect for commitment was found, F (1,88) = 74.43, p < .01. Participants in the high commitment condition (M = 4.28, SD = 1.41) experienced the number of bugs terminated in the initial round as significantly more compared to participants in the low commitment condition (M = 1.83, SD = 1.37). A 2 (commitment) X 2 (switch) ANOVA on experienced change in domain proved that the switch manipulation was successful. Only a main effect for the switch was found, F (1,88) = 44.22, p < .001. Participants in the domain switch condition (M = 3.76, SD = 1.34) experienced a bigger difference in the two situations compared to people in the non-domain switch condition (M = 1.93, SD = 1.27).
Hypothesis Tests
15
interaction was found, F (1,86) = .69, p = .41. In figure 2, the interaction effect between commitment and the domain switch is graphically displayed.
Figure 2, graph of interaction between commitment and a domain switch.
Because the results differed between the counter balance measure conditions, both the ‘first round slater’ and ‘first round mealworm’ conditions were individually subjected to a 2 (commitment) X 2 (switch) ANOVA1. The results for the ‘slater starting conditions’ indicate only a main effect for the switch, F (1,42) = 6.77, p < .05. Participants in the domain switch condition (M = 5.36, SD =1.56) terminated significantly more bugs during the second round compared to the non-domain switch condition (M = 4.38, SD = 1.06). There was a marginally significant main effect of commitment in the ‘slater starting conditions,’ F (1,42) = 3.04, p < .1. Participants in the high commitment condition (M = 5.16, SD = 1.25) exterminated more bugs compared to participants in the low commitment condition (M = 4.48, SD = 1.50). The ‘commitment X switch’ interaction effect turned out to be insignificant, F (1,42) = .24, p = .63. The results for the ‘mealworm starting conditions’ indicate as well only a significant main effect for the switch, F (1,42) = 4.16, p < .05. Now participants in the domain switch condition (M = 4.35, SD =1.04) terminated significantly less bugs during the second round compared to the non-domain switch condition (M = 5.08, SD = 1.38). No significant main effects for commitment and interaction were found.
As an alternative analysis to test the hypotheses a regression was used. Instead of the conditions themselves, participants’ feelings related to the commitment and the switch were used. These questions were the same as the ones used for the manipulation checks. The values were standardized prior to running the analysis. The hypotheses were tested in two regression steps. The
1 Controlling for the type of bug by making use of a covariate was not possible due to the violated required
assumption of homogeneity of variances. 0 1 2 3 4 5 6
Low commitment High commitment
Im m o ral b eh av io r
Interaction Commitment and Domain switch
16
results of this analysis are shown in Table 1. First, immoral behavior was regressed on the perceived commitment and the perceived domain switch. This model was not significant F (2,89) = .59, p = .56. No significant effects were found for the perceived commitment (B = -.07, p = .47) and the felt domain switch (B = -.06, p = .49). Second, the two-way interaction was included. This did not significantly change the model, F (3,88) = .49, p = .69. The interaction effect between perceived commitment and perceived domain switch did not significantly predict the amount of bugs terminated (B = .05, p = .60). The interaction is graphically depicted in Figure 3.
Figure 3, Regression analysis: graph of interaction between perceived commitment and perceived domain switch. -0.001 0.000 0.001 0.001 0.002 0.002 Low perceived commitment High perceived commitment Im m o ra l beha v io r
Interaction of perceived commitment and perceived domain switch
Low perceived switch in domain High perceived switch in domain
Table 1: Regression results for the effect of perceived commitment, perceived domain switch and interaction on immoral behavior.
17
Ancillary findings
In order to explain the different effects of the switch on immoral behavior in slater starting conditions and mealworm starting conditions the views with regard to the two different bugs were analyzed. Reasoning that the change in domain may have altered the behavior not because of the switch itself but because of the type of bug that was being used, participants’ opinion regarding the difference between mealworms and slaters was tested on different attributes.
First participants’ views on different general aspects of the two bugs were analyzed; i.e. the extent to which the bugs were considered to be; useful, harmful, creepy, unwanted, appealing, dirty, hygienic and the extent to which it was considered a ‘bug’. A paired-samples t-test was conducted to compare how people rated these aspects for mealworms and slaters. Analysis showed only a significant difference for harmfulness, t (91) = -2.59, p < .05. (two tailed). Mealworms (M = 3.12, SD = 1.55) were considered significantly more harmful than slaters (M = 2.67, SD = 1.45). Also, a 2 (commitment) X 2 (switch) ANOVA was performed on these different criteria, but no significant effects emerged.
Second, differences in moral views regarding killing the bugs were assessed. A paired-samples t-test was conducted to compare if peoples’ moral views differed between killing mealworms and killing slaters. The constructs of the moral views on slaters and mealworms were subjected to a paired samples t-test, but showed no significant effects, t (91) = -1.45, p = .15 (two tailed). The different items of the constructs (i.e. the extent to which the killing is viewed as; acceptable, ethical, justifiable, morally wrong, consistent with my norms, unfair, rightful, cruel and necessary) were individually subjected to the same paired samples t-test. The results reveal only a significant difference in the perceived level of ethicality of exterminating between the two bugs, t (91) = -2.61, p < .05 (two tailed). Killing mealworms (M = 4.01, SD = 1.61) was considered significantly more ethical justifiable than killing slaters (M = 3.71, SD = 1.63).
18 DISCUSSION
Effect of Commitment on Immoral Behavior
In the theoretical analysis was predicted that commitment to initial immoral behavior will lead towards an increase in the subsequent immoral behavior. Repeated initial killing was hypothesized to commit people to their previously displayed immoral behavior. As a means of justifying their immoral behavior people were suspected to increase in being immoral; leading to escalation of commitment. For the slater conditions, the results show a significant difference between participants who were high and those who were low committed. Participants who were high committed to the immoral behavior exterminated significantly more slaters compared to participants who were low committed. Therefore, for the slater starting conditions the first hypothesis can be confirmed. Commitment to immoral behavior fuels further immoral behavior, leading to escalation of commitment. Although killing is a morally disapproved behavior and individuals are motivated to inhibit the killing and thereby avoid cognitive dissonance, by committing people to this line of behavior (making them kill five slaters), people no longer try to reduce their cognitive dissonance by killing less slaters in the subsequent behavior. They instead try to reduce their cognitive dissonance by justifying the previous behavior by killing even more bugs. Cognitive dissonance reduction by justification is more likely to occur if an individual is committed to the immoral behavior. Because a participant already broke his or her moral standards (five times in the experimental design), reducing the killings would be a way of admitting that the previous behavior was immoral. Although the effect of commitment replicated the findings of Martens et al. (2007) for people who started with exterminating slaters, for people who started with mealworms the effect of commitment was insignificant. Only slightly more bugs were exterminated when people were committed to exterminating mealworms. The difference in the effect of commitment between the two bugs implies participants perceive the bugs differently.
19
well be caused by a difference in harmfulness. Finally, people who exterminated slaters used significantly more moral rationalizations compared to participants exterminating mealworms. According to Zyglidopoulos et al., (2009) a rationalization is a way of justifying the past immoral behavior. The rationalization has to cover the immoral behavior. Therefore the magnitude of the rationalization needs to be of an equal size as the displayed immoral behavior in order to justify it (Zyglidopoulos et al., 2009). In other words, the fact that moral rationalizations were used more when people worked with slaters indicates that the act of exterminating slaters was considered as more immoral.
This difference in morality is supported as well by the fact that the slaters displayed more liveliness than the mealworms did. The vividness of the slaters was larger; implying more movement within the cup. This difference in liveliness may have attributed to the different results in effect of commitment. The liveliness may have influenced the experienced morality of the behavior. Exterminating a bug that displays more signs of being alive can be perceived as much more unethical compared to exterminating a less vivid bug.
These results all support the rational that the experienced immorality of killing a mealworm is not as large as killing a slater. Therefore the commitment to the immoral behavior is less strong. People don’t perceive the behavior as being very immoral in the first place, which would imply people are less committed to the behavior. Because the behavior is perceived as less immoral, the need to justify the behavior is less strong; resulting in less escalation of commitment.
Contradicting Effects of Domain Switch
In the second hypothesis was theorized that commitment to immoral behavior could be diminished by changing the domain of the behavior. Because the commitment is not linked to the new domain individuals could change their behavior without threatening their moral self. The results show that people experienced the task of exterminating mealworms differently than killing slaters. More specifically, when mealworms had to be exterminated first and subsequently slaters; the amount of immoral behavior significantly reduced compared to people who continued to exterminate slaters. The opposite turned out to be the case for people who started with exterminating slaters and in the second round turned to mealworms; their level of immoral behavior significantly increased. The results show these increases/reductions in immoral behavior were not caused by the effect of the switch (i.e. the possibility to change the behavior without threatening the moral self) but were caused by the type of bug that was used. Therefore the second hypothesis cannot be confirmed. No evidence was found that a domain switch can moderate the effect of commitment on immoral behavior.
20
bugs. Instead of only switching the domain of the immoral behavior; the level of immorality associated with the killing itself was switched.
Another possibility may be that the effect of a domain switch does not exist. A possible explanation for this, deducted from the results of the experiment, is that the psychological difficulty of the killing may be central to the commitment people feel. People may experience the psychological difficulty of the killing as the most elementary criterion of their task. This psychological difficulty is the component to which the commitment is linked. The commitment is not linked to the domain in general but specifically to the experienced psychological difficulty of killing. If the psychological difficulty of killing is equal, people cannot change their behavior from immoral to moral without having to admit that the initial killing was unethical, even if the domains are switched. Consequently, they remain committed to the immoral behavior.
Limitations & Recommendations
As elaborated on in the discussion the primary limitation of the study is the inappropriateness of the mealworms as a mechanism to facilitate the domain switch. Exterminating mealworms is not perceived to be as immoral as exterminating slaters. To overcome this problem, a different bug should be used to function as a domain switch. In order to appropriately select the right bug an independent pre-analysis should be undertaken in which the differences can be clearly assessed. In this analysis, such a pre-analysis was undertaken. However, it failed to take into account the actual vividness of the bug which may be of great influence in determining the psychological difficulty of exterminating the bug. Instead of using pictures to prescribe the bugs, clearly alive bugs should be used to assess whether the psychological difficulty of killing the bugs is perceived as equal. A qualitative pre-analysis can help to determine which specific criteria determine the psychological easiness of killing the bug.
21
current experiment a hairdryer was not used because the heat did not affect the movement for mealworms as much as it did for the slaters.
Implications & Conclusions
The findings in this paper have several implications for organizations. First, recent corporate scandals highlight the importance of analyzing immoral behavior and the potency it has to escalate. Think for example of the recent horsemeat scandal, where large quantities of food advertised as containing beef were found to contain undeclared horse meat (Harris, 2013). In order to stop these immoral actions it is important to realize what causes people to continue and escalate in their immoral actions. This paper helps to understand the mechanisms that underlie the tendency that employees may have to continue and escalate in their immoral behavior. This thesis showed that commitment to immoral behavior is an important mechanism which facilitates further immoral behavior by justifying previous immoral actions. Organizations may actively use this knowledge by implementing it in ethics training. Current ethics training, mainly focused on the normative education, may fail to recognize that people are committed to their immoral behavior and therefore not abide by the rules. Second, this thesis could not find evidence for the moderating role of a domain switch. Future research is needed to correctly assess the influence of the effect of a domain switch on the level of commitment to immoral behavior. It may well be that people who are committed to their immoral behavior are more willing to change their behavior if they don’t have to admit that the initial behavior was wrong. A successful way of implementing this within organizations could be by making more use of job circulation. If people appear to be committed to their immoral behavior job circulation may offer them a possibility to change their behavior to moral without threatening their moral self.
22 REFERENCES
Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt.
Anand, V., Ashforth, B. E., & Joshi, M. (2005). Business as usual: The acceptance and perpetuation of corruption in organizations. Academy of Management Executive, 19, 9-23.
Bandura, A. (1999). Moral disengagement in the perpetuation of inhumanities. Personality and Social Psychology Review, 3, 193–209.
Blasi, A. (1983). Moral development and moral action: A theoretical perspective. Developmental Review, 3, 178–210.
Blasi, A. (2004). Moral functioning: Moral understanding and personality. In D. K. Lapsley & D. Narvaez (Eds.), Moral development, self and identity: Essays in honor of Augusto Blasi. Mahwah, NJ: Lawrence Erlbaum & Associates.
Deira, S. (2012, November 28). Diederik Stapel kon frauderen dankzij gebrek aan kritiek. Elsevier: meer dan een weekblad. Retrieved June 16, 2013, from http://www.elsevier.nl/ Wetenschap/nieuws/2012/11/Diederik-Stapel-kon-frauderen-dankzij-gebrek-aan-kritiek-ELSEVIER355752W/.
Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press.
Furr, R.M., & Funder, D.C. (2004). Situational similarity and behavioral consistency: Subjective, objective, variable-centered and person-centered approaches. Journal of Research in Personality, 38 421–447.
Galit, A. (2012). The Discursive Management of Financial Risk Scandals: The Case of Wall Street Journal Commentaries on LTCM and Enron. Qualitative Sociology, 35(3), 251-270.
Gino, F., & Bazerman, M. H. (2009). When misconduct goes unnoticed: The acceptability of gradual erosion in others' unethical behavior. Journal of Experimental Social Psychology, 45, 708-719.
23
Heider, F. (1958). The psychology of interpersonal relations. Hoboken, NJ US: John Wiley & Sons.
Jones, T. M. (1991). Ethical decision making by individuals in organizations: An issue contingent model. Academy of Management Review, 16, 366–395.
Jordan, J., Mullen, E., & Murnighan, J. K. (2011). Striving for the moral self: The effects of recalling past moral actions on future moral behavior. Personality and Social Psychology Bulletin, 37(5), 701-713.
Marlin, B. (2009). Common Woodlouse, Sow Bug, Pillbug. North American Insects and Spiders. Retrieved February 10, 2013, from http://www.cirrusimage.com/isopoda_sow_bug.htm.
Martens, A., Kosloff, S., Greenberg, J., Landau, M. J., & Schmader, T. (2007). Killing begets killing: Evidence from a bug-killing paradigm that initial killing fuels subsequent killing. Personality and Social Psychology Bulletin, 33(9), 1251-1264.
Messick, D. M., & Bazerman, M. H. (1996). Ethical leadership and the psychology of decision making. Sloan Manage. Rev. 37(2), 9–22.
Monin, B., & Miller, D. T. (2001). Moral credentials and the expression of prejudice. Journal of Personality and Social Psychology, 81, 33-43.
Sachdeva, S., Iliev, R., & Medin, D. L. (2009). Sinning saints and saintly sinners: The paradox of moral self-regulation. Psychological Science, 20(4), 523-528.
Sherman, R. A., Nave, C. S., & Funder, D.C. (2010). Situational Similarity and Personality Predict Behavioral Consistency. Journal of Personality and Social Psychology, 99(2), 330–343.
Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior & Human Performance, 16(1), 27-44.
Staw, B. M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577-587.
24
Street, M., Robertson, C., & Geiger, S. (1997). Ethical Decision Making: The Effects of Escalating Commitment. Journal of Business Ethics, 16(11), 1153-1161.
Street, M., & Street, V. (2006). The Effects of Escalating Commitment on Ethical Decision-Making. Journal of Business Ethics, 64(4), 343-355.
Winnen, P. (2007, December 24). Ooit werd doping de wielrenner vergeven. Trouw: de verdieping. Retrieved June 16, 2013, from http://www.trouw.nl/tr/nl/4324/Nieuws/archief/article/detail/ 1665865/2007/12/24/Ooit-werd-doping-de-wielrenner-vergeven.dhtml.
Zhong, C., Ku, G., Lount, R. B., & Murnighan, J. K. (2010). Compensatory ethics. Journal of Business Ethics, 92(3), 323-339.
25 APPENDICES
Appendix A
26 Appendix B
27 Appendix C
28 Appendix D
Moral rationalization statements (1-20)
.. By exterminating bugs I contribute to science.
.. The higher scientific purpose of this experiment justifies the extermination of bugs. .. The world would be a better place without bugs.
.. The bugs I exterminated were harmful. Therefore it was a good thing to kill them. .. The term ‘bugs’ is a suitable term to prescribe the little animals I killed.
.. Extermination is a suitable term to prescribe the killing of the bugs. .. The bugs I killed are only little bugs that do not matter much. .. The bugs I exterminated hardly feel any pain.
.. The way in which I killed the bugs was relatively painless. .. The experimenter forced me to exterminate the bugs.
.. In only carried out the tasks that were assigned by the experimenter.
.. The responsibility for killing the bugs lies more with the experimenter than with me. .. Killing bugs is something everybody does.
.. If I wouldn’t have killed the bugs somebody else would have.
.. I am not responsible for the death of the bugs because a lot of people are participating in this experiment.
.. Killing a few bugs is not a big deal compared to more severe things, like people who abuse horses or dogs.
.. Killing a few bugs is not a big deal compared to more severe things, like large scale accounting fraud.
