How Individual and Situation Factors In?uence the Use Of Neutralization Strategies

Description
How Individual and Situation Factors Influence the Use Of Neutralization Strategies: Lying In Negotiations?, In chemistry, neutralization is a chemical reaction in which an acid and a base react to form a salt. Water is frequently, but not necessarily, produced as well. Neutralizations with Arrhenius acids and bases always produce water where acid–alkali reactions produce water and a metal salt.

How Individual and Situational Factors In?uence the Use of Neutralization Strategies: Lying In Negotiations?

Summary

Lying in negotiations can cause negative emotions, so participants may use neutralization stra- tegies to reduce these feelings. We conducted a 2 (ethical versus non-ethical climate) Â 2 (low versus high negative consequences) experiment to examine how individual and situational factors affect the use of three such strategies: minimizing the lie, denigration of the target, and denial. Lying, psychological distress, and self-perceived moral attributes were measured as nonmanipulated independent variables. One hundred and ninety-two MBA students parti- cipated in a business negotiation in which they were provided with incentives to lie. As pre- dicted, higher distress was associated with greater denial of lies. In addition, climate and consequences interacted to affect minimization and liars engaged in less minimization than did participants who merely concealed information. Climate and moral attributes interacted to affect denigration. We believe these findings support further study of neutralization strategies in the workplace.

Introduction
In the wake of highly publicized allegations of lying and fraud committed by executives at companies like Enron, Tyco, and Arthur Anderson, it seems reasonable to be increasingly concerned about the trustworthiness of business leaders. It may be unfair, though, to single out business leaders as being untrustworthy because people from all walks of life tell lies (Kashy & DePaulo, 1996). It is true that company presidents do sometimes lie in reports to shareholders (Abrahamson & Park, 1994), but nurses also lie in documentations to physicians (Grover, 1993), police officers '?uff' evidence in crime reports (Barker & Carter, 1990), and most individuals would probably admit to occasionally telling lies to avoid hurting someone's feelings. Although lying appears to be pervasive, little research has exam- ined the dynamics of lying in organizations (Grover, 1997). Although a few studies have explored workplace antecedents of lying (e.g., Grover & Hui, 1994; Tenbrunsel, 1998) and its consequences

662

K. AQUINO AND T. E. BECKER

for personal relationships (Shapiro, 1991; Shapiro & Bies, 1994), less common are studies that exam- ine how on-the-job lies affect the person who tells them. For many people lying produces psychic distress—negative emotions such as shame and guilt—because telling lies threatens a person's self-perception as being moral and trustworthy (Festinger &Carlsmith, 1959; Taylor & Brown, 1988). Moreover, most people find it unpleasant to tell lies because lying is generally viewed as unethical (Anton, 1990; Grover, 1993). So when people lie or engage in other kinds of behaviors that deviate from normative standards of right and wrong, they often employ a variety of neutralization strategies to legitimize their actions and reduce psychic distress (Robinson &Kraatz, 1998; Sykes & Matza, 1957). The purpose of this study is to examine a subset of these stra- tegies in work-related settings. We investigate these strategies in the specific context of business nego- tiations because negotiations have been described as breeding grounds for lying and deception (Lewicki, 1983; Tenbrunsel, 1998). Therefore, negotiations are likely to motivate at least some people to use neutralization strategies as a way of relieving psychic distress when they have lied to their opponent.

Theoretical Background
The Oxford English Dictionary (1993) defines a lie as 'a false statement made with the intent to deceive,' a meaning shared by social psychologists (DePaulo & DePaulo, 1989; Ekman, 1985). Essen- tial to this definition are that the individual knows the information is false, wants to mislead another, and behaves proactively to do so (Bok, 1978). Lying falls within the broader category of deception. Other than lying, people can deceive by omitting facts, presenting information in a certain order, or framing statements to guide the listener away from the truth (Grover, 1997). Ekman (1985) refers to these methods as concealment. Concealment differs from lying because it does not involve active attempts to give false information and so it could be that the former is easier for most people to carry out than the latter. Some writers have argued that lying and other forms of deception are legitimate strategies for advan- cing one's interests during business transactions like negotiations (Beckman, 1977; Carr, 1968). Other writers assert that lying in particular is unethical because it threatens the integrity of the individual and the reputation of the firm (Becker, 1998; Haring, 1979; Locke & Woiceshyn, 1995). There is evidence suggesting that many people believe that lying in business contexts is unethical even in situations where some opportunism and deception are expected. Anton (1990) examined perceptions of negotia- tion behavior in a sample of MBA students, full-time managers, and clergymen. Not a single respon-dent in Anton's study believed making false statements during negotiation was ethical. In another study, Bies and Moag (1986) found that candidates who were lied to during a job interview or salary negotiation reported being angry at the guilty organization. Moreover, the honesty of company repre- sentatives was reported as being the most important standard for evaluating the fairness of recruitment. In sum, because many people seem to view lying in negotiations as an unethical and normatively devi- ant act, it is reasonable to assume that negotiators who lie sometimes employ cognitive strategies to rationalize their lies. Sykes and Matza (1957) introduced the concept of neutralization strategies to describe five common rationalizations people use to justify deviant behavior. These include (1) denial of responsibility, whereby one excuses his or her actions by claiming that they were accidental or uncontrollable; (2) denial of injury, whereby one downplays the harm resulting from an unethical act; (3) condemning the condemners, whereby one shifts the focus of his or her own actions to others by saying that other

NEUTRALIZATION STRATEGIES

663

people behave the same way; (4) appeal to higher loyalties, whereby one accepts responsibility for the behavior, but claims that it was done for a higher cause; and (5) denial of the victim, whereby the per- son justifies his or her behavior by assigning negative characteristics to the victim. Building on the work of Sykes and Matza (1957), Robinson and Kraatz (1998) proposed a typology of neutralization strategies used by people who engage in deviant behavior in organizations. They theorized that the most significant dimension along which neutralization strategies vary is the extent to which people openly defy or outwardly comply with existing organizational norms and values. The neutralization strategies falling along the defiant—compliant continuum include (1) the rejection of organizational norms whereby one reduces the dissonance between his or her acts and the norms of the organization by challenging the legitimacy of these norms, (2) the redefining of organizational norms where one modifies the interpretation of norms so that his or her behavior falls within acceptable bounds, (3) de?ecting comparisons where one attempts to rationalize incongruent acts based on their instrumental merits, (4) redefining of behavior whereby one modifies the interpretation of his or her actions to reduce the size of the discrepancy between behavior and the norm that was violated, and (5) changing perceptions of behavior by deceiving self or others by not disclosing the deviant act. The main difference among these strategies is that those on the compliant end attempt to shift the perception of the actor's behavior whereas those on the defiant end attempt to challenge the normative order of the organization. Robinson and Kraatz (1998) suggest that different strategies are likely to be used in different organizational settings and by different actors. This is because some strategies are more easily exe- cuted than others depending on individual (e.g., cognitive ability, motivation) and situational (e.g., clarity of organizational norms) factors. We do not assume that people will employ all of the neu- tralization strategies described above when they lie. Rather, they are likely to prefer the strategy that requires the least amount of cognitive and behavioral effort or that is least risky given their social environment. For this reason we do not make specific predictions about which neutralization strategy will be used following a lie and instead argue that at least one and perhaps more than one strategy will be used. Both Sykes and Matza (1957) and Robinson and Kraatz (1998) suggest that the primary function of neutralization strategies is to legitimize deviant behavior and reduce intra-psychic distress. The mostlikely source of distress is the violation of internalized standards that typically results from deviant behavior. Based on cognitive dissonance theory (Festinger, 1957), we expect that when people act contrary to how they believe they should behave, they experience negative emotions caused by the inconsistency between their standards and their conduct. This, in turn, motivates the desire to regain the sense of self-consistency. One way to achieve this goal is to execute one or more neutralization strategies. For example, by denying a lie to others, oneself, or both, one might avoid or escape self- condemnation and unpleasant feelings associated with performing a socially undesirable act. Based on this reasoning, we hypothesize the following: Hypothesis 1: Greater levels of psychic distress lead to the use of one or more neutralization strategies. It is difficult to examine all possible neutralization strategies in one study, so we examine a subset identified by both Sykes and Matza (1957) and Robinson and Kraatz (1998) that we believe may be prevalent in negotiations: (1) minimizing the consequences of the lie (denial of injury); (2) denigrating the target of the lie (denial of the victim); and (3) denying that the lie occurred (changing perceptions of behavior). Two points deserve mention here. First, in the context of negotiation, we considered the possibility that one way to minimize the negative consequences of a lie is for the liar to make larger concessions to the other party. This seems a plausible strategy because such a course of action can

664

K. AQUINO AND T. E. BECKER

potentially assuage some of their guilt and avoid self-condemnation. Second, the first two strategies described (minimization and denigration) are employed when a person publicly or privately admits to having lied. In contrast, to pursue the last strategy (denial) the person distorts the lie such that he or she interprets it as a 'minor' deviation from truth, or perhaps even as being objectively true (Grover, 1993). Unlike minimizing consequences and denigration, which recognize the lie for what it is, denial can lead to a fundamental change in one's view of reality such that the individual refuses to acknowledge having told a lie in the first place. We have argued that because lying is perceived by many people as a deviant and unethical act, tell- ing a lie can threaten a person's self-image as being moral, competent, and good (Steele, 1988). This can motivate him or her to execute one or more neutralization strategies to maintain a positive self- image (Kunda, 1990; Robinson & Kraatz, 1998). In addition, as we propose below, not all people are equally motivated to execute neutralization strategies after lying. This is because the strength of this motivation likely depends on how positively they evaluate their moral attributes and the context in which the lie occurs.

Positive self-evaluations and neutralization strategies
The maintenance of a positive self-concept is among the strongest and most persistent human goals (Hales, 1985). One of the ways people attain this goal is by trying to live up to their internal standards of right and wrong in the absence of external coercion (Baumeister, 1998; Cialdini, Finch, & DeNicholas, 1990; Hales, 1985). When a person fails to do so, it can threaten his or her self-esteem (Baumeister, 1998). This argument is consistent with theoretical conceptions of how people's moral identity—the mental representation they hold about their moral character—carries with it certain self- regulatory demands that motivate a desire for self-consistency (Aquino & Reed, 2002; Blasi, 1984). When these demands are ignored or violated, it can bring self-condemnation and cause psychic distress because the person has acted in a way that contradicts his or her moral identity. Neutralization strate- gies may allow people to avoid such outcomes by sustaining the experience of the self as being moral, stable, and capable of choice and control (Steele, 1988). We also suggest, though, that the motivation to execute neutralization strategies becomes stronger the more favorably a person evaluates him- or herself with respect to morally relevant attributes (e.g., being honest, fair, or benevolent). This will occur because more positive self-evaluations are likely to place stronger demands on people to maintain consistency between their moral identity and their actions in the world. In contrast, for people who do not view themselves as highly moral, violations of normative standards are less likely to threaten their self-concepts, so the need to execute neutralization strategies is weaker. Based on this reasoning, we hypothesize the following: Hypothesis 2: When negotiators lie, there is a positive relationship between the extent to which they evaluate themselves favorably with regard to their moral attributes and the use of one or more neutralization strategies. In addition to individual attributes such as one's self-evaluation, the use of neutralization strategies may be in?uenced by situational factors (Bersoff, 1999; Robinson & Kraatz, 1998). Thus, people who are highly motivated to execute neutralization strategies because they evaluate their moral attri- butes favorably are more likely to do so under certain conditions. We examined this possibility by experimentally manipulating two factors that might in?uence the relationship between self-perceived moral attributes and the use of neutralization strategies following a lie. The first was ethical climate perceptions.

NEUTRALIZATION STRATEGIES

665

The effects of an ethical climate
Victor and Cullen (1988, p. 101) defined ethical climate as 'the prevailing perceptions of typical organizational practices that have ethical content' or 'those aspects of work climate that determines what constitutes ethical behavior at work.' The concept of an ethical climate can be distinguished from the related concept of ethical culture because the former characterizes organizations 'in terms of broad normative characteristics that tell people what kind of organization this is—essentially what the orga- nization values' whereas the latter 'characterizes the organization in terms of formal and informal con- trol systems (e.g., rules, reward systems, norms) that are aimed more specifically at in?uencing behavior' (Trevino, Butterfield, & McCabe, 1998, p. 453). In our study, we focus on climate percep- ˜ tions because we are interested in how the normative characteristics of the organization might affect the use of neutralization strategies. Following Victor and Cullen (1988), we propose that ethical climate perceptions can in?uence indi- vidual attitudes and behaviors by providing information about the organization and guidance regarding appropriate conduct. We extend their argument by suggesting that these perceptions can also in?uence the motivation to execute neutralization strategies following a lie and that this motivation will be stron- ger among people who favorably evaluate their moral attributes. Our rationale for this prediction is that a person who views him- or herself as highly moral, but who nevertheless lies in the context of an ethical climate, will experience an especially strong threat to his or her self-concept because their behavior contradicts both the demands of their environment and their own personal standards. In con- trast, the same person might be somewhat less threatened if telling lies is not explicitly condemned or proscribed by the organization's ethical climate. This can occur because the behavior does not directly contradict the demands of the environment and, hence, it is less clear whether a lie is considered devi- ant in that particular organization. This is not to imply that the lie would be acceptable in this situation, but rather that it may not be inconsistent with the organization's normative standards. In other words, we expect the presence of an ethical climate that discourages lying and other forms of deception to moderate the relationship between self-perceived moral attributes and the use of neutralization strategies, as stated in the following hypothesis: Hypothesis 3: When negotiators lie, the positive relationship between their self-perceived moral attributes and the use of one or more neutralization strategies is stronger when the organization has an ethical climate than when it does not. A second variable that prior theory and research suggests might also in?uence the use of neutralization strategies among persons who view themselves favorably is the perceived consequences of the lie.

The moral consequences of lying
Our prediction regarding perceived consequences is based on Jones's (1991) assertion that ethical issues can be represented in terms of moral intensity. One element of moral intensity is the magnitude of consequences as perceived by the moral agent. Jones defines the magnitude of consequences as 'the sum of the harms (or benefits) done to the victim (or beneficiaries) of the moral act in question' (Jones, 1991, p. 374). This concept can be applied to negotiations because all agreements have consequences for both parties, and some agreements can produce very negative consequences for one party. There is evidence that people who perceive that serious, negative consequences will result from a moral act are more likely to act ethically (York, 1989). It follows that when a person lies and perceives that the lie can have negative consequences for the other party, the person who lies will be more motivated to employ neutralization strategies. This is because behaving in a way that causes more rather than less

666

K. AQUINO AND T. E. BECKER

harm to another is likely to be more threatening to a person's self-concept. Furthermore, extending the argument regarding the interaction between self-perceived moral attributes and ethical climate, moral consequences may moderate the relationship between these self-perceptions and neutralization strate- gies. Specifically, the motivation to execute neutralization strategies should be stronger when the per- son who lies perceives the moral consequences to the other party to be more severe. This should be especially true of people with positive perceptions of their moral attributes. Hypothesis 4: When negotiators lie, the positive relationship between their self-perceived moral attributes and the use of one or more neutralization strategies is stronger when the moral consequences to the other party are more rather than less severe.

Organizational Context

The MBA Population and their Academic Surroundings At the time of the study, approximately 2000 graduate students attended the University of Delaware, and about 500 of these were MBAs. MBA students spend most of their time in the Alfred Lerner College of Business and Economics, a modern and well-equipped building re?ecting the economic well-being consistent with the lifestyles of most students. Many MBAs live and work in the same town in which the university is located (Newark) or surrounding areas (e.g., Wilmington), although some travel from further away (e.g., Maryland, Pennsylvania). A substantial number of MBAs have their tuition paid for by their employers. The large majority of MBAs are white, and Asians are the second most common ethnic group. There are very few African Americans. The Lab The lab in which the study was conducted was built in 1998 with funds provided by a large pharmaceutical corporation. The four negotiation rooms consisted of a round table and two black chairs, and a miniature camera with audio recorder was mounted on a small shelf in a corner of each room. The rooms were recently painted and carpeted. The video and audio feeds led to a control room where researchers could view each negotiation and where the interactions were recorded via four VCRs. The Broader Environment During the time the study was conducted, considerable media attention was being directed at alleged management misdeeds and political and corporate scandals.

Method
Sample and procedures
Participants The participants were 192 MBA students enrolled in five management classes who participated as part of a class exercise on business negotiations. The average age of participants was 28.9 years and their average number of years of work experience was 6.7. Seventy-nine per cent of them were working

NEUTRALIZATION STRATEGIES

667

either full or part time while pursuing their degrees, and 62.2 per cent were male. The study was a 2 Â 2 factorial design with two levels of ethical climate (ethical versus non-ethical) and two levels of moral consequences (high versus low). Participants were put into 96 dyads and randomly assigned to condi- tions and to the negotiation roles of an agent or supplier. Owing to missing data, including those caused by impasses and a technical recording problem, data from 15 dyads could not be used.

Procedure
The study was conducted in a laboratory of a mid-Atlantic American business school. Approximately 2 weeks prior to the study, the participants' instructor gave them a measure of self-perceived moral attributes to complete at home. This was accompanied by instructions, including a request to bring the completed questionnaire to the lab on an assigned date. Upon entering the lab, participants turned in their completed moral attributes measure. They were then given background information common to all roles and confidential instructions unique to each. The task required subjects to take the role of either an agent for a large car and truck manufacturer ('General') or the supplier of a small firm spe- cializing in the stamping of metal parts ('Midwest'). Information presented to participants described how the agent and supplier were in the final stage of negotiating a multi-year agreement requiring the supplier's firm to produce aluminum hoods for the manufacturer. The remaining issue was the price of the hood. Because subjects negotiated over a single issue where one party's gain was the other's loss, the task was purely distributive. Participants were informed that their interaction would be videotaped. The instructions to agents provided them with information meant to lead them to consider whether to lie. Specifically, these instructions told agents that improvements in design would make the alumi- num hood obsolete in 3 years. Further, the agents' company planned to produce the new design in- house. Agents were also told that their company could unilaterally cancel contracts with its suppliers at any time and intended to do so in 3 years. Finally, they were told that the supplier did not know this, thus placing agents in a position where they could use the information strategically to elicit conces- sions from the other side. For this study, it was important to create a situation in which agents were likely to lie. However, for ethical reasons, participants were not told to lie. Instead, features were built into the exercise to tempt agents to lie while still giving them the freedom to refrain. First, agents' instructions introduced infor- mation designed to enhance the competitive nature of the exercise. They were told that due to down- sizing several agents would be terminated soon, and the most likely candidates were those unable to negotiate favorable contracts. They were further informed that their own past performance had not been exemplary, and that they would improve their chances of being retained if they could secure a favorable price for the hood. The probabilities of being terminated were described as a function of the negotiated price. This provided an incentive to reach a more favorable agreement at the expense of the other party. To discourage impasses, the probability of being fired was very high in the event of an impasse. Second, because incentives can encourage unethical behavior (Hegarty & Sims, 1978; Tenbrunsel, 1998), $50 was offered to the person in each role who reached the agreement most favorable to their company and, in the case of the agent, was not fired. The expectation was that this would increase the motivation to lie in order to reach a personally favorable agreement. In the event of a tie a random drawing would determine who received the prize. In addition, as an incentive and reward for participa- tion, two random $50 prizes were given based simply on completion of the study. Suppliers' information explained that they were anxious to negotiate a contract with the agent's company because they wanted to become a major supplier for the auto industry. Securing a favorable

668

K. AQUINO AND T. E. BECKER

quality rating from a large manufacturer would be an important first step for establishing their reputa- tion. Schedules showing the profits related to different price levels were included in instructions for each role. The agents' schedule showed that any price below $30 per hood would cause the supplier to lose money after 3 years. Suppliers were directly instructed to ask agents how long they expected to be under contract to produce the hood. They were told to ask this question prior to negotiating price. This was meant to ensure that the duration of the contract would be discussed, thus providing the opportu- nity for the agent to lie. Prior to negotiation, all participants completed a questionnaire assessing their understanding of the task and the effectiveness of the experimental manipulations. Then the experimenter reviewed the answers for task-related questions only and corrected any errors. After the experimenter answered clar- ification questions, participants were given 30 minutes to negotiate. They were told to reach agreement only on the unit price of the aluminum hood using one of the price levels shown in their instructions. When participants reached agreement, they completed a post-questionnaire. If they failed to agree in 30 minutes, they were told that the negotiation had reached an impasse. At the end of the experiment, we debriefed participants by supplying them with an information sheet explaining the purpose, design, and implications of the study. We answered any questions they had and gave them our contact infor- mation in case they had comments or concerns later. Finally, at the end of each semester during which the study was conducted, we awarded the four $50 prizes: one to each of the best performing persons in each role (agent and supplier), and two to the winners of the two random drawings for completing the study.

Experimental manipulations
Ethical climate Information about ethical climate in agents' organizations was varied by describing the organizational practices that either encouraged ethical behavior or offered no guidance as to how to behave. In the 'ethical climate condition,' agents were told: Although General expects its agents to negotiate the most favorable terms on all contracts, the company prides itself on being fair and honest in its business dealings. Consequently, it goes against General's corporate culture to withhold facts or 'stretch' the truth with a supplier simply to get them to produce a unique or difficult part at a low price. In fact, you know of several agents who have been ostracized by peers for not negotiating in good faith. Agents in the 'non-ethical climate' condition were given no information about the normative practices of the organization. Consequently, they were in?uenced solely by the reward contingencies that linked performance to settlement price. Moral consequences The negative consequences to the suppliers varied as a function of the final settlement price. The manipulation was such that settlements benefiting the agent were inversely related to the outcomes of the supplier. In the high negative consequences condition, agents were told: Your investigations (of Midwest) revealed that due to exceptionally unfavorable economic conditions in the industry that purchases most of Midwest's products, their profits have been steadily decreasing. Although this has not affected the quality of their products, Midwest's limited financial capabilities make it very unlikely that they will be able to stay in business should they experience any significant financial loss within the next few years.

NEUTRALIZATION STRATEGIES

669

Agents in the low negative consequences condition were told: Your investigations (of Midwest) revealed that due to exceptionally favorable economic conditions in the industry that purchases most of Midwest's products, their profits have been steadily increas- ing. Midwest's current financial stability makes it very likely that they will be able to stay in busi- ness even if they should experience significant financial losses within the next few years.

Measures
Lying Lying was assessed objectively by videotaping the negotiation and having two trained assistants rate whether agents lied. Prior to the negotiation, participants playing the role of supplier were instructed to ask the agent from General how long General intended to keep Midwest under contract. The suppliers were told to ask this question at the beginning of the negotiation, prior to discussing issues of price. During training, the raters (MBA students) were taught how to distinguish three categories of responses: (1) telling the truth (e.g., 'We will probably produce the hood in-house in 3 years'), (2) concealing (e.g., 'I can't really tell you at this point'), and (3) lying (e.g., 'I assure you that we'll be buying the hood from you for 9 years'). As part of the training, the assistants watched three video- tapes from the study, independently coded them, and then discussed their ratings with the second author. Over the next week, the assistants independently rated seven more tapes and then met again to discuss the results with the second author. This meeting focused on making sure that both assistants applied the same standards in evaluating lying. After rating the remaining dyads, the raters agreed in 76.7 per cent of the cases. In our analyses we only included cases for which the raters agreed. We used data on non-liars to help evaluate the effect of lying on the use of neutralization strategies. Specifically, we examined whether the use of minimization and denigration are in?uenced by the dif- ference between outright lies and what ethicists refer to as concealment (Bok, 1982). As noted in the introduction, concealment involves withholding information or presenting information in a way that guides the listener away from the truth (Grover, 1997). Concealment differs from lying in that conceal- ment does not involve an active attempt to give false information, and the extent of the deception is typically less (Ekman, 1985). This distinction is important because if the rationale for neutralization strategies is valid, people who conceal may experience less psychic distress than those who lie. As a result, concealers should be less motivated than those who lie to execute neutralization strategies. Therefore, a comparison of concealers versus liars provides a useful reference point for evaluating the latter's responses to their lies. In addition, we used the distinction between concealing and lying to remove concealers from our analysis of denial. This is because concealers, not having lied, do not have a reason to deny lying. Psychic distress Distress was evaluated via three items on the post-negotiation questionnaire. Specifically, agents were asked how guilty, nervous, and troubled they felt regarding the negotiation. They responded on a scale from 1 ¼ not at all to 5 ¼ a lot, and scores on the items were averaged. Cronbach's alpha for the scale was 0.78. Agents responded to these items prior to being given the opportunity to engage in denial and denigration. Self-perceived moral attributes Perceptions of moral attributes were measured on a pre-negotiation survey using 11 items from Becker's (2000) morality index. These items require respondents to use a 5-point scale (1 ¼ completely false of me to 5 ¼ completely true of me) to assess their own level of benevolence

670

K. AQUINO AND T. E. BECKER

(e.g., 'I show genuine concern for other people'), fairness (e.g., 'I treat other people fairly'), honesty (e.g., 'I am extremely honest'), pride (e.g., 'I demonstrate pride in my work'), productivity (e.g., 'I am very hard working'), and rationality (e.g., 'I use reasoned judgment in coming to decisions'). Items were averaged to form a single scale (ff ¼ 0.71). Neutralization strategies Denial was measured via a post-negotiation questionnaire by asking agents: 'Did you intentionally lie to the other party about how long they would be under contract to produce the aluminum hood?' Participants responded on a 2-point scale (1 ¼ yes, 2 ¼ no). Because there was a method of independently assessing lying, those who lied but responded 'no' to the above question were classified as using the denial strategy. To assess minimization of consequences, the final settlement price reached by nonimpasse dyads was analyzed. Possible prices were $16.67, $20, $25, $30, $32, and $35 per hood. The goal was to evaluate whether agents who admitted lying compensated the other party by foregoing potential gains more than did agents who concealed. Finally, denigration was assessed via five items included in the post-negotiation questionnaire. Agents used a 5-point scale (from 'not at all' to 'a lot') to describe how sincere, honest, and fair their negotiating partner was. We reverse-scored and averaged the items as our measure of denigration. The alpha for this scale was 0.83. Control variables Owing to the nature of our design and the order in which the denigration and denial measures appeared in our negotiation post-questionnaire, participants had the opportunity to execute some neutralization strategies before others. The first opportunity to execute a strategy was to agree to a higher price, which we used as our behavioral indicator of minimization; the second was to denigrate the opponent by evaluating them more negatively; and the third was to deny that a lie occurred. It may be that once a specific strategy is executed a person will feel less need to execute another strategy. We accounted for this possibility by controlling for any neutralization strategy that might have been executed before the specific strategy we were trying to predict. Because we asked participants to evaluate traits relevant to their own character (e.g., honesty, productivity), it is possible that some people's assessments were biased by socially desirable responding.To measure individuals' tendency to manage favorable impressions, we used the scale developed by Paulhus in his Balanced Inventory of Desirable Responding (Paulhus, 1988, 1991). Examples of items are (1) 'When I hear people talking privately, I avoid listening,' and (2) 'I have taken sick-leave from work or school even though I wasn't really sick' (reverse-coded). Responses were given on a scale from 1, 'not true,' to 7, 'very true.' The alpha for this measure was 0.84.

Results
Manipulation checks
Of the 96 participants in the agent role, 2 (2.0 per cent) told the truth, 41 (42.7 per cent) concealed, and 53 (55.2 per cent) lied. Hence, the procedures appear to have encouraged a good deal of deception. To assess the effectiveness of the ethical climate manipulation, on a pre-negotiation questionnaire agents were asked to indicate their agreement with this question: 'At General, it is acceptable for agents to ''stretch'' the truth if it will allow them to negotiate favorable prices with suppliers' (responded to on a scale from 1 ¼ strongly disagree to 5 ¼ strongly agree). Agents disagreed more with this statement in

NEUTRALIZATION STRATEGIES

671

the ethical climate condition (M ¼ 2.30) than in the non-ethical climate condition (M ¼ 4.20), F(1, 92) ¼ 78.76, p< 0.0001. To check the moral consequences manipulation, agents were asked to specify their agreement with the following item: 'Based on your investigation of Midwest, it is very likely that they will be able to stay in business if they experience a significant financial loss after signing a contract with General.' Agents disagreed more with this statement in the high moral consequences condition (M ¼ 1.43) than in the low moral consequences condition (M ¼ 4.31), F(1, 94) ¼ 289.36, p< 0.0001. Hence, both manipulations were effective.

Hypothesis tests
Results of the regressions for minimization and denigration are shown in Table 1. Using the procedures suggested by Aiken and West (1991), the variables forming the interactions were centered to minimize multicollinearity among the interaction terms and their individual components. Variables involved in the interactions were centered prior to running the analyses. As discussed earlier, we included both liars and concealers in this analysis because the comparison of concealers to liars provides a useful reference point for evaluating liars' responses to their lies. Because participants had the opportunity to use minimization before they could use denigration, we controlled for minimization in the model predicting denigration. As shown in the table, higher levels of psychic distress were associated with less denigration (fi ¼ À0.33, p< 0.01). This contradicts the prediction of Hypothesis 1. Because the opportunity for minimization occurred prior to the measurement of distress, Hypothesis 1 could not be tested for this neutralization strategy. Hypothesis 2 posits that people who lie are more likely to execute neutralization strategies if they are high in perceived moral attributes. Table 1 shows that self-perceived moral attributes interacted with
Table 1. Regression results: the minimization and denigration strategies Minimizing consequences Independent variables Impression management Lie Distress (denigration only) Minimizing consequences Self-perceived moral attributes Ethical climate Moral consequences Lie  moral attributes Lie  ethical climate Lie  consequences Climate  consequences Climate  moral attributes Consequences  moral attributes Lie  moral attributes  climate Lie  moral attributes  consequences R2 Adjusted R2 F d.f. N
*p< 0.05; **p< 0.01.

Denigration fi À0.01 0.15 À0.33 0.23 À0.11 0.05 0.08 À0.12 À0.09 0.09 À0.11 À0.30 0.03 À0.02 0.03 0.30 0.12 1.70 15, 61 77 t À0.09 1.19 À2.77** 1.77 À0.80 0.40 0.61 À0.85 0.76 0.73 À0.86 À2.46* 0.25 À0.14 0.22

fi 0.14 À0.18 — — À0.01 0.19 0.09 À0.33 0.13 À0.09 À0.27 À0.11 0.12 0.15 0.15 0.32 0.18 2.25* 13, 63 77

t 1.49 À1.46 — — À0.06 1.75 0.73 À2.48* 1.21 À0.75 À2.31* À0.99 0.99 1.22 1.26

672

K. AQUINO AND T. E. BECKER
29.00 28.00 Minimization (price) 27.00 26.00 25.00 24.00 23.00 22.00 Low Self-perceived moral attributes High Concealers Liars

Figure 1. Graph of the interaction between lying and self-perceived moral attributes on minimization

lying (fi ¼ À0.33, t ¼ À2.48, p< 0.05) to predict minimization. Following the procedures recommended by Aiken and West (1991), we graphed the nature of this interaction by running separate regressions of minimization on self-perceived moral attributes (a continuous variable) for concealers and liars (a dichotomous variable). Low and high levels of moral attributes were defined by values 1 standard deviation below and above the mean on this variable, respectively. The result is shown in Figure 1. The test of the interaction reported above indicates that the slope of the regression line for liars (fi ¼ À0.09, n.s.) is significantly different from that for concealers ( fi ¼ 0.31, p< 0.10). Notice that although the relationship for liars is not in the predicted direction, the pattern for concealers—people who were coded as not being entirely truthful in the negotiation —was consistent with Hypothesis 2. Hypothesis 3 predicts that when lying occurs the positive relationship between self-perceived moral attributes and one or more of the neutralization strategies will be stronger in an ethical compared to a nonethical climate. Hypothesis 4 predicts that when lying occurs the positive relationship between selfperceived moral attributes and one or more of the neutralization strategies will be stronger when moral consequences are high rather than low. The lying  self-perceived moral attributes  ethical climate interaction is not significant for minimization or denigration, thus failing to support Hypothesis 3. Contrary to Hypothesis 4, we did not find a lying  self-perceived moral attributes  moral consequences interaction. Logistic regression was used to test the hypotheses for denial. Only agents who lied were included in this analysis because only liars have reason to say they did not lie when, in fact, they did. Denial was dummy coded 0 ¼ admit lying, 1 ¼ deny lying. Since denial was the last strategy agents had an opportunity to execute, we controlled for minimization and denigration in our model. The results are shown in Table 2. As can be seen, the maximum likelihood estimate for the effect of psychic distress is significant and in the predicted direction ( ¼ 3.52, p< 0.05), indicating that liars who experienced greater levels of distress were more likely to deny lying. This supports Hypothesis 1. Self-perceived moral attributes did not predict denial, thus failing to support Hypothesis 2 for this neutralization strategy. We found no support for a self-perceived moral attributes  ethical climate interaction or a selfperceived moral attributes  moral consequences interaction. Hence, Hypotheses 3 and 4 were not supported for denial.

NEUTRALIZATION STRATEGIES

673

Table 2. Results of logistic regression: the denial strategy Independent variables Impression management Minimization Denigration Distress Self-perceived moral attributes Ethical climate Moral consequences Climate  moral attributes Consequences  moral attributes Estimate 1.067 0.003 À 0.140 3.522 À1.144 À0.040 À0.031 1.184 0.350 SE 1.013 0.002 0.673 1.635 1.029 0.552 0.517 0.860 0.915 _2 1.11 3.58 0.04 4.64* 1.24 0.01 0.00 1.89 0.15

*p< 0.05. N ¼ 44 (all liars who did not impasse). Estimate, maximum likelihood estimates; SE, standard error.

Ancillary findings
There were three findings that were significant, but not hypothesized. The interaction between ethical climate and moral consequences was significant (fi ¼ À0.27, t ¼ À 2.31, p< 0.05) in the model predicting minimization. Because both variables were categorical and because sample sizes were not equal in the various conditions, we graphed the interaction by plotting the least squares means. The results are shown in Figure 2. Tests of mean differences indicated that when moral consequences were low, being in an ethical rather than non-ethical climate led to greater minimization (Mdiff ¼ 2.82, p< 0.05). However, when moral consequences were high, there was no significant difference (Mdiff ¼ 1.23, n.s.). Thus, under certain conditions, it appears that negotiating within an ethical climate can lead to greater use of some neutralization strategies. We also found a significant two-way interaction between self-perceived moral attributes and ethical climate (fi ¼ À0.30, t ¼ À2.46, p< 0.05) on the denigration strategy. Because one of the central variables, moral attributes, was measured continuously, we graphed the nature of the interaction by running separate regressions of denigration on moral attributes for the ethical and non-ethical climate. As before, low and high levels of moral attributes were defined by values 1 standard deviation below and above the mean. The result is shown in Figure 3. The test of the interaction indicates that the slope of
27.00 26.50 26.00 Minimization (price) 25.50 25.00 24.50 24.00 23.50 23.00 22.50 22.00 Low High Moral consequences Non-ethical climate Ethical climate

Figure 2. Graph of the interaction between ethical climate and moral consequences on minimization

674

K. AQUINO AND T. E. BECKER
3.3 3 2.7 2.4 2.1 1.8 1.5 Low High Self-perceived moral attributes Non-ethical climate Ethical climate

Figure 3. Graph of the interaction between ethical climate and self-perceived moral attributes on denigration

the regression line for an ethical climate (fi ¼ À0.40, p< 0.05) is significantly different from that for a non-ethical climate (fi ¼ 0.09, n.s.). This pattern indicates that in an ethical climate, as self-perceived moral attributes increase, the inclination to denigrate others following negotiation tends to decrease. No such trend is found in a non-ethical climate. Finally, we found a nearly significant relation between minimization and denial ( ¼ 0.003, p ¼ 0.059). The direction of the relationship indicated that negotiators who lied but who agreed to a higher price were more likely to deny having lied.

Discussion
Our results provide a glimpse into the circumstances under which people execute various neutraliza- tion strategies, but they also suggest that some of our original theoretical assumptions need to be mod- ified. We comment on these modifications shortly, but first we consider one of our predictions that was supported by data. The positive relationship between distress and denial supported Hypothesis 1 and suggested that feelings such as guilt and shame can motivate the use of at least one type of neutralization strategy. Although it may at first seem odd that distress is inversely related to another strategy—denigration—this is not unexpected given our assumption that people may use different strategies under different circumstances. One interpretation of these findings is that participants in our study used denial when distress was high but denigration when distress was low because the former is a more effective strategy for alleviating strong feelings of guilt than the latter. This is because denigrating a victim's character right after harming them by telling a lie could increase guilt by, in effect, adding insult to injury. Denial does not appear to run such a risk. We found that behavioral minimization (foregoing financial gain to compensate for a lie) occurred most often under two conditions: (1) when people who had high self-perceived moral attributes con- cealed information from, but did not lie to, their negotiation partner; and (2) when the negotiation took place in the presence of an ethical climate where the negative consequences of lying on the target of the lie were not severe. Neither of these results was hypothesized, but they offer some insight into when

Denigration

NEUTRALIZATION STRATEGIES

675

people in our study might have attempted to use minimization as a neutralization strategy and how our theoretical arguments may need to be modified to account for this. Liars who evaluated themselves favorably did not try to minimize any more than liars who did not evaluate themselves this way. One interpretation of this result is that this particular neutralization strat- egy might not have been the preferred one among those who made the decision to lie. This makes sense if one assumes that people who chose to lie did so because they wanted to gain a bargaining advantage over the other party. If this assumption is valid, then it would seem irrational for them to give up the gains that may have resulted from employing this tactic. On the other hand, concealers may have been far less determined to use deception as a way of maximizing their outcome relative to the other party. Consequently, for concealers who perceived themselves favorably, giving up some gains to assuage their guilt might have been an attractive neutralization strategy for maintaining self-esteem. The preceding argument suggests that in the present context lying was directly linked to achieving a specific goal (i.e., reaching a personally favorable agreement) and so the 'cost' of minimization may have been higher for liars than concealers. This leads us to refine our theoretical arguments by suggest- ing that the motivation to execute different neutralization strategies will not always be higher for liars than those who engage in other more socially acceptable forms of deception. Rather, the motivation to execute a given strategy depends on how difficult it is to execute under the circumstances, and whether it seems promising as a means of achieving the goal the person is pursuing by acting unethically. In our study, the data suggest that liars may have found the denial strategy as potentially more effective than minimization of consequences. Turning to the climate-by-consequences interaction, we suspect that the central question confront- ing agents when negative consequences to the supplier were unlikely was: 'Nothing bad will happen to the other party if I negotiate a low price, so why should I give them more than I have to?' In the non- ethical climate condition the answer appears to have been, 'I shouldn't.' In the ethical climate condi- tion, however, participants had been told that their company 'prides itself on being fair and honest in its business dealings.' Such norms may have encouraged people to reach a fairer settlement price even when no certain and serious harm would result from reaching an agreement favoring oneself at the expense of the other party. On the other hand, when the possible negative consequences to the other party of agreeing to a low price were more severe, the range of agreements that could be reached with- out driving the other party into bankruptcy were more limited. Under these conditions the normative information provided by the ethical climate manipulation may have had little in?uence on negotiators' behavior because their strategy in the negotiation was driven primarily by considerations of whether the other party would be harmed by agreeing to a low price. This concern for not seriously harming the other party would explain why there was no effect of ethical climate on minimization when moral con- sequences were high. One of our ancillary findings showed that, in an ethical climate, people who evaluated their moral attributes favorably also evaluated their opponent positively. However, evaluations of the opponent did not differ as a function of moral attributes when no ethical climate was present. This finding leads us to extend our arguments by proposing that an ethical climate may in?uence social cogn itions such that people who view themselves in a positive light are more likely to perceive others the same way. It may be that people with favorable views of themselves come to believe that in an ethical climate most peo- ple, including perhaps those outside the organization, share ethical values as a result of the information provided by their social environment. People who view themselves more negatively, on the other hand, may be motivated more by self-serving biases to assume that others are like them—not so ethical. In a non-ethical climate, however, those viewing themselves positively may have more difficulty discerning whether others are likely to be ethical because the social information they receive from their environ- ment does not indicate that this is the case. Some, like themselves, may be driven by their own values to behave ethically, while others may not share such values. In contrast, those with more negative

676

K. AQUINO AND T. E. BECKER

views of themselves would seem to have the same motivation as before; that is, they assume others are no more ethical than themselves. This line of reasoning fits the pattern of our results, but our data do not allow a direct test. Future research might explore this possibility given that it seems theoretically plausible. Although Hypotheses 1 and 2 received some measure of support, financial compensation for lying and denigrating victims of a lie were not significantly affected by the predicted three-way interactions among lying, self-perceived moral attributes, and the two experimental manipulations. In retrospect, the lack of results involving the moral consequences variable may have been caused by insufficiently severe consequences. Given the experimental nature of the design, participants presumably knew that the consequences described in the background information would not truly occur and, therefore, may not have been adequately affected by these perceived outcomes. Another potential cause of the null findings is the combination of modest effect sizes and moderate sample sizes. Post hoc analyses revealed that power values for the tests in Tables 1 and 2 were such that small or medium effects in the population would likely go undetected. On the positive side, the population effect sizes for our significant findings are likely to be quite large because otherwise they would not have been detected. The current study has several limitations in addition to those already mentioned. First, there was not a consistent pattern of effects of the independent variables on neutralization strategies. We provided one reason for this earlier: certain strategies may be preferred under some circumstances but not others. Another reason may be that lying is not the most serious form of deviance in organizations. For example, theft and violence are probably perceived as more serious ethical violations. These kinds of violations may more vigorously motivate the use of neutralization strategies than does lying. To address this limitation, future studies could examine deviant behaviors that might induce a stronger drive to employ neutralization strategies. A second limitation is that although psychic distress was theorized to be a proximal cause of neutralization strategies, this variable was not measured at multiple points throughout the study. Ideally, psychic distress would have been assessed before the lie, immediately after it, and then again after the person had executed a neutralization strategy. In fact, although we measured distress in the postnegotiation survey, we did not assess it at the other relevant times. Still, given that deviant beha- vior can produce psychic distress (e.g., Festinger & Carlsmith, 1959), it seems reasonable to believe that lies told by participants in the current study generated negative emotions. Indeed, explaining the results of the study without reference to psychic distress seems difficult. Third, it is possible that social desirability bias may have in?uenced responses to the pre- and postquestionnaires. However, if this bias occurred it should have equally affected the results across conditions. At any rate, this concern cannot easily negate the significant findings of the study. Finally, any laboratory manipulation of ethical climate will fail to capture the full complexity and presence of an ethical climate in a real work organization. Given this, we were pleased that our manipulation of this variable was clearly detectable (per our manipulation check) and had interactive effects with moral consequences (on minimization) and with self-perceived moral attributes (on denigration). We suspect that because the ethical climate in an organization is likely to be more pervasive and powerful, the effects would be even stronger than those found in this study. Our findings have several implications for organizations. First, recent corporate scandals in the United States and elsewhere highlight the importance of studying lying and integrity in the workplace. Without adequate understanding in this area, ruined reputations, declining business outcomes, and unhappy customers and employees may be predicted to continue. Our general conclusion that lying and concealing are easy to induce and can motivate the use of neutralization strategies designed to cover up wrongdoing argues for further attention to this topic by organizational scientists and practitioners. Second, although ethical organizational climates are clearly desirable, the finding that those low in self-perceived moral attributes are prone to using neutralization strategies (denigration

NEUTRALIZATION STRATEGIES

677

in particular) in such climates is troubling. If more high-integrity people can be hired, or if training can reduce tendencies toward post-lie rationalization, the unintended effects of ethical climates on neutralization strategies might be reduced or eliminated. Finally, our finding that in a non-ethical climate high moral consequences resulted in more favorable outcomes for negotiation partners suggests that emphasizing the negative consequences of unethical acts may be ameliorative. This is because, as we discussed earlier, most people may be genuinely reluctant to lie or conceal information when they believe doing so will result in serious harm to others. In conclusion, we hope this study encourages further investigation of neutralization strategies so that we might gain a better understanding of when and how people who do wrong manage to convince themselves and others that what they did was right.

Acknowledgments
We thank Mike Gordon, Rebecca Hochstein, and Phil Oyerly for their assistance with the data collection.

Author biographies
Karl Aquino is an Associate Professor in the Department of Organizational Behaviour and Human Resources at the University of British Columbia. He conducts research on moral psychology, organizational justice, workplace victimization, revenge and forgiveness, antisocial behavior in organiza- tions, and the effects of status and group-based dominance on workplace outcomes. Thomas E. Becker is an Associate Professor of Management at the University of Delaware, where he teaches and conducts research on topics in organizational behavior and human resource management. He is especially interested in employee commitment, integrity, motivation, and job performance.

References
Abrahamson, E., & Park, C. (1994). Concealment of negative organizational outcomes: an agency theory perspective. Academy of Management Journal, 37, 1302-1334. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Anton, R. (1990). Drawing the line: an exploratory test of ethical behavior in negotiations. International Journal of Con?ict Management, 1, 265-280. Aquino, K., & Reed, A., II (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423-1440. Barker, T., & Carter, D. (1990). Fluffing up the evidence and covering your ass: some conceptual notes on police lying. Deviant Behavior, 11, 61-73. Baumeister, R. F. (1998). The self. In T. Gilbert, & S. T. Fiske (Eds.), The handbook of social psychology (Vol. 2, 4th edn, pp. 680-740). Boston, MA: McGraw-Hill. Becker, T. E. (1998). Integrity in organizations: beyond honesty and conscientiousness. Academy of Management Review, 23, 154-161.

678

K. AQUINO AND T. E. BECKER

Becker, T. E. (2000). Hallmarks and consequences of integrity in organizations: The employees' perspective. Paper presented at the meeting of the Academy of Management, Toronto, Canada, August 2000. Beckman, N. (1977). Negotiations. Lexington, MA: Lexington Books. Bersoff, D. M. (1999). Why good people sometimes do bad things: motivated reasoning and unethical behavior. Personality and Social Psychology Bulletin, 25, 28-39. Bies, R. J., & Moag, J. S. (1986). Interactional justice: communication criteria of fairness. In M. H. Bazerman, R. J. Lewicki, & B. H. Sheppard (Eds.), Research on negotiation in organizations (Vol. 3, pp. 43-55). Greenwich, CT: JAI Press. Blasi, A. (1984). Moral identity: its role in moral functioning. In W. Kurtines, & J. Gewirtz (Eds.), Morality, moral behavior and moral development (pp. 128-139). New York: Wiley. Bok, S. (1978). Lying: Moral choice in public and private life. New York: Vintage Books. Bok, S. (1982). Secrets. New York: Pantheon. Carr, A. Z. (1968). Is business bluffing ethical? Harvard Business Review, 46, 143-153. Cialdini, R. B., Finch, J. F., & DeNicholas, M. E. (1990). Strategic self-presentation: the indirect route. In M. J. Cody, & M. L. McLaughlin (Eds.), The psychology of tactical communication (pp. 194-206). Clevedon, UK: Multilingual Matters. DePaulo, P. J., & DePaulo, B. M. (1989). Can deception by salespersons and customers be detected through nonverbal behavioral cues? Journal of Applied Social Psychology, 19, 1552-1577. Ekman, P. (1985). Telling lies: Clues to deceit in the marketplace, politics, and marriage. New York: Norton. Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press. Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. Journal of Abnormal and Social Psychology, 58, 203-210. Grover, S. L. (1993). Lying, deceit, and subterfuge: a model of dishonesty in the workplace. Organization Science, 4, 478-495. Grover, S. L. (1997). Lying in organizations: theory, research, and future directions. In R. A. Giacalone, & J. Greenberg (Eds.), Antisocial behavior in organizations (pp. 68-84). Thousand Oaks, CA: Sage. Grover, S. L., & Hui, C. (1994). The in?uence of role con?ict and self-interest on lying in organizations. Journal of Business Ethics, 13, 295-303. Hales, S. (1985). The inadvertent rediscovery of self in social psychology. Journal for the Theory of Social Behavior, 15, 237-282. Haring, B. (1979). Free and faithful in Christ: Moral theology for priests and laity. New York: Seabury Press. Hegarty, W. H., & Sims, H. P. (1978). Some determinants of unethical decision behavior: an experiment. Journal of Applied Psychology, 63, 451-457. Jones, T. M. (1991). Ethical decision making by individuals in organizations: an issue-contingent model. Academy of Management Review, 16, 366-395. Kashy, D. A., & DePaulo, B. M. (1996). Who lies? Journal of Personality and Social Psychology, 70, 1037-1051. Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108, 480-498. Lewicki, R. J. (1983). Lying and deception: a behavioral model. In M. H. Bazerman, & R. J. Lewicki (Eds.), Negotiating in organizations (pp. 68-90). Beverly Hills, CA: Sage. Locke, E. A., & Woiceshyn, J. (1995). Why businessmen should be honest: the argument from rational egoism. Journal of Organizational Behavior, 16, 405-414. Oxford English Dictionary. (1993). New York: Oxford University Press. Paulhus, D. L. (1988). Assessing self-deception and impression management in self-reports: The Balanced Inventory of Desirable Responding. Manual, University of British Columbia, Vancouver. Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. Shaver, & L. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 17-59). San Diego: Academic Press. Robinson, S. L., & Kraatz, M. S. (1998). Constructing the reality of normative behavior: the use of neutralization strategies by organizational deviants. In R. W. Griffin, A. O'Leary-Kelly, & J. M. Collins (Eds.), Dysfunctional behavior in organizations (pp. 203-220). Stamford, CT: JAI Press. Shapiro, D. L. (1991). The effects of explanations on negative reactions to deceit. Administrative Science Quarterly, 36, 614-630. Shapiro, D. L., & Bies, R. J. (1994). Threats, bluffs, and disclaimers in negotiations. Organizational Behavior and Human Decision Processes, 60, 14-35. Steele, C. M. (1988). The psychology of self-affirmation: sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 261-302). New York: Academic Press. Sykes, G., & Matza, D. (1957). Techniques of neutralization: a theory of delinquency. American Sociological Review, 22, 664-670.

NEUTRALIZATION STRATEGIES

679

Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: a social psychological perspective on mental health. Psychological Bulletin, 103, 193-210. Tenbrunsel, A. E. (1998). Misrepresentation and expectations of misrepresentation in an ethical dilemma: the role of incentives and temptation. Academy of Management Journal, 41, 330-339. Trevino, L. K., Butterfield, K. D., & McCabe, D. L. (1998). The ethical context in organizations: in?uences on ˜ employee attitudes and behaviors. Business Ethics Quarterly, 8, 447-476. Victor, B., & Cullen, J. B. (1988). The organizational bases of ethical work climates. Administrative Science Quarterly, 33, 101-125. York, K. M. (1989). Defining sexual harassment in the workplace: a policy capturing approach. Academy of Management Journal, 32, 830-850.



doc_466780327.docx