Ethical Decision Making by Individuals i (1)

Ethical Decision Making by Individuals in Organizations: An Issue-Contingent Model Author(s): Thomas M. Jones Source: The Academy of Management Review, Vol. 16, No. 2 (Apr., 1991), pp. 366-395 Published by: Academy of Management Stable URL: http://www.jstor.org/stable/258867 . Accessed: 22/01/2014 23:06 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy of Management Review. http://www.jstor.org This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions ? Academy of Management Review 1991,Vol. 16, No. 2, 366-395. ETHICALDECISIONMAKINGBY INDIVIDUALS IN ORGANIZATIONS: AN ISSUE-CONTINGENT MODEL THOMASM. JONES University of Washington Existing theoretical models of individual ethical decision making in organizations place little or no emphasis on characteristics of the ethical issue itself. This article (a) proposes an issue-contingent model containing a new set of variables called moral intensity; (b) using concepts, theory, and evidence derived largely from social psychology, argues that moral intensity influences every component of moral decision making and behavior; (c) offers four research propositions; and (d) discusses implications of the theory. Reasons for increased societal focus on ethics in organizations are many. Insider trading on Wall Street; defense contractor scandals, involving both private and public sectors; rental car repair overcharges; and the resignation of over 100 Reagan administration officials have helped keep ethical issues in the public eye. Institutions have responded to these challenges in a variety of ways. Corporations have established or updated codes of ethics, and some business schools have responded with increased offerings in business ethics. Academe has also produced a greatly expanded literature on the subject of ethics, including textbooks and two Journal of Business Ethics and the Business and scholarly journals-the Professional Ethics Journal. An entire volume of Research in Corporate Social Performance and Policy has been devoted to business ethics and values (Frederick, 1987). Despite this increased attention to ethics in organizations, theoretical and empirical examinations of ethical decision making in organizations are in relatively short supply. Trevino (1986) offered a general theoretical model, whereas Ferrell and Gresham (1985), Hunt and Vitell (1986), and Dubinsky and Loken (1989) offered models that focus on marketing ethics. Rest (1986) presented a theory of individual ethical decision making that can easily be generalized to organizational settings. Among the empirical contributions to date are the works of Hegarty and Sims (1978, 1979), Fritzsche and Becker (1983), Frederick (1987), Laczniak and Inderrieden (1987), Fritzsche (1988), Dubinsky and Loken (1989), and Weber (1990). One reason for this relative The author wishes to thank Vandra Huber, who made useful suggestions on an earlier draft of this article. Marilyn Gist's thoughtful critique helped to polish the final draft. 366 This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 367 paucity of theoretical and empirical work in ethics may be that few scholars are interested in both ethics and organizational behavior and decision making. The models that have emerged are the products of scholars in psychology or psychology-based disciplines, including organizational behavior and marketing. In addition, organizational scholars may be reluctant to study value-based issues because of ideological reasons or because methodological problems are considered difficult to surmount. Although this article is grounded in social psychology, it also contains elements of moral philosophy and applied ethics. The purpose of this article is to introduce concepts not present in prior models and to offer a model that supplements, but does not replace, other models. The article argues that moral issues vary in terms of their moral intensity and that an issue-contingent model of ethical decision making and behavior can add significantly to the understanding of moral processes. This is an initial attempt to identify, not empirically validate, the issuerelated components of ethical behavior on which future research may be based. It attempts to build a nomological net of constructs and theory, which can be formally validated and tested in future studies. The validity of the proposed constructs is here limited to content validity based on logic, observation, and, in some cases, empirical analogy. The article also advances and discusses four general research propositions. DEFINITIONS Three definitions are central to the article. First, a moral issue is present where a person's actions, when freely performed, may harm or benefit others (Velasquez & Rostankowski, 1985). In other words, the action or decision must have consequences for others and must involve choice, or volition, on the part of the actor or decision maker. The definition is broad; decisions frequently have some consequences for others and volition is almost always present, although the costs of certain choices may be high. In sum, many decisions are moral decisions simply because they have a moral component. Second, a moral agent is a person who makes a moral decision, even though he or she may not recognize that moral issues are at stake. This feature of the definition is important because a central element of the moral decision-making model presented here is recognizing moral issues. (In this article, the terms moral and ethical are considered equivalent and will be used interchangeably, depending on context.) Third, an ethical decision is defined as a decision that is both legal and morally acceptable to the larger community. Conversely, an unethical decision is either illegal or morally unacceptable to the larger community. This definition follows from Kelman and Hamilton's (1989) definition of crimes of obedience and is consistent with the definitions used, either explicitly or implicitly, by some other authors in the field of ethics. Although the definition is admittedly imprecise and relativistic, it is adequate for the purposes of this article. Some authors, including Ferrell and Gresham (1985), Trevino This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 368 Academy of Management Review April (1986), Hunt and Vitell (1986), and Dubinsky and Loken (1989) did not provide substantive definitions of the terms ethical and unethical. Discussions regarding the difficulty of establishing substantive definitions for ethical beMoberg, and Velasquez (1981), havior can be found in Cavanagh, Beauchamp and Bowie (1979), and Jones (1980). EXISTINGMODELS Rest (1986) proposed a four-component model for individual ethical decision making and behavior, whereby a moral agent must (a) recognize the moral issue, (b) make a moral judgment, (c) resolve to place moral concerns ahead of other concerns (establish moral intent), and (d) act on the moral concerns. He argued that each component in the process is conceptually distinct and that success in one stage does not imply success in any other stage. For example, a person with a well-developed sense of moral reasoning (Component 2) will not necessarily have great resolve to act morally (Component 3). Much of the empirical research conducted in the context of this model has involved either Component 2, called moral development by Kolberg (1976) and Rest (1979, 1986), or the relationship between Components 2 and 4, moral development and action. Rest (1979) developed an instrument for measuring moral development that can be administered in groups and scored relatively easily, which probably accounts for the dozens of empirical studies involving this stage of the process. Although Trevino (1986) did not directly address Rest's model, she offered a competing model, which implicitly builds on it. Her person-situation interactionist model begins with the existence of an ethical dilemma and proceeds to a cognitions stage, wherein Kohlberg's cognitive moral development model becomes operative. Moral judgments made in the cognitions stage are then moderated by individual and situational factors. Individual factors include ego strength, field dependence, and locus of control. Situational factors include elements of immediate job context, organizational culture, and characteristics of the work. Moral judgments, thus moderated, affect ethical or unethical behavior. Ferrell and Gresham (1985) proposed a contingency framework for ethical decision making in marketing. In this model, an ethical issue or dilemma emerges from the social or cultural environment. The contingent factors that affect the decision maker are both individual (knowledge, values, attitudes, and intentions) and organizational (significant others and opportunity). The effect of significant others is supported in this model by differential association theory (Sutherland & Cressey, 1970) and role-set theory (Merton, 1957). Opportunity (to behave unethically) as a variable stems from the work of Cloward and Ohlin (1960) and, in Ferrell and Gresham's model, is related to the existence (or nonexistence) of professional codes, corporate policy, and rewards and punishment. The decision that emerges from this process leads first to behavior and next to evaluation of behavior, which, in turn, is the starting point for a feedback loop to individual and organizational factors. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 369 Hunt and Vitell (1986) proposed a general theory of marketing ethics that consists of several stages. A substantially simplified summary of this model is offered here. Environmental factors (cultural, industrial, and organizational) and personal experiences affect perceptions of the existence of an ethical problem, alternatives, and consequences. In turn, these perceptions, along with deontological norms and an evaluation of consequences, lead to both deontological and teleological evaluations, which, in turn, lead to ethical judgments. Judgment affects intentions, which, along with situational constraints, affect behavior. A feedback loop leads from behavior to actual consequences and back to personal experiences. Dubinsky and Loken (1989) presented an ethical decision-making model based on the theory of reasoned action (Fishbein & Ajzen, 1975). Their model begins with behavioral beliefs, outcome evaluations, normative beliefs, and motivation to comply. The first two of these variables affect attitude toward ethical or unethical behavior; the latter two variables affect subjective norms toward ethical or unethical behavior. Finally, attitude and subjective norms lead to intentions to engage in ethical or unethical behavior which, in turn, affect actual behavior, ethical or unethical. No feedback loop is present. Although they did not add to present theory, Ferrell, Gresham, and Fraedrich (1989) developed a five-stage synthesis of other models. Awareness (of ethical issues), cognitions (moral development), moral evaluations (deontological and teleological judgments), determination (intentions), and actions (ethical or unethical behavior) constitute the sequential order of their model. They also featured a feedback loop with behavioral evaluation of consequences leading to awareness, cognitions, moral evaluations, and determination. Brommer, Gratto, Gravender, and Tuttle (1987) also claimed a model of ethical decison making, but it actually distills to a catalog (albeit a thorough one) of factors that influence ethical decision makers. Environmental factors (work, personal, professional, governmental, legal, and social) join individual attributes to affect the ethical decision process. In all, over 20 variables are expected to be relevant to ethical decision making in this formulation. Each of these models has something to contribute to the understanding of ethical decision making. None, however, does more than hint that characteristics of the moral issue itself will affect the moral decision-making process. Ferrell and Gresham (1985) noted that the consensus regarding proper ethical conduct will be likely to change as the issue changes. They suggested that fewer people would endorse embezzling company funds than would endorse padding an expense account. Their model, however, includes no acknowledgement that issue differences affect ethical decision making. Hunt and Vitell (1986) added a teleological evaluation stage, wherein the consequences of the moral decison are evaluated, but they did not suggest a systematic relationship between consequences and subseand behavior. Dubinsky and quent elements of the model-intentions Loken (1989) implied that attitudes may vary according to the behavior This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions Academy of Management Review 370 April being evaluated, but they made no attempt to explain how this variation would occur. In sum, existing models do not adequately account for differences in ethical issues. A rough synthesis of existing models is useful for assessing their collective strengths and weaknesses. This synthesized model, shown in Figure 1, is necessarily simplified (e.g., feedback loops are omitted), and it uses Rest's (1986) four-stage model as a foundation. The process begins with the environment, which typically includes economic, social, cultural, and organizational factors (Ferrell & Gresham, 1985; Hunt & Vitell, 1986). From the environment emerge ethical issues (Ferrell & Gresham, 1985; Hunt & Vitell, 1986). Hunt and Vitell (1986) acknowledged that not all moral issues may be recognized by their use of the term per- FIGURE1 Synthesis of Ethical Decision-Making Models Characteristics of the Moral Issue (MoralIntensity) ,, ' 0' 0' Environment Social Cultural Economic Recognize Moral Issue -' / Jones's Model _ _ / Make Moral Judgment Establish Moral Intent R D/L H/V R D/L H/V Behavior Organizational T FIG FIG H/V Explicit R Cognitive Moral Development H/V R T Implicit T F/G Not Specified D/L Moral Evaluation D/L H/V Not Specified l DI Significant Others F/G uIndividual Moderators F/G F/G T Situationalb. Moderators Opportunity F/G Key: R T Engage in Moral = Rest (1986) = Trevino (1986) D/L = Dubinsky& Loken (1989) F/G = Ferrell & Gresham, (1985) H/V = Hunt & Vitell(1986) This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 371 ceived ethical problem, whereas Rest (1986) made recognition of moral issues an explicit element of his model. Trevino (1986) and Ferrell and Gresham (1985) left this step implicit, but Dubinsky and Loken (1989) did not include it. Four of the five models contain some form of moral judgment stage. In Rest's (1986) and Trevino's (1986) models, cognitive moral development is the critical element in the judgment phase. For Hunt and Vitell (1986) and Dubinsky and Loken (1989), moral evaluation (teleological and deontological) takes place. Ferrell and Gresham (1985) did not specify a process for this step. Rest (1986), Dubinsky and Loken (1989), and Hunt and Vitell (1986) explicitly included a step whereby the ethical decision maker establishes moral intent before engaging in moral behavior. Trevino (1986) and Ferrell and Gresham (1985) postulated a direct transition from the moral judgment phase to moral behavior. Moderating variables include significant others (Dubinsky & Loken, 1989; Ferrell & Gresham, 1985), individual moderators (Ferrell & Gresham, 1985; Trevino, 1986), situational moderators (Trevino, 1986), and opportunity (Ferrell & Gresham, 1985). Despite the fact that collectively these models are reasonably comprehensive, this synthesized model clearly shows that none of the previous models of ethical decision making explicitly includes characteristics of the moral issue itself as either an independent variable or a moderating variable. If the models making up this synthesized model are taken at face value, the moral decision-making and behavior process of individuals in organizations is identical for all moral issues. For example, people will decide and behave in the same manner whether the issue is the theft of a few supplies from the organization or the release of a dangerous product to the market. As the relationships represented by dotted lines (see Figure 1) indicate, the model presented here and explained in detail in the following sections explicitly rejects that view and formally includes characteristics of the moral issue itself as an independent variable affecting all four stages of moral decision making and behavior. Specifically, ethical decision making is issue contingent; that is, characteristics of the moral issue itself, collectively called moral intensity, are important determinants of ethical decision making and behavior. The issuecontingent model presented here owes its genesis to intuitive, observational, and empirical factors. Intuitively, people tend to become much more concerned about moral issues that affect those who are close to them rather than those with whom they have little or no contact. Individuals also seem to react more strongly to injustices that have immediate effects as opposed to those that have effects in the distant future. In terms of observational evidence, the quotation of a militiaman for one of Beirut's factional forces explains his reasoning that as long as the victims are strangers, the killing doesn't bother him: "Those I do not know, I don't care about," and "for bloody and tough battles, I take other fighters, not the ones I have here. These are my friends. I don't want a guilty conscience if This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 372 Academy of Management Review April something happens" (Newsweek, 1989: 58). Similarly, the Soviet invasion of Afghanistan, though recognized by Americans as horrible, registered "only at the fringes of our consciousness," according to Kaplan (1989: 29). This occurred, he argued, because "the war in Afghanistan-in addition to being difficult to report-happened too far away, to an alien people with few ethnic compatriots in America" (Kaplan, 1989: 29). These observations suggest that distance (physical, psychological, cultural, or social) affects the manner in which human beings view moral issues. In an organizational context, it can be noted that some employees, who would not consider stealing from individual strangers, pilfer supplies from their firms or make personal long-distance calls on company phones. Dispersion of effect seems to play a role in these moral decisions. Empirically, Fritzsche and Becker (1983: 297), who conducted a survey of managerial attitudes toward the appropriate responses to various ethical vignettes, rejected the hypothesis that "the behavior of marketing managers is invariant across categories of ethical problems." Further, respondents "would act more ethically in the dilemmas posing serious consequences than they would in less risky situations" (1983: 297). Similarly, Fritzsche (1988) found that marketing managers responded differentially to ethical vignettes where the magnitude of the consequences was varied. In some of the vignettes, the variation was related to the magnitude of the consequences. Additionally, Weber (1990) offered evidence that corporate managers use different modes of moral reasoning for different types of moral issues. The previous evidence suggests that human beings may respond differentially to moral issues in a way that is systematically related to characteristics of the issue itself. This article, drawing on theory from social psychology, argues that six characteristics of the moral issue (magnitude of consequences, social consensus, probability of effect, temporal immediacy, proximity, and concentration of effect) will be positively related to moral decision making and behavior. These characteristics of moral issues, collectively called moral intensity, are integral parts of an issue-contingent model of moral decision making and behavior. This model is discussed in detail, following an elaboration of the moral intensity construct. MORALINTENSITY Central to the issue-contingent model presented here is the notion of moral intensity. Moral intensity is a construct that captures the extent of issue-related moral imperative in a situation. It is multidimensional, and its component parts are characteristics of the moral issue such as magnitude of consequences, social consensus, probability of effect, temporal immediacy, proximity, and concentration of effect. Moral intensity does not include traits of moral decision makers, such as moral development (Kohlberg, 1976);ego strength, field dependence, or locus of control (Trevino, 1986);or knowledge or values (Ferrell & Gresham, 1985). It also does not include organizational This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions Jones 1991 373 factors, such as organizational culture (Trevino, 1986) or corporate policies (Ferrell & Gresham, 1985). In sum, moral intensity focuses on the moral issue, not on the moral agent or the organizational context. Moral intensity is likely to vary substantially from issue to issue, with a few issues achieving high levels and many issues achieving low levels. The probable reliability and stability of moral intensity are unknown, but these parameters could be established empirically. The construct of moral intensity is not found in the literature of descriptive models of moral decision making. It is derived, in part, from the normative arguments of moral philosophers who differentiate levels of moral responsibility based on proportionality. Proportionality is related to: 1. 2. 3. 4. 5. The type of goodness or evil involved. The urgency of the situation. The certainty or probabilityof effects. The extent of the moral agent's influence on events. The availability of alternate means. (Garrett,1966:9-10) Wirtenberger (1962) offered a similar expansion of proportionality in terms of cooperation in evil acts. Legal concepts also serve as precedents for the concept of moral intensity. One of the functions of penalties in criminal law is retribution (Packer, 1968), and the extent of retribution is often proportional to the evil perpetrated. Thus, the range of sentences for murder is more severe than the range of sentences for petty larceny. This legal principle is analogous to moral intensity in ethical decision making. The argument for validity of the moral intensity construct is based on logic, analogy, and, in the case of some of its components, observations from prior research. According to Schwab, construct validation is often a sequential process. The scientist typically begins with a construct, probably ill defined. She/he suspects (hypothesizes)that this constructis related to other constructsin some sort of theoretical model which is probably also ill defined. At this point, a measure of the construct is typically developed. (1980:9) Because both the moral intensity construct and the issue-contingent model are in preliminary stages of development, the validation process suggested by Schwab will be approximated. Only content validity can currently be claimed for the moral intensity construct. The argument for content validity is based on the observations that (a) moral intensity varies from issue to issue, (b) individuals can make judgments of moral intensity, and (c) these judgments, although often subject to error and systematic bias (Kahneman, Slovic, & Tversky, 1982), are sufficiently accurate for a person to make critical distinctions. Moral philosophers are not the only ones to make judgments of proportionality on moral issues; ordinary citizens do so as well. The legal system of the United States provides evidence that human beings can and do make This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 374 Academy of Management Review April such distinctions. Trained individuals (judges) and untrained individuals (jury members) are repeatedly called upon to render legal judgments regarding guilt, liability, sentencing, and damages in the court system. Although legal issues and moral issues do not completely overlap, legal principles are often based on moral principles. Judgments of moral intensity are certainly analogous to judgments that are routinely made in courts of law. If human beings were unable to make such judgments reasonably well, the legal system would have collapsed long ago. Another approach to construct validation is elaboration of the theoretical framework that includes the construct. According to Schwab (1980), this elaboration serves two purposes. The theorized interconstruct linkages provide clarification of the construct of interest and serve as input for subsequent establishment of validation procedures. Finally, empirical testing of the hypothesized relationships among constructs can strengthen the case for construct validity of the focal construct (Schwab, 1980). Discussion of the issue-contingent model in later sections of this article will also serve to help validate the moral intensity construct. COMPONENTSOF MORALINTENSITY This article postulates that every ethical issue can be represented in terms of its moral intensity, a construct that includes six components: magnitude of consequences, social consensus, probability of effect, temporal immediacy, proximity, and concentration of effect. Definitions and examples of these components and a rationale for their inclusion in the construct follow. Magnitude of Consequences The magnitude of consequences of the moral issue is defined as the sum of the harms (or benefits) done to victims (or beneficiaries) of the moral act in question. For example: 1. An act that causes 1,000 people to suffer a particular injury is of greater magnitude of consequence than an act that causes 10 people to suffer the same injury. 2. An act that causes the death of a human being is of greater magnitude of consequence than an act that causes a person to suffer a minor injury. The inclusion of magnitude of consequences in the moral intensity construct is based on common-sense understanding and observation of human behavior and empirically derived evidence. First, the definition of moral issue is broad; decisions involving consequences for others and volition on the part of the moral agent have a moral component. However, many moral issues are quite trivial in terms of consequences. For example, most people are unlikely to become morally outraged when a co-worker is denied a desired vacation at a time when others also want to take their vacations; the consequences don't warrant it. Further, because moral issues are present in most organizational decisions, people concerned with minor issues would be morally agitated most of the time. Because people are not constantly This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions Jones 1991 375 agitated over moral issues, it is assumed that many moral issues fail to reach a threshold of magnitude of consequences. Empirically derived clues (described above) include Fritzsche and Becker's (1983) judgment that when moral dilemmas are faced, serious consequences are more likely to prompt ethical behavior than are modest consequences. Further, Fritzsche (1988) found some support for a positive link between serious consequences and the ethical responses of marketing managers to vignettes containing moral dilemmas. Also, Weber (1990) discovered a link between decision consequences and moral reasoning patterns. Finally, York (1989) determined that subjects were more likely to make judgments of sexual harassment where job consequences for the victim were more severe. Social Consensus The social consensus of the moral issue is defined as the degree of social agreement that a proposed act is evil (or good). For example: 1. The evil involved in discriminating against minority job candidates has greater social consensus than the evil involved in refusing to act affirmatively on behalf of minority job candidates. 2. The evil involved in bribing a customs official in Texas has greater social consensus than the evil involved in bribing a customs official in Mexico. (Nehemkis, 1975) Social consensus is included in the moral intensity construct for logical and empirical reasons. Logically, it is difficult to act ethically if a person does not know what good ethics prescribes in a situation; a high degree of social consensus reduces the likelihood that ambiguity will exist. Empirically, Laczniak and Inderrieden (1987) determined that subjects in an ethical judgment experiment rejected illegal decisions with far greater frequency than they rejected unethical (but not illegal) decisions. Although this result may suggest that legal penalties play a role in moral decision making, it may also be that the social consensus that is implied by legal prohibition of a practice reduces moral ambiguity for the moral agent. Indeed, these authors seemed to agree: "In order for individuals to respond appropriately to a given situation, agreement must exist as to whether or not the behavior is appropriate" (Laczniak & Inderrieden, 1987: 304). Probability of Effect The probability of effect of the moral act in question is a joint function of the probability that the act in question will actually take place and the act in question will actually cause the harm (benefit) predicted. For example: 1. Producing a vehicle that would be dangerous to occupants during routine driving maneuvers has greater probability of harm than producing a vehicle that endangers occupants only during rear-end collisions. 2. Selling a gun to a known armed robber has greater probability of harm than selling a gun to a law-abiding citizen. Probability of effect is included in the moral intensity construct for reasons of logic. The expected value of, for example, a financial gain is the product of This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions Academy of Management Review 376 April the magnitude of the gain and its probability of occurrence. Similarly, the expected consequences of a moral act would be the product of the magnitude of consequences, the probability that the act will take place, and the probability that the act will cause the harm (benefit) predicted. Moral acts of given magnitude of consequences will thus be "discounted" if either of the probabilities mentioned is substantially less than 1.00. To be sure, individuals are not good at estimating probabilities (Kahneman, Slovic, & Tversky, 1982), but imperfect estimates may be adequate to make rough assessments of expected consequences of moral acts. Temporal Immediacy The temporal immediacy of the moral issue is the length of time between the present and the onset of consequences of the moral act in question (shorter length of time implies greater immediacy). For example: 1. Releasing a drug that will cause 1 percent of the people who take it to have acute nervous reactions soon after they take it has greater temporal immediacy than releasing a drug that will cause 1 percent of those who take it to develop nervous disorders after 20 years. 2. Reducing the retirement benefits of current retirees has greater temporal immediacy than reducing retirement benefits of employees who are currently between 40 and 50 years of age. Temporal immediacy is a component of the moral intensity construct for two related reasons. First, as economists know well, people tend to discount the impact of events that occur in the future. The value of a dollar today is greater than the value of a dollar promised in two years. The greater the time period, the greater the discount. Hence, the magnitude of consequences will be discounted in accordance with the temporal distance of the predicted effects. Second, as the time period between the act in question and its expected consequences expands, the probability that the act will actually cause the predicted harm declines. Assuming that all else remains constant, additional time creates additional possibilities for moral interventions, by either the moral agent or by another person and, hence, reduces the moral urgency of the immediate problem. Proximity The proximity of the moral issue is the feeling of nearness (social, cultural, psychological, or physical) that the moral agent has for victims (beneficiaries) of the evil (beneficial) act in question. For example: 1. Layoffs in a person's work unit have greater moral proximity (physical and psychological) than do layoffs in a remote plant. 2. For U.S. citizens, the sale of dangerous pesticides in U.S. markets has greater moral proximity (social, cultural, and physical) than does the sale of such pesticides in Latin America. The moral intensity construct includes proximity for intuitive and empirical reasons. Intuitively, people care more about other people who are close to them (socially, culturally, psychologically, or physically) than they do for people who are distant. The words of the Beirut militiaman quoted previ- This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 377 ously are evidence of this claim. The proximity element of the construct is highlighted in the historical novel, Schindler's List, by Thomas Keneally (1982). In his book the author recounts the courageous efforts of the German industrialist Oskar Schindler to move the Jews who had worked at his factory near Crakow to his new factory in Moravia, where he could continue to protect them. The women prisoners had been routed to the death camp at Auschwitz, where camp authorities offered him 300 "fresh" inmates for his factory. Schindler, however, insisted on employing his original workers, despite the fact that most, after weeks in Auschwitz, had lost all value as industrial workers. Empirically, a series of obedience experiments by Milgram (1974) supports the inclusion of proximity in the moral intensity construct. Milgram's subjects ("teachers") were ordered by the experimenter to administer (what the teacher thought were) increasingly powerful shocks to a "learner" (an actor working with the researcher) when the learner failed to answer certain questions correctly. The experiment was designed "to find out when and how people would defy authority in the face of a clear moral imperative" (1974: 4). Milgram found that increased physical proximity of the teacher and the learner significantly reduced the incidence of complete obedience. In an experimental variation that required actual physical contact with the victim, complete subject obedience dropped from 62.5 percent in the baseline condition to 30 percent. Proximity also plays a role in relationships in legal contexts. The legal scholar Charles Fried (1976) argued that not only do attorneys often tend to develop close (proximate) relationships with their clients, but also that these relationships are morally appropriate. He paraphrased Mill (1961) and Sidgwick (1907) in the following sentence: "Our propensity to prefer the interests of those who are close to us is in fact perfectly reasonable because we are more likely to be able to benefit those people" (Fried, 1976: 1067). This utilitarian argument is part of Fried's "lawyer as friend" analogy in justification of zealous pursuit of client interests by attorneys. Proximity seems to be linked to morality in legal relationships as well. Thus, the case for including proximity in the moral intensity construct becomes even more compelling. It must be conceded that proximity is really four variables; that is, social, cultural, psychological, and physical proximity could be separately analyzed. These variables are combined here because of their conceptual similarities and in order to simplify the discussion of components in this exploratory paper. Concentration of Effect The concentration of effect of the moral act is an inverse function of the number of people affected by an act of given magnitude. For example: 1. A change in a warranty policy denying coverage to 10 people with claims of $10,000has a more concentratedeffect than a change denying coverage to 10,000 people with claims of $10.00. 2. Cheating an individual or small group of individuals out of a given sum has a This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 378 Academy of Management Review April more concentrated effect than cheating an institutional entity, such as a corporation or government agency, out of the same sum. Concentration of effect has been included in the moral intensity construct mainly for intuitive reasons. People who have a sense of the paramount importance of justice for the individual (Rawls, 1971)will abhor immoral acts that result in highly concentrated effects. This sentiment is well captured in Ursula LeGuin's "The Ones Who Walked Away from Omelas" (1975), where some inhabitants of a mythical paradise reject a social order that depends on the abject suffering of a single individual. Concentration of consequences is also included in the moral intensity construct for the sake of completeness. MORALINTENSITY AND ITS COMPONENTPARTS Because the intent of this article is to identify some possible components of ethical decision making and behavior for future research, it is impossible to precisely specify (a) the relationships between the moral intensity construct and its components, including their relative importance, and (b) the relationships among the components. Such determinations must be made empirically at a future date. A few comments are in order, however. First, there are two reasons for aggregating these components into a single construct: (a) the components are all characteristics of the moral issue itself and (b) the components are expected to have interactive effects, at least at some levels, as suggested by the expected relationships that will be described in the following sections. Second, moral intensity is generally expected to increase (monotonically) if there is an increase in any one (or more) of its components, and it is expected to decrease if there is a decrease in any one (or more) or its components, assuming the remaining components remain constant. Interactive effects among components are quite likely, however. For example, some threshold of proximity may have to be reached before differences in magnitude significantly affect moral intensity; the precise death toll of violence in Azerbaijan is probably of little consequence to most Americans because the proximity of the event is so low. Indeed, it is expected that threshold levels of all components must be reached before moral intensity begins to vary significantly. Measurement of moral intensity and its components is probably possible only in terms of relatively large distinctions. For example, acts resulting in death will have greater magnitude than acts resulting in injuries, all else being equal. Magnitude of economic harm is an exception, however, where the continuous "money metric" (harm measured in dollars) can be employed. AN ISSUE-CONTINGENT MODEL Because the primary purpose of this article is to introduce a new construct into the discourse on ethical decision making, the underlying framework presented here includes only the major components of ethical decision This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 379 making present in earlier models. Other scholars have adequately made the case for the relevance of other components, and their arguments will be repeated or expanded upon only as they become germane to the formulation presented here. Rest's (1986) four-component model (recognizing moral issues, making moral judgments, establishing moral intent, and implementing moral actions) is a worthy starting point. It is parsimonious, yet it contains all the key elements of moral decision making and behavior. Important contributions by other theorists will be noted and discussed, as the explanation of the issue-contingent model requires. The model is graphically depicted in Figure 2; its component parts and the research propositions derived from it are discussed in the following sections. Much of the theoretical foundation of the issue-contingent model presented here is the complex set of theories and relationships grouped under the general heading of social cognition (Fiske & Taylor, 1984). The elements and processes of social cognition are not fully understood. Models of various processes and relationships overlap, and semantic differences among them An Issue-Contingent FIGURE 2 Model of Ethical Decision Making in Organizations Moral Intensity Magnitude of Consequences Social Consensus Probability of Effect Temporal Immediacy Proximity Concentration of Effect Recognize Moral Issue a Make Establish Moral Moral Judgment Engagein ------0 IBehavior Organizational Factors Group Dynamics Authority Factors Socialization Processes This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions Moral Academy of Management Review 380 April make integration difficult; no single model is universally accepted. This article assumes a simplified model of cognitive processes that retains elements critical to the understanding of single-event moral decision making and eliminates elements that may shape moral decision making over time. The discussion of the model also includes some individual difference variables, which, though important to moral decision making in general, are not vital to the issue-related factors of concern here. In the simplified model assumed here, stimuli from the environment vie for attention through an encoding process. Attention influences attributions, inferences, memory, affect, judgments, intentions, and behavior. Attributions underlie inferences, judgments, intentions, and behavior. Elements of social cognition that are assumed to remain constant over the course of single-event moral decision making include schemata and attitudes. The following analysis points out the effects of moral intensity on various elements of social cognition and, in turn, on moral decision making and behavior. The Moral Issue Human decision-making processes are often activated by the presence of a problem that requires a solution or response and often some form of action (Bazerman, 1986). Moral decision making is no exception; the process begins with a problem, which includes a moral component. The moral component of the problem, or moral issue, can be characterized in terms of its moral intensity. Recognizing Moral Issues For the moral decision-making process to begin, a person must be able to recognize the moral issue. Although many decisions are moral decisions, decision makers do not always recognize the moral element of their decisions. Recall that, according to Velasquez and Rostankowski (1985), moral issues are present if a person's actions, when freely performed, may harm or help others. Therefore, recognizing moral issues involves two elements. A person must recognize that his or her decision or action will affect others (decisions or actions have consequences for human beings) and some choice must be involved (the person has volition). In sum, the person must recognize that he or she is a moral agent. As discussed more fully at the end of this section, a person who fails to recognize a moral issue will fail to employ moral decision-making schemata and will make the decision according to other schemata, economic rationality, for example. Moral intensity will affect the recognition of moral issues through its impact on the individual's recognition of the consequences of decisions. Moral intensity will affect the selective aspects (as opposed to the effort aspects) of attention, that is, salience and vividness. Stimuli are salient to the extent that they stand out from their backgrounds. Moral issues of high intensity will be more salient than those of low intensity because (a) their effects are more extreme (greater magnitude of consequences), (b) their This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 381 effects stand out (higher concentration of effect), or (c) their effects involve significant others (greater social, cultural, psychological, or physical proximity) (Fiske & Taylor, 1984: 187). Stimuli are vivid to the extent that they are emotionally interesting, concrete and imagery provoking, and proximate in a sensory, temporal, or spatial way (Nisbet & Ross, 1980). Fiske and Taylor (1984) argued that, although the empirical case for vividness effects is not strong, vivid stimuli may well dominate pallid stimuli under certain real-world conditions where differential attention is important. Moral issues of high intensity will be more vivid than those of low intensity because (a) their effects are emotionally interesting (greater magnitude of consequences or greater concentration of effect), (b) they are more concrete (more extensive social consensus or higher probability of effect), or (c) they are more proximate, that is, socially, culturally, psychologically, physically (proximity), or temporally (temporal immediacy). In sum, because high-intensity moral issues are salient and vivid, they will be more likely to catch the attention of the moral decision maker and will be recognized as having consequences for others, a vital component of recognizing moral issues. Vivid information also elicits more information from memory than does pallid information. Because the information added from memory is likely to be similar to the vivid stimulus, according to the laws of association (Nisbett & Ross, 1980), inferences of consequences for others are likely to be magnified. Therefore, recognition of moral issues is rendered more probable. A study by Rothbart, Fulero, Jensen, Howard, and Birrell (1978) suggests that magnitude of consequences and salience are linked. These researchers found that subjects judged the proportion of criminals in a sample to be greater if the severity of the individual offenses was greater. Apparently severe offenses were more salient than mild offenses. Recognition of moral issues should be similarly affected by magnitude of consequences. Another factor that may exaggerate the impact of moral intensity on moral judgment and behavior is the tendency of people to utilize simplifying heuristic principles to evaluate the likelihood of uncertain events. In particular, the availability heuristic, in which people use the ease with which instances can be brought to mind to assess the probability of an event, causes people to overestimate the risks present in various situations if these risks are vividly portrayed (Tversky & Kahneman, 1982). Because issues of high moral intensity tend to be more vivid, people will tend to exaggerate the probability of their effect, further heightening the moral intensity and subsequent impact on moral decision making and behavior. The salience of high-intensity moral issues may have a similar effect (Nisbett & Ross, 1980; Taylor, 1982). Volition is another element in the recognition of moral issues; a person must acknowledge that he or she has choice. Heider's (1958) work on perceptions of responsibility for outcomes is relevant here. Heider postulated four levels of responsibility for an individual's actions. The first three (intentional, forseeable, and causal) are of no concern here, but the most This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 382 Academy of Management Review April removed level of responsibility, associational, is relevant. Associational responsibility occurs when a person is held accountable for an action, even though he or she is not causally involved. For example, if, in a fit of anger, a worker verbally abuses his boss and is fired, he may hold a co-worker responsible for failing to restrain him, even though the co-worker saw that he was angry. The co-worker's attribution of personal responsibility will be associational in this case. Moral intensity will influence attributions of associational responsibility, and, hence, perceived volition, in three ways. A person will assume little responsibility (a) if the consequences affect someone psychologically or physically removed from him or her (low psychological and physical proximity), (b) if the consequences are expected to occur in the distant future (low temporal immediacy), and (c) if the consequences are unlikely to occur (low probability of effect). Volition and moral intensity may also be linked by what has been called the defensive attribution hypothesis: greater personal responsibility is attributed to perpetrators of accidents that hold severe, rather than mild, consequences (Fiske & Taylor, 1984). Moral issues of greater magnitude of consequences or concentration of effects would therefore elicit attributions of greater responsibility. Research by Walster (1966) and Shaver (1970a,b) supports the defensive attribution hypothesis. Volition will also be related to self-attributions of freedom. Attributions of freedom are related to "perceptions that an action was chosen from a set of available options and not forced on one by circumstances" (Fiske & Taylor, 1984: 103). Hence, attributions of freedom depend on context. Contextual factors that enhance attributions of freedom include the availability of positive outcome options and certainty on the part of the decision maker regarding a desirable outcome (Wortman, 1975). Two components of moral intensity (magnitude of consequences and social consensus) are relevant here. Moral issues of great magnitude will make clear the availability of positive options (enhanced benefit or decreased harm). High levels of social consensus will enhance the certainty of the decision maker (moral agent) regarding the desirability of the chosen outcome. Thus, moral intensity will influence self-attributions of freedom and, hence, perceived volition. Individual difference variables will also play a role in the recognition of moral issues. Rotter's (1966, 1971) concept of locus of control posits that some people, called internals, credit themselves with substantial control over events, whereas others, called externals, see events as largely under the control of luck, chance, or other individuals. Internals are masters of their fate; externals are pawns of other forces. Locus of control may be related to perceived volition and, hence, to recognition of moral issues. Individuals also differ in terms of their responses to unpleasant events, which moral choices often are. Repressors avoid psychological threats and often ignore the initial signs of unpleasant situations. Sensitizers tend to investigate such threats, mull them over, and explain them intellectually (Fiske & Taylor, 1984). The repression-sensitization dimension also may affect a person's ability to recognize moral issues. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 383 Overarching all of these attributional factors that may be theoretically linked to volition is the actor-observer effect, whereby people rely heavily on situational factors, as opposed to dispositional factors, in explaining their own behavior (Fiske & Taylor, 1984). They tend, in effect, to underestimate their own volition. Presumably, this bias will be constant over moral issues of all levels of intensity; moral intensity will still affect volition differentially as described above. These theoretical and empirical observations lead to Proposition 1. Proposition 1: Issues of high moral intensity will be recognized as moral issues more frequently than will issues of low moral intensity. The recognition of moral issues is related to moral decision making and behavior in an important way. Moral decision making and behavior can be thought of in terms of schemata. Schemata are cognitive structures "that represent organized knowledge about a given concept or type of stimulus. A schema contains both the attributes of the concept and the relationships among the attributes" (Fiske & Taylor, 1984: 140). Event schemata, or scripts, are "structures that describe appropriate sequences of events in well-known situations" (Fiske & Taylor, 1984: 167). Role schemata are norms and behaviors appropriate to certain social roles. The moral decision-making process is an event schema; the moral decision maker is a role schema. The recognition of moral issues triggers schemata that are relevant to moral issues, that is, the moral decision-making process (event) and the moral decision maker (role). Issues of great moral intensity will positively affect the recognition of moral issues and, therefore, will increase the likelihood that moral decision-making schemata will be employed. Moral Development and Moral Judgments Once a person recognizes that a moral issue exists, he or she must make a moral judgment. Component 2 of Rest's (1986) model relates to the question, "How do people make moral judgments?" At this point, Kohlberg's (1976) model of moral development becomes relevant. Kohlberg postulated that human beings make moral judgments in some combination of six analytically distinct ways. Children and morally immature adults have predominantly preconventional orientations. In Stage 1, Obedience and Punishment, the individual obeys rules to avoid punishment; in Stage 2, Instrumental Purpose and Exchange, the individual obeys rules only to further his or her own interests. Most adults operate at conventional levels of moral development. In Stage 3, Interpersonal Accord, Conformity, and Mutual Expectations, the individual adapts to the moral standards of his or her peers; in Stage 4, Social Accord and System Maintenance, the individual adopts the moral standards of society, particularly its laws. Some adults reach a postconventional level of moral development. In Stage 5, Social Contract and Individual Rights, the individual is aware of the relativity of values and upholds rules because they conform to the social contract; in Stage 6, Universal Ethical Principles, the individual chooses his or her own This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 384 Academy of Management Review April ethical principles and follows them, even if they run counter to laws. Rest (1979) subdivided the basic categories somewhat, but mainly followed Kohlberg's model. These authors agreed that individuals vary considerably in terms of their approaches to moral issues and that some form of cognitivedevelopment perspective is appropriate. Although many researchers have regarded moral development as a relatively stable individual difference variable, some, including Rest (1979), argued that many individuals operate within a range of moral development stages. Higgins, Power, and Kohlberg (1984: 103) offered empirical evidence of the effect of context ("moral atmosphere") on individual moral reasoning: "These students did not use their highest or best moral reasoning when thinking about real-life dilemmas in the context of their own schools." Weber (1990) also found evidence that organizational context affects moral reasoning. Levine (1979) argued that individuals may perform at moral development levels lower than their potential and discussed conditions under which this phenomenon may occur. Finally, Trevino (1986) postulated that managers will use lower levels of cognitive moral development in actual work environments compared to hypothetical situations, such as those found on tests designed to measure moral development. These authors have suggested that moral development, or at least the levels at which people actually reason, may be context dependent. This article argues that moral reasoning is issue dependent. The argument is intuitive, theoretical, and empirical. From an intuitive perspective, because moral reasoning takes time and energy (e.g., gathering facts, applying moral principles, and making moral judgments [Velasquez, 1982]), it is likely that moral agents will economize on efforts devoted to moral reasoning when moral stakes are low. Fiske and Taylor (1984: 146) captured this view well in terms of social cognition in general: "People dedicate more effort to social understanding when the stakes are higher. They think more, if not more accurately." Theoretically, social cognition provides a number of supporting perspectives for the postulated link between moral intensity and stage of moral reasoning. Some of these theoretical perspectives are based on the "cognitive miser" principle, whereby people adopt cognitive strategies that simplify complex problems. "The capacity-limited thinker searches for rapid adequate solutions, rather than slow, accurate solutions" (Fiske & Taylor, 1984: 12). Efficiency is stressed at the expense of accuracy. The argument presented here focuses on conditions under which efficiency will be sacrificed for more thorough understanding. A study by Taylor (1975) suggests that the magnitude of consequences of decisions influences the amount of time and information that a person will bring to bear on cognitive processes. People may make use of external cues when stakes are low, but they will rely on self-perception processes more fully when the stakes are higher. Moral cognitive processes should be similarly affected by this component of moral intensity. Weber (1990) presented empirical evidence that suggests a link be- This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 385 tween moral intensity and stage of moral reasoning. He noted that subjects responded differentially to three moral dilemmas that vary in what this article calls moral intensity. Mean stage level for a dilemma that pits human life against obedience to the law (Kohlberg's "Heinz" dilemma) was 3.84; two other dilemmas (of his own design), involving personal career goals versus personal integrity, and professional duty versus obedience to a superior, elicited mean stage scores of 3.22 and 3.35, respectively (1990: 695). Although Weber conceded that organizational context factors might account for part of these differences, he added that "the nature of the moral issue confronting the individual may influence the respondent in determining if postconventional or conventional moral reasoning is appropriate" (1990: 698). This empirical evidence strongly suggests that moral reasoning patterns not only vary from issue to issue, but also may vary in rough proportion to moral intensity as postulated in this article. Issues of high moral intensity may elicit more sophisticated moral reasoning. This empirical evidence and the theoretical considerations discussed above lead to Proposition 2. Proposition 2: Issues of high moral intensity will elicit more sophisticated moral reasoning (higher levels of cognitive moral development) than will issues of low moral intensity. Cognitive moral development has been linked to ethical behavior in several studies. Blasi (1980: 11) reviewed 15 studies of moral reasoning and delinquency and found "a degree of congruency" between moral reasoning and delinquent behavior. Also, with respect to 17 studies on cheating, he determined that "it is not unreasonable to conclude that the hypothesis of a significant positive relation between level of moral thinking and resistance to temptation is supported" (Blasi, 1980: 25). Similarly, he found a positive link between moral stage and resistance to conformity, based on an examination of five studies. Blasi also found that moral development was not conclusively linked to altruistic behavior or "real-life behavior." Overall, he concluded that considerable support exists for a positive link between moral reasoning and moral action. In a review of several studies of moral development and behavior that utilized the Defining Issues Test (Rest, 1979), Thoma and Rest (1986: 135) concluded that "generally there is a link between moral judgment and behavior," although "the strength of the relationship is only moderate." Trevino (1986: 602) strengthened her case for a person-situation interactionist model of ethical decision making in organizations by arguing that cognitive moral development "strongly influences" ethical judgments. Further, she marshaled empirical evidence that linked moral development and cheating (negative relationship) (no reference given), obedience to harmful authority (negative relationship) (Kohlberg, 1969), and helping behavior (positive relationship) (Kohlberg & Candee, 1984) to support her contention that moral development and ethical behavior are linked. This evidence This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 386 Academy of Management Review April helped to validate her interactionist model. Similarly, it should help to validate the issue-contingent model presented here. Moral Intent Once a person has made a moral judgment, a process that is dependent on his or her cognitive moral development (Kohlberg, 1976; Rest, 1986), he or she must decide what to do. A decision about what is morally "correct," a moral judgment, is not the same as a decision to act on that judgment, that is, to establish moral intent. The term intent is functionally equivalent to the word intentions, which is found in some of the social psychology literature (e.g., Fishbein & Ajzen, 1975). At this stage, the moral agent balances moral factors against other factors, notably including selfinterest (Rest, 1986). For example, a supervisor may determine that refusing to fire a senior employee is the "right" thing to do (a moral judgment), but may decide to fire him or her anyway (failure to establish moral intent) for reasons of career advancement or organizational pressures. In his autobiographical account of the "Aircraft Brake Scandal," Kermit Vandivier (1972) never considered "blowing the whistle" on his own company, even though the firm was about to deliver a dangerously unsafe product. He knew what was "right," but intended to do nothing meaningful about it. Moral intensity may also play a role in establishing moral intent. Attributions of responsibility are related to perceived control over an event; the greater the perceived control, the greater the attributed responsibility (Fiske & Taylor, 1984). Proximity, an element of moral intensity, is likely to affect perceived control and, in turn, attributions of responsibility. For example, a person usually cannot be held responsible for events that are physically distant. Because people seek to avoid negative attributions of responsibility, they will establish positive moral intent more frequently when the moral issue is proximate. Further, this desire to avoid aversive consequences will influence moral intent in situations where social consensus is high. Because low social desirability of behavior often reveals underlying dispositions (Jones & McGillis, 1976), people will attempt to behave appropriately when the social consensus is high regarding the desirability of certain moral behavior. Moral intensity may also influence moral intent profoundly through its influence on affect (emotions, feelings, and mood). Stimuli that are vivid and salient often heighten emotions and feelings, which, in turn, intensify both cognitive and behavioral responses (Fiske & Taylor, 1984). In particular, people may commit themselves to (moral) actions (establish moral intent) while emotionally excited, and they may retain the commitment even after the intense feelings have died down (Bryant & Zillman, 1979). The moral commitment established in an emotionally aroused state may carry through to actual moral behavior, the next step in the process. Indeed, affect may be a critical element in overcoming the organizational impediments to moral action (discussed in a later section). These theoretical observations lead to Proposition 3. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 387 Proposition 3: Moral intent will be established more frequently where issues of high moral intensity are involved than where issues of low moral intensity are involved. The establishment of moral intent is important to the moral decisionmaking and behavior model presented here because intentions are important determinants of behavior. Fishbein and Ajzen argued that "the best predictor of a person's behavior is his intention to perform the behavior" (1975: 381), especially when the intention and behavior are measured at the same level of specificity. Moral intent and moral behavior should be similarly related. Indeed, Dubinsky and Loken (1989) and Hunt and Vitell (1986) explicitly included a link between moral intent and moral behavior in their models. Moral Behavior The fourth component of Rest's model involves acting on a person's moral intentions, that is, engaging in moral behavior. In Rest's words, "Executing and implementing a pla[n] of action ... involves ... working around impediments and unexpected difficulties, overcoming fatigue and frustration, resisting distractions and allurements, and keeping sight of the original goal" (1986: 15). Establishing moral intent is not enough; colloquially, "The road to Hell is paved with good intentions." Social cognition is also useful in establishing theoretical links between moral intensity and moral behavior. Attributions of controllability influence how people respond to others, particularly those in need of help (Weiner, 1979, 1980). A person is more inclined to help someone whose predicament is uncontrollable than one whose predicament is controllable through, for example, increased effort. Help is less likely to be forthcoming to those who are responsible for their predicaments. Similar research by Ickes and Kidd (1976) on the effect of intentionality on helping behavior and Berkowitz (1969) on the effect of internal (personal) versus external (environmental) factors on helping behavior supports this conclusion. These findings, when coupled with research results examined by Burger (1981), also link moral intensity with behavior. Burger concluded that subjects attribute less responsibility to perpetrators of accidents as the severity of consequences increased when subjects and perpetrators were situationally and personally similar. When subjects and perpetrators were dissimilar, more responsibility was attributed to perpetrators as accident severity increased. These findings suggest that proximity (personal and situational similarity), a component of moral intensity, is negatively linked to attributions of responsibility. If attributions of responsibility are negatively linked to helping behavior, as discussed in this section, it is likely that proximity will positively influence helping behavior. Even attributional biases may favorably affect the link between moral intensity and behavior. The fundamental attribution error is the tendency to attribute the behavior of others to dispositional qualities, as opposed to situational factors. People tend not to see that situational forces influence the This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 388 Academy of Management Review April behavior of others; instead, they see the behavior of others as freely chosen. "Victims of situational forces may be held more accountable for their situations than they should be" (Fiske & Taylor, 1984: 74). In sharp contrast, people explain their own behavior more in situational terms and attribute it much less to dispositional factors. In sum, they hold themselves less accountable for their behavior than they would hold others in the same situation. This self-serving attributional bias is thought to be related to two factors: differential salience (a person cannot observe him- or herself behaving) and differential information (Fiske & Taylor, 1984). The latter of these factors is relevant here. People know their own attitudes, feelings, and intentions toward an event and, hence, can attribute their behavior to situational factors about which outside observers have no knowledge. The observer therefore tends to attribute behavior more to dispositional factors than does the actor. This phenomenon is called the actorobserver effect (Fiske & Taylor, 1984). Moral intensity may play a role in the distribution of attributions between dispositional and situational factors in the following way. Information about a person and a situation will be influenced significantly by proximity. People tend to know more about other people and situations that are proximate (socially, culturally, psychologically, and physically) compared to people and situations that are distant. Greater knowledge would tend to reduce the gap between actor attributions (the actor him- or herself being the ultimate in proximity) and observer attributions. Hence, the tendency to make dispositional attributions would decline as proximity increased; greater knowledge would increase the incidence of situational attributions. Further, as discussed in this section, people are more inclined to help those whose predicaments are uncontrollable (attributed to situational factors) compared to those whose predicaments are controllable (attributed to dispositional factors). It follows that proximity, a component of moral intensity, will positively influence helping behavior. Kelman and Hamilton, in their examination of destructively obedient behavior, Crimes of Obedience (1989), argued that an individual's tendency to challenge authority is based on the interplay of two opposing forces: binding forces, which tend to reinforce the authority structure, and opposing forces, which heighten resistance to authority. When opposing forces are stronger than binding forces, the person will tend to challenge authority. Important among these opposing forces are physical and psychological distance; as these distances increase, opposing forces decrease. In moral intensity terms, proximity is inversely related to distance. Therefore, as proximity (physical and psychological) increases, opposing forces increase, challenges to authority become more probable, and the incidence of immoral behavior in authority situations should decrease. Milgram's (1974) obedience studies also demonstrated the importance of physical distance to obedience. His subjects resisted authority more fully when the physical distance between learner and teacher was reduced (ultimately, to actual contact). This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 389 Kilham and Mann (1974), using a variant of the Milgram experiment, divided the teacher's role into two parts: a person who was a "transmitter" and a person who was an "executant." The transmitter informed the executant when to push the switch. Obedience among transmitters was significantly higher than that among executants. This result lends further credence to the importance of proximity in moral behavior; the hierarchically more proximate executants showed greater moral restraint. These empirical findings, coupled with the theoretical arguments discussed above, lead to Proposition 4. Proposition 4: Ethical behavior will be observed more frequently where issues of high moral intensity are involved than where issues of low moral intensity are involved. BIASESIN ASSESSINGMORALINTENSITY The preceding discussion has highlighted the differential effects of various cognitive processes on moral issues of varying intensity. Some cognitive processes will affect moral decision making and behavior in general, without regard to the moral intensity of the issue itself. Some of these cognitive processes create biases against the recognition of moral issues and, hence, against the engagement of moral decisionmaking processes. Included in this first group is the inability of people to conceptualize events that have not occurred (Ross & Anderson, 1982). In the case of moral issues, where events are often prospective, bias in risk perception (Slovic, Fischoff, & Lichtenstein, 1982) may be particularly important; people may fail to recognize risky future situations because they are less imaginable. According to Nisbett and Ross, "Such 'null' information tends to be overlooked and underappreciated" (1980: 48). Individuals are also poor at detecting covariation. True covariation plays a limited role in the perception of covariation in the social domain; preexisting theories play a much larger role (Jennings, Amabile, & Ross, 1982; Nisbett & Ross, 1980). The failure to detect covariation may affect a person's ability to recognize early symptoms of various problems (Jennings, Amabile, & Ross, 1982), a particularly important element in moral decision making. Even when covariation is detected, its causes may be difficult to analyze correctly; "prior theories of causality may override the implications of the covariation pattern" (Nisbett & Ross, 1980: 10). Prior theories may not include moral elements, thus biasing the causal inference process against moral considerations. Individuals may tend not to perceive themselves as independent agents in moral situations. Kelman and Hamilton (1989), in their analysis of the Milgram (1974) obedience studies, argued that some subjects felt that they did stop administering shocks, even though their hands were still manipulating the switches. These subjects regarded the decisions and the responsibility for those decisions as being entirely in the hands of the experimenter. This ceding of responsibility was total in some cases; one subject could not This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 390 Academy of Management Review April imagine conditions under which he would have stopped administering shocks (Kelman & Hamilton, 1989). Apparent ceding of responsibility was also present in the responses of subjects in a study conducted by Derry (1987). Many (33%) of her subjects reported that they never faced a moral conflict at work, despite evidence to the contrary. Some of this group of subjects accepted the organization's authority structure in moral matters, whereas others saw their roles as employees as limited to matters of efficiency and effectiveness. Clearly, not all people see themselves as independent moral agents in work situations. Bias favorable to the recognition of moral issues may also be present in cognitive processes. Langer (1982) discussed the illusion of control, wherein people overestimate their personal control in situations involving substantial chance. Where moral issues are involved, this tendency would bias individuals toward attributions of personal responsibility and away from situational attributions. In essence, this tendency would bias moral agents toward judgments of personal volition, an important element of moral decision making and behavior. On balance, the net effect of these biases in cognitive processes on moral decision making is unknown. However, they are expected to affect all moral situations equally and not apply differentially to moral issues of differing intensity. Organizational Factors Organizational settings present special challenges to moral agents. Moral decision making and behavior at the individual level, though often difficult, at least are not complicated by major organizational factors. Trevino discussed several such factors under the heading of "situational variables" (1986: 603). Ferrell and Gresham (1985), Ferrell, Gresham, and Fraedrich (1989), and Hunt and Vitell (1986) included organizational factors in their models under significant others and opportunity, organizational culture, and organizational environment, respectively. Higgins, Power, and Kohlberg (1984) provided empirical evidence that moral atmosphere affects moral reasoning and moral judgment. Smith and Carroll (1984) presented a detailed argument that organizational factors often create impediments to individual ethical behavior. In their view, socialization processes, environmental influences, and hierarchical relationships collectively constitute a "stacked deck," which impedes moral behavior. Research conducted on small group conformity behavior (Asch, 1951, 1955, 1956), obedience to authority (Milgram, 1963, 1974), and groupthink (Janis, 1972) also suggests that organizational factors may distort the ethical intentions of individuals. Kelman and Hamilton (1989) described the dynamics of challenging organizational authority in considerable detail. In order to overcome macrolevel obstacles to challenging authority (those anchored in institutional and social structures), an individual must redefine the authority relationship and the demand as illegitimate. Conflicts between individual morality and This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 391 macro-level authority are often acted out on the micro level, however. At this level, the interplay of binding forces (immediate presence of authority, consequences of disobedience, fear of embarrassment, and the actions of comparable others) and opposing physical and psychological forces (proximity of the victim and perceptions of personal causation) help determine the individual's behavior (Kelman & Hamilton, 1989). Organizational factors are likely to play a role in moral decision making and behavior at two points: establishing moral intent and engaging in moral behavior. Implicit organizational pressures may be sufficient to determine the person's moral intent (e.g., the previous Vandivier [1972] example). Explicit organizational factors may cause unethical (or ethical) behavior to result despite good (or bad) intention. CONCLUSIONSAND IMPLICATIONS Existing theoretical models have ignored the effect of characteristics of the moral issue itself on ethical decision making and behavior in organizations. Taken at face value, these models suggest that individuals will decide and behave in the same manner regardless of the nature of the moral issue involved. An employee of a drug manufacturer would view the release of a dangerous drug by his or her firm with the same alarm (or lack of alarm) that he or she viewed the theft of a few diskettes from the company supply cabinet by a fellow employee. The issue-contingent model proposed here explicitly rejects this view and suggests that the moral intensity of the issue itself has a significant effect on moral decision making and behavior at all stages of the process. If this model is found to have empirical support, the testing of other models would be significantly affected. Controlling for issue traits would become an integral part of a meaningful test of Trevino's (1986) person-situation interactionist model, for example; the relative importance of personal factors and situational factors might vary considerably, from issue to issue. Similarly, issue characteristics could alter the balance of teleological and deontological considerations in the moral evaluation stage of Hunt and Vitell's (1986) general theory model of marketing ethics. Perhaps the most important potential impact of an empirical finding that ethical decision making and behavior are issue contingent involves the applicability of the models themselves. Moral intensity is expected to play a major role in the recognition of moral issues and, hence, in the actual engagement of moral decision-making processes instead of, or in addition to, other decision-making schemata. Simply stated, the details of moral decision-making and behavior processes become irrelevant if the person does not recognize that he or she is dealing with a moral issue. Future models of ethical decision making should include some consideration of the effect of the moral agent's failure to recognize the moral issue. Moral intensity is also relevant to the general applicability of Kohlberg's (1976) theory of cognitive moral development. If moral development is issue contingent, as this article and some emerging empirical evidence suggest, then Kohlberg's theory would have to be substantially revised, and much of This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 392 Academy of Management Review April the research based on it would have to be reappraised. Future research based on his developmental theory would have to control for traits of the moral issues involved. From a practical point of view, issue contingency is important to normative judgments of moral decisions and of the people who make them. Many of the elements of moral intensity (magnitude of consequences, probability of effect, temporal immediacy, and concentration of effect) are directly related to judgments of the importance of moral issues. If these elements of moral intensity are found to be positively linked to moral behavior, it can be concluded that people generally behave better when the moral issue is important than they do when it is unimportant. Regardless of a person's views regarding the overall moral tenor of society or its alleged decline in recent years, he or she could easily be encouraged by the finding that people's best moral behavior is inspired by issues of substantial importance. REFERENCES Asch, S. E. 1951.Effectsof group pressure upon the modificationand distortionof judgments. In H. S. Guetzkow (Ed.), Groups, leadership and men: 177-190. Pittsburgh:Carnegie Press. Asch, S. E. 1955.Opinions and social pressure. Scientific American, 193:31-35. Asch, S. E. 1956. Studies of independence and conformity:I. A minority of one against a unanimous majority.Psychological Monographs, 70(9),No. 416: 1-70. Bazerman, M. H. 1986.Judgment in managerial decision making. New York:Wiley. Beauchamp, T. L., &Bowie, N. E. (Eds.). 1979.Ethical theory and business. Englewood Cliffs, NJ:Prentice-Hall. Berkowitz,L. 1969. Roots of aggression: A reexamination of the frustration-aggression hypothesis. New York:Atherton. Blasi, A. 1980.Bridging moral cognition and moral action: A critical review of the literature. Psychological Bulletin, 88: 1-45. Brommer,M., Gratto,C., Gravender, J., & Tuttle,M. 1987.A behavioral model of ethical and unethical decision making. Journal of Business Ethics, 6: 265-280. Bryant,J., &Zillman,D. 1979.Effectof intensificationof annoyance through unrelated residual excitationon substantiallydelayed hostile behavior. Journal of Experimental Social Psychology, 15:470-480. Burger, J. M. 1981. Motivationalbiases in the attributionof responsibilityfor an accident: A metaanalysis of the defensive-attributionhypothesis. Psychological Bulletin, 90: 496-512. Cavanagh, G. F., Moberg, D. J., & Velasquez, M. 1981.The ethics of organizational politics. Academy of Management Review, 6: 363-374. Cloward, R. A., & Ohlin, L. E. 1960.Delinquency and opportunity. Glencoe, IL:Free Press. Derry, R. 1987.Moral reasoning in work-relatedconflicts. In W. C. Frederick& L. E. Preston (Eds.), Research in corporate social performance and policy, vol. 9: 25-49. Greenwich, CT:JAIPress. Dubinsky,A. J., & Loken,B. 1989.Analyzing ethical decision making in marketing.Journal of Business Research, 19(2):83-107. Ferrell, 0. C., & Gresham, L. G. 1985. A contingency frameworkfor understanding ethical decision making in marketing.Journal of Marketing, 49(3):87-96. Ferrell, 0. C., Gresham, L. G., &Fraedrich,J. 1989.A synthesis of ethical decision models for marketing.Journal of Macromarketing,9(2):55-64. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 393 Fishbein, M., & Ajzen, I. 1975.Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA:Addison-Wesley. Fiske, S. T., & Taylor, S. E. 1984.Social cognition. New York:Random House. Frederick,W. C. (Ed.). 1987.Research in corporate social performance and policy (vol. 9). Greenwich, CT:JAIPress. Fried, C. 1976.The lawyer as friend:The moral foundationsof the lawyer-client relation. Yale Law Journal, 85: 1060-1091. Fritzsche, D. J. 1988.An examination of marketingethics: Role of the decision maker, consequences of the decision, management position, and sex of the respondent. Journal of Macromarketing,8(2):29-39. Fritzsche,D. J., & Becker, H. 1983.Ethical behavior of marketingmanagers. Journal of Business Ethics, 2: 291-299. Garrett,T. M. 1966.Business ethics. Englewood Cliffs,NJ:Prentice-Hall. Hegarty, W. H., & Sims, H. P., Jr. 1978.Some determinantsof unethical decision behavior: An experiment. Journal of Applied Psychology, 63: 451-457. Hegarty, W. H., & Sims, H. P., Jr. 1979. Organizational philosophy, policies and objectives related to unethical decision behavior: A laboratoryexperiment.Journal of Applied Psychology, 64: 331-338. Heider, F. 1958.The psychology of interpersonal relations. New York:Wiley. Higgins, A., Power, C., & Kohlberg, L. 1984.The relationshipof moral atmosphere to judgments of responsibility.In W. M. Kurtines&J. L. Gewirtz(Eds.),Morality, moral behavior and moral development: 74-106. New York:Wiley. Hunt,S. D., &Vitell,S. 1986.A general theoryof marketingethics. Journalof Macromarketing, 6(1):5-16. Ickes, W. J., & Kidd, R. F. 1976. Attributionalanalysis of helping behavior. In J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New directions in attribution research, vol. 1: 311-334. Hillsdale, NJ:Erlbaum. Janis, I. L. 1972.Victims of groupthink. Boston:HoughtonMifflin. Jennings, D. L., Amabile, T. M., &Ross, L. 1982.Informalcovariationassessment: Data-based versus theory-based judgments. In D. Kahneman, P. Slovic, &A. Tversky(Eds.),ludgment under certainty: Heuristics and biases: 211-230. Cambridge: Cambridge University Press. Jones, E. E., & McGillis,D. 1976.Correspondentinferences and the attributioncube: A comparative appraisal. In J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New directions in attribution research, vol. 1: 389-420. Hillsdale, NJ:Erlbaum. Jones, T. M. 1980.Corporatesocial responsibilityrevisited, redefined. California Management Review, 22(3):9-67. Kahneman, D., Slovic, P., &Tversky,A. (Eds.). 1982.Judgment under certainty: Heuristicsand biases. Cambridge: Cambridge UniversityPress. Kaplan, R. D. 1989.Afghanistan postmortem.The Atlantic, April:26, 28, 29. Kelman, H. C., &Hamilton,V. L. 1989.Crimes of obedience. New Haven, CT:Yale University Press. Keneally, T. 1982.Schindler's list. New York:Penguin Books. Kilham, W., & Mann, L. 1974.Level of destructiveobedience as a functionof transmitterand executant roles in the Milgram obedience paradigm. Journal of Personality and Social Psychology, 29: 696-702. Kohlberg, L. 1969. Stage and sequence: The cognitive-developmental approach to socializa- This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 394 Academy of Management Review April tion. In D. A. Goslin (Ed.),Handbook of socialization theory and research: 347-480. Chicago: Rand McNally. Kohlberg,L. 1976.Moralstages and moralization:The cognitive-developmentapproach. In T. Lickona (Ed.), Moral development and behavior: Theory, research and social issues: 31-53. New York:Holt, Rinehart& Winston. Kohlberg,L., &Candee, D. 1984.The relationshipof moral judgmentto moral action. In W. M. Kurtines&J. L. Gewirtz(Eds.),Morality,moral behavior, and moral development: 52-73. New York:Wiley. Laczniak,G. R., &Inderrieden,E. J. 1987.The influence of stated organizationalconcern upon ethical decision making. Journal of Business Ethics, 6: 297-307. Langer, E. J. 1982. The illusion of control. In D. Kahneman, P. Slovic, & A. Tversky(Eds.), Judgment under uncertainty: Heuristics and biases: 231-238. Cambridge: Cambridge UniversityPress. LeGuin, U. K. 1975.The ones who walked away from Omelas. The wind's twelve quarters: 275-284. New York:Harper & Row. Levine, C. G. 1979. Stage acquisition and stage use: An appraisal of stage displacement explanations of variation in moral reasoning. Human Development, 22: 145-164. Merton,R. K. 1957.The role set. BritishJournalof Sociology, 8(June):106-120. Milgram,S. 1963.Behavioralstudy of obedience. Journalof Abnormaland Social Psychology, 67: 371-378. Milgram, S. 1974.Obedience to authority. New York:Harper & Row. Mill,J. S. 1961.Utilitarianism.In M. Cohen (Ed.),Thephilosophy of John StuartMill. New York: ModernLibrary. Nehemkis, P. 1975. Business payoffs abroad: Rhetoricand reality. California Management Review, 18(2):5-20. Newsweek. 1989.Beirutdays: Life and death. November 13:58-59. Nisbett, R., & Ross, L. 1980.Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs, NJ:Prentice-Hall. Packer, H. L. 1968. The limits of the criminal sanction. Stanford, CA: Stanford University Press. Rawls, J. 1971.A theory of justice. Cambridge, MA:HarvardUniversityPress. Rest, J. R. 1979.Development in judging moral issues. Minneapolis:Universityof Minnesota Press. Rest, J. R. 1986.Moral development: Advances in research and theory. New York:Praeger. Ross, L., & Anderson, C. A. 1982.Shortcomingsin the attributionprocess: On the origins and maintenance of erroneous social assessments. In D. Kahneman, P. Slovic, & A. Tversky (Eds.),Judgment under uncertainty: Heuristics and biases: 129-152. Cambridge: Cambridge UniversityPress. Rothbart,M., Fulero, S., Jensen, C., Howard, J., & Birrell,B. 1978.From individual to group impressions:Availability heuristics in stereotype formation.Journal of Experimental Social Psychology, 14:237-255. Rotter,J. B. 1966. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs:General and Applied, 80(1),Whole No. 609: 1-28. Rotter,J. B. 1971.Externalcontroland internalcontrol.Psychology Today, June:37-42, 58-59. Schwab, D. P. 1980.Constructvalidity in organizationalbehavior. In B. M. Staw &L. L. Cummings (Eds.),Research in organizational behavior, vol. 2: 3-43. Greenwich, CT:JAIPress. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions 1991 Jones 395 Shaver, K. G. 1970a.Defensive attribution:Effectsof severity and relevance on the responsibilityassigned foran accident. Journalof Personality and Social Psychology, 14:101-113. Shaver, K. G. 1970b. Redress and conscientiousness in the attributionof responsibility for accidents. Journal of Experimental Social Psychology, 6: 100-110. Sidgwick, H. 1907.The methods of ethics (7thed.). Indianapolis:Hackett. Slovic, P., Fischoff, B., & Lichtenstein,S. 1982.Facts versus fears: Understanding perceived risk. In D. Kahneman, P. Slovic, & A. Tversky(Eds.),Judgment under uncertainty: Heuristics and biases: 463-489. Cambridge: Cambridge UniversityPress. Smith, H. R., & Carroll,A. B. 1984.Organizationalethics: A stacked deck. Journal of Business Ethics, 3: 95-100. Sutherland, E., & Cressey, D. R. 1970.Principles of criminology (8th ed.). Chicago: Lippincott. Taylor, S. E. 1975. On inferring one's own attitudes from one's behavior: Some delimiting conditions. Journal of Personality and Social Psychology, 31: 126-131. Taylor, S. E. 1982.The availability bias in social perception and interaction.In D. Kahneman, P. Slovic, & A. Tversky(Eds.),Judgment under uncertainty: Heuristics and biases: 190200. Cambridge: Cambridge UniversityPress. Thoma, S. J., &Rest, J. R. 1986.Moraljudgment, behavior, decision making, and attitudes. In J. R. Rest (Ed.), Moral development: Advances in research and theory: 133-175. New York:Praeger. Trevino,L. K. 1986.Ethicaldecision making in organizations:A person-situationinteractionist model. Academy of Management Review, 11:601-617. Tversky,A., & Kahneman, D. 1982.Judgmentunder uncertainty:Heuristicsand biases. In D. Kahneman, P. Slovic, & A. Tversky(Eds.),ludgment under uncertainty: Heuristics and biases: 3-20. Cambridge: Cambridge UniversityPress. Vandivier, K. 1972.The aircraftbrake scandal. Harper'sMagazine, April:45-52. Velasquez, M. G. 1982.Business ethics: Concepts and cases. Englewood Cliffs, NJ:PrenticeHall. Velasquez, M. G., &Rostankowski,C. 1985.Ethics:Theoryand practice. Englewood Cliffs,NJ: Prentice-Hall. Walster, E. 1966. Assignment of responsibility for an accident. Journal of Personality and Social Psychology, 3: 73-79. Weber, J. 1990.Managers' moral reasoning: Assessing their responses to three moral dilemmas. Human Relations, 43: 687-702. Weiner, B. 1979. A theory of motivationfor some classroom experiences. Journal of Educational Psychology, 71: 3-25. Weiner, B. 1980.Human motivation. New York:Holt, Rinehart& Winston. Wirtenberger,H. J. 1962.Morality and business. Chicago: Loyola UniversityPress. Wortman, C. B. 1975. Some determinants of perceived control. journal of Personality and Social Psychology, 31: 282-294. York, K. M. 1989. Defining sexual harassment in workplaces: A policy-capturing approach. Academy of Management Journal, 32: 830-850. Thomas M. Jones received his Ph.D. from the University of California-Berkeley. He is an associate professor of organization and environment at the University of Washington. His current research interests include ethical decision-making models, the effect of educational programs on student ethics, and business and society paradigms. This content downloaded from 2.176.192.83 on Wed, 22 Jan 2014 23:06:35 PM All use subject to JSTOR Terms and Conditions