doi:10.1016/j.obhdp.2005.09.001 Organizational Behavior and Human Decision Processes 100 (2006) 110–127 www.elsevier.com/locate/obhdp Justice and personality: Using integrative theories to derive moderators of justice eVects � Jason A. Colquitt a,¤, Brent A. Scott a, Timothy A. Judge a, John C. Shaw b a University of Florida, USA b Mississippi State University, USA Received 29 July 2004 Available online 10 November 2005 Abstract Although organizational justice has been shown to have behavioral consequences, there remains a surprising amount of variation in how individuals react to fair and unfair treatment. The present study drew on three integrative theories in the justice literature— fairness heuristic theory, uncertainty management theory, and fairness theory—to identify personality traits that could explain such variation. From these theories, we identiWed trust propensity, risk aversion, and morality (rooted in circumplex models of personal- ity) as potential moderators. A laboratory study provided some support for the prediction that the three traits moderate the eVects of procedural, interpersonal, and distributive justice on task performance and counterproductive behavior. The moderating eVects of the three traits explained more variance in the outcomes than moderators based in the justice literature (equity sensitivity, sensitivity to befallen injustice) or the Wve-factor model of personality. Taken together, the results suggest that the three integrative theories can inform the search for personality-based moderators of justice eVects. © 2005 Elsevier Inc. All rights reserved. Keywords: Organizational justice; Fairness; Trust; Personality Fair treatment impacts an organization’s members in a number of ways. It provides evidence that organiza- tional authorities are trustworthy, reducing fears of exploitation while enhancing the legitimacy of organiza- tional actions (Lind, 2001; Tyler & Lind, 1992; Van den Bos, 2001a). Fair treatment also makes future events more predictable and controllable, reducing some of the uncertainty experienced in day-to-day working life (Lind & Van den Bos, 2002; Thibaut & Walker, 1975). Finally, fair treatment signals an adherence to moral and ethical standards on the part of authorities, potentially bringing � An earlier draft of this manuscript was presented in T.A. Judge and J.A. Colquitt (Chairs), The how and why of fairness: Mediators/modera- tors of justice eVects. Symposium conducted at the annual meeting of the Society for Industrial and Organizational Psychology, Chicago, IL. * Corresponding author. Fax: +1 352 392 6020. E-mail address: colquitt@uX.edu (J.A. Colquitt). 0749-5978/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.obhdp.2005.09.001 more meaning to working life (Cropanzano, Byrne, Bob- ocel, & Rupp, 2001; Folger, 1998). The members of an organization can evaluate fairness along a number of dimensions. Distributive justice refers to the perceived fairness of decision outcomes and is judged by gauging whether rewards are proportional to costs (Homans, 1961), whether outcomes adhere to expectations (Blau, 1964), and whether outcome/input ratios match those of a comparison other (Adams, 1965). Procedural justice refers to the perceived fairness of deci- sion-making procedures and is judged by gauging whether procedures are accurate, consistent, unbiased, and correctable (Leventhal, 1980), and open to employee input (Thibaut & Walker, 1975). Interactional justice refers to the perceived fairness of the enactment or implementation of procedures (Bies & Moag, 1986) and has two subfacets. Interpersonal justice captures the sin- cerity and respectfulness of authority communication, mailto: colquitt@ufl.edu mailto: colquitt@ufl.edu mailto: colquitt@ufl.edu mailto: colquitt@ufl.edu J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 111 while informational justice concerns the use of honest and adequate explanations for decisions (Colquitt, 2001; Greenberg, 1993a). These justice dimensions have been associated with a number of behavioral reactions on the part of employees (for meta-analytic reviews, see Cohen-Charash & Spec- tor, 2001; Colquitt, Conlon, Wesson, Porter, & Ng, 2001), including counterproductive behavior (Green- berg, 1990, 1993b, 2002; Skarlicki, Folger, & Tesluk, 1999) and task performance (Konovsky & Cropanzano, 1991; Van den Bos, Vermunt, & Wilke, 1996). However, meta-analytic results suggest that a substantial amount of variation exists in these relationships, and that moder- ators could explain much of that variation (Colquitt et al., 2001). Indeed, the relationships between the justice dimensions and performance are some of the most inconsistent eVects in the literature. The purpose of the present study was to identify per- sonality-based moderators that could explain some of the inconsistencies in the eVects of justice on task perfor- mance and counterproductive behavior. According to the interactional psychology perspective, behavior is determined by a complex interplay of personal and situ- ational variables such that personality alters the cogni- tive construction of an individual’s environment and shapes the meaning of the various responses to that envi- ronment (Schneider, 1983). Applied to the study of reac- tions to fair and unfair treatment, this perspective would acknowledge that personality alters individuals’ percep- tions of their treatment while also shaping the cognitive and behavioral reactions triggered by those perceptions. Thus, we focus on personality moderators of justice eVects, though both personal and situational diVerences could potentially serve as boundary conditions. If personality variables indeed are capable of explain- ing variation in justice reactions, the key question becomes which personality traits are most worthy of study. To make that critical choice, we drew on the set of theories that has been introduced in the justice literature over the past Wve years: fairness heuristic theory (Lind, 2001; Van den Bos, 2001a), uncertainty management theory (Lind & Van den Bos, 2002; Van den Bos & Lind, 2002), and fairness theory (Folger & Cropanzano, 2001). These theories represent what Colquitt, Greenberg, and Zapata-Phelan (2005) termed the “integrative wave” of the justice literature, described as “a phase in which scholars began building models and theories that exam- ined the eVects of multiple justice dimensions in combi- nation” (p. 35). The theories are termed “integrative” because, unlike other theories that focus on only one type of justice (e.g., equity theory, Adams, 1965), the integrative theories consider multiple forms of justice in concert. Because these three integrative theories capture much of the current thinking on why justice matters to people and why it impacts their behavior, we reasoned that the mechanisms in those theories could highlight potentially impactful moderators. Moreover, given that Lind (2001) suggested that the most relevant dependent variables in fairness heuristic theory are those variables that capture the distinction between cooperative and antisocial behaviors, and Folger and Cropanzano (2001) argued that fairness theory can explain the presence ver- sus absence of retaliatory behaviors, we felt that task performance and counterproductive behavior were appropriate outcomes for examining personality moder- ators of justice eVects. Past research on moderation of justice eVects Distributive justice research Concerns about individual diVerences in justice reac- tions are not new. In one of the Wrst reviews of the sub- ject, Major and Deaux (1982) stated that research on individual diVerences in justice behavior could be divided into two categories: (a) individual diVerences in reward allocation decisions, and (b) individual diVer- ences in reactions to inequity. The authors noted that the study of personality moderators of equity reactions was relatively rare. However, such work was bolstered by the introduction of equity sensitivity (Huseman, HatWeld, & Miles, 1987), a construct intended to capture sensitivity to diVerences in outcome/input ratios. Equity sensitivity can be conceptualized as a continuum ranging from “Benevolents” to “Entitleds.” Huseman et al. (1987) originally conceptualized Benevolents as individuals who prefer their outcome/input ratios to be less than a comparison other’s (underreward) and Entitleds as indi- viduals who prefer their outcome/input ratios to be greater than a comparison other’s (overreward). King, Miles, and Day (1993) redeWned Benevolents as having greater tolerance for underreward, with Entitleds having greater tolerance for overreward. In between Benevo- lents and Entitleds are “Equity Sensitives,” who adhere to Adams’s (1965) original conceptualization of equity reactions—experiencing distress when their ratios diVer in either direction. Procedural justice research Though the introduction of procedural justice pro- vided a potential new direction for research on personal- ity moderators of justice eVects, very few studies have pursued this direction. Early studies examined traits such as locus of control (Sweeney, McFarlin, & Cotton, 1991) and delay of gratiWcation (Joy & Witt, 1992) but did not explore their moderating eVects on attitudinal or behavioral reactions. However, subsequent studies have explored personality traits as moderators of both behav- ioral and attitudinal reactions to procedural justice. Sch- mitt and Dorfel (1999) found that sensitivity to befallen 112 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 injustice (SBI) moderated the relationship between pro- cedural justice and self-reported health. Individuals high in SBI are expected to more frequently recall unjust events, become more angry about them, be more likely to be preoccupied by them, and have a stronger urge for punitivity (Schmitt, Neumann, & Montada, 1995). Other studies have drawn from various theories to identify personality moderators of procedural justice eVects such as self-esteem (Brockner et al., 1998) and exchange ideology (Witt, Kacmar, & Andrews, 2001), the latter reXecting an employee’s expectation of (and likely response to) exchange relationships with his or her organization (Eisenberger, Cotterell, & Marvel, 1987). In addition, Hagedoorn, Buunk, and van de Vliert (2002) examined the belief in a just world as a moderator of procedural and distributive justice eVects, and a recent study by Van den Bos, Maas, Waldring, and Semin (2003) found support for the moderating role of aVect intensity on emotional reactions to procedural and dis- tributive justice. Interactional justice research The introduction of interactional justice provided another new direction for research on personality mod- erators of justice eVects. We are aware of only two stud- ies that have explored moderators of such eVects. Skarlicki et al. (1999) found that the combination of low interactional justice and low distributive justice was more likely to result in retaliation when individuals were high in negative aVectivity and low in agreeableness. Heuer, Blumenthal, Douglas, and Weinblatt (1999) reported that interactional justice had a stronger eVect on fairness perceptions for high esteem individuals, a result which provided support for the relational model of justice (Tyler & Lind, 1992). Applying integrative justice theories to personality moderators of justice eVects The articles discussed in the prior section illustrate that scholars have begun to identify personality modera- tors of justice eVects. However, two points should be noted about the extant literature. First, few studies have explored justice moderators with behavioral reactions, with outcomes instead consisting of either fairness per- ceptions (e.g., Joy & Witt, 1992; Sweeney et al., 1991) or attitudinal outcomes (e.g., Schmitt & Dorfel, 1999; Witt et al., 2001). Second, studies of moderators of procedural and interactional justice eVects remain relatively rare, with more work focusing on distributive eVects. To move the literature examining personality mod- erators of justice eVects forward, we believe it would be useful to establish a more direct linkage to the cur- rent state of theorizing in the justice literature. As noted above, fairness heuristic theory, uncertainty management theory, and fairness theory capture the current thinking about organizational justice issues. With their focus on multiple justice dimensions, these theories could hold potential for identifying personal- ity moderators that cut across more than one type of justice. Creating a more direct linkage between research on moderators of justice eVects and these integrative theories could have two primary beneWts. First, the theories could be used to establish a concep- tual framework that explains how and why a given trait could alter an individual’s response to fair or unfair treatment. Second, the theories could be used to guide the critical question of which traits to examine next, given the potentially endless list of traits avail- able for study. A brief review of integrative justice theories Fairness heuristic theory Fairness heuristic theory, which grew out of earlier work on the relational model of justice (Tyler & Lind, 1992), suggests that individuals in organizations are con- tinually faced with the “fundamental social dilemma” (Lind, 2001; Van den Bos, 2001a). Although cooperating with organizational agents can lead to better outcomes in the long term, it also raises the potential of exploita- tion. To cope with that dilemma, individuals use a “fair- ness heuristic”—a psychological shortcut used to decide whether to cooperate with authorities. Lind (2001) argued that the fundamental social dilemma highlights the importance of trust, where trust is deWned as a willingness to accept vulnerability to another based on positive expectations of that person’s intentions and actions (Mayer, Davis, & Schoorman, 1995). Unfortunately, the trustworthiness of authorities can be diYcult to judge, as it depends on assessments of unobservable concepts like integrity, benevolence, and ability. In contrast, fairness perceptions depend on rela- tively more observable phenomena such as met expecta- tions (Blau, 1964), consistency of procedures (Leventhal, 1980), and respectfulness of communication (Bies & Moag, 1986). Thus, fairness heuristic theory argues that justice is used as a proxy for trust, with fair treatment signaling a trustworthy authority (Lind, 2001; Van den Bos, 2001a). The theory suggests that the fairness heuristic is formed quickly during a “judgmental phase” using whatever fairness information is Wrst gathered or is most interpretable. Once the heuristic has been formed, the theory argues that individuals will use it as a proxy for trust to guide day-to-day actions (see Van den Bos, Wilke, & Lind, 1998a, for empirical support), a period termed the “use phase” (Lind, 2001). Individuals con- tinue to employ the heuristic until a “phase shifting event,” such as a particularly important or unexpected J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 113 change at work, causes the individual to reconsider fair- ness levels and return to the judgmental phase. Uncertainty management theory In subsequent work, Lind and Van den Bos deempha- sized uncertainty about trust per se, instead focusing on more general forms of uncertainty. For example, Van den Bos (2001b) asked participants to describe the emo- tions they typically feel when they are uncertain or not in control. Results showed that justice had a stronger eVect on reactions when uncertainty was high than when uncertainty was low. Such studies led to the formal introduction of uncertainty management theory, which was cast as a successor to fairness heuristic theory (Lind & Van den Bos, 2002; Van den Bos & Lind, 2002). Uncertainty management theory recognizes that many aspects of work and family life may contain uncer- tainty. According to the theory, fairness can remove trust-related uncertainty and mitigate the discomfort associated with other forms of uncertainty—even if they have nothing to do with authorities. The authors sum- marized this key tenet as follows: “What appears to be happening is that people use fairness to manage their reactions to uncertainty, Wnding comfort in related or even unrelated fair experiencesƒ” (Lind & Van den Bos, 2002, p. 216). Fairness theory Fairness theory argues that individuals engage in counterfactual thinking to determine the fairness of a particular event and whether authorities should be blamed for that event (Folger & Cropanzano, 2001). More speciWcally, an individual makes three diVerent counterfactual determinations: would, could, and should. An individual assesses what would have happened when he or she imagines plausible alternative states of being to the current situation. An individual assesses what could have happened when he or she determines whether events were under the discretion of an authority. When an individual assesses what should have happened, he or she considers whether the authority’s behavior was mor- ally appropriate. According to the theory, individuals typically react to a decision-making event by Wrst judging whether welfare has been reduced or threatened. If individuals determine that an injury has occurred, the next question is whether the authority can be blamed. That judgment of blame depends on the presence of other feasible alternatives, as authorities cannot be blamed if they had no control over their choice of actions. Folger and Cropanzano (2001) further suggested that blame can only be placed if some ethical principal of social conduct has been violated. The characteristics of fairness theory can be distin- guished from fairness heuristic theory and uncertainty management theory on a number of levels. First, the counterfactual thinking described in fairness theory is triggered by a discrete event, whereas the fairness heuris- tic is used and maintained across a number of events. Sec- ond, the process of judging fairness is more deliberate in fairness theory, relative to the almost subconscious short- cuts described in the other two theories. Third, the mech- anisms that guide behavioral reactions are decidedly diVerent. With its emphasis on blame, fairness theory is ideally suited to explaining counterproductive reactions (Folger & Cropanzano, 2001). In contrast, fairness heu- ristic theory’s emphasis on cooperation makes it more relevant to explaining prosocial behaviors (Lind, 2001). Which traits worthy of study can be derived from the integrative theories? The integrative theories reviewed above can be used to create a conceptual framework that describes how and why a given personality trait could moderate the eVects of certain justice dimensions. Each of the theories considers three distinct questions in some form: (a) Why do individuals care about justice—what triggers justice concerns in the Wrst place? (b) How carefully and inten- sively do individuals ruminate on justice information once their concerns have been triggered? (c) What actions could individuals take once they believe that unfairness has been experienced? As summarized in Fig. 1, these three questions provide three diVerent mechanisms that could be aVected by an individual’s personality. Consider a personality trait that makes individuals more sensitive to justice, causes them to ruminate on jus- tice information more deliberately, and increases the likelihood that they will react to injustice with some action. That trait should amplify the eVects of justice on behavioral reactions, and that result should be fairly robust across contexts because it is being realized through three distinct mechanisms. Judge and Larsen (2001) used similar logic in their stimulus–organism– response (S–O–R) model of personality inXuences on job satisfaction. They argued that personality variables should have a signiWcant eVect on job satisfaction to the extent that they aVect sensitivity to environmental stim- uli, inXuence the cognitive and emotional processing of those stimuli within the organism, and heighten the like- lihood of an eventual response. The subsequent sections of this manuscript review traits, derived from the inte- grative theories, that are likely to inXuence more than one of the mechanisms shown in Fig. 1. Trust propensity Given the central role of trust in fairness heuristic the- ory (Lind, 2001; Van den Bos, 2001a), trust propensity stands out as a personality variable that could moderate justice eVects. Trust propensity is a generalized expecta- tion about the trustworthiness of others (Mayer et al., 1995), which should impact all three of the mechanisms 114 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 in Fig. 1. Suspicious individuals should attend more to their environment when forming trust judgments, and fairness concerns should be more easily triggered as sus- picions about ability, benevolence, and integrity govern day-to-day interactions (Mayer et al., 1995). Suspicious individuals should also ruminate on fairness-relevant information more deliberately, as the use of the fairness heuristic continues until an important event causes a “phase shift” that bumps the individual back to the judgmental phase (Lind, 2001). The threshold for phase shifting should be lower for suspicious individuals given their generally wary reactions to stimuli. If so, individu- als low in trust propensity will more frequently engage in careful analysis of justice information. Finally, individuals low in trust propensity should be more likely to respond to unfair treatment with behav- ioral repercussions. Lind (2001) noted that “the fairness heuristic is an imprecise algorithm for deciding what is the best thing to do” (p. 66). Relying on the heuristic during the use phase will inevitably cause individuals to gloss over daily or weekly Xuctuations in actual fairness levels. Suspicious individuals, on the other hand, will be unlikely to gloss over such Xuctuations. Their more care- ful and frequent analysis of actual fairness levels should allow their behaviors to be based on a cognitive ledger of treatment versus contributions. Fairness heuristic theory deemphasizes the distinc- tions among the various justice dimensions, arguing that the timing and interpretability of fairness experiences is more relevant than their actual content. Nevertheless, we would argue that trust propensity will be more likely to moderate procedural and interactional justice eVects, as opposed to distributive eVects. Models in the trust litera- ture explicitly recognize procedural principles like con- sistency, bias suppression, and ethicality as forms of integrity (Mayer et al., 1995). Those same models also recognize interactional principles like respectfulness, supportiveness, and openness as forms of benevolence (Mayer et al., 1995). In contrast, distributive principles have much less presence in trust theorizing. Trust pro- pensity should therefore be more likely to impact the consideration of (and reaction to) procedural and inter- actional information because these justice dimensions tap trustworthiness issues to a greater extent. We there- fore predicted: Hypothesis 1. Propensity to trust moderates the eVects of (a) procedural justice, and (b) interactional justice, on task performance and counterproductive behavior, such that the relationships are stronger for individuals low in trust propensity and weaker for individuals high in trust propensity. Risk aversion The shift from fairness heuristic theory to uncertainty management theory supplanted trust as a central con- cern in favor of more general uncertainty. In discussing the importance of uncertainty, Lind and Van den Bos (2002) suggested that “tolerance of risk” could be an important determinant of the importance of justice (p. 215). The personality variable that most closely captures such tolerance is risk aversion. Risk aversion is an Fig. 1. Conceptual model of personality moderators of justice eVects. J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 115 individual diVerence that captures diVerential attention to stimuli in potentially risky situations, along with the tendency to react to risk with anxiety and eventual with- drawal (e.g., Cable & Judge, 1994; Maehr & Videbeck, 1968). Though risk levels are, to a large extent, driven by situational characteristics, highly risk-averse individuals view and react to those situations diVerently than less risk-averse individuals. Risk aversion should be associated with increased sensitivity to justice concerns. Because risk-averse indi- viduals react to uncertainty with more anxiety, they should pay close attention to any environmental cues that can help them manage or mitigate such uncertainty (Lind & Van den Bos, 2002). That same sense of caution should prompt risk-averse individuals to resist relying on an “imprecise algorithm” like the fairness heuristic (Lind, 2001) in favor of a careful, reasoned analysis of authority behavior. The linkage between risk aversion and the likelihood of behavioral response is less straight- forward. It seems likely that risk-averse individuals will respond to an injustice with a behavioral response that does not create a great deal of anxiety. Because risk averse individuals react to adverse stimuli with more fre- quent withdrawal (e.g., Cable & Judge, 1994; Maehr & Videbeck, 1968), passive forms of retaliation (like decreased eVort or task performance) could be likely. Individuals’ likelihood of engaging in more active retali- ation (like counterproductive behavior) should depend on the risk involved in such acts. When the risk involved is low, counterproductive behavior should provide another means of retaliation for risk-averse individuals. Our predictions for trust propensity were limited to procedural and interactional justice because of their spe- cial relevance to trust—does the same type of distinction need to be made for risk aversion? Van den Bos and Lind (2002) deWned uncertainty quite broadly, framing it as the unpredictability of future events that deprives one of con- Wdence about what to expect from the environment. Unfair procedures should create an especially high level of uncertainty because procedures have a systemic char- acter—they remain in place over the long-term (Sweeney & McFarlin, 1993). Unfair treatment by an authority Wgure also has long-term implications on uncertainty, so long as the authority remains in his or her position for an extended period of time. In contrast, concerns about unfair outcome distributions are more short-term in ori- entation, as outcomes tend to have a discrete, episodic character. Nevertheless, distributive unfairness deprives an individual of conWdence in how to behave and what to expect. UnfulWlled expectations (Blau, 1964), inconsistent social comparisons (Adams, 1965), and divergence in rewards and costs (Homans, 1961) prevent the develop- ment of the behavior-outcome contingencies that are endemic to so many models of volitional behavior (Kan- fer, 1991). We therefore made risk aversion moderation predictions for all three justice dimensions: Hypothesis 2. Risk aversion moderates the eVects of (a) procedural justice, (b) interactional justice, and (c) dis- tributive justice on task performance and counterpro- ductive behavior, such that the relationships are stronger for individuals high in risk aversion and weaker for indi- viduals low in risk aversion. Morality In the years since fairness theory was Wrst introduced, a great deal of attention has focused on the should com- ponent of the theory, which captures whether an author- ity’s actions have complied with moral or ethical principles. This attention can be seen in the “moral vir- tue” and “deontic” perspectives on justice, which argue that justice is valued because it signiWes adherence to prevailing moral standards (Cropanzano et al., 2001; Folger, 1998, 2001; Folger, Cropanzano, & Goldman, 2005). The “deontic” term comes from the Greek word deon, meaning duty or obligation (Folger, 2001). Folger et al. (2005) argued that violations of moral principles can trigger “deontic anger,” which may prompt retalia- tory behaviors even when such actions are not rational (Folger et al., 2005). Of course, there are likely to be individual diVerences in the degree to which individuals experience deontic anger, as well as their responses to such violations. Fol- ger (1998) hinted that an individual’s morality might be one variable that captures such variation. Within the personality literature, trait morality can be operationally deWned as a combination of high agreeableness and high conscientiousness. Hofstee, de Raad, and Goldberg (1992) introduced an integration of the Big Five and cir- cumplex approaches to trait structure. The authors described 10 diVerent circumplex models, each of which illustrates various combinations (or blends) of two Big Five factors. Taken together, these 10 circumplex models were used to provide operational deWnitions for 540 spe- ciWc traits. One of those traits was “moral”, which was positioned in the high agreeableness, high conscientious- ness area of one of the circumplex models (see also de Raad, Hendriks, & Hofstee, 1992). Other studies have similarly shown that trait morality, and related concepts such as trait integrity, seem to represent a blend of high agreeableness and high conscientiousness (Ones, 1994; Saucier & Goldberg, 1996). The circumplex origin of trait morality is important in that the intersection of agreeableness and conscien- tiousness is believed to capture a behavior pattern not explainable by simple additive eVects of those two Big Five dimensions. However, we should acknowledge that this trait-based version of morality diVers from Kohl- berg’s (1984) conceptualization of the stages of moral development. A trait-based view emphasizes static or cross-sectional diVerences in morality across individuals, whereas Kohlberg’s (1984) emphasis was on within-indi- vidual transitions over time (Walker, 2002). To highlight 116 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 this critical diVerence and reduce potential confusion, our manuscript intentionally uses the term ‘‘trait moral- ity” when describing our moderator variable. High levels of trait morality should be associated with increased sensitivity to justice concerns. From a moral virtue perspective, justice can help bring meaning to working life (Cropanzano et al., 2001; Folger, 1998). Highly moral individuals may be more likely to search for that meaning and should also be less likely to engage in unjust behaviors themselves. They should therefore react to others’ instances of injustice with more deontic anger, leading to more deliberate rumination on justice- relevant information. In a fairness theory sense, moral individuals should have a heightened interest in the should counterfactual portion of the blame assessment process (Folger & Cropanzano, 2001). The linkage between trait morality and the likelihood of behavioral response would seem to depend on the nature of the speciWc reaction. On the one hand, highly moral individuals are particularly likely to hold others responsible for their actions, as morality is associated with a tendency to ascribe responsibility (Zuckerman & Reis, 1978) and sanctions often follow from ascriptions of blame (Pizarro, Uhlmann, & Bloom, 2003). Moreover, Folger et al. (2005) suggested that retribution against the source of an injustice is itself a “moral remedy” that is consistent with social exchange theories of interaction (Blau, 1964). Bies and Tripp (1995) went so far as to describe revenge as a “moral imperative,” noting that perpetrators of revenge often report a strong belief that they were “doing the right thing” (pp. 258–259). On the other hand, individuals high in trait morality should be very resistant to the prospect of engaging in counterproductive behaviors like theft or retaliation. Indi- rect support for this assertion can be drawn from a series of studies by Turillo, Folger, Lavelle, Umphress, and Gee (2002). The studies gave participants the opportunity to either self-sacriWcially punish an individual who had exhibited unfair intentions in a past phase of the study, or self-sacriWcially reward an individual who had exhibited fair intentions. Punishments and rewards occurred using a payout matrix, where participants chose their own payout while also allocating payouts to others. Importantly, the results of one of the studies revealed that participants decided to punish past transgressions only when those actions did not harm the payouts received by other cur- rent participants. The authors wrote, “These results sup- port our reasoning that individuals will only engage in retributive justice to punish unfairness in the past when, by doing so, they are not themselves being unfair in the present.” (p. 852). Because a counterproductive response means “being unfair in the present,” a moral individual should be more likely to respond to injustice with less task-related eVort as opposed to more overt forms of retaliation. A reduction in performance can amount to a brand of “civil disobedience” that is relatively harmless to others. As a result, our hypotheses for trait morality are limited to task performance. As with the other integrative theories, fairness theory emphasizes the commonalities among the diVerent forms of justice rather than the diVerences. For example, the would counterfactual considers an event’s aversiveness, with no particular importance given to whether the event is distributive, procedural, or interactional in nature (Folger & Cropanzano, 2001). The could counter- factual assesses whether other sequences of events or actions might have been feasible, again with little emphasis on modalities of justice. However, discussions of the should counterfactual and writings on the moral virtue and deontic approaches to justice do seem to pri- oritize procedural and interactional justice violations (Folger, 1998, 2001; Folger & Cropanzano, 2001). Proce- dural and interactional justice rules, such as bias, ethical- ity, respectfulness, and honesty seem more “morally charged” than distributive concepts like met expecta- tions, reward-cost proportionality, and outcome/input ratio comparisons. Indeed, in Folger’s (2001) “deontic diVerentiation” of distributive, procedural, and interac- tional justice (pp. 22–25), he speculated that moral accountability for justice tends to grow stronger in ascending order from distributive to procedural to inter- actional (see also Turillo et al., 2002). We therefore made moderation predictions for only the latter two dimen- sions of justice: Hypothesis 3. Trait morality moderates the eVects of (a) procedural justice and (b) interactional justice on task performance, such that the relationships are stronger for individuals high in trait morality and weaker for individ- uals low in trait morality. Alternative approaches to moderation of justice eVects Though the hypotheses advanced in Fig. 1 are derived from the three theories that form the core of the so-called “integrative wave” of the justice literature (Colquitt et al., 2005), there are other potential approaches to the study of moderation of justice eVects that have been used in past research. One approach is the study of personality traits that are speciWcally designed to moderate justice eVects, as in the cases of equity sensitivity (Huseman et al., 1987) and sensitivity to befallen injustice (SBI; Sch- mitt et al., 1995). Both traits might have moderating potential, though research on equity sensitivity is limited to distributive justice eVects and research on SBI has yet to extend beyond the tests by Schmitt and colleagues (e.g., Schmitt & Dorfel, 1999; Schmitt et al., 1995). We therefore included equity sensitivity and SBI so that their moderating eVects could be compared to the eVects of trust propensity, risk aversion, and trait morality. A second alternative approach to moderation of jus- tice eVects is the Wve-factor model of personality J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 117 (Costa & McCrae, 1992; Goldberg, 1990). The “Big Five” dimensions of conscientiousness, agreeableness, neuroticism, openness to experience, and extraversion subsume many of the narrow traits shown in Fig. 1. The Big Five may also moderate justice eVects, as research by Skarlicki et al. (1999) showed that agreeableness and neuroticism moderated the eVects of speciWc justice com- binations. However, as Hogan and Roberts (1996) noted, broad traits are not always best suited for narrow predictions, and the reaction to an injustice is clearly a more speciWc criterion relative to day-to-day job perfor- mance. Nevertheless, we also included the Big Five as alternative moderators of justice eVects to provide a sec- ond frame of comparison for the eVects predicted in Fig. 1. Method Sample Participants were 238 undergraduates from a large, southeastern university recruited from an introductory management course. Females composed 45% of the sam- ple. In exchange for participation, participants were given course credit and earned a $5 cash prize. Procedure Upon entering the laboratory, participants were seated at one of Wve tables. Each table had a pen holder containing seven expensive-looking pens. The tables were arranged so that participants’ views of one another were obstructed during the session. The experimenter provided a cover story, informing participants that the purpose of the study was to validate assessment tools that would be used in the future hiring of research assis- tants. Included in these assessment tools were a proof- reading task and a reading comprehension task, and participants were told that their performance on the Wrst of these (the proofreading task) would determine whether they would receive a $5 cash prize. Participants were told that around 66% of the participants would perform well enough to receive the cash prize. This instruction ensured that all participants began the exper- iment with the same expectations for being rewarded. Participants were then told that they could choose a pen to use during the study and that they could keep the pen they chose. They were then given seven minutes to complete the proofreading task, which contained a pas- sage describing various types of work groups (e.g., for- mal, informal, virtual). Next, the experimenter informed participants that he would leave the room to grade the tasks and then would return to give the participants feedback on their performance. During the grading period, participants completed a questionnaire assessing the Big Five, risk aversion, trust, and equity sensitivity, which contained the following instructions: “This ques- tionnaire assesses various aspects of your personal and general attitudes. Please answer the questions in each section as honestly as you can.” Unless otherwise noted, all items on all questionnaires utilized a 5-point scale, where 1 D Strongly Disagree and 5 D Strongly Agree. After 10 min, the experimenter called each partici- pant individually to another room to deliver the justice manipulations. Procedural, distributive, and interac- tional justice were then manipulated by the experi- menter via eight possible statements, resulting in a 2 £ 2 £ 2 between-subjects design with participants randomly assigned to conditions. Table 1 provides the exact wording of the justice manipulations. The proce- dural manipulation varied Leventhal’s (1980) accuracy and consistency criteria for justice in a manner similar to Van den Bos, Bruins, Wilke, and Dronkert (1999). Interactional justice was manipulated using the inter- personal justice subfacet, which captures the sincerity and respectfulness of authority communication (Col- quitt, 2001; Greenberg, 1993a). The remainder of this manuscript uses the interpersonal justice label in the interest of precision. SpeciWcally, the manipulation var- Table 1 Summary of justice manipulations Justice dimension Level Text of manipulation Procedural High In the past, I’ve always graded the whole proofreading task in order to be as accurate and consistent as possible. So I took the time to carefully grade all the paragraphs that you corrected. Low In the past, I’ve always graded the whole proofreading task in order to be as accurate and consistent as possible. I didn’t do that here though. I Wnished grading everyone else’s but ran out of time on yours, so I just graded the last paragraph. Interpersonal High I understand that students are very busy, and there’s a lot of other things you could be doing right now besides helping us out. We really appreciate your time. Thanks a lot. Low Whatever. I don’t give a damn whether you get paid or not. That’s the way it is. I’ve got better things to do than grade these things. Distributive High Anyway, (from that one paragraph/based on those paragraphs) I have determined that you should be paid. Low Anyway, (from that one paragraph/based on those paragraphs) I have determined that you should not be paid. 118 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 ied the respect and rudeness components of the con- struct (Bies & Moag, 1986). Distributive justice was manipulated by meeting or not meeting participants’ expectations about the receipt of the $5 reward, similar to the approach used by Greenberg (1993b) and Van den Bos, Wilke, Lind, and Vermunt (1998b). Partici- pants were either paid immediately, thereby meeting expectations, or were told they would not be paid, thereby failing to meet expectations. Next, participants were given 10 min to complete a reading comprehension test, which served as the measure of task performance. This test consisted of a short pas- sage describing carpal tunnel syndrome followed by eight questions which required participants to examine the speciWc details and main points of the passage to obtain the correct answers. Using reading comprehen- sion as a measure of task performance is appropriate in this context because the test was designed to be largely eVort-driven as opposed to ability-driven. The questions were not diYcult but did require a careful perusal of the passage on the part of the participant. SpeciWcally, we included a brief paragraph at the end of the passage that corrected previous statements. To answer the questions correctly, participants had to take the time to carefully read the entire passage. Participants then completed a Wnal questionnaire that contained items for the manipu- lation checks, sensitivity to befallen injustice, and trait morality. Near completion of the Wnal questionnaire, the experi- menter interrupted and stated: “Um, I have to make a phone call. When you guys are done just leave your ques- tionnaire on your desk and you’re free to go. Oh, also: I know I read from the script earlier that you could keep the pen, but we’re running some more sessions and it looks like we’re getting a little low. So, if you would not keep the pen I’d appreciate it.” The experimenter then left the room, giving participants the opportunity to take pens from the pen holder. The set up of the room prevented participants from being able to access the pen containers of others, so they could only take pens from their individ- ual pen holder. When participants exited the room, the experimenter asked each one to wait for the others. Once all had exited, the experimenter then brought the partici- pants into another room to debrief them about the true nature of the experiment. Finally, participants who did not receive the $5 during the study were paid. Integrative theory traits Trust propensity We measured trust propensity using 5 items from the International Personality Item Pool (2001). The items included: “I trust others,” “I trust what people say,” “I am wary of others (R),” “I suspect hidden motives in others (R),” and “I distrust people (R).” CoeYcient � for this scale was .81. Risk aversion Risk aversion was also assessed using 6 items from the International Personality Item Pool (2001). Items included: “I enjoy being reckless (R),” “I take risks (R),” “I seek danger (R),” “I seek adventure (R),” I would never go hang-gliding or bungee jumping,” and “I would never make a high risk investment.” CoeYcient � for this scale was .82. Trait morality This trait was measured using 6 items from the Inter- national Personality Item Pool (2001). The items included: “I would never cheat on my taxes,” “I turn my back on others (R),” I scheme against others (R),” “I act at the expense of others (R),” “I respect the privacy of others,” and “I respect authority.” CoeYcient � for this scale was .86. ConWrmatory factor analysis We conducted a conWrmatory factor analysis on the measures of the three integrative theory traits to deter- mine whether the items seemed to be tapping three dis- tinct constructs. We included a reverse-wording factor with loadings on the items worded in the opposite direc- tion from the majority of the scale, given that responses to these sorts of items can be inXuenced by careless respondents (Schmitt & Stults, 1985). The trait morality scale was balanced in wording, so the items reXecting a lack of morality were loaded onto the wording factor. We also allowed the errors for the two “I trustƒ” items to covary as their item similarity resulted in an observed correlation signiWcantly higher than the model repro- duced correlation. The results of a three-factor model provided an acceptable Wt to the data (�2 (108) D 174.50; CFI D .954; RMR D .048; RMSEA D .052). The loadings for the three personality factors were all statistically sig- niWcant and averaged as follows: trust propensity (.62), risk aversion (.63), and trait morality (.59). The correla- tions among the three latent variables were moderate in magnitude, with an absolute average of .25. Taken together, these results support the assertion that the scales are measuring three distinct factors. Alternative moderators Equity sensitivity We measured equity sensitivity using the 5-item scale developed by King and Miles (1994). Participants were asked to divide 10 points between two opposing state- ments by “giving the most points to the choice that is most like you and the fewest points to the choice that is least like you.” All Wve items began with the phrase “In any organization I might work for:” An example item is “It would be more important for me to (A) get from the organization or (B) give to the organization.” The instrument was scored by adding items reXecting J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 119 Entitledness; thus high scores represent a high sensitivity to perceived underreward. The potential range of scores was zero to 50. CoeYcient � for this scale was .77. Sensitivity to Befallen injustice We measured SBI using the 10-item scale developed by Schmitt and Maes (2000). A sample item is: “It both- ers me when others receive something that ought to be mine.” CoeYcient � for this scale was .87. Big Five dimensions We measured the Wve-factor model using the Big Five Inventory (BFI; John, Donahue, & Kentle, 1991), with dimensions as follows: neuroticism (eight items, e.g., “I can be moody,” � D .86), extraversion (eight items, e.g., “I have an assertive personality,” � D .89), openness to experience (10 items, e.g., “I am curious about many things,” � D .86), agreeableness (nine items, e.g., “I have a forgiving nature,” � D .75), and conscientiousness (eight items, e.g., “I do a thorough job,” � D .80). Outcome measures Task performance Task performance was assessed by scoring each par- ticipant’s answers to the reading comprehension test. Task performance ranged from a possible score of zero (no correct answers) to eight (all answers correct). Counterproductive behavior We measured counterproductive behavior by the number of pens taken from each participant’s pen holder upon leaving the experiment (pen holders held seven pens in total). Counterproductive behavior ranged from 0 (no pens taken) to 3 (the maximum number of pens taken by a participant). The act of tak- ing a pen when asked not to do so likely taps two diVerent counterproductive behaviors that range in severity. At its most severe, taking a pen represents a form of theft. At its least severe, taking a pen repre- sents a form of intentional noncompliance with an authority Wgure. Manipulation checks We assessed whether participants perceived the jus- tice manipulations as intended using three items for each dimension of organizational justice. The procedural jus- tice items included “The grading procedures used on the proofreading task seemed accurate.” (� D .94). The inter- personal justice items included “The experimenter spoke to me with sincerity and respect.” (� D .97). The distribu- tive justice items included “I received the $5 reward for the proofreading task.” (� D .96). We also veriWed that the manipulations aVected perceptions of fairness by including the following four-item global fairness percep- tions scale (e.g., “In general, this was a fair experiment”) (� D .96). Results Manipulation checks A MANOVA with the three manipulation check scales as dependent variables revealed signiWcant main eVects for all three manipulations (F D 903.40, p < .001, for distributive justice; F D 36.24, p < .001, for interper- sonal justice; F D 10.10, p < .001, for procedural justice) with no interaction eVects. ANOVA results showed that the distributive manipulation had a strong eVect on the distributive check (F D 2724.93, p < .001, �2 D .92, M D 4.67 vs. 1.21) and weaker eVects on the procedural (F D 20.85, p < .001, �2 D .08, M D 3.17 vs. 2.50) and inter- personal checks (F D 23.03, p < .001, �2 D .09, M D 3.99 vs. 3.33). The interpersonal manipulation had a strong eVect on the interpersonal check (F D 92.65, p < .001, �2 D .29, M D 4.25 vs. 2.84), a weaker eVect on the procedural check (F D 6.13, p < .01, �2 D .03, M D 2.99 vs. 2.58), and no eVect on the distributive check (F D .12, p D .73, �2 D .00, M D 2.77 vs. 2.85). The procedural manipulation had a strong eVect on the procedural check (F D 30.56, p < .001, �2 D .12, M D 3.20 vs. 2.36), a weaker eVect on the interpersonal check (F D 10.42, p < .001, �2 D .04, M D 3.89 vs. 3.33), and no eVect on the distributive check (F D .40, p D .53, �2 D .00, M D 2.84 vs. 2.76). All three manipulations also inXuenced global perceptions of fair- ness, explaining a total of 17% of the variance (F D 16.07, p < .001). Unique eVects were as follows: distributive jus- tice (B D .67, p < .001), procedural justice (B D .28, p < .05), and interpersonal justice (B D .54, p < .001). The ANOVA results above indicate some spillover, with a manipulation of one form of justice aVecting checks of another, which is likely due to attempting orthogonal manipulations of variables that are so highly correlated on a perceptual level (Cohen-Charash & Spector, 2001; Colquitt et al., 2001). However, the degree of spillover does not seem serious enough to impair an unambiguous evaluation of the experiment’s results (Perdue & Summers, 1986). The F values for the intended eVects were 53 times stronger than the F values for the unintended eVects, and the �2 values for the intended eVects were 12 times stronger than �2 values for the unintended eVects. Nevertheless, we statistically con- trolled for all spillover by partialling out the contaminat- ing variance from each justice manipulation. For example, we partialled out the shared variance between the procedural manipulation and the interpersonal check by regressing the latter onto the former and using the residual as the new procedural manipulation vari- able. This process resulted in a completely “clean” set of ANOVA results. We should emphasize that this form of 120 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 statistical control did not alter the tests of any of our hypotheses but is presented here to provide as clear a picture of our results as possible. Results for integrative theory traits Table 2 provides the descriptive statistics and zero- order correlations among the variables. The moderated regressions used to test our hypotheses are shown in Table 3. The Wrst two steps assess the main and interac- tive eVects of the justice manipulations, which did not have main or interactive eVects on task performance, with the main eVects falling within the conWdence inter- vals from Colquitt et al.’s (2001) meta-analytic review. The justice manipulations did have signiWcant main and interactive eVects on counterproductive behavior, con- sistent with past research (Greenberg, 1990, 1993b, 2002), with the procedural and distributive manipula- tions reducing counterproductive behavior (though the latter eVect only approached signiWcance at p < .07). The interaction eVects showed that the combination of low procedural justice and either low interpersonal or low distributive justice created particularly high counterpro- ductive behavior levels. The third step of the regressions assesses the main eVects of the three personality traits, with none reaching statistical signiWcance. The fourth step of the regressions enters the justice £ personality product terms used to test the hypotheses. Consistent with the recommendations of Cohen, Cohen, West, and Aiken (2003), we mean-cen- tered all moderators before computing the product terms. The set of justice £ personality product terms explained signiWcant incremental variance in both task performance and counterproductive behavior. Hypothe- sis 1a and b predicted stronger eVects for procedural and interpersonal justice on task performance and counter- productive behavior when trust propensity was low. The interpersonal justice £ trust propensity interaction approached signiWcance for task performance, with the pattern in the predicted direction (p < .07, see Fig. 2). Though not predicted, the results also revealed a similar interaction with distributive justice for task performance (p < .06, see Fig. 2). A signiWcant interpersonal justice £ trust propensity interaction was also observed for counterproductive behavior, but was opposite to Table 3 Moderated regression results for trust propensity, risk aversion, and trait morality Note. n D 238. �R2 values may not sum exactly to R2 values due to rounding error. ¤ p < .05, two-tailed. ¤¤ p < .10, two-tailed. Regression step Task performance Counterproductive behavior R2 �R2 � R2 �R2 � 1. Procedural Justice (PJ) .00 .00 .05 .05¤ .05¤ ¡.19¤ Interpersonal Justice (IJ) .02 ¡.08 Distributive Justice (DJ) .00 ¡.12¤¤ 2. PJ £ IJ .01 .01 ¡.05 .10¤ .06¤ .20¤ PJ £ DJ ¡.04 .15¤ IJ £ DJ ¡.09 .07 3. Trust Propensity .03 .02 .00 .11¤ .01 ¡.02 Risk Aversion .11 .04 Trait Morality ¡.10 ¡.06 4. PJ £ Trust Propensity .12¤ .09¤ ¡.01 .23¤ .12¤ ¡.11 IJ £ Trust Propensity ¡.14¤¤ ¡.21¤ DJ £ Trust Propensity ¡.14¤¤ ¡.05 PJ £ Risk Aversion .14¤ ¡.22¤ IJ £ Risk Aversion ¡.05 ¡.19¤ DJ £ Risk Aversion ¡.06 ¡.08 PJ £ Trait Morality ¡.11 .11 IJ £ Trait Morality .25¤ ¡.02 DJ £ Trait Morality .24¤ ¡.00 Table 2 Descriptive statistics and zero-order correlations Note. n D 237. SBI D Sensitivity to Befallen Injustice. ¤ p < .05, two-tailed. Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1. Procedural Justice .00 .49 — 2. Interpersonal Justice .00 .49 ¡.09 — 3. Distributive Justice .00 .48 ¡.08 ¡.12 — 4. Trust Propensity 3.79 .54 .05 .01 ¡.05 — 5. Risk Aversion 3.08 .66 ¡.00 .12 .08 .09 — 6. Trait Morality 4.15 .46 .04 .10 .00 .47¤ .40¤ — 7. Equity Sensitivity 22.45 6.12 ¡.06 .07 .04 ¡.23¤ .08 ¡.22¤ — 8. SBI 2.92 .65 .11 ¡.01 .06 ¡.32¤ .07 ¡.21¤ .19¤ — 9. Conscientiousness 4.04 .53 ¡.03 .02 ¡.03 .35¤ .09 .40¤ ¡.28¤ ¡.12 — 10. Agreeableness 3.84 .50 ¡.05 .06 ¡.03 .51¤ .12 .43¤ ¡.24¤ ¡.28¤ .28¤ — 11. Neuroticism 2.80 .73 .08 ¡.05 ¡.03 ¡.45¤ .17¤ ¡.17¤ .11 .39¤ ¡.33¤ ¡.34¤ — 12. Openness 3.64 .67 ¡.08 .07 ¡.08 .16¤ ¡.10 .17¤ ¡.12 ¡.14¤ .21¤ .23¤ ¡.13¤ — 13. Extraversion 3.58 .72 ¡.01 ¡.10 ¡.01 .41¤ ¡.10 .14¤ ¡.14¤ ¡.05 .20¤ .22¤ .24¤ .13¤ — 14. Task Performance 6.17 1.42 .05 .01 ¡.00 .03 .05 ¡.05 .09 ¡.04 .07 .08 ¡.14 .10 .03 — 15. Counterproductive Behavior .10 .30 ¡.18¤ ¡.05 ¡.10 ¡.03 ¡.04 ¡.10 .06 .07 ¡.12 ¡.02 .03 ¡.01 ¡.08 ¡.01 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 121 predictions. The eVect of interpersonal justice on coun- terproductive behavior was stronger for trusting individ- uals. Neither of the procedural justice interactions reached statistical signiWcance. Fig. 2. Justice £ trust propensity interactions for task performance and counterproductive behavior. High Trust Propensity Low Trust Propensity High Trust Propensity Low Trust Propensity High Trust Propensity Interpersonal Justice Interpersonal Justice T a s k P e rf o rm a n c e T a s k P e rf o rm a n c e C o u n te r p ro d u c ti v e B e h a v io u r Distributive Justice Low Trust Propensity High Low High Low High Low 7.75 7.00 6.25 5.50 4.75 7.75 7.00 6.25 5.50 4.75 0.40 0.30 0.20 0.10 0.00 Hypothesis 2a–c predicted stronger eVects for proce- dural, interpersonal, and distributive justice on task per- formance and counterproductive behavior when risk aversion was high. The procedural justice £ risk aversion interaction was signiWcant for both task performance and counterproductive behavior, with the patterns in the predicted direction (see Fig. 3). A signiWcant interper- sonal justice £ risk aversion interaction was also observed for counterproductive behavior, again in the predicted direction (see Fig. 3). In all three cases, the beneWcial eVects of fair treatment were stronger for indi- Fig. 3. Justice £ risk aversion interactions for task performance and counterproductive behavior. High Risk Aversion Low Risk Aversion High Risk Aversion Low Risk Aversion High Risk Aversion Low Risk Aversion High Low High Low High Low 7.75 7.00 6.25 5.50 4.75 0.40 0.30 0.20 0.10 0.00 0.40 0.30 0.20 0.10 0.00 Interpersonal Justice Procedural Justice Procedural Justice T a s k P e rf o rm a n c e C o u n te r p ro d u c ti v e B e h a v io u r C o u n te r p ro d u c ti v e B e h a v io u r 122 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 viduals who were highly risk-averse. Contrary to predic- tions, neither of the distributive justice interactions was signiWcant. Hypothesis 3a and b predicted stronger eVects for procedural and interpersonal justice on task perfor- mance when trait morality was high. The interpersonal justice £ trait morality interaction was signiWcant for task performance, with the pattern in the predicted direction (see Fig. 4). Though not predicted, the results also revealed a similar interaction with distributive jus- tice for task performance (see Fig. 4). In both cases, the eVects of fair treatment on task performance were more positive for individuals high in trait morality. Contrary to predictions, the procedural justice interaction was not statistically signiWcant. Results for alternative approaches to moderation of justice eVects Table 4 provides the moderated regression results for equity sensitivity and SBI. The Wrst two steps reproduce the justice manipulation results from Table 3. The third step examines the main eVects of the two traits, and nei- ther had signiWcant eVects. The fourth step examines the justice £ personality product terms, and the step was not Fig. 4. Justice £ trait morality interactions for task performance. Table 4 Moderated regression results for equity sensitivity and sensitivity to Befallen Injustice Note. n D 238. �R2 values may not sum exactly to R2 values due to rounding error. ¤ p < .05, two-tailed. ¤¤ p < .10, two-tailed. Regression step Task performance Counterproductive behavior R2 �R2 � R2 �R2 � 1. Procedural Justice (PJ) .00 .00 .05 .05¤ .05¤ ¡.19¤ Interpersonal Justice (IJ) .02 ¡.08 Distributive Justice (DJ) .00 ¡.12¤¤ 2. PJ £ IJ .01 .01 ¡.05 .10¤ .06¤ .20¤ PJ £ DJ ¡.04 .15¤ IJ £ DJ ¡.09 .07 3. Equity Sensitivity .03 .02 .11 .12¤ .01 .04 SBI ¡.08 .11 4. PJ £ Equity Sensitivity .05 .02 ¡.01 .15 .03 ¡.13 IJ £ Equity Sensitivity .10 .06 DJ £ Equity Sensitivity ¡.01 ¡.07 PJ £ SBI ¡.06 ¡.06 IJ £ SBI ¡.08 .02 DJ £ SBI ¡.09 ¡.01 Table 5 Moderated regression results for Big Five dimensions Note. n D 238. �R2 values may not sum exactly to R2 values due to rounding error. ¤ p < .05, two-tailed. ¤¤ p < .10, two-tailed. Regression step Task performance Counterproductive behavior R2 �R2 � R2 �R2 � 1. Procedural Justice (PJ) .00 .00 .05 .05¤ .05¤ ¡.19¤ Interpersonal Justice (IJ) .02 ¡.08 Distributive Justice (DJ) .00 ¡.12¤¤ 2. PJ £ IJ .01 .01 ¡.05 .10¤ .06¤ .20¤ PJ £ DJ ¡.04 .15¤ IJ £ DJ ¡.09 .07 3. Conscientiousness .04 .03 .01 .13¤ .02 ¡.12 Agreeableness .02 .01 Neuroticism ¡.15 .01 Openness .08 ¡.01 Extraversion ¡.02 ¡.08 4. PJ £ Conscientiousness .10 .06 .00 .20¤ .07 .05 IJ £ Conscientiousness .00 ¡.10 DJ£ Conscientiousness ¡.03 .04 PJ £ Agreeableness ¡.07 ¡.13 IJ £ Agreeableness ¡.15 ¡.18 DJ £ Agreeableness .04 ¡.09 PJ £ Neuroticism .04 .00 IJ £ Neuroticism ¡.03 ¡.04 DJ £ Neuroticism ¡.06 ¡.10 PJ £ Openness ¡.00 .06 IJ £ Openness .18 .05 DJ £ Openness .01 .02 PJ £ Extraversion .05 .10 IJ £ Extraversion ¡.04 .02 DJ £ Extraversion .02 .07 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 123 signiWcant for either task performance or counterpro- ductive behavior. Table 5 presents the moderated regres- sion results for the Wve-factor model. As with the other personality traits, the third step of the regressions revealed no main eVects of the Big Five dimensions on task performance or counterproductive behavior. The fourth step examines the justice £ personality product terms, and the step was not signiWcant for either task performance or counterproductive behavior. Discussion Taken together, our results illustrate that the integra- tive theories that have emerged within the justice litera- ture over the past Wve years can be fruitful sources of potential moderators of justice eVects. Although fairness heuristic theory, uncertainty management theory, and fairness theory were not created for the express purpose of identifying personality moderators, the traits exam- ined here Xow neatly from the core propositions of each theory. The signiWcant interactions for trust propensity, risk aversion, and trait morality were observed across types of justice, supporting the replicability of the eVects. With the exception of trait morality, the interactions were also observed for both task performance and coun- terproductive behavior. The task performance eVects are notable, as the interactions help to clarify the inconsis- tent results in past studies of justice and performance (Cohen-Charash & Spector, 2001; Colquitt et al., 2001). Of the moderators examined in our study, risk aver- sion yielded the most promising results, as it ampliWed the beneWcial eVects of both procedural and interper- sonal justice. Indeed, it was the only trait examined in our study that altered procedural eVects. Risk aversion was also the only trait that moderated both task perfor- mance and counterproductive behavior eVects in the pre- dicted manner. In contrast, risk aversion did not alter the eVects of distributive justice on either task performance or counterproductive behavior. Although uncertainty management theory deemphasizes the diVerences among the justice dimensions (Lind & Van den Bos, 2002; Van den Bos & Lind, 2002), it may be that distributive justice is less relevant to the management of uncertainty than procedural or interpersonal justice because it is more episodic, particularly when referenced to a one-time out- come or decision event. With regards to trust propensity, given that justice is used as a proxy for trust, we reasoned that justice would have less value to individuals who trusted “by default,” instead taking on more importance to the dispositionally suspicious. Indeed, our results showed that the eVects of interpersonal and distributive justice on task perfor- mance were more positive for individuals lower in trust propensity. Those eVects aside, two other aspects of our trust propensity results were contrary to predictions. First, trust propensity failed to moderate procedural jus- tice eVects on either task performance or counterproduc- tive behavior, which is somewhat surprising because many procedural justice rules (e.g., consistency, bias sup- pression, and ethicality) overlap conceptually with the integrity dimension of trustworthiness (Leventhal, 1980; Mayer et al., 1995). It may be that trust propensity would emerge as a signiWcant moderator of procedural justice eVects in Weld settings, where long-term relation- ships are prevalent. Second, the eVects of interpersonal justice on counterproductive behavior were actually stronger for individuals high in trust propensity, oppo- site to the task performance results. This result may be akin to Wndings by Brockner, Tyler, and Cooper-Schnei- der (1992) that individuals who were highly committed to an organization experienced greater negative reac- tions as a result of unfair treatment than those less com- mitted. That is, “the higher they are, the harder they fall” (Brockner et al., 1992). However, given that the results for task performance were in the opposite pattern, rather than non-signiWcant, this explanation may not be ade- quate. Future research is needed to provide insight into explaining these conXicting results. The third moderator examined in our study was trait morality. Using a trait-based form of the variable drawn from trait models of personality (de Raad et al., 1992; Hofstee et al., 1992), we found that the eVects of inter- personal and distributive justice on task performance were more positive for individuals high in trait morality. In contrast, trait morality did not moderate the eVects of procedural justice on task performance, contrary to our predictions. Folger (2001) argued that the moral accountability of justice tends to grow in ascending order from distributive to procedural to interactional. Thus, although distributive justice does have moral rele- vance (Folger, 1998, 2001), one would have expected the procedural interaction eVects to be stronger than the dis- tributive. It may be that our Wndings might have diVered had we manipulated procedural rules that were more morally charged than accuracy and consistency, such as bias suppression or ethicality (Leventhal, 1980). To gauge the signiWcant eVects for risk aversion, trust propensity, and trait morality, we explored two alterna- tive approaches to moderation of justice eVects. First, we examined two traits introduced by justice scholars: equity sensitivity and SBI. Both traits failed to moderate the eVects of justice, and their moderating eVects were much weaker than those of the three integrative theory traits. However, it should be noted that equity sensitivity was never intended to moderate the eVects of procedural or interpersonal justice. Similarly, the content of SBI, as measured using Schmitt and Maes’s (2000) scale, is decidedly distributive in nature, with the “befallen injus- tice” usually consisting of others receiving more of something of value. It therefore remains unclear how relevant SBI is to procedural and interpersonal justice. 124 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 Second, we investigated the traits of the Big Five; how- ever, results were weaker compared to the results for the three integrative theory traits, with the justice £ Big Five product terms explaining less variance in the outcomes overall. Although Skarlicki et al.’s (1999) results found signiWcant interactions with Big Five variables, their results involved three-way interactions rather than the two-way eVects explored in our study. Still, it is notable that 6 years have passed since Skarlicki et al.’s (1999) study was published without further published work on justice £ Big Five interactions. It may be that such work has been conducted but yielded non-signiWcant results, creating a potential Wle drawer problem. Practical implications Our results oVer a number of practical implications. Task performance and counterproductive behavior clearly are important outcomes in any organization, as both impact bottom-line costs. Our results add to past research linking justice dimensions to those behaviors (Greenberg, 1990, 1993b, 2002; Konovsky & Cropanz- ano, 1991; Van den Bos et al., 1996). Inaccurate and inconsistent procedures, disrespectful interpersonal treatment, and unexpectedly low rewards impacted counterproductive behavior and task performance, though often in an interactive rather than main eVect fashion. Although justice principles can be trained (e.g., Skarl- icki & Latham, 1996), training initiatives often fail because of a lack of awareness of speciWc person and organizational variables that impact training eVective- ness. The “person analysis” phase of the training needs assessment is used to identify participant characteristics that make training more or less eVective, whereas the “organization analysis” explores contextual variables that can do the same. Our results could inform the per- son analysis, suggesting that justice training interven- tions would have particularly strong eVects when the leaders participating in the training oversee units with predominantly suspicious, cautious, and moral individu- als. However, organizations do not typically collect data on trust propensity, risk aversion, and trait morality, and it would be diYcult to characterize an entire unit on such personality traits. Our results might therefore inform the organization analysis, as justice trainers could consider contextual cues that trigger or amplify concerns about distrust, uncertainty, or moral issues. Limitations As in many laboratory studies, the framing of instruc- tions, the sequence of measurement, and the timing of measurements could impact the generalizability of results. For example, Harrison, McLaughlin, and Coalter (1996) showed that measures that reference fair- ness concepts are susceptible to scale ordering and con- text eVects. Our personality measures, some of which themselves referenced justice concepts, may be suscepti- ble to such biases. Context eVects could also have accounted for the small amount of manipulation spill- over that occurred among the justice manipulation checks. Although we removed this contamination by residualizing the aVected manipulations, a procedural remedy is usually preferred over a statistical one. In addition, our distributive justice manipulation varied fulWllment of expectations rather than outcome/input ratio comparisons. Although expectation fulWllment matches Blau’s (1964) conceptualization of distributive justice and has been used extensively in past research (Greenberg, 1993b; Van den Bos et al., 1998b), that type of manipulation may have constrained equity sensitiv- ity’s eVects, given that it is based on Adams’s (1965) theorizing. Other limitations center on the measures of some of our variables. Van den Bos et al. (1998a) showed that knowledge of trust moderated the eVects of justice, with justice being more impactful when trust was uncertain. Our measure of trust propensity is incapable of separat- ing trust uncertainty from trust valence, as suspicious individuals may either be uncertain about the trustwor- thiness of others or be certain that trustworthiness is lacking. In addition, our measure of counterproductive behavior could be interpreted in two ways: as intentional noncompliance with authority (one of the less serious forms of the construct) and outright theft (one of the more serious forms of the construct). Our results for counterproductive behavior may therefore fail to gener- alize to contexts where the retaliation is clearly serious in nature (e.g., theft of actual money). Finally, the most serious limitation of our study cen- ters on our trait morality measure. Recall that we chose to deWne morality as an intersection of high agreeable- ness and high conscientiousness (Hofstee et al., 1992). However, there are some interpretational ambiguities in trait models, mostly due to the presence of measure- ment error. For example, the adjectives “conscien- tious” and “agreeable” do not themselves fall squarely on the circumplex poles used to deWne those respective Big Five factors. Such ambiguities may hinder the con- struct validity of adjectives like “morality” that reside at the intersection of the conscientiousness and agree- ableness poles. Although future work can clarify many of these issues, there remains a limited amount of research on the Big Five circumplex. Moreover, it is important to note that Hofstee et al.’s (1992) model does not focus on morality per se, instead describing a large number of traits that represent blends of multiple Big Five factors. It is also important to note that the results for our cir- cumplex form of morality cannot be generalized to other forms of the construct, such as a Kohlbergian stage J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 125 approach. As stated previously, a trait approach diVers from a stage approach in that the former emphasizes diVerences across individuals while the latter emphasizes diVerences within individuals over time (Walker, 2002). However, it may be that stage models such as Kohl- berg’s would also be useful for examining moderating eVects of justice (for an example, see Greenberg, 2002). Future researchers utilizing such an approach would beneWt from using appropriate stage measures such as the DeWning Issues Test 2 (Rest, Narvaez, Thoma, & Bebeau, 1999). Suggestions for future research Perhaps the most fruitful area for future research con- cerns one additional limitation of our study. Our concep- tual model in Fig. 1 includes three diVerent mechanisms through which personality traits could moderate justice eVects: sensitivity to justice concerns, deliberate rumina- tion on justice information, and likelihood of behavioral response. Our study did not assess those mechanisms, however. It therefore remains an empirical question whether those mechanisms do explain moderation eVects at all, or whether only one or two of the processes is neces- sary for interactions to emerge. Although our study was fairly comprehensive in its coverage of multiple justice dimensions and its inclusion of several personality traits, some gaps in our coverage highlight other areas for future research. For example, the “belief in a just world” may be relevant to the should counterfactual and to the mechanisms in Fig. 1. It may be that individuals with strong just world beliefs attend more to justice because they, like moral individuals, are more concerned with fairness in general. However, it may also be that individuals with strong just world beliefs attend less to issues of fairness because they believe that things will “work out” in the end. Future research should also explore personality mod- erators in conjunction with the other sub-facet of interac- tional justice: informational justice. We did not include informational justice in the present study for both practi- cal and conceptual reasons. Practically speaking, includ- ing informational justice would have doubled the number of experimental conditions in an already complex study. Conceptually, we felt that some of our personality traits were more relevant to interpersonal justice, given that it is morally charged and holds special relevance to the integ- rity and benevolence forms of trustworthiness (Folger, 2001; Mayer et al., 1995). However, informational justice is germane to the could and should portions of fairness theory (Folger & Cropanzano, 2001; Shaw, Wild, & Colquitt, 2003) and its eVects could be moderated by traits that capture the likelihood of asking “why?” ques- tions (e.g., curiosity, need for cognition). Future research should also explore personality mod- erators of the two-way justice interactions often observed in the justice literature. Skarlicki et al.’s (1999) results showed that neuroticism and agreeableness mod- erated the interaction between distributive and interac- tional justice, while Hagedoorn et al.’s (2002) showed that just world beliefs moderated the interaction between distributive and procedural justice. Such work allows justice scholars to assess the generalizability of such interactions across personality types. It may be that more complex interactions could help explain some of the unexpected patterns in our two-way results. For example, some of the interactions were symmetric in nature, with fair treatment resulting in lower perfor- mance or more counterproductive behavior at certain personality patterns. Although such patterns are surpris- ing in the context of two-way interactions, they may make more sense in conjunction with higher-order inter- action patterns. Finally, it is important to point out that the interac- tions observed in our study could be interpreted in a diVerent way—that justice dimensions moderate the relationships between personality traits and counter- productive behavior and task performance. The signiW- cant product terms can be interpreted in either fashion, as can the plots represented in the Wgures. It may be that organizational justice can be used to clarify incon- sistent behavioral eVects within the trust propensity, risk aversion, or morality literatures. For example, unfair treatment may be relevant to the principle of trait activation (e.g., Tett & Gutterman, 2000), with fairness violations creating situational cues that enhance the likelihood of trait expression. To our knowledge, the justice dimensions have rarely been cast as moderators in past research, leaving open a poten- tially fruitful area for future theorizing and empirical study. References Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 267–299). New York: Academic Press. Bies, R. J., & Moag, J. F. (1986). Interactional justice: Communication criteria of fairness. In R. J. Lewicki, B. H. Sheppard, & M. H. Bazer- man (Eds.), Research on negotiations in organizations (Vol. 1, pp. 43–55). Greenwich, CT: JAI Press. Bies, R. J., & Tripp, T. M. (1995). Beyond distrust: “Getting even” and the need for revenge. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 246–260). Thou- sand Oaks, CA: Sage. Blau, P. (1964). Exchange and power in social life. New York: Wiley. Brockner, J., Heuer, L., Siegel, P. A., Wiesenfeld, B., Martin, C., Grover, S., et al. (1998). The moderating eVect of self-esteem in reaction to voice: Converging evidence from Wve studies. Journal of Personality and Social Psychology, 75, 394–407. Brockner, J., Tyler, T. R., & Cooper-Schneider, R. (1992). The inXuence of prior commitment to an institution on reactions to perceived unfairness: The higher they are, the harder they fall. Administrative Science Quarterly, 37, 241–261. 126 J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 Cable, D. M., & Judge, T. A. (1994). Pay preferences and job search decisions: A person-organization Wt perspective. Personnel Psychol- ogy, 47, 317–348. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, NJ: Erlbaum. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in orga- nizations: A meta-analysis. Organizational Behavior and Human Decision Processes, 86, 278–321. Colquitt, J. A. (2001). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86, 386–400. Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C., & Ng, K. Y. (2001). Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychol- ogy, 86, 425–445. Colquitt, J. A., Greenberg, J., & Zapata-Phelan, C. P. (2005). What is organizational justice?: A historical overview. In J. Greenberg & J. A. Colquitt (Eds.), The handbook of organizational justice (pp. 3– 56). Mahwah, NJ: Erlbaum. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO- FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Cropanzano, R., Byrne, Z. S., Bobocel, D. R., & Rupp, D. E. (2001). Moral virtues, fairness heuristics, social entities, and other denizens of organizational justice. Journal of Vocational Behavior, 58, 164–209. de Raad, B., Hendriks, A. J., & Hofstee, W. K. (1992). Towards a reWned structure of personality traits. European Journal of Person- ality, 6, 301–319. Eisenberger, R., Cotterell, N., & Marvel, J. (1987). Reciprocation ideol- ogy. Journal of Personality and Social Psychology, 53, 743–750. Folger, R. (1998). Fairness as a moral virtue. In M. Schminke (Ed.), Managerial ethics: Morally managing people and processes (pp. 13– 34). Mahwah, NJ: Erlbaum. Folger, R. (2001). Fairness as deonance. In S. W. Gilliland, D. D. Steiner, & D. P. Skarlicki (Eds.), Theoretical and cultural perspec- tives on organizational justice (pp. 3–34). Greenwich, CT: Informa- tion Age Publishing. Folger, R., & Cropanzano, R. (2001). Fairness theory: Justice as accountability. In J. Greenberg & R. Cropanzano (Eds.), Advances in organizational justice (pp. 89–118). Stanford, CA: Stanford Uni- versity Press. Folger, R., Cropanzano, R., & Goldman, B. (2005). What is the rela- tionship between justice and morality. In J. Greenberg & J. A. Col- quitt (Eds.), Handbook of organizational justice (pp. 215–245). Mahwah, NJ: Erlbaum. Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five factor structure. Journal of Personality and Social Psychology, 59, 1216–1229. Greenberg, J. (1990). Employee theft as a reaction to underpayment nequity: The hidden cost of pay cuts. Journal of Applied Psychol- ogy, 75, 561–568. Greenberg, J. (1993a). The social side of fairness: Interpersonal and informational classes of organizational justice. In R. Cropanzano (Ed.), Justice in the workplace: Approaching fairness in human resource management (pp. 79–103). Hillsdale, NJ: Erlbaum. Greenberg, J. (1993b). Stealing in the name of justice: Informational and interpersonal moderators of theft reactions to underpayment inequity. Organizational Behavior and Human Decision Processes, 54, 81–103. Greenberg, J. (2002). Who stole the money, and when? Individual and situational determinants of employee theft. Organizational Behavior and Human Decision Processes, 89, 985–1003. Hagedoorn, M., Buunk, B. P., & van de Vliert, E. (2002). Do just world believers process unfair authoritative decisions diVerently? Applied Psychology: An International Review, 51, 126–145. Harrison, D. A., McLaughlin, M. E., & Coalter, T. M. (1996). Context, cognition, and common method variance: Psychometric and verbal protocol evidence. Organizational Behavior and Human Decision Processes, 68, 246–261. Heuer, L., Blumenthal, E., Douglas, A., & Weinblatt, T. (1999). A deserv- ingness approach to respect as a relationally based fairness judgment. Personality and Social Psychology Bulletin, 25, 1279–1292. Hofstee, W. K., de Raad, B., & Goldberg, L. R. (1992). Integration of the Big Five and circumplex approaches to trait structure. Journal of Personality and Social Psychology, 63, 146–163. Hogan, J., & Roberts, B. W. (1996). Issues and non-issues in the Wdelity-bandwidth debate. Journal of Organizational Behavior, 17, 627–637. Homans, G. C. (1961). Social behaviour: Its elementary forms. London: Routledge and Kegan Paul. Huseman, R. C., HatWeld, J. D., & Miles, E. W. (1987). A new perspec- tive on equity theory: The Equity Sensitivity construct. Academy of Management Review, 12, 222–234. International Personality Item Pool (2001). A scientiWc collabora- tory for the development of advanced measures of personality traits and other individual diVerences (http://ipip.ori.org/). Internet Web Site. John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big FiveInven- tory—Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research. Joy, V. L., & Witt, L. A. (1992). Delay of gratiWcation as a moderator of the procedural justice-distributive justice relationship. Group and Organization Management, 17, 297–308. Judge, T. A., & Larsen, R. J. (2001). Dispositional aVect and job satis- faction: A review and theoretical extension. Organizational Behav- ior and Human Decision Processes, 86, 67–98. Kanfer, R. (1991). Motivation theory and industrial and organiza- tional psychology. In M. D. Dunnette & L. M. Hough (Eds.), Hand- book of industrial and organizational psychology (Vol. 1, pp. 75– 170). Palo Alto, CA: Consulting Psychologists Press. King, W. C., & Miles, E. W. (1994). The measurement of equity sensi- tivity. Journal of Occupational and Organizational Psychology, 67, 133–142. King, W. C., Miles, E. W., & Day, D. D. (1993). A test and reWnement of the equity sensitivity construct. Journal of Organizational Behav- ior, 14, 301–317. Kohlberg, L. (1984). The psychology of moral development. San Fran- cisco, CA: Harper and Row. Konovsky, M. A., & Cropanzano, R. (1991). Perceived fairness of employee drug testing as a predictor of employee attitudes and job performance. Journal of Applied Psychology, 76, 698–707. Leventhal, G. S. (1980). What should be done with equity theory? New approaches to the study of fairness in social relationships. In K. Ger- gen, M. Greenberg, & R. Willis (Eds.), Social exchange: Advances in theory and research (pp. 27–55). New York: Plenum Press. Lind, E. A. (2001). Fairness heuristic theory: Justice judgments as piv- otal cognitions in organizational relations. In J. Greenberg & R. Cropanzano (Eds.), Advances in organizational justice (pp. 56–88). Stanford, CA: Stanford University Press. Lind, E. A., & Van den Bos, K. (2002). When fairness works: Toward a general theory of uncertainty management. In B. M. Staw & R. M. Kramer (Eds.), Research in organizational behavior (Vol. 24, pp. 181–223). Boston, MA: Elsevier. Maehr, M. L., & Videbeck, R. (1968). Predisposition to risk and persis- tence under varying reinforcement-success schedules. Journal of Personality and Social Psychology, 9, 96–100. Major, B., & Deaux, K. (1982). Individual diVerences in justice behav- ior. In J. Greenberg & R. L. Cohen (Eds.), Equity and justice in social behavior (pp. 43–76). New York: Academic Press. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20, 709–734. http://ipip.ori.org/ http://ipip.ori.org/ J.A. Colquitt et al. / Organizational Behavior and Human Decision Processes 100 (2006) 110–127 127 Ones, D. S. (1994). The construct validity of integrity tests. Unpublished doctoral dissertation, University of Iowa. Perdue, B. C., & Summers, J. O. (1986). Checking the success of manip- ulations in marketing experiments. Journal of Marketing Research, 23, 317–326. Pizarro, D. A., Uhlmann, E., & Bloom, P. (2003). Causal deviance and the attribution of moral responsibility. Journal of Experimental Social Psychology, 39, 653–660. Rest, J., Narvaez, D., Thoma, S. J., & Bebeau, M. J. (1999). DIT-2: Devising and testing a new instrument of moral judgment. Journal of Educational Psychology, 91, 644–659. Saucier, G., & Goldberg, L. R. (1996). Evidence for the Big Five in analyses of familiar English personality adjectives. European Jour- nal of Personality, 10, 61–77. Schmitt, M. J., & Dorfel, M. (1999). Procedural injustice at work, jus- tice sensitivity, job satisfaction and psychosomatic well-being. European Journal of Social Psychology, 29, 443–453. Schmitt, M. J., & Maes, J. (2000). Scales for measuring justice sensitivity from three perspectives. Unpublished manuscript, University of Trier, Germany. Schmitt, M. J., Neumann, R., & Montada, L. (1995). Dispositional sen- sitivity to befallen injustice. Social Justice Research, 8, 385–407. Schmitt, N., & Stults, D. M. (1985). Factors deWned by negatively keyed items: The result of careless respondents? Applied Psychological Measurement, 9, 367–373. Schneider, B. (1983). Interactional psychology and organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in orga- nizational behavior (Vol.5, pp. 1–31). Greenwich, CT: JAI Press. Shaw, J. C., Wild, R. E., & Colquitt, J. A. (2003). To justify or excuse?: A meta-analysis of the eVects of explanations. Journal of Applied Psychology, 88, 444–458. Skarlicki, D. P., Folger, R., & Tesluk, P. (1999). Personality as a moder- ator in the relationship between fairness and retaliation. Academy of Management Journal, 42, 100–108. Skarlicki, D. P., & Latham, G. P. (1996). Increasing citizenship behav- ior within a labor union: A test of organizational justice theory. Journal of Applied Psychology, 81, 161–169. Sweeney, P. D., & McFarlin, D. B. (1993). Workers’ evaluations of the “ends” and “means”: An examination of four models of distribu- tive and procedural justice. Organizational Behavior and Human Decision Processes, 55, 23–40. Sweeney, P. D., McFarlin, D. B., & Cotton, J. L. (1991). Locus of con- trol as a moderator of the relationship between perceived inXuence and procedural justice. Human Relations, 44, 333–342. Tett, R. P., & Gutterman, H. A. (2000). Situation trait relevance, trait expression, and cross-situational consistency: Testing a principle of trait activation. Journal of Research in Personality, 34, 397–423. Thibaut, J., & Walker, L. (1975). Procedural justice: A psychological analysis. Hillsdale, NJ: Erlbaum. Turillo, C. J., Folger, R., Lavelle, J. J., Umphress, E. E., & Gee, J. O. (2002). Is virtue its own reward? Self-sacriWcial decisions for the sake of fairness. Organizational Behavior and Human Decision Pro- cesses, 89, 839–865. Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. In M. P. Zanna (Ed.), Advances in experimental social psy- chology (Vol. 25, pp. 115–191). San Diego, CA: Academic Press. Van den Bos, K. (2001a). Fairness heuristic theory: Assessing the infor- mation to which people are reacting has a pivotal role in under- standing organizational justice. In S. Gilliland, D. Steiner, & D. Skarlicki (Eds.), Theoretical and cultural perspectives on organiza- tional justice (pp. 63–84). Greenwich, CT: Information Age Publish- ing. Van den Bos, K. (2001b). Uncertainty management: The inXuence of uncertainty salience on reactions to perceived procedural fairness. Journal of Personality and Social Psychology, 80, 931–941. Van den Bos, K., Bruins, J., Wilke, H. A. M., & Dronkert, E. (1999). Sometimes unfair procedures have nice aspects: On the psychology of the fair process eVect. Journal of Personality and Social Psychol- ogy, 77, 324–336. Van den Bos, K., & Lind, E. A. (2002). Uncertainty management by means of fairness judgments. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 34, pp. 1–60). Boston, MA: Elsevier. Van den Bos, K., Maas, M., Waldring, I. E., & Semin, G. R. (2003). Toward understanding the psychology of reactions to perceived fairness: The role of aVect intensity. Social Justice Research, 16, 151–168. Van den Bos, K., Vermunt, R., & Wilke, H. A. M. (1996). The consis- tency rule and the voice eVect: The inXuence of expectations on procedural fairness judgments and performance. European Journal of Social Psychology, 26, 411–428. Van den Bos, K., Wilke, H. A. M., & Lind, E. A. (1998a). When do we need procedural fairness? The role of trust in authority. Journal of Personality and Social Psychology, 75, 1449–1458. Van den Bos, K., Wilke, H. A. M., Lind, E. A., & Vermunt, R. (1998b). Evaluating outcomes by means of the fair process eVect: Evidence for diVerent processes in fairness and satisfaction judgments. Jour- nal of Personality and Social Psychology, 74, 1493–1503. Walker, L. J. (2002). The model and the measure: An appraisal of the Minnesota approach to moral development. Journal of Moral Edu- cation, 31, 353–367. Witt, L. A., Kacmar, K. M., & Andrews, M. C. (2001). The interac- tive eVects of procedural justice and exchange ideology on supervisor-rated commitment. Journal of Organizational Behavior, 22, 505–515. Zuckerman, M., & Reis, H. T. (1978). Comparison of three models for predicting altruistic behavior. Journal of Personality and Social Psychology, 36, 498–510. Justice and personality: Using integrative theories to derive moderators of justice effects Past research on moderation of justice effects Distributive justice research Procedural justice research Interactional justice research Applying integrative justice theories to personality moderators of justice effects A brief review of integrative justice theories Fairness heuristic theory Uncertainty management theory Fairness theory Which traits worthy of study can be derived from the integrative theories? Trust propensity Risk aversion Morality Alternative approaches to moderation of justice effects Method Sample Procedure Integrative theory traits Trust propensity Risk aversion Trait morality Confirmatory factor analysis Alternative moderators Equity sensitivity Sensitivity to Befallen injustice Big Five dimensions Outcome measures Task performance Counterproductive behavior Manipulation checks Results Manipulation checks Results for integrative theory traits Results for alternative approaches to moderation of justice effects Discussion Practical implications Limitations Suggestions for future research References