Procedural Justice Model - Draft 1

    

RUNNING HEAD:

THE FOUR QUADRANT MODEL OF PROCEDURAL JUSTICE

THE FOUR QUADRANT MODEL OF PROCEDURAL JUSTICE: A proposed research model to test the effects of self-affirmation and uncertainty management theory within the context of a proposed model for procedural justice determinations Christopher V. Langone Cornell University

Understanding social justice is a crucial aspect of understanding human behavior. Tyler and Blader (2003) have stressed, justice “has an impact; it is substantial in magnitude; it is consistently found across a wide variety of group and organizational contexts; and it is distinct from judgments of self-interest or personal/group gain.” (p. 349.) Social justice has also been described as “one of the most important norms and values in human society. (Van Prooijen, Karremans & Beest, 2006, citing Folger, 1984). Broadly speaking, there are two kinds of justice: distributive and procedural. (Lind & Tyler, 1988; Tyler & Lind, 1992). Distributive justice tends to focus on outcomes or “ends;” whereas procedural justice, most famously espoused by Rawls (1971), tends to focus on procedures, or “means.”1 The lines between procedural and distributive justice can sometimes be fuzzy (Folger, 1996), and judgments about procedural fairness are oftentimes influenced by the outcome. (Daley & Tripp, 1996).2 This paper will review the existing state of procedural justice research, set forth and explain a proposed “Four-Quadrant” model of justice research and then set forth a specific research program designed to contrast and test emerging theories in justice research. I.

PROCEDURAL JUSTICE RESEARCH Initially, empirical research on justice focused on distributive aspects, in a body of research

that came to be known as “equity theory.” (Adams, 1965). But after Rawls penned his Theory of Justice in 1971, social science research in the area of procedural justice began in earnest. Thibaut and Walker’s (1975) seminal work is perhaps best known in this regard due to their rather striking                                                              1

Folger (1996) critiques the distinction, explaining that the line between the two conceptions is often blurred in practice. Indeed, after proposing a third construct of “interactional justice,” he quickly dismisses it, noting it too does not create a clear demarcation of boundaries. For other work on “interactional justice” see, Bies & Shapiro, 1987; Blader & Tyler, 2003). 2

Daley and Tripp (1996) cite as an example of this the fact that prior to announcement of the O.J. Simpson verdict only 40% of African Americans stated they thought he had gotten a fair trial, but after the “not guilty” verdict was announced 80% viewed the trial as “fair.” 1   

findings that suggested that people may care about fair procedures even more than favorable outcomes. As they noted, “the use of a fair procedure can increase the satisfaction of all concerned without any increase in the real outcomes available for distribution.” (p. 29, emphasis in original). The above effect has been dubbed the “fair process effect.” Most research in the area of procedural research has come from one of four theoretical perspectives: (1) the mainly instrumental approach involving control over outcome stemming from the original Thibaut & Walker research; (2) a relational approach developed in the labs of Tyler and Lind, which developed two relational models (group value & authority) that they now seek to synthesize into a more general “group engagement model;” (3) the pioneering application of uncertainty management theory into the procedural justice context (Van den Bos, 2001) and (4) the often overlooked deonance and moral virtue theory advanced primarily by Folger (1996). This last theory provides some theoretical links to emerging research in self-affirmation theory. 1.

Quadrant 1: Structural Inputs – Thibaut & Walker (1975); Process Control and the “Instrumental” Approach

In the Thibaut and Walker (1975) procedural justice studies, procedures were viewed as mechanisms for making decisions about the allocation of outcomes. In particular they focused on “formal procedures that related to decision-making processes in legal settings.” (Tyler & Blader, 2003, p.350). Thibaut & Walker (1975) identified three variables that mediate the degree to which a process is deemed fair: the perceived neutrality of the decision-maker, the degree to which trust is developed in the decision-maker and the process itself, and the degree to which the process allows participants to exercise “control over the development and selection of information that will constitute the basis of resolving the dispute.” (p. 546). This last variable, which Thibaut and Walker called process control, has been named the “voice effect.” (Folger 1977). Although Lind, Kanfer, and Earley (1990) have described the voice

2   

effect as “probably the best-documented phenomenon in procedural justice research;” (p. 952), it is also, as Shapiro and Brett (1993) “probably the most debated too.” (p. 1167). The debate focuses on whether voice is important for primarily instrumental or non-instrumental reasons. The instrumental explanation of voice, preferred by Thibaut and Walker (1975), posits that voice is important to the extent it allows participants the opportunity to exercise influence over the process by bolstering one’s own case and disputing the adversary’s contentions. (p. 109). Tyler and Lind (1992), however, have emphasized the non-instrumental importance of voice. Under this view, voice is important not because it can influence outcomes, but because it results in “value expressive” feelings that increase participant satisfaction and perceived fairness regardless of the effect on the outcome. Other views have suggested that voice has instrumental value to the extent it reduces uncertainty (Van den Bos & Lind, 2002), or affirms moral values and principles (Folger & Cropanzano, 1998). These views are discussed in more detail later in this paper. Fondacaro, et al. (2006), note that Thibaut and Walker’s work proved to be of great heuristic value because: (1) it shows that procedural justice is one mechanism by which outcomes are accepted, thus serving an important legitimization function; and (2) it was the first empirical demonstration that procedural justice has an effect independent of outcome (p. 988). Nevertheless, they criticize the research program as being too narrow in that it was focused on laboratory studies using legal-decision making scenarios. This criticism was echoed by Tyler and Blader (2003) who explained that because Thibaut and Walker’s procedural models were rooted in an era in which distributive justice dominated, their focus was natural. Tyler and Blader (2003) argued that this focus, along with the use of legal scenarios, limited Thibaut and Walker’s theory development because they linked desires for fair procedures with desires to achieve equitable outcomes. In other words, procedures were valued only insofar as they could affect the associated outcomes. Tyler and Blader conclude, “this focus on 3   

decision making in allocation contexts is no longer true of procedural justice research. Researchers have increasingly moved their attention away from an exclusive focus on the decision-making function of procedures to include attention to the more interpersonal aspects of procedures.” (p. 350). By looking at legal decision-making, Thibuat and Walker focused on formal structures – such as advance notice, neutrality, opportunity to be heard, and accuracy checks such as appeals – in making a measure of procedural fairness. Indeed, Leventhal (1980) expanded the Thibaut and Walker inquiry and ascertained what Fondacaro, et al. (2006) call the “6 structural components that every procedure should have: (1) consistency; (2) suppression of biases; (3) accuracy of information; (4) correctibility; (5) representation or voice; and (6) ethicality.” These are structural procedural inputs and form the basis of what I will call “quadrant I” in the 4-quadrant model that I am proposing in this paper. I will explain the 4-quadrant model in more detailing after reviewing the further developments in procedural justice research. 2.

Quadrant II: Relational & Interpersonal Processes; Tyler’s Development of the Group Engagement Model

Tyler and Blader claim that an example of the “shift from an exclusive focus on decision making to a focus that includes attention to the interpersonal quality of the interaction can be found in the literature on voice or process control.” (Tyler & Blader, pp. 350-351). A great deal of empirical research in the social psychological literature has been directed towards attempting to understand whether voice effects are primarily due to instrumental or non-instrumental motivations. An extensive literature review and summary of the early research can be found in Lind, Kanfer and Early’s (1990) discussion and report of their surprising findings that voice effects are so significant that they increase perceptions of procedural fairness even when people cannot express their views until after the decision is made. This effect also suggests that voice may somehow be related to

4   

catharsis (or possibly some form of self-affirmation), and will be discussed in more detail in conjunction with my discussion of self-affirmation theory. Tom Tyler, one of the leading researchers in this area, noted that the early voice research resulted in the group-value theory, the primary contribution of which was to help us realize that non-instrumental factors influence judgments about procedural justice. Along similar lines, was the parallel and subsequent development of the authority-relations model, which posits that relational cues from procedural treatment by authorities provide important information about social standing and influence perceptions of authority legitimacy vis-à-vis neutrality, trustworthiness and standing. (Tyler & Blader, 2003, p. 352, Table 1). Most recently, Tyler and Blader have posited the “group engagement model,” which they claim “integrates the insights of the earlier group value (Lind & Tyler, 1988) and relational models (Tyler & Lind, 1992) and extends those insights into explanations for why procedural justice shapes cooperation in groups, organizations and societies.” (p. 352). Before examining the experimental and theoretical implications of the group engagement model, it is important to review the predecessor models. The group-value model extends upon the social identity theory hypothesis that people prefer to belong to positively-valued groups or social categories because positive social identities contribute to general feelings of self-worth. (Smith & Tyler, 1996, citing Tajfel & Turner, 1986). Specifically, the model proposes that “treatment by authorities communicates two symbolic messages about group membership” – respect (i.e., that the person is a valuable and respected member of the group) and pride (i.e., that the group as a whole is a worthy group to belong to). (Ibid., p. 147, citing Tyler, Degoey & Smith, 1996). Feelings of value and respect are linked to whether authorities are perceived as trustworthy, neutral and respectful. While it is hypothesized that treatment by authorities communicates information about group memberships that would bolster self-esteem, “evidence for a direct relationship between group status and personal self-esteem has been inconsistent.” (p.149, 5   

citing conflicting studies). Smith & Tyler (1996) suggest this may be due to the failure to draw a distinction between personal and collective self esteem. In this regard they suggest that personal selfesteem measures “feelings of self-worth derived from personal identity,” whereas collective selfesteem focuses on “feelings of self-worth derived from social identity.” (Ibid., p. 150,citing Crocker & Luhtanen, 1990).3 The relational model of authority (Tyler & Lind, 1992) builds on this thesis and asserts that because voice procedures are usually implemented by authorities, and these authorities are perceived as representatives of the entire group; then an authority, by using a fair procedure, communicates that recipients are respected, and thus included, within the social group, which has implications on self-worth (Van Prooijen, et al. 2005). Unfair procedures, in contrast, communicate disrespect and exclusion from the relevant group. (Ibid.) The model would thus predict a difference in interpretation of feedback from in-group versus out group authority figures. It was this hypothesis that was tested by Smith, Tyler, Huo, Ortiz & Lind (1998). Smith, et al., (1998), experimentally examined treatment by both in-group and out-group authorities. They found that negative treatment by an “in-group” authority had an effect on feelings of respect and a marginal effect on self-esteem (no effect in study 1; p<.05 in study 2). They also reported “unexpectedly,” that “rude treatment by an out-group authority led to feelings of greater respect than did polite treatment, but only when outcomes were negative.” (p. 483). They speculated that this result might be due either to reactance or that “when the authority represents and out-group, alternative sources of self-validation may be more salient and available to people” due to the opportunity to “affirm their competence and integrity in some unrelated area as a method for diffusing feelings of dissonance.”

                                                             3

 In this regard, Smith & Tyler (1996) found in their studies that both pride and respect correlated with collective  self‐esteem, but only respect correlated with personal self‐esteem. 

6   

These finding also suggests the involvement of self-affirmation processes – an observation I will return to later in this paper. At this point, it is sufficient to note that the group-value and authority relations models operate exclusively within Quadrant II of what I call the 4-quadrant model of procedural justice. I will further describe the 4-quadrant model later, when describing my proposed research program. 3.

Quadrant III: Decision and Resolution of Uncertainty

One finding that Smith and Tyler (1996) noted was “unexpected” was the influence of outcome favorability (rather than treatment quality) on self-esteem. (p. 484). But they also noted, citing Van den bos 1997, that it was possible that the reporting of outcome results before procedural information might have increased the significance of the outcome effect. This result might not be surprising given the predictions of uncertainty management theory. Uncertainty management theory hypothesizes that the costs of social isolation are significant. Accordingly, before self-identifying with a group we assure ourselves of the “fairness” of the procedures to defend against arbitration exclusion from the group. (Thau, et al., 2007). According to uncertainty management theory (“UMT”) people rely on fairness most when they are confronted with uncertainty. The central tenant of uncertainty management theory is that people have a fundamental need to feel certain about the world and their place within it, that stable self-views and constructs help maintain certainty and that people avoid, eliminate or create cognitive defenses to the self-related threats posed by uncertainty. (Van den Bos, 2001; Vand den Bos & Lind 2002). Van Den Bos, et al, 2007, note “many uncertainty management theories propose that people want to protect themselves from being in or thinking of situations in which they were uncertain about themselves. One way in which people can do this is by adhering to cultural norms and values.” (p. 202). Experiences supportive of worldviews, they suggest, lead people to be less uncertain or better able to tolerate the uncertainty. The theory thus hypothesizes that when 7   

uncertainty is made salient, certainty-supportive procedures such as voice (and fair process generally) are desired and appreciated. Unfair events, in contrast, are certainty-threatening and thus elicit highly-negative affect and evaluation. (Ibid.) This line of research has shown that reminding people of personal uncertainty lead to more intense and valenced reactions to procedural justice manipulations. (Ibid, citing DeCremer & Sedikides, 2005) There is even research starting to investigate whether these kinds of threats activate a neurological “human alarm system.” (p. 203) To the extent uncertainty stems from a challenge to one’s moral values and beliefs, selfaffirmation may play a role in either eliminating, alleviating or reversing feelings of uncertainty. It does not appear that any research has been conducted specifically in the procedural justice arena relative to uncertainty and affirmation. Some research has suggested that having the opportunity to voice one’s opinion has a stronger impact when there is high level of uncertainty about the trustworthiness of the authority. In the 4-quadrant model uncertainty plays a strong role in quadrant II and possibly in quadrant I. By the time the decision is announced (Quadrant III), all uncertainty is alleviated. According to UMT, one function of fairness is to reduce uncertainty in people’s lives. (Lind & Van den Bos, 2002; Van den Bos & Lind, 2002). Thus, UMT theorists predict that “people experiencing high uncertainty should be motivated to direct more of their attention to fairness cues in their environment than should people who experience low uncertainty.” (p. 252). UMT posits that sources of uncertainty can be either environmental or self-imposed. An example of environmental uncertainty is “job termination.” Self-imposed uncertainty, in contrast, involves situations where people are uncertain in relevant self-domains such as competence or interpersonal relations, such as love and respect. Social comparison theory suggests that people who are uncertain about themselves possess a greater propensity to evaluate their beliefs, attitudes and behaviors relative to others. (Thau, et al., 8   

2007, p. 252). The “social comparison orientation” construct was designed to measure this tendency. A person high in social comparison is “sensitive to the behavior of others and has a degree of uncertainty about the self, along with an interest in reducing self-uncertainty.” (Gibbons & Buunk, 1999, p. 138). 4.

Quadrant IV: Behavior and Legitimization

Quadrant IV focuses on a key attitudinal and behavioral consequences of procedural justice evaluations: legitimacy and legitimation. Tyler (2006) explains, “legitimation refers to the characteristic of being legitimized by being placed within a framework through which something is viewed as right and proper.” (p. 376). Legitimation is crucial because it is a key moderator of the degree to which people will accept and comply with decisions. Nevertheless, until recently legitimation has been neglected in the social psychology field. (Tyler, p. 379). Tyler describes legitimacy as a “system level attribute,” which means Quadrant IV behaviors feed back into the Quadrant I structural inputs by providing a “reservoir of support” for existing authorities, groups and institutions. (p. 381). In this regard, Tyler makes a strong assertion that is one of the core inspirations for the proposed research program. Tyler states: In legal settings, people are found to be more willing to defer to the decisions of formal and informal legal authorities when those decisions are made fairly. (citations omitted). One important recent development is the widespread use and study of restorative justice conferences in lieu of trials. These conferences, which are experienced by those involved as procedurally fairer than trials, lead to greater cooperation with the law. This finding mirrors the earlier finding that mediation, which is viewed as procedurally fairer, leads to greater decision acceptance by disputants. (p. 379-380). There are some bold assertions here. First, the view that mediation is viewed as procedurally fairer needs further examination. Mediation, for instance, is non-binding – so there may be less uncertainty and a greater sense of control that explains satisfaction rather than voice. No direct comparisons have been made between mediation and arbitration relative to voice from the 9   

theoretical perspective of uncertainty management theory. Second, the assertion that courtsponsored settlement conferences are experienced as procedurally fairer should also be subjected to empirical scrutiny. Oftentimes, these proceedings are dominated by lawyers and judges who leave litigants feeling “pressured” to settle under fear of the great “uncertainty” of trial. Ironically, satisfaction for mediated settlements is increased by de-legitimizing status quo trial procedures. Although the quoted statement is limited to “legal settings,” Tyler significantly broadens his assertions when he and Blader write: “Those interpersonal aspects of procedures arise because procedures are settings within which people are involved in a social interaction with one another. This is true regardless of whether the procedure involves bargaining, a market exchange, team interaction among equals, or a third party procedure with a decision maker, such as mediation or a trial.” (Tyler & Blader, 2003, p. 350). This too is a strong and interesting assertion. Nonetheless, the focus of the claim involves only the “interpersonal” – what I have dubbed Quadrant II – interactions. My proposed research would test this theory by varying the Quadrant I structural inputs and measuring Quadrant II interactions. Before setting forth the specific research proposal, I want to further describe the Four Quadrant Model and some of the hypothesized interactions. II.

THE FOUR QUADRANT MODEL OF PROCEDURAL JUSTICE The four-quadrant model was inspired by and attempts to expand upon the “four

component” model proposed by Blader and Tyler (2003). The four component model contemplates two orthogonally related dimensions – procedural function and source. This result in four components: (1) formal decision making, (2) formal quality of treatment; (3) informal decision making and (4) informal quality of treatment. By “formal,” Blader and Tyler mean abstract and structural. In this regard they note, “Importantly, the other two components of the model have not been explicitly recognized in the literature. One is the formal quality of treatment, or the influence of structural factors on the quality of treatment experienced in the context of one’s group 10   

membership.” (p. 749). By structural factors, Tyler and Blader indicate that they mean things like the “Bill of Rights.” A. Quadrant I Quadrant I examines structural inputs. Recall from the above discussion that these inputs may be altered by aggregation of the individual legitimization decisions and behaviors that may or may not provide a sufficient reservoir of support to sustain the decision-making body or process. Overall, these inputs are abstract, universal and idealized. They include core “values” and norms like conceptions of due process (notice and opportunity to be heard) or the right against selfincrimination. But they also include the specific form of resolution procedure, whether it is a union grievance hearing, a private arbitration, a court-annexed mediation, a jury trial, a bench trial, or a meeting with a professor or university ombudsperson. The structural inputs sometimes carry with them written procedures, or unwritten norms; they can be private (like ADR), or public; sometimes they can occur in on-line or computer mediated environments. In any event, systematic investigation needs to be made regarding the structural inputs involved in any given procedural justice determination. Structural inputs also include variables like number of disputants, a factor I will discuss in further detail when describing pmy roposed research. Structural inputs are almost entirely formal. Quadrant I encompasses the “formal quality of treatment” component that Tyler has suggested is currently understudied. (Blader & Tyler, 2003). Formal quality of procedure can be operationalized in codified rules and procedures such as the American Arbitration Associations Rules for Consumer Disputes, the Federal Rules of Civil Procedure, or grievance-resolution procedures required by a collective bargaining agreement. These rules vary to the extent they provide structural procedural protections such as right to cross examine, advance notice of charges and proceedings, rights to take pre-hearing discovery (which often does not exist in arbitration, for instance), and formal rules of evidence (which often structurally constrain 11   

voice). To a large extent, structural components also capture the six components described by Leventhal (1980). B. Quadrant II Quadrant II is where the decision-making process itself occurs. It is the quadrant in which almost all existing research – especially the group relations model proposed by Tyler – has been conducted. It is here that the disputant(s) interact with the decision-maker. It is here that “authority” figures supply the informational cues that group engagement theorists suggest are so important for self-constructs. It is here that people are either treated politely or rudely, listened to or ignored, treated neutrally or in a biased manner. It is here that pure procedural judgments are formed. There are both formal and information aspects to the Quadrant II interaction. Depending on structural inputs the decision maker may (or may not) apply differing degrees of treatment. Some decisions are discretionary and ad hoc (and thus informal treatment under Tyler’s model) and others are formally required. Tyler’s formal and informal treatment components both operate within this Quadrant. One alteration and suggested improvement I believe the four-quadrant model makes over the four-component model is that is places both formal and informal treatment within 1 quadrant. While many Quadrant II interactions are understood as a result of extensive existing research, emerging questions remain. Significantly, I believe in the area of uncertainty management theory. UCM effects occur to some extent in Quadrant I (people are generally apprehensive going into to a procedure, but their concerns may (or may) not be alleviated by clearly set forth and disclosed rules and protections. But when people actually enter into the process itself, and are interacting with the decision-maker (Quadrant II), uncertainty is expected to be at its highest. Thus when people cross the threshold from Quadrant I into Quadrant II is a robust area to refine some 12   

of the existing questions being posed by uncertainty management theorists. This is further discussed in part III of the paper, below, where I set forth my research proposal. C.

Quadrant III

Quadrant III involves the announcement and justification of the decision. I have not found any research that examines the process/effect of justification as a variable. Nonetheless, justification can be hypothesized to be a very important part of the procedural justice model. It is here that outcomes also come into play – the outcome is announced and it is either favorable or unfavorable. There are four primary conditions that can occur in Quadrant III: a favorable decision with a justification; a favorable decision without a justification; an unfavorable decision with a justification and an unfavorable decision without a justification. Previous research that looked at the effect of outcome (e.g., Tyler & Lind, 1992) did not explicitly manipulate and examine a “justification” variable. Justification may play a very important role to the extent it interacts with and mediates perception of outcome. Favorable outcomes with a justification may have a different cognitive, affective and behavioral consequence than a favorable outcome alone because of the “affirming” or esteem-boosting effect of the favorable justification. Unfavorable outcomes with no justification may leave disputants feeling angry and robbed. Justification may play an important role in legitimating unfavorable outcomes, but the justification may also be linked to the perceived trustworthiness of the authority based on Quadrant II interactions. Unfavorable outcomes with justification are a particularly interesting Quadrant III interactions because they can be construed as communication of a persuasive message from the decision maker to the “losing” disputant. And as such, research would be informed by all the existing literature on phenomena such as reactance or cognitive processing under the elaborationlikelihood model of persuasion. Acceptance of the validity/persuasiveness of the proffered 13   

justification is a Quadrant III measure that could be developed and would be informative in a variety of contexts. Quadrant III interactions are also important to the extent the justification has an effect on self-related constructs. For example, a justification made on a credibility basis, such as “I am going to rule for the Defendant because I thought the Plaintiff was a liar,” would be expected to have a dramatically different effect than “I hate to do it, but the law compels me to rule for the Defendant.” There are explicitly-stated evaluative processes in justifications that may have crucial effects depending on both the qualities of the person or entity who makes the justification and individual differences amongst the disputants. Examining justification may also help explain the “unexpected” finding in the Smith, et al., (1998) research about why rude treatment from out-group authority figures might be construed differently only under unfavorable outcomes. There are numerous and interesting experimental manipulations that can be created (such as the you’re-a-liar/bound-by-the-law dichotomy explained above. These justifications might have an effect to the extent they are affirming (yet another point of intersection for Steele’s (1983, 1988) affirmation theories.. Another interesting line of Quadrant III research involves examining justifications from the perspective of self-verification theory. (Wiesenfeld, Swann, Brockner & Bartel (2007). Selfverification theory reasons that “because people with low self-esteem doubt they are worthy of respectful treatment, they are ambivalent about the positive regard conveyed by fair treatment. Consequently, they may fail to show the clear preference for just treatment that people with high self-esteem display.” (p. 1235). Wisenfeld et al. contend that self-verification research is important to help specify the boundary conditions of procedural justice, explore the underlying psychological mechanisms to help disambiguate self-enhancement (e.g., Baumeister, 1989) and self-verification theories and to try to demonstrate links between self-verification and organizational and group 14   

behaviors. (Ibid.) The authors explain that to the extent relational theory posits that procedural fairness communicates self-evaluative information; it may vary on account of individual differences such as self-esteem or other “self-related motives.” This is likely true. But it not just Quadrant II relational cues that may be involved in this process, Quadrant III justifications likely play an important role in confirming or threatening various self-related constructs. Weisenfeld, et. al, noted that self-enhancement and self-verification theories lead to conflicting theoretical predictions with respect to interpersonal (what I call Quadrant II) interactions. There is the compensatory self-enhancement view, represented by sociometer theorists such as Leary and Baumeitster, 2000), which would predict that people become hungry for and seek out positively evaluative information in procedurally fair treatment. (Weisenfeld, et al., 2007). Yet self-verification theory would predict that fair treatment disconfirms self-views in people with low self-esteem and thus “people with low self-esteem should be less inclined than their high self-esteem counterparts to exhibit the typically positive response to procedural justice. (Ibid.) These predictions apply with equal, if not stronger force, in the context of Quadrant III justifications, where information is made explicit. People do not need to seek out cues; the reasoning is stated outright and is either respectful and affirming or demeaning and threatening. (Perhaps is can be truly “neutral,” but that would seem very rare – there is likely a scale that could be developed here.) D.

Quadrant IV

Quadrant IV is where the outward behavioral consequences of the process can be observed. People either accept or reject the decision – and they can manifest their decision in a variety of ways: they can do nothing, they can pay a judgment or force collection, appeal or not appeal, in smaller groups and organizations they can leave the group or remain. These can all be operationalized as measures of behavior that either accept or reject decisions and increase or diminish existing “reservoirs of support.” They are tangible outcome-based behaviors that either legitimate or de15   

legitimate procedural structures, institutions and schemes. Psychological research in this area can provide a crucial link between the psychological and the sociological fields of research in social justice. Moreover, to the extent these legitimizing outputs feed back into the model as structural inputs, we can attempt to alter outputs by manipulating procedures in view of Quadrant II and III research. Self-affirmation theory seems like it can play an important role in all four quadrants of the model, which is why I would propose that a research program be designed from the perspective of self-affirmation theory. A comprehensive literature review of affirmation theory has been done elsewhere, and is beyond the scope of this paper. General understanding of the theory is assumed. I have also commented elsewhere on the paper where I feel affirmation theory may shed some light. In the remaining section of the paper I want to outline my proposed research program and specifically specify some of the theoretical predictions that implicate affirmation theory. I will also note where other theories may play a role in developing and advancing the research program. III.

PROPOSED RESEARCH PROGRAM The Quadrant I focus is on structural inputs. As part of this examination I see several

important variables. First involves the ex ante certainty that is predicted from the procedure. Some procedures, such as mediation, are “high control” and thus more predictable and certain. They are more certain, for instance, because disputants retain control over whether to accept or reject any result, so under no circumstances are participants worse off than they are under the status quo. Other procedures, such as a jury decision would be “low control” because there exist numerous variables, many unpredictable, which play a role in jury decision-making. Moreover, juries unlike judges are not required to explain or justify their decision. This would play an important part in Quadrant III interactions, as the effect of the no-justification decision can be compared to justification scenarios. 16   

The identity of the decision-maker is another important Quadrant I variable. A judge is an authority figure literally clothed in robes that express significant authority at the core level of societal values. The judge is a personification of state sanction. This contrasts with an arbitrator, who is typically a retired judge or private lawyer. Arbitrators do not wear robes, and preside over much more informal proceedings that do not take plain in a courtroom with bailiffs and flags. The salience of the symbolic trappings is much less. Given the reported effects of an experimenter wearing a flag pin (or as in Smith, et al, a Stanford tee-shirt), the core symbolic trappings of a courtroom versus a conference room seem very significant. Arbitrators are interesting for another reason. Under the rules of may arbitral forums the arbitrator is not required to justify their decision. This provides another level of comparison for Quadrant III interactions. For my initial pilot study I would propose four Quadrant I conditions: mediation, arbitration, judge, and jury. Further research might vary these input variables to fit work or union contexts, grade disputes, etc. I believe that the four legal inputs are a good place to start for several reasons. First, the four variables break down between a public and private distinction, with the judge and jury representing traditional, “public” and status quo authority and the mediator and arbitrator representing “private” mechanisms where authority relations may play less of a role. Pre-study perceptual measures make be taken, with an attitudes to law measure that I developed elsewhere. Second, the inputs vary justification in the public sphere between judge and jury and vary locus of control in the private sphere between arbitrator and mediator. These manipulations are hoped to provide data that will help further refine uncertainty management theory, as well as provide a starting point to formulate further hypothesizing about Quadrant III interactions. Third, legal input variables allow similar disputes to be framed in both gain and loss frames within the context of a similar dispute. Van Prooijen, Karremans & van Beest (2006) have recently 17   

noted that “people’s more positive responses to the grating, as opposed to the denial of voice have been described as one of the most robust findings in procedural justice literature, and evidence for these voice effects has been found in both applied and experimental settings.” (p. 686). Nonetheless, the note, most of these studies have been done in a “gain-framed” rather than a loss-framed context. The only cited exception was Brockener et al.’s (1998), notable study on procedural justice in the context of plant layoffs. Most studies use pay distributions, student grades and lottery tickets as the manipulated “outcome” variable. Yet all of these outcomes are gain-framed. Relying on the hedonic principle and Higgin’s (1997) research on the manner in which approach or avoidance motivational state affects the salience and cognitive processing of socially congruent (or dissonant) information – the authors hypothesize that significant differences may occur in loss-framed contexts. The advantage of using legal inputs is that the same fact pattern is a gain-frame from the perspective of the Plaintiff, and a loss frame from the perspective of the Defendant. Pre-Study Measures: Initially, I propose a series of pre-study measures; including, the Rosenberg Self-Esteem scale, and attitude towards law measure (which includes an authority subscale) and a newly-devised measure of attitudes toward dispute resolution procedures using the Leventhal (1980) criteria. People would also be asked about their views regarding the strength of the dispute and their certainty of winning. Disputes would be drawn from small claims courts in Chicago, Illinois where people are already required to participate in court-annexed non-binding arbitration. All would be given the opportunity to participate in a “free” pre-trial mediation event (free in the sense that they are providing data in exchange for the services of a free mediator). Some participants will be selected to participate in a non-binding “jury” procedure that would be filmed and conducted as a focus-group. This would be a substitute for the non-binding arbitration already offered by the Cook County Judicial System, but would allow for the same kind of acceptance versus 18   

rejection measures. These attitudes might also provide insight into the extent to which participates possess or relate to the core procedural values of the society providing a baseline against which affirmations could be assessed and compared. Disputes would also be characterized along two orthogonal dimensions, involving dignitary torts (such as defamation or injury) or contract/commercial disputes and categorized along an amount-in-controversy. This data is already maintained by the Circuit Court of Cook County from which the subject pool would be recruited. Experimental Manipulations: Disputes would be handled initially by mediators and either resolved or not resolved. Outcome-satisfaction measures would be administered based on resolved disputes and statistical comparisons analyzed along a post-dispute satisfaction measure. Some disputes would not be resolved by mediation and would continue through the experimental process. A “resolution” measure would also be taken that would compare mediation experiences along existing group-value dimensions (e.g., voice, polite treatment, perceived neutrality of the mediator). Some existing hypothesis could thus be replicated and further explained. Admittedly, some people by choosing to accept mediation would self-select out of the remaining process, nonetheless valuable insights could be gained. People that remain in the process would then participate in the non-binding adversarial phase within either arbitration or mock-jury conditions. The justification variable could be satisfied by asking arbitrators to either explain their decisions or simply (as is currently done under the status quo) announce their decisions. Here, the effect of justification could be analyzed. Because decisions are still non-binding at this phase, acceptance or rejection of the award would provide a valid postmanipulation measure. After the proceeding an uncertainty measure could be taken to further test and refine the predictions made by uncertainty management theory. Levels of uncertainty could be compared to the pre-hearing measure and contrasted regarding the effect of the manipulations of independent 19   

variables such as justification. It is at this time (post-hearing/pre-decision) that uncertainty is expected to be at a high point. Crossing the threshold from Quadrant II into Quadrant III resolves the uncertainty, as the outcome is announced. It is here that cognitive processes and specific processes related to selfconstructs are most likely to come into play. It is here that I would also apply and test the crucial effect of a self-affirmation. Affirmation research suggests that affirmation in unrelated domains help buffer the self-system against threats from dissonant messages, health-threats and other negative stimuli that are perceived to threaten the self-system. Presumably, adverse outcomes (especially those justified and legitimized by negative, or self-threatening, justifications) would be buffered if the predictions of affirmation theory are accurate. Thus, affirmations would be expected to increase both decision acceptance and the perceived legitimacy of the rationale. Comparisons could be made within condition based on individual differences like self-esteem and attitude toward law, as well as across conditions related to degree of uncertainty. For the first time, an effect of stating legitimizing rationale could also be detected. Because these are non-binding there would still be a high degree of control across all conditions. Control is relinquished in stage 3 of the study when people proceed to their actual trial. In the small claims division there are both bench and jury trials, significant percentages of which actually proceed to trial relative to other disputes in the Circuit Court of Cook County. (This is a function of both amount in controversy and volume). Thus, final comparisons could be made in comparing satisfaction with the experimentally manipulated variables and actual real world outcomes. This phased combination of a field experiment that allows some degree of variable manipulation, in the context of real-world disputes, would be expected to have a more significant degree of external validity relative to a traditional laboratory experiment. Although research within particular phases or quadrants could be more directly examined in lab contexts. 20   

The primary findings I expect to emerge would be consistent with self-affirmation theories predictions – people who affirm will be more likely to accept outcomes across all conditions, but the effect would be particularly strong in the negative-outcome, hostile justification context. Uncertainty management theory also predicts some confounds, but nonetheless the phased design across quadrants allows for systematic and periodic isolation and examination of the effect of this variable. The use of legal scenarios creates a gain-frame for the plaintiff and a loss-frame for the defendant, which allows for further empirical testing of the concerns raised by Van Prooijen, et al. (2006). Also, the use of the four quadrant models allows for a superior method of specifying, isolating and situating various research contexts. It seems an improvement on Blader and Tylers (2003) four component model because it expressly incorporates structural inputs and sees the justice system operating at a level where legitimization outputs feed back into system inputs, thus creating a more dynamic model. The lessons of justice research over the past 20 years are as follows: it is not so easy to separate the procedural and distributive justice effects, any meaningful model for further research needs to integrate both. I propose a research program that returns to the “old school” roots of the original Thibuat and Walker studies. We should embrace, not reject, legal decision-making frameworks in advancing justice research. Legal decision making frameworks allow for close examination of both outcome-oriented and procedural variables and see how they interact. The function of “authority” is central in justice research, it forms the core of Tyler’s authority model as is central to “interactional” models as well. Unlike in the 1970s we have seen a rise in ADR, both private and court annexed. Judges, as authority, preside over both trials and inchambers mediations. Private mediators are deciding binding disputes. The difference between black-robed judges and private arbitrators is a good way to attempt to test relational theories.

21   

The proposed research program allows us to look at distinction between disputes, such as dignitary torts (like slander) versus contract disputes, which are more relational in nature. The varying amounts in controversy also allow for statistical manipulations based on level of threat – there is a big difference between a small claim and a foreclosure (or ‘bet-your-business’ case – both of which are in extreme demand. A back-to-basics approach will also help shed some light on the more cutting-edge issues meeting existing calls for further research. 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., & Shapiro, D.L., (1987). Interactional fairness judgments: The influence of causal accounts. Social Justice Research, 1, 199-218. Blader, S.L. & Tyler, T.R. (2003). A four-component model of procedural justice: Defining the meaning of a ‘fair process.’ Personality and Social Psychology Bulletin, 29, 747-753. Brockner, J., Heuer, L., Siegel, P.A., Wiesenfeld, B., Martin, C., Grover, S., Reed, T., & Bjorgvinsson, S. (1998). The moderating effect of self-esteem in reaction to voice: Converging evidence from five studies. Journal of Personality and Social Psychology, 75(2), 394-407. Cohen, G.L., Sherman, D.K., Bastardi, A., Hsu, L., McGoey, M., & Ross, L. (2007). Bridging the partisan divide: Self-affirmation reduces ideological closed-mindedness and inflexibility in negotiation. Journal of Personality and Social Psychology, 93 (3), 415-430. Daly, J.P. & Tripp, T.M. (1996) Is outcome fairness used to make procedural fairness judgments when procedural information is inaccessible? Social Justice Research, 9(4): 327-349. De Cremer, D., & Sedikides, C. (2005). Self-uncertainty and responsiveness to procedural justice. Journal of Experimental Social Psychology, 41, 157-173. Fondacaro, M.R., Brank, E.M., Stuart, J., Villanueva-Abrahan, S., Luescher, J., McNatt, P.S., (2006). Identity Orientation, Voice and Judgments of Procedural Justice During Late Adolescence, Journal of Youth and Adolescence, 35 (6), 987-997. Folger, R. (1984) Emerging issues in the social psychology of justice. In R. Folger (Ed.) The sense of injustice: Social Psychological perspectives (pp. 3-24). New York: Plenum Press. Folger, R., (1996). Distributive and Procedural Justice: Multifacted meanings and interrelations. Social Justice Research, 9(4): 395-416. Gibbons, F.X., & Buunk, B.P., (1999). Individual differences in social comparison: Development of 22   

a scale of social comparison orientation. Journal of Personality and Social Psychology, 76, 129-142. Howard, E.S., Gardner, W.L. & Thompson, L. (2007). The role of self-concept and the social context in determining the behavior of power holders: Self-construal in intergroup versus dyadic dispute resolution negotiations. Journal or Personality and Social Psychology, 93(4), 614-631. Hui, M.K., Au, K. & Zhao, X., (2007). Interactional justice and the fair process effect: The role of outcome uncertainty. Journal of Experimental Social Psychology, 43, 210-220. Lind, E.A., & Tyler, T.R. (1988). The social psychology of procedural justice. New York: Plenum Press. Lind, E. A., Kanfer, R., & Earley, P. C. (1990). Voice, control, and procedural justice: Instrumental and noninstrumental concerns in fairness judgments. Journal of Personality and Social Psychology, 59,952-959. Smith, H. & Tyler, T.R., (1996). Choosing the right pond: The impact of group membership on selfesteem and group-oriented behavior, Journal of Experimental Social Pyschology, 33, 146-170. Smith, H., Tyler, R., Huo Y., Ortiz, D., & Lind, E., (1998). The Self-Relevant Implications of the Group-Value Model: Group Membership, Self-Worth, and Treatment Quality, Journal of Experimental Social Pyschology, 34, 470-493. Steele, C.M., & Liu, T.J. (1983). Dissonance processes as self-affirmation. Journal of Personality and Social Psychology, 48, 18-34. Tajifel, H., & Turner, J. (1986). The social identity theory of intergroup behavior. In S. Worchel (Ed.) Psychology of intergroup relations. Chicago: Nelson Hall. Thau, S., Aquino, K. & Wittek, R., (2007). An extension of uncertainty management theory to the self: The relationship between justice, social comparison orientation, and antisocial work behaviors. Journal of Applied Psychology, 92(1), 250-258 Thibaut, J. & Walker, L. (1975). Proedural Justice: A psychological analysis. Hillsdale, N.J. Erlbaum Tyler, T.R., & Blader, S.L., (2003). The group engagement model: Procedural justice, social identity, and cooperative behavior. Personality and Social Psychology Review, 7, 349-361. Tyler, T.R., Degoey, P., & Smith, H. (1996). Understanding why the justice of group procedures matters: a test of the psychological dynamics of the group-value model. Journal of Personality and Social Psychology, 70, 913-930. Tyler, T.R., & Lind, E.A. (1992). A relational model of authority in groups. Advances in Experimental Social Psychology, 25, 115-191. Tyler, T.R., (2006). Psychological Perspectives of Legitimacy and Legitimation, Annual Rev. Psychol. 57, 375-400. Van den Bos, K., Ham, J., Lind, E.A., Simonis, M., van Essen, W.J., & Rijpkema, M., (2007). Justice and the human alarm system: The impact of exclamation points and flashing lights on the justice 23   

process. Journal of Experimental Social Psychology, 44, 201-219. Van den Bos, K. (2001). Uncertainty management: The influence of uncertainty salience on reactions to perceived procedural fairness. Journal of Personality and Social Psychology, 80, 931-941. 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). San Diego, CA: Academic Press. Van Prooijen, J., Karremans, J., van Beest, I. (2006). Procedural Justice and the Hedonic Principle: How approach versus Avoidance Motivation Influences the Psychology of Voice, Journal of Personality and Social Psychology, 91(4), 686-697. Van Prooijen, J., Van den Bos, K., & Wilke, H.A.M., (2005). Procedural justice and intragroup status: Knowing where we stand in a group enhances reactions to procedures. Journal of Experimental Social Psychology, 41, 664-676. Wiesenfeld, B.M., Swann, W.B., Brockener, J., & Bartel, C.A. (2007). Is more fairness always preferred? Self-Esteem moderates reactions to procedural justice. Academy of Management Journal, 50(5), 235-253.

24