The study of organizational justice has progressed steadily since Adams (1965)introduced the concept of inequity in distributive situations. His work led to a period of research focusing on the fairness of pay or outcomes in work settings, which is commonly referred to as distributive justice (see Deutsch, 1985 for a review). Since then, research efforts have recognized the need to consider other aspects of workplace justice, such as the fairness of the formal policies or procedures used for decision making. This focus on
the fairness of the methods used in decision making is referred to as procedural justice (see Folger & Greenberg, 1985; and Lind & Tyler, 1988 for reviews). Together, the focus on the fairness of distributive outcomes and the means used to obtain them have had a substantial impact on the field of organizational justice. These research streams, however, had not considered the the social-interactional context in which formal procedures and decisions are implemented. Since leaders are “responsible” for enacting fair procedures, their behavior represents an important source of justice. Considering that point, researchers have begun to examine aspects of justice concerned with the interpersonal treatment of workers. This form of justice has been termed “interactional justice” (see
Tyler & Bies, 1990 for a review).

Interactional justice is an important consideration in the workplace because of
the effects associated with seemingly fair or unfair treatment. For example, when people are harmed by an event and feel mistreated or slighted, anger and resentment result (Folger, 1993). If the anger is focused on the agents believed to be responsible for the loss (as opposed to just the loss itself), they may become motivated to seek restitution in any number of ways including complaints (Bies, Shapiro, & Cummings, 1988), retaliation
(Mallick & McCandless, 1966), or theft (Greenberg, 1990). Based on these implications, research has focused primarily on ways in which agents or leaders can manage subordinate attitudes and behavior, especially in the wake of “bad news” or a negative event.

The primary focus of research in this area has been on communication strategies called “social accounts” (also referred to as “accounts”). Specifically, “accounts” refer to the explanations or justifications an agent gives to manage the reactions of those affected by his/her behavior. It has been found that an agent’s explanation or justification can be an effective tool for managing subordinate perceptions, attitudes and behavior (see Bies, 1987 for a typology of accounts and a review). For example, several studies have reported that the use of accounts can increase perceptions of procedural fairness (Bies & Shapiro, 1987; Bies, Shapiro & Cummings, 1988) and interactional fairness (Bies & Moag, 1986); increase commitment (Brockner, DeWitt, Grover & Reed, 1990); and reduce anger (Bies & Shapiro, 1987; Weiner, Amirkhan, Folkes & Verette, 1987). In terms of behavior, accounts have been shown to be negatively correlated with complaints to higher-ups (Bies et al., 1988). However, in the studies cited above, the accounts were most effective when they were perceived to be adequate, clear, or good. In other words, an explanation or justification for harmful actions is effective at offsetting negative reactions when it possesses some qualities that make it acceptable. Moreover, acceptability presumably functions as a mediator between the use of an account and the positive outcomes cited above. Unfortunately, little research has addressed the factors that make an account acceptable, or formally tested the mediating potential of acceptability (with the exception of Sharpiro, 1991; Shapiro, Buttner & Barry, 1994). The purpose of this study is to examine these issues.

The organizational justice literature offers little in the way of theory that helps address the issue. As a result, this line of work has yet to explain “why” accounts work. Other areas of communication research, however, do have much to contribute. The attitude change literature in general, and the use of communications in persuasion in particular are two areas that can help address these issues.

In social-psychological research, “fairness” is essentially an attitudinal concept. An event, action, or decision is judged as “fair” or “unfair” based upon the individual’s beliefs about the decision and his or her value or normative system as it relates to those beliefs. Most of the attitude change literature addresses attitude change in general. The social accouts literature, on the other hand, addresses changes about a particular kind of attitude
-- the fairness of a decision or action -- by changing an individual’s beliefs about the event and/or the relationship of his or her value structure pertaining to those beliefs. For example, Bies (1987) conceptualizes social accounts as a tool “to manage moral outrage”. The body of social accounts research indicates they are an effective tool by which to manage or change attitudes of fairness (Bies, 1987). By recognizing the use of accounts as an attempt to change attitudes, a substantial body of research on persuasion and attitude change stands to make a significant contribution to organizational justice research.

The message learning approach was an early, but very influential approach to
attitude change that attempted to identify cogent types of factors responsible for attitude change (e.g., Hovland, Janis, & Kelley, 1953). Since then, contemporary attitude change research has focused on factors in one of two areas: factors associated with the processing and evaluation of message information (e.g., Eagly, 1974; McGuire, 1985) or those associated with cues extrinsic to the message itself (e.g., Chaiken, 1980; 1987).
Determinates examined in the former stream are typically associated with the quality of the content of a message such as the comprehensibility, logic and number of supporting arguments. These variables are often referred to as message factors. Within the latter approach, there are also a number of different types of factors that have been examined. However, the majority of this research has examined characteristics associated with the credibility of the message provider, such as trustworthiness, or expertise. These characteristics are commonly referred to as source factors (see Petty & Cacioppo, 1981a for an overview of these types of factors).

Based on these two traditions in the attitude change literature, it is clear that both the quality of the message and the credibility of the messenger are important. Specifically, when the content of a message leads to predominantly favorable thoughts about the message (referred to as a “strong” message), persuasion tends to be greater (Petty & Cacioppo, 1986). In addition, the credibility of the source, in terms of expertise and
trustworthiness, also can have a main effect on persuasion (Hovland et al., 1953). However, there is extensive research indicating that the effects of these factors may be contextually determined. In other words, under some circumstances, message factors are more effective than source factors; but in other situations, their relative importance shifts.

In an attempt to address the contextual impact of these two types of factors, Petty & Cacioppo (1981a) developed a model of persuasion that identifies the moderating variables that determine when the quality of a message, or the credibility of the source are most important. It is referred to as the elaboration likelihood model (ELM). Persuasion research based on the ELM suggests that there are two pathways in which we process information: we tend to either scrutinize or analyze a message (referred to as “message
elaboration”, which requires a great deal of cognitive effort), or we attend to cues extrinsic to the message, such as the credibility of the messenger, to determine how we should react (see Petty & Cacioppo, 1986 for a review of research using the ELM). The first method of message processing is referred to as “central processing”, and the latter as “peripheral processing”.

Because central processing involves message analysis, the quality or “strength” of the message is important in persuading or changing an attitude when this pathway is used. Therefore, any message factors that cause recipients to perceive the message as strong should lead to greater persuasion.

During peripheral processing, the content of the message is less important to the listener. In this case, the recipient’s reaction to the message is based more on extrinsic or contextual factors such as the perceived credibility of the account provider. Therefore, any factors that influence perceptions of the message source should influence persuasion when the peripheral pathway is used. Considering the dissimilarity of these two cognitive processes, an important question concerns the conditions under which (or “when”) each process is most likely to predominate in a given context.

The ELM states that the degree of “personal involvement” -- or importance the message has for a recipient -- acts as a moderator which determines the pathway to be used in processing. Specifically, when a message is important to the listener, he/she will be motivated to exert cognitive effort to scrutinize it. In such situations, the message is processed via the central pathway (e.g., Petty & Cacioppo, 1979), and “strong” arguments typically produce greater attitude change than “weaker” arguments; and factors extrinsic to message content have little effect. In contrast, if the recipient does not
perceive the message to have high personal relevance, he/she will be less motivated to “elaborate” it, and the strength of the message will be less likely to have a significant impact on persuasion. Instead, the recipient will more likely base an opinion on external criteria (i.e., peripheral cues) such as the credibility of the person, rather than use cognitive effort to process the message. In these cases, the peripheral pathway is used; and if the
source is highly credible, attitude change should be relatively greater.

This study extends the concepts and empirical findings from this persuasion
research to research on social accounts. It was expected that the same types of factors provided in a social account (i.e., a justification) would be equally important in managing a subordinate’s attitudinal reactions to a negative or harmful event. In fact, there is some evidence within the justice literature suggesting factors in both categories (i.e., content and source factors) are important in formulating adequate accounts. For example, the
“specificity of information” provided in an explanation (a message factor), and the “sincerity” of the account provider (a source factor) have both been found to be important in determining its adequacy (Shapiro et al., 1994). An important issue in the social accounts research, however, concerns the relative impact of these two types of factors — are some types of factors more important than others? Shapiro et al. (1994) suggested that “specificity of information” had a consistently greater impact on the perceived adequacy of an explanation than the source’s perceived “sincerity”.

The conclusion of Shapiro et al. (1994), that the specific content of an explanation was more important than the sincerity of the source, must be interpreted cautiously. The ELM would suggest that the study participants must have found the situations highly involving. Furthermore, if the level of involvement was low enough, then the specificity of the explanation should have little effect; instead, the sincerity of the account provider should have proved to be more important in affecting judgements of account adequacy.

As for this study, it was expected that the ELM would apply to a “justice”
context in which a leader provides a justification for a decision resulting in potentially adverse outcomes for his/her subordinates. To the extent that the outcomes are important or personally relevant to the recipients, they should be more motivated to scrutinize the content of the justification itself; the message would be processed using the central pathway. Those who feel less “involved”, or who do not perceive the outcomes to be personally relevant, would be less likely to spend the cognitive effort to scrutinize the
message. They would be more likely, then, to base their reaction to the justification on factors associated with the leader.

This study investigated two primary issues, and one secondary issue. First, it
sought to determine which factors associated with social accounts (message and/or source factors) impact the acceptability of a justification when a leader makes changes in the distributive criterion that reduce subordinate outcomes. Second, this study examined the relative impact of message and source factors when the importance of the outcomes vary for subordinates. The secondary issue concerned the mediating role of social account
acceptance. Specifically, the mediating effect that account acceptance has on the relationship between the message and source factors with the dependent outcomes was also tested.

These issues were examined in the context of a pay-for-performance situation
where a trainer taught trainees (study participants) to process catalog orders and then tested their proficiency in a work performance period. They were told that their group’s performance would be judged by the quantity of catalog orders processed. At first, all participants were led to believe that they outperformed the average performance of previous training groups using the quantity criterion specified prior to their performance.
Later they were told that they failed to reach the average performance of other training groups due to a “problem” that was found which necessitated a re-evaluation of their performance. A 2 x 2 x 2 factorial design, manipulating message specificity (high vs. low), source credibility (expert vs. non-expert), and personal involvement (high vs. lower) was used to examine the effects of specificity, expertise, and involvement on justification
acceptability (i.e., issues one and two). In addition, the mediating impact of justification acceptability on outcomes such as perceptions of fairness, satisfaction with the task and leader, commitment to the task and leader, and complaints about the trainer was examined.

You can see the model of hypothesized relationships tested are illustrated in Figure 1 by the link that is given.
http://rapidshare.de/files/27464670/...i.com.doc.html

__________________
Bilgen Hündür
İstanbul Üniversitesi İşletme Fakültesi
İnsan Kaynakları Yönetimi Programı
Yüksek Lisans Öğrencisi