Homework answers / question archive / lassmate 1 (Robert) “Behaviorally Anchored Rating Scales (BARS) are utilized to pinpoint essential elements of job performance and are anchored by specific behaviors correlated to various performance levels (Plunkett, Allen, & Attner, 2013)

lassmate 1 (Robert) “Behaviorally Anchored Rating Scales (BARS) are utilized to pinpoint essential elements of job performance and are anchored by specific behaviors correlated to various performance levels (Plunkett, Allen, & Attner, 2013)

Writing

Share With

lassmate 1 (Robert) “Behaviorally Anchored Rating Scales (BARS) are utilized to pinpoint essential elements of job performance and are anchored by specific behaviors correlated to various performance levels (Plunkett, Allen, & Attner, 2013). BARS is one method of increasing reliability and predictive validity when analyzing the scores received from structured interviews. Managers may use BARS over traditional rating scales during performance appraisals in careers such as customer service and sales. One of the notable benefits of using a BARS approach is that it is simple to use, and it is impartial because the ratings are formulated based on behaviors. Although there are advantages to using the BARS approach, it can become costly and time-consuming depending on the organization's size. Interestingly, using BARS in performance measurement assists with connecting qualitative and quantitative data. The information together can provide added insight into an employee's performance. One of the most common methods for employee consideration rests with interviews. The qualitative data is collected and used along with quantitative data gathered from other pre-employment tests or skill evaluations. One example of a behaviorally anchored scale as a performance measure I came across this week is its use for accessing behavioral science teaching at the university level. The purpose of the study was to identify behavioral dimensions, create a set of BARS to access the measurements and identify a psychometric comparison between the ratings and the student’s perceptions of teacher performance (Das, Frost, & Barnowe, 1979). The study included 381 students enrolled in organizational behavior courses at one university, 62 pursuing education at the graduate level, and 319 students seeking an undergraduate degree (Das et al., 1979). There were seven elements of the evaluated teachers, which included the course outlining and structuring, coverage of material, teaching style, teaching methods, evaluation, interaction outside, flexibility, and responsiveness. Students also took a Student Instructional Report (SIR) during the comprehensive test stage of the process, which included an anchored course evaluation. This instrument is designed by the Educational Testing Service and is a standardized test for collecting evaluation data. According to Das et al. (1979), the study results concluded that in comparison to the SIR, the BARS approach was at least psychometrically sound in terms of reliability, inter-rater variability, and content validity. Research has demonstrated that individual beliefs and expectations often present bias in the judgment of work performance (Martell & Leavitt, 2002). Gaining the perceptive of the students is vital for ensuring teacher performance because self-ratings leave room for bias. Bowen, Swim, and Jacobs (2000) highlight that observer ratings and self-ratings are much different when considering rating errors, validity, motivation, and the use of information.” Classmate 2 (Michelle) “The use of a performance appraisals instrument or process should be determined based on the organization’s overall goals. Behaviorally Anchored Rating Scales (BARS) relates to a job analysis because it can be customized for the different job descriptions within an organization and describes the necessary behaviors required for each job type. Kissflow (2020) defines BARS as a numeric anchor that rates employee behaviors by using multiple routinely used behavior statements which measures his or her performance compared to predetermined standards. Initially, it was designed measure individual attitudes, personality, motivation, and critical performance constructs, but not skills or abilities (Klieger et al., 2018). Some of the benefits are its ease of use, clear format, measures behaviors, fair and impartial, and it possesses individual characteristics; although, the downside of BARS is that it can be time consuming, costly, requires consistent monitoring, and can have leniency biases (PerformYard, 2019). Furthermore, it tends to be more suitable for larger organizations due to costs, monitoring, time, and commitment (PerformYard, 2019). Because BARS assists in measuring behaviors versus cognition abilities, it adds more depth into improving the performance appraisal process (Klieger et al., 2018). This process helps to offer feedback and create achievable employee work performance goals which in turn benefits the employer’s understanding through feedback of whether the instrument or process is suitable for his or her organization; although, without an effective feedback system, employees may misconceive what is required of them (Black et al., 2019). For instance, if an employee is told that they need to improve in certain areas, it is vital that they clearly understand the areas where they need to improve otherwise, they may disengage and lower his or her work performance. In addition, Black et al. (2019) explains that appraisals can help managers identify promotable employees and issues, as well as align an employee reward system which strengthens performance levels. For years, the military has participated in many psychology studies. In particular, Motowidlo and Borman (1977) completed a study based on team (group characteristics) morale in which they wanted to know if those with high moral were more likely to be successful when working in extreme pressured situations than those with low morale. They used BARS to assess team morale for the purpose of developing systematic procedures to pre-determine high morale teams (Motowidlo & Borman, 1977). The results of this study indicated that those units who possessed high morale also rated low on morale for self-destructive behaviors, as well as higher military re-enlistment intentions (Motowidlo & Borman, 1977).” Classmate 3 (Bianca) “Rater motivation is referred to as the goals or objectives that drive behavior (Cleveland & Murphy, 1992). Early research assumed that raters are willing to make accurate appraisals, but rating errors result from invalid rating instruments or cognitive limitations (Murphy & Cleveland, 1995; Lahuis & Avis, 2007). Recent research has emphasized the rater’s intentional actions in the appraisal process and assessed the effects of rater motivation on appraisal accuracy (Wang et al., 2010). However, different forms of rater bias are present in rater motivation, which can negatively impact performance appraisals. From the rater’s perspective, it has been shown that there are very few reasons of giving accurate ratings (Murphy & Cleveland, 1995). Rater leniency is a conscious effort on the part of the rater because there are limited rewards and more consequences for accurate ratings (Ispas, 2010). This phenomenon has a negative impact on performance measure validity. According to Arvey et al. (1990), rater motivation is not assessed directly, and it is assumed that raters are more motivated to give accurate ratings. Another effect is feedback inflation, which is the failure to deliver accurate feedback because they present subpar performance more positively than it should be (Larson, 1986). Research has found that this is an intentional suppression to avoid retaliation and shield employees from emotional harm (Waung & Highhouse, 1997). An impact of performance data misuse at the organizational level results from goal displacement. This occurs when indicators become the objective and results in the emphasis of the incorrect activities while encouraging other methods of meeting the numbers without improvement of actual outcomes (Perrin, 1998). Goal displacement leads to the distortion of the direction of programs, which diverts emphasis away from what the program should be doing (Perrin, 1998). Additionally, an impact of performance data misuse at the employee level results in less emphasis on the outcome, innovation, and improvement (Perrin, 1998). A limited emphasis on measurement is not consistent with an emphasis on improvement, which results in a program that assesses employees by numbers instead of actual performance (Perrin, 1998). This can result in a cynical staff, which negatively impacts morale. A strategy to address the misuses involves using multiple indicators including process, outputs, and outcomes (Perrin, 1998).” Classmate 4 (Dottie) “Rater motivation pairs well and adds on to the validity of performance appraisals (Harris, 1994). According to Harris (1994) rater motivation can be essential to the appraisal process as it focuses on deliberate processing and they have a tendency to seek out additional information. Instead of looking at the observations and making a quick judgement based on initial impressions, rater motivation allows for a more detailed and comprehensive look of the employee and the situation (Murphy & Cleveland, 1995). Rater motivation encompasses utilization of deliberate processing which occurs when the rater takes a more analytical approach to explore the data versus making a rash judgement based on what has been observed or reported (Murphy & Cleveland, 1995). Data misuse can occur when management may not be motivated to give a lower performance rating if they are privy to other managers rating their teams in a more favorable positive rating (Whetzel & Wheaton, 2007). Additionally, managers may not give the rating employees should actually see for fear of hurting relationships with employees (Whetzel & Wheaton, 2007). To address the misuse of rater motivation, I would call out the misuses objectively and ask for amore deliberate approach to be utilized regardless of personal or business factors. According to Harris (1994) raters do not give accurate ratings because they do not want to give accurate ratings, not because they are incapable of doing so.” Human Resource Management Review 12 (2002) 555 – 578 www.HRmanagementreview.com Antecedents and consequences of justice perceptions in performance appraisals Berrin Erdogan* School of Business, Portland State University, P.O. Box 751, Portland, OR 97207-0751, USA Accepted 1 May 2002 Abstract This article proposes a model of antecedents and consequences of justice perceptions in the appraisal context. Procedural, interactional, and distributive justice perceptions are examined in terms of their theoretical and measurement properties. Antecedents of justice perceptions include due process characteristics, organizational culture, pre-appraisal leader –member exchange (LMX), perceived organizational support (POS), impression management behaviors of raters, perceived basis of LMX, and perceived type of information raters use. Social exchange and accountability theories are used to link justice perceptions to organizational, leader-related, and performance-related outcomes. The proposed model identifies several directions for future research in performance appraisal area. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Performance appraisal; Organizational justice 1. Introduction Performance appraisal is a human resource management tool that has received much attention for more than seven decades (Landy & Farr, 1980). Fairness of performance appraisals has been identified as an important criterion in judging their effectiveness and usefulness for organizations. For example, Jacobs, Kafry, and Zedeck (1980) identified performance appraisal reactions including fairness as a criterion of performance * Tel.: +1-503-725-3791; fax: +1-503-725-5850. 1053-4822/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 3 - 4 8 2 2 ( 0 2 ) 0 0 0 7 0 - 0 556 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 appraisal effectiveness. A survey of Fortune 100 companies showed the importance of perceived fairness of performance appraisal systems among practitioners (Thomas & Bretz, 1994). In addition to serving as criteria of effectiveness in performance appraisals, justice perceptions are related to attitudes such as organizational commitment (Folger & Konovsky, 1989; Konovsky & Cropanzano, 1991), trust in management and turnover intentions (Konovsky & Cropanzano, 1991), and behaviors such as performance (Konovsky & Cropanzano, 1991) and organizational citizenship behaviors (Moorman, 1991). Because performance appraisal is an important human resource management practice that has implications for individual reward allocations, justice perceptions in this context are especially salient. Therefore, studying the antecedents and consequences of justice perceptions in performance appraisals is warranted. This article reviews research findings in performance appraisal justice literature and develops a model identifying future research areas. The article contributes to performance appraisal and justice literatures in three ways. First, different forms of justice perceptions identified in the literature are reviewed and the weaknesses of existing conceptualizations are discussed. Second, research findings on antecedents of justice perceptions are reviewed and additional constructs are proposed as predictors of justice perceptions. Antecedents of justice perceptions discussed in this article include system characteristics, rater behaviors, contextual factors such as organizational culture and interpersonal exchanges before the performance appraisals, and performance ratings. The inclusion of different categories of antecedents emphasizes the importance of considering a broad range of factors influencing justice perceptions in performance appraisals. Finally, social exchange and accountability theories are used as mechanisms by which performance appraisal justice is related to outcomes. Social exchange theory explains the implications of justice perceptions for future exchanges between the individual and the organization, whereas accountability theory explains how a fair performance appraisal can be a mechanism for increasing accountability for maintaining and improving performance. The article attempts to contribute to the literature by identifying future research avenues for performance appraisal justice literature. Performance appraisal is defined as the formal process of evaluating organizational members. Although informal appraisals may occur on a daily basis, the definition in this article is limited to formal appraisals that are periodically conducted. Performance appraisal refers to the whole procedure, including establishment of performance standards, appraisalrelated behaviors of raters within the performance appraisal period, determination of performance rating, and communication of the rating to the ratee. The focus of the article is limited to appraisals conducted by supervisors. In recent years, appraisals by peers and 360° appraisals have become more popular. However, supervisors are still common raters in organizations. Thomas and Bretz (1994) found that immediate supervisors were the most influential raters in Fortune 100 companies. Due to their unique power relationship with subordinates, the mechanisms operating for supervisor performance appraisals may not be generalizable to team member or subordinate appraisals. Consequently, this article focuses on appraisals conducted by supervisors. B. Erdogan / Human Resource Management Review 12 (2002) 555–578 557 2. Types of justice The taxonomy that is most often used to describe organizational justice is distributive and procedural justice (Cropanzano & Folger, 1991). More recently, interactional justice has been introduced as a third type of justice (Bies & Moag, 1986). Even though these dimensions are correlated, they are accepted as distinct dimensions of the justice construct. In the next section, the types of justice are defined, the problems with these constructs are identified, and an attempt is made to clarify and differentiate these constructs in a manner consistent with their respective underlying theories. 2.1. Procedural and interactional justice Procedural justice refers to the fairness of procedures by which performance is evaluated. Even when the outcome of the appraisal is fair, procedures used to arrive at those outcomes may be unfair. The importance of procedural justice is explained by two theories. According to control theory by Thibaut and Walker (1975), individuals have a desire to control what happens to them. Fair procedures are valuable because they allow individuals control over outcomes. Secondly, the group-value model of Lind and Tyler (1988) argues that individuals have a desire to be valuable members of their groups and fair procedures are desirable because they indicate that individuals are valued. Interactional justice has been defined as the fairness of interpersonal treatment received during the execution of a procedure (Bies & Moag, 1986) and emphasizes the importance of truthfulness, respect, and justification as fairness criteria of interpersonal communication. Researchers have recently differentiated procedural from interactional justice by examining the source of the justice perception (supervisor or organization). Moorman (1991) differentiated procedural and interactional justice by assuming that supervisor’s behaviors can affect only interactional justice and not procedural justice. Masterson, Lewis, Goldman, and Taylor (2000) argued that procedural justice is the fairness of the organization and interactional justice is the fairness of the supervisor. This definition equates the source (supervisor versus organization) and type (procedural versus interaction) of justice, which may be problematic for performance appraisals. In performance appraisals, raters enact procedures. For example, the organization can develop fair appraisal criteria, but if raters do not actually apply these criteria, procedures cannot be fair. Even though raters apply the criteria, fairness of appraisal criteria is a procedural element, and should be captured under procedural justice. If raters do not actually execute the procedures developed by the organization, procedural justice will be reduced into a hollow construct, because without rater behaviors performance appraisal procedures cannot exist. In order to resolve this issue, procedural justice may be thought of as a two-dimensional construct. Rater procedural justice refers to perceived fairness of procedures raters use during performance appraisals, whereas system procedural justice refers to perceived fairness of the performance appraisal procedures adopted by the organization. Rater and system procedural justice are likely to be related, but distinct constructs. Erdogan, Kraimer, and Liden (2001) provided the initial construct and discriminant validity of these two dimensions, and showed that these dimensions had different 558 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 predictors. However, they did not show whether these dimensions were different from other forms of justice. Using a two-dimensional conceptualization of procedural justice allows researchers to separate procedural justice of rater behaviors from interactional justice, and is therefore theoretically more accurate. Interactional justice can be defined as the fairness of interpersonal interaction during the performance appraisal session. Performance appraisal requires substantial amounts of communication between raters and ratees. During this interaction, individuals expect to be treated respectfully. The fairness of interpersonal communication during appraisal process constitutes interactional justice perceptions. Interactional justice differs from rater procedural justice in that rater procedural justice is concerned with how fairly the rater implements procedures, whereas interactional justice is concerned with the fairness of the communication process. For example, the extent to which the rater uses the appraisal system as it is intended by the organization is part of rater procedural justice. The extent to which the rater is considerate, kind, and respectful is a component of interactional justice. These definitions are likely to make procedural and interactional justice more consistent with underlying theories. 2.2. Distributive justice Distributive justice has its origins in equity theory (Adams, 1965), which argues that individuals compare their input–output ratios with those of others in order to determine the level of fairness. When individuals perceive inequity, they modify their effort, or change their perceptions of inputs or outcomes. In performance appraisals, individuals compare their efforts with the rating they received and the fairness of the rating constitutes distributive justice perceptions. Even though distributive justice is based on equity theory, it is not clear whether studies of distributive justice capture all forms of inequity. Equity theory allows for unfavorable inequity in which individuals receive less than they deserve and favorable inequity in which they get more than they deserve. Operationalizations of the distributive justice do not distinguish between different types of inequity. In two of the frequently used measures of distributive justice, individuals are asked to indicate their agreement with statements such as ‘‘my pay level is fair’’ and ‘‘I am fairly rewarded considering my responsibilities’’ (Moorman, 1991; Niehoff & Moorman, 1993). It is likely that individuals disagree with these statements only when they experience unfavorable inequity. Supporting this argument, outcome level has been found to be positively related to distributive justice perceptions (Conlon, 1993; Greenberg, 1987; Stoffey & Reilly, 1997; Tyler, 1989, 1994). This indicates a possible failure of distributive justice measures to capture favorable inequity. It is even possible that distributive justice captures outcome satisfaction rather than justice. With existing measures of distributive justice, unobjectively high and inaccurate ratings may be viewed as distributively fair. In order to achieve accuracy and fairness together, distributive justice perceptions need to be measured with items that separate favorable inequity from equity, and fairness from outcome satisfaction. Even though distributive justice is not a strong predictor of performance and broader organizational perceptions (Konovsky & Cropanzano, 1991; Robbins, Summers, & Miller, 2000; Sweeney & McFarlin, 1993), it may be premature to conclude that distributive justice is not important. B. Erdogan / Human Resource Management Review 12 (2002) 555–578 559 Fig. 1. Antecedents and consequences of justice perceptions in performance appraisals. In summary, it is possible to differentiate between four types of justice perceptions in performance appraisals. System procedural justice is the perception that the appraisal procedures adopted by the organization are fair. Rater procedural justice is the perception that the application of procedures by raters is fair. Interactional justice is the degree to which the interpersonal communication during the appraisal process is fair. Distributive justice is the perceived equity of the performance rating. Future studies need to separate distributive justice from outcome satisfaction, and better differentiate procedural justice originating from different sources and interactional justice. Using these definitions, a model that describes antecedents and consequences of justice perceptions is presented in Fig. 1. The model introduces several new antecedents of justice perceptions. Attribution theory, social exchange, and accountability theory are used as mechanisms by which justice perceptions are related to outcomes. 3. Antecedents of justice perceptions 3.1. Antecedents of procedural justice 3.1.1. Due process characteristics Antecedents of procedural justice are factors that enable the ratees to exert control over the process and make them feel as valuable members of the group. Folger, Konovsky, and Cropanzano (1992) summarized fair system characteristics under the label of ‘‘due process 560 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 appraisals.’’ Due process appraisals are based on three basic principles: Adequate notice, fair hearing, and judgment based on evidence. The due process phenomenon is borrowed from the legal literature, where it is used to refer to fair decision-making processes. Therefore, due process characteristics are likely to predict procedural justice perceptions. The only empirical studies of due process appraisals were conducted by Taylor, Masterson, Renard, and Tracy (1998) and Taylor, Tracy, Renard, Harrison, and Carroll (1995) in which a categorical variable was used to represent whether the system was due process or not. In these quasi-experiments, ratees and raters reacted more positively to due process appraisals. However, a due process appraisal is not an objective reality. Ratees in the same organization may have different perceptions of due process characteristics. Therefore, it would be more accurate to measure individual perceptions of due process principles and relate them to individual level outcomes. Each due process principle consists of several different components. To my knowledge, researchers have not yet studied whether these components fall under three due process dimensions. The due process construct is useful in summarizing what is known about appraisal system characteristics, but the predictive validity of each dimension is questionable as the components comprising the dimensions are not homogeneous. As a result, it may be more meaningful to discuss components of due process dimensions, and use these components in predictive models, instead of three due process dimensions. Adequate notice principle states that ratees need to know performance appraisal objectives and standards, need to be allowed input to standard development, and need to be provided with feedback during the appraisal period. Research shows that knowledge of standards and mechanisms (Levy & Williams, 1998), and understanding of the process (Tang & SarshfieldBaldwin, 1996) are related to procedural justice. Field research suggests that participation in the development of standards (Inderrieden, Keaveny, & Allen, 1988) and in goal setting (Renn, 1998) is related to fairness perceptions. Finally, frequent feedback from supervisors during the appraisal period is related to perceived fairness (Inderrieden et al., 1988). Adequate notice components are likely to predict different dimensions of procedural justice. Communicating appraisal criteria to employees is often the role of organizations. Organizations can disseminate information about the system to employees before the appraisal period. Involving employees in system development is also an organizational decision, and will contribute to system procedural justice. However, providing frequent feedback during the appraisal period is the decision of each individual rater, and therefore will predict rater procedural justice. Therefore, I propose that: Proposition 1a: Components of adequate notice will be differentially related to system and rater procedural justice such that, communication of appraisal criteria and involvement in development of appraisal criteria will be positively related to system procedural justice, whereas frequent feedback during appraisal period will be positively related to rater procedural justice. The second due process dimension is fair hearing. This principle argues that raters should be knowledgeable of ratee’s work, and ratees should be allowed to present their explanations B. Erdogan / Human Resource Management Review 12 (2002) 555–578 561 and provide input before the decision is made. Rater familiarity of ratee work is related to fairness perceptions (Evans & McShane, 1988; Fulk, Brief, & Barr, 1985; Landy, Barnes, & Murphy, 1978) and procedural justice (Greenberg, 1986). In addition, the ability to exert voice during the decision-making process is a consistent predictor of procedural justice (e.g., Lind, Kanfer, & Earley, 1990; Skarlicki, Ellard, & Kelln, 1998), even when it does not allow individuals to influence outcomes (Cawley, Keeping, & Levy, 1998; Korsgaard & Roberson, 1995). Components of fair hearing are likely to be differentially related to procedural justice dimensions. Selecting a rater who is familiar with ratee’s work is likely to be the responsibility of the organization, and therefore will predict system procedural justice. However, allowing the ratees voice in the decision is often at the rater’s discretion. Even if the organization attempts to allow voice to the ratee, the extent to which ratee input will be taken seriously or the extent to which the ratee feels free to voice his/her opinions will depend on the rater. Therefore, the following is proposed. Proposition 1b: Components of fair hearing will be differentially related to system and rater procedural justice such that, having a rater familiar with ratee’s work will be positively related to system procedural justice, whereas allowing ratees input in decision-making will be positively related to rater procedural justice. The third due process dimension is judgment based on evidence, which argues that appraisal standards should be applied consistently across all individuals, the decision must be explained after citing evidence, and appeal mechanisms allowing employees to challenge the decision must exist. Consistent application of standards is an important determinant of procedural justice (Greenberg, 1986). Diary keeping, which may be used to provide ratees with examples of performance incidents, results in better differentiation between ratees (DeNisi & Peters, 1996). Finally, the perception that ratings are correctable is related to procedural justice perceptions (Barclay & Harland, 1995). Components of judgment based on evidence are likely to differentially predict dimensions of procedural justice perceptions. Designing appraisal standards is the responsibility of the organization, but applying them consistently across ratees is the responsibility of raters. Explaining the decision to the ratee is also a task that belongs to raters. On the contrary, developing and maintaining effective appeal systems depends on organizational efforts. As a result, the following is proposed. Proposition 1c: Components of judgment based on evidence will be differentially related to system and rater procedural justice such that, existence of effective appeal mechanisms will be positively related to system procedural justice, whereas consistent application of standards and explaining the decision to the ratee will be positively related to rater procedural justice. Predictions in Propositions 1a, 1b, and 1c may need to be modified depending on the characteristics of the performance appraisal system in the organization under study. For example, in some organizations, communicating the appraisal criteria to individuals may be 562 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 the role of raters rather than the organization. When studying such organizations, Proposition 1a would predict that knowledge of appraisal criteria would be positively related to rater procedural justice. Similarly, the predictions of Proposition 1b needs to be changed in organizations in which the selection of raters are not centralized and leaders have a role in determining who should rate each individual. In this case, having a rater knowledgeable about ratee performance would be positively related to rater procedural justice. 3.1.2. Contextual factors In addition to the characteristics of the performance appraisal system, factors outside the appraisal process may influence justice perceptions of performance appraisals. Contextual factors include perceived organizational support (POS), organizational culture, and leader– member exchange quality (LMX). 3.1.2.1. Pre-appraisal POS. Eisenberger, Huntington, Hutchison, and Sowa (1986) defined POS as the global perception of employees that the organization values their well-being and cares about them. POS construct is based on the idea that individuals perceive the organization as an entity with whom they interact. Employee perceptions of POS are shaped by developmental experiences provided to the individual, promotions (Wayne, Shore, & Liden, 1997), job condition favorableness (Eisenberger, Cummings, Armeli, & Lynch, 1997), and other factors signaling to the individual that the organization cares about their well-being. Once individuals enter into the organization, they gradually develop POS based on several organizational practices and their own experiences with the organization. Previous studies established a relationship between POS and procedural justice in cross-sectional studies, and argued that POS would be a consequence of procedural justice (Masterson et al., 2000; Moorman, Blakely, & Niehoff, 1998). However, it is likely that pre-appraisal POS is an antecedent of perceived procedural justice because, by the time the individuals reach their first performance appraisal, perceptions of organizational support will already have been formed. Pre-appraisal POS is likely to have an impact on system procedural justice perceptions. When individuals have a positive exchange with the organization, they may expect the organizational procedures to be fair. This expectation may influence later justice perceptions, by making individuals attend to cues that suggest fairness, and ignore the cues that suggest unfairness. Therefore, the following is proposed: Proposition 2a: POS before the appraisal will be positively related to system procedural justice perceptions during the appraisal. 3.1.2.2. Organizational culture. Organizational culture has been defined as ‘‘a pattern of basic assumptions, invented or discovered by a group as it learns to cope with problems of adaptation and integration, that has worked well enough to be considered valid, and therefore are taught to new members as correct ways of perceiving, thinking, and feeling in relation to those problems’’ (Schein, 1990). In other words, organizational culture refers to socially constructed and historically determined cognitions that include basic assumptions of B. Erdogan / Human Resource Management Review 12 (2002) 555–578 563 members, and shape their thinking and behavioral patterns. Some researchers argue that performance appraisal systems must fit the organizational culture (Evans & McShane, 1988; Maroney & Buckely, 1992). Because organizational culture constitutes the context in which performance appraisal occurs, it is likely to influence justice perceptions. Organizational culture has been conceptualized in different ways, such as a collection of values (O’Reilly, Chatman, & Caldwell, 1991), behavioral norms (Cooke & Rousseau, 1988), or myths and stories (Trice & Beyer, 1984). Stories and myths may be too general to influence perceptions of justice in the context of performance appraisals. Consequently, studying culture as values or behavioral norms may be more meaningful in relating organizational culture to justice perceptions. Organizational culture may be conceptualized in different ways, and the approach adopted in this article is relatively narrow in focus. Because behavioral norms are likely to be a visible reflection of the underlying organizational culture and allow studying culture quantitatively, conceptualizing organizational culture as behavioral norms is a practical approach that may facilitate research in this area. Even though the propositions in this section conceptualized culture as behavioral styles, similar propositions may be developed by conceptualizing culture as values. According to Cooke and Rousseau (1988), organizational culture consists of three behavioral styles: constructive, passive– defensive, and aggressive– defensive styles. In constructive cultures, there is an emphasis on achievement and individual development (Cooke & Szumal, 1993), and therefore appraisal systems may be perceived as less threatening. Performance appraisals may have a greater likelihood of being used as performance enhancing tools in these cultures. In passive–defensive cultures in which individuals try to minimize interpersonal conflict and protect status quo, raters may avoid conducting realistic appraisals in order to avoid confrontation. Finally, in aggressive–defensive cultures where gaining and maintaining power is an important concern, performance appraisal may serve as a political tool. Because appraisal fairness and objectivity may not be major concerns, procedural justice perceptions in passive–defensive and aggressive–defensive cultures may be lower. Organizational culture is likely to affect rater procedural justice rather than system procedural justice. When constructive behavioral styles are prevalent, raters are more likely to follow fair procedures during appraisals. When culture is more political or conformist appraisals serve purposes other than performance management and raters may be less likely to pay attention to objectivity or fairness of appraisals. Therefore, Proposition 2b: Organizational culture will be related to perceptions of rater procedural justice such that, in constructive cultures compared to passive–defensive or aggressive–defensive cultures, the highest levels of rater procedural justice will be observed. 3.1.2.3. Pre-appraisal LMX quality. Performance appraisal occurs within the context of an ongoing relationship between the leader and member. Therefore, LMX quality may influence perceived procedural justice. LMX literature showed that supervisors establish differential relationships with their subordinates (see Liden, Sparrowe, & Wayne, 1997 for a review). A high-quality relationship goes beyond which is specified in the employment 564 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 contract, and is characterized by trust, respect, affect, and openness. LMX quality is related to attitudes and perceptions such as job satisfaction, turnover intentions, and organizational commitment (Gerstner & Day, 1997; Liden et al., 1997). LMX has also been related to behaviors such as objective performance (Graen, Novak, & Sommerkamp, 1982) and organizational citizenship behavior (Settoon, Bennett, & Liden, 1996; Wayne & Green, 1993; Wayne et al., 1997). The growing LMX literature suggests that LMX quality is related to whether subordinates are satisfied and productive members of the organization. The overall quality of the exchange is likely to have implications for rater and ratee behaviors, and resulting perceptions of fairness. In performance appraisal context, LMX quality is likely to be related to rater procedural justice perceptions. High LMX employees tend to perceive higher decision influence in general (Scandura, Graen, & Novak, 1986), and therefore may feel that they have more control over appraisal procedures because of their exchange quality. In these exchanges, the rater may exert more effort to make the procedures more fair, in order to protect one’s relationship with the ratee. As a result, the following is proposed: Proposition 2c: LMX quality before the performance appraisal will be positively related to perceptions of rater procedural justice. This proposition needs to be tested longitudinally. Masterson et al. (2000) hypothesized LMX to be an outcome of justice in a cross-sectional study. They found that all effects of procedural justice on outcome variables were mediated through POS, rather than LMX. The insignificant relationship between procedural justice and LMX may be because of the procedural justice measure used in this study, which captured fairness of the organization. When system and rater procedural justice are differentiated, LMX is likely to be related to rater procedural justice. In addition, I argued that pre-appraisal LMX will be an antecedent of rater procedural justice. LMX tends to develop very early in the leader–member relationship and shape individual experiences following relationship development (Liden, Wayne, & Stilwell, 1993). Therefore, pre-appraisal LMX may be thought of as an antecedent of procedural justice perceptions. Thus, future research examining the link between LMX and procedural justice using longitudinal study designs is warranted. In summary, procedural justice perceptions are likely to be a function of due process components and contextual factors. Due process characteristics will be positively related to system and rater procedural justice. Pre-appraisal POS will be positively related to system procedural justice, whereas organizational culture and pre-appraisal LMX will be positively related to rater procedural justice. 3.2. Antecedents of interactional justice In performance appraisals, interactional justice refers to the fairness of appraisal-related communication and interaction. Even when procedures are fair, individuals may find the communication and interpersonal treatment involved in the process unfair. Interactional justice is determined by supervisor behaviors in the performance appraisal setting. Being a B. Erdogan / Human Resource Management Review 12 (2002) 555–578 565 good listener and letting subordinates talk are important determinants of fairness (Burke, Weitzel, & Weir, 1980). Session behaviors of raters that are courteous, respectful, and providing two-way communication are likely to lead to interactional justice perceptions. In addition to session behaviors communicating openness, respect, and consideration, raters may use other behaviors that may influence interactional justice perceptions. More specifically, just as ratees manage their impressions, raters may use impression management tactics, which may influence justice perceptions. 3.2.1. Impression management Impression management refers to behaviors that alter or maintain a person’s image in the eyes of others (Villanova & Bernardin, 1989). Greenberg (1990) argued that fairness is at least partly a result of impression management behaviors. Studies showed that employee impression management tactics were successful in influencing performance ratings (Gundersen, Tinsley, & Terpstra, 1996; Wayne & Ferris, 1990; Wayne & Kacmar, 1991; Wayne & Liden, 1995). However, the effects of leader impression management behaviors received little attention. In one exception, Kacmar, Wayne, and Wright (1996) conducted an experiment and manipulated the use of impression management of the supervisor during performance appraisal, using a confederate. They found that subjects playing ratees evaluated the impression-managing supervisor more positively. These findings need to be replicated with respect to justice perceptions, and using a more complete conceptualization of impression management. Wayne and Ferris (1990) classified employee impression management tactics as supervisor focused, subordinate focused, and job focused. Their definitions can be adapted to describe supervisor impression management behaviors. Supervisor-focused tactics are behaviors attempting to convey the impression that the supervisor is a likeable person. Exerting a conscious effort to be polite and friendly during the appraisal interview may be considered as supervisor-focused behaviors because the rater regulates one’s own behaviors and appearance in order to achieve the image of a likeable person. Subordinate-focused tactics can be defined as doing favors or praising the subordinate. Praising the ratee during the performance appraisal interview, inquiring about the personal reasons behind poor performance episodes, offering to do a personal favor for the ratee, agreeing with the ratee during the discussion may be thought of as subordinate focused tactics because these behaviors involve a more active approach to gain the favor of the ratee. Job-focused tactics may be defined as behaviors trying to convey the image that the rater is doing a good job as a supervisor and rater. For example, during performance appraisal interview, the rater may take responsibility of an event positively affecting the subordinate or try to reduce responsibility in an event negatively affecting the employee, and thereby convey the impression that the rater is effective in his or her job. Impression management behaviors are likely to influence interactional justice perceptions. Job focused tactics may have a negative effect on interactional justice perceptions because they involve the supervisor promoting one’s self-appearance directly. By trying to be perceived as a good supervisor, the rater may unintentionally be perceived as less interactionally fair because these behaviors are unlikely to promote two-way communication. Supervisor and subordinate 566 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 focused tactics may indicate kindness and understanding, and therefore may be positively related to interactional justice perceptions. As a result, the following is proposed: Proposition 3: The rater’s use of job focused impression management tactics will be negatively related to interactional justice perceptions, whereas the use of supervisor and subordinate focused tactics will be positively related. 3.2.2. Pre-appraisal LMX quality I argued previously that LMX quality would influence procedural justice perceptions. LMX quality may also influence interactional justice perceptions. Fairhurst and Chandler (1989) showed that in high-quality relationships, leaders behave in a less authoritarian manner and use their formal authority less frequently. As a result, if overall exchange quality is high, ratees may feel that they are being treated fairly and respectfully during the performance appraisal because the rater may behave more sensitively. Recent research argued that LMX is a consequence of interactional justice, and found support for this model in a cross-sectional study (Masterson et al., 2000). However, because LMX tends to form very early in the interaction of leader and member (Liden et al., 1993), appraisal behaviors of raters leading to interactional justice may be a function of the overall exchange quality before the appraisal. Therefore, the following is proposed. Proposition 4: Pre-appraisal LMX quality will be positively related to perceptions of interactional justice during performance appraisal. In summary, rater behaviors and contextual factors are expected to influence interactional justice perceptions. More specifically, rater impression management tactics and LMX quality are argued to be predictors of interactional justice perceptions. Longitudinal designs are more appropriate in testing these relationships because interactional justice perceptions are expected to develop as a result of the existing organizational and interpersonal context before the appraisal. 3.3. Antecedents of distributive justice In the performance appraisal context, distributive justice refers to perceived fairness of performance ratings. Distributive justice is based on equity theory (Adams, 1965), which argues that individuals compare their inputs and outcomes with those of referent individuals, and perceive fairness if these comparisons are proportional. Antecedents of distributive justice include the actual performance rating as well as comparisons made by individuals. 3.3.1. Performance rating Researchers who related actual ratings to perceptions of distributive justice found that higher ratings were related to higher perceived fairness (Evans & McShane, 1988; Inderrieden et al., 1988) and distributive justice (Stoffey & Reilly, 1997). However, Greenberg (1986) conducted a study in which he asked subjects the characteristics of fair B. Erdogan / Human Resource Management Review 12 (2002) 555–578 567 performance appraisals, and subjects mentioned that ratings needed to reflect actual performance. With this reasoning, high ratings should not necessarily be related to distributive justice. Subjects in the Greenberg’s study may have demonstrated social desirability bias. It is also possible that observed positive relationships between ratings and distributive justice reflect the limitations of distributive justice measurement. For example, the studies of Inderrieden et al. (1988) and Stoffey and Reilly (1997) did not differentiate favorable and unfavorable equity and just asked whether the performance evaluation was fair. It is possible that this measure captured satisfaction with evaluation, rather than equity. A positive relationship is expected between ratings and distributive justice because individuals are likely to overestimate their contributions to the group (Ashford, 1989; Harris & Schaubroeck, 1988), and therefore may perceive high ratings as more fair. However, the relationship between favorableness of ratings and distributive justice perceptions needs to be demonstrated using a distributive justice measure that captures favorable inequity as well as unfavorable inequity. 3.3.2. Pre-appraisal LMX quality LMX quality is likely to increase the expectations of employees from their supervisors. For example, Mansour-Cole and Scott (1998) showed that when employees hear the news of layoffs through sources other than their supervisor, high LMX employees perceive lower levels of procedural justice, suggesting that high LMX members expect to hear the news from their leaders. High LMX employees also perceive high levels of decision influence in general (Scandura et al., 1986). Even when employees are aware that their performance is sub par, they may expect the leader to make these suggestions informally, rather than bringing it up in the formal appraisal, which will influence promotion and pay decisions. Therefore, high LMX employees may be more sensitive to their rating level and low ratings may disturb them more than low LMX employees. As a result, the following is proposed. Proposition 5: The relationship between ratings and distributive justice perceptions will be moderated by LMX quality such that, for high LMX employees, there will be a stronger positive relationship between ratings and distributive justice perceptions. 3.3.3. Perceived basis of LMX Rating systems used in many organizations prevent individuals from making direct comparisons with their coworkers, because they usually do not know the ratings of their peers. However, employees may still engage in comparisons. Individuals may compare their ratings with internal standards or with past ratings to determine equity perceptions. More interestingly, they may make external comparisons by guessing the performance rating that will be given to other employees in their groups. Individuals may use the interpersonal exchanges the leader forms with other group members to guess the performance rating these members will receive. Individuals accurately observe with whom the leader has a high- and low-quality relationship (Duchon, Green, & Taber, 1986; Graen & Cashman, 1975). In addition, LMX quality predicts outcomes such as career progress (Wakabayashi, Graen, Graen, & Graen, 1988; Wakabayashi & Graen, 1984) and salary progress (Wayne, Liden, 568 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 Kraimer, & Graf, 1999). It is likely that individuals will observe the actual linkage between LMX and positive outcomes, and assume that there is a linkage even when there is not. Individuals may believe that the closeness between the leader and member will result in higher ratings for high LMX members, because leaders may want to protect their high-quality exchanges with these members even when they realize that high LMX members are not performing at high levels. In addition, individuals may believe that leaders will perceive high LMX members as higher performers, and therefore may not pay attention to information suggesting poor or mediocre performance, or may not weight this information heavily. Therefore, I propose that the following. Proposition 6a: When ratees do not know the performance ratings of their coworkers, they will believe that those with higher LMXs are more likely to receive higher performance ratings. If individuals use coworker LMXs to predict their performance ratings, distributive justice perceptions will be affected by the perceived degree to which the leader forms high-quality LMXs based on work-related factors. Studies showed that leaders do not always form highquality relationships with high performers (Rosse & Kraut, 1983). In fact, factors such as impression management behaviors of employees (Wayne & Ferris, 1990) and perceived similarity (Liden et al., 1993) are related to LMX quality. If ratees believe that the leader forms high LMXs based on factors unrelated to work performance, they are more likely to perceive lower levels of distributive justice in performance appraisals. In this case, the ratee will assume that low performers are going to be rewarded with high ratings because of their high LMX levels, which will create inequity in one’s own ratings. Therefore, the following is proposed. Proposition 6b: The perception that the leader forms LMXs based on work-related factors will be positively related to distributive justice perceptions. 3.3.4. Rater information use Williams, DeNisi, Blencoe, and Cafferty (1985) adopted Kelley’s (1967, 1971) attribution theory to identify the types of information raters use during performance appraisals. They argued that raters could look for three types of information—consistency, distinctiveness, and consensus. Consistency refers to how the ratee behaves in the same setting across different occasions. Distinctiveness refers to how the ratee behaves in different settings. Consensus refers to how other people behave in the same setting. Williams et al. (1985) showed that raters seek different types of information in differing amounts, and according to DeNisi, Cafferty, and Meglino (1984), how a rater searches for information determines which behaviors are observed. Rater information seeking is an important cognitive variable that has not been integrated with performance appraisal reactions. Raters using consistency information look at how a certain ratee behaves in the same setting across different occasions. The perception that the rater is using consistency information may be positively related to distributive justice, because the use of B. Erdogan / Human Resource Management Review 12 (2002) 555–578 569 consistency information suggests that the rater is looking at performance across a period of time, rather than focusing on isolated incidents. Likewise, the use of distinctiveness information may be positively related to distributive justice because it allows behaviors across different settings to be included in the appraisal. Consensus information may be negatively related to distributive justice, because consensus information involves comparisons across different individuals in the same setting. A person performing at satisfactory levels may be given a less than average rating, because of higher performing individuals in the same group. The rating may become a reflection of an individual’s ranking in the group and therefore distributive justice perceptions may be lower. Therefore, the following is proposed: Proposition 6c: The perceived type of information used in appraisals will be related to distributive justice perceptions, such that the use of consistency and distinctiveness of information will be positively related to distributive justice perceptions, whereas the use of consensus information will be negatively related. In summary, distributive justice perceptions will be a function of characteristics of performance ratings. More specifically, it is argued that performance ratings and distributive justice will be strongly and positively related for high LMX members. The perception that the leader is forming LMXs based on work-related factors will influence distributive justice perceptions. Finally, the perceived types of information raters are using when rating the individual will influence distributive justice perceptions. 4. Outcomes of justice perceptions in performance appraisals Outcomes of justice perceptions in performance appraisals are summarized in three groups: organization-related outcomes, leader-related outcomes, and performance-related outcomes. This categorization is not intended to be exhaustive, but they represent some of the important behaviors and attitudes related to justice perceptions in performance appraisals. 4.1. Organizational outcomes Organizational outcomes are behaviors individuals use to express their reactions to the organization. Organizational commitment and turnover intentions indicate the attachment of the individual to the organization. Prosocial behaviors may be defined as social acts carried out to produce and maintain the well-being and integrity of others (Brief & Motowidlo, 1986). Prosocial behaviors targeting the organization may be considered as an organizational outcome. Social exchange theory predicts that, when a person or entity does a favor to another, the recipient of the favor will be obliged to reciprocate (Gouldner, 1960). Based on this reasoning, individuals will reciprocate to sources of fairness through behaviors that benefit the source. Organizational outcomes may be mechanisms through which individuals use to reciprocate to organizational fairness. Procedural justice has been accepted as the 570 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 strongest predictor of organizational outcomes. For example, Folger and Konovsky (1989) showed that procedural justice explained a larger variance in organizational commitment compared to distributive justice. Procedural justice is negatively related to turnover intentions (Dailey & Kirk, 1992; Konovsky & Cropanzano, 1991; Taylor et al., 1995). Finally, organizational citizenship behavior, the degree to which the individual behaves in ways that are not specified in role descriptions but are beneficial to the organization (Organ, 1988) is related to procedural justice (Farh, Earley, & Lin, 1997; Moorman, 1991). When procedural justice is conceptualized as two dimensional, organizational outcomes are likely to follow system procedural justice. POS has been argued to be a mediator of procedural justice–outcome relationship. After Fasolo (1995) suggested that POS would be a consequence of procedural justice, Masterson et al. (2000) showed that POS mediated the relationship between procedural justice and outcomes such as organizational commitment, citizenship behaviors targeting the organization, turnover intentions, and job satisfaction. These studies assume that procedural justice will be attributed to the efforts of the organization. However, when procedural justice of organization and rater are distinguished, system procedural justice is likely to be related to post-appraisal POS, and thereby organizational outcomes. Therefore, the following are proposed. Proposition 7a: System procedural justice will be positively related to post-appraisal POS. Proposition 7b: Post-appraisal POS will mediate the relationship between system procedural justice and organizational outcomes. 4.2. Leader-related outcomes Leader-related outcomes are attitudes reflecting satisfaction with the leader, and behaviors benefiting the leader such as prosocial behaviors targeting the leader. Based on social exchange theory, individuals may reciprocate fairness of raters by behaving in ways that will benefit the raters. When the rater is one’s supervisor, leader-related outcomes are some of the ways in which individuals can reciprocate. Raters enforce performance appraisal regulations, and therefore they contribute to rater procedural justice perceptions. By communicating with the ratees during the appraisal session in a constructive manner, they shape perceptions of interactional justice. Finally, they decide the final performance rating, and therefore they are responsible for perceptions of distributive justice. As a result, these three forms of justice perceptions may create a desire to reciprocate through leader-related outcomes. Supporting this argument, Tang and Sarshfield-Baldwin (1996) found that perceptions of procedural justice are related to satisfaction with the supervisor. Similarly, Folger and Konovsky (1989) used items that include procedural and interactional justice and found a relationship with trust in the supervisor. Taylor et al. (1995) showed that due process appraisal was related to satisfaction with manager. Finally, Masterson et al. (2000) related procedural and interactional justice perceptions to leader-related B. Erdogan / Human Resource Management Review 12 (2002) 555–578 571 outcomes such as citizenship behaviors benefiting the leader, and found that only interactional justice was related to these outcomes, through LMX. This study did not explore distributive justice as an antecedent of LMX, and used a procedural justice measure focusing on system justice. It is likely that all three forms of justice perceptions will be related to leader-related outcomes through their effects on LMX quality following the appraisal, provided that these justice perceptions are attributed to leader behaviors. From past research, the relationship between interactional justice and LMX is established. Similar relationships would be expected for rater procedural justice and distributive justice perceptions. Proposition 8a: Rater procedural justice and distributive justice perceptions would be positively related to post-appraisal LMX. Proposition 8b: The relationship between rater procedural justice, interactional justice, distributive justice, and leader-related outcomes will be mediated by post-appraisal LMX. 4.3. Performance-related outcomes A meta-analysis of LMX literature shows that there is a significant relationship between LMX quality and individual performance (Gerstner & Day, 1997). Consequently, procedural, interactional, and distributive justice perceptions may be indirectly related to individual performance, through their effects on LMX quality. In the performance appraisal context, justice perceptions attributed to the leader may improve the exchange between the leader and member and may motivate the individual to exert more effort. In addition, leaders may provide more financial and informational resources to employees closer to them, and therefore leaders may help improve focal employee performance. Another mechanism by which justice perceptions may influence performance is perceived accountability. Accountability can be defined as the implicit and explicit belief that one is expected to justify one’s actions to others (Scott & Lyman, 1968). Performance appraisal has been viewed as a way of ensuring accountability for performance (Frink & Ferris, 1998). In experimental studies, accountability is sometimes operationalized by manipulating the existence of performance appraisals (e.g., Staw & Boettger, 1990). In other words, when individuals are given performance reviews, it is assumed that they will perceive the need to justify their performance-related actions, and therefore feel more accountable toward the organization and leader. Perceived accountability resulting from performance appraisals is expected to motivate individuals to maintain their high performance or increase their performance. However, the existence of a formal performance appraisal system may not ensure that individuals perceive accountability. Therefore, it is necessary to identify characteristics of performance appraisals that lead to higher levels of accountability. Perceived accountability after a performance appraisal is likely to be higher when individuals perceive distributive justice during the performance appraisal. When ratings are equitable and reflect their actual performance, individuals are likely to perceive higher levels 572 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 of accountability for performance in the future. Favorable inequity in ratings may create the feeling that the supervisor does not really perceive one’s weaknesses, or poor performance is not being punished, and therefore may reduce the perception of accountability toward the leader. Unfavorable inequity will likewise weaken the perceived relationship between performance and outcomes, and therefore it will reduce accountability perceptions for maintaining and increasing high performance. As a result, it is argued that: Proposition 9a: Distributive justice perceptions in performance appraisals will be positively related to perceived accountability. Perceived accountability is likely to be one of the mechanisms by which distributive justice perceptions are related to performance-related outcomes. Studies show that accountability is related to more stringent decision-making (Ordonez, Benson, & Beach, 1999), increased productivity (Tausky & Chelte, 1983), and increased normative behaviors in individuals (Gelfand & Realo, 1999). It is likely that when distributive justice perceptions are high, individuals perceive higher levels of accountability, and thereby engage in behaviors that improve performance. As a result, the following is proposed: Proposition 9b: The relationship between distributive justice perceptions and performance related outcomes will be mediated by perceived accountability. In summary, justice perceptions will be related to organization-related, leader-related, and performance-related outcomes, through improved exchanges with the organization and the leader, and through increased accountability pressures. Therefore, performance appraisal is argued to be a critical incident, which shapes future interactions between the leader and member as well as the leader and organization, and influences future attitudes and behaviors. Differential relationships are expected between different types of justice and outcomes, but all types of justice perceptions are argued to be important for organizational effectiveness. 5. Discussion Performance appraisal is a performance management mechanism that has broad implications for attitudes and behaviors in organizations. The model presented identifies areas in which there is a need for more research. First of all, there is a need to differentiate between different forms of justice perceptions and to distinguish justice perceptions from related concepts. At the moment, procedural justice is often equated with justice attributable to the organization, which is not in line with the definition of procedural justice. In order to separate source and type of justice, rater and system procedural justice dimensions were proposed. With the addition of a two-dimensional view of procedural justice, the overlap between procedural justice and interactional justice may be reduced, and higher conceptual clarity will be achieved. In addition, the need to separate distributive justice from outcome satisfaction, and to incorporate favorable inequity into measures of distributive justice were pointed out. B. Erdogan / Human Resource Management Review 12 (2002) 555–578 573 Secondly, antecedents of justice perceptions were discussed. Regarding procedural justice, it is argued that due process characteristics is not likely to constitute a three-dimensional construct because the dimensions of due process principles are not homogeneous. Instead, it may be more meaningful to examine the relationships between the components of each due process dimension and procedural justice perceptions. Due process characteristics are likely to be summary labels for components of performance appraisals. Their predictive validity needs to be established before using dimensions of due process in performance appraisal research. Pre-appraisal POS and LMX are argued to be contextual antecedents of justice perceptions. This proposition is novel because it does not assume that justice will follow POS and LMX as most empirical studies have. In reality, we do not know whether justice follows, or causes LMX. One way to resolve this issue is to adopt a longitudinal model. It is argued that perceptions during the appraisal are shaped by exchanges before the appraisal, whereas perceptions during the appraisal will be important enough to be incorporated into the exchanges following appraisals. Rater impression management behavior is another addition to the literature. Impression management behaviors used by subordinates were adapted for use by raters. Raters are likely to engage in impression management during appraisals and these behaviors may have implications for interactional justice. It is argued that instead of looking at the effects of the frequency of overall impression management behaviors, it is necessary to consider the type of impression management raters use. Impression management may positively or negatively influence interactional justice perceptions, depending on its effects on the twoway communication during the appraisal interview. Several new constructs were introduced as antecedents of distributive justice. First of all, LMX quality is argued to be a factor influencing the importance of rating level in determining justice. It is argued that in high-quality exchanges, members expect higher ratings, and will be more dissatisfied with lower ratings. This relationship may explain why raters may be reluctant to give low or moderate ratings to subordinates close to them. Secondly, it is argued that even when employees do not know their coworkers’ ratings, they will attempt to guess these ratings to determine the fairness of their own rating. If supported, this proposition would suggest the need for leaders to justify their exchanges with group members to other members and the need for leaders to form higher quality exchanges based on work-related factors. Thirdly, findings of past research regarding the outcomes of justice perceptions are summarized and future research directions are identified. In addition to emphasizing the importance of social exchange theory in explaining justice–outcome relationships, perceived accountability is introduced as a mechanism linking justice to outcomes. Performance appraisal is generally accepted as a human resource application increasing individual accountability, but it is not clear which aspects of appraisal systems lead to accountability. Distributive justice is introduced as a mechanism by which performance appraisals increase perceived accountability in organizations and thereby influence individual performance. The model developed in this article has practical applications as well. The model does not emphasize one form of justice perception at the expense of others and instead argues that different forms of justice perceptions are related to different attitudes and behaviors. 574 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 Therefore, practitioners need to exert effort to increase all types of justice perceptions. The model also indicates the need to consider the bigger picture when trying to increase justice perceptions. For example, high performance ratings may increase distributive justice perceptions, and may lead to higher performance in the future because of a need to reciprocate to the leader. However, this does not mean that leaders should give high ratings to everyone. When the leader gives undeservedly high ratings to an individual, accountability perceptions may suffer, leading to reduced performance. In other words, the model may help practitioners predict the possible consequences of their behaviors during the performance appraisal and suggest alternative ways in which they can improve actual and perceived fairness of performance appraisals. In summary, justice perceptions are important reactions in the performance appraisal context. Several studies focused on what makes appraisals fair and the consequences of appraisal fairness. Still, there is a need to pay attention to the operationalizations of justice perceptions used by researchers and to modify these measures in line with their respective theories. Different research traditions have implications for performance appraisal justice literature and, after 70 years of research, performance appraisal literature continues to be important and interesting. Acknowledgements The author wishes to thank Maria Kraimer, Sandy Wayne, Bob Liden, Renata Jaworski, and Monica Gavino for their comments on the manuscript. This article is based on a paper written to satisfy the qualifying paper requirement of the doctoral program at University of Illinois at Chicago. An earlier version of the article was presented at the 43rd Midwest Academy of Management meeting, March 2000 in Chicago, IL. 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. Ashford, S. (1989). Self-assessment in organizations: a literature review and integrative model. Research in Organizational Behavior, 11, 133 – 174. Barclay, J. H., & Harland, L. K. (1995). Peer performance appraisals: the impact of rater competence, rater location, and rating correctability on fairness perceptions. Group and Organization Management, 20, 39 – 60. Bies, R. J., & Moag, J. S. (1986). Interactional justice: communication criteria of fairness. Research on Negotiation in Organizations, 1, 43 – 55. Brief, A. P., & Motowidlo, S. J. (1986). Prosocial organizational behaviors. Academy of Management Review, 11, 710 – 725. Burke, R. J., Weitzel, W. F., & Weir, T. (1980). Characteristics of effective interviews of employees’ performance review and development: one more time. Psychological Reports, 47, 683 – 695. Cawley, B. D., Keeping, L. M., & Levy, L. M. (1998). Participation in the performance appraisal process and employee reactions: a meta analytic review of field investigations. Journal of Applied Psychology, 83, 615 – 633. Conlon, D. E. (1993). Some tests of the self-interest and group-value models of procedural justice: evidence from an organizational appeal procedure. Academy of Management Journal, 36, 1109 – 1124. B. Erdogan / Human Resource Management Review 12 (2002) 555–578 575 Cooke, R. A., & Rousseau, D. M. (1988). Behavioral norms and expectations: a quantitative approach to the assessment of organizational culture. Group and Organization Studies, 13, 245 – 273. Cooke, R. A., & Szumal, J. L. (1993). Measuring normative beliefs and shared behavioral expectations in organizations: the reliability and validity of the organizational culture inventory. Psychological Reports, 72, 1299 – 1330. Cropanzano, R., & Folger, R. (1991). Procedural justice and worker motivation. In R. M. Steers, L. W. Porter (Eds.), Motivation and work behavior (5th ed.) (pp. 131 – 143). New York: McGraw-Hill. Dailey, R. C., & Kirk, D. J. (1992). Distributive and procedural justice as antecedents of job dissatisfaction and intent to turnover. Human Relations, 45, 305 – 317. DeNisi, A. S., Cafferty, T. P., & Meglino, B. M. (1984). A cognitive view of the performance appraisal process: a model and research propositions. Organizational Behavior and Human Performance, 33, 360 – 396. DeNisi, A. S., & Peters, L. H. (1996). Organization of information in memory and the performance appraisal process: evidence from the field. Journal of Applied Psychology, 81, 717 – 737. Duchon, D., Green, S. G., & Taber, T. D. (1986). Vertical dyad linkage: a longitudinal assessment of antecedents, measures, and consequences. Journal of Applied Psychology, 71, 56 – 60. Eisenberger, R., Cummings, J., Armeli, S., & Lynch, P. (1997). Perceived organizational support, discretionary treatment, and job satisfaction. Journal of Applied Psychology, 82, 812 – 820. Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Perceived organizational support. Journal of Applied Psychology, 71, 500 – 507. Erdogan, B., Kraimer, M. L., & Liden, R. C. (2001). Procedural justice as a two-dimensional construct: an examination in the performance appraisal context. Journal of Applied Behavioral Science, 37, 205 – 222. Evans, E. M., & McShane, S. L. (1988). Employee perceptions of performance appraisal fairness in two organizations. Canadian Journal of Behavioral Sciences, 20, 177 – 191. Fairhurst, G. T., & Chandler, T. A. (1989). Social structure in leader – member interaction. Communication Monographs, 56, 215 – 239. Farh, J. L., Earley, P. C., & Lin, S. C. (1997). Impetus for action: a cultural analysis of justice and organizational citizenship behavior in Chinese society. Administrative Science Quarterly, 42, 421 – 444. Fasolo, P. M. (1995). Procedural justice and perceived organizational support: hypothesized effects on job performance. In R. S. Cropanzano, & K. M. Kacmar (Eds.), Organizational politics, justice and support: managing the social climate of the workplace (pp. 185 – 195). Westport, CT: Quorum Books. Folger, R., & Konovsky, M. A. (1989). Effects of procedural and distributive justice on reactions to pay raise decisions. Academy of Management Journal, 32, 115 – 130. Folger, R., Konovsky, M. A., & Cropanzano, R. (1992). A due process metaphor for performance appraisal. Research in Organizational Behavior, 14, 129 – 177. Frink, D. D., & Ferris, G. R. (1998). Accountability, impression management, and goal setting in the performance evaluation process. Human Relations, 51, 1259 – 1283. Fulk, J., Brief, A. P., & Barr, S. H. (1985). Trust-in supervisor and perceived fairness and accuracy of performance evaluations. Journal of Business Research, 13, 301 – 313. Gelfand, M. J., & Realo, A. (1999). Individualism – collectivism and accountability in intergroup negotiations. Journal of Applied Psychology, 84, 721 – 736. Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader – member exchange theory: correlates and construct issues. Journal of Applied Psychology, 82, 827 – 844. Gouldner, A. W. (1960). The norm of reciprocity: a preliminary statement. American Sociological Review, 25, 161 – 177. Graen, G., & Cashman, J. F. (1975). A role-making model of leadership in formal organizations: a developmental approach. In J. G. Hunt, & L. L. Larson (Eds.), Leadership frontiers (pp. 143 – 165). Kent, OH: Kent State University Press. Graen, G., Novak, M. A., & Sommerkamp, P. (1982). The effects of leader member exchange and job design on productivity and satisfaction: testing a dual attachment model. Organizational Behavior and Human Performance, 30, 109 – 131. 576 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 Greenberg, J. (1986). Determinants of perceived fairness of performance evaluations. Journal of Applied Psychology, 71, 340 – 342. Greenberg, J. (1987). Reactions to procedural injustice in payment distributions: do the means justify the ends? Journal of Applied Psychology, 72, 55 – 61. Greenberg, J. (1990). Looking fair vs. being fair: managing impressions of organizational justice. Research in Organizational Behavior, 12, 111 – 157. Gundersen, D. E., Tinsley, D. B., & Terpstra, D. E. (1996). Empirical assessment of impression management biases: the potential for performance appraisal error. Journal of Social Behavior and Personality, 11, 57 – 76. Harris, M., & Schaubroeck, J. (1988). A meta-analysis of self-supervisor self-peer and peer – supervisor ratings. Personnel Psychology, 41, 43 – 61. Inderrieden, E. J., Keaveny, T. J., & Allen, R. E. (1988). Predictors of employee satisfaction with the performance appraisal process. Journal of Business and Psychology, 2, 306 – 310. Jacobs, R., Kafry, D., & Zedeck, S. (1980). Expectations of behaviorally anchored rating scales. Personnel Psychology, 33, 595 – 640. Kacmar, K. M., Wayne, S. J., & Wright, P. M. (1996). Subordinate reactions to the use of impression management tactics and feedback by the supervisor. Journal of Managerial Issues, 8, 35 – 53. Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska symposium on motivation. Lincoln: University of Nebraska Press. Kelley, H. H. (1971). Attribution of social interaction. Morristown, NJ: General Learning Press. 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. Korsgaard, M. A., & Roberson, L. (1995). Procedural justice in performance evaluation: the role of instrumental and non-instrumental voice in performance appraisal discussions. Journal of Management, 21, 657 – 669. Landy, F. J., Barnes, J. L., & Murphy, K. R. (1978). Correlates of perceived fairness and accuracy of performance evaluation. Journal of Applied Psychology, 63, 751 – 754. Landy, F. J., & Farr, J. L. (1980). Performance rating. Psychological Bulletin, 87, 72 – 107. Levy, P. E., & Williams, J. R. (1998). The role of perceived system knowledge in predicting appraisal reactions, job satisfaction, and organizational commitment. Journal of Organizational Behavior, 19, 53 – 65. Liden, R. C., Sparrowe, R. T., & Wayne, S. J. (1997). Leader member exchange theory: the past and the potential for future. Research in Personnel and Human Resources Management, 15, 47 – 119. Liden, R. C., Wayne, S. J., & Stilwell, D. (1993). A longitudinal study on the early development of leader – member exchange. Journal of Applied Psychology, 78, 662 – 674. Lind, E. A., Kanfer, R., & Earley, P. E. (1990). Voice, control and procedural justice: instrumental and noninstrumental concerns in fairness judgments. Journal of Personality and Social Psychology, 59, 952 – 959. Lind, E. A., & Tyler, T. R. (1988). The social psychology of procedural justice. New York: Plenum. Mansour-Cole, D. M., & Scott, S. G. (1998). Hearing it through the grapevine: the influence of source, leader relations and legitimacy on survivors’ fairness perceptions. Personnel Psychology, 51, 25 – 54. Maroney, B. P., & Buckely, M. R. (1992). Does research in performance appraisal influence the practice of performance appraisal? Regretfully not! Public Personnel Management, 21, 185 – 196. Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000). Integrating justice and social exchange: the differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43, 738 – 748. Moorman, R. H. (1991). Relationship between organizational justice and organizational citizenship behaviors: do fairness perceptions influence employee citizenship? Journal of Applied Psychology, 76, 845 – 855. Moorman, R. H., Blakely, G. L., & Niehoff, B. P. (1998). Does perceived organizational support mediate the relationship between procedural justice and organizational citizenship behavior? Academy of Management Journal, 41, 351 – 357. Niehoff, B. P., & Moorman, R. H. (1993). Justice as a mediator of the relationship between methods of monitoring and organizational citizenship behavior. Academy of Management Journal, 36, 527 – 556. Ordonez, L. D., Benson, L., & Beach, L. R. (1999). Testing the compatibility test: how instructions, account- B. Erdogan / Human Resource Management Review 12 (2002) 555–578 577 ability, and anticipated regret affect prechoice screening options. Organizational Behavior and Human Decision Processes, 78, 63 – 80. O’Reilly, C. A., Chatman, J., & Caldwell, D. F. (1991). People and organizational culture: a profile comparison approach to assessing person – organization fit. Academy of Management Journal, 34, 487 – 516. Organ, D. W. (1988). Organizational citizenship behavior: the good soldier syndrome. Lexington, MA: Lexington Books. Renn, R. W. (1998). Participation’s effect on task performance, mediating roles of goal acceptance and procedural justice. Journal of Business Research, 41, 115 – 125. Robbins, T. L., Summers, T. P., & Miller, J. L. (2000). Using the group-value model to explain the role of noninstrumental justice in distinguishing the effects of distributive and procedural justice. Journal of Occupational and Organizational Psychology, 73, 511 – 518. Rosse, J. G., & Kraut, A. I. (1983). Reconsidering the vertical dyad linkage model of leadership. Journal of Occupational Psychology, 56, 63 – 71. Scandura, T. A., Graen, G. B., & Novak, M. A. (1986). When managers decide not to decide autocratically: an investigation of leader – member exchange and decision influence. Journal of Applied Psychology, 71, 579 – 584. Schein, E. H. (1990). Organizational culture. American Psychologist, 45, 109 – 119. Scott, M. D., & Lyman, S. (1968). Accounts. American Sociological Review, 33, 46 – 62. Settoon, R. P., Bennett, N., & Liden, R. C. (1996). Social exchange in organizations: perceived organizational support, leader – member exchange, and employee reciprocity. Journal of Applied Psychology, 81, 219 – 227. Skarlicki, D. P., Ellard, J. H., & Kelln, B. R. C. (1998). Third-party perceptions of a layoff: procedural, derogation, and retributive aspects of justice. Journal of Applied Psychology, 83, 119 – 127. Staw, B. M., & Boettger, R. D. (1990). Task revision: a neglected form of work performance. Academy of Management Journal, 33, 534 – 559. Stoffey, R. W., & Reilly, R. R. (1997). Training appraisees to participate in appraisal: effects on appraisers and appraisees. Journal of Business and Psychology, 12, 219 – 239. Sweeney, P. D., & McFarlin, D. B. (1993). Workers’ evaluations of the ‘‘ends’’ and the ‘‘means’’: an examination of four models of distributive and procedural justice. Organizational Behavior and Human Decision Processes, 55, 23 – 40. Tang, T. L., & Sarshfield-Baldwin, L. J. (1996). Distributive and procedural justice as related to satisfaction and commitment. SAM Advanced Management Journal, 61, 25 – 31. Tausky, C., & Chelte, A. F. (1983). Accountability and productivity: some longitudinal data. Work and Occupations, 10, 207 – 220. Taylor, M. S., Masterson, S. S., Renard, M. K., & Tracy, K. B. (1998). Managers’ reactions to procedurally just performance management systems. Academy of Management Journal, 41, 568 – 579. Taylor, M. S., Tracy, K. B., Renard, M. K., Harrison, J. K., & Carroll, S. J. (1995). Due process in performance appraisal: a quasi-experiment in procedural justice. Administrative Science Quarterly, 40, 495 – 523. Thibaut, J., & Walker, L. (1975). Procedural justice: a psychological analysis. Hillsdale, NJ: Earlbaum. Thomas, S. L., & Bretz, R. D. (1994). Research and practice in performance appraisal: evaluating employee performance in America’s largest companies. SAM Advanced Management Journal, 59, 28 – 34. Trice, H., & Beyer, J. (1984). Studying organizational cultures through rites and ceremonials. Academy of Management Review, 9, 653 – 669. Tyler, T. R. (1989). The psychology of procedural justice: a test of group value model. Journal of Personality and Social Psychology, 57, 830 – 838. Tyler, T. R. (1994). Psychological models of the justice motive: antecedents of distributive and procedural justice. Journal of Personality and Social Psychology, 67, 850 – 863. Villanova, P., & Bernardin, H. J. (1989). Impression management in the context of performance appraisal. In R. A. Giacalone, & P. Rosenfeld (Eds.), Impression management in the organization (pp. 299 – 313). Hillsdale, NJ: Erlbaum. 578 B. Erdogan / Human Resource Management Review 12 (2002) 555–578 Wakabayashi, M., Graen, G., Graen, M., & Graen, M. (1988). Japanese management progress: mobility into middle management. Journal of Applied Psychology, 73, 217 – 227. Wakabayashi, M., & Graen, G. B. (1984). The Japanese career progress study: a 7-year follow-up. Journal of Applied Psychology, 69, 603 – 614. Wayne, S. J., & Ferris, G. R. (1990). Influence tactics, affect, and exchange quality in supervisor – subordinate interactions: a laboratory experiment and field study. Journal of Applied Psychology, 75, 487 – 499. Wayne, S. J., & Green, S. A. (1993). The effects of leader – member exchange on employee citizenship and impression management behavior. Human Relations, 46, 1431 – 1440. Wayne, S. J., & Kacmar, K. M. (1991). The effects of impression management on the performance appraisal process. Organizational Behavior and Human Decision Processes, 48, 70 – 88. Wayne, S. J., & Liden, R. C. (1995). Effects of impression management on performance ratings: a longitudinal study. Academy of Management Journal, 38, 232 – 260. Wayne, S. J., Liden, R. C., Kraimer, M. L., & Graf, I. K. (1999). The role of human capital, motivation, and supervisor sponsorship in predicting career success. Journal of Organizational Behavior, 20, 577 – 595. Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organizational support and leader – member exchange: a social exchange perspective. Academy of Management Journal, 40, 82 – 111. Williams, K. J., DeNisi, A. S., Blencoe, A. G., & Cafferty, T. F. (1985). The role of appraisal purpose: effects of purpose on information acquisition and utilization. Organizational Behavior and Human Decision Processes, 35, 314 – 339. Journal of Applied Psychology 2002, Vol. 87, No. 6, 1032–1041 Copyright 2002 by the American Psychological Association, Inc. 0021-9010/02/$5.00 DOI: 10.1037//0021-9010.87.6.1032 Reducing the Performance-Cue Bias in Work Behavior Ratings: Can Groups Help? Richard F. Martell and Keith N. Leavitt This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Montana State University The authors examined whether the performance-cue bias can be reduced by relying on groups as raters. Study participants (N ? 333) were provided with feedback regarding the performance of a workgroup and, after observing the group, assigned to an individual or group rater condition to complete a behavioral rating instrument. Results revealed that when provided with positive (vs. negative) feedback, individuals attributed more effective and fewer ineffective behaviors to the workgroup; however, group ratings were unaffected by the feedback. In addition, feedback biased the decision criteria and false alarm rates of individuals but not of groups. Discussion of when groups may attenuate versus amplify bias in performance appraisal judgments emphasizes 2 key elements— bias magnitude and task perception. credit individuals with critical work-related behaviors and is used for employee evaluation, training, feedback, and assessment center judgments. This research has found that observers led to believe that a group performed well (vs. poorly) attributed more effective and fewer ineffective behaviors to the group (e.g., Martell & Guzzo, 1991; Martell & Willis, 1993). Moreover, performance-cue biases have been observed in the judgments of experienced managers and in field studies (Fichman, 1991; Gardner, Wright, & Gerhart, 2000; Golden, 1992). The importance of the performancecue bias was recently affirmed in a study designed to resolve a long-standing methodological issue regarding whether the bias is an artifact, triggered by the desire to render behavioral ratings consistent with experimenter-provided feedback. In contrast to previous studies, Martell, Guzzo, and Willis (1995) provided no performance-based feedback and instead allowed observers to develop their own self-generated evaluations of the group’s performance. Results indicated that self-generated evaluations biased observers’ behavioral ratings in precisely the same manner as experimenter-provided feedback, demonstrating that the bias is genuine and not an artifact. Despite the implications of the performance-cue bias for theory development and work performance appraisal, the design of interventions aimed at reducing the effects of performance-based expectations on raters’ judgments has received scant attention. The sole exception is a recent study by Baltes and Parker (2000), which found that a halo error training program and a structured recall memory intervention eliminated the performance-cue bias. This finding is encouraging and suggests that the effects of performance cues on work behavior ratings can be circumvented. To date, all previous research has relied on individuals as raters—little is known of the performance-cue bias in group-based judgments. Given the increasing frequency with which groups are employed as decision makers in organizational settings and the finding that groups are less sensitive to some types of judgmental biases, attention to group-based judgments is warranted. The goal of the research reported in this article was to examine whether the performance-cue bias in behavioral ratings can be reduced by relying on groups as opposed to individuals as raters. An abundance of research demonstrates that individuals’ beliefs and expectations can introduce bias in judgments of work performance (Feldman, 1994; Ilgen, Barnes-Farrell, & McKellin, 1993; Motowidlo, 1986). One well-researched example is the performance-cue bias that was first investigated by Staw (1975). In this study, Staw led business students to believe that their workgroup had performed “very well” or “very poorly” on a business simulation exercise. Group process ratings (e.g., cohesiveness, task conflict) and self-evaluations (e.g., ability, motivation) collected at the conclusion of the exercise revealed that the bogus performance feedback produced biased evaluations. Specifically, members of high performing groups rated their performance and the quality of the group’s interaction more favorably than did members of low performing groups. These findings raised the possibility of causal reversals in which knowledge of group performance serves as a cue, causing individuals to ascribe cue-consistent attributes to the target of their assessment. Indeed, Staw concluded that theory development in organization behavior could be seriously compromised because “significant correlations between performance and self-report data may only be reflecting the respondents ‘theories’ of organizational performance rather than actual events” (p. 417). More recent research on the performance-cue bias has examined the effects of observers’ expectations on behavioral rating accuracy. As discussed by Murphy (1991) and Lord (1985a), behavioral rating accuracy is an important criterion measure of work performance; it yields a measure of the accuracy with which raters Richard F. Martell and Keith N. Leavitt, Department of Psychology, Montana State University. Keith N. Leavitt conducted this study under the direction of Richard F. Martell as part of the requirements for his master’s degree in psychology at Montana State University. We gratefully acknowledge Wesley Lynch and Charles Pierce, who served as committee members, and James Robison-Cox for their helpful advice. Eric Szemes’s untiring assistance in conducting this research is much appreciated. Correspondence concerning this article should be addressed to Richard F. Martell, Department of Psychology, Montana State University, Bozeman, Montana 59717-3440. E-mail: martell@montana.edu 1032 PERFORMANCE-CUE BIAS This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Groups as Raters Organizations are increasingly relying on groups to do the work typically reserved for individuals (see Borman, Hanson, & Hedge, 1997; Dumaine, 1994; Guzzo & Shea, 1992; Ilgen, 1999; Katzenbach & Smith, 1993; Shea & Guzzo, 1987), including, among other things, human resource management functions. Reviewing the potential for groups to perform various human resource management functions, Shea and Guzzo (1987) concluded that “it is time for the field of personnel and human resources management to discover formal groups as resources in their own right” (p. 323) and pointed to performance appraisal as one such function. The notion that multiple raters may convey an advantage over individuals in the performance appraisal process can be traced to a theory of rating proposed decades ago (Wherry & Bartlett, 1982). The primary aims of the theory were to examine the effects of factors other than actual ratee performance on performance appraisals and to identify methods for eliminating rater-induced bias. The theory devotes particular attention to the impact of plural raters on performance appraisal judgments (Wherry & Bartlett, 1982). A key proposition is that the use of multiple raters can reduce bias and error in performance appraisals (Wherry & Bartlett, 1982; see Theorems 25–27). This prediction rests largely on the belief that multiple raters can bring nonredundant information to the judgment task. More recent thinking suggests that multiple raters, if assembled into a group, may correct individual member errors and biases and, perhaps, bolster individual rater motivation. Consistent with this proposition, group discussion among multiple assessors is a critical component of assessment centers and is thought to be helpful in heightening rater accuracy and decreasing bias (Howard, 1997; Lowry, 1997; Spychalski, Quinones, Gaughler, & Pohley, 1997; Zedeck, 1986; but see Sackett & Wilson, 1982). Research on social judgment biases, including the correspondence bias (Wright & Wells, 1985), the theory-perseverance effect (Wright & Christie, 1989), the hindsight bias (Stahlberg, Eller, Maas, & Frey, 1995, Experiment 2) and the consensus underutilization effect (Wright, Lu?u?s, & Christie, 1990), provide further reason to believe that groups may render less biased assessments than individuals. In each of these studies, individual-level bias was attenuated (and often eliminated) by groups. Consider, for example, the correspondence bias whereby perceivers tend to ignore compelling situational explanations when determining the cause of a person’s behavior and instead wrongly assume that the behavior is a direct manifestation of a person’s underlying dispositional state. The correspondence bias reflects a type of cognitive shortcut in which perceivers too readily succumb to an implicit theory that behavior is caused by the underlying trait or disposition to which it corresponds (Gilbert & Malone, 1995); it reveals heuristic thinking and is a consequence of faulty information processing. However, as with many social judgment biases, the magnitude of the correspondence bias is not large; not all individuals necessarily succumb to the bias. This suggests that in a group context, it is likely that one or more individuals are unaffected by the correspondence bias and thus are in a position to impose their (accurate) beliefs on the group (Hinsz, Tindale & Vollrath, 1997; Wright & Wells, 1985). In addition, given the public nature of group deliberation in which members are exposed to contrary beliefs and alternative positions, groups may more often abandon heuristic 1033 thinking and engage in a deeper, more systematic processing of information than would individuals, especially if group members believe there is a necessarily correct answer to the task at hand (Stasser & Stewart, 1992). These features of group discussion provide reason to believe that reliance on implicit theories and the biases they produce in work performance ratings may .