Fuel for litigation? Links between procedural justice and multisource feedback.

Approximately 40 percent of organizations use multisource feedback (MSF) in some capacity (Bracken et al., 2001). Flatter organizations, formalized leadership development efforts, and cross-functional work arrangements have contributed to the prevalence of MSF processes (Green, 2002; Toegel and Conger, 2003). Rather than performance assessment resting solely with the supervisor, MSF (also referred to as 360-degree feedback) prescribes having multiple sources (e.g., direct reports, customers, peers, supervisor, self) provide performance feedback to employees on behaviors or competencies important for organizational success (Bracken and Timmreck, 1999; Tornow, 1993). The hallmark of MSF, and an obvious departure from traditional performance appraisal, is incorporation of data from these different perspectives. The feedback from multiple constituencies assists the feedback recipient with performance improvement efforts and may also serve as input to organizational decision making, such as staffing, compensation, and succession planning (Bracken and Timmreck, 1999).

Given its increasing administrative use (Toegel and Conger, 2003), one would expect MSF to be examined with the same legal scrutiny as that given to traditional performance appraisal processes. Several published studies have offered guidelines for appraisal components that the courts tend to review favorably (e.g., Werner and Bolino, 1997). However, this work is just now extending to multisource feedback (Bernardin and Tyler, 2001). Thus, the existing literature is lean in regards to the legal fortitude of MSF and remains mostly silent concerning areas that are unique to these programs. Furthermore, to our knowledge, no research has examined how aspects of a multisource feedback process may affect employee attitudes about litigation, most notably intention to file an employment discrimination claim against the company in the first place. This article conceptually explores the relationship between multisource feedback and litigation intentions, rather than judicial decisions. We incorporate the literature relating to procedural justice as the primary theoretical basis for our propositions. As indicated in Figure 1, we propose that design features or decisions made about an MSF system influence employee perceptions of procedural justice, subsequently affecting employee litigation intentions.

[FIGURE 1 OMITTED]

MULTISOURCE FEEDBACK

A typical MSF process involves choosing raters, gathering feedback (usually through a confidential or an anonymous questionnaire format), compiling the results, and providing an integrative report to the feedback recipient, with average results reported by item or construct for each rater group. Self-ratings and those from other rater groups typically diverge (see Harris and Schaubroeck (1988) for a meta-analytic review). Presumably, these discrepancies increase feedback recipients' self-awareness, enabling them to identify opportunities for positive changes in their behavior.

Research has investigated the extent to which discrepancies between self-ratings and ratings from other sources (e.g., supervisor, peers, subordinates) influence the behavior of feedback recipients. Evidence suggests that leaders who overestimate their behavior in reference to their subordinates (overraters) improve their leadership behavior as measured by follow-up performance ratings made by subordinates (Atwater et al., 1995; Johnson and Ferstl, 1999). Overraters also tend to decrease their self-ratings to better align with their subordinates' ratings (Atwater et al., 1995, 2000; Johnson and Ferstl, 1999) and seek out additional feedback from their subordinates (Waldman and Atwater, 2001). In contrast, leaders rating themselves lower relative to their subordinates (underraters) tend to retain their current behaviors, but increase their self-ratings to be more consistent with those made by their subordinates (Atwater et al., 1995). Generally, underraters and those with consistent, high ratings tend to be better performers than overraters based on such criteria as supervisor ratings (Atwater et al., 1998; Ostroff et al., 2004) and assessment center results (Atkins and Wood, 2002). Thus, although the use of MSF may benefit both overraters and underraters, managers whose subordinates rate their performance as low or moderate tend to show the most increase in subsequent subordinate ratings, even beyond the effects of regression to the mean (Walker and Smither, 1999).

Some research suggests primarily positive outcomes from MSF, including improvements in self-awareness, team interactions, organizational performance (Green, 2002), and development planning (Mabey, 2001). For example, university employees who participated in MSF reported higher satisfaction with career development and rated their employer more favorably compared with other employers (Mabey, 2001). Other research has focused on the effects of multiple feedback sources, as well as recipients' personal characteristics. For example, managers receiving supervisor (Waldman and Atwater, 2001) and peer ratings (Maurer et al., 2002) that were lower than their self-ratings were more likely to view the MSF positively, presumably because they view this feedback as reflective of the organization and instrumental in helping them succeed in the organization (Maurer et al., 2002; Waldman and Atwater, 2001). Peers and subordinates, in contrast to other sources, seem to wield the most influence on ratees, with positive ratings from these sources associated with ratees' positive attitudes toward MSF (Maurer et al., 2002). Managers receiving low subordinate ratings may be more likely to seek additional feedback from their subordinates (Waldman and Atwater, 2001). Manager-initiated discussions with subordinates tend to improve the manager's performance, especially when the feedback is unfavorable (Walker and Smither, 1999).

MSF processes are not always effective. Atwater et al. (2000) found that only half of the police leaders in their study improved after participating in MSF and proposed that other factors, such as cynicism and commitment to subordinates, may influence the extent to which feedback is effective. Supervisors receiving low subordinate ratings were less likely to exhibit loyalty to or help their subordinates, while those receiving positive ratings were more likely to demonstrate a higher level of commitment to their subordinates. Furthermore, leaders with more cynicism were less committed to their subordinates and were less likely to improve their behavior (Atwater et al., 2000). In addition to individual factors, the organization must create the context for MSF to be used effectively. For example, post-MSF developmental activity is more likely to occur when the work context is supportive of learning, the feedback recipients believe it is possible to improve skills, the recipients have higher self-efficacy (Maurer et al., 2002), and the recipients receive leader and organizational support (e.g., progress checks) for their development activities (Mabey, 2001). Maurer and Palmer (1999) found that managers' intentions to improve after receiving feedback depended on perceived benefits, social pressures from others in the workplace, and their sense of control over improvement; that is, the greater the feedback recipients felt they could improve, the greater their intention to improve based on their feedback.

Interestingly, comparatively little research exists on how the design of an MSF process influences employee attitudes and behaviors. Design decisions may include such factors as: how the resulting data will be used (developmental or administrative purposes), what items to include on the MSF questionnaire (e.g., task-oriented questions, management competencies, open-ended questions), whether a sample of feedback providers will be selected (versus an entire population), who selects the sample (e.g., feedback recipient, organization, human resources department, supervisor), and whether or not the raters' identities will remain anonymous. Although a majority of companies use MSF for employee development (Rogers et al., 2002), combining developmental and administrative objectives potentially bolsters return on investment (Waldman et al., 1998). In addition, purposes may be difficult to disentangle, with information from developmentally-oriented processes leaking out to decision makers (Ghorpade, 2000). Therefore, MSF likely reflects a continuum between the two approaches, with developmental data being used in some capacity for employee evaluation.

EMPLOYMENT DISCRIMINATION

Evaluating individuals for selection, promotion, compensation, termination, or other human resource decisions falls within the realm of federal legislation. The use of multisource feedback as a means for making decisions about employees underlies the need for the MSF process to be non-discriminatory in reference to protected characteristics. Workers are protected from discrimination on the basis of race, sex, color, religion, national origin (Title VII, Civil Rights Act of 1964, and its amendments), age (Age Discrimination in Employment Act, 1967), and disability (Americans with Disabilities Act, 1990). Therefore, any intentional or unintentional discrimination based on these characteristics is deemed illegal. Possible defenses for discrimination include acceptable validation evidence or the exception of a bona fide occupational qualification (BFOQ) to the extent that it rises to the level of a business necessity.