|
Article Excerpt
We examined 145 leaders' recall of their multisource feedback (MSF) 9 months after receiving the feedback. Leaders recalled more strengths than weaknesses, but these memories had only a small relationship with the actual feedback received. Leaders were more likely to recall feedback related to their consideration of employees and performance orientation than feedback related to developing and recognizing employees, and they were more likely to recall feedback from supervisors and direct reports than peers. Self versus other rating discrepancies were unrelated to recall, and recall of MSF was not related to subsequent improvement in MSE
Keywords: multisource feedback; performance management
**********
Multisource feedback (MSF) is expected to influence leader development because feedback from multiple sources (supervisor, peers, and direct reports) provides a more complete perspective of a leader's performance than traditional top-down appraisals. The leader is expected to use the MSF to set improvement goals and enhance his or her effectiveness. But a recent meta-analysis of 24 longitudinal studies (Smither, London, & Reilly, 2005) found that although effect sizes were generally positive, the magnitude of improvement in direct report, peer, and supervisor ratings over time was generally small (d = .15, .05, and .15 for direct report, peer, and supervisor feedback, respectively). Also, a large percentage of the variance in effect sizes was not explained by sampling error, even after accounting for the effects of moderator variables, thereby suggesting that other factors might affect the extent of behavior change associated with MSE Moreover, Smither, London, and Reilly (2005) argued that performance improvement should be more likely for some feedback recipients than others, and they discussed factors that might play a role in determining the extent of behavior change and performance improvement following MSF (e.g., initial reactions to feedback, personality, beliefs about change, goal setting, taking action).
Although the Smither, London, and Reilly (2005) model suggests several promising areas for research, it did not address an unstated assumption regarding MSF that, to date, appears to be taken for granted. That is, it is assumed that leaders will attend to and later recall key aspects of their MSF. If leaders do not recall (or do not accurately recall) key aspects of their MSF, then MSF is unlikely to have any effect on developmental actions and subsequent performance. Thus, the extent to which leaders can later recall their MSF might play a key role in determining whether they subsequently improve their performance.
We believe that there are good reasons to believe that leaders might have difficulty recalling their MSE Leaders who receive MSF are presented with an enormous, perhaps overwhelming, amount of data. For example, consider an MSF report that presents the mean and distribution (e.g., standard deviation, range, or frequency distribution) of ratings from each of three sources (supervisor, peers, direct reports) on 40 items. In this example, the MSF report would contain 240 data points (i.e., Mean and Distribution x 40 items x 3 sources), excluding any narrative comments that might also be included in the report. No one expects feedback recipients to attend to or recall 240 data points. Instead, feedback recipients are usually asked to read through their data (with or without the help of a facilitator) and attempt to extract the most important information. Presumably, subsequent goal setting and development efforts (if they occur at all) will be guided by whatever feedback recipients subsequently recall from their MSE However, no study of MSF has examined what feedback recipients actually recall about their feedback or whether what they remember is related to any subsequent changes in behavior.
In this article, we address this gap in the literature by examining what feedback recipients recalled about their MSF 9 months after receiving their feedback. For example, are people more likely to recall positive rather than negative feedback? Are they more likely to recall feedback from some rater sources than others? Are they more likely to recall feedback about some dimensions or aspects of their behavior than others? Are they likely to recall feedback about specific behaviors, or are they more likely to recall broad, trait-oriented impressions? We also examine whether initial emotional reactions to MSF are related to recall of the feedback and whether recall of the feedback is related to changes in behavior.
Are Feedback Recipients More Likely to Recall Positive or Negative Feedback?
In most applications of MSF, the mean rating on all or nearly all items is above the midpoint of the rating scale. For example, in the current study, the mean rating on a 9-point rating scale ranged from 6.6 to 7.7 across rater sources and performance dimensions. From this perspective, most feedback recipients can say that, overall, they received favorable feedback. Hence, feedback recipients might be expected to recall more positive than negative feedback.
However, cognitive processing and biases might affect memory for positive versus negative feedback. For example, prospect theory (Kahneman & Tversky, 1979) shows that, in decision making, losses have more impact than comparably sized gains. An extensive literature review by Baumeister, Bratslavsky, Finkenauer, and Vohs (2001) concluded that bad information is processed more thoroughly than good across a broad range of psychological phenomena. They argue that this is adaptive from an evolutionary perspective because ignoring a negative threat could result in harm whereas ignoring a positive opportunity could lead only to regret.
Baumeister and Cairns (1992) examined how individuals processed and remembered negative feedback. The highest memory scores in their experiment were obtained for the small amount of negative feedback embedded in the generally positive feedback. In a study of student reactions to feedback from teachers, Coleman, Jussim, and Abraham (1987) found that negative feedback had a stronger effect on students' perceptions of their own performance than positive feedback. These studies would suggest that feedback recipients might be more likely to remember negative than positive feedback.
However, negative information about the self might be processed differently than negative information about others. For example, Baumeister and Cairns (1992) found that repressors (i.e., people who are likely to display defensive or self-deceptive responses) sometimes devoted less processing time to unfavorable personality feedback and later recalled positive feedback better than negative. In one study, Sedikides and Green (2000) found that people recalled central (to self-concept), negative self-referent behaviors poorly relative to central, positive self-referent behaviors and to central, negative other-referent behaviors. In another study, they showed that this finding was due in part to individuals allocating limited processing time to central, negative self-referent behaviors. That is, people appear to protect themselves against self-threat by allocating minimal processing resources to central negative self-referent information (Sedikides & Green, 2000). D'Argembeau, Comblain, and Van der Linden (2005) found that people recalled positive trait information better than negative trait information when the information referred to the self but not when the adjectives described a celebrity. They also found that this effect was present only in a recall (but not a recognition) condition, thereby suggesting that the effect is in part due to protective control exerted on memory retrieval.
Baumeister et al. (2001) noted that the motivation to have a positive view of ourselves can reflect self-protection (avoiding a bad opinion of ourselves) or self-enhancement (gaining a good opinion of ourselves). Baumeister and Cairns (1992) examined how individuals processed and remembered negative feedback. They found that negative feedback elicited defensive responses (e.g., avoiding exposure, thoughts to refute the feedback) during encoding. Based on this and other studies, Baumeister et al. concluded that we are more motivated to avoid bad self-definitions than to pursue good ones. But both self-protection and self-enhancement can be expected to result in greater recall of positive than negative feedback.
Based on the preponderance of evidence about recall of self-relevant information (such as MSF), we hypothesize
Hypothesis 1: Feedback recipients will recall more positive than negative feedback.
Are Feedback Recipients More Likely to Recall Feedback From Some Rater Sources Than Others?
Feedback recipients might think that some rater sources are better able to offer useful feedback than others (Farr & Newman, 2001). For example, a supervisor's judgments concerning a subordinate manager's performance are likely to directly affect the manager (e.g., influencing formal performance appraisals and allocation of rewards to the manager).
Feedback from direct reports might be seen as especially valuable because a manager's success often requires the cooperation of direct reports and because feedback from direct reports might be less readily available than feedback from one's supervisor. Even if a manager were to ask a direct report to provide face-to-face feedback, the manager might wonder whether the feedback reflects the direct report's true opinions (e.g., wondering whether direct reports withhold their true perceptions for fear of retaliation by the manager). Of course, managers might believe that direct reports are better able to evaluate some dimensions of performance (e.g., delegation, leadership, interpersonal skills) than others (e.g., budgeting, decision making, quantity of work) (McEvoy, 1990). Some managers might be concerned that direct reports will give low ratings to demanding managers (Bernardin, 1986). Research has shown that managers' reactions to direct reports' feedback are generally favorable (e.g., Smither et al., 1995; Smither & Walker, 2001). Bernardin, Dahmus, and Redmon (1993) found that feedback recipients had more positive attitudes toward upward feedback from their direct reports when they also received feedback from their supervisors (as is common in many but not all applications of MSF).
The perceived value of peer feedback is likely to depend on the feedback recipient's situation. In some instances, feedback recipients might have little interaction with peers. Therefore, peers would often have to base their ratings on indirect information (rather than direct observation of the manager's behavior). Also, in some situations, peers might have less impact on the manager's success and rewards than do supervisors and direct reports. In other situations, peers might work closely with the feedback recipient (e.g., as members of a project team) and their feedback might therefore be especially valuable. In other situations, the perceived value of peer feedback might depend on the feedback recipient's degree of respect for his or her peers (e.g., the extent to which the peers are knowledgeable, competent, trustworthy, and so on). In the current study, our feedback recipients were leaders in an elementary school district (e.g., principals, assistant principals, staff directors) and a retail organization (e.g., district and regional sales managers). In both organizations, leaders were usually not colocated with their peers, interacted with them only occasionally, and were not dependent on peers for their own work success. Based on the above, we hypothesize
Hypothesis 2: Feedback recipients will be more likely to recall feedback from direct reports and supervisors than from peers.
Are Feedback Recipients More Likely to Recall Feedback About Some Performance Dimensions or Aspects of Their Behavior Than Others?
Just as beauty is said to be in the eye of the beholder, some aspects of a...
|
