Inherent Difficulties in thePerformance Appraisal Process

The performance management cycle begins with objective setting where the appraisee ought to be oriented about performance expectations for the given performance period. Throughout the period, performance is measured officially through the performance appraisal exercise, where the appraiser is given the chance to give performance feedback through an interview. The performance management system is meaningfully linked to rewards based on outcomes or results. Following all these, modifications are implemented to objectives and activities, including the drafting of an individual development plan for closing competency gaps (Clark, 1998).

Types of Performance Appraisal Systems

Traditionally, employee performance has been evaluated solely by supervisors. Recently, however, organizations have realized that supervisors see only certain aspects of an employee’s behavior. For instance, a manager might see only 30% of his staff’s behavior; the rest is observed by customers, peers, and support staff in other parts of the organization. Furthermore, the staff might behave differently around her supervisor than around other people. Consequently, to obtain an accurate view of the staff’s performance, these other sources should provide feedback. The buzzword for using multiple sources to appraise performance is 360-degree feedback (Gruner, 1997). Sources of relevant information include supervisors, peers, subordinates, customers, and self-appraisal. According to Conway and Huffcutt (1997), there is often very little agreement in the way that two supervisors evaluate an employee or that a supervisor and a peer might rate an employee. Interestingly, supervisors whose self-ratings agree with others’ ratings tend to be better performers than supervisors whose ratings are not consistent with others’ (Witt, 1996).

 Supervisor Appraisals

By far, the most common type of performance appraisal is the supervisor rating. In fact, Bernardin & Beatty (1984) estimated that over 90% of all performance appraisals are conducted using supervisors’ ratings of performance. Supervisors are best able to evaluate the extent to which an employee contributes to the overall success of the organization. Through supervisors may not see every minute of an employee’s behavior, they do see the end result. A superior may not actually see a staff sign up customers but will review the overall output for the day.

 Peer Appraisals

Whereas supervisors see the results of an employee’s efforts, peers often see the actual behavior. Peer ratings usually come from employees who work directly with an employee. An employee may be rated by those in the same level or position. However, other employees in the organization, those who often come in contact with the employee, can also provide useful information.

Research has shown that peer ratings are fairly reliable only when the peers who make the ratings are similar to and sell acquainted with the employees being rated (Mumford, 1983). Most important, peer ratings have been successful in predicting the future success of promoted employees as they correlate highly with supervisor ratings (Cederbloom, 1989). But even through peer ratings appear promising, few organizations use them. One reason could be that peer ratings are lenient when used for evaluation purposes bit not when they are used only to provide feedback (Farh, Cannella, & Bedeian, 1991).

Research suggests that certain employees are more lenient in their peer ratings than are other employees. Saavedra & Kwun (1993) found that high performers evaluate their peers more strictly than do low performers. This difference in ratings is probably because employees compare others to themselves. Thus, the average employee does not appear impressive to a high performer but may to a less productive employee. Though peers may provide a unique view of performance, employees tend to react worse to negative feedback from peers than they do to feedback from experts (Albright & Levy, 1995). Employees who score high in self-esteem, high in self-monitoring, and low in individualism react most favorably to peer ratings (Long, Long & Dobbins, 1998).

 Subordinate Appraisals

Subordinate feedback, also called upward feedback is an important component of 360-degree feedback, as subordinates can provide a very different view about a supervisor’s behavior (Whetstone, 1994). However, with the exception of students rating teachers, formal methods are neither common nor well regarded by managers (McEvoy, 1990). Subordinate ratings can be difficult to obtain because employees fear a backlash if they unfavorably rate their supervisor, especially when a supervisor has only one or two subordinates. However, subordinates’ feedback can be encouraged if supervisors appear open to employee comments (Baumgartner, 1994) and if the ratings are made anonymously (Antonioni, 1994). Interestingly, subordinate ratings correlate highly with upper management ratings of supervisors’ performance (Furnham & Stringfield, 1994). Research indicates that subordinate feedback can enhance managerial performance, especially that of poorly performing managers (Walker, 1997). This improvement in performance holds especially for areas targeted for improvement (Clarke, Rogers, & Miklos, 1996).

 Customer Appraisals

Though it would be unlikely that an organization would ask customers to fill out a performance appraisal instrument on an employee, organizations do value customer feedback. Informally, customers provide feedback on employee performance by filing complaints or complimenting a manager about one of his subordinates. Formally, customers provide feedback by completing evaluation cards (Farh, Canella, & Bedeian, 1991). 

Self-appraisal

Allowing an employee to evaluate her own behavior and performance is a technique used by 12% of a sample of organizations (Laser & Wisdom, 1977). Research on self-appraisal, however, has demonstrated that self-appraisals tend to suffer from leniency (Holdback, 1978; Meyer, 1980) and correlate moderately (r=.29) with actual performance (Maybe & West, 1982) and poorly with subordinate ratings (London & Wholes, 1991). However, when evaluations are made with clear rating standards and social comparison information, agreement is increased between self- and supervisor ratings (Keeping & Sulky, 1996).

The leniency found in the self-ratings of US workers may not generalize to other countries. Farh, Dobbins, and Cheng (1991) found that the self-ratings of Taiwanese workers suffered from modesty rather than leniency However, Furnham & Stringfield (1994) and Yu and Murphy (1993) found leniency in the self-ratings of Mainland Chinese employees. Further research is still needed to investigate potential cultural differences in Mainland Chinese ratings.

Self-appraisals of performance appear to be the most accurate when the purpose of the self-appraisal is for either research or performance appraisal review interviews rather than for such administrative purposes as raises or promotions (Williams & Levy, 1992), and when employees believe that an objective record of their performance is available with which the supervisor can compare the self-appraisal (Farh & Werbel, 1986).

 Systems for Evaluating Performance

Trait-focused appraisal systems. A trait-focused system concentrates on employees’ attributes such as their dependability, assertiveness, and friendliness. Though commonly used, trait focused performance appraisal instruments are not a good idea because they provide poor feedback and thus will not result in employee development and growth (Kingstrom & Bass, 1981). For example, in a performance review meeting in which the supervisor tells an employee that she received low ratings on responsibility and friendliness, the employee is likely to become defensive. Furthermore, the employee will want specific examples the supervisor may not have available (Kingstrom & Bass, 1981).

Behavior-focused performance appraisal systems. Behavior-focused instruments focus on what an employee does. Instead of rating them on personal traits, a behavior-focused instrument would rate him or her on specific behaviors. For example, in the case of a bank teller, some behaviors that may be rated on are as follows: “Knows customers’ names”, and “Thanks customer after each transaction.” The obvious advantage to a behavior-focused system is the amount of specific feedback that can be given to each employee. Further, the focus on behavior rather than traits does not only reduce employee defensiveness but reduces legal problems (Kingstrom & Bass, 1981).

There are various methods for rating behavior, as follows:

Graphic rating scales. The most common rating scale is the graphic rating scale. Such scales are simple, with 5 to 7 points accompanied by words such as good and poor anchoring the ends of the scale. The obvious advantage  to graphic rating scales is their ease of construction and use, but they have been criticized because of their susceptibility to such rating errors as halo and leniency (Kingstrom & Bass, 1981)

Behaviorally anchored rating scales. P.C. Smith and Kendall (1983) developed behaviorally anchored rating scales (BARSs), which use critical incidents (samples of behavior) to provide meaning to the numbers on a rating scale. To use the scale when actually rating performance, the supervisor compares the incidents she has recorded for each employee to the incidents on the scale. This can be done in one of two ways. The most accurate (and time consuming) method compares each of the recorded incidents to the anchors and records the value of the incident on the scale that most closely resembles the recorded incident. The value for each incident is summed and divided by the total number of incidents recorded for that dimension; this yields an average incident value, which is the employee’s rating for that particular job dimension (Smith & Kendall, 1983). In the second method, which is easier but is less accurate, all of the recorded incidents are read to obtain a general impression of each employee. This general impression is compared to the incidents that anchor each scale point. The scale point next to the incident that most closely resembles the general impression gained from the incidents then becomes an employee’s score for that dimension (Smith & Kendall, 1983).

Evaluation of Performance Appraisal Methods

We now come to the question of assessing which appraisal method is best. Research has shown that more complicated techniques such as BARS, forced-choice scales, and mixed standard scales are only occasionally superior to inexpensive and uncomplicated graphic rating scales (Giffin, 1989). In fact, behavioral anchors sometimes bias supervisors’ ratings by forcing them to concentrate on specific behaviors (Murphy & Constans, 1987). Yet graphic rating scales are seldom superior to these more complicated rating methods.

Although the more complicated techniques are only more psychometrically sound, they still have some advantages over graphic rating scales. Because employees are directly involved in creating techniques such as BARS, they tend to see performance evaluation results as being more fair. Furthermore, many supervisors who make such ratings prefer many of the more complicated behavioral approaches. Finally feedback from BARS may lead to greater increases in future performance than feedback from graphic rating scales (Hom, DeNisi, Kinicki, & Bannister, 1982).

Though many of the behavioral methods yield similar results, the same is not true when comparing subjective and objective ratings. A meta-analysis by Bommer, Johnson, Rich, Podsakoff, and Mackenzie (1995) indicated that objective and subjective results are only slightly correlated (r=.39). Interestingly, there was a stronger relationship between objective and subjective ratings of quantity (r=.38) than between objective and subjective ratings of quality (r=.24).

From a legal perspective, courts are more interested in the due process afforded by a performance appraisal system that in its technical aspects. After reviewing 295 circuit court decision regarding performance appraisal, Werner & Bolino (1997) concluded that performance appraisal systems are most likely to survive a legal challenge if they are based on job analysis, if raters receive training and written instructions, if employees are allowed to review results, and if ratings from multiple raters are consistent.

Rating Errors

Some of the errors that may be committed in appraising performance are discussed below:

Distribution errors. A common type of error in evaluating employee performance involves the distribution of ratings on  a rating scale; such errors are known as distribution errors. One kind of distribution error is called leniency error because certain raters tend to rate every employee at the upper end of the scale regardless of the actual performance of the employee. A related error is central tendency error, which results in a supervisor’s rating every employee in the middle of the scale. Still another error, strictness error, rates every employee at the lower end of the scale.

These types of errors pose problems for an organization because two employees doing equal work will receive different ratings if one employee is supervised by a lenient rater and another by a strict rater. This problem can be eliminated partly by having several people rate each employee (Kane & Lawler, 1979), although this is not often feasible, especially in small brand offices with only one manager or supervisor.

Halo errrors. A halo error occurs when a rater allows either a single attribute or an overall impression of an individual to affect ratings that she makes on each relevant job dimension. Halo effects occur especially when the rater has little knowledge of the job and is less familiar with the person being rated (Kozlowski, Kirsh, & Chao, 1986).

Usually, halo error is statistically determined by correlating the ratings for each dimension with those for the other dimensions. If there are highly correlated, halo error is often said to have occurred. However, some authors argue that many times consistent ratings across several dimensions indicate not error but actual employee performance. Halo errors may or may not be a serious problem, but they can be reduced by having supervisors rated each trait at several times. That is, the supervisor might rate the employee on attendance one day and then rate her on dependability the next day (Balzer & Sulzky, 1992).

Proximity errors. Proximity errors occur when a rating made on one dimension affects the rating on the dimension that immediately follows it on the rating scale. With proximity error, only the dimensions physically located nearest a particular dimension on the rating scale are affected; the reason for the effect, in fact, is the close physical proximity of the dimension rather than the overall impression (Balzer & Sulzky, 1992). 

Contrast errors. The performance ratings one person receives can be influenced by the performance of the previously evaluated person. These errors can occur between separate performance evaluations of the same person. That is, the ratings received by one person on one performance appraisal will affect the ratings made on an appraisal six months later (Bravo & Kravitz, 1996).

Contrast effects occur only when the person making the evaluation actually sees the employee perform and rates the employee during both rating periods. Even if a new supervisor reads that an employee’s previous evaluations were excellent but observes poor performance by the employee, she will probably continue to give excellent ratings – even though the employee’s performance deteriorated. Smither et al (1988) call this rating error assimilation.

 Sampling Problems

Recency effect. Performance appraisals are typically conducted once or twice a year. The evaluation is designed to cover all of the behaviors that have taken place during the previous 6 months to a year. Research has demonstrated, however, that recent behaviors are given more weight in the performance evaluation than behaviors that occurred during the first few months of the evaluation period. Such an effect penalizes workers who performed well during most of the period but tailed off toward the end, and it rewards workers who save their best work until just before the evaluation (Bravo & Kravitz, 1996).

Infrequent observation. Another problem that affects performance appraisals is that many managers or supervisor do not have the opportunity to observe a representative sample of employee behavior. Infrequent observation occurs for two reasons. First, managers are often so busy with their own work that they often have no time to “walk the floor” and observe their employees’ behavior. Instead, they make inferences based on completed work or employee personality traits (Conway & Huffcutt, 1997).

This problem can be alleviated somewhat by having several raters evaluate the employee. Other raters can be other supervisors, peers, and even customers. A meta-analysis conducted by Conway and Huffcutt (1997) indicated that supervisor ratings on the average correlate .34 with peer ratings. Thus, even though the two groups tend to agree with one another, the agreement is certainly not perfect.

Cognitive Processing of Observed Behavior

Observation of behavior. Just because an employee’s behavior is observed does not guarantee that it will be properly remembered or recalled during the performance appraisal. Cooper (1981) indicates that raters recall those behaviors that are consistent with the general impression of an employee (a halo). And the greater the time interval between the actual behavior and the performance rating, the greater the probability that halo and distortion errors occur. Furthermore, raters who are familiar with the job being evaluated recall more judgments about performance but fewer behaviors than do raters who are unfamiliar with the job (Cooper, 1981).

But even though memory-based ratings lead to more distortion, in many circumstances they are more accurate than ratings made immediately after the behaviors occur (Murphy & Blazer, 1986). The reason for these increases in halo and accuracy is not clear. Supervisors perhaps realize that it will be a long internal between observation of employee behavior and the formal evaluation of that behavior and that they will not be able to remember specific behaviors. Thus, they form an overall impression of the employee and an ideal and a poor employee and evaluate the employee on the basis of comparison with the ideal (Murphy & Blazer, 1986).

Emotional state. The amount of stress under which a supervisor operates also affects her performance ratings. Srinivas and Motowidlo (1987) found that raters who were placed in a stressful situation produced ratings with more errors than did raters who were not under stress. This finding is important because performance evaluations are often conducted hurriedly  as supervisors evaluated employee performance so that they can return to their “real” work.

Raters who like the person being rated may be more lenient and less accurate in rating employees than are raters who neither like nor dislike their employees. But this does not mean that a person who is liked will always receive higher ratings than someone who is disliked. The rater may overcompensate in an effort to be “fair”. The rater’s feelings or affect, toward an employee may interfere with the cognitive processing of actual performance information (Srinivas and Motowidlo, 1987).

Research has also indicated that racial bias exists in performance evaluations. Kraiger and Ford (1985) conducted a meta-analysis of 74 studies and found that White raters gave higher performance ratings to White employees and that Black raters gave higher ratings to Black employees. Interestingly, this bias occurred only with studies involving real organizations; laboratory research seldom reveal racial bias in rating.

Communicating Appraisal Results to Employees

Perhaps the most important use of performance evaluation data is to provide feedback to the employee and assess his or her strengths and weaknesses so that further training can be implemented. Although this feedback and training should be an ongoing process, the semi-annual evaluation might be the best time to formally discuss employee performance. Furthermore, holding a formal review interview places the organization on better legal ground in the event of a lawsuit (Field & Holley, 1982).

Normally, in most organizations a supervisor spends a few minutes with employees every six months to tell them about the scores they received during the most recent performance evaluation period. This process is probably the norm because most managers do not like to judge others; because of this dislike, they try to complete the evaluation process as quickly as possible (Field & Holley, 1982).

 Furthermore, seldom does evaluating employees benefit the supervisor. The best scenario is to hear no complaints, and the worst scenario is a lawsuit. In fact, one study demonstrated that dissatisfaction and a decrease in organizational commitment occurs even when an employee receives an evaluation that is “satisfactory” but not outstanding (Pearce & Porter, 1986). Finally, in the “tell and sell” approach to performance appraisal interviews, a supervisor “tells” an employee everything she has done poorly and then “sells” her on the ways in which she can improve. This method, however, accomplishes little.

There are certain techniques that will make the performance appraisal interview more effective, and these are discussed in the following sections.

Both the supervisor and the employee must have time to prepare for the review interview. Both should be allowed at least 1 hour to prepare before an interview and at least 1 hour before the interview itself (Pearce & Porter, 1986).

The interview location should be a neutral place that ensures privacy and allows the supervisor and the employee to face one another without a desk between them as a communication barrier (Rhoads, 1997). Performance appraisal review interviews should be scheduled at least once every 6 months for most employees and more often for new employees. Review interviews are commonly scheduled 6 months after an employee begins working for the organization. It is important to note that while formal performance review interviews occur only twice a year, informal “progress checks” should be held throughout the year to provide feedback (Rhoads, 1997).

While preparing for the interview, the supervisor should review the ratings she has assigned to the employee and the reasons for those ratings. This step is important because the quality of feedback given to employees will affect their satisfaction with the entire performance appraisal process (King, 1984). Meanwhile the employee should rate her own performance, using the same format as the supervisor. The employee also should write down specific reasons and examples that support the ratings she gives herself (King, 1984).

At the outset of the interview, the supervisor should communicate the following: 1) the role of the performance appraisal – that making decisions about salary increases and terminations is not its sole purpose; 2) how the performance appraisal was conducted; and 3) how the evaluation process was accomplished. It is advisable that the supervisor also communicate her own feelings about the performance appraisal process (Kelly, 1984).

The review process is probably best begun with the employee communicating her own ratings and her justification for those ratings. Research indicates that employees who are actively involved in the interview from the start will be more satisfied with the results. The supervisor then communicates his ratings and his reasons for them (King, 1984).

At the conclusion of the interview, goals should be mutually set for future performance and behavior, and both supervisor and employee should understand how these goals are met (Cederbloom, 1982).

On the whole, the performance appraisal process, with its inherent difficulties, may be leveraged on to improve organizational productivity. Proper management of this process shall help motivate employees, and ultimately, bring more to the enterprise’s coffers.

References

Albright, M. & Levy, P. (1995). The effects of source credibility and performance discrepancy on reactions to multiple raters. Journal of Applied Social Psychology, 25(7), 537-548.

Antonioni, D. (1994). The effects of feedback accountability on upward appraisal ratings. Personnel Psychology, 47(2), 349-356.

Balzer, W. K., & Sulzky, L. M. (1992). Halo and performance appraisal research: A critical examination. Journal of Applied Psychology, 77(6), 971-986.

Baumgartner, J. (1994). Give it to me straight. Training: Development, 48(6), 49-51.

Bernardin, H. & Beatty, R. (1984). Performance appraisal: Assessing human behavior at work. Boston: Kent.

Bommer, W. H., Johnson, J. L., Rich, G. A., Podsakoff, P. M., & Mackenzie, S. B. (1995). On the interchangeability of objective and subjective measures of employee performance: A meta-analysis. Personnel Psychology, 48(3), 587-605.

Bravo, 1. M., & Kravitz, D. A. (1996). Context effects in performance appraisals: Influence of target value, context polarity, and individual differences. Journal of Applied Social Psychology, 26(19),

Cederbloom, D. (1982). The performance appraisal interview: A review, implications, and suggestions. Academy of Management Review, 7, 219-227.

Cederbloom, D. (1989). Peer and supervisor evaluations: An underused promotion method used for law enforcement. Proceedings of the 13^th Annual Meeting of the International Personnel Management Association Assessment Council.

Clarke, D., Rogers, V. & Miklos, S. (1996). Upward appraisal: Does it make a difference? Paper presented at the 11^th annual meeting of the Society for Industrial and Organizational Psychology, St. Louis, MO.

Clarke, D., Rogers, V. & Miklos, S. (1996). Upward appraisal: Does it make a difference? Paper presented at the 11^th annual meeting of the Society for Industrial and Organizational Psychology, St. Louis, MO.

Conway, J. & Huffcutt, A. (1997). Psychometric properties of multi-score performance ratings: A meta-analysis of subordinate, supervisor, peer, and self-ratings. Human Performance, 10(4), 331-360.

Conway, J. M., & Huffcutt, A. I. (1997). psychometric properties of multisource performance ratings: A meta-analysis of subordinate, supervisor, peer, and self-ratings. Human Performance, 10(4), 331-360.

Cooper, W. H. (1981b). Ubiquitous halo. Psychological Bulletin, 90, 218-244.

Fahr, J. & Werbel, J. (1986). Effects of purpose of the appraisal and expectation of validation on self-appraisal leniency. Journal of Applied Psychology, 71, 527-529.

Farh, J., Canella, A., & Bedeian, A. (1991). Peer ratings: The impact of purpose on rating quality and user acceptance. Group and Organization Studies, 16(4), 367-386.

Farh, J., Canella, A., & Bedeian, A. (1991). Peer ratings: The impact of purpose on rating quality and user acceptance. Group and Organization Studies, 16(4), 367-386.

Farh, J., Dobbins, G., & Cheng, B. (1991). Cultural relativity in action: A comparison of self-ratings made by Chinese and US workers. Personnel Psychology, 44(1), 129-147.

Field, H. S., & Holley, W. H. (1982). The relationship of performance appraisal system characteristics to verdicts in selected employment discrimination cases. Academy of Management Journal, 25, 392-406.

Furnham, A. & Stringfield, P. (1994). Congruence of self and subordinate ratings of managerial practices as a correlate of superior evaluation. Journal of Occupational and Organizational Psychology, 67(1), 57-67.

Furnham, A. & Stringfield, P. (1994). Congruence of self and subordinate ratings of managerial practices as a correlate of superior evaluation. Journal of Occupational and Organizational Psychology, 67(1), 57-67.

Giffin, M. E. (1989). Personnel research on testing, selection, and performance appraisal. Public Personnel Management, 18, 127-137.

Gruner, S. (1997) Feedback from everyone: Are 360-degree performance reviews a silly fad – or a smart management tool? Inc., 19(2), 102-103.

Holzbach, R. (1978). Rater bias in performance ratings: Supervisor, self- and peer ratings. Journal of Applied Psychology, 63, 579-588.

Hom, P. W., DeNisi, A. S., Kinicki, A.J., & Bannister, B. D. (1982). Effectiveness of performance feedback from behaviorally anchored rating scales. Journal of Applied Psychology, 67, 568-576.

Kane, J. S., & Lawler, E. E. (1979). Performance appraisal effectiveness: Its assessment and determinants. In B. M. Staw (Ed.), Research in organizational behavior (Vol. 1., pp. 425-478). GreenWich, CT: JAI Press.

Keeping, L. & Sulsky, L. (1996). Examining the quality of self-ratings of performance. Poster presented at the 11^th annual meeting of the Society for Industrial and Organizational Psychology, San Diego, CA.

Kelly, C. M. (1984). Reasonable performance appraisals. Training: Development, 38(1), 79-82.

King, P. (1984). Performance planning and appraisal. New York: McGraw-Hill.

Kozlowski, S., Kirsch, M. P., & Chao, G. T. (1986). Job knowledge, ratee familiarity, conceptual similarity and halo error: An exploration. Journal of Applied Psychology, 71 45-49.

Kraiger, K., & Ford, J. K. (1985). A meta-analysis of ratee race effects. Journal of Applied Psychology, 70, 56-65.

Lazer, R. & Wikstrom, W. (1977). Appraising managerial performance. New York: The Conference Board.

London, M. & Wohlers, A. (1991). Agreement between subordinate and self-ratings in upward feedback. Personnel Psychology, 44, 375-390.

Long, W., Long, E., & Dobbins, G. (1998). Correlates of satisfaction with a peer evaluation system: Investigation of performance levels and individual differences. Journal of Business and Psychology, 12(33), 299-317.

Mabe, P. & West, S. (1982). Validity of self-evaluation of ability: A review and meta-analysis. Journal of Applied Psychology, 67, 280-296.

McEvoy, G. (1990). Public sector managers’ reactions to appraisals by subordinates. Public Personnel Management, 19(2), 201-212.

Meyer, H. (1980). Self-appraisal of job performance. Personnel Psychology, 33, 291-296.

Mumford, M. (1983). Social comparison theory and the evaluation of peer evaluations: A review and some applied implications. Personnel Psychology, 36, 867-881.

Murphy, K. R., & Blazer, W. K. (1986). Systematic distortions in memory-based behavior ratings and performance evaluations: Consequences for rating accuracy. Journal of Applied Psychology, 71, 39-44.

Murphy, K. R., & Constans, J. (1987). Behavioral anchors as a source of bias in rating. Journal of Applied Psychology, 72, 573-577.

Pearce, J. L., & Porter, L. W. (1986). Employee responses to formal performance appraisal feedback. Journal of Applied Psychology, 71, 211-218.

rating scales (BARS) and other rating formats. Personnel Psychology, 34, 263-289.

Rhoads, C. (1997). A year-round schedule said to take sting out of performance reviews. American Banker, 162(28), 6.

Saavedra, R. & Kwun, S. (1993). Peer evaluation in self-managing work groups. Journal of Applied Psychology, 78(3), 450-462.

Smith, P. C., & Kendall, L. M. (1963). Retranslating expectations: An approach to the construction of unambiguous anchors for rating scales. Journal of Applied Psychology, 47, 149-155.

Smither, J. w., Reilly, R. R., & Buda, R. (1988). Effect of prior performance information on ratings of recent performance: Contrast versus assimilation revisited. Journal of Applied Psychology, 73,487-496.

Srinivas, S., & Motowidlo, S. J. (1987). Effects of rater’s stress on the dispersion and favorability of performance ratings. Journal of Applied Psychology, 72, 247-25l.

Walker, A. (1997). Upward feedback: Incremental improvement in managers’ performance over five years. Poster presented at the 12^th Annual Conference of the Society for Industrial and Organizational Psychology, St. Louis, MO.

Werner, J. M., & Bolino, M. C. (1997). Explaining U.S. Courts of Appeals decisions involving performance appraisal: Accuracy, fairness, and validation. Personnel Psychology, 50(1), 1-24.

Whetstone, T. (1994). Subordinates evaluate supervisory and administrative performance. Police Chief, 61(6), 57-62.

Williams, J. & Levy, P. (1992). The effects of perceived system knowledge on the agreement between self-ratings and superior ratings. Personnel Psychology, 45, 835-847.

Witt, L. (1996). Listen up! Your upward feedback results are speaking. Poster presented at the 11^th Annual Conference of the Society for Industrial and Organizational Psychology, San Diego, CA.

Yu, J. & Murphy, K. (1993). Modesty bias in self-ratings of performance: A test of the cultural relating hypothesis. Personnel Psychology, 46(2), 357-363.