The ideal performance appraisal is a format, not a form.

Kondrasuk, Jack N.

INTRODUCTION

The latest edition of one of the most prominent human resource management textbooks (Dessler, 2011) points out that "every manager needs some way to appraise employees' performance" (p. 306), that performance appraisal (PA) will be done in each case--whether by the supervising manager or others (e.g. peers), and "few things managers do are fraught with more peril than appraising subordinates; performance" (p. 321). The appraisal of employee job performance is one of the most important, most common, and probably the most disliked human resource management activity. Others have echoed these points.

Thomas and Bretz (1994) state that PA, as typically conducted, "has remained a largely unsatisfactory endeavor" for years even though it is a very important HRM area; "both managers and employees tend to approach appraisal feedback sessions with fear and loathing" (p. 28). Thomas and Bretz state that managers and employees dislike the PA process because neither was involved in developing the forms nor processes, neither's suggestions for changes are solicited nor acted upon, managers don't like to give nor do subordinates like to receive negative messages, negative PA ratings have negative effects on employee careers and perceptions of their managers, and there are no rewards for taking the manager's valuable time to appropriately conduct the PA. Performance appraisal has been said to be "one of six deadly diseases" that keep organizations from performing at their peak (Staff of Employee Recruitment & Retention, 2010). However, Grote (2010) points out that PA has more influence on individual careers and work lives than any other management process. Performance appraisal can both make a business more efficient and help keep employees motivated. Evaluating people at regular intervals, appraisals help firms show where their employees excel, where they can improve, and how well they have followed the goals set by the firm.

PURPOSE OF THE PAPER

What would be an ideal performance appraisal (PA)? From the supervisor's perspective, it would probably be an appraisal that would be accurate and helpful in improving the employee's job performance and making administrative decisions (e. g. pay raises) about the employee. From the employee's perspective it would probably be an appraisal that would fully capture all that the employee has contributed in the job to the employer and enable continued career growth for the employee. From society's view it would probably be an appraisal that fairly assesses the employee's performance and is used justly in the employment situation to make the organization more useful to society. The purpose of this paper is to develop an ideal PA--or come as close as possible to a panacea in this area. To accomplish this we need to define PA and its goals, understand how PA is usually conducted, list the problem areas encountered in typical performance appraisals, and then propose an ideal PA to meet the concerns. The ideal PA system proposed here will seek to rectify the problem areas and achieve the goals of PA. The ideal PA system is meant to be theoretical but with enough details to make it pragmatically useful now. We start with the general background and the specific definition of PA.

BACKGROUND OF PA

Finding that the typical PA is disliked by both the appraiser and the appraise is not unexpected as it started as a negative procedure. The use of performance appraisals became institutionalized as a way of monitoring organizational output during the Industrial Revolution when bureaucratic organizations proliferated (Fandray, 2001). Use of performance appraisals during this epoch was usually linked to reactivity and punishment for poor performance (Kennedy & Dresser, 2001). In other words, the PA mechanism focused on the threat of punishing employees for poor performance as a means for motivating them to achieve higher performance standards. As industrialization continued and bureaucratic organizations proliferated, however, the PA system similarly began to evolve. Kennedy and Dresser (2001) told how "organizations gradually adopted more refined methods for seeking improvement in workplace performance ... eventually championing rewards over punishment, forsaking the stick for the carrot, arguing that performance should not only be appraised but also managed, and devising new and sometimes complex methods to improve performance" (p. 8). Within the last thirty years scholars and professionals alike have vigorously analyzed and critically examined the use and effectiveness of performance appraisals within the organizational context. Unfortunately, however, no consensus exists as to what type of PA system best meets the desired objectives.

Definition of PA

The term "Performance Appraisal" (PA) has been synonymous with performance evaluation, performance review, and other similar terms. PA has, at various times, referred to 1) an instrument or form to assess an employee's job performance, 2) an interview where an employee's job performance is assessed and feedback is given to the employee, 3) a system of setting employee job expectations/employee actual job performance/assessing that performance/feedback to the employee on the performance assessment and how to improve it in the future/setting new goals and expectations for another period, or 4) performance management with job performance appraisal a part of it (Dessler, 2011). More recently a fifth entry has been Integrated Organizational Performance Management with vertical and horizontal loadings and strategic/operating plans and individual goals and metrics as described by McGrath (2010). At the present time, PA typically refers to more of a systems approach as stated in #3 in the preceding. That is the definition of PA that we will use in this study.

PROBLEMS ASSOCIATED WITH CONDUCTING PA

It is much easier to find problems in doing PA than to find solutions or suggestions for improvement. PA systems have been criticized in many areas. It would seem that the present problems could be ascertained by surveying the research and practitioner literature about PA. Such a survey was completed which led to 76 different problems with PA as it is typically conducted. The list of problems seeks to be a representative, comprehensive list of PA problems, not an exhaustive list of all references to those problems (Kondrasuk, 2010).

It still constitutes an overwhelming, confusing list of problems regarding the typical PA system. The 76 problems found in present PA systems can probably be reduced to four categories (Kondrasuk et al, 2008). Those categories would be problems with: 1) the purpose of PA, 2) those involved with PA, 3) what is measured and how, and 4) the system and process of PA. The major complaints within each of these areas should provide a clearer understanding of the hurdles to overcome to develop an ideal PA system.

Problems with the Purpose and Goals of PA

Most authors in the field have indicated that there are two main purposes of a PA system: 1) Administrative and 2) Developmental (Kondrasuk, Crowell, Dillon, Kilzer, & Teeley, 2008). One stated goal of PA is to learn what the employee is/is not doing as well as possible and help the employee to improve her job performance. This is basically a counseling or guidance role that the evaluator plays in this role. The second goal or purpose is to use the PA results to help make administrative decisions such as if and how much to award in pay increases, what training is necessary or helpful to improve employee performance, and other uses such as test validation criteria. This second goal places the evaluator in the role of judge. Roberts (1998) states that the supervisor needs to achieve both goals. However, trying to achieve both goals simultaneously can create conflicts in the evaluator and appraisee. It is very difficult for the supervisor to concurrently be a counselor/guide while trying to be a judge at the same time. An evaluatee is likely to be very open and admit shortcomings to a counselor who could help him but NOT be candid to a judge who may cut his pay raise or reduce his promotional opportunities ... or fire him! Evaluators may feel they are placed in conflicting roles by having to be both a coach and a judge of subordinate performance (Eichel, & Bender, 1984; Grote, 1996).

Inconsistent evaluator perceptions are another issue with the purpose of performance appraisals. Inconsistent perceptions as to the purpose of the performance appraisal can throw the entire performance appraisal system off. If evaluators have different views on the purposes of their specific performance appraisal, the process will be conflicting, as well as what to do with the results. A supervisor who believes that the purpose of the appraisal is to determine which team members need to develop additional skills to better achieve organization goals may conduct the appraisals in a completely different way than a supervisor who believes the purpose of the appraisal is to determine which employees deserve a raise. It gets even more problematic when we add in more participants. For instance, additional problems in this area occur when appraisers and evaluatees both have different, conflicting views of the purpose of the PA.

Problems Involving the Participants in PA

Three categories of participants are usually involved in PA: 1) appraisers, 2) evaluatess, and 3) other users. The evaluator can be a variety of individuals or groups of individuals. Traditionally, the evaluatee's direct supervisor evaluates the individual because s/he is in the best position to observe the behavior and evaluate (Kondrasuk, Riley, & Hua, 1999). However other approaches may also be used so that "the evaluator" could be a panel of managers, an employee's peer, subordinate, customers or any person the appraisee interacts with. Regardless of who does the evaluation, they need the support of the total organization.

Reading through the research conducted on performance appraisals, a major issue that arises time and time again is the integration of the PA within the organization's daily functions and culture. Without the full support of integrating the PA process from the top all the way down, the PA has little chance of being implemented successfully. If the organization lacks commitment to the process of performance appraisals, then evaluators do not take the process seriously enough (Roberts, 1998; Eichel, & Bender, 1984; Grote, 1996). When performance appraisals are used as a mechanism of power, domination, or control over underlings, the individual's growth and the effectiveness of the PA system both deteriorate (Roberts, 1998; Neck, Stewart, & Manz, 1995; Wilson, 1991; Eichel, & Bender, 1984; Grote, 1996).

If those who are involved do not have sufficient skills to conduct PA, the results will be less than ideal. Evaluators are frequently not trained in the PA process nor given the necessary training to perform the PA effectively and consistently (Roberts, 1998; Wilson, 1991; Fletcher, 2001; Vinson, 1996; Neck et al; Gray, 2002; Odiome, 1985; Eichel, & Bender, 1984; Grote, 1996)). One specific example of appraisal ineptitude is seen in the way performer behaviors are vaguely determined or not weighted properly in the process (Fletcher, 2001; Eichel, & Bender, 1984; Grote, 1996). Also, upper management seldom is trained properly in how to use this data for organizational improvement (Roberts, 1998).

Biases

Evaluators are often criticized for being biased in the PA process. Common biases include: Central tendency, leniency, severity, recency effect, primacy/first actions effect, favoritism, halo or horns effect, attributional bias, giving evaluations/ratings to justify prejudged actions (e.g. pay raise), and the Hawthorne Effect. Personal bias is apparent in different ways. Evaluators may give either satisfactory or unsatisfactory appraisals to individuals who do not deserve them (Roberts, 1998; Gray, 2002; Fletcher, 2001). Favoritism and subjectivity play a major role in these undeserved appraisals (Roberts, 1998; Kane & Kane, 1992; Bernardin, Crooke, & Villanova, 2000; Gray, 2002; Crow, 1996; Eichel, & Bender, 1984; Grote, 1996). When an evaluator appraises someone who they like, they may be more apt to give them a superior evaluation than someone they do not like. Leniency may also play a role in unreliable performance appraisals (Roberts, 1998; Bernardin et al., 2000). Instead of directly dealing with evaluatees who may be difficult to reprimand, an evaluator may rate them more leniently in hopes of avoiding the wrath of the difficult persons. Leadership styles, personality, mood characteristics, and personal disposition can cause fluctuations in the effectiveness of performance appraisals (Neck et al., 1995; Villanova, Bernardin, Dahmus, & Sims, 1993; Fletcher, 2001). An evaluator's propensity to allow personal bias into the process of giving performance appraisals will cause problems in the effectiveness of the system.

Evaluatees' Perceptions

Related to biases, perceptions and expectations of appraisees in the PA process may be a significant inhibitor of PA success. Evaluatees may refuse to agree to evaluators' assessments and conclusions because they do not meet the evaluatees' expectations of the PA process (Blau, 1999; Roberts, 1998; Fletcher, 2001; Eichel, & Bender, 1984; Grote, 1996). When evaluatees do not perceive fairness and trust in the process of performance appraisals, they are quick to deny the accuracy of the system (Wilson, 1991; Roberts, 1998; Blau, 1999; Eichel, & Bender, 1984; Grote, 1996). Furthermore, evaluatees may argue with the evaluation if it does not match the results of past satisfactory performance appraisals (Gray, 2002; Neck et al., 1995). It appears that some evaluatees are often reluctant to take the evaluation process seriously (Vinson, 1996; Kondrasuk et al., 1999). In many instances, PA systems do not provide for effective communication. For example, some evaluatees feel they are given inaccurate information on the performance criteria (Roberts, 1998; Gray, 2002; Crow, 1996; Vison, 1996; Fletcher, 2001). Appraisals are made weighting behaviors evaluatees did not know were essential for satisfactory appraisal (Roberts, 1998; Eichel & Bender, 1984; Grote, 1996). At the conclusion of a performance appraisal, many times evaluatees are not given instructions on how to use this feedback to improve future job performance (Vinson, 1996; Neck et. al., 1995; Fletcher, 2001; Eichel, & Bender, 1984; Grote, 1996). Additionally, even if instruction is given for performance improvement, evaluatees may express dissatisfaction with the amount and the type of feedback they receive (Roberts, 1998; Eichel & Bender, 1984; Grote, 1996). It appears that many evaluatees feel uncomfortable because they have little or no opportunity to influence the process of performance appraisal (Fletcher, 2001; Wilson, 1991; Eichel & Bender, 1984; Grote, 1996). In order to develop a working PA system, the lack of subordinate support must be addressed (Gray, 2002; Crow, 1996; Kondrasuk et al., 1999; Eichel & Bender, 1984; Grote, 1996).

Problems Involving What is Measured and How It is Measured

Employee's individual goals must mesh with and support the higher-level goals of the organization for performance appraisals to be effective for an organization,. Many firms use inappropriate assessments because they do not use the correct tools for designing the system. The two basic considerations in designing the actual appraisal tool are 1) what to measure and 2) how to measure it (Dessler, 2011). What to measure refers to the focus for measuring the employee's performance, such as quantity, quality and timeliness of work. It may also be measured in respect to developing one's competencies or achieving one's goals. Assessing employee performance is a very difficult task. While employee performance in some jobs, such as selling shoes in a shoe store, is clearly measurable, assessing performance in many other professions, such as that of a nurse, can be less evident. The process may be measuring a person rather than the person's job performance (Kane & Kane, 1992). Appraisals may have difficulty assessing jobs that have immeasurable outcomes or abstract natures (Wilson, 1991; Roberts, 1998). Another problem with typical PA systems is that they only seem to accurately and reliably measure extreme performances and do not reliably differentiate middle-range performances (Roberts, 1998; Kane & Kane, 1992). Behaviors may not be weighted properly to give an accurate evaluation of an individual's overall performance (Eichel, & Bender, 1984; Grote, 1996). In sum, many PA systems have low reliability (rating errors), which must be taken into consideration when analyzing them for their efficacy (Roberts, 1998; Kane & Kane, 1992).In general, we can measure traits, behaviors, and results.

An employee's job performance can be measured by numerous techniques (Dessler, 2011). Evaluators can use graphic rating scales, forced choice distributions, and ranking (paired comparison, alternation, straight rankings) to measure traits. They can use critical incidents, narratives, BARS, BOS, and electronic monitoring to assess behaviors. They can use MBO to assess results. For instance, Thomas and Bretz (1994) stated that an MBO instrument was by far most common for assessing the job performance of both managers and non-manager exempt personnel. The most common appraisal instrument for non-exempt employees varied from MBO (31%) to graphic rating scale (32%) to "other" (23%). However, how to choose the instrument for a particular situation and how to score a combination of measuring instruments is problematic.

Problems with the Basic System and the Process of PA

When reading through the research conducted on performance appraisals, a major issue that arose numerous times was the integration of the PA within the organization's overarching system of functioning. The PA process must be integrated from the top all the way down the ladder for the PA to be implemented successfully. Some critics maintain that there is typically a lack of resources provided by the organization to adequately institute the process (Roberts, 1998). Furthermore, the assessments made by management are often not comparable across the organization (Roberts, 1998; Bernardin et. al., 2000; Fletcher, 2001) and do not fit into the preexisting job descriptions and developmental and administrative systems established by the organization (Eichel, & Bender, 1984; Grote, 1996; Roberts, 1998; Fletcher, 2001). An organization's goals are not always considered when a PA system is designed (Roberts, 1998). Another apparent challenge for PA to be effective is that the PA components are not given enough time to be completed (Roberts, 1998; Kondrasuk et al, 1999).

Time and Timing of PA

Timing of the events in the PA process has been a big issue that has caused much trouble for implementers. When does one conduct the PA? Yearly? Half yearly? Performance Appraisals take time. There is not always enough time to allow for a full appraisal of an employee even for the commonly-prescribed semi-annual approach. Most firms schedule their appraisals according to either when an employee was hired or at a set date for all employees such as at the end of the year. Rating employees according to the date they were hired allows managers to allow enough time to pass in order to have a productive appraisal. Opposed to that, rating all employees at the same time, sometimes called the focal point method, allows managers to compare employees to each other more easily and make broad changes in the direction of individuals and the entire firm (Grote, 2002). Ideally the PA would be conducted more often than they are in most businesses. In most cases any meeting between a manager and a subordinate is helpful to the firm and can increase productivity, but the fact of the matter is that managers and employees are busy with their work and end up putting administrative tasks like performance appraisals on a list of tasks to do if there is extra time.

Deciding when to bring in new objectives has been another problem with the PA process. An evaluator may believe that the evaluatee should be given new objectives to further advance their progress or working success, but when and where to introduce these objectives is unclear. Are practitioners expected to wait to add recent important tasks until the next cycle starts because the present PA cycle is already established and in motion? Much of what causes the problem here is that the appraisal allows for feedback, but does not specifically give a time and place to create and set new objectives in a fast-changing world. Both the evaluator and evaluatee may deem it necessary to be done at different times and in different ways, which ultimately generates an unsettling problem.

Use of Results

After gathering job performance information, another big problem with PA is deciding what to do with the information gathered from the appraisal. Without proper implementation of results, the appraisal is useless. A major issue with results is that managers may go through the entire process and ignore the results all together (Allen, 1994). Over time, the system will lose whatever credibility it may have had. Another issue with results is deciding how to effectively use the information gathered about the employee. In other words, what will happen to the employee based on the results of the evaluation? Once the evaluation is completed, it is unclear what the managers should do with the results. If the evaluatee receives a good appraisal, should the evaluatee automatically receive a pay increase? If the evaluatee receives a bad appraisal, should they be fired or demoted? Because there is a lack of standardization most managers don't know how to go about implementing the results. So let us look at what authors have recently been suggesting to rectify problems like these just listed.

RECENT RECOMMENDATIONS FOR IMPROVEMENT OF PA

There have been many suggestions stated by academic scholars and industrial practitioners as to what aspects to change, add, or tweak. Grote (2010) recently stated that the ideal PA should consist of a 5-step process: 1) Employee performance planning where the manager meets with each employee for an hour at the beginning of the year. to discuss goals for the year. 2) Employee performance execution where the employee performs his job and seeks to achieve his established goals. At the middle of the year, the manager and subordinate meet to discuss progress toward achieving those goals. 3) Employee performance assessment where the manager fills out the PA form and discusses it with his boss before discussing it with the subordinate. The manager also decides administrative decisions like compensation at this time. 4) Employee PA interview where the manager meets with the appraisee for an hour to discuss the manager's appraisal, the subordinate's self-appraisal, and how to improve. 5) They set a date to reconvene to set next year's goals/start the process over. Grote (2010) and others have stated that individual development/improvement plans should also be employed. Grote adds that there should be more responsibility placed on the employee such as being responsible for seeking coaching and feedback and doing a self-evaluation at the end of the period. Other authors have recommended trying to produce more measurable goals to begin with, give more frequent feedback on performance to the operating employee, reduce biases in appraising employee achievements, and periodically and continually reviewing the effectiveness of the PA system. Some have stated that the goals of performance appraisal should include retaining employees and aligning organizational goals (Mayhew, 2011). There is also unanimity among authors that the ideal PA should be pragmatic, cost-effective, and legal (especially regarding EEO requirements).

Regarding what is measured and how. Thomas and Bretz (1994) suggested that managers and subordinates both be more involved in the development of the general PA system/process as well as the PA forms used. They also suggested that there should be more rewards for appropriately using the PA system. Strive for clear, specific, measurable expectations. Use techniques as free from biases as possible. Use evidence-based techniques that are shown by quality research to be valid and reliable. In fact, some, like Jafari, Bourouni, & Amiri (2009), have even developed mathematical approaches to comparing and selecting the best PA technique to use in given circumstances.

A strong emphasis is being placed on better training of those involved in the PA system. All managers who currently perform performance appraisals, or any who would potentially do so, should be required to participate in PA training. A trained HR specialist, with particular training in the performance appraisal field, should conduct the training programs. The HR specialist can better convey the corporation's overarching culture and values as well as the legal aspects of PA throughout the training. The training can tackle the major aspects of language use, objectivity, legal aspects, psychological concerns, evaluative criteria and listening skills. New employees should also be trained in PA as part of their corporate orientation. Present employees could go through refresher training annually to learn new and improved elements of the PA process as appropriate. This learning should never stop. Nor should we stop attempts to improve the performance appraisal in general.

THE MAIN ELEMENTS OF THE IDEAL PA

To have an "Ideal Performance Appraisal System" it is assumed that many present performance appraisal (PA) system components should be retained in general. The components to retain include the basic process of the PA system: 1) establishing expectations for employee performance, 2) allowing the employee the resources to perform the job, 3) appraising that employee's job performance, 4) feedback and discussion of the appraiser's evaluation of the evaluatee's job performance, and 5) continuing the cycle of steps 1-4. However, there are six additional aspects where important changes could be made to produce a more "ideal" performance appraisal system. Those aspects are: 1) Performing the changes recently recommended by researchers, 2) clarifying the goals and role of performance appraisal, 3) focusing on both results and behaviors, 4) adding an appraisal category, 5) aligning the timing of the PA process, and 6) involving more constituencies in this process.

1. Include the Changes Recently Recommended

As we have seen, recent articles have recommended improvements to conduct PA such as the suggestions mentioned above. Those, as well as some others listed here, should be employed if possible. For instance, regarding the evaluator, it has often been recommended that the appraisers be trained in the process of appraisal. This way the manager can have more motivation and more skills to do a better job of appraising her subordinates' job performances. Many have recommended that there be more training of appraisers on how to use the results. There should be PA manuals (preferably online) available for further review and as-needed information. The system should make sure that users of the PA system are rewarded for properly conducting the PA system and applying the results. Make the PA system clear and relatively easy to perform so appraisers are not overwhelmed and over-extended. The ideal PA system must be practical in the sense that it is easy to use, understand, and administer. It must be useful for making decisions, and it must be cost effective. Make sure that audits of the PA system are conducted--that the system is reviewed frequently to spot problems and to make improvements.

With the greater use of computerization, the PA is more likely to get input from additional sources (e. g. 360-degree feedback) for a more well-rounded and accurate view of the appraisee's job performance. Responses to different achievement levels should be relatively standardized. For example, meeting standards gets a 3 % raise while exceeding standards obtains a 6 % pay increase. Have the performance appraisals flow through the total organization including, and having support from, the top management of the organization. It has been recommended that all PA's be conducted at the same time in the calendar year--not on each employee's anniversary date--to be more consistent in standards used for judgment. Assess the evaluatee's job performance on a more frequent basis--continuously if possible. This could include daily progress reports and/or feedback sessions between supervisor and subordinate. Also have specialists from the human resource (HR) department review the PA results for the same reason and to pick up biases (Staff of Employee Recruitment & Retention, 2010).

2. Clarify and Separate the Goals of Performance Appraisal

As previously stated, the ideal PA system could refer to a specific instrument, the PA interview, a PA system, or performance management. We have chosen to focus on PA as a system as described above. It is very important to realize that the ideal PA system is a concept and not a specific instrument. As Russell and Russell (2011) recently pointed out about performance management (which applies to PA as a system), it is a process and not a single event--a format and not a form.

The purpose of the PA system needs to be clarified and delineated. Many years ago it was asserted that organizations typically try to concurrently achieve two goals in their performance appraisals which produces a conflict and less than ideal results. Organizations seek to use the PA to a) make administrative decisions (such as whether to fire/retain/promote, the level of pay increase, training needs, etc.) and b) improve employee performance (by learning the shortcomings of the employee and seek to help the employee improve in those areas). The first is a judicial process where the latter is a counseling process. The appraiser must act as a judge in the former and a guide/counselor in the latter. It is very difficult to be a judge and counselor at the same time. On the other side of the desk, the appraisee tends to selectively hide potentially damaging information that could hurt his being judged highly but tends to openly state weaknesses that could be rectified when the appraiser is acting as a counselor. So there are conflicts within both the appraiser and the appraisee in a typical PA. The proposed best way to resolve these conflicts is to clearly separate the two goals (administrative and developmental) so that both the appraisee and the appraiser know when each purpose is occurring. It should be clear when the appraiser is evaluating the evaluatee on administrative standards (tied to organization goals) or on developmental goals (tied to what the employee personally wants to achieve in that work setting). This separation also has implications for the training of appraisers; they should be trained in how to be a good counselor as well as a good evaluator.

3. Focus on Both Results and Behaviors

If we assess both objective aspects like results and subjective aspects like attitudes, we get a fuller and clearer picture of the employee's performance. Likewise, if we assess both specific end results and also the process/behaviors that led to the results, we get a fuller picture of the employee's performance. But let's take that another step further. If we go back to the basic goals of the PA, we start with goals of making administrative decisions and improving employee performance. Now let's separate those two goals and tie them into the results and process dimensions. If our goal is to make better administrative decisions (e.g. regarding employee retention and promotion, compensation, training), focus on objective appraisals like performance results. If the goal is to_develop the employee and improve that employee's job performance, focus on the subjective/process elements like behaviors. Administrative decisions, such as employee termination, should and often are legally required to be based on objective data--not subjective opinions. The number of items produced or sold, the revenues or profits obtained, even the number of hours worked to achieve the end result are all objective results and can be defended to the employee or the judge/jury in a court setting. However, when we talk about improving an employee's performance, we tend to take the objective results as givens and focus on what the employee could do differently; the employee must behave differently to achieve different (better) results. Doing the same thing (behaviors) should get the same results (less than perfect performance results). So to improve job performance, the behaviors (and their motivation, attitude, etc.) should be changed. The employee should do something differently/behave better to get better objective results.

4. Add a Situation Appraisal Category

Improving job behaviors may not always be sufficient to increase job performance. We assume that if the employee changes/improves his job performance behaviors, that his job performance results will improve. However, that assumption misses one very important ingredient of job performance--the situation. If the economy deteriorates or the salesperson is suddenly assigned a territory with a dearth of prospects or the engineer's computer breaks down or the store check-out clerk has vastly fewer customers, the job performance will diminish in all cases even if the employee adds job skills and increases her motivation to perform better. Consequently, the ideal PA must measure the situation--the opportunity to perform and the organization's support to perform well.

For an example of how to assess the appraisee's situation, one could look at the instrument to assess the situation in Fiedler's (1977) contingency theory of leadership. His "situation favorableness" assessment instrument is not necessarily a panacea to measure the situation for performance appraisals, but it can be a starting point for developing such an instrument.

5. Alter the PA Process

Carefully delineating the timing and arrangement of the elements in the PA process is very important. The typical PA has been conducted by setting goals with each employee at the beginning of the year, giving the employee time to perform and meet those goals, evaluating those performances relative to the goals/expectations, meeting to discuss the employee's evaluation, and continuing the cycle. This is laborious and also contentious for a number of reasons. The typical conflicting appraiser PA goals of guiding and judging cause the skeptical appraisee to withhold information and resist suggestions or demands by the evaluator for the evaluatee to establish certain goals. Since the supervisor, at whatever level, tends to meet with each subordinate one at a time and each session averages about an hour, it is very time consuming. Considering that the supervisor has other responsibilities to attend to such as producing products or services for customers, PA interviews can be drawn out over lengthy periods of time. Likewise at the end of the cycle when the supervisor sits down with the evaluatee to discuss the annual or semi-annual appraisal of the employee's performance, that also takes a great deal of time and can have even more subordinate resistance whenever the employee perceives himself to be judged for administrative decisions such as pay raises or promotions. After all, the average employee believes he is above average--or at least 75% of the employees believe they are above average. So how do we deal with these challenges?

To alleviate some of these aforementioned problems and at the heart of the new, ideal PA, it is recommended that the PA be split into two parts with 1) quick decisions being made regarding administrative decisions at the front end in setting the standards for the position for the year (or other time period) and at the end where the PA is done based on achieving the standards based on objective (measurable) results) evaluations. This will reduce the "limbo" time between ending one period and starting another as well as increase the consistency of evaluations across appraisees. 2) Then, in a second series of meetings, take more time to establish the individualized developmental goals at the beginning and use subjective/behavioral criteria to measure individual goal achievement at the end of the period. The discussions of extent of goal achievement here are to help the subordinate look at how to meet his personal goals for improving his work performance to achieve what he wants to on the job (e.g. be rated "excellent," to be promoted, or just to get by with an adequate work output until retirement). The supervisor does not have to act as a vengeful judge because all administrative decisions have already been made. The supervisor can focus on helping the self-motivated subordinate achieve her personal goals--what she wants to accomplish--and increase the job performance of that employee. Focusing on employee goals encourages more involvement and engagement of the employee--thus increasing employee motivation to do the work on the job.

6. Involve More Constituencies in the Process

The more people that are involved in the PA process, the more chances for better ideas and fewer mistakes--to a point. If more sources can make suggestions to improve the job standards and goals, they should be "better" standards and goals--more accurate, more challenging, more measurable, etc. Therefore, besides the supervisor and the subordinate(s) developing the goals, reviewing the goals by specialists in the organization's human resource department (experts who knows the qualities required to write good objectives and who have a system-wide view of what needs to be done and what others are doing) should result in "better" employee goals. Also, at the other (performance appraisal) end, having input from others (basically a 360-degree appraisal) should give a more complete picture of the actual achievements in comparison with the expectations of what to accomplish (the goals, standards). Getting evaluations toward the goals from the supervisor and self-evaluations from the subordinate are commonplace. Getting evaluations from interacting and knowledgeable peers and the subordinate's subordinates are atypical but recommended as a worthwhile addition here. It is also recommended here to get assessments from the subordinate's customers.

Even with all of these recommendations for changes to improve the PA, it is still questionable if it will work in all situations for all people. Can the same, ideal PA be applied the same way in all situations? Could there still be problems with different types of appraisers, evaluatees, jobs and levels of jobs, companies, sizes (large versus small companies), types (public and private; local, regional, national, international), industries, geographical locations, cultures, and/or countries? Many questions still remain unanswered. More will be known about that after these recommendations are applied and further research is conducted.

CONCLUSION

The ideal PA system is a concept or format, not a specific form. It is a process that involves setting expectations (of the supervisor and subordinate), having the subordinate perform to achieve the expectations, of appraising and feeding back the results, and applying the results of the assessment in ways that benefit the organization, the supervisor, and the subordinate involved. Remember that the ideal PA system has two separate purposes (administrative and developmental)--which must be separated and not attempted to be achieved simultaneously. It appraises both standards applied to many as well as goals applied uniquely to each individual. Administrative decisions, based on standards and objective results, should be made first and quickly; developmental aspects, based on individual goals, are made later and may take more time. Both may assess objective and subjective aspects of the employee's job performance. The appraisal considers the appraisee's accomplishments within the context of a changing job situation to accurately judge the appraisee's job performance. The process and techniques applied are based on evidence-based management that applies valid and reliable approaches. Implementation of the ideal performance appraisal may not be feasible, or possible, for all organizations. But for those who can and do use the PA system proposed herein, it should be a definite improvement.

REFERENCES

Allen, Peter. (1994). Designing and Implementing an Effective Performance Appraisal System. Review of Business, 16(2), pp. 3-8.

Bernardin, H. J. (2003). Human Resource Management. Upper Saddle River, NJ: Prentice Hall.

Bernardin, H. J., Cooke, D. K. & Villanova, P. (2000). Conscientiousness and agreeableness as predictors of rating leniency. Journal of Applied Psychology, 85(2), 232-234.

Blau, G. (1999). Testing the longitudinal impact of work variables and performance appraisal satisfaction on subsequent overall job satisfaction. Human Relations, 52(8), 1099-1113.

Crow, R. (1996). You Cannot Improve My Performance By Measuring It! Journal for Quality and Participation, 91(1), 62-65.

Dessler, Gary. (2011). .Human Resorce Management. 12th ed., Upper Saddle River: Pearson Education Inc.

Eichel, E., & Bender, H. E. (1984). Performance appraisal: A study of current techniques. An AMA research study (pp. 22-57). New York: Research and Information Service, American Management Association. Fandray, D. (2001). The new thinking in performance appraisals. Workforce, 80(5), 36-40.

Fiedler, F. E., Chemers, M. M., and Mahar, L. (1977). Improving Leadership Effectiveness: The Leader-Match Concept. New York: Wiley.

Fletcher, C. (2001). Performance appraisal and management: The developing research agenda. Journal of Occupational and Organizational Psychology, 74, 473-487.

Gold, Jeremy. (2010). Personal interview about General Mills performance appraisal approach by e-mail. March 5. Gray, G. (2002). Performance appraisals don't work. Industrial Management, 44(2), 15-18.

Grote, Dick (2010). Employee performance appraisal--An ideal system. Buzzle.com, Retrieved 7 April 2000 at http://www.buzzle.com/articles/employee-performance-appraisal-ideal- system.html. Grote, D. (2002). Performance appraisal. Executive Excellence, 19(12), 12-13.

Grote, Dick, and Grote, Richard C. (1996). The Complete Guide to Performance Appraisal. New York: AMACOM Div American Management Association.

Jafari, Mostafa, Bourouni, Atieh, and Amiri, Roozbeh Hesam. (2009). A new framework for selection of the best performance appraisal method. European Journal of Social Sciences, 7 (3), 92-100.

Kane, J. S. & Kane, K. F. (1992). The analytic framework: The most promising approach for the advancement of performance appraisal. Human Resource Management Review, 2(1), 37-70.

Kennedy, P. W. & Dresser, S. G. (2001). Appraising and paying for performance: Another look at an age-old problem. Employee Benefits Journal, 26(4), 8-15.

Kondrasuk, Jack N. (2010). So what would an ideal performance appraisal look like? Proceedings of the National Business and Economics Society International Conference. Kauai, HI, March.

Kondrasuk, J. N.; Crowell, Emi; Dillon, Kelly; Kilzer, Steven; & Teeley, Jarad (2008). Appraising performance appraisal: The problems. Proceedings of the 16th International Conference 2008 of the Association on Employment Practices and Principles, Chicago, IL.

Kondrasuk, J., Riley, M. & Hau, W. (1999). If we want to pay for performance, how do we judge performance? Journal of Compensation and Benefits, 15(2), 35-40.

Mayhew, Ruth (2011). Objectives of an annual performance review. EHow, Retrieved February 9, 2011, from http://www.ehow.com/info_7756080_objectives-annual- performance-review.html

McGrath, Chris (2010). Integrated organizational performance management. The Industrial-Organizational Psychologist. 47 (3), 17-19.

Neck, C. P., Stewart, G. L., & Manz, C. C. (1995). Thought self-leadership as a framework for enhancing the performance of performance appraisers. Journal of Applied Behavioral Science, 31(3), 278-302.

Odiorne, G. S. (1985). Performance appraisal: What three companies are doing. Chicago, IL: Commerce Clearing House, Inc.

Roberts, G. E. (1998). Perspectives on enduring and emerging issues in performance appraisal. Public Personnel Management, 27(3), 301-319.

Rupp, Holly (2010). Personal interview by telephone regarding Umpqua Bank's performance appraisal system. February 19.

Russell, J., and Russell, L. (2011). The next level of performance management. T + D, 64 (4), 42-48.

Staff of Employee Recruitment & Retention (January, 2010). Performance appraisals: Improve the process and avoid the problems. Employee Recruitment & Retention, Retrieved April 12, 2010 from http://www.managementresources.com/ME2/Sites/dirmod.asp?sid=&nm=&type= Publishing&mod=Publications::Article&mid= 8F3A7027421841978F18BE895F87F791&SiteID=47341 8383454426DA53DD03044F159BD&tier=4&id=5C3C747F192D44528E2412F439748CA7,

Thomas, S. L., and Bretz, R.D., Jr. (1994). Research and practice in performance appraisal: Evaluating employee performance in America's largest companies. SAM Advanced Management Journal, Spring, 59(2), 28-34.

Villanova, R., Bernardin, H. J., Dahmus, S. A., & Sims, R. L. (1993). Rater leniency and performance appraisal discomfort. Educational and Psychological Measurement, 53(3), 789-799.

Vinson, M. N. (1996). The pros and cons of 360-degree feedback: Making it work. Training and Development, 50(4), pp. 11-12.

Wilson, J. (1991). Performance appraisal for non-management staff. Successfully developing and implementing a system. Industrial and Commercial Training, 23(4), 28-31.

Jack N. Kondrasuk, University of Portland