In spite of this fact, however, there are some elements which are common to all effective performance appraisal systems, regardless of the actual method(s) used in the system. These elements will be discussed shortly. However, before examining these common links, a brief overview of performance appraisal as it is currently practised in American organisations is in order. Current Trends in Performance Appraisal As previously noted, controversy over the “best” performance appraisal system continues.
The dilemma was highlighted in the 19 May 1980 issue of Business Week where the editors concluded that managers want a system “that will pinpoint specific marginal behaviour that should be reinforced or discontinued, serve as a personnel development tool, provide a realistic assess ment of an employee’s potential for advancement, and — a particularly hot issue in the 1980s — stand up in court as a valid defence in discrimination suits. ” Has the search for a “best” system affected what companies actually do in performance appraisal? A study conducted by Taylor and Zawacki in 1981 set out to answer this question y sending a mail questionnaire to 200 firms located throughout the United States — these companies were selected at random from the Fortune 1000. Eighty-four (42 per cent) were returned and used in the study. The size of respondent firms ranged from less than 1,000 employees (nine), 1,000-5,000 employees (63), and more than 5,000 employees (12). Non-respondent firms did not vary significantly in terms of size. This study, which duplicated a previous one conducted in 1976, asked what kind of performance appraisal system was used for management and blue-collar employees.
It also asked for the interval between ratings, productivity and employee reaction to the appraisal system, anticipated changes and respondent satisfaction to the present system. While it is not possible to go into all the detailed findings of this study, some of the most pertinent information is summarised below. ? While in 1976 43 per cent of the respondent firms had used a traditional performance appraisal system (e. g. , forced distribution) and 57 per cent had used a collaborative system (e. g. , MBO), in 1981 these figures had changed to 53 per cent and 47 per cent respectively.
In other words, the proportion of companies using a traditional approach to performance appraisal had increased while the proportion of those using a collaborative approach had decreased. Several respondents provided written comments stating that they had changed to quantitative (i. e. traditional) systems in recent years in reaction to legal challenges to their previous collaborative system. In 1981, 39 of the 41 organisations using a traditional system used a graphic rating scale. Of the collaborative forms, 23 firms used MBO and 11 used a BARS system.
The percentage of firms not satisfied with their current appraisal system increased from only nine per cent in 1976 to 47 per cent in 1981. In addition, those with collaborative systems were more likely to be satisfied, while the majority of firms with traditional systems expressed dissatisfaction. As far as the effect of the type of system used on employee attitudes went, 37 per cent of the ? IMDS January/February 1988 13 ? companies using a traditional approach felt that it had improved employee attitudes while 63 per cent felt it had not.
Of those companies using a collaborative approach, 77 per cent felt it had improved employee attitudes and 23 per cent felt it had not. ? Of the 22 firms indicating that they anticipated changing their performance appraisal system in the near future, 12 were moving from a collaborative system to a traditional system. This is especially interesting in light of the fact that, in the 1976 study, the majority of firms indicating that they were considering a change said that the move would be from a traditional to a collaborative approach.
While the 1981 study did not delve into the reasons behind this shift in attitude, Taylor and Zawacki conjectured that it was due to governmental and legal pressures for precise (i. e. , quantitative) measures which overwhelmed a desire to help people develop and grow towards becoming more effective employees. Of the firms surveyed, 49 per cent felt that their performance appraisal system had improved employee performance (roughly the same proportion found in 1976).
However, the number of firms that did not believe employee performance had improved as a result of the appraisal process had gone from four per cent in 1976 to 19 per cent in 1981 — and none of these firms anticipated changing their system! (5) The appraiser should be given feedback regarding his/her effectiveness in the performance appraisal process. (6) The performance appraisal system, regardless of the methodology employed, must comply with legal requirements (notably, Equal Employment Opportunities guidelines).
Since the factors listed above are consistently highlighted in the literature as essential elements of an effective performance appraisal system, each of them warrants individual attention. Performance Goals Must Be Clearly and Specifically Defined Special emphasis should be placed on this phase of performance appraisal, since the lack of specifically defined performance goals will undoubtedly undermine the effectiveness of the entire performance appraisal process. The key performance areas need to be identified, assigned priorities and stated in quantifiable terms whenever possible.
The mutual goal-setting process between a manager and subordinate associated with Management by Objectives is a particularly beneficial way to foster acceptance and internal motivation on the part of the employee. As is often the case, if multiple goals are established, they should be ranked so that the employee has a clear understanding of which areas may warrant more attention and resources than others. Furthermore, every attempt should be made to describe performance goals in terms of their time, quality, quantity, and monetary dimensions.
This will reduce the opportunity for misinterpretation about what is to be accomplished and what limitations there are. The quantification of goals will also make it easier for the manager and the employee to measure the employee’s progress towards achieving the objectives. The need for quantifying objectives is succinctly summed up by George Ordione: “If you can’t count it, measure it, or describe it, you probably don’t know what you want and can often forget it as a goal. There is still too much, ‘do your best’, or ‘I’ll let you know when it’s right’, going around in today’s organisations.
If you can’t define the desired type and level of performance in detail, then you have no right to expect your subordinate to achieve it. “ ? To summarise, it would appear that while most firms wish to use a collaborative form of performance appraisal, they feel thwarted by outside forces (notably Equal Employment Opportunities requirements) in their attempts to implement such a system within their organisations. The dilemma, then, is finding a workable solution which will meet both constraints. The remainder of this article will take a look at these two seemingly conflicting areas (effectiveness vs. efensiveness) and how they can be integrated into a meaningful performance appraisal system. Elements of an Effective Performance Appraisal System While various authors use different names and modified descriptions for them, the following factors seem to be universally accepted by most authorities on the subjects as requisites for an effective performance appraisal system: (1) Performance goals must be specifically and clearly defined. (2) Attention must be paid to identifying, in specific and measurable terms, what constitutes the varying levels of performance. 3) To be effective, performance appraisal programmes should tie personal rewards to organisational performance. (4) The supervisor and employee should jointly identify ways to improve the employee’s performance, and then establish a development plan to help the employee achieve his/her goals. The Varying Levels of Performance While setting performance goals is a crucial first step in the process, managers also need to concentrate more attention on identifying what constitutes the varying levels of performance.
If the organisation uses the typical “poor, fair, good, very good and excellent” scale of performance, the manager has a responsibility to identify at the beginning what levels of performance will produce a “very good” or “excellent” rating. However, setting specific goals for organisational performance is not enough — managers also need to relate performance to the individual’s rewards. Agreeing on what is to be accomplished and what varying levels of performance represent in terms of evaluation and rewards is crucial for the performance appraisal process to be effective.
Since the first two steps of this process (i. e. , defining performance goals and setting performance standards) IMDS January/February 1988 14 are closely connected, an example of how these steps might be achieved is warranted. A prerequisite for setting performance goals is to establish job tasks. To measure performance realistically, objectively and productively, we must base our reviews on job content rather that job constructs. Constructs are broad, often self-evident terms which describe a general task, activity or requirement. Richards refers to them as “garbage words” in terms of their usefulness as performance standards). An example might be “communication skills”. While few would argue the need for skills in communication for many employees, the problem is how to define the term in light of the requirements of the specific job in question. Will the employee be required to: ? ? ? ? ? ? ? ? ? ? ? ? Write memos? Write letters? Conduct interviews? Deliver public speeches? Present proposals to clients? Describe features and benefits of a product? Resolve face-to-face conflicts?
Handle customer complaints? Write job descriptions? Describe and define job standards? Manage meetings? Present ideas to top management? Initiative: Resourceful in taking necessary or appropriate action on own responsibility. Unsatisfactory Poor A routine Often waits unnecessarily worker; usually for direction. waits to be told what to do, requiring constant direction. Satisfactory Good Excellent Seeks and gets added tasks for self; highly selfreliant. Assumes responsibility. Does regular Resourceful; work without alert to waiting for opportunities directions. or Follows improvement directions with of work. little follow-up Volunteers suggestions. Table I. drinks per bottle, etc. In turn, these indicators should be broken down into measurable standards, as shown in Table II. As shown, when identifying what constitutes the varying levels of performance, we need to decide what we can expect in terms of outstanding performance, what is satisfactory and what is the minimum level of performance we can tolerate. One could argue that these are subjective determinations, and this is of course true.
What is important, however, is that once these determinations have been made, performance can be measured objectively against the standard. It is important to keep in mind that standards should be set based on what we require or need in the performance of a job and not on our assessment of a specific individual’s ability to do the job. Unless we specify the behaviour we want in the context of job content requirements, it will be near impossible objectively to measure someone’s performance under the generic construct of “communication”.
We must determine the sort of communicating the job requires of the employee. Some organisations attempt to aid supervisors by providing rating scales which are anchored to descriptions of performance (i. e. , the BARS approach), such as the one shown in Table I. While this type of scale is certainly a vast improvement over those that offer no anchors (rating descriptions) at all, we could still argue over the ratings. The standards are subjective and unmeasurable, both undesirable traits in any performance appraisal system.
To overcome these problems, the job should be broken down into responsibilities, with a series of performance indicators provided for each responsibility. In turn, these indicators should be accompanied by objective and measurable performance standards. An example will help illustrate the process. A bartender’s job can be broken down into several responsibilities, including mixing drinks, cost control, inventory control, house keeping, safety, law enforcement, supervision, customer relations, etc. In turn, each of these responsibility areas can be broken down into several performance indicators.
For example, performance indicators of the job responsibility “mixing drinks” might include complaints, returns, brands used, appearance, speed, number of Personal Rewards and Organisational Performance To be truly effective, performance appraisal programmes should tie personal rewards to organisational performance. Too many reward systems are based on time on the job, are divided evenly among employees, or offer too little incentive to increase motivation significantly. As noted by Harper, performance appraisal systems need to be designed with the three “E’s” of motivation in mind.
The first ” E ” refers to the exchange theory, which states that people tend to contribute to the organisation’s objectives as long as they believe they will be rewarded. The second ” E ” refers to the equity theory, which states that motivation is tied to the relative, rather than the absolute, size of the reward. For example, if person A does 25 per cent better than person B, but gets only five per cent more in a “merit” increase, then person A is likely to feel that management has actually punished him or her for doing noticeably better than person B.
The third ” E ” is the expectancy theory of motivation, which asserts that motivation is a combination of the person’s perceived probability (expectancy) of receiving a reward and the worth of the reward. Even when the reward is great, motivation may in fact be quite low if the employee does not believe that he or she has a reasonable chance of achieving the necessary level of performance to get the reward. Conversely, if the employee believes that the probability of receiving the reward is high, there will be little motivation if he or she does not need or value the reward. IMDS January/February 1988 15
Job: Bartender Job responsibilities Mix drinks, etc. Indicators Complaints Returns Measurements used (recipe) Brands used Appearance Time No. of drinks per bottle, etc. feedback to managers about the quality of their performance appraisal ratings would seem to have several advantages: ? ? It is relatively inexpensive and easy to develop and implement. The feedback is based on ratings made by each manager as part of the formal performance appraisal process. This enables the feedback to be tailored to the individual. The feedback can provide managers with a basis upon which to compare their ratings with those made by other managers.
This normative type of feedback is rarely available to managers; as a result, there is very little information upon which they can evaluate how lenient or strict they are. A feedback system should help to ensure comparability of ratings among managers, which in turn may increase employee satisfaction with the appraisal process. That is, employees are more likely to perceive that their performance has been evaluated equitably since managers are using the same standards when evaluating performance. ? Job: Bartender Standards Job responsibilities Mix drinks Indicators Minimum Complaints 4/week Satisfactory 2/week Outstanding 0 ?
Table II. In summary, then, for a performance appraisal programme to be successful in this area, it must: (1) Tie rewards to performance (2) Offer a high enough level of reward (3) Have the level of reward reflect the relative differences in the various levels of performance (4) Tailor the rewards to the needs and desires of individual employees. Development Plans Ideally, the performance appraisal programme should be comprised of two separate sessions between the manager and the employee. In the first session the manager and employee review the level of performance from the previous period — what went well, what did not, and why.
This session also identifies the employee’s strengths as well as the areas that need to be improved. The manager then encourages the employee to prepare a development plan to be discussed at the second meeting. The development plan is intended to identify areas that should be improved upon during the coming period. The subordinate should be encouraged to: (1) Concentrate on those areas that will affect results (2) Select three or four particular areas for improvement rather than an unrealistic and unmanageable number (3) Set improvement goals that are specific and measurable.
Whatever the end result happens to be, the employee needs to be the principal author (although the manager should offer help and suggestions) since people tend to be more motivated to accept and implement a plan of their own making. IMDS January/February 1988 16 Indications of the usefulness of such a feedback system were documented in a study by Davis and Mount in which managers were provided feedback vis a vis the ratings they gave to employees.
In response to a questionnaire distributed one week after they had received feedback regarding the quality of their performance ratings, 79 per cent of the managers indicated they were either satisfied (seven per cent) or very satisfied (72 per cent) with the feedback; 93 per cent said they considered it when making subsequent performance evaluations; 70 per cent said it influenced their ratings either appreciably (47 per cent) or substantially (23 per cent), and 79 per cent said the feedback had utility for making managers’ ratings more comparable.
The test results from this study indicated that the feedback also significantly reduced the presence of leniency error (the tendency to skew the rating distribution towards the higher rating categories) in the managers’ ratings. This is significant from an organisational perspective because of the multiple uses of performance ratings in organisations. Often, performance ratings are the criterion on which selection tests are validated and often provide the basis on which merit pay increases are determined.
According to Davis and Mount, improving the psychometric quality of the ratings may enable the tests to be validated more effectively and provide a more equitable method for distributing pay increases — an important consideration, as previously discussed. Conforming to Guidelines Obviously, in addition to the other factors which have already been discussed, another practical consideration which must be taken into account is that any performance appraisal system, regardless of the methods employed, must comply with all Equal Employment Opportunity guidelines.
While a complete discussion of this important area is beyond the scope Feedback Regarding Effectiveness It is surprising how infrequently organisations provide their managers with information about their performance appraisal ratings. However, providing of this article, the Uniform Guidelines on Employee Selection Procedures, put together by the Equal Employment Opportunity Commission (EEOC) and several other agencies in 1978, deserve special mention.
These procedures were meant to clarify the exact requirements which appraisal and other selection systems must meet, and include the following points: (1) To continue using an appraisal system that has adversely affected one or more protected groups, the company must demonstrate that the system is “valid”, that it is job related, and that it accurately measures significant aspects of job performance. (2) The company must establish that there is no other available method of achieving the same necessary business purpose that would be less discriminatory in its effects, and none can be developed.
According to the courts, the plaintiff (employee), rather than the defendant (company) must show the availability of the alternatives. The EEOC has told employers what they cannot do, but it has not provided them with definitive guidelines for solving the performance appraisal puzzle. However, some help in this regard was provided in the Autumn, 1980 issue of EEO Today. (1) Base your appraisal on a comprehensive job analysis. EEOC guidelines dictate that you measure job performance against specific, clearly defined standards of performance.
The performance you appraise, says the EEOC, “must represent major critical work behaviours as revealed by a careful job analysis. ” Without a clear, written statement of job responsibilities, you increase your risk of EEO liability. (7) Submit the appraisal to several reviewers, especially if it is negative. To prevent conscious or unconscious bias from creeping into the appraisal process, develop a multilevel review system. Have your superior review and sign the appraisal. This system of checks and balances will reduce the risk of losing a court action. Final Comment
As can be seen from the foregoing discussion, an effective performance appraisal system involves much more than a mere annual or biennial evaluation of an employee’s past performance. Nonetheless, astute managers are becoming increasingly aware of the value of their human resources, viewing them as an investment rather than merely an expense or overhead to be minimised. Accordingly, many organisations are taking the time and effort necessary to develop an effective performance appraisal system in order to help their people achieve their personal goals, which in turn allows the organisation to meet its own objectives.
Unfortunately, many managers still object that they just do not have the time to make performance review and development an ongoing process. However, if management is defined as “the ability to get things done through people”, and if we accept the fact that an effective performance evaluation process helps in getting the most important and productive things accomplished, then what else should managers spend their time doing? References 1. Fletcher, C. , “What’s New in Performance Appraisal? “, Personnel Management, February 1984, pp. 20-2. 2. Taylor, R. L. and Zawacki, R. A. “Trends in Performance Appraisal: Guidelines for Managers”, Personnel Administrator, March 1984, pp. 71-80. (2) Know the details of your company’s 3. Harper, S. C. , “A Development Approach to Performance nondiscriminatory policies. You and every other Appraisal”, Business Horizons, September-October 1983, pp. manager in the company should aim for the 68-74. uniform application of all appraisal guidelines. 4. Mellenhoff, “How to Measure Work by Professionals”, Management Review, November 1977, pp. 39-43. (3) Avoid subjective criteria. According to the Albemarle Paper Co. v.
Moody decision, subjective 5. Richards, R. C. , “How to Design an Objective PerformanceEvaluation System”, Training, March 1984, pp. 38-43. supervisory appraisals of job performance are 6. Kellogg, M. S. , What to do About Performance Appraisal, inherently suspect if they produce adverse impact American Management Association, New York, 1975. against a protected group. To stand up to the 7. Davis, B. L. and Mount, M. K. , “Design and Use of a scrutiny of the courts, these judgements must Performance Appraisal Feedback System”, Personnel be considered fair and job-related. Administrator, March 1984, pp. 1-7. 8. Block, J. R. , Performance Appraisal on the Job: Making it (4) Document! Keep records. That is the only way Work, Prentice-Hall, Inc. , Englewood Cliffs, New Jersey, 1981. you can support whatever subjective judge 9. Butler, R. J. and Yorks, L. , “A New Appraisal System as ments creep into the appraisal process. (They Organizational Change: GE’s Task Force Approach”, are inevitable. ) Personnel, January-February 1984, pp. 31-42. (5) Aim for a group of appraisers who have common demographic characteristics with the group being appraised. This criterion was established in Rowe v.
General Motors. When only white males appraise blacks, Hispanics, women and other protected groups, the courts question the fairness of the. system. Once a system is challenged and shown to have adverse impact, the company must prove its validity. (6) Never directly or indirectly imply that race, colour, religion, sex, age, national origin, handicap, or veteran status was a factor in your appraisal decision. Making any disciminatory statement, orally or in writing, will make your organisation subject to court action. Additional Reading Kaye, B. L. and Krantz, S. , “Preparing Employees: The
Missing Link in Performance Appraisal Training”, Personnel, May-June 1982, pp. 23-9. “Performance Appraisal: Curre. ” Practices and Techniques”, Personnel, May-June 1984, pp. 5799. Heneman, R. L. and Wexley, K. W. , “The Effects of Time Delay in Rating and Amount of Information Observed on Performance Rating Accuracy”, Academy of Management Journal, December 1983, pp. 677-86. “The Trouble with Performance Appraisal”, Training, April 1984, pp. 91-2. Gehrman, D B. , “Beyond Today’s Compensation and Performance Appraisal Systems”, Personnel Administrator, March 1984, pp. 21-33. IMDS January/February 1988 17