CHAPTER ELEVEN

"Appraising Employee Performance"

---

LEARNING OBJECTIVES

CHAPTER OUTLINE

• The concepts of performance, performance appraisal and performance appraisal system.
• Why is performance appraisal important?
• The performance appraisal process.
• Methods for appraising performance.
• Biases in appraisal.
• The appraisal feedback interview.
• Performance appraisal and TQM

1. THE CONCEPTS OF PERFORMANCE, PERFORMANCE APPRAISAL AND PERFORMANCE APPRAISAL SYSTEM

After an employee has been selected for a job, has been trained to do it, and has worked on it for a period of time, his or her performance should be reviewed. This chapter is devoted to one of the most significant topics in HRM: performance appraisal. Let us begin by defining our basic terms:

Performance refers to an employee's accomplishment of assigned tasks¹.

Performance appraisal [PA] is the process of evaluating how well employees do their jobs compared with a set of standards and communicating that information to employees². While,

A Performance Appraisal System [PAS] involves the performance appraisal and the entire setting in which the appraisal takes place³.

2. WHY IS PERFORMANCE APPRAISAL IMPORTANT?

Some major organisational purposes are served by performance appraisal systems. Such purposes are often condensed into two general categories: evaluative and developmental⁴.

o The evaluative category includes:

- Performance measurement, which establishes the relative value of an individual's contribution

to the company, and thus provides information that can be used in decisions on layoffs,

terminations and promotions.

- Compensation, which determines appropriate pay for performance and equitable salary and

bonus incentives based on merit or results.

- Motivation, which is the end result of an effective evaluation.

o The developmental category includes:

- Management development, which identifies and prepares individuals for increased

responsibilities and thus provides a framework for future employee development.

• Identification of potential, which identifies candidates for promotion.
• Feedback, which informs employees of the organisation's standards and how their

performance measures up.

- Human resource planning, which audits management talent and evaluates the present supply

of human resources for replacement planning.

- Communications, which provide a format for dialogue between superior and subordinate and

improve understanding of personal goals and concerns. This can also have the effect of

increasing the trust between the rater and ratee.

- Performance improvement, which encourages continued successful performance and

strengthens individual weaknesses to make employees more effective so that organisations

can successfully implement strategies such as quality enhancement.

- Research on legal compliance, which establishes the validity of employment decisions made

on the basis of performance-based information and provides a defense for management

actions, such as selection, demotion or termination.

3. THE PERFORMANCE APPRAISAL PROCESS

In establishing a performance appraisal system, decisions regarding what to measure should be made very early in the development of the system. Other important decisions centre around the timing of appraisals and the choice of appraisers.

o Establishing job criteria and performance standards.

Job analysis helps identify the most important duties and tasks of jobs. The job criteria are elements of a job to be evaluated during performance appraisal and they should be specified⁵. For example, one criterion for a typist's job might be typing speed; another might be typing accuracy.

Each job criterion should be compared with a performance standard, which is the expected level of performance. For the typist's job, the standard might be 50 words per minute and no more than two errors. Multiple job criteria are the rule rather than the exception for all but the simplest jobs. The various criteria for a given job should also be weighted to reflect the relative importance of criteria. For example, in the typist's job speed might be twice as important as accuracy, and accuracy might be equally as important as getting to work on time and being at work every day.

Realistic, measurable, clearly understood performance standards benefit both the organisation and the employee. It is important to establish standards before the work is performed so that all involved will understand the level of accomplishment expected. Standards often are established for the following⁶:

• quantity of output
• quality of output
• timeliness of results
• manner of performance
• effectiveness in use of resources

o Appraisals timing

Timing of appraisals is important. Systematic appraisals typically are conducted once or twice a year⁷. For new employees, an appraisal 90 days after employment, again at six months, and annually thereafter is common timing.

o Choice of appraisers

Performance appraisal can be done by anyone familiar with a person's performance including the following⁸: supervisors, subordinates, peers, customers and self-appraisal.

- Supervisor rating of subordinates.

Rating of employees by supervisors is based on the assumption that the manager is the most qualified person to evaluate the employee's performance realistically, objectively and fairly. A manager's appraisal typically is reviewed by the manager's superior to make sure that the manager has done a proper job of appraisal and, if the review and merit discussion is not separated, that the recommended pay increase is justified.

Appraisal by superiors has certain drawbacks⁹. First, the superior may have an ethical bias against 'playing God' . Second, the superior may not have the necessary interpersonal skills to give good feedback. And, lastly, the superior - having reward and punishment power - may make the employee feel threatened and alienated.

- Employee rating of superiors.

The concept of having superiors rated by subordinates [upward or reverse appraisals] is being used in a number of organisations today. Major firms, including Chrysler, Du Pont and others use subordinates' ratings of how their bosses manage to improve operations, to make their organisations less hierarchical, and to develop better managers¹⁰. While subordinates often do not have access to information about all dimensions of supervisory performance, they do have access to information about supervisor-subordinate interactions. One drawback is that subordinates may not always evaluate performance objectively or honestly - particularly, if subordinates feel threatened.

- Peer ratings

Peer ratings are especially useful when supervisors do not have the opportunity to observe each employee's performance, but other work group members do. Common performance dimensions on which team members have evaluation expertise include¹¹:

• attendance and timeliness: attends scheduled group meetings.
• interpersonal skills: willing to give and take on issues; not unreasonable stubborn.
• group supportiveness: offers ideas/suggestions for the group to use on the project, supports

group decisions.

- planning and coordination: contributes input to assist other team members in performing their

assignments.

Peer evaluations have not been widely used however- with the exception of TQM organisations-because usually team members resist evaluating colleagues since this can damage relationships.

Thus, it may be that peer evaluations are best used for developmental purposes rather than for administrative purposes.

- Appraisal by Customers

Another source of appraisal information comes from customers or clients of the employee. For salespeople and other service jobs, customers may provide the only really clear view of certain behaviours. The information that customers provide can serve as useful input for employment decisions, such as those regarding promotion, transfer, and need for training. It can also be used to assess the impact of training, or as a basis for self-development¹².

- Self Appraisal

Self appraisal works in certain situations. Essentially it is a self-development tool that forces employees to think about their strengths and weaknesses and set goals for improvement. Thus if an employee possesses a unique skill, the employee may be the only one qualified to rate his/her own behaviour. Yet, employees may not rate themselves as supervisors would rate them, using quite different standards.

4. METHODS FOR APPRAISING PERFORMANCE

Performance actually can be appraised by a wide variety of methods and techniques. The most commonly used performance appraisal methods can be distinguished into three major categories: comparative appraisals; behavioural appraisals; and output-based appraisals.

I. Comparative Appraisals.

Comparative appraisals require that managers directly compare the performance of their subordinates against one another. For example, a data-entry operator's performance would be compared with that of other data-entry operators by the computing supervisor. Comparative techniques include ranking, paired comparisons, and forced distribution¹³.

[i]. Ranking. In this method, the supervisor lists all subordinates in order, from the highest to the lowest in performance. Rankings such as this are appropriate only in small companies. As the number of employees increases, it becomes gradually more difficult to discern differences in their performance.

The primary drawback of the ranking method is that the size of differences among individuals is not well defined¹⁴. For example, there may be little difference between individuals ranked second and third, but a big difference in performance between those ranked third and fourth. This drawback can be overcome to some extent by assigning points to indicate the size of the gaps existing among employees.

[ii]. Paired Comparisons. The paired comparison method involves comparing each employee to every other employee in the rating group - one at a time - to determine who is 'better'. A rank order is obtained by counting the number of times each individual is selected as being the better of a pair. Figure 11.2 illustrates the paired comparison method with three individuals.

Fig. 11.2 "Paired comparisons", Mathis and Jackson, p. 337

If the number of employees to be ranked is large, the number of comparisons that have to be made may be unmanageable. The large number of comparisons that must be made is the major drawback of this method¹⁵.

[iii]. Forced Distribution.

In the third comparative method, forced distribution, the supervisor must assign only a certain proportion of his/her subordinates to each of several categories on each evaluative factor¹⁶. A common forced distribution scale is divided into five categories. A fixed percentage of all subordinates in the group fall within each of these categories. Typically, the distribution follows a normal distribution such as the one presented in Figure 11.3.

Fig. 11.3 "Forced distribution on a Bell-Shaped Curve", Mathis and Jackson, p. 338

Thus, for example, if a supervisor had to appraise the performance of 50 subordinates using the forced distribution method, he/she should have to rate: 5 employees [10 percent] as 'unsatisfactory'; 10 employees [20 percent] as 'below average'; 20 subordinates [40 percent] as 'average'; 10 individuals [ 20 percent] as 'good'; and 5 people [ 10 percent] as 'excellent'.

A problem with the forced distribution method is that a group of subordinates may not conform to the fixed percentage17. Another drawback of forced distribution is that a supervisor may resist placing any individual in the lowest, or the highest, category¹⁸. In fact, generally, the distribution of performance appraisal ratings does not approximate the normal distribution of the bell-shaped curve, since it is common for 60% to 70% of the organisation's workforce to be rated in the top two performance levels.

II. Behavioural Appraisals

With comparative appraisals, the supervisor is forced to evaluate each employee relative to other employees. In contrast, behavioural appraisals allow supervisors to evaluate each person's performance independent of other employees but relative to important job-related behaviours, which when exhibited can lead to job success. For example, a salesperson who can exhibit the behaviour of 'verbal persuasion' appropriately has satisfied a behaviour-based criterion.

The simplest methods for appraising job performance using behavioural criteria are those that require a manager to mark an employee's level of performance an a specific form, namely: the graphic rating scale and checklist.

[i]. Graphic Rating Scales.

The graphic rating scale allows the rater to mark an employee's job performance on a continuum - that is, on a five-point or seven-point Likert-type scale . This method identifies certain subjective character traits, such as 'pleasant personality', 'initiative' or 'creativity' to be used as basic job performance criteria. Because of its simplicity, the graphic rating scale is the most frequently used performance appraisal method.

However, there are some obvious drawbacks to the graphic rating scale¹⁹. Apart from the fact that the traits used for performance evaluation may be unrelated to the job itself, such traits often tend to be ambiguous and too vague to be used as the basis for employee performance appraisal. Another drawback is that the descriptive terms used in such scales may have different meanings to different raters. Factors such as 'initiative' and 'cooperation' are subject to many interpretations, especially in conjunction with words such as 'outstanding', 'average', or 'poor'. A sample of a graphic rating scale is presented in Figure 11.1

Fig. 11.1 "A graphic rating scale", Mathis and Jackson, p. 335

[ii]. Checklist.

The checklist uses a list of statements or words that are checked by raters²⁰. Raters check statements most representative of the characteristics and performance of an employee. The following are typical checklist statements²¹:

• ...................... can be expected to finish work in time
• ...................... seldom agrees to work overtime
• ...................... is cooperative and helpful
• ...................... accepts criticism
• ...................... strives for self improvement

The checklist can be modified so that varying weights are assigned to the statements or words. The results can then be quantified. Usually, the results are not known by the rater and are tabulated by someone else, such as a member of the HR unit²².

As with the graphic rating scale, the words or statements used may have different meanings to different raters, thus causing severe practical problems to the effective evaluation of employees.

Apart from checklists and graphic rating scales, some managers are required to provide written appraisal information - in an narrative form. These records describe an employee's actions rather than indicating an actual rating. Among the most common narrative appraisal methods are included: the critical incident method; and the essay.

[iii]. Critical Incidents.

In the critical incident method, the manager keeps a written record of the highly favourable and unfavourable actions in an employee's performance. When something happens [a 'critical incident' involving a particular employee] the manager writes it down. Thus, a list of critical incidents is kept during the entire rating period for each employee. The critical incident method can be used with other methods to document the reasons why an employee was rated in a certain way.

The critical incident method also has its unfavourable aspects²³. First, what constitutes a critical incident is not defined in the same way by all supervisors. Next, producing daily or weekly written reports about each employee's performance may be extremely time-consuming. And lastly, employees may become overly concerned about what the supervisor writes and begin to feel 'threatened'.

[iv]. Essays.

The essay, or 'free-form', appraisal method requires the manager to write a short essay describing each employee's performance during the rating period²⁴. The rater usually is given a few general headings under which to categorise comments. The intent is to allow the rater more flexibility than other methods do. As a result, the method is often combined with other methods.

[v]. Behaviourally Anchored Rating Scales {BARS].

A major breakthrough in utilising critical incidents to evaluate performance is the development of behaviourally anchored rating scales [BARS], which focus appraisal on employee behaviours that can be changed. Thus, a BARS system describes examples of 'good' or 'bad' behaviour. These examples are 'anchored', or measured, against a scale of performance levels. Figure 11.4 shows a BARS that rates a university professor's attitude toward students. What constitutes various levels of performance is clearly defined in the figure. Spelling out the behaviour associated with each level of performance helps minimise some of the problems noted earlier.

Construction of BARS begins with the identification of important job dimensions. The dimensions are the most important performance factors in an employee's description. To continue with the university professor example, assume the major job dimensions associated with teaching are:

• course organisation
• attitude toward students
• fair treatment
• competence in subject area

Short statements, similar to critical incidents, are developed that describe both desirable and undesirable behaviours [anchors]. Then they are 'retranslated' or assigned to one of the job dimensions. This task is usually a group project and assignment to a dimension usually requires the agreement of 60% to 70% of the group. The group, consisting of people familiar with the job, then assigns each 'anchor' a number, which represents how 'good' or 'bad' the behaviour is. When numbered, these anchors are fitted to a scale²⁵. Figure 11.5 shows a flow diagramme of the BARS construction process.

Behaviourally anchored rating scales require extensive time and effort to develop and maintain. Also, separate BARS forms are necessary to accommodate different types of jobs in an organisation.

Fig. 11.5 "Flow Diagram of BARS Construction", Mathis and Jackson, p. 340

III. OUTPUT-BASED APPRAISALS

While the methods described above focus on job behaviours or processes, output-based appraisals focus on job products as the primary criteria. The most commonly used output-based appraisal is Management-by-Objectives {or MBO}.

MBO specifies the performance goals that an individual hopes to attain within an appropriate length of time . The objectives that each manager sets are derived from the overall goals and objectives of the organisation. Implementing an MBO appraisal system comprised four basic stages ²⁶. These stages are shown in Figure 11.6.

Fig. 11.6 "MBO Process", Mathis and Jackson, p. 342

o Job Review and Agreement.

In the first phase the employee and the supervisor review the job description and the key activities that comprise the employee's job. The idea is to agree on the exact makeup of the employee's job.

o Development of Performance Standards.

Specific standards of performance must be mutually developed. This phase specifies a satisfactory level of performance that is specific and measurable.

o Guided Objective Setting.

Objectives are established by the employee in conjunction with, and guided by, the supervisor. Objectives should be realistically attainable and may be different from the set performance standard.

o Continuing Performance Discussions.

The employee and the supervisor use the objectives as bases for continuing discussions about the employee's performance. Although a formal review session may be scheduled, the employee and the manager do not necessarily wait until the appointed time for performance discussion. Objectives are modified mutually, and progress is discussed during the period.

Despite the many advantages of MBO [e.g. measurable performance objectives that clearly define expected outcomes; higher employee productivity; and enhanced employee commitment and satisfaction], no management tool is perfect. Thus, MBO is not appropriate for jobs with little or no flexibility. Therefore, the MBO process seems to be most useful with managerial personnel and employees who have a fairly wide range of flexibility and control over their jobs²⁷.

5. BIASES IN APPRAISALS

There are many possible sources of error in the performance appraisal process²⁸. One of the major sources is mistakes made by the rater. There is no simple way to eliminate these errors, but making raters aware of them is helpful. The most common errors committed in performance appraisal include: the halo effect; leniency; strictness; the central tendency error; the recency effect; and the constrast error.

o The Halo Effect.

The halo effect occurs when a manager rates an employee high or low on all items because of one characteristic. For example, if a worker has few absences, her supervisor might give her a high rating in all other areas of work, including quantity and quality of output, because of her dependability. The manager may not really think about the employee's other characteristics separately. The opposite of a halo error is a horn error, where negative performance in one dimension influences any positive aspects of the employee's performance.

An appraisal that shows the same rating on all characteristics may be evidence of the halo or horn effect. Clearly specifying the categories to be rated, rating all employees on one characteristic at a time, and training raters to recognise the problem are some means of reducing the halo and horn effects.

o The Leniency Error.

A second common and often intended rating error is called leniency error - or the process of being 'too easy'. Leniency bias may exist because supervisors are concerned about damaging a good working relationship by giving an unfavourable rating. Or they may wish to avoid giving negative feedback, which is often unpleasant, so they inflate the ratings.

o The Error of Strictness.

At the opposite extreme of leniency is the error of strictness in which ratees are give unfavourable ratings regardless of performance level. Raters with low self-esteem, or raters who have personally received a low rating are most likely to rate strictly. Rater training, which includes reversal of supervisor subordinate roles and confidence building, will reduce this error.

o The Central Tendency Error.

Rather than using extremes in ratings, there is a tendency on the part of some raters to evaluate all ratees as average even when performance actually varies. This bias is referred to as the error of central tendency. Raters with large spans of control and little opportunity to observe behaviour are likely to rate the majority of employees in the middle of the scale, rather than too high or too low. This is a 'play-it-safe' strategy. Central tendency can also be a by-product of the rating method. The forced-distribution format requires the most employees be rated 'average'.

o The Recency Effect.

As the typical appraisal period [six months to a year] is far too long for any rater to adequately remember all performance-relevant information. As the appraisal interview draws near, the rater searches for information cues as to the value of performance. Unfortunately, recent behaviours or outputs are more salient. As a result, recent events are weighted more heavily than they should be. Called the recency of events error, this bias can have serious consequences for a ratee who performs well for six months or a year but then makes a serious or costly error in the last week or two before evaluations are made.

Employees and managers can minimise this error by keeping ongoing behavioural or critical incident files in which good and poor behaviours and outputs are recorded. Although time consuming, they ensure that information for the entire period is incorporated into the appraisal.

o The Contrast Error.

Rating should be done on the basis of standards that are established before the rating. The contrast error is the tendency to rate people relative to other people rather than to performance standards. For example, if everyone else in a group is doing a mediocre job, a person performing somewhat better may be rated as excellent because of the contrast effect.

But in a group performing well, the same person might have received only a poor rating. Although it may be appropriate to compare people at times, the rating should reflect performance against job requirements, not against other people.

6. THE APPRAISAL FEEDBACK INTERVIEW

Once appraisals have been made, it is important to communicate them so that employees have a clear understanding of how they stand in the eyes of their immediate superiors and the organisation. It is fairly common for organisations to require that managers discuss appraisals with employees. The appraisal feedback interview can clear up misunderstandings on both sides. In this interview, the manager should emphasise counceling and development, not just tell the employee, 'Here is how you rate and why'. Focusing on development gives both parties an opportunity to consider the employee's performance and its potential for improvement.

The appraisal interview presents both an opportunity and a danger²⁹. It is an emotional experience for the manager and the employee because the manager must communicate both praise and constructive criticism. A major concern is how to emphasise the positive aspects of the employee's performance while still discussing ways to make needed improvements. If the interview is handled poorly, the employee may feel resentment and conflict may result, which probably will be reflected in future work.

7. PERFORMANCE APPRAISAL AND TOTAL QUALITY MANAGEMENT

As suggested in Chapter 3, TQM emphasises the continuous improvement of products and processes to ensure long-term customer satisfaction. Its group problem-solving focus encourages employee empowerment by using the job related expertise of the workforce. Cross-functional teams develop solutions to complex problems, often shortening the time taken to design, develop, or produce products and services. Since a team may not include a representative of management, the dividing line between labour and management often becomes blurred in practice, as workers themselves begin to solve organisational problems³⁰.

According to W. E. Deming, one of the 'founding fathers' of TQM, appraisal systems that tie individual performance to salary adjustments should be eliminated³¹. In his view, such systems hinder teamwork, create fear and mistrust, and discourage risk-taking behaviour, thereby stifling innovation. This occurs, Deming argues, because most appraisal systems are based on the faulty assumption that individuals have significant control over their own performance - that is that most individuals can improve if they choose to do so by putting forth the necessary effort.

How can organisations incorporate these basic TQM principles in the design of their job performance appraisal systems? Here follow three suggestions for harmonizing these two processes³²:

o Let customer expectations generate individual or team performance expectations.

Start by identifying customer expectations by product or service. Customers may be internal or external. Then individuals and teams can begin to assess their performance against these expectations. Using this baseline of achievement, individuals, teams, and managers can develop continuous improvement targets. Comparing actual performance against expected performance helps avoid detrimental intrateam competition, because individuals or teams are compared against their own benchmarks, rather than the accomplishments of others.

o Include results expectations that identify actions to meet or exceed those expectations. Individuals, teams and supervisors together consider these customer expectations in conjunction with the business plan and begin to establish priorities for improvement opportunities.

o Include behavioural skills that make a real difference in achieving quality performance and total customer satisfaction. For example, effective customer service requires: 'attention to detail'; 'initiative'; and 'listening skills'. These continuous improvement skills are as important to total quality as are results-oriented targets.

CASE

Joe Miller sat at his desk, looking over the performance appraisal form that he had just completed on Bill Cox, one of his insurance underwriters. Bill was on his way to Joe's office for their annual review session. Joe dreaded these appraisal meetings, even when he did not have to confront employees with negative feedback.

A couple of years before, Essex Insurance Company, which had experienced very rapid growth, decided to implement a formal appraisal system. All supervisors were presented with the new appraisal form, which included five different categories in addition to an overall rating. They were asked to rate employees on each dimension, using a scale from 1 [unacceptable] to 5 [exceptional]. They were also advised to maintain a file on each employee into which they could drop notes on specific incidents of good or poor performance during the year to use as 'documentation' when completing the appraisal form. The supervisors were then told that they could only give a rating of 1 or 5 if they had 'substantial' documentation to back it up.

Joe had never given ratings of 1 or 5 because he was not diligent about recording specific incidents for employee fields; he believed that writing up all of the documentation necessary to justify such a rating was to time-consuming. In Joe's opinion, a couple of employees in his department deserved a 5 rating, but so far none of them had complained about his appraisal.

Bill was one of Joe's 'exceptional' workers. Joe had three or four specific examples of exceptional performance in Bill's file, but looking over the form he could not clearly identify the category in which they belonged. 'Oh, well', Joe said to himself, I'll just give him 3's and 4's. I do not have to justify those and Bill has never complained before'. One of the categories was 'Analysing Work Materials'. Joe had never understood what that meant or whether it was relevant to the job of insurance underwriter. He checked 4 [satisfactory] for Bill, as he did on all the evaluations he did. He understood the meaning of the other categories - Quantity of Work; Quality of Work; Improved Work Methods; and Relationships with Co-workers - although he was confused as to what a 3 or 4 indicated about each category.

Bill knocked on Joe's door and came in. Joe looked up and smiled. 'Hi Bill. Sit down. Let's get through this thing so we can get back to work. O.K.?'

QUESTIONS

1. What problems do you see with the appraisal system Joe is using?

2. What are Bill's likely reactions to being told by Joe that he scored 3s and 4s even though he is

one of Joe's exceptional workers?

3. What suggestions do you have for improving the performance appraisal system?

Click here to send your answers

Source: Schuler, R.S. [1995]. Managing Human Resources. Min/St. Paul : West Publishing.

NOTES