Discuss performance appraisal instruments, give examples of four types of rating scales, and describe three types of rating errors.

Question

What will be an ideal response?

Answer 1

• Most organizations carry out formal evaluations of employees’ job performance, called performance appraisal. Usually, each employee’s supervisor or manager completes a performance appraisal form that requires assigning numerical values to employees’ performance. These performance appraisals qualify legally as tests.

• One in five Fortune 500 companies uses forced ranking as a method of performance appraisal.
o To rank employees, the supervisor must decide who is the “best” employee, the “next best” employee, and so on based on predetermined dimensions or criteria.
o Forced Distribution, requires the supervisor to assign a certain number of employees to each performance category, such as “poor,” “below average,” “average,” “above average,” and “outstanding,” so that the appraisals are distributed in a way that resembles a normal curve. Using this method, no matter how well each person performs, some employees will always be ranked as “outstanding” and some will always be ranked as “poor.”
• For annual or semiannual performance assessment, most organizations prefer to rate employee performance using a scale that specifies job dimensions or job behaviors. The traditional method for assessing employees is to ask supervisors or managers to rate employee performance.

• Graphic rating scales that provide visual indicators to help the manager rate employees. Guided by the numbers or words, called anchors, the rater chooses the category that best represents the employee’s performance on the specified dimension. Any type of rating format can be used with any rating dimension.

• Behaviorally anchored rating scales (BARS) are another rating method that uses on-the-job behaviors as anchors for rating scales that represent job dimensions. Using BARS, the rater chooses the rating category by reading the behaviors and placing a mark on the scale that is most representative of the employee’s performance on that dimension. Research on the BARS method suggests that these scales provide more accurate ratings of employee behavior than do traditionally anchored rating scales. However, they are difficult and time-consuming to develop and require concentrated rater training because of their complexity.

• Rating errors:
o Leniency errors result when raters give all employees better ratings than they deserve.
o Severity errors result when raters give all employees worse ratings than they deserve.
o Central tendency errors result when raters use only the middle of the rating scale and ignore the highest and lowest scale categories.
o A halo effect occurs when raters let their judgment on one dimension influence judgments on other dimensions. For instance, an employee who receives a low rating on “quality of work” may also be rated low on “quantity of work” even though the employee actually meets the performance standards for quantity output.