Performance Appraisal Rater Errors
Posted on November 14, 2011 by Gail Danneman
As human beings, we are subject to unintentional errors that we are often unaware of. Performance appraisals are a common area where our human tendencies can appear full force. Because we know that rating errors can occur, what do we do about it? Obviously we cannot change people’s unconscious thoughts and ways of thinking, but what we can do is train raters to be aware of his or her own biases. Once raters are aware of biases to avoid, performance appraisals can be much more accurate.
One of the most common rating errors is known as the halo effect. Halo effect occurs when one particular trait about someone causes us to either rate that person very high or very low on performance appraisals. For example, physical appearance often impacts the perception a manager may have of a subordinate. Therefore, because an employee is physically attractive, this individual may receive all high scores throughout the performance appraisal system. Why is this a problem? When a rater gives ratings to an employee because of a trait or characteristic, the rating is not accurate. Just because someone is physically attractive does not mean they deserve high scores across the board. Unfortunately, the halo effect is typically an unconscious judgment. As a manager, I may not even know that I am rating employees differently based on personality or physical appearance. However, if I receive the right training, I am more likely to think about the scores I am giving employees.
A second common error is known as the leniency error. This error is often made in an attempt to avoid conflict. Performance appraisals are an uncomfortable situation for both managers and employees. Managers do not always enjoy giving negative feedback and employees do not like receiving negative feedback. To avoid the awkward situation, some managers will not rate employees accurately. Instead, managers give high ratings to all employees to avoid looking like the bad guy. Although performance appraisal meetings induce anxiety on both the manager and the employee, giving an employee high ratings when they are not deserved does not help employees improve his or her performance. A poor performer that receives high ratings will not change his or her behavior because areas of improvement are not addressed.
Central tendency error occurs when a rater does not give high or low ratings, but tends to stay in the middle of the rating scale. Similar to the leniency error, managers that rate employees in the middle do so to avoid conflict with employees. Rather than rate a poor performer at the lowest spectrum of the scale, many managers feel they are being more fair if they rate the individual in the middle of the scale. Again, employees are not getting a true sense as to how his or her performance is rated.
Lastly, similarity error can have a very negative impact on certain employees. Social psychology tells us that we tend to gravitate toward people that are similar to us or birds of a feather flock together. We like people who are like us. In conducting performance ratings, managers may be giving higher ratings to employees who are similar to them rather than giving an accurate rating. Another way this error can be interpreted is through in-groups and out-groups. In-groups can form based on personality similarities or common interests. Out-groups are those individuals that do not seem to fit into the norm or in-group. A manager may unknowingly rate a member of their in-group higher compared to a member of the out-group.
When thinking about performance appraisals, it is essential to look at the rater. If performance appraisals are to be helpful, they must be accurate. When a manager is subject to one of the rater errors, the rating is not a clear indication of the employee’s performance. Therefore, training raters to consider these rating errors may help to eliminate inaccurate performance appraisals.
Have you seen these errors in practice? How can we combat them?