Rating Errors in Performance AppraisalSunday, Sep 23, 2007
Performance appraisal is a critical HR process yet often yields unnerving experience for both employees and their supervisors. Though several systematic methods and approaches are available to assess employee performance, the element of subjectivity cannot be completely eliminated from the process. Human judgment is far from perfect and this weakness is an important factor behind the controversies associated with performance appraisals.
Human performance is largely a qualitative phenomenon and cannot be precisely measured quantitatively. Though quantifying performance elements can make the overall process more objective, but not exact or perfect. After all, human behaviour is a complex phenomenon - far more perplexing than the measurement problems in Physics - weight and length etc. High precision in performance measurement for most of the jobs is a mere illusion.
Apart from the inexactness of measurement, a few cognitive and perceptive problems often cause the raters to make significant errors in judgment. Being aware of these pitfalls can make the process more objective and could also reduce the unpleasantness that is caused as a consequence of a poorly administered performance appraisal. These common rating errors are summarized below:
Halo effect occurs when a rater attaches too much significance to a single factor of performance and gives similar ratings on other performance elements. Thus overall evaluation is significantly influenced by a single factor. Such a perception undermines the importance of other elements and leads to an unbalanced performance assessment of the individual.
For example a manager rates a worker very high on quality because of her immaculate attention to details and lack of defects in her work. Then assuming the individual to be an overall high performer based on the quality of her work output, the manager rates her very high on efficiency, responsibility, punctuality, etc. without taking an objective look at her performance in these areas.
Strictness, Leniency and Central Tendency Bias
People differ in their tendency to evaluate people or performance. Some supervisors are very strict or conservative in their ratings and generally give low scores in their evaluations. This tendency may make high performers attain somewhat average ranking and average performers appear as poor performers. Raters with such tendency are known to have a strictness bias.
On the other hand, some supervisors demonstrate a leniency bias and rate their subordinates very liberally which may make even average performers seem like star performers, attaining very high performance scores.
And yet, others "play safe" by rating around the golden mean - the average. This may be done to avoid the necessity to justify scoring across the two extremes as some systems expect managers to specify additional comments as they give too high or too low ratings to employees. This rating error is known as the central tendency bias.
Performance appraisal involves assessment of employee performance for a specific period - quarterly, annually etc. People may not perform uniformly throughout that period. We all face highs and lows and demonstrate variance in performance due to numerous factors. It is therefore very important to review performance demonstrated throughout the period under consideration.
Often however, recent events tend to overshadow the overall performance. People do have "short memories". Thus a person who has worked very hard and excelled throughout the year, but for some inadvertent reasons had faced performance issues in the last weeks or month may at times get a poor appraisal from the supervisor, showing a recency bias.
When supervisors rate employees one after another, rating of an exceptional performer or a very poor performer could affect the subsequent ratings of other individual(s). This phenomenon is known as the contrast effect. For example, let's suppose that a supervisor has just rated an outstanding performer and is now evaluating an individual who is also a good performer but there seems to be a contrasting difference between the abilities and output of the two individuals. This significant difference in performance or employee competencies could lead the supervisor to rating the second individual as an average performer.
Personal beliefs, attitudes, assumptions, experiences, preferences and lack of understanding about a person, class or a phenomenon can lead to an unfair evaluation which is off from reality. We all suffer from these shortcomings, consciously or unconsciously, while making everyday judgments about people, things, events etc.
It is especially important to be aware and sensitive to possible biases, prejudices and stereotypes while making judgments about employee performance. While many of the prejudices operate covertly and unconsciously, others strike us through conscious thoughts and feelings. Understanding common biases and being on guard while appraising can significantly raise the objectivity of the evaluation process.
Examples of personal bias include a bias against a race, ethnicity, religion, age, sex, or assuming that certain type or class of people are not suitable to perform a specific job or function etc. If you believe for instance, that women are emotional and men are rational, then chances are that you would not employ a female worker for a role that involves making objective decisions. Similarly, an assumption that young workers are faster and more efficient than old workers would make it more probable to give a higher efficiency rating to a younger worker than an old worker.