09 Nov Human Resources Individual Evaluation
The appraisal of individual employees is almost inevitably dysfunctional. Organizational performance is improved through refinements to the overall system or process of production, which must be the focus of attention. According to this line of thought, individual performance appraisals typically divert attention from more important tasks; they focus people’s attention on alleviating symptoms of poor performance rather than identifying root causes, and they serve only to demoralize workers who find the evaluations unfair or inequitable. Deming argues that performance can usefully be appraised at the system level only, because it is system improvements that must be stressed.
One can legitimately ask why summary evaluations are necessary. If the point of performance appraisal is to help the individual to comprehend what the organization desires or to aid in self-improvement, then a vague, multi-criterion evaluation -“You did well in sales level, but you need to pay more attention to established customers, at least as measured by repeat sales”-is often adequate and even superior to a single precise summary statistic, such as “Overall, you are a 3 on a scale of 1 to 6.”
There is no one-size-fits-all cure for this conundrum, but three compromises can help in many instances. First, even when a summary statistic is computed, it is sensible to share with the individual his ranking along the different aspects of performance that go into that ranking, together with at least an informal sense of how he can improve his summary statistic. “You did well in sales level, but you were below average in attention to established customers. To move from an overall evaluation of 3 up to a 4, you can either…”
Second, it is often effective to establish minimally acceptable standards in each of several aspects and, if all the thresholds are met, evaluate the individual according to a weighted average of the different performance indices. This approach works well for guardian jobs, where the minimums are set at levels that avoid disasters. When jobs mix star and guardian elements and it is desirable that individuals star in different ways, it can be effective to measure performance on the basis of the best single aspect instead of a weighted average, as long as minimum standards are met in all aspects. For instance, think of physicians who are engaged in clinical research, teaching, and treating patients in a university hospital. There may be good reasons why a university hospital would want to have physicians doing all three of these activities, especially because of complementation among them; that is, physicians doing applied research and teaching medical students will benefit from having an ongoing practice treating patients in their area of expertise, and doing research and teaching keeps clinicians plugged into the state-of-the-art, which is likely to improve their effectiveness in treating patients.
But particular physicians may be better able to deliver star performances as researchers, others as teachers, and others still in particular areas of clinical practice, such as mastery of an esoteric type of organ transplantation that differentiates the hospital from its competitors. A hospital in these circumstances might want to adopt an assessment scheme for physicians in whom they are evaluated based on the task research, teaching, or clinical practice-that they have chosen to make their primary concentration, subject to maintaining acceptable performance at the other two tasks. (And because treating patients has much stronger guardian aspects than do research or teaching, we would expect to see particular emphasis on maintaining minimum standards in clinical practice among all physicians at the hospital.)
Third, if multi-criterion evaluations are desirable on other grounds-such as providing richer and more accurate feedback, guiding corporate training efforts, or communicating multiple organizational goals – you might consider a system in which, except for truly exceptional performers, salary distinctions are not made at all or are based on seniority and rank, and promotion is detached to some extent from formal annual performance reviews. The point is to remove contention from the evaluation process to the greatest extent possible, so that evaluators are as comfortable as possible providing candid feedback and the individuals being evaluated can approach the assessment process with a relatively open and non-hostile mind.