Around the third or fourth century, the Chinese philosopher Sin Yu complained that the Imperial Rater of the Royal Court was showing favoritism in his ratings. It has been an uphill battle ever since.
Most people throughout history have despised appraising or being appraised at work. What’s despised isn’t so much the idea of a performance appraisal. Judging other people’s performance, after all, is almost instinctive, and most of us probably sense the need for performance appraisals in organizations. What’s despised is the way it’s done, which is usually badly.
Long ago I played Captain Appraisal and took audiences on a "time capsule" tour of at least 2,000 years of human foibles in the ways, some quite bizarre, that performance appraisals have been done. They include two of the most problematic yet enduring ways--ratings andrankings. When you can get an audience to laugh about the matter, you know that a deep-seated frustration has been tapped that needs venting.
Performance ratings, although almost universally loathed, still exist. When will we ever learn that dishonest ratings are inevitable because they’re so easy to fudge and so tempting to fudge when they’re tied to consequences such as bonuses and promotions? I’ve seen a rating distribution with 98 percent of the people rated "outstanding," the only people who "stand out" being those who didn’t get the fudged rating.
Rubber bands and bells
History is replete with doomed rating schemes and policies intent on foiling the Imperial Rater and placating the ghost of Sin Yu. A few examples are "rubber-band rating scales," periodically stretched in trying to keep ahead of ratings creeping upward; the "cracked bell," where ratings are forced into a bell-curve distribution; and "group fudge," in which it’s hoped that a truer rating will emerge from a circle of raters. These and other schemes are all part of the futile search for the Holy Grail of Honest Ratings.
When will we ever learn that dishonest ratings are inevitable?
Some organizations use rankings instead of--or in addition to--ratings. Rankings aren’t even performance appraisals, however, because appraisals in their correct form don’t rank people, but instead compare the performance of an individual against the expectations of performance by that individual. The worst flaw of rankings, though, isn’t that they’re not really performance appraisals but that forced judgments are required, and forced judgments will always have an arbitrary and unfair element to them. Thus, besides being invalid, rankings could also be considered an unethical practice.
Where does all of this leave us? Some critics argue for abolishingperformance appraisals altogether, but that’s not a viable option. Performance appraisals are an integral part of managing performance, and unmanaged performance simply can’t be allowed to happen. The process of managing performance is essentially one of holding people accountable for their performance. It starts with the setting of performance expectations and ends with sanctions--rewards and penalties--on appraised performance.
If we want to be responsible and get good performance, therefore, we’re stuck with performance appraisals. What we sorely need is a more tolerable and honest way of doing them. One alternative that has been successfully pilot-tested by this Captain is asking a series of questions about performance. The questions require yes-or-no answers. Since performance has two parts--behavior and results--the questions must relate to each part. For example,
- Did you exceed any of your objectives?
- Which objectives, if any, were clearly exceeded, and what were the extra gains?
- Were any unexpected and very tough obstacles encountered?
- Are there any skills you want and/or need developed?
In all, about 20 short questions are needed, some of which are required to identify any star performers.
While 100 percent honesty in performance appraisal isn't possible no matter how well it’s done, answers to yes-or-no questions ought to be closer to the truth than ratings are. Here are four reasons why.
First, answering yes when the true answer is no is more than a fudge; it’s a big, bald-faced lie, which requires considerable audacity to make, and isn’t likely to be made.
Second, appraisals should be reviewed and certified, and people should know that the appraisals are being checked. This reduces the incentive to distort what they’ve accomplished.
Third, many of the answers are readily verifiable. For instance, you can verify whether an objective has been met or even exceeded by comparing the documented results obtained with the criteria that were stated in the written objective. As for the more sensitive and more subjective questions relating to behavior, there should be a paper trail showing any problems that the company and the employee have been working on as part of the performance-management process.
Fourth, a person’s performance is often visible to colleagues--and it would likely take a conspiracy of two or more people to conceal a falsehood. This is possible, though not likely.
Whether yes-or-no answers are themselves indeed the answer to a 2,000-year-old vexation is open to further debate and experience. In the meantime, remember to put performance appraisals in perspective. While they’re necessary, they’re less critical than the setting of performance expectations from which everything else follows.
Workforce Online, July 2003 --Register Now!