This question always seems to get more than its share of attention and emotion: “should we use a three-point scale, a four-point scale or a five-point scale?.” The answer depends on your company’s performance rating history by scale point and on what is intended to be accomplished by any change. Here are the possibilities and the suggested approaches:
Existing Rating Scale Has Three Points and Managers Do Not Apply the Lowest Rating Even Where Appropriate – In many companies’ cultures, using two ratings of decent performance and a third rating reflecting sub-par performance is not workable. Managers want to be able to assign a less-than-expected score without placing the individual in the lowest rating category. They will work hard to avoid the proper rating by using the middle rating with the need for improvement expressed in the text of the evaluation or by attempting to add a minus sign to the middle rating for weaker performers. In this case, the best practice is to switch to a four-point rating scale where the point added just above the lowest possible rating reflects need for improvement but not unsatisfactory work.
Existing Rating Scale Has Four Points Resulting in an Absence of an Appropriate Rating Point for Solid Performance – Many companies want the bulk of employees clustered in a central performance rating with only those demonstrating truly exceptional or sub-par performance receiving a higher or lower rating, respectively. It is difficult to achieve this result using a four-point rating scale. In this case, the best practice is to switch to a five-point rating scale where the middle rating point reflects solid performance. Use of a three-point rating scale is also a possibility, but the result will often be the previously discussed reluctance to apply the lowest rating properly.
Existing Rating Scale Has Five Points and Managers Overwhelmingly Assign the Middle Rating Showing an Unwillingness to Differentiate Performance – This is not an uncommon situation. It results where, because a company’s leadership wants strongly differentiated performance, there is pressure on managers to avoid assigning too many ratings to the upper rating categories, but managers themselves have a reluctance to use the lower rating categories. In this case, the best practice is to switch to a four-point scale, forcing managers to make a decision that separates performers in the top two categories from those in the lower two categories.
It is important to keep in mind that all of these issues can also be addressed by imposition of a forced distribution rather than a rating-scale change. Forced distributions, however, have other implications as explained in the numerous media discussions of rank-and-yank such as the Conference Board’s blog piece, It’s Time to Bury Forced Ranking Once and for All and my blog, After Rank-And-Yank: Achieving Balance in Performance Management. In addition, managers prefer a solution which directly addresses the weaknesses of an existing rating scale over a solution which makes them assign ratings that they believe are not fair. Whichever approach is taken, it will need to be accompanied by good leadership communications and education. It also should not be disturbing if this issue needs to be revisited periodically. Both business needs and leadership rating-distribution philosophies can be expected to change over time.
II. Useful materials for performance review scale