Performance evaluations

I read a very interesting article in the Sunday New York Times about performance evaluation. As an arbitrator and mediator, I am frequently confronted with the results of a performance evaluation. I do have some basic problems with them and, in particular, with numeric scoring systems.

The most common system has points awarded often on a scale from one to five. Two issues arise quickly for me. One, if there is a forced distribution of scores (for instance - there must be 20% 1 and 20% 2 and so on), the system is based on a belief that in any organization there will be twenty per cent of the performers who are exceptional. What if in reality this organization happens to have 30% exceptional performers or 10%? Does that mean because of forced distribution there will not be an accurate scoring for some individual? Even though the person is by any objective standard exceptional, the score can not be given because the grouping is already full? I think that it is inevitable and inevitably leads to an inaccuracy in scoring. Second, if there is not a forced distribution system, then you encounter the "wine buying experience" problem. Let me explain. I was in my local liqour store on the weekend looking for a bottle of wine and I noticed that there is a marked increase in the advertisements of ratings that various wines have achieved. I assume that the scale is from 1-100 yet all of the wines that I saw had a rating between 88 and 93! This is just like the organization where everyone is a 4.

A third problem that I have with numeric job evaluation systems is that I am never sure what they mean - is a 2 twice as good as a 1 while a 3 is only fifty percent better than a 2?

Unrelated to my mediation and arbitration practice, I do not think that putting a number on a person helps them improve. If you tell an employee that she is a star performer - it is likely that the person already knew or at least thought that. If you tell an employee that she is an average performer, it is likely that you are telling her something less than what she perceives - is that likely to make her want to do better? I think that the best performance evaluations do not focus on numbers but instead focus on performance - what are you doing well? what else could you be doing? what do you need to do better/

The next time you present me with a performance evaluation - you should know that I am skeptical about what exactly it is telling me (or the person who was being evaluated in the first place).