This content is locked. Please login or become a member.
Fostering Mutual Understanding: Rethink Performance Evaluations, with Valerie Purdie-Vaughns, Associate Professor of Psychology, Columbia University
Performance evaluations are something that are happening more and more in companies, and the irony of performance evaluations is that companies are actually becoming more and more technical in their performance evaluations where the science is telling us that there’s more and more bias in performance evaluations. So one question is, where are the biases? One of the biases is, when you are observing a person’s performance – so if I am observing, say, a female executive, I’m going to remember when they’re doing things that are consistent with gender stereotypes. So if they’re having a bad day at work, or something happened, or they’ve been diagnosed with cancer and you just happen to see them tearing up, you’re going to remember that and say, “This woman’s emotional.” What you’re not going to remember are all of the hours that she was there at 7 p.m. or 8 p.m., or 5 a.m., or working on Saturdays and Sundays. If a woman has a picture of her children at work, you’re going to remember that and think about those types of instances. They become more salient when six months later you’re doing performance evaluations. So the first step is, we need to retrain the brain to pick up on the whole range of what people are doing. That doesn’t happen at the performance stage; that happens all throughout the year. So that’s the first thing.
The second thing is, during the actual performance evaluation, one of the findings that is really important in psychology is that people are making the wrong kinds of comparisons. So there’s say, for example, again, you take a woman in the workplace, you’re saying, “This woman is actually doing really well, given that she’s a woman in finance.” So what happens is, we actually are starting to compare what we think are apples to apples and orange to oranges. The problem with that is, when you move up and up in a company and say you’re a high-level executive. There’s only one woman and there’s like ten other men – there’s no other comparison. So now you’re saying, “This woman is okay for a man, but would be really awesome relative to a woman.” So the problem is, performance evaluations, ironically, particularly for underrepresented groups, get worse and worse and worse the higher they go up because they have solo status.
360s are extremely popular, where each person in the workplace is evaluating each other person. But if there is bias in the workplace – if there is a very strong stereotype about say, women in investment banking – then each person is going to still pick up on the same cues: this woman is emotional, this woman isn’t following oil in the same technical way that this man is. So there’s all of these really sort of important dynamics that happen at the point of how you observe people, but also at the performance evaluation stage. And these are things that people don’t know about, so what happens is you’re looking at data, or what you think is “data”, what you think are objective qualities of a person, but what’s baked into that are stereotypes.