How good of a driver are you? Pretty good? Pretty great? Maybe the next Jeff Gordon, if you only had some training and a jumped-up Chevy? Well, perhaps -- but probably not, and you'll probably never know.
It's difficult, almost impossible, for us to accurately evaluate our competencies. So says David Dunning, Ph.D., professor of psychology at Cornell University and author of several books and papers on accuracy and illusion in human judgment.
Dr. Dunning focuses on the difference between people's perceptions of their abilities and reality -- a gap that can make all the difference in business. Why? Because our own incompetencies blind us to our incompetence. Employees are often asked to tackle something new at work; some will excel while others will fail. Is there a way to predict that failure in advance and avoid it? Furthermore, decision makers must have confidence in their fundamental ability to do their job, but when that confidence is misplaced, even once, the whole organization can suffer. Most business missteps and mistakes are essentially errors in judgment of judgment.
There are ways, however, to objectively evaluate your competence before you fall on your face. In this interview, Dr. Dunning discusses how to work around blind spots, how to make critiques more effective, and what to do when a coworker fails to accurately assess his or her competence.
GMJ: Why do people tend to overestimate their abilities?
Dr. Dunning: There are many, many reasons. The first is the spin we tend to give the feedback we receive about ourselves from the outside world. That is, we claim credit for our successes and lay blame for our failures elsewhere. Second, what people tell you to your face is never exactly what they're saying behind your back. That will give you an inaccurate idea of your abilities. And finally, people just don't have all the information they need to be able to see themselves accurately, and what they miss tends to leave them overconfident.
When we're incompetent, we're not often in a position to recognize that incompetence. Often we make errors of omission because we're not aware of how we could have done a task in a better or a different way. But because we are unaware of these alternatives, we think instead that we've done just fine.
So there are just a whole host of reasons why people generally, but not always, are left with a sense of confidence that may not be justified.
GMJ: There's a lot of research into gender differences in self-appraisal of competence. For example, some research shows that men tend to overrate and women under-rate their ability to pick stocks. Both views are inaccurate, though. Is that same gender dichotomy true in other aspects?
Dunning: No. At least in American culture, you find that both genders tend to be overconfident, but the tendency will differ depending on what area of life you're talking about. So it may be true that men are overconfident about their ability to pick stocks, but if you move to, say, knowledge about literature or aesthetics, the gender difference may go away or reverse.
If you take a look at teenage kids, boys will be more positive and overconfident in their ability to deal with science than girls are. But if you move to English, that gender bias goes away. In the North American culture, if there's going to be a bias that people on average tend to have, it's to be overconfident, though that obviously doesn't happen all the time. And that's not necessarily true in other cultures.
GMJ: What's the danger of being overconfident? What's wrong with being wrong?
Dunning: There are some areas where it could be right to be wrong, but I think we all can easily imagine areas where overconfidence can certainly get you into trouble. I wouldn't want to be an overconfident gambler. I wouldn't want to be an overconfident airplane pilot. I wouldn't want to be a doctor who doesn't know when he or she has to call in a specialist for a consult.
There are a number of areas where overplaying your expertise can have bad effects for you and the people around you. Now there might be some areas, and I think this is underexplored in psychology, where being overconfident and being unrealistic may actually be helpful. Those areas tend to be where people are facing the extremes of life -- like you're putting your life together after your country has gone through a civil war, or you're facing a cancer diagnosis.
GMJ: Shelley Taylor [a professor of social psychology at UCLA] recently won the Clifton Strengths Prize for, among other things, researching just that -- she studied breast cancer patients and found that positive illusions are enormously therapeutic. [See "A Positive Approach to Workplace Stress" in the "See Also" area on this page.]
Dunning: That's right. I think the way to think about psychology is that everything is true -- but only in its specific context. So my take is that there are areas where overconfidence may be helpful. Something that energizes you, even if it's unrealistic, would be good for you in a difficult situation.
On the flipside, though, there are other contexts or situations where the exact same tendency is not something you want to display. I would not want to be a person with foremost confidence in my poker ability going up against a professional like Gus Hansen, for example. That's a situation where you're much more likely to be dead money than a winner. The consequences of overconfidence do depend on the exact situation or the exact task you're facing.
GMJ: People need to accurately judge their abilities, sometimes more than others, as you said. How can you figure out what you can't figure out?
Dunning: One of the pet phrases I have is "The road to self-insight runs through other people." Other people can often give us invaluable feedback that can really correct an illusion that we're suffering from.
One of my favorite, but most chilling, findings is from a study that surveyed surgical residents. They were asked about their surgical skills, and then they were given the standardized board exam. The residents' views of their skills didn't predict at all how well they did on the exam. But the impressions of their peers and their supervisors strongly predicted how well they did. Thus, there are times when what other people think of you can be an invaluable source of what you need to work on and what you're already good at.
And it can happen in different ways -- you don't always necessarily need formal feedback. If you just observe other people and see how they handle situations that you come across as well, you can more accurately judge your own skill in that situation. It's called benchmarking. I mean, it's no secret. Often you find different or better ways to deal with situations that you just hadn't thought of.
As I mentioned, people can't be expected to be aware of their errors of omission. But if you see how other people handle the same situation, it may clue you in to things that you didn't know you didn't know. And that can make you more accurate about yourself and more competent.
GMJ: Are there other ways to become better at self-evaluation?
Dunning: You can self-test, though it's easier to do it in some areas than others. When you're doing a task, evaluate yourself, and then try to get an evaluation from some outside source just to see if your evaluation agrees with more objective evaluations from the outside world.
One interesting thing for organizations to consider is, when employees are being trained, such as in technical skills, give them tests to evaluate their progress, but also have them estimate how well they did on the test. That may go a long way to alerting people to deficits that they didn't know that they had -- both in their skills and their ability to judge those skills.
You really do need some outside agent to point out that you have a deficit that you weren't aware of. You can't depend on your own devices; you really do have to seek help to get a better, accurate image of where your shortcomings lay.
GMJ: When you critique others, particularly people who work for you or maybe even peers, what do you do? Do you give an accurate critique or a tactful one?
Dunning: Giving feedback is a tricky business, and nearly 40% of feedback programs actually demotivate people. There is a skill to be learned here, and there are two things we can do to give feedback that's motivating, accurate, and tactful. The first thing is to give feedback that is concrete, as opposed to feedback that's about the person's character. You want to talk at the behavioral level. Feedback should not feel like a character attack, but rather a helpful suggestion. The other thing is to not only point out the bad, but point out the good, at a behavioral level.
So when you give people feedback, give them feedback that's both positive and negative. If all the feedback is just negative, negative, negative, they might develop some psychic calluses against that feedback.
The last thing I would mention, though, is that feedback becomes more risky and the consequences are higher if you receive it only rarely. Instead, to the extent that feedback is a small event that happens frequently, every piece of feedback carries less of a threat. You don't want managers and employees to be giving everybody feedback every five minutes, but giving feedback often and in small doses removes or reduces the threat level associated with it. You might want to spend more time with employees, giving them explicit goals for the week, the month, and the year.
GMJ: Because everybody hates annual reviews, right?
Dunning: Oh, everybody hates reviews -- giving as well as receiving them. And that creates a problem that managers have to avoid, which is waiting until you're angry to give the feedback you really want to give.
Think of what it's like to hear feedback from an angry person. Why on earth would a person listen to you when you're yelling at him? You're a crazy person; you are not giving objective feedback. One of the reasons to give feedback in frequent small doses is so you're not waiting until you blow.
GMJ: How do you measure improvement in subjective areas? How can you tell whether you're getting better at things like social situations or thinking creatively?
Dunning: That's a very good point, because social situations are subject to a lot of nuance. Those are inherently difficult areas to judge yourself, but even a little feedback can be tremendously helpful. And so even in these difficult situations -- and they are difficult, and people should just recognize that they're difficult -- getting outside feedback can still be very, very helpful.
Ninety-five times out of one hundred, seeking outside sources of feedback probably is going to be more helpful than confusing. Now, mind you, it's not going to work in all instances, but over the long term with a lot of people, it will be helpful. Even though it's stressful to go through, it can be some of the most useful information that we ever receive.
Nothing is ever going to be a panacea that works perfectly in all subjective circumstances. But we have things that can, even in the face of all that, make us better and make our situation better.
GMJ: How should you deal with someone who has gravely overestimated his or her own abilities and made a horrible mistake?
Dunning: Avoid something that can be read as a character attack. We're all prone to see a character attack -- that's a human tendency -- even if none is meant. You want to talk about the specific behavior, the specific consequences it created, and then point out, unless you're firing them, what they can do differently in the future. Suggest a way of improvement or a way of repair.
Some mistakes are so big that the person has to be fired; there's just no way around it. But if you're going to keep the person on, don't wander into character attack, especially if there's an angry undertone to your voice. I think the thing that you want to do is be concrete and behavioral as opposed to, "My God, you really screwed up big, didn't you?" That just won't do.
-- Interviewed by Jennifer Robison