In the second experiment, 63 students were asked to estimate the cost of tuition — as calculated by U.S. News & World Report — for each of seven public and private U.S. universities that appeared in random order on a computer screen. The participants answered questions about how confident they were in their responses and how powerful they felt in their relationships with others. The participants were then provided with another student’s estimates, which, unbeknownst to them, happened to be the correct numbers, and were given a chance to change their responses. Just as in the first study, women were significantly more receptive to outside input than men, and power correlated with confidence and an aversion to incorporating others’ views.
In the third experiment, more than 250 people ranging in age from 18 to 65 were asked to estimate the value of coins that filled three jars. The participants were then “primed” to be in either a high-power or a low-power group by being asked to recall and describe in detail a situation in which they held a great deal of influence or very little. A third (control) group was merely asked to describe all their meals for the past 72 hours.
After this “power manipulation,” the participants rated their level of confidence in their answers about the coins and were invited to change them; as in the second experiment, they were apprised of an estimate from a peer advisor that happened to be correct. Once again, participants in the higher-power group placed significantly less weight on others’ input. “Power elevates confidence in the accuracy of one’s judgment, which in turn reduces advice taking,” the authors write.
In the final experiment, the researchers examined whether the effects of power on advice taking reduced the accuracy of the participants’ ultimate judgment. In this study, which was similar to the second one, 126 undergraduate business students were asked to estimate the cost of tuition at each of seven American universities and were primed to be in a high-power, low-power, or control group. Again, they rated their confidence before being offered any advice. But this time, rather than providing generally accurate estimates, the peer advisors gave participants random answers that varied in accuracy. After submitting their final answers, the participants were asked to rate their confidence in their post-advice estimates.
In addition to confirming the previous experiments’ finding that more powerful people were less likely to take advice and were more likely to have high confidence in their answers, this final experiment showed that high-power participants were less accurate in their answers than low-power participants. By calculating the mean deviation between respondents’ initial estimates and the true answers, the researchers showed that low-power participants came significantly closer in their final estimates to the real tuition numbers because they “averaged” their initial guesses with the input from the advisors.
The researchers propose that their findings have troubling implications for organizations — and that power could negatively affect not just advice taking, but also an individual’s approach to seeking help or accepting performance feedback. But because power and confidence are so interrelated, there are ways to mitigate the problem. By “directly addressing the inflated confidence levels of powerful individuals,” the researchers write, “organizations may be able to help people with power take (and/or seek) advice when it is valuable to do so.”
For one thing, organizations could formally include advice gathering at the earliest stages of the decision-making process, before powerful individuals have a chance to form their own opinions. Encouraging leaders to refrain from commenting on decisions publicly could also keep them from feeling wedded to a particular point of view.