French and Raven’s final source of power is expert power. Experts have knowledge or information, and conforming to those whom we perceive to be experts is useful for making decisions about issues for which we have insufficient expertise. Expert power thus represents a type of informational influence based on the fundamental desire to obtain valid and accurate information, and where the outcome is likely to be private acceptance. Conformity to the beliefs or instructions of doctors, teachers, lawyers, and computer experts is an example of expert influence; we assume that these individuals have valid information about their areas of expertise, and we accept their opinions based on this perceived expertise (particularly if their advice seems to be successful in solving problems). Indeed, one method of increasing one’s power is to become an expert in a domain. Expert power is increased for those who possess more information about a relevant topic than others do because the others must turn to this individual to gain the information. You can see, then, that if you want to influence others, it can be useful to gain as much information about the topic as you can.
Research Focus: Does Power Corrupt?
Having power provides some benefits for those who have it. In comparison to those with less power, people who have more power over others are more confident and more attuned to potential
opportunities in their environment (Anderson & Berdahl, 2002). They are also more likely than are people with less power to take action to meet their goals (Anderson & Galinsky, 2006;
Galinsky, Gruenfeld, & Magee, 2003). Despite these advantages of having power, a little power goes a long way and having too much can be dangerous, for both the targets of the power and
the power-holder himself or herself. In an experiment by David Kipnis (1972), college students played the role of “supervisors” who were supposedly working on a task with other students (the
“workers”). According to random assignment to experimental conditions, one half of the supervisors were able to influence the workers through legitimate power only, by sending them messages
attempting to persuade them to work harder. The other half of the supervisors were given increased power. In addition to being able to persuade the workers to increase their output through
the messages, they were also given both reward power (the ability to give small monetary rewards) and coercive power (the ability to take away earlier rewards). Although the workers (who were
actually preprogrammed) performed equally well in both conditions, the participants who were given more power took advantage of it by more frequently contacting the workers and more
frequently threatening them. The students in this condition relied almost exclusively on coercive power rather than attempting to use their legitimate power to develop positive relations with
the subordinates. Although it did not increase the workers’ performance, having the extra power had a negative effect on the power-holders’ images of the workers. At the end of the study, the
supervisors who had been given extra power rated the workers more negatively, were less interested in meeting them, and felt that the only reason the workers did well was to obtain the
rewards. The conclusion of these researchers is clear: having power may lead people to use it, even though it may not be necessary, which may then lead them to believe that their subordinates
are performing only because of the threats. Although using excess power may be successful in the short run, power that is based exclusively on reward and coercion is not likely to produce a
positive environment for either the power-holder or the subordinate. People with power may also be more likely to stereotype people with less power than they have (Depret & Fiske, 1999)
and may be less likely to help other people who are in need (van Kleef et al., 2008).
Although this research suggests that people may use power when it is available to them, other research has found that this is not equally true for all people—still another case of a person-situation interaction. Serena Chen and her colleagues (Chen, Lee-Chai, & Bargh, 2001) found that students who had been classified as more self-oriented (in the sense that they considered relationships in terms of what they could and should get out of them for themselves) were more likely to misuse their power, whereas students who were classified as other-oriented were more likely to use their power to help others