Public Release: 

Confident Committee Not Always Best At Solving Problem, Scholars Say

University of Illinois at Urbana-Champaign

CHAMPAIGN, Ill. -- A decision made with beaming confidence by a committee may not necessarily be the best solution, researchers say. One may need to consider the type of problem that was addressed and the influence of individuals who were in the group.

Confidence and accuracy appear to reflect social influence exerted by individuals within a group, report University of Illinois psychologists in a paper published in the July issue of the Journal of Experimental Social Psychology. The findings resulted from a study of decision-making by individuals, by groups of two members and by groups of five members.

In general, researchers found that groups perform better than individuals, and that the five-member groups were the most effective for subjective tasks involving judgment. Where the picture blurred, however, was in the confidence levels: Group confidence of a solution was stronger than that held by individuals, and group confidence can be justified but also sometimes be unwarranted.

Confidence should be placed in context of the type of problem being solved, said Janet A. Sniezek, a professor of psychology. With an objective issue, such as a math problem, it is likely that someone in a group can clearly demonstrate an answer that others will accept. Accuracy isn't so clear when the problem could have alternative solutions, Sniezek said.

In the study, the groups considered tasks that varied in the difficulty of demonstrating why a solution is a good one. Each individual was asked to report his or her confidence level in solving the problem at both the beginning and conclusion of the task.

Groups were more effective than individuals in solving subjective tasks, such as definitively forecasting a result. In fact, large groups often displayed unrealistic overconfidence. Such a possibility can occur when individuals who are not confident entering the decision-making process end up latching on to a well-presented explanation for a suggested solution, Sniezek said.

"Confidence is an imperfect indicator of accuracy. We may have few other clues to go on in group settings, but it's good to push and to explore to get the reasons behind someone's opinion," she said. "Don't let someone's certainty be a proxy for explanation. This may not always be possible. People should recognize that an individual's confidence can be influential above and beyond the reasons."

A subsequent, as-yet unpublished study looked closer at the impact of individual influence on a committee's decision-making. In a report in May to the Midwest Psychological Association in Chicago, co-investigator Paul Zarnoth, a U. of I. doctoral student in psychology, said that a person with high confidence and strong arguments wields substantial influence.

However, the study also found that if the highly confident person's arguments are not demonstrated convincingly, then the influence can be totally lost, Sniezek said. "It is reassuring to know that there appears to be limits to confidence's influence," she said.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.