ACCOUNTABILITY KEEPS DECISIONS MORE HONEST –SOMETIMES– SAYS STUDY OF TWO PROFESSIONS (March 14, 2000)

"Decision makers should not believe things simply because they want them to be true," says Professor J. Edward Russo of the Johnson Graduate School of Management, Cornell University. "As obvious as this principle may be, however, it is often violated.

"We found that while making a decision, professionals distort information to support whichever alternative is currently leading. Holding decision makers accountable reduced distortion, but for salespersons only, and even then to a limited extent."

The study, "Predecisional Distortion of Information by Auditors and Salespersons," is by Prof. Russo; Margaret G. Meloy, Virginia Polytechnic Institute and State University, Blacksburg, Virginia; and T. Jeffrey Wilks, Cornell University. It appears in the current issue of Management Science, an INFORMS publication.

Distortion of Information is Costly
As people are deciding between two alternatives they may distort new information to support whichever alternative is tentatively preferred.

"Predecisional distortion of information is costly," says Prof. Russo. "If an important business decision has to be made and new information can’t get through, how do you make the right choice?"

The presence of predecisional distortion of information was tested by the researchers in decisions made by the two groups of professionals. Both exhibited substantial distortion of information, the study found, with little reduction for professional decisions compared to nonprofessional ones.

Significantly, auditors’ distortion was substantially less than that of salespersons. Holding professionals accountable for their decisions, akin to a supervisory review, lowered distortion somewhat for salespersons but not at all for auditors. The latter seemed to act as if they were always being held accountable.

For those seeking practical remedies, the researchers sound a note of caution. The researchers write that because people seem unaware that they are distorting information, at least at the moment this bias is occurring, they are fully convinced of the soundness of their choices. This may make it difficult for distortion to be detected by decision makers themselves or even by supervisors who cannot completely duplicate their subordinate’s knowledge.

The Experiment
The auditors who participated in the study were 90 employees of KPMG Peat Marwick attending a three-day training session. The 76 salesperson who participated worked for a large pharmaceutical company whose representatives asked that the company name not be disclosed. The experiment was administered to this group during a three-day company quarterly meeting.

The researchers selected auditors as a group of professionals less likely to distort information, among other reasons, because of legal liability associated with biased audits. The researchers chose salespeople as a group more likely to distort information, in contrast, because successful salespeople are characterized by their sense of conviction; they posited that this tendency toward conviction in order to persuade buyers might make salespeople more likely to distort information.

Members of each group were presented with three scenarios, one of which was related specifically to their profession, one which was a business question that could be answered by members of either profession, and one whose responses could be compared to those from a control group of students. The experiment also included variations that were used to test the effectiveness of accountability measures.

The decision specific to auditors involved a scenario in which the subjects had to prioritize responsibilities to two publicly traded client companies. The decision specific to salespeople involved a scenario in which the subjects had to decide which of two physician clients to visit during a brief trip to acity in the salesperson’s territory.

The second professional decision involved choosing between two restaurants for a business dinner. The third decision, which could be compared to a control group, involved the professionals acting as consumers choosing between the merits of two dry cleaners.

The Institute for Operations Research and the Management Sciences (INFORMS®) is an international scientific society with 12,000 members, including Nobel Prize laureates, dedicated to applying scientific methods to help improve decision-making, management, and operations. Members of INFORMS work in business, government, and academia. They are represented in fields as diverse as airlines, health care, law enforcement, the military, the stock market, and telecommunications. The INFORMS website is at http://www.informs.org.