Persuasive Advertising - J. Scott Armstrong [201]
A process called “the devil’s advocate” (DA) involves gaining a group’s acceptance to arbitrarily select a group member whose task is to find the negative aspects in a given proposal, such as an advertising campaign. Once the defects are identified, the group can consider improvements.
The DA procedure has a long history. In 1587, the Roman Catholic Pope, Sixtus V, instituted the DA to help evaluate whether a person should be canonized as a saint. The Church used this procedure until the 1980s. In the latter half of the 20th century, it was recommended for evaluating proposals in organizations. In practice, however, people use the term to apply to any meeting in which they evaluate the merits of a situation.
A literature review found no support for the use of the DA even when properly applied. What appears to happen is that those presenting a plan believe that there is no hope of convincing someone who has been instructed to be completely negative (the DA). As a result, group members are led to reinforce their arguments for the current plan instead of using the criticisms to effect change (Nemeth and Nemeth-Brown 2003).
Experimental studies have shown that authentic dissent (that is, when people are arguing for what they believe) is more effective than the DA procedure. This occurs partly because the group realizes that an authentic dissenter might change, whereas the role of the DA is not to change. Furthermore, the assignment of the role did little to protect the people playing DA; they became unpopular. While this also happened to authentic dissenters, people tended to respect authentic dissenters for their courage.
Those with authentic beliefs do a better job in arguing their positions than do those who are merely assigned to the role. Based on their experiments, Nemeth et al. (2001) found that authentic dissenters were able to get group members to focus more on opposing thoughts than supporting ones, thus contributing to more opinion change.
With authentic dissent, potential dissenters realize that they might become unpopular; so they are usually reluctant to speak up. In addition, as with the DA, it is time-consuming because only one person dissents, while the rest defend; it can lead groups to bolster their arguments rather than to think about how to overcome objections.
These problems can be overcome by a procedure that I refer to as multiple anonymous authentic dissent (MAAD). This involves asking experts, including all those involved with a given proposal, to act as dissenters. Each person independently writes all potential defects that they perceive in a proposed campaign. They then send their defect lists (unsigned) to an administrator who organizes, edits, and circulates the list to the group. Each expert then assumes that each of the objections has merit, and describes ways to deal with the objections, then sends them (again unsigned) to the administrator. Finally, the suggested improvements are then summarized by the administrator and provided to the group so that they can make appropriate revisions to the campaign. This process can be repeated.
To ensure that group members respond in a timely fashion, the MAAD process can be conducted during a meeting by taking a time-out for individual work.
Judging from related research on group processes, MAAD should be effective at improving proposals. Because many people can be involved in finding defects, the likelihood that important defects will be discovered increases. This process is analogous to a scientific peer review. As is known from research on peer reviews, individual reviewers catch only some of the errors when they review papers for publication. For example, in an experiment on medical research, 68 percent of the “reviewers” of a fictitious paper with intentional errors did not realize that the conclusions in the paper were not supported