Role Playing: A Method to Forecast Decisions

By J. Scott Armstrong

The Wharton School, University of Pennsylvania

Summary:

Armstrong describes role-playing to be when an administrator asks people to play roles and use their “decisions” as forecasts.  These exercises produce a realistic simulation of the interactions among conflicting groups. Armstrong then gave a few situations were role play has been used.

1.     A special interest group considers a sit-in to convince the government to provide subsidies to its members. The government believes the subsidy to be unwise and is willing to make only minor concessions. How likely is it that a sit-in would succeed?

2.     A firm selling industrial products to a small number of customers plans major changes in its product design. The changes are risky but potentially profitable. It wants to make the changes without its competitors finding out.

3.     A law firm is considering strategies for a defendant. Which defense would be most persuasive to the jury?

All of the situations above are dependent on the interactions of two parties. Role-play can be used to help accurately forecast one’s own decisions and the reactions of other parties. Armstrong argues that role-playing is most effective for predictions when two conflicting parties respond to large changes. To use role-play a forecaster asks subjects to put themselves in specified roles. Then they can either imagine how they would act, act out their responses alone, or interact with others in the situation. The decisions that are made in a situation can vary based on the designated role.  An experiment by Cyert, March and Starbuck in 1961 proved that. The subjects were presented with the same data, however they made substantially different forecast depending if they were “cost analyst” or “market analyst.”

In the next section Armstrong listed conditions where the use of role-playing is favored. There is a balance in the amount of parties that interact in effective role-playing. It is easiest to copy situations in which only two parties interact. When many parties involved different viewpoints, matching the role-play to the situation becomes more difficult forecast. Role-playing is useful in situations where the interacting parties are in conflict. It is also useful in situations involving large change.

Role-playing is the preferred when predicting decisions in situations in which parties interact. It is especially important that when trying to forecast the outcome of a decision-making situation, the analyst ensures that the role-playing matches the actual situation. The key is well thought out casting, role instructions, descriptions of the situation, administrative procedures, and interaction among groups for effective role-playing. It can provide the most realistic representation of interactions among different parties. It’s a low-cost and confidential alternative to experimentation. Evidence showed that out of the situations listed above that role-playing was more accurate than expert opinions for predicting decision-making when there were conflicts between groups and when large changes were involved.

Role-playing is transferable across many disciplines.

Critique:

The author spends a lot of time going through the steps of setting up role-play in this article and only does a shallow dive in to results of role-play. Also, the article is written on qualitative case studies. The argument for the value of role-playing would have been much stronger with statistics to support it. This is something that is hard to measure the value. Role-playing as a methodology is valuable because it can be used across most disciplines. In the situations the author laid out he was able to show that it cold be applied to businesses, polices and law makers, educations, and in healthcare. The author on touches quickly on role-play without interaction of multiple parties.  

https://repository.upenn.edu/cgi/viewcontent.cgi?article=1175&context=marketing_papers