טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentDruyan Meir
SubjectOn Individual Decision Making in a Group Context: When
Does the Group Context Facilitate Risk Taking?
DepartmentDepartment of Industrial Engineering and Management
Supervisor Professor Eldad Yechiam
Full Thesis text - in Hebrew Full thesis text - Hebrew Version


Abstract

 This study focus on the effect of making repeated decisions in a social context. Here it is argued and shown that in the social context there is an additional unique information component. When seeing the outcomes of others’ choices decision makers occasionally receive information concerning outcomes from unselected alternatives. This latter information, called foregone payoffs were found to increase risk taking when the risky alternative produces low-probability high-outcome penalties (Yechiam & Busemeyer, 2006).

This prediction is contrasted with alternative accounts concerning the impact of the presence of other people on individual choice behavior. Under one theoretical account the presence of others’ increases conformity to accepted norms, which amongst young individuals often includes risk taking (Ariely & Levav, 2000). The main difference between the predictions based on foregone payoffs and the alternative approaches is that the former imply an interaction between the social context and the incentive structure.

Experiment 1 presents a problem called “1/20 Problem”, in this problem the risky alternative produces favorable outcomes most of the time along with rare but highly negative penalties. In the control condition participants made decisions individually and without foregone payoffs. In the social context condition participants made decisions individually but saw each others’ choices and outcomes. In addition, I included a foregone payoff condition. As predicted, making individual decisions in a social context increased risk taking. Moreover, the choice proportions in the social context condition were very similar to the proportions in the foregone payoffs condition, suggesting that the decision makers behave similarly in these two situations.

Experiment 2 presents a problem called; the 1/2 Problem, where the risky alternative produces more frequent but smaller negative outcomes. As predicted, the social context and the foregone payoffs did not lead to a significant increase in risk taking. Accordingly, the effect of making decisions in a social context on the degree of risk taking appears to be very similar to the effect of foregone payoffs and to be moderated by risk type.

The goal of experiments 3 and 4 was to compare the effect of viewing others’ choices to the effect of being watched by others. In experiment 3 I used the 1/20 Problem, as predicted the effect I found in experiment 1 emerged only in the viewing condition. While in experiment 4 I used the 1/2 Problem, and as predicted, the effect I found in experiment 1 and 3 did not emerge at all.