The Bomb Risk Elicitation Task (Crosetto and Filippin, 20131) is a straightforward way to measure a student’s risk preferences. A bomb is randomly placed in one of 100 boxes, and each student chooses the number of boxes she will open. She earns zero if the bomb is in one of the opened boxes. Otherwise, her payoff equals the number of boxes she opened.
Individuals differ in their tolerance for risk (there is no “right” answer). Risk preferences displayed is this game may predict behavior in other risky environments (e.g., a MobLab first-price auction).
Expected payoff is maximized by opening 50 boxes. Opening fewer boxes is consistent with risk aversion, while opening more is consistent with risk seeking behavior.
All students will play this game individually in their own group. This means that the randomization of the bomb location will be different for each student. The expected payoff from opening b boxes is b(100-b)/100, meaning expected payoff is maximized at b*=50 which consistent with risk neutrality. Opening fewer than 50 displays risk aversion, and more displays a preference for risk.
The results screen presents a table and a graph for each group, both showing the heterogeneity of risk preferences.
The table (Figure 1) summarizes each group’s choices, presenting the min, max, and median choices for each group. The graph generated for each group (Figure 1)—with choice on the horizontal axis and fraction of students making that choice on the vertical—presents the full distribution of choices.