Morality Meets Risk: Predicting Others’ Excuse-Driven Behaviors

The experiment consists of two parts and adopts a within-subject design. The first part is a decision task for subjects to elicit their excuse-driven risk preferences. The second part is a prediction task to investigate participants’ predictions of others’ excuse-driven behavior under the same risk scenario. Two parts are presented in random order at the individual level.
In each part, there are four blocks of decision tasks including 12 price lists. Price lists in each block only differ in 3 different probabilities, i.e. {90%, 50%, 10%}, and four blocks are presented in a random order for each participant. The price lists include similar decisions as elaborated in Dong and Zheng (2023).

2023-08-07

2023-09-30

- excuse-driven risk preferences: participants' valuations of different lotteries
- predictions of others’ excuse-driven behavior

Building upon the work of Dong and Zheng (2023), we conduct a laboratory experiment consisting of two essential parts: a decision task and a prediction task. The decision task will assess participants’ excuse-driven risk preferences, while the prediction task will delve into their predictions of others’ excuse-driven behavior under the same risk scenario. In the decision task, participants complete a series of price lists involving risk. The price lists include similar decisions as elaborated in Exley (2016). Participants’ choices reveal their certainty equivalences of different lotteries such that we can estimate their excuse-driven risk preferences. In the prediction task, participants face the same decision tasks as in the decision task. Instead of making decisions for themselves, participants are asked to predict the responses to these tasks by other participants in their session. Two parts are presented in random order at the individual level.

In each part, participants encounter twelve price lists. Each price list consists of 21 binary decisions. Participants make binary decisions between Option A and Option B for participants themselves or for the other participants in the same session. Option A columns always involve a consistent self-lottery or others-lottery given to themselves or other participants. The lottery yields 10 yuan with probability P and 0 with probability 1 − P, P ∈ {0.90, 0.50, 0.10}. Option B columns always involve a self-certain amount or an others-certain amount given to the corresponding participant. The amount increases evenly from 0 to 10 yuan along the 20 rows of the price list. Participants always start by selecting Option A and then switch to Option B at a switch point where Participants are indifferent between the lottery and a certain amount. As such, we can estimate the certainty equivalents of self-lotteries or others-lotteries in terms of self-valuations or others-valuations. The estimation process closely follows the derivations of Exley (2016) and Dong and Zheng (2023).

randomization done by a computer.

individual.

No

100 subjects in Beijing (50 males and 50 females), and 200 full-time students from Renmin University of China.

100 subjects in Beijing (50 males and 50 females), and 200 full-time students from Renmin University of China.

300 subjects in the decision task and 300 subjects in the prediction task.