Higher order risk preferences for money and food

How to cite

Drichoutis, Andreas; Cerroni, Simone; Kazemekaityte, Austeja; Vassilopoulos, Achilleas. 2024. "Higher order risk preferences for money and food." AEA RCT Registry. AEARCTR-0014075. https://doi.org/10.1257/rct.14075-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Andreas Drichoutis* , Simone Cerroni , Austeja Kazemekaityte , Achilleas Vassilopoulos

* Primary investigator

Registry ID
AEARCTR-0014075
DOI
https://doi.org/10.1257/rct.14075-1.1
First registered
Last updated
Status
In development
Trial dates
2024-10-09 – 2024-12-20
Countries
Private
JEL codes
C91, D81
Registry record
http://www.socialscienceregistry.org/trials/14075

Abstract

This study seeks to explore how people make decisions involving risk when it comes to two different types of goods: money and food. In particular, we are interested in understanding higher-order risk preferences, such as prudence (the tendency to be cautious about potential risks) and temperance (the ability to moderate risk-taking), when individuals are faced with choices that involve either monetary rewards or food items in a controlled laboratory setting.

In this study, participants are asked to make binary decisions between lotteries. We use the risk apportionment tasks of Deck et al. (2023) to measure prudence and temperance along with risk aversion. Subjects are asked to make lottery choices between either monetary rewards or food item rewards on a within-subjects basis. To study how risk preferences differ based on the type of food, participants will be randomly assigned to one of two groups. One group will make decisions in lotteries where the food rewards are healthy items (such as cheese), while the other group will face lotteries where the food rewards are unhealthy items (such as chocolates).

The results of this study will help us understand how people weigh risks when it comes to different types of goods, which could provide insights into economic and health-related decision-making.

Deck, C., Huang, R.J., Tzeng, L.Y. and Zhao, L. (2023) A Simple Approach for Measuring Higher-Order Risk Attitudes.

Intervention

The intervention consists of exposing participants to decision-making tasks where they make choices between binary lotteries offering either monetary or food rewards. Participants are randomly assigned to one of two conditions based on the type of food rewards:

Healthy Food Condition: Participants make lottery choices with potential rewards in units of cheese.

Unhealthy Food Condition: Participants make lottery choices with potential rewards in units of chocolate.

All participants will also participate in monetary reward tasks. The risk apportionment tasks of Deck et al. (2023) will be used.

These interventions are designed to compare risk preferences across healthy and unhealthy food items, and between food and money as reward categories.

Deck, C., Huang, R.J., Tzeng, L.Y. and Zhao, L. (2023) A Simple Approach for Measuring Higher-Order Risk Attitudes.

Experimental design

The experimental design consists of two within-subjects tasks. In each task, participants are asked to make decisions between two options presented as binary lotteries. These lotteries are designed to evaluate participants' risk aversion, prudence and temperance under two distinct conditions: for monetary rewards and for food rewards. Each participant will complete a series of binary lotteries for both types of rewards, but the order of the tasks (money or food first) will be randomly assigned for each participant to ensure there is no systematic bias based on task sequence.

Participants are randomly assigned to one of two treatment groups:

Group 1: This group will be presented with a series of binary lotteries in which the rewards are paid in healthy food items (specifically, units of cheese).
Group 2: This group will face a similar series of binary lotteries, but their rewards will be in the form of unhealthy food items (specifically, units of chocolates).

Moreover, all subjects will make decisions over monetary lottery choice tasks, thus providing a direct comparison on how they choose when it comes to food and money.

Lottery choice tasks are based on the risk apportionment tasks of Deck et al. (2023).

At the end of the study, a randomized payment procedure will be used. The computer will randomly select one of the choices made by the participant from either the monetary or food task to determine their bonus payment. This ensures that participants approach each decision with the possibility that it may affect their final earnings. The potential bonus will range from 0 to 26 units, where each unit represents either euros (for the monetary task) or units of food (for the food task).

In addition to any bonus they may earn, all participants will receive a flat rate participation fee of 5 euros, ensuring that every participant is compensated for their time, regardless of the lottery outcomes.

Design details

Randomization method
Randomization done by computer (using python "random" module) at the time when participant enters the experimental software.
Randomization unit
Randomization at the individual level.
Clusters
0
Observations
Each subject will make 27 choices for money and 27 choices for food. Therefore, the total N of observations will be 200*54 = 10800
Arms
100 subjects in the Money + Unhealthy food treatment 100 subjects in the Money + Healthy food treatment
Primary outcomes
Our key outcome variable is switching points in a series of binary lottery choice tasks for money and food.

Cited by

No citation data available for this registration yet.

Keywords