Effect of different peer information on dynamic optimization decisions

How to cite

Gunawardena, Sembukutti Arachchige Bhagya Niroshini; Kalayci, Kenan; Friesen, Lana. 2021. "Effect of different peer information on dynamic optimization decisions." AEA RCT Registry. AEARCTR-0008602. https://doi.org/10.1257/rct.8602-1.1

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

Sembukutti Arachchige Bhagya Niroshini Gunawardena* , Kenan Kalayci , Lana Friesen

* Primary investigator

Registry ID
AEARCTR-0008602
DOI
https://doi.org/10.1257/rct.8602-1.1
First registered
Last updated
Status
In development
Trial dates
2021-12-06 – 2022-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/8602

Abstract

This study investigates how different peer information affect individuals' dynamic optimization decisions over repeated lifecycles. We compare individuals who make decisions alone over the lifecycles to the others who receive either a selected peer's decisions only, decisions along with outcome, or decisions along with confidence in a selected sequence. For this purpose, this study uses a simple deterministic decision making environment with no savings or borrowings where an individual's current decisions affect the future outcomes through the utility function. The outcomes of this study would help to identify how and what peer information can be used to nudge individuals in optimal decision making paths.

Intervention

This study provides a selected peer's information for the subjects in treatment groups. The treatments are as follows.
1. Decisions only
2. Decisions with the outcome
3. Decisions with confidence

All the subjects participate in the experiment through the online platform Prolific and the intervention will be undertaken while they are performing the main task.

Experimental design

Subjects start with instructions to the main task in which they need to decide how much of their endowment to allocate as input in production. Subjects are informed that this task consists of 5 sequences of decision making, where each sequence consists of 10 periods. At the end of each sequence, subjects need to elicit their confidence about their decision. After the main task of decision making, subjects face two attempts of a task which measures the ability of backward thinking, a risk elicitation task, and a demographic survey. Except for the demographic survey, all the additional tasks are incentivised. At the end of these tasks, subjects receive information on their earnings in each task and total earnings.

Design details

Randomization method
Randomization is done by the computer program (oTree). Assuming subjects arrive in the experiment in a random order, the first subject register in the study will be assigned to treatment zero (control), the second subject into treatment one, etc. If a subject drops out before completing the experiment, a new place is opened by the Prolific, and a new subject can register. The treatment for this newly registered subject will be the next treatment from the last treatment. That is, if the last subject (before someone drops out) was assigned to treatment x, then this newly registered subject will be assigned to treatment x+1. The treatments are allocated iteratively (i.e. 0,1,2,3,0,1,...) to minimize treatment imbalance.
Randomization unit
Individual
Clusters
500 individuals
Observations
500 individuals
Arms
125 per each treatment - cluster is an individual
Primary outcomes
Individuals' earnings in each sequence of the main task and their confidence in earnings
Secondary outcomes
Individuals' decisions in each period of repeated lifecycles (sequences), the difference between the optimal decisions and individuals' decisions, and the difference between the myopic decisions and individuals'

Cited by

No citation data available for this registration yet.

Keywords