Probability and Decision Making

Last registered on March 28, 2019

Pre-Trial

Trial Information

General Information

Title
Probability and Decision Making
RCT ID
AEARCTR-0004049
Initial registration date
March 25, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 28, 2019, 7:54 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Primary Investigator

Affiliation
National Taiwan University

Other Primary Investigator(s)

Additional Trial Information

Status
In development
Start date
2019-03-26
End date
2019-12-31
Secondary IDs
Abstract
This study involves the relationship between probability and a particular feature of decision making. Because of the sensitive nature of the research question, please wait for the study to be completed or contact the author, as subjects searching the trial registry are still able to see the abstract, even before the study is complete.
External Link(s)

Registration Citation

Citation
DeJarnette, Patrick. 2019. "Probability and Decision Making." AEA RCT Registry. March 28. https://doi.org/10.1257/rct.4049-1.0
Former Citation
DeJarnette, Patrick. 2019. "Probability and Decision Making." AEA RCT Registry. March 28. https://www.socialscienceregistry.org/trials/4049/history/44314
Experimental Details

Interventions

Intervention(s)
Please wait until trial is completed to see more information about the intervention, given the sensitive nature of the study (and subjects availability to see the trial details). You may also contact the PI for details in advance of the completion.
Intervention Start Date
2019-03-26
Intervention End Date
2019-06-30

Primary Outcomes

Primary Outcomes (end points)
The main outcome of interest is the statistical distribution of lying (or lack of lying). In this case, we look at the average number of subjects who are statistically likely to have lied in the first stage vs statistically likely to have told the truth.
Primary Outcomes (explanation)
The primary outcomes are constructed using the distribution of 1, 2, 3, 4, 5, and 6 on the 6 sided dies given the sample size. We can then compute the number of subjects who were likely to have been lying with 95% confidence.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Please wait until trial is completed to see more information about the intervention, given the sensitive nature of the study (and subjects availability to see the trial details). You may also contact the PI for details in advance of the completion.
Experimental Design Details
This study investigates to what extent lying aversion is the result of the deterministic nature of the lie. In other words, if lying does not guarantee success, but rather improves the chance of success, does this encourage subjects to engage in lie? There are two competing theories that could predict a difference in this sense.

In the first, fear of "karmic retribution" could cause subjects to be concerned whether lying now may cause a "bad" probabilistic outcome. This may lead to more lying averison. In the second, existing lie aversion may be more closely related to the disutility from "stealing". In this sense, a lie that deterministically benefits me may be quite different from a lie that only benefits me on average.

In other words, the intervention is a two stage 'game' where players roll dice. In the first stage, the subjects (privately or non-privately) roll a 6-sided die. In the second stage, subjects (privately or non-privately or commonly) roll a 10-sided die. If the second roll is less than the first, they get a large bonus in addition to the experimental show up fee.

The treatments have two primary arms, a 2 by 3 design (with one cell omitted). The first treatment arm is whether or not the first roll is private (experimenter cannot confirm or deny) or non-private (experimenter can view the outcome). The second treatment arm is whether or not the second roll is private (experimenter cannot confirm or deny) or non-private (experimenter can view the outcome) or common (a single roll, by the experimenter, for all subjects). The omitted cell is non-private in first round and non-private in the second round, as there is no decision making going on in that cell and it can be analytically estimated or simulated with monte carlo techniques.

One secondary treatment arm is the size of the reward for rolling the second die lower than the first die. This is to put the "value" of the effect on lying aversion in monetary terms.
Randomization Method
Within the experiment, dice rolls (6 sided dice and 10 sided dice).

For treatment, randomization is done at the session level (for details made clear in the experimental design).
Randomization Unit
Session level
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
15 sessions, depending on subject enrollment rate. 3 Sessions of each of the 15 cells
Sample size: planned number of observations
250 Subjects, approximately 17 subjects per session.
Sample size (or number of clusters) by treatment arms
3 sessions of each of the 5 cells in the 3 by 2 (omit 1).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Research Ethics Committee
IRB Approval Date
2019-01-18
IRB Approval Number
201810HS013

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials